Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Models And Determinants For The Economic Allocation Of Retail Space In Urban Planning
(USC Thesis Other)
Models And Determinants For The Economic Allocation Of Retail Space In Urban Planning
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Copyright by
PHILIP HOWARD FRIEDLY
1966
MODELS AND DETERMINANTS FOR THE ECONOMIC ALLOCATION OF
RETAIL SPACE IN URBAN PLANNING
by
Philip Howard Friedly
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Economics)
January 1966
U NIVERSITY O F SO U TH ER N C ALIFO R N IA
THE GRADUATE SCHOOL
U N IV E R S ITY PARK
LOS ANGELES, C A LIFO R N IA 8 0 0 0 7
This dissertation, written by
Philip Howard Friedly
under the direction of h..i,U..Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R O F P H IL O S O P H Y
Dean
L 9 m i"
DISSERTATION COMMITTEE
TABLE OF CONTENTS
LIST OF TABLES
Chapter
I INTRODUCTION.........................
Purpose and Scope of the Study
Analytical Methods Employed
Definitions
Plan of the Study
II LAND USE PLANNING, THE MARKET MECHANISM
AND SOCIAL COST .....................
Local Government Planning and Land
Use
Land Use Planning and the Market
Mechanism
Resource Allocation and Social Cost
in the Urban Economy
III ECONOMIC BASE ANALYSIS ............
Historic Review
Measuring and Forecasting Local
Economic Activity Levels
Limitations of Economic Base
Analysis
IV TRADE AREA ANALYSIS .................
Historic Review
Measuring and Evaluating Local
Expe nd i tur e s
Limitations of Trade Area Analysis
Page
iv
1
18
52
130
TABLE OF CONTENTS— Continued
Page
Chapter
V SPACE REQUIREMENTS ANALYSIS ........... 199
Historic Review
Measuring the Productivity of
Retail Space
Retail Operation
Market Structure and Retail
Location
VI RETAIL SPACE ALLOCATION ............... 257
Local Public Planning Needs
A Model of Retail Space Allocation
Data Requirements of the Model
VII EVALUATION OF THE MODEL AS PLANNING
ANALYSIS........................... 321
The Model and Social Cost
Relationship of the Model to the
Municipal Planning Process
VIII SUMMARY AND CONCLUSIONS OF THE STUDY . . 342
Summary
Conclusions
APPENDIX..........................................350
BIBLIOGRAPHY......................................362
iii
LIST OF TABLES
Page
Table
1 Distribution of Consumer Expenditures
by Income Level Nationally, 1958 . . . 173
2 Distribution of Disposable Personal
Income Among Selected Retail
Purchases, 1929-1954 ................. 175
3 Distribution of Consumer Expenditures
Among Selected Retail Store Groups,
1935-1960 ........................... 176
4 Distribution of Disposable Personal
Income in California Among
Expenditures in Selected Retail
Store Categories...................... 179
5 Shopping Center Retail Expenditure
Potential..............................184
6 Input-Output Table ...................... 2 03
7 Norwalk Regional Shopping Center
Sales Potential and Space
Requirements ......................... 211
8 Department Store Space Productivity,
1928-1963 ........................... 222
9 Range of Dollar Sales Per Square Foot
of Department Store Space.............225
10 Sales Capacities Per Square Foot of
Selling Space for Proposed Shopping
Center Facilities ................... 229
11 Retail Center Classifications
and Locations......................... 232
iv
LIST OF TABLES— Continued
Page
Table
12 Composition and Character of Major
Shopping Center Types .............. 233
13 Typical Tenant Composition of a
Regional Shopping Center ............ 237
14 Department Store Sales as a Per Cent
of Total Retail Sales .............. 242
15 Distribution of the Locations of the
Most Profitable Department Stores
Among Unplanned and Planned
Shopping Centers ..................... 256
16 Population Densities in Redondo Beach
and Selected Adjacent Areas ........ 278
17 Occupational and Industrial
Distribution of the Redondo Beach
Labor Force in 1962 ................. 281
18 Per Capita Balance of Payments Index
for the Redondo Beach Economy .... 283
19 Population Growth in Redondo Beach
and its Surrounding Retail Trade Area
in the Los Angeles Region from
1940 to 1962 ......................... 284
20 Population and Income Forecasts
for Redondo Beach and South Bay Area
to 1980 ............................. 287
21 Proximity of Trade Area Segments
to Redondo Beach Regional Shopping
Center............................... 294
22 Characteristics of Redondo Beach
Regional Shopping Center ............ 295a.
v
LIST OF TABLES— Continued
Table
23
24
25
26
27
28
• 1
Redondo Beach Regional Shopping
Center Market Potential by
Trade Area Segments ................
Expected Annual Export Sales
Potential for the Redondo Beach
Regional Shopping Center in 1980 . .
Forecast of Annual Sales Potential
for Redondo Beach Regional Shopping
Center Facilities to 1980 ......
Retail Space Productivity Standards
for Redondo Beach Regional Shopping
Facilities .........................
Space-Input Requirements of Regional
Shopping Facilities in Redondo Beach
1960 and 1980 .......................
Land Allocation Requirements for
Regional Retail Facilities in
Redondo Beach 1960 and 1980 ........
Page
296
. 299
303
. 306
309
313
vi
CHAPTER I
INTRODUCTION
Purpose and Scope of the Study
Purpose.— This study has three basic aims; present;
bring together? and relate to the municipal government land
use planning process, three types of local area economic
analysis. Clear and concise presentations will be made of
economic base, trade area, and space productivity or re
quirements methods of analysis. These three analytical
approaches to municipal study will then be brought together
into a single analytical model for the study of local areas.
This approach will be related to municipal land use plan
ning analysis— specifically the study of allocating physical
space within a local jurisdiction to retail use.
Often, in municipal planning studies, an economic
base analysis is conducted by the public authority to pro
vide general economic background information for subsequent
planning decisions. Usually, in private shopping center
ventures, some sort of geographically oriented market demand
analysis is carried out in the form of a trade area study.
Frequently, in either public or private planning studies,
an analysis of the productivity or efficiency of retail
space in supplying expected levels of local expenditure is
2
ignored entirely.
In rather haphazard and inefficient manner, these
three analytical approaches have been used in conjunction
with one another on rare occasion. Seldom, if ever,
though, has full advantage been taken of the analytical
insight into space planning problems offered from a more
complete drawing together of these three types of analysis.
The purpose of this study, then, is to bring to bear on
municipal retail space allocation problems, the full ana
lytical potential deriving from a fusion of economic base,
trade area, and retail space productivity studies.
Scope.— The range of this study is limited to detail
ed discussion of the analytical approaches under considera
tion, their relationship to one another, and their com
bined relationship to solving problems of allocating re
tail space in the municipal government land use planning
process.
Little empirical analysis or verification of the
proposed approach is undertaken in this study. The scope
of the study includes discussion of suggested types of
local area economic analysis, and suggestions for ways in
which they, used in closer relationship to one another,
can usefully be applied to municipal land use planning
problems. Brief and relevant treatment is given, for back-
3
ground purposes, to a description of the local land use
planning process as it relates to retail development.
Hypothesis of the Study.— The basic hypothesis of
this study is that the macro-level information derived
from an economic base study, and the micro-level informa
tion derived from trade area and retail space productivity
studies, supply municipal land use planners with the data
that they need to construct a simple retail space alloca
tion model accounting for many determinants of local retail
land use.
The location and amount of land required for retail
or shopping center facilities in a municipality is deter
mined largely by the geographic distribution of consumer
spending units. Trade area analysis studies in great
detail the location, amount, and geographic direction of
consumer expenditure flows. Economic base analysis, on the
other hand, can provide a determination of present, and
forecast the future, aggregate levels of income and expend
iture which can be expected to occur within municipal
boundries. Finally, retail space productivity analysis
provides the technical link with which to join retail
physical space supply requirements to the anticipated
compositions and levels of consumer expenditure.
All of this leads to the hypothesis that a space
allocation model; having for its components economic base,
trade area, and retail space productivity analyses, would
provide a relatively complete explanation of municipal re
tail or shopping center space requirements.
Hypothesis to be tested.— While this is not primarily
an empirical study, some data are presented on the product
ivity of department store space, to indicate that retail
space coefficients can be derived with reasonable accuracy,
and that their future size can be anticipated.
The hypothesis implies that standards or "norms" of
retail space productivity can be established for stores
of various type and size. Depending on the level and
character of expected expenditure within a trading area,
it should be possible to determine the retail floor space
requirements to serve that demand.
The data offered in Chapter V of this study indicate
that fairly definite space requirements can be established
for serving different types and amounts of consumer expend
iture. Expenditure patterns and propensities of consumers
in a trading area may imply certain types and sizes of
retail facilities, given the space productivity performance
of different hinds of retail establishments of varying
scale. Predictable variations occur in retail space pro
ductivity with changes in store size. From historical
space productivity data, then, future space requirements
can be anticipated, assuming no radical departures from
current and recent retailing technology are taking place.
Analytical Methods Employed
Macro-demand analysis.— When a municipality is
involved in the continuous process of comprehensive
planning for local development, it must have information
pertaining to current and expected future overall levels
of local econ .nic activity. Economic base studies are
often called for by local planning agencies.
The economic base study supplies the analytical
content for what local government planners refer to as
"community economic analysis." The Committee for Economic
Development has suggested that economic base studies are
an essential starting point in the urban planning process.
The CED feels that "Knowledge about the economic base is
essential to sound public decisions in the form of master
plans, zoning ordinances, transportation plans, renewal
programs and other public improvement plants.
Economic base analysis views the community at a
macro-level, and with an assumption of full local employ
ment. All economic activity in a geographically delimited
^Committee for Economic Development, Guiding
Metropolitan Growth, (New York: Committee fer Economic
Development, August, 1960), p.31.
area of study is aggregated into two categories— basic and
local. Basic activity arises in response to demand sources
emanating from outside the study area. Local activity
serves the demands of residents with the community.
Basic activity, then, is analagous to exports in inter
national or interregional trade.
When the economic base framework is infused with
income multiplier analysis, a community income analysis
can be conducted. Future export levels can be forecast
(by any one of a number of methods) and local income levels
can be predicted through the local multiplier process.
Exports are the exogenous factor in the local model, as
investment is in the national income model. This approach
permits a portion of total income to be "determined" by
the logical processes of a mathematical model of resident
consumer behavior.
The planning insights offered by an economic base
study consist of identifying and partially explaining the
demand sources of aggregates of local economic activity.
More specifically, the investigations in an economic base
study concentrate on: (1) finding the precise sources of
employment, income, and output in the past; (2) ascertain
ing which sources have served and will be influenced by
external markets; (3) ascertaining which sources have
served local markets; (4) estimating the prospects for
future growth, change, or decline in these economic
activities; and (5) estimating what, if any, new industries
may be expected to emerge in the next decade or two.
The function of economic base studies is to provide
a general macro-economic analysis of community demand and
growth characteristics. Overall levels of demand, emana
ting both from within and without the planning area, can
be estimated and forecast from the income and employment
levels indicated by the base study. The basic-local
dichotomy of economic activities supplies an insight into
the process of community economic growth which is superior
to simple projections of the totality of local economic
activity, extrapolated from the past.
Micro-demand and supply analysis.— Once the general
levels of local income and expenditure have been deter
mined, more precise knowledge is required as to the spatial
distribution within the community (and inside the
community's retail trade area, too) of the families earning
incomes and making expenditures. Economic base analysis
says little about this, except to indicate the proportion
of income and expenditure originating outside the local
area1s boundries.
Trade area analysis views the community at a micro
level, both geographically and economically. The spending
behavior, i.e., patterns and propensities, of resident and
non-resident families making retail expenditures in the
community is analyzed in detail. Field surveys of shopp
ers and gravity, or market potential, models determine
what, and where, retail goods and services are purchased
in the community.
Trade area analysis involves specific micro-analysis
of family composition, income, and expenditure patterns
and propensities within the community, or trade area if
the two are not spatially coincident. Specific levels of
demand, in terms of dollar expenditures, can be allocated
to the various categories of retail activity in the
community.
On the supply side, this local micro-analysis deals
with a study of the productiveness with which various types
of retail space are utilized. From empirical study, it
can be observed that an efficient department store should
require x number of square feet of enclosed selling space
to efficiently supply retail goods and services to con
sumers spending y number of dollars per year in that de
partment store. The productivity of retail space is
measured in dollars of sales per square foot of selling
space per year.
An attempt is being made to determine the retail
space-input required to serve prescribed levels of final
consumer demand. The approach followed in this study is
loosely analagous to that followed in construction of an
input-output table, from which production input co
efficients can be derived for given levels of final and
intermediate demand. It should be useful to attempt to
derive constants of retail space input, i.e., retail space
coefficients, which can be applied to expected levels of
dollar expenditure for various types of retail stores.
In this manner, expected levels of dollar sales of retail
stores could be translated directly into square feet of
retail selling space required to supply these sales levels.
Retail space requirements are then accounted for by link
ing potential expenditure or sales levels to the technical
conditions of retail operation via retail space co
efficients.
Significance of the space allocation approach to
local government planning.— There are numerous
specific areas of significance that this broader approach
to local economic analysis has for municipal government
planning. Basically, however, bringing trade area and
retail space analyses together with traditional economic
base studies adds a spatial dimension to community eco
nomic analysis. The economic base study is essentially
"spaceless," in the sense that it does not precisely re
late economic activity to the amount and location of land
10
required for that activity.
Since a very important aspect of municipal govern
ment planning involves land use planning and zoning,
analysis relating directly to the economic determinants
of land use is a prerequisite for intelligent planning
decisions. The land use planning process must be capable
of economically allocating and restricting terrestrial
space within the community for retail use. Private and
social costs can be reduced by avoidance of excessive or
insufficient and poorly located commercial business zones.
If local public planning agencies have some idea of
the retail demand and supply conditions existing in their
communities, they will be in a better position to make
firm decisions on present and future land use allocations.
Local planners will be better able to relate their com
munities ' social needs with the private retail or shopping
center developers' investment criteria. It would then be
more difficult, for example, for private commercial in
terests to justify zoning variances and exceptions so
that excessive commercial development might take place
during a period of rapid residential expansion.
The proposed approach attempts to draw together
three partial analyses of economic activity, to provide a
more complete form of economic analysis on which to base
land use planning decisions. Isard says that as urban
11
analysts
...scan the list of possible types
of studies, such as regional income
studies, commodity flow studies, balance
of payments studies, economic base studies,
multiplier studies, industrial location
studies, etc., they are perplexed as to
which one to attempt. They see the virtues
and limitations of each. Often they clearly
perceive the partial character of each and
are dissatisfied...if they have the resources
to do more than one partial study, they seek
ways by which these studies may be inter
related and may be conceived and formulated
to contribute to one another.2
The proposed model of community retail space allocation
attempts to supplement the land use planning process'
analytical requirements. The significance to municipal
planners is that selected methods of urban economic
analysis are being brought to bear on practical problems
frequently encountered in making land allocation decisions.
To the extent that local planning analysis is being
improved, the aim of the study is, more precisely stated,
to construct a spatial coefficient model for retail
activity. This spatial coefficient model is ideally con
ceived as a constant which converts numbers of dollars
of potential retail expenditure, in a given combined trade
area, into numbers of square feet of selling space re
quired in retail establishments within a given municipal
ity.
^Walter Isard, Methods of Regional Analysis.
(Cambridge, Mass.: The M.I.T. Press, 1960), p.309.
Definitions
Economic terms and concepts.— A few of the frequent
ly occurring terms and concepts of the study, which are
more or less unique to urban economies, are briefly
defined here. The concepts of economic base, trade area,
and retail space productivity have been touched on above.
In the process of allocating physical space within
a community to retail use, reference to the market mech
anism indicates non-governmental allocation decisions.
The market mechanisms operating to allocate physical
space to retail use in urban areas are embodied in local
real estate and retail commodity markets.
Since virtually all local retail activity in the
United States is initiated by private individuals and
organizations, reference to retail or shopping center
developers, or development, indicates private non-govern
mental spheres of activity. Generally, "Either real
estate developers or retailer-owned real estate enter
prises are primarily responsible for the planned shopping
i O
center." Local government planning agencies are assumed
not to participate directly at all in the financing or
; ownership, i.e., development, of retail shopping centers.
i
■^William Applebaum and Saul B. Cohen, "Store
Trading Areas in a Changing Market— Part II,"
Journal of Retailing, (Fall, 1961), 34.
13
Social costs discussed in this study refer to those
arising from malallocation of the scarce land resource at
the local level. Social costs in retailing arise from
many specific causes, such as excessive amounts of store
space serving a trading area, increased traffic congestion
arising from "strip" or "string" retail development or
other poor physical relationships, and so forth. Many of
the additional costs caused by improper local retail
development are borne by retail firms as private costs.
Some of these, undoubtedly, are not passed on to the
consumer. However, the community suffers from a generally
poor quality, and either surfeit or insufficiency of
retail goods and services.
Planning terms and concepts.— The planning terms
and concepts defined here are those unique to urban or
municipal government planning.
In practice, the local planning dealt with in this
study occurs at the municipal, or sub-metropolitan, level
of government. As a matter of fact, there are few, if
any, metropolitan legal jurisdictions in the United States
with a wide range of planning authority. Special dis
tricts, e.g., mass transit, may exist, but these cannot
be considered as general city planning bodies. The most
common decision making unit having broad land use planning
14
powers is the municipality.
The specific concern of this study is with the local
land use planning and zoning process at the municipal
level of government. Local public planning authorities
plan for and restrict the use to which land may be put in
various areas of a community. In the retail sphere, these
planning authorities, under normal circumstances, never
actually initiate and bring to fruition shopping center
developments. Municipal land use planners are concerned
with anticipating the temporal and spatial sequences of
retail center development in their communities; and then
regulating these privately owned and operated facilities
in such a manner that their timing, location, and site
layout all conform to community needs and desires.
This kind of local planning process provides an
investment infra-structure for a local area, within which
private investment decisions can be made in conformance
with the public interest. Public planning authorities,
of course, must be aware of present and future demands on
retail facilities in their communities, and informed as
to the investment criteria and technical conditions of
retail activity.
Plan of Study
Economic analysis and land use planning.— The study
15
will first attempt to explain in detail the planning and
market processes to which the economic analysis of the
synthetic retail space allocation model is related.
Chapter II discusses the land use planning process,
and retail and land market structures at a local level.
An attempt is made to discern the analytical requirements
of the land use planning process from a discussion of some
of the problems encountered in economically allocating
land to retail use.
The operation and inadequacies of private market
mechanisms in economically allocating physical space to
retail use is also analyzed in Chapter II. Public plann
ing supplements to the private decision making process
are offered.
Finally, there is discussion of the significance
that economic retail space allocation has to the reduction
of social cost in a community.
Approaches to local economic analysis.— There are
three main analytical elements which are brought together
in the retail space allocation model of this study.
Detailed historical and critical treatment of each type
of analysis is presented in Chapters III, IV, and V, for,
respectively, economic base, trade area, and retail space
requirements analysis.
16
Chapter III consists of a complete and separate
historic review of the development of economic base theory
and analysis, evaluation of economic base analysis, and
treatment of the limitations of the base approach.
Chapters IV and V are developed along similar lines for
trade area and space requirements analysis. At this stage
in the study, no special effort is made to specifically
relate these analytical approaches to one another, unless
comparisons arise naturally in the course of presentation.
Each analytical approach is discussed, evaluated, and
criticized largely on the basis of its own merits and
deficiencies.
A more comprehensive approach.— In Chapter VI an
attempt is made to draw economic base, trade area, and
retail space productivity studies together.
Initially, the analytical requirements of local
planning authorities are established in greater detail
than in Chapter II. Next, a formal relating of the
approaches, or construction of the model, is carried out.
Finally, the data requirements of the model are discussed.
The efficient generation of both primary (or field) and
secondary data is examined.
In sum, Chapter VI forms the core of this study.
The relevant types of analysis are brought together and
17
applied, in a hypothetical situation, to a shopping center
planning problem.
Evaluation, summary, and conclusions.— Chapter VII
relates the model to the problem of reducing or avoiding
social cost in retail development. The model’s relation
ship to the municipal land use planning process is also
discussed.
Chapter VIII summarizes the findings of the study,
and offers its broad conclusions.
CHAPTER II
LAND USE PLANNING, THE MARKET MECHANISM
AND SOCIAL COST
Local Government Planning and Land Use
Objectives of land use planning.— This chapter is
intended to provide a brief background revealing outlines
of some of the basic problems and purposes involved in
contemporary municipal land use planning. The economic
determinants and models relating to retail space allocation
discussed and developed later in this study provide ana
lytical tools useful to land use planning. Since the
present effort is directed toward relating these allocative
mechanisms to local planning needs, the planning concepts
with which they are associated and the problems toward
which they can be directed must be discussed.
I
What are the basic objectives of local planning for
community land use? This is not an easy question to
answer, for a number of reasons. Often, objectives are
vaguely stated, or, when clearly expressed, urban planners
may still disagree on objectives. This problem stems part
ly from the fact that urban planning draws from numerous
social and physical scientific disciplines for its know
ledge and methods. On the other hand, urban economics is
19
in a considerably more favorable situation. It is working
within the confines of a unified discipline and can draw
from a relatively large body of established theory and
knowledge for its analytical needs.
Fortunately, there is no strict dichotomy between
urban economics and urban planning. Many economists write
about land planning problems in terms of economic object
ives and criteria. These economic objectives and criteria
are widely accepted as constituting factors of overriding
importance in the land use planning process. This offer
ing in planning literature by economists, and its nodding
acceptance on the part of planners, is not, lamentably,
indication of intimate understanding by economists of the
specific analytical needs of planners, any more than it is
indicative of an appreciation by urban planners of the
analytical value of portions of that offering.
According to neo-classical economic theory, the
ideal allocation of land resources within a community
would result from the operation of a perfectly competitive
market mechanism. According to the market criterion, the
basic objective of land use planning would be to attain
the same land use pattern resulting from the natural
processes of the urban real estate market under perfect
competition. A perfect land market would yield a pattern
of land uses in a community resulting in a minimum aggre
20
gate of land value. Good planning, through increasing
access and inducing population mobility, diminishes the
scarcity value of prime locations and reduces the rent
which they can command.^ From this viewpoint the role of
the land use planning process is essentially limited to
helping perfect the operation of the market mechanism.
If the city is viewed as an economic and social
mechanism, demand for urban land uses is seen as arising
from the need for space for concentrated socio-economic
activities. The spatial pattern of functional areas with
in the community; the street system, the buildings, the
people, etc., can all be viewed as constituting items of
economic equipment. The physical and ecological forms of
the community are then products of the demand and supply
forces working within the framework of the real estate
market and conditioned by economic, social, and legal
institutions. The spatial distribution of various land
uses evolves from functions performed through the process
of competition for preferred locations among these uses
in the land market. Community land use planning might
then begin by forecasting the pattern of land uses evolv
ing from the unimpeded social forces operating within the
market framework. Under this approach much of the plan
^Charles M. Haar (ed.), Land Use Planning,
(Boston: Little Brown & Company, 1959), p.76.
21
itself would merely express the unmodified result of these
forces. The planning process would be left to determine
what (if any) modifications of this market growth pattern
are socially warranted.
While economic efficiency is a basic goal of land
use planning in achieving an ideal community land use
pattern, it is not always obtainable through unfettered
operation of the market's space allocative mechanisms.
More generally, it can be said that a basic objective
of planning is rational action. Planning is a means of
reducing waste or producing the greatest return from the
employment of resources. Efficient utilization of re
sources (in a democratic context) would be that which most
closely satisfied the particular preferences of individuals
within a community.
Efficiency must be measured in terms of the purpose
it serves. One task of urban planning is to provide in
formation to decision makers (public and private) and
identify the best of all existing alternatives in terms
of a consensus on goals. The social interests of the
community may sometimes be in conflict with the aims of
private activity operating in response to market stimulii
or suppresants. Haar states that "Since land-use policy
in the United States is always an uneasy marriage between
the market mechanism and the dictates of public policy,
22
many of the conflicts over the ordering of land uses stems
from such differing outlooks."2
As seen in the context of this study,
The changing retail structure confronts the
planner with his eternal problem: is the
emerging natural pattern the most economic,
or the most satisfactory solution for all
the citizens, or is there a better alterna
tive? How to balance reality with the
need for land-use control?^
The objective of the urban land use planning process is to
rationalize retail development occuring in the market
place with community goals of land use and development.
Certain social constraints should be placed on local re
tail activity's use of land. Any public allocations of
land to retail use, however, must be made in light of
realistic appraisal of local economic conditions now and
in the future.
The economics of land use planning.— Let us define
community planning, as does Benjamin Higgens, "...as the
analysis of land use in its relation to the human satis
factions to be obtained from land use,...and the alloca
tion of land among various uses in such a way as to
2Ibid.. p.75.
3James Simmons, The Changing Pattern of Retail
Location. (Chicago: University of Chicago Press, 1964),
p.169.
23
maximize satisfaction of the community as a whole."4
Community planning proposals then are made for the purpose
of increasing the community's welfare and particular
proposals made would be those most likely, as Nathaniel
Lichfield says, "...to yield the largest attainable
combination of net benefits, as compared with the costs
of attaining them."5
If, attempting to achieve greater community welfare,
local planners program development of physical facilities
without relating them to community economic development,
the results will be illusory. It is equally illusory to
attempt general community planning while neglecting to
establish economic objectives. Among specific economic
objectives of community planning are: minimization of
travel times; attainment of optimum population and build
ing densities; and maximization of the net social product
of land use through the optimum allocation of land among
residential, commercial, and industrial uses.
Even though economic forces may be of overriding
importance to community planners as they attempt to achieve
these ends, this does not necessarily imply that their role
^Benjamin Higgens, "Towards a Science of Community
Planning," Journal of the American Institute of Planners,
(Fall, 1949), 6.
^Nathaniel Lichfield, "Cost-Benefit Analysis in
City Planning," Journal of the American Institute of
Planners, (November, 1960), 273.
is passive and that they are merely data gatherers and
information dispensers for those making decisions within
the private market framework- A belief in the possibility
of effective land uee planning rests on the assumption
that man controls his environment, either by affecting the
rate and direction c?£ ongoing processes or by initiating
such processes, planning is often relied on to gain such
control. This means that i0cal planning should be viewed
as a process of whici* perfecting the market allocative
mechanism in the latter's determination of local land uses
is only one of several alternatives for attaining the goal
of increased community welfare.
For local planning to t>ecome sufficiently rich in
alternatives more should be icnown about the nature of the
urban economy and hoV it can manipulated. How does it
change? How can thiss change be controlled and directed?
The greater importance of economic model building in urban
planning stems from its role as a substitute for experiment
in the initial stages of plan development and as a guide
to experiment in the latter stages. A model of urban
economic functioning is n° mofe than a scientific hypothe
sis as to how cities operate and grow. An economic model
for projection of land, use space requirements should be
able to make alternative assumptions in the realm of taste,
technology and public policy.
25
Retail space allocative mechanisms of existing
models of the urban economy may be more helpful to planners
in the rational attainment of goals than would limited
attempts at perfection of the real estate market mechanism.
Urban planners should at least be aware of the fact that
urban economics has yielded models containing allocative
devices which can approximately determine space require
ments for different land uses in accordance with criteria
of economic efficiency, and within the framework of social
and other non-economic policy goals.
Urban planning has given explicit recognition to the
problem (and sometimes dilemma) of resolving questions of
economics while in the process of attaining overall plann
ing goals of an essentially non-economic nature. A
prominent author expresses her view in the statement that
The fundamental question in all planning is
not so much the development of a Utopian
community with all of the advantages of our
technological age as the striking of a
balance between what is economically possible
and socially desirable. Any skillful and
imaginative planner may be able to devise
an ideal plan, but its translation into
reality must depend upon the capacity of
the community to meet the cost and avoid
economic losses from the dislocation of
community activity. Fundamentally planning
and carrying out of a plan must be resolved
by answers to questions of economics...The
budgeting of values to be realized should
be related to the actual and potential
economic resources which the community can
reasonably command...6
^Carol Aronovici, Community.Building, (Garden Citv
Doubleday & Co. , inc.; 1956), p.63. y
26
If a community land use plan is to contain legally enforce
able controls on the degree and type of economic activity
in various areas of a community, then local planning must
have at its disposal analytical techniques permitting it
to determine efficient levels of space needs for each line
of retail and other activity in each area of a community.
Implementation, as Higgens notes, requires that the
objectives of community planning be reduced to measurable
quantities so that they can provide empirical tests of
the extent of improvement sought through planning and the
extent to which particular policies bring actual conditions
closer to desired conditions. To answer economic ques
tions, municipal planning must know more about the means
of allocating scarce land resources efficiently.
Allocation of land space to specific use.— Land use
planning is concerned with the future space needs of the
community and, says F. Stuart Chapin, as a
...summary of future land use requirements...
describes requirements of both a qualitative
and quantitative character. In the former
category is a statement of principles setting
forth relationships to be observed in setting
aside areas for various land uses....In the
second category is a statement of space
requirements for each class of useTv
^F. Stuart Chapin, Urban Land Use Planning,
(New York: Harper & Brothers, 1957), p.282.
27
The analytical discussion of this study is concerned
specifically, and almost exclusively with Chapin's second
or quantitative category of land use planning. It does
not deal as such with the problem of physical location,
even though approximate areas within which retail facili
ties should be located may be implied by the analysis.
Roland Artie, an urban economist, emphasizes the
positive aspects and potentials of planning for an urban
area through the use of decision models. The passive
attitude of many urban planners, who merely forecast local
economic activity with the implication that very little
can be done to influence the course of events, is a result,
Artie feels, of their failure to realize the potential of
economic models.
As an example, Artie postulates a decision model in
general form as: y = f(s, u, x); where y_ is a set of out
come variables, _s is a set of instrumental variables, u is
a set of stochastic (random) influences, x is a set of
exogenous variables, and f is a set of rules describing
correspondence between predetermined variables and de
pendent outcome variables. In other words, _f explains the
ways in which outcomes are linked to instrumental and
exogenous variables. Structural relations can be deter
mined through empirical research to quantify the linkages
28
between inputs and outputs.® Later phases of this study
are pointed toward showing how economic models similar to
the type postulated by Artie can be useful in allocating
relatively precise amounts of space in a local area to
retail use.
Land Use Planning and the Market Mechanism
Market allocation processes.— Land is selected and
purchased or set aside for retail and other commercial uses
on the basis of present and future sales potential. The
market conditions of both retail trade and urban land
transfers have relevance to a discussion of the allocative
processes by which physical space in a community is
secured for retail activity.
In urban real estate markets demand for retail sites
derives from demand for completed development by retail
developers. Bidding from a variety of private developers
interested in particular sites constitutes the demand side
of the market. The opportunity cost of a site to the
entrepreneur (private or public) is its value if put to
that alternative use which would earn the maximum yield.
Developers must consider what price they are willing to
pay, given the return that they can expect. Price will
®Roland Artie, "Public Policy and the Space
Economy of the City", Cities and Space, (ed.) Lowdon Wingo
(Baltimore: Johns Hopkins Press, 1963), pp.161,162.
29
usually be no higher than that required given the
competition for land and the expected return.
The supply side of urban land markets is character
ized by a largely fixed amount of land available for uses
of all sorts. Land can be fundamentally divided into
rural and urban uses according to demand. As demand for
sites increases, some land formerly undeveloped or under
developed becomes available as sites for various uses.
Institutional and legal constraints such as zoning and
planning ordinances result in a more rigidly fixed supply
of sites available for particular uses in highly urbanized
areas.
Lichfield conveniently lists a number of factors
characterizing the urban land market: (1) this market is
made up of a generality of owners who wish to sell or lease
and purchasers who wish to acquire; (2) sales require a
relatively longer period of time to complete than in other
markets; (3) this market is spatially localized in that
demand is satisfied at a particular site or area; (4) there
are few transactions relative to the amount of property
in existence; (5) details of transactions are hard to
come by; (6) there is no readily available index of
current or past prices; (7) there is no central body in
fluencing price levels; (8) many purchases are made for
non-economic reasons; (9) economic motives of purchasers
30
differ— investment, speculation, development, etc.; and,
finally, (10) sites change hands in this market as a
result of the supply and demand for them.®
This list of factors indicates that markets for
urban development sites are much less than pure and
perfect. The typical secrecy about the details of trans
actions and the localized nature of the commodity itself
result in a dearth of market information. For purposes
of this analysis, purchaser's motives are assumed to be
rational, economic, and concerned primarily with site
development for shopping center and other retail use.
Retail market structure.— Economists differ in their
descriptions of the structure of retail trade markets.
While all seem to agree that they do not conform to the
criteria of perfect competition some, such as Melvin
Greenhut, describe them as oligopolistic, and others, such
as Jane Aubert-Krier, refer to them as imperfectly com
petitive. While Greenhut points out that some retail
markets do have competitive processes of an oligopolistic
nature, most authors feel that retail markets conform to
a type of market containing spatially (and otherwise)
differentiated products and services, which are sold under
^Nathaniel Lichfield, The Economics of Planned
Development, (London: Estates Gazette, 1956), pp.21-27.
31
conditions approximating those set forth by Chamberlin
for monopolistic competition. The writer feels, except
for specified cases, that retail markets for shopping
centers can generally be referred to as monopolistically
or imperfectly competitive.
Greenhut's theoretical analysis leads him to
state that
An oligopolistic spatial market is the case...
in the retailing fields in which consumers are
invariably located in a circular area around
the individual retailing shops. The limit to
the retailer's sales radius is not so much the
dollar-and-cent freight cost to the consumer
as it is the transport burden (mainly time-
convenience) to him.10
In this type of market, encroachment on the trading area
by competitors through price competition is sporadic,i.e.,
not usual. The relationship between individual retail
enterprises is indicative of heterogeneous oligopoly with
a live and let live price system. Since the retail trade
market is spatial it does not conform to the requirements
of non-spatial market types in traditional economic ana
lysis, and a chainlike or zone based price policy may be
induced by space.
For Aubert-Krier, retail firms operate under condi
tions of monopolistic competition in which equilibrium
Greenhut, .Microeconomics and the Space
Economy, (Palo Alto: Scott Foresman and Co., 1963),
pp. 287,288.
32
involves firms (even without monopoly profits) being more
numerous and smaller than under pure and perfect competi
tion and operating with excess capacity. Equilibrium
from free entry does not then necessarily secure optimum
conditions for consumers even when transport cost savings
accruing from dispersion of selling places are accounted
for. Aubert-Krier divides the market for retail goods
into two sub-markets; one for convenience goods which are
purchased incidentally during trips for other purposes or
»
without much selectivity, and one for shopper's goods
which are purchased with more discrimination. In the
former sub-market, ease of entry creates an attenuated
competitive atmosphere in which the density of retail
facilities must, during some time period, reach a maxi
mum not permitting new retailers to enter the market.
At that point, potential retailers will hesitate to
establish themselves in a trading area where they will
have to wrest away from their competitors a share of the
existing market. In the latter sub-market, the variety
and variation of needs and imperfections permit newcomers
to establish themselves more readily, while creating for
them illusions about potential profitability. This situa
tion gives rise to lively competition from new entry.
11Jane Aubert-Krier, "Monopolistic and Imperfect
Competition in Retail Trade," Monopoly and Competition
and their Regulation, (ed) Edward H. Chamberlin
(London: MacMillan & Co..Ltd., 1954), p.293.
33
The markets for urban land and the products of retail
trade are infused with imperfections arising from their
spatially localized nature. More detailed analysis of
some of the reasons for these monopolistic elements will
be made in subsequent chapters of the study. At this
point, it should be noted that land use planning would
be of no value in an environment where an open, fully
competitive market, operated perfectly. Before discussing
the ways in which the public planning process can supplant,
suppress, or support the private market efficiency and the
manner in which the market process might fail to attain
an economically efficient or otherwise satisfactory
allocation of resources will be examined.
Market efficiency and welfare.— The connection
between market efficiency and community welfare is ex
plained by Francis Bator in the following manner:
The central theorem of modern welfare economics,
the so-called duality theorem, asserts a corres
pondence between Pareto efficiency and market
performance.... Correspondence between Pareto-
efficiency and market performance implies,
at the least, that decentralized decisions
in response to..."prices" by atomistic
profit-and satisfaction-maximizers sustain
just that constellation of inputs, outputs
and commodity-distribution, that the maximum
of the specified social welfare function
calls for. It implies, in other words,
that decentralized market calculations
correctly account for all "economic"
costs and benefits to which the relevant
W-function is sensitive.12
Paretian efficiency requires that utility combinations
are maximized with given preference functions, resource
endowments and technology. A community can be said to be
on its Paretian frontier if it is impossible to make any
one better off (in terms of his own ordinal preference
function) without making someone else worse off. Associa
ted with the utility possibility frontier, in turn, is a
production possibility frontier denoting maximal alterna
tive output combinations.
The prices established in competitive markets provide
private retail developers with guides established by the
wishes of consumers, who, in turn, face prices reflecting
the costs of production. It is through prices that the
market mechanism brings the tastes of consumers and the
technical conditions of production together and produces
an allocation of resources. Given the wishes of consumers
(or preferences of the community), an efficient allocation
of resources implies, under perfectly competitive condi
tions, that outputs are produced with a minimum use of
resources, i.e., at least cost, and that marginal cost
equals price. As a result of these competitive conditions,
Bator, "The Anatomy of Market Failure,"
Quarterly Journal of Economics, (August, 195a), 353.
private developers making the highest bids for particular
land sites would secure the right to develop them. It is
market bids, if permitted to operate, which primarily
determine the uses to which sites are put. Free operation
of a perfectly competitive market would generate a land
use pattern reflecting the "highest and best" use of each
parcel, yielding the largest net return to owners in terms
of money and amenities over a specified period of time.
The uses which come about as a result of competitive bids
for alternative uses will yield the closest approximation
of the preferences of the community.
Market failure in the allocation of urban land.—
How can the market mechanism fail to minimize resource
costs and maximize community welfare? In a very general
sense, Bator tells us that in allocation theory market
failure means, "...the failure of a more or less ideal
ized system of price-market institutions to sustain
'desirable' activities or to estop 'undesirable'
activities."13 There are numerous reasons in the real
world why a Pareto-efficient allocation of resources
might be violated— imperfect competition, inertia and
resistance to change, businessmen's desire for a "quiet
life," uncertainty and inconsistent expectations, etc.
13Ibid., p.351.
36
The condition our analysis is most concerned with
is that which Bator terms failure by structure, where
prices are determined by market forces, they will not
conform to a Pareto welfare maximum unless pure and perfect
competition prevail in all markets. If some markets are
saturated by a few efficient firms, the full welfare-maxi
mizing solution of inputs, outputs, and prices will not be
sustained.
Urban real estate markets often fail to bring about
changes in land use as a result of market bids. Sometimes
it is because speculators, rather than developers, bid on
undeveloped sites for other than development purposes.
Sometimes in the re-use of urban land, property owners
refuse to sell at prices appropriate to redevelopment if
these are less than the value of the property in its
current use, or if they are speculating themselves by
waiting for a higher bid from a more profitable redevelop
ment scheme. Often, mere ignorance of the real value of
a site by either owner or would-be developer is sufficient
to prevent its being used in a rational manner.
Analysis of the market mechanism's failure to attain
proper changes in land use has occurred in connection with
the federal urban renewal program. Both A.H. Schaaf-^ and
l^A.H. Schaaf, "Public Policies in Urban Renewal:
"An Economic Analysis of Justifications and Effects, "
Land Economics, (February, 1964), 67-78.
37
Otto Davis-1-5 have made cogent presentations of the problem.
These two authors discuss the way in which "externalities"
impair proper functioning of the market mechanism. Ex
ternal diseconomies in a blighted area reduce the income
potential for re-uses, so external changes from renewal,
i.e., reduction of diseconomies and increasing of eco
nomies, result in a new set of environmental linkages and
re-use values.
External diseconomies in the form of poor public
and private services and improvements around a particular
parcel remove incentive for the owner to improve it (the
same amount of dollars invested in a "nicer" section of
town would yield a higher return). As Davis says, "The
removal of externalities causes social and private products
to be equated and, insofar as renewal eliminates external
ities, the urban property market is moved toward the Pareto
optimum."1^ Lichfield makes the point more broadly when
he says that:
The market should not always be allowed to
operate freely, and be permitted to establish
uses by the highest bidder, because of the
divergence of social and private costs...
Planning must intervene, therefore, to minimize
possible social costs...and...to supplement market
forces if the best use of land is to be achieved.17
l^otto A. Davis, "A Pure Theory of Urban Renewal,"
Land Economics, (May, 1960), 220-226.
16Ibid.. p.225.
^Lichfield, Economics of Planned Development, 310.
38
As discussed above, the market mechanism may fail to
yield an optimum resource allocation because of structural
deficiencies. It may also fail to achieve an allocation
of resources that minimizes social costs. Even a purely
and perfectly competitive market may require governmental
planning supplementation to minimize social costs.
Retail competition and land use.— The conditions of
competition in retail trade may often lead to a rnalalloca-
tion of resources (including land space) in this sector of
a local economy. The absolute amount of consumer expend
iture for any segment of retailing is relatively fixed
within any given trade area by the parameters and consumer
expenditure pattern of that trade area. Retail facilities
are established on the basis of both existing potential
and future growth anticipations. Retail facilities are
often introduced ahead of the market potential required for
their support. Marketing analysis indicates that a typical
situation is one where all retailers anticipate growth and
all retailers establish facilities in the growth area.
This pattern of competition frequently results in permanent
over-saturation prior to maturity of the growth area. One
marketing analyst describes this situation as "market
myopia." While suffering from this malady, retailers may
feel that everyone else is building too many facilities in
the area but that they are just keeping up with their share
39
of the market. When this opinion is held by all (or most)
competitors in a trade area, over-saturation usually
*
results.
In retailing, as noted above, any given geographic
market contains a relatively fixed amount of market poten
tial in terms of consumer expenditures. The resources
required to secure shares of retail markets display wide
fluctuations in different areas. While these fluctuations
could be caused by other factors, market researchers feel
they.are primarily the result of intensity of competition
for available market potential.18 As the radius of a
business center's trade area increases, because of more
rapid transportation, etc., activities in the center
become increasingly more differentiated as it attempts to
meet new and increased demands. The point of market
saturation approaches, the migration of retail facilities
to nearby areas becomes expedient. To avoid private and
social costs of over-expansion under the private market
mechanism, multiple objective long-range planning for an
entire area should supplement separate income and expend
iture forecasts of individual firms.
Public planning and land use.— Planning authorities
l®William D. Stevens, "The Social Responsibilities
of Marketing, ’ ! Proceedings of the American Marketing
Association. (1961), 572.
are concerned with looking as far as possible into the
future, while private developers are usually concerned
only with their own immediate future. The interest
horizon of local government planning is the entire plann
ing area, while that of private retail developers may be
only one or a few sites in the same locality. Developers
may lose some things under planning but they may gain
other things. They can plan their specific projects
within the framework of a local planning process. Since
total demand for all kinds of retail development in a
trade area is derived from individual and collective ex
penditures made by consumers living both inside and out
side a community, this total spending will place bounds
on the amount of commercial development of all types
which can occur in a community. Even though it is not
practical or useful to estimate demand for the totality
of heterogeneous development in a community, it certainly
is feasible to make checks on the separate totals of
segments of that overall demand. Estimates can be made,
for example, of the number of retail stores of various
types required, on the basis of the expected future
number of families and their composition, and their future
expected incomes and expenditure patterns.
The writer's opinion coincides with that of John
Dyckman, namely, that "Planning is itself an alternative
mechanism, which may be invoked either in conjunction
with or as an alternative to other social mechanisms.
Municipal government planning should not trust its alloca
tion decisions to the hopefully unimpaired operation of
diffused automatic market forces alone but, instead, base
them in part on the operation of economic models con
sciously constructed to explain determinants of local
land use and retail development. Government planning can
supplement private market land use allocations made by
the price system.
There is an idea, running through current urban
planning literature, that the planning process should
provide a guiding hand or framework for private invest
ment and development decisions. Lichfield feels that
planning agencies can affect the costs and benefits of
both private and public developers through providing in
formation about the area being planned— the activities
and intentions of other developers, insights into po
tential markets, and so forth.
Costs that are otherwise indirect can be made direct
as municipal planning strives to reduce the divergence
between private and social costs. Planning can increase
direct private costs by increasing standards, or decrease
l^John w. Dyckman, “Planning and Decision Theory,"
Journal of the American Institute of Planners,
(November, 1961), 335.
42
direct private benefits by decreasing density.
Under normal conditions, the private sector accounts
only for the costs it must bear and the benefits it can
appropriate; but not for the diffused costs suffered by
others which it need not compensate, nor benefits enjoyed
by others for which it cannot exact payment. A planning
agency acting as a supra-investment body, redistributing
social costs and benefits so as to increase aggregate
community welfare, would require adequate analysis for
both a welfare test and a rational resource allocation.
For the former, Lichfield suggests that planners adapt
cost-benefit analysis to their needs. This type of
analysis uses market criteria on which to base judgements.
For the latter, this study suggests a combination of eco
nomic base, trade area, and space productivity analysis.
These planning allocative mechanisms can supplement the
private market mechanism.
Resource Allocation and Social Cost in the Urban Economy
Measurement and evaluation of social costs and
benefits.— The analysis in this study concentrates
primarily on the problem of allocating scarce land re
sources (i.e., those required for the production of retail
goods and services) in the urban economy in such a way that
community welfare is maximized efficiently. This implies
that, in addition to firms .operating in a least cost
43
manner, there be as great a reduction as possible of
social costs arising from private and public activities,
and a more equitable distribution of remaining social costs
and benefits. Since the problems of social cost and
benefit are important to urban planning, these concepts,
their measurement, and the relationship to them of the
planning space allocation mechanism developed later in
this study must be considered.
The traditional view of social costs and benefits is
characterized by the assumption that private returns and
outlays constitute a theoretically adequate measure of the
total benefits and costs of productive activity. However,
X.W. Kapp tells us that
...the treatment of social costs as a
minor and exceptional disturbance rather
than as a characteristic phenomenon of
the market economy reflects merely the
very imperfect way in which these costs
are taken into consideration in the
present system of economic calculation
of costs and returns.20
Social costs can be defined as all direct and in
direct losses suffered by third parties or the general
public as the result of economic activities. Social bene
fits are all direct and indirect gains enjoyed by third
parties or the general public from economic activities.
20K.W. Kapp, The Social Costs of Private Enterprise.
(Cambridge: Harvard University Press, 1950), p.9.
44
For purposes of analysis, these concepts are not necessar
ily quantifiable. They serve their purpose, however, if
they aid in tracing and revealing a substantial portion
of the social losses and gains generated by individual
producers.
The magnitude of social costs and gains is ultimately
a matter of social evaluation. The magnitude of social
costs, for example, depends on the importance which the
community attributes to both the tangible and intangible
values involved. And, as Kapp reminds us, "Whether or not
a system of economic planning would avoid the social costs
of production depends upon whether the planners wish to
avoid or neglect them."2^ -
Current planning literature not only indicates that
planners are aware of social costs and benefits but that
they are actively engaged in seeking ways to measure and
account for them.22 Frequent mention is made of the need
for a welfare test in community planning because the
planning process is so subjective that it is often
difficult to move rationally to the attainment of goals.
A welfare test would analyze the implications of
benefits accruing to a community and the cost (in re
sources) at which they would be bought. Such analysis
2- * -Ibid., p. 24.
22Lichfield, Journal of the American Institute of
Planners, (November, 1960), 273-279.
45
would test proposals and plans, at various stages in the
planning process, for their likely effects on community
welfare. Comparison of planning alternatives could be
made at various stages of plan development, or the case
for and against planning itself could be tested.
Lichfield suggests that a planning balance sheet be
drawn up for a community. It would list benefits or gains
on one side and the offsetting costs or losses on the
other, thus reducing the chance of arbitrarily ignoring
the costs associated with a particular set of benefits and
vice versa. Items in the balance sheet should be sharply
defined and expressed as quantities to the greatest extent
possible. Where practicable, losses and benefits should
be measured in monetary terms to facilitate comparison.
Urban planning can force private developers to
absorb certain social costs with a resultant rise in
private costs, which will either be met by the land develop
er or the consumer, depending on market conditions. But
proper land use planning should simultaneously aim at a
reduction in private as well as social costs, rather than
mere absorption of the latter into the former. Lichfield's
intention is that "...land planning aims at a reduction in
costs, both private and social, and at an apportionment
between private and social costs which is in accord with
current social conscience."23
0 3 " — ------------------------------------------------------------------------- — — - ■■ ■
*JLichfield, The Economics of Planned Development. 277.
46
Benefit-cost analysis as a tool to evaluate land
use decisions.— As some of the above terminology
might indicate, the current effort in urban planning to
measure community welfare borrows heavily from the benefit-
cost analysis used to evaluate large multi-purpose govern
ment programs and projects, the products and services of
which are, to a large extent, not priced in the market.
Otto Eckstein explains the purpose of this analysis in the
following manner:
Benefit-cost analysis...is a method of
approximating the results of consumer
voting in the market place by means of
computations which use the prices that
consumers establish when they choose
marketable commodities.24
Even though community planners may often deal with
marketable goods and services (and not collective goods
and services provided directly by a public authority), the
framework of analysis provided by benefit-cost type studies
can be very useful in evaluating the effects of discrete
planning decisions. For it is not strictly at the margin
that local planning authorities allocate land space for
specific uses within the framework of a community general
plan. This kind of allocative decision involves the
equivalent of blocks of marginal decision units, and is
^Otto Eckstein, Water-Resource Development.
(Cambridge: Harvard University Press, 1958), p.39.
47
not comparable to the decision of an individual developer
to appropriate one or a few parcels of land for a particu
lar use out of the total land area of a community in which
there are no land use restrictions.
The intention of land use planning ought to be to
maximize the social product of each type of land use. The
net social product of a land use is its contribution to
community satisfaction. Maximum net social product is
achieved when marginal cost equals price for each product
of land use. If benefits are equivalent to the value of a
commodity to consumers and their willingness to pay for it,
it may be assumed that benefit equals price. If costs are
equivalent to costs of production plus the monetary value
assigned to any social costs, then it can be said that the
condition for maximizing community satisfaction is that
marginal cost equal marginal benefit.
As Higgens points out, local planners cannot achieve
these conditions because they do not control the necessary
economic data, i.e., level of employment, market structure,
etc. However, planners can determine what distribution of
land among different uses will come closest to maximizing
community satisfaction over the foreseeable future, by
comparing the costs of various types of use with the amounts
that the community is individually or collectively willing
to pay for the use.
48
The aim is to give the community the type
of land use it wants. These wants are
expressed in the amount of money they
are willing to pay, and will be willing
to pay in the future,...25
Lichfield points out that in municipal planning,
the measurement of costs and benefits, is not absolutely
necessary. Their precise description, with an indication
of incidence, would of itself aid decision making, because
it would provide a framework within which relevant argu
ments for and against a planning proposal could be organ
ized.
Benefit-cost studies are a method of evaluating
particular resource allocation proposals, such as those
for the allocation of land space to retail use, with re
gard to their welfare implications for the community.
Economic base and trade area studies can supply detailed
estimates of existing and future potential consumer demand
(in terms of monetary expenditures) for retail facilities
within a particular community or other planning area.
This present and future estimated demand can then be
linked, through retail space coefficients, to the corres
ponding space requirements in terms of retail facilities.
The controlled rates of retail development that could be
achieved by use of more precise estimates of present and
^Higgens, Journal of the American Institute of
Planners, (Fall, 1949), 12.
49
future retail demands, and their translation into land
resource requirements, would permit avoidance of many un
necessary private and social costs presently incurred by
private retail activity and a community as a whole.
There seems to exist little doubt among observers,
economists and urban planners alike, that unnecessary
social cost and economic waste arise in retailing as the
result of the free and unabridged operation of market
forces. One planner says that the excessive allocation
of land to business use poses a most serious menace to
residential property values. A too extensive zoning of
business districts tends to create an unwarranted antici
pation of business expansion preceding any rational
evidence of possible expansion. Such excessive retail
zoning tends to reduce land values because the supply of
retail space exceeds the demand.
Running counter to the above problem, a prominent
retail location analyst notes that
Modern zoning regulations...have tried to
establish some relationship between the
market for retail facilities in the
community and the amount of land which is
zoned for that purpose... The trend toward
reduction of the extensive retail zoning
in some areas results from an appreciation
that, if much of the land were actually
built on, ruinous competition would result
among retailers, and the community as a
whole would suffer.2^
26Richard L. Nelson, The Selection of Retail Locations.
(New York: F.W. Dodge Corp., 1958), p.316.
50
As will be demonstrated later in the study, the planning
mechanism can allocate land resources to match retail needs
with adequate precision.
Because elements of spatial monopoly play an im
portant role in the emergence and maintenance of duplica
tion and excess capacity in retailing, the opening of new
retail facilities where existing firms are sufficient has
the effect of forcing both the old and new retailers to
operate below capacity without being able to reduce their
fixed costs. The result is a continuous effort to shift
these costs to consumers as higher prices.
Since these monopolistic elements are inherent and
ubiquitous in retailing enterprise from a policy view, it
might be more advantageous to consumers, and more socially
efficient, to base investment and physical location de
cisions in retailing on a pragmatic concept of social
optimum, which in turn is based on technical criteria
accounting for long-run perspectives and objective
standards of capacity.
It seems obvious then, as Simmons says, that
The complex inter-relationships between the
retail structure and the whole urban system
require constant study to provide parameters
of interaction, together with accurate
forecasts of population and income distri
bution. Flexibility must be an important
requirement for the planning of a retail
51
needs since so many important factors—
innovation, consumer values, and
technological change are virtually
unpredictable.27
It is hoped by the writer that the combined analysis offer
ed in this study will supply a practical tool with which
urban land use planning can anticipate some of these
dynamic factors.
27Simmons, The Changing Pattern p.170.
CHAPTER III
ECONOMIC BASE ANALYSIS
This chapter is intended to provide an overview of
the development, content, and deficiencies of economic
base and income multiplier analysis. Economic base studies
constitute the macro-demand analytical component of the
retail space allocation model presented in Chapter VI.
Historic Review
The economic base concept.— The theory of the urban
economic base, as presently defined, had its inception in
the early 1920's. The economic base of a specifically
delimited geographic area consists of all the economic
activities of that area which arise from a demand that is
generated outside the boundries of the local area. All
economic activities in a geographically defined area that
serve a demand emanating from within the boundries of the
area, are nonbasic. In other words, activities consti
tuting the economic base of an area export to other areas,
while nonbasic activities in an area serve local needs,
and are not involved in interarea trade.
Initial reference to the idea of the economic base
53
in urban economics literature is somewhat vague.1 Homer
Hoyt, during the 1930's clearly defined the economic base
concept in operational terms, and applied it to community
analyses in the form of the economic base study.2
The insight into urban economic functioning offered
by the economic base concept is a dynamic one. City growth
is postulated to derive directly and largely from in
creases in local exports. As Palmer puts it
It is supposed that as basic income grows,
so grows the city; little account need be
taken of other factors in the growth of
derivative activity, since the latter may
be expected to grow automatically with
the increase in basic activity.3
In order to grow, or at least not to shrink in size,
a city or metropolitan area must improve its position, or
hold its own, in export markets. Historically speaking,
-*-For early mentions of the modern concept of the
urban economic base see; M. Aurousseau, "The Distribution
of Population, A Constructive Problem," 2. Geographical
Review, (1921), 574; Robert M. Haig, Regional Survey of
Hew York and Its Environs, (New York; New York Regional
Planning Committee, 1927), pp.42,43; and P.L. Nussbaum,
A History of the Economic Institutions of Modern Europe,
(New York: F.S. Crofts & Co., 1933), p.36.
^Some of Hoyt's earliest writing on application of
the economic base concept includes: The Structure and
Growth of Residential Neighborhoods in American Cities,
(Washington: FHA, 1939), p.131, a book written with
Arthur Weimer, Principles of Urban Real Estate,
(New York: Ronald Press, 1939), chapter 3, and "Economic
Background of Cities," The Journal of Land and Public
Utility Economics, (May, 1941), 188-195.
^Edgar Z. Palmer (ed.), The Community Economic Base
and Multiplier, (Omaha: University of Nebraska, Business
Research Bulletin No. 63, 1958), p.l.
54
export demand is an exogenous factor, beyond the control
of even a large metropolitan area. However, internal
costs are not necessarily beyond local control. This may
mean, that as cities and other types of regions grow up
around an export base, external economies will develop
which improve the competitive cost position of export
goods and services. These external economies may be in
the form of a diversity of local services on which other
firms can draw on short notice for a multiplicity of needs.
Or, they could be in the form of the availability of face-
to-face contact between decision making personnel in
national market firms. Whatever their specific form,
these kinds of external economies derive from the exist
ence of a very large concentration or agglomeration of
people and activities in one place, e.g., in downtown
Manhattan.
Since the external economies attracting new export
activities, by reducing the costs of export enterprise,
often arise as a result of the growth of the nonbasic
sector of the local economy, it might be concluded that
the nonbasic sector also makes a material contribution to
city growth. This conclusion would be quite valid, but
only if it were explained that nonbasic activities
contribute to city growth by having previously induced
export activities to locate in a city. This is so because,
55
while local market activities are necessary to the growth
of a city, it is the export market activities that fulfill
the conditions of both necessity and sufficiency in
initiating and maintaining local economic growth.
Theoretical foundations of the economic base
concept.— Why is it that export activity has great
significance for certain kinds of geographic areas? At
the national level, in the United States for example, it
is apparent that changes in the level of export activity
are not the prime movers behind national economic growth.
The export sector of the U.S. economy accounts for only
about five per cent of Gross National Product each year.
The extensive resource base, including natural, capital,
and human resources, of the U.S. permits the national
economy to grow from internal investment sources, with
relatively small reliance on exogenous factors such as
export markets.
The smaller in geographic extent is a nation, the
narrower, i.e., smaller and less diversified, its resource
base is likely to be. As Weimer and Hoyt have said:
An urban region, unlike a nation, does not
possess extensive farming areas, mines and
forests which enable it to be self-sufficient.
The residents of a city cannot support them
selves by taking in their own washing. They
must produce goods or services for export
56
beyond the limits of their urban region
to pay for the food and manufactured
products which they consume but do not
produce themselves.4
In order to maintain a relatively high standard of living,
a small nation in the international economy, or a region
within a nation, must specialize in accordance with local
comparative advantage and engage in trade with the "rest
of the world."
The level of output in an economy is limited by the
quantity and productivity of the resources it contains.
Crosson says that,
Base theory deals with the mechanism by
which these limitations may be partially
overcome. The mechanism, of course, is
'foreign' trade. The area may 'import*
capital goods and materials to supplement
'domestic' supplies of these and thereby
permit resource combinations which yield
a higher level of income than would be
possible in the absence of these imports...
It is for this reason that base theory
places such emphasis on exports: aside
from improvements in productivity, the
ability of an economy to grow beyond
some (relatively low) level is set by
its ability to export.5
The smaller in geographic size and narrower in resource
base is an area, the greater the applicability of the
economic base concept as an analytical framework, down to
4Arthur M. Weimer and Homer Hoyt, Principles of
Real Estate. (New York: Ronald Press Co., 1960), p.362.
^Pierre R. Crosson, “Further Comment on Economic
Base Theory," Land Economics. (May, 1960), 198.
57
a certain point. A geographic area would, of course, have
to be a minimum size such as to be a functioning economy
in a useful sense of the word.
The theoretical content, then, of the economic base
concept derives from international or interregional trade
theory. The economic base framework of analysis views
viability and growth potential of local economies purely
in terms of ability to engage in satisfactory trade
relations with the "rest of the world." The term, "rest
of the world," is used because trade relations between
specified trading bodies are not analyzed.
Any increase in a city's imports are assumed to have
no discernable effects on income levels in areas with
which a city trades. If interregional trade effects were
deemed significant enough to take into account, a "feed
back" effect might be measured in the following manner:
if a city's imports from a surrounding hinterland rise,
then hinterland income rises because of increased sales
to the city, and hinterland residents spend some of this
higher income on imports from the city, i.e., city exports
rise. It is this kind of feedback process that economic
base analysis does not account for.
Economic base theory views cities in much the same
manner that international trade theory views a small
nation. The small nation case is one where the trading
58
body accepts its level of exports as given, i.e., their
level is exogenously determined. Since a city can do
nothing to influence the level of its exports, the
fortunes of the entire local economy are dependent on
national, and perhaps regional, forces.
In any event, at this point it can be observed that
the greater the economic specialization of a city, when
associated with high levels of living, the greater the
importance of export activity, especially if the resource
base is narrow. One authority notes, in reference to the
operational usefulness of the economic base concept, that
"Many if not all counties in the United States are far
from being self sufficient and it is reasonable to assume
that many of these areal units are particularly suited to
the use of the basic-nonbasic concept."6
Economic location theory can help to explain why
basic activities locate in a particular area in the first
place. Basic activities may originally be attracted to
an area because it has certain locational advantages,
perhaps a superior competitive cost advantage over other
areas, for particular basic activities. Cheaper trans
portation costs have led to concentration of national and
regional market activities in a few large cities. Agglom-
^Morgan D. Thomas, "Economic Activity in Small
Areas," Land Economics. (May, 1960), 169.
59
eration-type external economies add to the differential
internal cost advantages of larger cities in a cumulative
manner over time. Historically, the increasing concentra
tion of economic activity in cities has given rise to a
greater division of labor within cities and increasing
specialization between cities. These latter two phenomena
have promoted higher levels of export-import activity,
which, in turn, yield higher standards of living for
residents of separate geographic areas each having re
source bases too narrow to support self-sufficiency at
high levels of living.
Emphasizing the high degree of geographic special
ization that must be accounted for in city economic ana
lysis, Palmer notes that "The reason for the heavier
dependence on the basic sphere is the nature of the
exchange economy which is characterized by a high degree
of division of labor and specialization."7 The less self-
sufficient is a local economy, the greater is the extent
to which it must rely on imports to maintain a high stand
ard of living. The greater the imports required, the
greater is the leakage of income from the local economy.
The local economy must, therefore, have more basic activi
ties to supply new local income if it is to import more,
at least in the long-run.
^Palmer, The Community Economic Base..., 4.
60
Economic Base Ratios.— In its traditional form, an
economic base study nearly always included a formal calcu
lation of the "base ratio" for a community under study.
Since most earlier, and some recent, economic base studies
have used employment as the unit of measure, a base ratio
can be specifically defined as a ratio of nonbasic employ
ment to basic employment in a local economy.
During the 1930's, Hoyt devised an application of
economic base analysis for the Federal Housing Administra
tion to test the economic viability and strength of local
communities for mortgage lending purposes. As late as
1939, Hoyt assumed that there existed a typical ratio
relationship of 1 : 1 between basic and nonbasic employ
ment. The assumption was that, after any given increase
in basic employment, the ratio of 1 : 1 would be restored
as nonbasic employment automatically adjusted to any in
crease. Under these assumptions, the ultimate increase
in total local employment would be twice the original
increase in basic employment. In discussing a weighting
procedure that could be applied to the predominantly
export-oriented industries of a local economy, Hoyt
pointed out that,
Since it is estimated that the persons in
basic sources of employment support on the
average an equal number of persons in the
service activities...double weight should
61
be given to the industry and specialty
(basic) categories. Thus, if fourteen
per cent of all gainfully employed
persons are engaged in industry, the
weight for the industry category should
be twenty eight per cent.8
If this ratio remained constant over time, expected
changes in the level of total employment could be derived
from knowledge of present and expected future levels of
employment in the basic or export sector of a local eco
nomy. A local employment multiplier (multipliers are
discussed in the following section) could be expressed as
a ratio of total employment to basic employment. In
Hoyt's case, the multiplier would be 2 r 1, or two.
In subsequent writings, Hoyt, and virtually all
other economic base theorists and practitioners, have
adopted considerably more sophisticated views relating to
the size, character, and behavior of economic base ratios
through time and space. Locational and temporal vari
ability of economic base ratios and local multipliers
limit, but do not invalidate, the utility of the economic
base concept as a framework for analyzing local economies.
The local multiplier process.— A local area multi
plier effect arises as a result of the leverage exerted
by changes in basic activity on changes in total economic
8Hoyt, The Structure and Growth..., 131.
62
activity of an area. Isard notes that, in a general sense,
local multiplier analysis
...stresses the interrelations of sectors
within a regional economy and the spread
of impulses originating in any one sector
to all other sectors either directly or
indirectly. Such spreading in essence
has a multiplying result.9
Economists normally think of multipliers in terms of
a dynamic process, or, at least, in terms of comparative
statics. However, one economic base theorist defines both
a static and a dynamic community multiplier.
Statically, it is the ratio of the total
economy to the basic economy at a given
time. Dynamically, it is the change in
the total economy compared with the change
in the basic economy over a given interval
of time.l°
Pfouts believes that the incongruency in thought on this
matter arises because economists and city planners use
the word multiplier in different ways.
To the base theorists, the multiplier is a
ratio, such as the ratio of basic to total
employment. To the economist, the multiplier
is an expression of change in income that
arises from an autonomous change in some
variable such as investment or exports.H
^Walter Isard, Methods of Regional Analysis,
(Cambridge, Mass.: The M.I.T. Press, 1960), p.189.
■^Palmer, The Community Economic Base..., 6.
llRalph W. Pfouts, "Reply to Harris on Testing the
Base Theory," Journal of the American Institute of
Planners. (November, 1958), 241.
63
The multiplier is used in this study in the usual
dynamic sense, in which local income is generated by an
iterative process of successive rounds of spending by
local residents through time, because of an initial in
crease in exogenously determined export income.
Depending on whether the unit of measure is employ
ment or income, a local multiplier could be either an
employment or an income multiplier. Numerical values for
these two multipliers
...would tend to be identical except for
changes in wage-levels and price-levels,
changes in income distribution, and the
occurrence of decreasing (or increasing)
returns which would increase output
proportionally less (or more) than
employment.12
It is generally conceded by economists that the income
multiplier is a more useful concept than the employment
multiplier. The former reveals more about income and
employment determination, and, it provides a more useful
measure of community welfare.
Multipliers can, of course, be classified in many
ways. Multipliers are classified according to the unit of
measurement of the variables undergoing change in the eco
nomic base ratio; i.e., income, spending, output, or
employment multipliers, or according to the independent
l2Palmer, The Community Economic Base.... 41,42.
64
variable or exogenous element in the system; i.e., invest
ment, foreign trade, consumption, or government spending
multipliers. All of these exogenous variables may, of
course, have induced components. Empirically relevant
classifications are numerous and need not detain us
further. It need only be mentioned that the type of
multiplier model considered most useful in local area
analysis by the writer can be termed an economic base
income multiplier model.^
Local multiplier theory assumes that a chain reaction
process is set off when basic income increases. A sympa
thetic reaction is presumed to be automatic, though not
necessarily instantaneous. The sympathetic nonbasic re
action is viewed as a process of readjustment, by which
local economic activity will ultimately attain a new level
of equilibrium. A local economy is in equilibrium when a
predetermined "normal" economic base ratio relationship is
reached again. This assumes, of course, that a judgement
can be made as to what basic-nonbasic relationship should
be maintained after a local economy has experienced a
dynamic process of growth and, perhaps, structural change.
Of all the multiplier concepts developed in national
income determination analysis, the local multiplier is most
l^Philip Howard Friedly, "An Economic Base and
Income Multiplier Study of Redondo Beach, California,"
65
closely related to the foreign trade multiplier. In
adapting foreign trade multiplier analysis to the study
of local economies, the following designations are made:
Y is local income; C is consumption by local residents;
X is local export of goods and services; E is net export
income; M is local imports; and B is the balance of trade
on current account.
Local disposable income is defined as
Y = C + X + E
The current account trade balance is defined as
B = X + E - M
The following propensities of local residents are
defined as
c = ip (average propensity to consume locally)
m = ^ (average propensity to import)
g
b = — (average propensity to lend to nonresidents)
Local income can be allocated by consuming it local
ly, exchanging it for imports, or lending it to nonresi
dents. So, according to the above definitions,
c + m + b = 1. In this model, the ratio of local output
to export output, i.e., the base ratio, is c / m + b.
Since c + m + b = 1, c and (m + b) must move in opposite
directions.
A local area's propensity to trade with other areas
66
is related to its diversity of products and services
relative to its diversity of demand. The more specialized
is an area's output, the higher is its propensity to trade,
given its diversity of demand. Or, the more diversified
is an area's demand, given its output diversification,
the higher is its propensity to trade. Also, as income
rises, propensity to import rises.
When b is greater than zero (a surplus trade balance
■with all other areas) it represents an average propensity
not to spend. This indicates that local residents are,
on balance, lending to other areas rather than receiving
imports now. If b is less than zero (a deficit trade
balance) it represents an average propensity to borrow
from nonresidents. In this case local residents are, on
balance, borrowing from other areas to receive more
imports now than they can pay for out of current export
income.
Sirikin says that
An area can have a surplus or deficit in
the current account of its balance of
payments (i.e., b can differ from zero)
either as a temporary disequilibrium
condition or as a continuing dis
equilibrium position.
The imbalance is short-run when it is being financed by
!4g . Sirikin, "The Theory of the Regional Economic
Base," Review of Economics and Statistics,
(November, 1959), 427.
67
payments from the deficit to the surplus area. Eventually
the financial reserves of the deficit area will be ex
hausted. The analagous situation in international trade
is where a deficit country exhausts its supply of con
vertible foreign exchange and its gold reserves.
There is an equilibrating process to restore balance,
in which monetary payments reduce prices and incomes in a
deficit area, thereby reducing imports and giving local
exports a competitive cost advantage, and increase prices
and incomes in the surplus area, causing a rise in imports
and putting exports at a competitive disadvantage. This
equilibrating process has little, or no, relevance to an
economic base study, however, because any interregional
trade feedback effects were previously assumed away. For
very large metropolitan areas, a diluted feedback effect
might be discernable, but it would not be strong enough
to operate as an equilibrating mechanism in the sense
just described.
An imbalance is of a long-run nature when it is
being covered by financial lending from the surplus to
the deficit area, so that no net monetary payments are
required between areas.
Sirikin observes that
Empirical studies of the behavior of base
ratios will be required to ascertain whether
the key relationships indicated in this
model are sufficiently stable or follow
68
sufficiently predictable paths that the
base-ratio concept can be a useful tool
to planners.15
The problems of multiplier stability through time are
dealt with in subsequent sections of this chapter. It can
be seen at this point, though, that local planning agencies
can derive much knowledge about the economic intentions of
local residents from behavioral implications in this kind
of model.
Delimiting a geographic area of study.— What criteria
should be applied in selection of a local area for eco
nomic base analysis? Since the purpose of this study is
to relate economic base analysis to the needs of the local
government planning process, criteria of study area
selection must be related to the purposes and goals of
local planners. This conclusion is in general agreement
with Tiebout's opinion that
No single geographic area is most appropriate
for a base study. Many areas are appropriate,
such as the labor supply area, the trading
area, and the metropolitan area. Each has
some merit. Yet, what is to be included
is a matter of judgement.15
Nonetheless, an acute awareness must exist of the effects
15Ibid., 429.
■^Charles m . Tiebout, The Community Economic Base
Study, Supplementary Paper No. 16, (New York: Committee
for Economic Development, 1962), p.21.
69
of selecting alternative types of geographic areas for
economic base study.
Let us assume that an operational criterion is the
location of local government jurisdictional boundries.
This criterion is non-economic in nature. It is operation
al in terms of availability of data and geographic co
incidence of study areas with local decision making and
planning bodies. Because municipal government has author
ity over a specific geographic area, it may want to study
that politically defined area as a local entity for
planning purposes.
This operational criterion must be compared with
economic criteria, if reasonable conclusions are to be
drawn from an economic analysis.
The base ratio tends to vary directly with size of
city population because of the greater self-sufficiency
of a large and heterogeneous population. The base ratio
also varies directly with the geographic extent of a city.
The "foreign" trade sector of a local economy depends on
other things than geographic size; but there is a definite
and common-sense relationship between the two. One
authority points out that
...as an area grows in size it supplies
and consumes internally a growing range
of services and goods; and it imports a
70
lessening percentage of the inputs
required for the goods and services
it does supply.17
Quite obviously, if the borders of a defined area
are extended farther and farther outward by successive
steps, each outward extension encompasses sources of de
mand that were previously basic. These previously basic
demands become nonbasic. In other words, as an area
grows in geographic size, it is progressively encompass
ing first, metropolitan, then regional, and, finally
national market demand forces within the economic system.
Eventually, when the resource base of an area becomes
very broad, export demand is neither necessarily an ex
ogenous nor a significant factor in relation to economic
growth.
Generally speaking, economic base analysis is most
relevant and useful when applied to a geographic area for
which interarea trade is the most important source of
growth and development. This type of area is usually
relatively small in geographic extent and contains a
rather narrow resource base, e.g., a municipality.
Price structures of small and large cities are
different. Cost of land, for example, increases with city
size. The effect of changing price structure might be to
■^Charles T. Stewart, Jr., "Economic Base Dynamics,"
Land Economics, (November, 1959), 328.
71
reduce the size of the base ratio as city size increases.
This would result because the bulk of items whose costs
rise are local service activities. The costs of many
manufacturing products with regional or national markets
may decrease because of scale economies in the distribution
industry.
Economic base ratios appropriate to particular city-
sizes, in terms of either population or geographic extent,
or both, have to be indicated. Even then, these appro
priate ratios can be specified only for given levels of
economic specialization and welfare existing in cities of
various size. The conclusion, then, is that even though
measured base ratios have displayed wide variation as
between cities and over time, they may still be subject
to generalization once account is taken of the systematic
causes of variation.
Some writers have suggested that degree of economic
interdependence should be a criterion for judging where
boundries are drawn. This interdependence would mark, in
a rough manner, outer limits for an area's economy. Eco
nomic interdependence would be measured in several ways?
(1) noting the points where transport routes begin to
veer off toward other metropolitan areas, (2) mapping the
radius served by the metropolitan press, (3) noting the
degree of dependence of outlying financial institutions
72
on the center for clearances and reserves, and (4) noting
whether or not it is from the center that outlying retail
ers procure their supplies. According to this approach,
it is the purpose of area delimitation to reveal an urban
community as an economic and social entity whose product
ive and distributive parts are interdependent to a high
degree.
In reference to metropolitan areas, it might be
desirable to define a region as a labor market area. The
labor market area for most industries in a metropolitan
area is often conceived as being roughly coincident with
the metropolitan area itself. On the other hand, the
different product market areas, i.e., trade areas, of
various local industries may vary from a small fraction of
the metropolitan area itself, to the total area, to the
entire national economy. This approach assumes, at least
in the short-run, that the spatial immobility of labor
temporarily isolates the labor supply of a metropolitan
area from the rest of the economy, and closely identifies
its economic welfare with that of the metropolitan area
within which the labor supply resides.
The particular interest of the present study is in
the delimitation of the market or trading areas of products
and services produced in a previously defined municipal
planning area. The purpose is to derive a market area for
73
the totality of economic activity concentrated within a
particular planning area, e.g., a small city in or adjacent
to a larger metropolitan area.
There are problems involved in delimiting a market
area for any particular urban concentration or node. Market
areas served by different economic activities operating in
a city will vary greatly among themselves. This is what
leads Mayer to say that
The boundries of the service area of a city,
therefore, are not only arbitrary in most
instances, but are composites of numerous
boundries, each in itself capable of being
delimited only arbitrarily.18
The trading area boundry of a city which serves as
an economic focus for a surrounding area can be determined
for each economic activity, or for the composite of all
activities, by measurement of the amount of "gravitational
pull" that the city exerts on the surrounding area.
Mayer notes that studies indicate "hierarchies" of
urban local market areas related to hierarchies of urban
places, each successively higher level of local market
area in the hierarchy serving a more geographically exten
sive area, which includes local market areas lower on the
scale and with less spatially extensive functions. In a
metropolitan area there are, then, numerous levels of
^Harold M. Mayer, "Urban Nodality and the Economic
Base, " Journal of the American Institute of Planners.
(Summer, 1954), 117.
74
economic function, each serving populations in different
trade areas. Mayer laments the fact that
Economic base studies of the largest urban
agglomerations have generally neglected any
investigation of the relations of the city
to the region, or, rather, the hierarchy
of regions, for which the city serves as
the primary focus.
According to this trade area approach, it would be
a good idea to determine the immediate geographic hinter
land, say of a large city such as Chicago, that is the
local market area (and does not involve a true export
function) of the city. For the city of Chicago the local
market area might roughly consist of the Chicago Standard
Metropolitan Area, i.e,, the contiguous counties surround
ing Chicago. Economic activities in Chicago serving this
market area would constitute the nonbasic sector of the
local economy. Economic activities serving midwestern
regional markets, the nation, and foreign countries, would
constitute the basic sector of Chicago's economy.
Tiebout suggests that economic base studies be con
ducted of metropolitan areas as units, and, in addition,
that separate studies be made of local communities and
cities within metropolitan areas. A metropolitan area
study will enable local communities to evaluate their
positions within a larger context, while community studies
19Ibid.. 117.
will enable metropolitan areas to observe the operation
of their components in greater detail.
Specifically, Tiebout says that
The economic base concept also applies to
single suburban communities. The suburban
community is part of an urban complex.
Typically, it is commuter oriented with
only a smattering of local industry other
than retail trade.20
This is probably the type of city for which the model
developed in Chapter VI of this study is most appropriate.
Concern is with an efficient allocation of retail space,
and this economic problem is of proportionally greater
concern to a city in which the predominant basic activity
is retailing.
Many suburban cities contain retail facilities
having trade areas extending far beyond municipal
boundaries. Demand for retail goods and services origina
ting outside the municipality is of a basic nature. This
is so, because, unlike a metropolis such as Chicago, a
small suburban city exerts no influence or control on the
export demand for its retail goods and services, even
though that demand emanates from an adjoining or surround
ing metropolitan area. That is to say, a small city cannot
affect regional or national economic forces affecting jobs,
incomes, and spending propensities of people residing in
20Tiebout, The Community Economic Base Study. 22.
that portion of a retail trade area lying outside city
boundaries. So, retail expenditures in a small city by
nonresidents can be considered as giving rise to basic
income for residents and local businesses.
Economic base models for local analysis.— Economic
base models are constructed in order that significant
sectors of local economies can have their behavior
"explained" given changes in other "unexplained" or
exogenous sectors. The assumption is that small cities
do not affect sizes or rates of change in their export
market demand sources.
Depending on whether analysis is of a short-run or
long-run nature, more or fewer sectors of a local economy
may be unexplained by the model. As Tiebout puts it,
Exports depend on external market demands.
The community is such a small part of this
total market that it does not affect this
market but is affected by it. Year-to-year
variations in local investment are also
assumed to depend on outside forces;
interest rates, general credit conditions,
and other such non-local factors. All of
this means that, for short-run analysis,
export and local investment income are
taken as given, that is, they are measured
but not explained by a base study. This
leaves the local consumption sector which
is "explained".21
In long-run analysis only export income remains an unex-
^1Ibid., 58.
77
plained factor.
For this kind of model to be useful, it must, of
course, be a valid assumption that some sectors of a local
area's economy are truly endogenous, because
This implies that given the level of income
generated by basic activities, i.e., income
arising out of basic activities, one is
theoretically able to predict total community
income.^2
Specifically, the process by which local income is created
as a result of an unexplained increase in export income
of residents is as follows: (1) residents spend some
proportion of their increased income locally; (2) these
local expenditures create local retail sales; and (3)
some proportion of these sale's revenues are received by
residents as local income.
Of relevance to the present study is the observation
that, in some local areas, "...retail sales may be exogen
ous if the demand is not determined by the city itself."23
If incomes of nonresidents making local retail purchases
are not earned in that local area, then the latter's retail
export sales constitute an exogenous variable in the model.
Determination and analysis of the geographic market served
by local retail establishments through trade area analysis,
22charles M. Tiebout, "A Method of Determining
Incomes and their Variations in Small Regions, "
Regional Science Association Papers. (1955), F-8.
23Ibid.. F-7.
78
would help explain more thoroughly the behavior of an
important segment of a local economy's export demand. A
complete explanation of the behavior of retail export de
mand is, of course, impossible without analyzing regional
and national trends affecting jobs and incomes of a trade
area population. Nonetheless, a detailed knowledge can be
gained of expenditure propensities and patterns of a trade
area population, given assumptions about national and
regional factors affecting trade area jobs and incomes.
Economic base models can be formulated in many ways.
As noted previously, income as a unit of measure has
greatest utility in terms of the writer's approach. This
is so because income analysis reveals in much greater de
tail the behavioral characteristics and attitudes of resi
dents in a community, than does an employment study. In
come analysis permits observation of the process by which
local income is generated from rising export income.
In addition to these important reasons, income ana
lysis is also compatible with the other analytical com
ponents of the retail space allocation model developed in
Chapter VI. Trade area analysis measures expenditure and
sales potentials for retail facilities. It focuses on
family income and allocation of that income, both geo
graphically and by product or service. Space requirements
analysis is formulated in terms of allocating economic
79
quantities of physical space to serve certain expected
levels of retail sales, measured in dollars.
The following definitions are for the various ele
ments of a local economic base income multiplier model:
B = balance of trade; X = exports and M = imports
Yx = export income (independent variable)
Ym = local income (dependent variable)
Y = total income (Yx + Yn)
m = marginal propensity to import
s = marginal propensity to save
c = marginal propensity to consume locally
h = local income created per dollar of local sales
(i.e., income accruing to residents employed
in local business establishments)
b = local income creating propensity of local
expenditures (c^ h)
k = local income multiplier
d = "an increase in"
Suppose that X is greater than M leading to dYx, i.e.,
there is an "injection" of new money into the community.
If both m and c are positive fractions and m is greater
than zero while c is less than one, then the proportion of
new money remaining in the community after importing is
1 - m, and after saving is 1 - s. Total leakage from the
local income stream at each round of spending is 1 -
(m + s). The multiplication, dYx (1 - m) c, must be
repeated for each successive round of spending until the
80
iteration is completed. The resulting geometric progress
ion of the iterative process is summarized by the follow
ing equation:
dY = dYx ------~ ^ ----- = kdYx
1 - (1 - m) c 1-c-cm
Traditionally, in economic base theory a community
would have to attain a favorable trade balance (i.e., X
greater than M during some specified time interval) to
achieve net growth in its total income. While this is a
fundamentally valid conclusion, it led traditional eco
nomic base theory to often place undue stress on export
income as virtually the only source of income growth. As
an alternative to the older economic base ratios empha
sizing exports, Pfouts, Tiebout, and others have offered
new formulations in terms of local income multiplier ana
lysis. These models account for variables such as imports
and savings. As Pfouts points out, "...it can be argued
that imports, since they represent money leaving the
community, should be given a place in considering the
income stream within the community."24 pfouts continues
by emphasizing that if the multiplier is greater than one
"...it seems incorrect to emphasize exporting industries
to the exclusion of industry that causes the circulation
24Ralph W. Pfouts, "An Empirical Testing of the
Economic Base Theory," Journal of the American Institute
of Planners, (May, 1957), 58.
81
of money within the community."25
There also is a close link between imports and the
location of economic activity. Given its present level
of production and consumption, the entrepreneurs of a city
or metropolitan area face the problem of whether to rela
tively increase or decrease its local production of a
certain commodity or service, at the expense or gain of
local imports of this commodity or service. This view
point sees the problem of interarea exchange as one of
substitution. This
...means that we conceive of the exports from
a metropolitan area as being dependent on the
economic possibilities of substitution of
imports for local production, or vice versa,
in all other regions (including their metro-
politan areas).*6
This would involve a model which accounted for inter
regional trade relation effects. It has already been
indicated that an economic base model assumes away all
feedback effects arising from interregional trade because
they are not significant on a municipal scale. Although,
imports and local production should neither be neglected
nor relegated to a subordinate position in urban economic
analysis, generally.
25Ibid., 69.
2^Roland Artie, "On Some Methods and Problems in the
Study of Metropolitan Economics," Regional Science
Association Papers. (1962), 81.
82
To rid the analysis of implication that nonbasic
activity is less productive or essential to city growth,
Palmer states that
All labor producing any kind of utility,
whether engaged in the basic or the
derivative sphere is productive. The
derivative activities are complementary
to the basic activities and both activities
are essential and indispensable parts of
the functioning of the community's life.27
The essential nature of local market activities is revealed
in the local income multiplier process, where increases in
export income generate even greater increases in total
income. The significance of local importing activity
lies in the substitution process where local productive
facilities supply a city with goods and services previous
ly imported, thereby relieving the export sector to some
extent in maintaining economic welfare.
Tiebout's formulation of a local income multiplier
model reveals more clearly the multiplier process by which
local income is generated from local retail expenditures,
or sales. Using definitions from the above economic base
model, Tiebout's multiplier model is expressed as:
dY = dYx ---—--- = — -— = kdYx
1 - (ch) 1 - b
This is a short-run model with only local consumption act
ivity assumed to be explained by the model. Other factors,
27Palmer, The Community Economic Base..., 14.
83
local investment activity, for example, are assumed to be
independent variables in the model during the short-run.
In a long-run model, a marginal propensity to invest
locally, and a propensity of a dollar of local investment
to create local income, could be added to the endogenous
part of the model. The assumption then being that both
local consumption and investment have their levels ex
plained by total income.
In local area studies Tiebout feels that short-run
and long-run time spans should be defined in the following
manner:
Short-run time spans, as a rule of thumb,
refer to periods from a few months up to
around two years. Long-run analysis is
usually concerned with five, ten and/or
twenty five year time p e r i o d s . 28
Over a period of two years or less, export and local
investment demand largely determine the level of total
income. For periods of time exceeding two years, export
demand alone determines the level of total income, while
total income itself determines the level of local invest
ment demand and the other endogenous sectors.
For study of larger areas (metropolitan, regional,
state, and multi-state economic areas) Hansen and Tiebout
have developed intersectoral flows analysis. They offer
this approach as an alternative to economic base multi-
28Tiebout, The Community Economic Base Study. 57.
84
plier, and input-output studies. Actually, intersectoral
flows analysis utilizes both economic base and input-
output frameworks of analysis. The Hansen-Tiebout model
is operational in the sense that necessary data can be
acquired relatively easily and cheaply, as compared with
input-output studies.
Hansen and Tiebout used employment as a measure in
their intersectoral flows study of the California eco
nomy. using this approach, industry classifications can
be made as fine as desired. Hansen and Tiebout divided
demand sources into seven final demand sectors. These
were: local consumption; local business and housing in
vestment; local government current operations and invest
ment; and private exports and exports to the federal
government.
An intersectoral flows table was then constructed
showing sales by each industry to all other industries
(intermediate demand), and sales to all of the final
demand sectors. The sales of each industry, then, can be
allocated according to the per cent going to other indus
tries and the per cent going to satisfy final demands.
As in an input-output study, employment coefficients can
29w.Lee Hansen and Charles M. Tiebout, "An Inter
sectoral Flows Analysis of the California Economy,"
Review of Economics and Statistics, (November, 1963),
408-418.
85
be developed for each industry by dividing the employment
input from a particular selling industry by the total
employment inputs from all other industries for a
purchasing industry.
The total per cent of all industry sales allocated
to export demand can be determined by tracing sales
through the intermediate demand sector. That portion of
any industry's sales that ends up in an exported product
or service is allocated to final export demand.
Three kinds of multiplier effects on local employ
ment were noted by Hansen and Tiebout. These are:
(1) direct, i.e., the producer sells directly to the
exogenous final demand sector; (2) indirect, i.e., the
producer is linked to the final demand sector via inter
mediate producer demands; and (3) induced, i.e., the
operation of the local multiplier process within the
economic base framework. The behavior of these multi
pliers is discussed in the following section.
An objective of economic base studies is to dis
cover functional relationships between the basic and non-
basic sectors, and their effects on economic life in
cities. Economic base studies aid in understanding
current sources of income and employment, and in pinpoint
ing areas of weakness or strength in local economies.
They also point out industrial mixes of export activity.
86
They trace local sources of demand for products and
services of local firms. They describe factors affecting
local costs of both export and local market firms.
As a result of information generated about local
economic structure and behavior in an economic base study,
local government planning decisions can be more rational
in terms of both efficiency and welfare. Economic base
studies are often useful as starting points for many other
kinds of specialized local planning studies. As Tiebout
says, and as is demonstrated in Chapter VI of this study,
Frequently base studies also provide the
necessary background for more specialized
studies. Thus, base studies lead to or
are often coupled with research into
marketing, urban renewal, land use,
transportation, water supply, and
similar fields of special interest.30
Measuring and Forecasting Local Economic Activity Levels
Identification of the economic base.— An initial
problem faced in conduct of an economic base study is that
of accurately identifying basic activity. Ideally, perhaps,
it would be desirable to draw boundries for an area so
that economic activities in it could each be classified as
either entirely local or entirely export. This implies
that markets for goods and services produced locally fall
either wholly inside or wholly outside the area itself.
3®Tiebout, The Community Economic Base Study, 15.
I
87
In practice, of course, there is nearly always a mixture
of export and local markets served by local enterprise;
therefore, some allocation of sales and income between
basic and nonbasic sectors is necessary.
Total income of a local area can be classified,
first, by the industry from which it derives, and, second
ly, cross-classified for each industry by whether or not
it derives from export or local sales. The sales of each
industry to export and local sectors can be traced by
noting its direct, indirect (intermediate demand), and
thus, its total sales to each sector. "The rule of thumb
in classifying exports is whether or not the community
receives dollars from private outside sources."31 The
source of demand for the products must arise from outside
the local area.
The economic base of a local economy can be identi
fied either directly or indirectly. Direct measures of
basic activity involve field surveys to generate primary
data on the study area. Indirect measurement of basic
activity utilizes secondary data sources, and makes
certain assumptions as to the characteristics of the local
economy. The indirect approach has advantages of being
easier and cheaper to implement. However, a margin of
error must be allowed for when assumptions are applied to
31Ibid., 40.
88
often inadequate secondary data.
There are several important types of indirect eco
nomic base measurement. The first of these is referred
to as the assumption approach. The study area is observed
and note is made of the various kinds of economic activity
operating in it. An arbitrary assumption is then made as
to what is local and what is export activity. The basis
for such an assumption usually rests on the belief that
most, or all, agricultural and manufacturing activity,
especially in a relatively small area, serves export
markets and that all else is local. Use of this method
often results in gross error in estimating the size of
the economic base, because of wide variations in the eco
nomic structure and character of different geographical
areas.
Another widely used method of estimating the size of
the economic base is the location quotient technique. A
location quotient is derived from the following formula:
Number of employees in industry i in local area
Lq _ Number of employees in all industries in local area
Number of employees in industry ~ j nationally
Number of employees in all industries nationally
= Per cent of local employment in industry j
Per cent of national employment in industry j
When the location quotient equals one, local industry is
assumed to be just serving local needs with no exports.
If the location quotient is greater than one, the excess
89
activity over unity is allocated to the export sector, and
vice versa if the location quotient is less than one.
The purpose of this technique is to indicate local
specialization in particular industries relative to na
tional, state, county, or other benchmark regions. An un
derlying assumption of location quotients is that the local
study area has the same demand patterns, productivity, and
general economic structure as the benchmark area. Most of
the time this is not a valid assumption. A city or metro
politan area is not a microcosm of the nation, or even a
state. As pointed out earlier, a relatively small geo
graphic area tends to contain a narrower resource base,
and is more dependent on trade than are larger more di
versified regions. This is apparently the reason that the
location quotient, when compared with direct survey re
sults, usually understates exports by a significant
amount.
Because regional and local area analysis emphasizes
local differences rather than similarities, while location
quotients assume similarity among areas, sophistications
of this identification technique have been attempted.
These improvements on the location quotient are generally
referred to as the minimum requirements approach to eco
nomic base identification. This approach tries to account
for the differences between different kinds and sizes of
90
local economies.
In essence,
The minimum requirements method...yields
a quantitative statement which closely
approximates the minimum percentage of
a labor force required in various
sectors of its economy to maintain the
viability of an urban area.32
Employment in any industry in excess of the minimum re
quired is automatically allocated to the export sector
of a local economy.
Calculation of the minimum employment required in
a particular industry to serve the needs of a city of a
certain type, would be conducted in the following manner.
Take, say, 100 cities similar to the one under study and
compute the per cent of labor force employed in each local
industry for each city. Cities with the smallest per cent
employed in each industry provide the per cent figures
used as the minimum required employment to serve local
markets for cities of the type represented.
A refinement of the minimum requirements technique
is the development of k values. The k value indicates a
minimum per cent of economic activity, or employment in
that activity, that would be expected in a particular in
dustry in a normal city of a given type. The k value is
^Edward L. Ullman and Michael P. Dacey, "The
Minimum Requirements Approach to the Urban Economic Base,"
Regional Science Association Papers, (1960), 176.
91
found at the fifth percentile, rather than at the bottom,
of an array of per cents employed in a particular industry
in different cities of the same type. One counts up five
cities, from the lowest per cent employed, to remove the
chance of any fluke arising from an atypical city. Indus
tries with zero k values are termed "sporadic", while
those with positive k values are termed "ubiquitous".
Along these same lines, a method used by the writer
in his economic base study of Redondo Beach, offers a use
ful short-cut in estimating the approximate amount of re
tail exports of a smaller city. A retail balance of pay
ments index can be constructed for a city, indicating
whether or not the city serves retail patrons other than
residents. When per capita city income, expressed as a
per cent of county per capita income, is less than city
per capita retail sales, expressed as a per cent of
county per capita retail sales, the city is assumed to be
engaged in some amount of retail exporting activity.
Index values are computed as follows:
per capita city retail sales
REI = per capita county retail sales x 1Q_
per capita city income
per capita county income
When the index value exceeds 100, the assumption is
that a city is engaging in export of retail goods and
services. The higher the index value, the greater the
volume of retail exports. Though not a precise measure
92
of the retail export base of a city, the index does pro
vide some indication of the relative degree of trade
dependence that a city has in retail trade.
Recently, Hoyt suggested a different approach to
calculating the size of an area's economic b a s e . 33 He
would attempt to estimate the amount of an area's imports
during a given time period. Hoyt's assumption is that in
the long-run exports equal imports. Therefore, an equiv
alent amount of exports would have to be made to pay
imports. Hoyt's purpose in arriving at an estimate of
economic base size by this roundabout method is the great
er ease with which imports can be measured as compared
with exports.
Hoyt would measure local merchandise imports by
wholesale expenditures on goods from outside the area.
The retail mark-up from wholesale on merchandise includes
coverage of: local wages, local real estate taxes,
interest and depreciation on buildings or rents, local
delivery charges, sales taxes, and profits of local store
proprietors. Therefore, the retail mark-up cannot be
considered in estimating local import expenditures.
An assumption that the wholesale cost of imported
33Homer Hoyt, "A Method for Measuring the Value of
Imports into an Urban Community, " Land Economics,
(May, 1961), 150-161.
93
merchandise is about 65 per cent of local retail sales,
is made by Hoyt as a rule of thumb. A rough check on the
accuracy of the estimated value of imports can be made by
estimating the value of exports. An exact bookkeeping
balance would not be expected since neither of these
magnitudes can be measured precisely.
A distinction should be made. Residents in a metro
politan area usually make most of their retail purchases
within the area because of relative diversity of choice.
Consequently, they pay local wages and taxes included in
retail mark-ups of stores serving the local market. In
this case, only the wholesale cost of retail items not
produced locally constitutes imports. In smaller cities,
however, where residents may patronize retail stores which
are themselves outside the local area, 100 per cent of
their purchases represent imports, and not just wholesale
costs.
The difficulty in estimating the economic base from
imports lies in the fact that exports do not necessarily
equal imports, on current account, during any given period
of time. As discussed earlier, imports may be financed
in several different ways for varying lengths of time.
As a result, a deficit trade balance on current account
could be sustained indefinitely. It would be necessary
to investigate financial as well as merchandise trans-
94
actions to balance overall export activity with imports.
Direct measures of the economic base of a small
city are often a necessity because of lack of secondary
data sources. One method of generating primary data in
volves measurement of an area's commodity and money flows.
The volume, or dollar value, of goods and services flowing
across an area's borders must be ascertained. The trans
portation industry can provide information on the physical
movement of goods and people. Even here, except for
samplings of waybill data by the Interstate Commerce
Commission on railroads (which is difficult to convert
to dollar value) information is hard to come by. Data is
nonexistent for other modes of transportation, such as
trucks, etc. A survey of the study area, such as the one
undertaken by Fortune in Oskaloosa, I o w a ,34 is required
to generate sufficient information on all real and
financial transactions occurring within and between areas.
However, to gain accurate and complete local financial
data, an expensive and time consuming exhaustive survey
of a study area is a prerequisite.
Local consumer and business field surveys can be
conducted at relatively small expense, and they generate
sufficient information to supply data for an economic base
34"oskaloosa Versus the United States," Fortune,
(April, 1938), 55-62.
95
income multiplier model. Financial and commodity flows
are not measured. Rather, total family and firm incomes
are estimated, and classified according to geographic
origin, i.e., inside or outside a study area. Local
consumption and intermediate purchases by resident families
and firms are investigated. A statistical sampling of
families, and a complete survey of firms (for a small
city), will yield the necessary information for a local
income multiplier analysis.
Several authors strongly emphasize need for field
survey work in connection with economic base studies.
As McGovern says,
Two of the short-cut methods of identifying
the exporting, or basic, industries— the
location quotient and the index of surplus
workers— were tested against detailed
market survey data in the Vancouver
metropolitan area. They were found to
have identified only two-thirds of the
export industries. The findings suggest
that there may be no adequate substitute
for a comprehensive economic survey, since
the short-cut methods of identifying the
i ' ’ ovide information of only
Another authority feels that, "Simple multipliers of the
basic-service type can most accurately be constructed
35p. d . McGovern, "Identifying Exporting Industries,"
Journal of the American Institute of Planners,
(May 1961), 144.
through interview techniques together with the use of
financial data."^6
From a functional and analytical standpoint, Leven
recommends that the national economy be disaggregated and
studied in terms of nodes of urban economic activity,
rather than merely in terms of spatially exhaustive and
non-functional regions and areas such as states, counties,
cities, and so forth. The present study attempts just
such a functional disaggregation of local study areas
according to trade areas served by nodes of retail activi
ty. While a city may be the study area for planning
purposes, trade areas of local retail facilities probably
lie partly within and partly without a city's legal
boundaries.
To generate sufficient data on a retail trading
area, which may or may not correspond to municipal plann
ing boundaries, there must be a field survey. Data
reporting on the types of geographic areas to which the
: present analysis is likely to be applied, are either
: insufficient or nonexistent.
Local area income analysis.— A set of accounts
j should be set up for a study area to accurately identify
^Britton Harris, "Projecting Industrial Growth
of Regions," Regional Science Association Papers,
(1956), 247.
97
various magnitudes according to a framework of analysis.
As Leven says,
In the case of income and product accounting
the goal is to approximate as nearly as
possible a segregation into exogenous and
endogenous segments, i.e., income determining
as opposed to income determined activities.
Such a segregation is the backbone of
multiplier analysis. At the national level
these roles are played by investment and
consumption, respectively. At the local
level a different dichotomy is used as a
starting point; production for export
as opposed to production for the local
market.
An export account, for purposes of this analysis, might
be considered exogenous in both the short-run and long-
run. A local capital formation account, on the other
hand, might be exogenous in the short-run and endogenous
in the long-run. Finally, a local consumption account
might be endogenous in both the short-run and long-run.
In local area study, Leven feels that income multi
plier effects from exogenous change in export income are
much more significant than interindustry multiplier
effects measured by an input-output matrix. Therefore,
he feels that income accounts are the most useful tabula
ting device to measure local multiplier effects.
In determining interindustry effects the input-
3^C.L. Leven, "A Theory of Regional Social
Accounting, " Regional Science Association Papers.
(1958), 222.
98
output technique is highly suitable,
However, the very small size of the inter
industry effect, at least in an area of
60,000 people, gives one pause to reflect
whether such an intricate mechanism is
called for in analyzing it. Also, other
research results would suggest that
interindustry effects are not much
bigger in areas of even a few hundred
thousand. ...The relative importance
of the export-multiplier vs the linked-
industry effect, at least for medium
sized industrial areas, should clearly
indicate the relatively greater applica
bility of income and product accounts,
particularly for normative economic
considerations.38
The indication is that the input-output approach, in its
usual form, is not only too powerful a technique in many
cases, but is often inapplicable. There are elements of
the input-output approach, however, which can prove use
ful in analysis of even a small, simply structured, and
mainly retail trade-oriented local economy.
Units of measure.— There are a large number of
variables from which a unit of measure can be selected
for economic base analysis. A few of the more significant
units of measure are discussed here, with recommendations
for an appropriate selection in terms of the analytical
approach adopted in this study.
In the past, employment was frequently used to
38Ibid.. 236, 237.
measure levels, and changes in levels, of basic and non-
basic activity. The basic component of this measure is
the job. Some method has to be devised to ascertain the
number of full-time job equivalents existing in an area
at a given time. Employment was often used in the past
because of ready availability of employment data, in,
say, selected areas of a state. However, it is not as
sensitive a measure of economic change as other variables.
Employment changes can be misleading in measuring short-
run local economic adjustments, unless refinements are
made. The implications for income and output, and there
fore local economic and social welfare, of changes in
employment levels, depend also on local price and product
ivity levels vis-a-vis other areas.
Total sales in an area offer another unit of measure.
These count the total transactions occuring during a given
time period. Double counting takes place because some
products are involved in intermediate sales as inputs
into finished products, which were in turn recorded as
final sales. This measure does not reveal the absolute
level of income since total income will be somewhat less
than total sales.
Value added is a more accurate unit of measure in
determining local output and sales contributions. Value
added is the sales of a firm less the cost of materials
100
purchased from other firms. The value added of a geo
graphic area is the sales of all local firms less the cost
of materials purchased outside of the area. At the local
level, value added is roughly equivalent to national
income originating in a specific industry or area. Value
added, of course, avoids the problem of double counting
encountered in total sales. The main problem in its use
arises from the fact that many national firms operate
branch plants in several different locations. Corporate
income (profit) and sales data are seldom published or
released for individual plants. How than are corporate
sales to be divided for a multi-plant firm as between
sales of a local plant and sales of other plant units?
Accuracy is impossible to maintain in this case.
Income accruing to individuals,families, or business
firms, from any source is the most useful unit of measure
in terms of the present analysis. It is similar to per
sonal income at the national level. Though not as all-
inclusive a measure as value added, income accruing is
more easily and accurately determined than the former.
Also,
Per capita income provides a measure of the
average citizen's well-being. Base studies
carried on over time will measure changes
in per capita income and help to unearth
the reasons for such c h a n g e s . 39
^Tiebout, The Community Economic Base Study, 15,16.
The concept of income accruing to individuals aids in
measuring the economic and social welfare of people.
Similarly it aids in determining local market sales
potentials. As Adamson says,
In attempting to measure income in any
area, particularly in small areas, the
research worker necessarily will be
guided to a very large extent by the
nature of...the needs of those using
the estimates. For example, in
determining market potentials, primary
emphasis is placed upon income which
is available for commitment, that is,
"spendable income". 40
Disposable personal income is an important concept in the
analytical approach presently under consideration because,
in the following chapter, trade area analysis is developed
as a tool to more precisely determine present and expected
future potential expenditure and sales levels for local
retail facilities.
Empirical support for economic base theory.— There
have been a few empirical studies which, their authors
claim, invalidate economic base theory as a useful concept
in local area analysis.41 On the other hand, many com
petent authorities in the field of urban economics have
40W.M. Adamson, "Measurement of Income in Small
Geographic Areas," Southern Economic Journal,
(April, 1942), 480.
41see especially Pfout's article: "An Empirical Testing
of the Economic Base Theory," in the May, 1957 issue of the
Journal of the American Institute of Planners, 81-86.
given thoughtful reflection to this subject, and, in light
of other supporting empirical studies, feel that economic
base theory does have a valid foundation in fact.
One of the most significant economic base studies,
measuring changes in employment, is the one conducted in
Los Angeles County in 1949 by Hildebrand and Mace. The
purpose of their study was
...to divide the industries of the
given community between those essentially
producing for external markets ("non
localized" or "export" industries) and
those primarily serving the community
itself ("localized" or "home" industries).
Given a reliable procedure for making such
division...it then becomes possible to
correlate changes in localized community
employment with changes in non-localized
community employment, and so to derive an
employment multiplier for the "export"
trades in a local area.42
In the Hildebrand and Mace study, the local multiplier
hypothesis says that an increase in exports leads to a
rise in basic employment. This, in turn, leads to an
increase in community income and expenditures which cause
a rise in nonbasic employment.
Statistically, Hildebrand and Mace feel that
...the multiplier may only be used safely
in community employment estimation where
the regression line involves at least a
correlation coefficient of + .90. Values
42ceorge Hildebrand and Arthur Mace, Jr., "The
Employment Multiplier in an Expanding Industrial Market:
Los Angeles County, 1940-1947, "Review of Economics and
Statistics, (August, 1950), 241.
below this level suggest the strong
presence of changing independent factors,
operating to scatter the figures for
localized and nonlocalized employment
too widely to give a satisfactory fit.43
Their correlation in Los Angeles County are sufficiently
high to indicate that the local employment multiplier is
statistically significant.
A location quotient approach was used by Hildebrand
and Mace to identify basic employment. They compared Los
Angeles County with several different benchmarks; the West
(a specified regional grouping of states), Southern Calif
ornia (a specified regional grouping of counties), and the
United States. After allocating employment into export
and local categories, they made monthly estimates of
employment in Los Angeles County for each month was
designated by and export employment by X2- With all
of the x-j^'s and x2 ' s for each month covered by the analysis
a regression of x^ on x^ was found, i.e., a regression co
efficient or multiplier.
Hildebrand and Mace found a sample regression co
efficient of local on basic employment in Los Angeles
County for 1940-1947 of 1.248. This means that if basic
employment increases by, say, 10,000 jobs, then the esti
mated increase in local employment would amount to 12,840
43Ibid.. 243.
r~ ' -
104
jobs, or, a total employment increase of 22,840 jobs.
In similar fashion, the rate of change in local
income could be correlated with the present level of basic
income. The regression coefficient represents a dynamic
income multiplier which is useful for predicting local
income levels.
Hoyt feels that,
The fundamental question is whether the
economic base has a value in determining
city growth even if it is not an infallible
device for predicting the exact amount of
that growth. Can an urban region grow
without any increase in basic activities,
or are basic activities necessary at all?44
Hoyt's thesis is that some basic activities are necessary
in every urban community without which it cannot exist.
Strong and pronounced population growth cannot occur
without corresponding growth in basic employment. The
population of an area will likely decline, if there is a
reduction in the level of basic income below an amount
required to pay for its imports.
Hoyt offers empirical evidence that valid relation
ships between changes in basic activity and total popula
tion did exist for most rapidly growing U.S. cities from
1950 to 1960.45 fje mainly compares population increases
^Horner Hoyt, "The Utility of the Economic Base
Method in Calculating Urban Growth," Land Economics.
(February, 1961), 52.
45Ibid., 51-58.
with manufacturing employment increases during a specified
time period. In Los Angeles County, for example, Hoyt
notes that population increased from four to six million
between 1947 and 1957. Concurrently, manufacturing employ
ment rose from 387 to 772 thousand. Hoyt assumes that
manufacturing activity is predominantly export oriented.
As another example, the State of California experienced
population growth of over three and one half million
persons between 1947 and 1957. Concurrently, the State's
manufacturing employment rose by 494 thousand.
Because of these and similar data, Hoyt thinks it
is reasonable to extrapolate past trends in the relation
ship between increases in basic employment and increases
in population, for a given city into the future. However,
if a city grows rapidly without any increase in exporting,
then there is no logical reason to use the economic base
method as a forecasting tool. In the opinion of Dean
Gillies of UCLA, though, expected population increases are
not possible without requisite increases in basic manu
facturing opportunities. A region may reduce its reliance
on imports through greater diversification of activity,
but a new impetus to growth must come from basic employ
ment.
Since metropolitan areas and large cities have
diversified basic economic supports, it is a complex
problem to show the exact effect of a decline or rise in
any single basic support. Hoyt says that the "Most
important contribution of the economic base theory, in
my opinion, is its function in furnishing a sound method
of analyzing the economic structure of an urban r e g i o n . "46
Since the ratio between basic and nonbasic supports in a
city is variable over time, economic base studies should
anticipate and measure these changes.
Many authorities in urban economics have confidence
in economic base theory as a useful conceptual approach
to local area analysis. They place reliance on its empiri
cal validity in the conduct of economic base studies.
Thoughtfully interpreted, many of the spatial and temporal
variabilities of the economic base can be explained and
accounted for in a systematic manner.
An economic base model would be more useful if it
could be demonstrated that similar communities have similar
proportions of nonbasic out of total activity. This
situation demands that consumption habits not vary among
communities, and that productive activities also be
similar. Even where consumption habits are not the same,
if the differences are known they can be accounted for,
I and the level and distribution of nonbasic activity can
still be predicted.
46Ibid., 58.
107
Forecasting local exports.— One of the most important
tasks in conducting an economic base study is making an
accurate forecast of future levels of export activity for
an area under study. The importance of export forecasting
is emphasized by Tiebout:
An essential step in forecasting is the
identification of key factors which are
important to future levels of activity...
For communities the key factor is the
basic or export market. How will it
grow or decline over the next decade?
As the basic market changes so too will
the nonbasic market. In essence, base
analysis concentrates much of the
forecaster's task. Detailed attention
needs to be paid primarily to the
export market.47
The precise manner in which future export levels should be
estimated is debatable. Because of the influence of non
economic random factors within a metropolitan area, it is
especially difficult to make export forecasts for a small
city on a purely economic basis. Certain kinds of basic
activity may be oriented to a particular metropolitan area
for economic reasons, but their exact location within a
metropolitan area is often not determined by rational
economic calculation alone. One small city within a metro
politan area might be as economically feasible a choice as
several others. The ultimate choice for location of the
^Tiebout, The Community Economic Base Study. 16.
basic activity may be based on whims of firm decision
makers.
The type of basic activity referred to above is
largely that serving national markets. In a small metro
politan community these basic activities would be largely
represented by manufacturing firms. In some small cities
manufacturing is an important local industry. However,
incidence of large proportions of manufacturing activity
in small city economies varies widely in most areas.
Manufacturing activity tends to concentrate in certain
sections of metropolitan areas, thereby giving cities
located in those areas of concentration the largest
proportions of manufacturing activity.
On the other hand, incidence of large proportion of
retail activity in small city economies is high and evenly
distributed. Retail trade is usually an important segment
of most small city economies. In many cases, retail trade
provides an important source of basic income. Some small
cities contain large regional shopping complexes which
serve not only local residents, but also serve outside
residents scattered about the metropolitan area. The
combined trade area, situated outside a city, of all local
retail establishments constitutes the regional export
market for retail goods and services.
Since local retail facilities are serving spatial
109
markets located within a metropolitan area, a more useful
and accurate economic projection of future levels of retail
activity can be made. The composition and character of
metropolitan trade areas can be intensively studied
through trade area analysis, and specific sections of
rapid population growth can be delineated and retail
expenditures and sales potentials estimated.
An economic base model is open as far as exports are
concerned. "A formula which forecasts the level of the
export sector would complete, in a sense, the model."48
At the present state of development, however, some kind
of rough forecast must be relied upon.
A simple investigation could be made of present
export industries and a forecast made of their future
potential. Also, investigation could be made of the
possibility of new export industry in the future. A
feasibility study would compare location factors of various
industries, and, in turn, compare these with the economic
potential (resource base) of the local area, in terms of
holding existing exporters and attracting new ones in the
future. This approach accounts for differences between
local areas. It is a difficult approach to implement,
though, because future local industry input requirements
cannot be accurately determined.
48Ibid., 61
A national, or regional, share approach to forecast
ing exports provides a less troublesome method than does
detailed analysis of a local area. Under a share approach,
a local area is assumed to maintain its present proportion
of basic activities, relative to the nation (for national
market activities, e.g., manufacturing), or the region
(for regional market activities, e.g., retail trade), for
the time span covered by the forecast.
The methodology of the share approach usually in
volves obtaining previously made national and regional
forecasts of activity levels for each basic industry
located in a local area. There are often several national
forecasts available for various types of industries which
have been made by competent research organizations. In
the case of national market basic industries, determination
is first made of the share that a metropolitan area can
expect of projected national activity levels for selected
industries. Then, determination is made of a community's
expected share of the metropolitan area's national share.
With regional market basic industries, a community's
expected share would be determined directly from projected
metropolitan area activity levels.
The share approach allows a check on consistency in
that it sees local areas as parts of a total, and assigns
■ certain shares to each area in the future. In this way,
all local areas cannot view themselves as gaining an
ever larger share of a national or regional total.
But, Harris reminds us that,
The use of this method expresses the fact
that we have not systematized our knowledge
of the basic forces governing the shifts
of population, economic activity, and
resource exploitation...We are not,...,
in possession of a full explanation of
these past movements nor do we know how
they will be affected in the future by
rising levels of income; increasing
mobility of factors; increasing trans
portability of certain materials and
products and increasing market orientation
of others...49
If a projection of national income is accepted, and an
assumption made that regional income will grow at the same
rate (both total and per capita), a proportionality
approach is being followed. Maintenance of proportional
shares of basic activity by local areas, assumes away
dynamic changes in local economies. Regional economic
differences are not specifically analyzed.
The proportionality assumption of the share approach
to export forecasting can be modified by infusion of some
location analysis into the forecast. Detailed location
factors should be included affecting each major local
export industry. In this manner, some insight can be
obtained into specific locational factors which may
4®Harris, Regional Science Association Papers,
(1956), 246.
influence future growth of existing and potential export
activities in an area.
A good example of competent economic base forecast
ing is offered by the economic base study of Denver, con
ducted by the Denver Planning Office in the early 1950's.50
Two basic purposes of the Denver study were: (1) to
determine the quantity and quality of future commercial
and manufacturing land needs; and (2) to estimate future
Denver employment and population.
Resources, industries, and markets were all measured
in terms of their current supporting roles in the Denver
economy. Basic activity was analyzed in terms of its
growth potential, and a projection of basic activity
levels was made to 1970.
As Kurtz says,
The basic character of each major segment of
the economy was measured, and its prospects
evaluated, in terms of past history of the
industry and of the data provided from the
interviews, questionnaires, and letters from
home offices of branch firms. This intensive
analysis was limited to the basic employment
factor since other phases of the population
projection could be developed accordinq to
formula.51
Estimation was made of the population size which could be
50Maxine Kurtz, "Denver Economic Survey,"
Journal of the American Institute of Planners,
(Winter, 1953), 62-77.
51lbid., 66-68.
! 113
supported by the projected economic base in 1970. At
the time of the survey the following ratios prevailed in
Denver:
Basic employment: Nonbasic employment = 100 : 156
Total employment: Population = 100 : 140
Basic employment: Population = 100 : 660
In the projections of basic activity levels,
Denver's export markets were broken down into three geo
graphical areas. Primary, secondary, and tertiary market
areas were designated, and Denver's basic industries were
classified according to which of these markets they served.
Each market area extended farther out than the preceding
one. The primary market area included the Denver metro
politan area, the secondary market area encompassed the
Rocky Mountain region, while the tertiary market area
embraced the entire nation and foreign countries. It
could then be discerned whether export growth was emanating
from regional or national markets. In the case of Denver,
growth was observed to derive mainly from regional exports.
The following basic, nonbasic, and total employment
and population estimates were calculated for Denver:
Year
1940
1950
1970
The 1970 forecasts were based on a long-run trend line
(ruling out cyclical analysis) extrapolated from past
Basic
Employment
52.000
86.000
124,000
Nonbasic
Employment
88,000
132.000
191.000
Total
Employment
140, 000
218,000
315,000
Total
Population
408.000
564.000
819.000
114
observed trends, and, on the present (i.e., at the time
of the study) basic: nonbasic: total employment:
population relationships or ratios.
The conclusion was that "An economic base study is
useful in forecasting community economic l e v e l s . " 5 2 But,
specifically, what does an economic base study contribute
to a local economic forecast? Primarily, it focuses on
the volatile sectors separately. Attention is concentrat
ed on export and local investment activity as the deter
mining factors in local income creation. In addition,
economic base analysis insures consistency among the
sectors. That is, there can be no high forecast for the
local consumption sector unless exports are expected to
do well. Even if forecasts are wrong, it is possible to
go back and locate places of error specifically and
analytically, because each industry and sector of a local
economy is analyzed in an economic base study.
Limitations of Economic Base Analysis
Macro-analysis.— Economic base models divide all
economic activity in an area into two aggregative categor
ies— activity serving export markets and activity serving
local markets. The economic base concept provides
52Tiebout, The Community Economic Base Study, 81.
"...a new type of economic theory, namely a macro-micro
economics, the economics of the community."53 ^ macro-
analytical process is being applied to micro-geographic
areas.
Palmer says that,
...the local multiplier, as defined by the
derivative-basic ratio, is a summary measure
of the causal connection, or supporting
capacity, of a local community's economy.
Thus it embraces the totality of the
economic functioning of a local community,
showing the causal and functional relation
ship between the basic activity and the
derivative activity which together make
up the whole economy of the community.54
The economic base ratio is determined by many institution
al and economic factors, so that in the long-run, varia
tions in basic activities cause parallel fluctuations in
nonbasic activities. "In principle, therefore, this
approach attempts to summarize in a single measurement
the relationship between externally and internally
generated economic activity."55
An economic base study is essentially a "spaceless"
model of local economic behavior. It abstracts from any
micro-analytical statements about the spatial configura
tion and nature of market areas, except to the extent that
^^Palmer, The Community Economic Base..., 3.
54ibid.. 27.
55sue S.Moyerman and Britton Harris, "The Economics
of the Base Study," Journal of the American Institute of
Planners, (Spring-Summer, 1955), 89.
it defines basic activities as those selling to demand
sources outside a local area.
This gross macro-division of all local activities
into basic and nonbasic categories suppresses many import
ant economic relationships existing in a community. No
thing more than levels of local demand and income are
determined by an economic base model. No precise state
ments are made about the implications that present or
future levels of income may have, in terms of potential
retail expenditures, for local commercial establishments.
Neither can a reasonably exact estimate be made, merely
from income levels, of types, sizes, and locations of
retail establishments required to supply given levels of
demand.
Demand orientation.— An economic base model is
primarily demand oriented. Division of all local activity
is according to the location of its demand sources.
Where, in a geographic sense, do local firms sell their
products? Is income earned from employment inside or
outside a local community?
Tiebout is careful to point out that,
A base study in a sense deals with the
demand side of the economy, that is,
where does it sell its products. In
considering the future of the community
it is also important to study the
supply side. The supply side in
effect deals with the nature of
the local economy as an economic
environment.56
Studies of local supply factors direct attention to
matters such as availability of land sites. This includes
making inventories of a community's assets and liabilities.
The space requirements approach developed in Chapter V
focuses on local retail space needs in meeting various
kinds and levels of consumer demand.
Relationship to other studies.— Local government
planning agencies usually undertake economic base studies
as preliminary background analyses against which more
specific kinds of studies can be judged. Often economic
base studies are made in conjunction with land use
planning studies.
Community planning is concerned with
spatial allocation of activities within
the community. The interest is in
patterns of growth. Patterns imply
groups or aggregates, such as
...retail-commercial, service-
commercial, ...and...residential uses.
The first use of an economic base
study in planning is to help provide
"control totals" for these aggregates.57
The planning process involves allocating totals of
-^Tiebout, The Community Economic Base Study. 18.
57ibid., 17.
various kinds of uses within a given land area. "For
example, what will be the volume of retail activity and
what will its space requirements be? A base study with a
projection can help to answer such questions."5® The
process could be as follows: (1) forecast export income
for year x; (2) via the local income multiplier estimate
local income for year x; (3) estimate land use (space)
requirements in terms of dollars of retail sales per
square foot, in export and local retail establishments;
(4) with the aid of other studies, e.g., trade area and
input requirements analyses, allocate this space within
the community; and (5) see if these space requirements
can be met with available land sites and under existing
zoning and land use regulations.
What is the most useful level of aggregation in
economic base studies? It depends on the nature of the
local economy. "Except in extremely small communities,
it is simply not feasible to consider the details of each
and every economic unit. And yet, the loss of detail can
hurt."59
Harris tells us, though, that
...the problem of industry detail cannot
be avoided, since different types of
commercial and industrial activity
58Ibid.. 17.
59Ibid.. 28.
require land with different character
istics and different locations within
the metropolitan structure.60
In terras of input-output coefficient tables at the
metropolitan level, Harris can "...see that projecting
land-use inputs for the full range of metropolitan
activities requires a much more complicated and difficult
breakdown.Long-range projection of input coefficients,
e.g., technical spatial coefficients indicating retail
sales space input requirements, in terms of square feet of
net sales space, for different expenditure levels, requires
knowledge of technical and productivity change in retail
trade.
An alternative suggestion to traditional input-
output analysis, for retail trade, is to use some form of
correlation analysis based on past trends, which would
implicitly project expenditure patterns and technical
relations, which have in the past influenced space re
quirements for different levels of operation (in terms of
gross sales) of retail establishments of different types.
For this kind of analysis in a small area an exhaustive
input-output matrix is not necessary.
^Harris, Regional Science Association Papers,
(1956), 241.
61Ibid., 241.
120
Time-space limitations.— Economic base analysis is
definitely less applicable to larger regions, and less
accurate in longer-run predictions. The quantitative
importance of exports as an explanatory factor in regional
income determination depends, in part, on the size of the
region under study. In fact,
...a region may grow with exports at a
constant level, if internal autonomous
activities (business investment, govern
ment expenditures, residential construction,
etc.) are on the upswing. The larger the
region, the more the dynamic forces causing
income change will be found inside its
borders.62
The significant consideration is not which boundries are
chosen but the effects of this choice on the variables
being analyzed. When there is awareness of direction of
change in variables as a function of regional boundries,
the question of boundries per se becomes less significant.
The fact is, that if a larger region is under study,
an investment multiplier of some sort may be the relevant
analytical tool, rather than a basic income multiplier.
For present considerations, economic base analysis is
being discussed mainly in relation to its use in study of
small or medium sized cities or communities located within
62c.M. Tiebout, "Exports and Regional Economic
Growth," Journal of Political Economy,
(April, 1956), 161.
or adjacent to urbanized regions.
Generally speaking, multiplier effects are of the
same order of magnitude for small as for large areas,
qualified by adjustments for differences in import leakage
from the income stream. What varies with the size of an
area is the degree of self-sufficiency, not the multiplier
effects. It would be false to assume that all local
activity is geared to the needs of the export base. More
and more activities become internal as an area is enlarged,
and the export base becomes a progressively smaller
proportion of the whole. Multiplier effects, however,
depend on the functional relations between sectors of a
local economy as well as on a division between local and
nonlocal economic activity. These functional relations
are not affected by widening the borders of a local area,
but the base ratio is. Doubling the export base in a
small city may (after a time-lag) double total income, but
it would not do the same for the country.
For a number of reasons economic base predictions
are most valid in the short-run. Isard sums up the fail
ings inherent in economic base forecasts with his follow
ing statement:
The use of a multiplier to estimate the
results of future changes in basic
activities of a city or region is an
attempt at prediction. This prediction
is based on past or present data and is
122
subject to error because of future
qualitative changes in social, tech
nological, and economic conditions,
the influences of many of which cannot
be crudely, let alone precisely,
estimated.
It might be pointed out, of course, that these unpredict
able future changes plague any kind of forecast. The
problem is, that if there is too large an amount of un
predictable future change, the value of a sophisticated
analytical prediction device is obviated. Simple extra
polation of past trends of total activity levels would be
useful and accurate as economic base predictions in such
a case.
The accuracy of an economic base forecast lies with
the dependability of the two elements in a prediction of
future total income levels. First, the forecast of export
activity must be accurate. Secondly, changes in the size
of the multiplier must be predicted with some certainty.
Since export forecasts are usually derived from other
sources, an assumption is made that they are generally
dependable in terms of accuracy. As far as the economic
base model is concerned, though, questions of the accuracy
of short-term and long-term forecasts hinge on an ability
to anticipate future changes in the size of the local
multiplier.
63walter Isard, Methods of Regional Analysis.
(Cambridge, Mass.: The M.I.T. Press, 1960), p.199.
J
123
Predictive value: local consumption patterns
and economic structure.— According to economic base
theory, when any change in basic activity occurs, either
in terms of income, employment, or output, a change in
nonbasic activity is expected to occur. If a certain
ratio of basic nonbasic activity could be assumed normal
for a community of given size and economic structure, then
a predictable multiplier would exist. The problem is that
cities change in size and economic structure through time.
The result is that local income multipliers cannot be
expected to remain constant over long periods of time.
There is some degree of dynamic instability in local
income multipliers.
It is behavioral patterns of local residents which
are most significant in determining the size of a local
income multiplier, and changes in its size through time.
Specifically, geographic spending patterns of residents
are of prime concern. Import leakages are the most
important cause of reduction in local multiplier size
in small areas.
There are a number of reasons for local multipliers
being different in size. Magnitude of consumption expend
iture depends on income level. Empirical data indicate
that the higher the income level of local residents, the
! 124
lower the rate of local purchase. The more diversified
are consumer tastes, the lower is the rate of local
purchase. Newer residents tend to spend less locally.
Geographical location of a city vis-a-vis neighboring
urbanized areas affects rate of local purchase. Small
cities within or next to large metropolitan centers tend
to lose local customers to central shopping areas.
Responsiveness of local business enterprise in meeting
new local demands affects the level of local expenditure.
If new retail facilities are not provided, it is difficult
for rates of local purchase to rise. The extent to which
local business firms meet their payrolls to nonresidents,
and import material, causes a depressive effect on local
multiplier size.
In the local sector there is a need for adequate
anticipations data on probable changing consumer expend
iture patterns in the future, as well as on responsive
ness of local retail trade to increases in demand.64 it
must be remembered that as income levels rise, local eco
nomic structure changes too. With larger local markets,
thresholds, may be reached in terms of economies of scale
which permit new types of retail firms to establish them-
^Friedly, "An Economic Base and Income Multiplier
Study...,"
Chapter 4
selves locally (i.e., specialty shops, etc.) thus raising
the rate of local retail purchase.
No simple projections of future local income levels
can be made using a present multiplier. As Ferguson says,
in reference to the results of an initial increase in
expor t income,
...the increase in income will lead to
a change in the marginal propensity to
consume and probably in the marginal
propensity to import. That is, the
consumption function is not linear?
there is neither theoretical nor
empirical evidence to support the
assumption that it is...thus the
income generated by the initial
change in basic activity will not
correspond to the change predicted
by the least-squares regression
method.65
In other words, it is a spurious effort to make simple
straight line projections of local income based on recent
income multipliers. More sophisticated assumptions must
be made concerning future changes in local consumption
habits. Even if a rigorous model accounting for these
expected changes cannot be constructed, rough approxima-
i tions would add enormously to the accuracy of local
income projections.
There is much regularity and predictability in these
65Ralph W. Pfouts (ed.),
The Techniques of Urban Economic Analysis,
(West Trenton, N.J.: Chandler-Davis Publishing Co.,
1960), pp.337,338.
local and temporal variations in multiplier size. From
the import side, it is true that the propensity to import
changes with time because of structural and behavioral
change in an area.
Properly projected and taking these factors
into consideration, the basic-service ratio
based on employment should enable us to
translate any estimate of export employ
ment into an estimate of total employment
and hence of total population.66
A number of generalizations can be made concerning
variations in local multiplier size. Some were noted
above in relation to changes in local spending habits over
time. In relation to spatial variation, Tiebout's con
clusion is that,
Since consumers' propensities to spend locally
vary in the same manner, the following general
relation emerges: increased size and geographic
isolation tend to raise the local consumption/
total ratio. Or, stated another way, the non-
basic/total ratio uses with city size and
isolation.67
Tiebout's more general conclusion
...is that the local consumption/total ratio,
or more broadly interpreted the 'base ratio',
is not subject to willy-nilly fluctuations.
Positively stated, it is a meaningful, useful
ratio.68
In addition to internal causes of variation in
multiplier size, there are external forces which may
66Harris, Regional Science Association Papers,
(1956), 248.
67Tiebout, The Community Economic Base Study. 69.
68ibid.. 70.
) 127
distort the picture of the economic base at any one time.
National business cycles in durable producer and consumer
goods affect the basic sector most heavily. Since basic
! elements in a local economy tend to fluctuate with greater
amplitude than local elements, the base ratio may increase
during cyclical downswing and decrease during cyclical
upswing. The particular ratio calculated at any given
time depends on the state of the business cycle.
The overall conclusion is that regardless of the
i
existence of variations in the relative importance of the
elements of the base ratio at different times and/or
places, sufficient information can be obtained to antici
pate most important probable variations in the sizes of
base ratio elements. Information can be generated about
dynamic behavior of consumer expenditures from trade area
studies. Roterus and Calef feel that,
Hoyt is absolutely right in his implication
that the use of a multiplier factor (which
grows out of the basic-nonbasic idea) will
give a more rational prediction of future
population or future space requirements
than would be made without such a factor.
However, it is perfectly clear that the
accuracy of his population predictions
rests ultimately on the accuracy of his
analytical guesses about (a) the future
structure and size of industry in the
community and (b) the future basic-
nonbasic ratio.69
6 9 V i c t o r Roterus and Wesley Calef, "Notes on the
i Basic-Nonbasic Employment Ratio", Economic Geography,
(January, 1955), 19.
So, even though the economic base multiplier is a rational
and superior tool for local area prediction, greater
accuracy must be achieved in anticipating changes in size
of the local multiplier, i.e., changes in underlying
patterns of consumer behavior.
Inadequate information.— For purposes of local land-
use planning and space allocation, the main deficiency of
economic base analysis lies with its inability to generate
sufficient information. Its demand-and macro-orientations
reveal nothing of retail space requirements to serve in
dicated demand levels. In addition, there is little in
formation generated on specific expenditure patterns and
propensities which can be translated into accurate
potential future retail sales levels.
There are, of course, means by which we
can simplify such a model and modify it
for the uses which we have in mind...
Perhaps the most useful type of model...
is an input-output model for the
metropolitan economy,...70
The additions to economic base analysis offered in
Chapter VI of this study, attempt to fill the information
gap of traditional "community economic studies" which use
only an economic base approach. Efficient retail space
70Harris, Reqional Science Association Papers,
244,245.
supply standards can be approximated without employing
the complete input-output approach. Micro-analytic study
can be made of local (and some basic) expenditures, to
more precisely determine the character and location of
local demand and its space requirements.
The micro-demand and space analytical deficiencies
of an economic base study are remedied by fusing its
analytical structures with those of trade area analysis
and retail space requirements analysis. These latter
approaches are presented in the following two chapters.
CHAPTER IV
TRADE AREA ANALYSIS
This chapter is intended to provide an overview of
the development, content, and deficiencies of trade area
and market potential analysis. Trade area studies consti
tute the micro-demand analytical component of the model
presented in Chapter VI.
Historic Review
The theory of market and trading areas.— The type of
market under consideration in this study is spatial in
nature. It is the geographic extent of territory served
by various types of retail activity. The geographical
location, or spatial distribution, of demand sources for
various kinds of retail commodities is the basic factor in
market and trade area analysis.
A market area is a spatially defined area of demand,
say, for retail goods and services of all types. It may
include existing and/or potential purchasers of goods and
services within clearly defined geographical limits.
A market area may, or may not, be spatially coinci
dent with one or several trading areas of local shopping
centers or other separate retail stores. Whereas a market
area is more generally defined as an area of demand for
retail commodities, a trading (or trade) area defines
the geographic area from which a particular retail store,
or grouping of stores, draws most of its customers.
August Loesch was probably the first theorist to
provide exhaustive analysis and explanation of the nature
of market areas and their arrangement over space. Loesch
initially postulates a broad, homogeneous plain with
uniform transport features in all directions, an even
scatter of production inputs sufficient for any desired
output, and a uniform distribution of population with a
uniform set of tastes and preferences.
In isolation, the ideal market form would be a
circle, according to Loesch. The average demand in a
small circle is greater than that of any polygon of equal
area. But when several market areas are considered in
juxtaposition to one another, circles fail to exhaust
the entire area of a plain. Because circles leave empty
areas between themselves, when they are tangent over an
area, demand per unit of the entire area is greater in
the case of a hexagon than in the case of either a circle,
square, or equilateral triangle.
In other words, among all the possibilities
of realizing the same total demand, the
132
most land is required with a triangle,
and the least with a regular hexagon.1
A retailer selling products would find his market
area approaching circular form, if he were a monopolist.
However... the force of competition, by
eliminating all excess profits, not only
will contract the market area of the
original producer but also will transform
the circular shape of the market area into
a hexagon. The hexagon is the ideal economic
form of market area, it is maintained.2
Since a hexagon deviates least from a circle, consumer
travel costs are minimized in supplying a given demand,
or, conversely, the demand of the population of a given
area is maximized.
It is because competition cuts down the size of a
retailer's market area through elimination of undeserved
profits that empty spaces are removed and the smallest
possible hexagon approached. A smaller circle or hexagon
yields a greater demand per square mile than does a larger
circle or hexagon. Loesch says that,
...per unit of area of the market region,
the demand of the small circle not that
of the large circle, is greatest because
in the latter case the average for the
area is reduced by the small demand near
the limits for shipment.3
1August Loesch, The Economics of Location,
(New Haven: Yale University Press, 1954), p.112.
2Walter Isard, Location and Space-Economy,
(Cambridge: The M.I.T. Press, 1956), p.44.
^Loesch, The Economics..., 112.
In a trade area, say for a regional shopping center,
it should be expected that near the periphery a much low
er proportion of families will be visiting the shopping
center than closer in toward the center of the trade area
(by moving the periphery farther out from the center)
yields a proportionately smaller increase in total expend
itures made at the shopping center. As the outer bounds
of a trade area are approached, costs to consumers in
reaching a shopping center in time and travel expense,
continually rise and diminish the frequency of their trips
to it. Finally, a point is reached where virtually no one
can be considered a patron of the shopping center. Com
petition also reduces, with increasing force, the number
of consumers visiting a center as distance increases.
Loesch observes further, that in these competitive
situations "... the advantage of the hexagon becomes
greater, the more elastic the demand at the boundries
of the region, and smaller the more inelastic is this
demand.With an elastic demand curve, demand falls
relatively rapidly with increasing distance from a
shopping center— a fall that becomes increasingly greater
to the extent that larger portions of the market area,
e.g., corners of triangles, lie relatively far from the
shopping center.
4Ibid., 113.
! 134
Since retail trade areas are assumed to have rather
elastic demand toward their peripheries, because of vigor
ous competition in urbanized areas, hexagonal shapes
would seem to offer an ideal market form for retail goods
and services, other things being equal.
Theoretical versus actual market area shapes
and sizes.— Under the varied physical, political,
social, economic, and other conditions existing in the
real world, market and trading areas assume shapes and
achieve sizes at variance with theoretically ideal
conditions. Applebaum and Cohen note that,
Theoretically, the hexagonal, triangular,
or square-shaped trading areas offer the
best possibilities for covering a market
most thoroughly. In practice, however,
street patterns play the most significant
role in moulding such shapes, and frequently
run counter to what would be theoretically
desirable.5
Even though physical features are the most obvious
factors in determining actual trade area configurations,
the density and income level of consuming units in the
surrounding area affects size and shape. The greater is
residential density, and the higher are income levels,
j the smaller in geographic extent can a trading area be
^William Applebaum and Saul B. Cohen,
Store Trading Areas in a Changing Market— Part I,"
Journal of Retailing, (Fall, 1961), 22,23.
to support a shopping center of given size and diversity.
In a metropolitan area, because of the relatively high
population density nearer the center, Isard points out
that "...the size of a market area in square kilometers
necessary to generate sufficient demand for a commodity
to justify production is much smaller at the core than
at a great distance from the core."6
Uneveness of consumer spatial distribution, in
equalities in effective purchasing power, and differences
of taste among consumers distort the "natural" circular
or hexagonal regularity of trade areas. Isard says,
However, the essential condition that
the market boundry be a locus of
consumers who are just willing to
pay for the first unit of product
a price which is equal to the factory
price plus transport cost to the point
of consumption still obtains.7
In terms of retail shopping centers, where consumers
directly incur travel costs to make purchases, the market
boundry would ideally consist of a locus of consumers who
are just willing to travel to a particular shopping center
to make retail purchases on some regular basis.
Retail market structure and competition also affect
the extent and shape of shopping center trading areas.
^Isard, Location..., 271.
7Ibid., 145,146.
i Xn the language of central place theory,® purchasing
power thresholds and geographic ranges can be established
I for individual retail goods or classes of goods. The
threshold of a particular retail commodity is the minimum
amount of purchasing power required to support its supply
! from a specified place. The range of a retail good equals
the trading area of a shopping center for that good. The
lower limit of this trade area is threshold purchasing
power for the retail good, while the upper (or outer)
limit is the distance beyond which the good cannot be
sold because demand for it is zero, i.e., the ideal limit;
or a competing shopping center is more convenient to
consumers, i.e., the real limit.
Each shopping center attempts to extend the limits
of and increase its control over the retail trading area
surrounding it. Technical and financial conditions of
retail supply may largely determine the threshold level
for establishing a retail store or center, initially.
Once in operation, however, shopping centers
...typically are able to influence price,
to discriminate among consumers, to induce
consumers by advertising, price cuts, or
other means to shift their allegiance from
Q
°Central place theory views economic and social
functions as occuring at various points in space, and
arranges them in a hierarchical manner. Shopping centers,
for instance, could be arranged in ascending order in a
retail hierarchy as their trading functions become more
geographically extensive.
r
137
a competitor...All these types of mono
polistic and oligopolistic behavior tend
to invalidate the simple, clear-cut boundry
lines customarily depicted. At best boundry
lines are blurred and tend to degenerate
into overlapping zones.9
Monopolistic control may enable retail competitors
to extend, at various places and times, the range over
which they can sell their products to consumers. Trade
areas, then, are not necessarily well-defined, neatly
outlined areas on a map. As Loesch says,
When we drop the assumption of a uniform
plain, the simple market and supply areas
assume irregular shapes and sizes because
of the natural, personal, and political
differences already mentioned. Further
more, market areas for the same product
overlap, since it is seldom finished in
exactly the same way, and each area is
fringed in consequence.
Division of trade areas between retail competitors.-
There are numerous factors limiting the geographic area
that a retail store or center can effectively serve. As
one moves farther away from a trade area's center, in
terms of both physical distance and travel time, con
sumers become increasingly disinterested in making frequ
ent trips from residence to shopping center for the
I purpose of retail purchase. Most of the rising costs of
^Isard, Location..., 264.
lOLoesch, The Economics..., 215.
138
doing business at ever greater distances from a retail
location result from the "friction of distance."
Some of the costs of overcoming distance are borne
by consumers, but some are also borne by retailers. Time
and travel costs incurred by consumers increase with
distance. On the other hand, selling and delivery costs
incurred by retailers rise with distance of consumers.
The cost of sales promotion rises with distance because
consumers usually become more thinly distributed with
increased distance from a retail store or center. The
cost of delivering goods purchased during prior shopping
trips limits the area that can be served without making
additional delivery charges and thereby raising purchase
costs of consumers. Also, business risks rise with
increased distance because of the difficulty in getting
reliable information on the credit reliability and needs
of purchasers.
In the process of overcoming spatial friction,
consumers tend to become increasingly indifferent toward
visiting a particular shopping center, the farther away
from it they are. For this reason, retail centers be
come less and less competitive as the peripheries of
their trade areas are approached. That is, an increas
ingly smaller proportion of the consuming units in the
distant portions of a trade area visit a shopping center,
than in the closer in portions. Competing shopping
centers or individual retail stores are attracting con
sumers ever more effectively as distance from the subject
shopping center increases.
Isard makes a formal statement of the condition of
consumer indifference which defines a trading area boundry
He sets forth the following equation expressing a trade
area boundry condition:
F
rs + ^irisi = T
i = A
where r is the transport rate per unit of product (or
consumer travel time and cost) to the boundry line along
any radius; s is the radius of the circle which defines
the boundry line; A, ...F are raw material inputs; b^ is
a constant coefficient of the number of units of raw
material i used per unit of product; r^ is the transport
rate on a unit of raw material input i; T is PQ minus p;
PQ is the maximum price a consumer is willing to pay for
the product; and p is the per unit production cost (ex
cluding transport or travel time costs).
According to Isard, this boundry condition equation
...states that at the market boundry the
sum of the transport costs on the unit
product and on the raw materials required
to yield the unit product is just equal
to the difference between unit costs of
production and the maximum price the
consumer is willing to pay.I1
This would imply that consumers located on a retail trade
area boundary line, between similar shopping centers which
sell their goods at competitive prices, are indifferent
as to which shopping center they select, because the
money and time costs incurred in traveling the distances
from their homes to either shopping center are equal
(i.e., both the objectively measurable costs and any non-
quantifiable psychic costs).
The trade area boundry described above should not
be interpreted in such a manner that all consumers located
within its confines are assumed to frequent only the
subject shopping center. The final boundary line of a
retail trading area is usually defined as one beyond which
only a negligible number of consumers come to the subject
center. As one approaches the final boundary line from
within the trade area, concentric zones are passed through
which contain progressively smaller proportions of
resident consumers who regularly purchase at the subject
center. The progressively effective competition of
surrounding shopping centers and retail stores is discern-
able as the frequency of consumer visits to the subject
center declines with increases in distance from it.
H-Isard, Location.... 231, 232.
More will be said later about accurately defining
the extent of competing shopping centers' trade areas.
For the present it is sufficient to note that a retail
trade area boundry line should consist of a locus of
points of equal driving time-distance cost between com
peting shopping centers. In the case where any other
"...boundry exists, it becomes economic for certain
consumers to shift their buying from one producer to
another."12 in effect, consumers are thereby substituting
travel expenditures involved in the purchase of goods
from one retailer for travel expenditures involved in the
purchase of goods from a second retailer, and they should
continue to do so until a boundry comprising points of
equal driving time-distance costs is attained.
The gravity or market potential concept.— In re
tailing, economies of large scale distribution of goods
and services at one place favor concentration of retail
activity at one location in space, while consumer trans-
i portation costs favor dispersion of retail activity over
space. Some analytical framework is required which con
ceptualizes the forces operating for and against agglomera-
| tion of retail activity at one location. A gravity or
I potential model provides just such an analytical framework
f
- — — - _ _ __ .
12Ibid., 147.
142
within which to study the spatial distribution, and
changes in that distribution, of retail markets and trad
ing areas.
A gravity model postulates that an attractive force
of interaction between two areas of human activity is
created by the population masses of the two areas, and a
friction against interaction is created by intervening
space over which the interaction occurs. This is referred
to generally as an "interactance hypothesis."
For interaction to take place individuals must be
in some form of communication with one another. Generally,
it is assumed that an individual, as a unit in a large
group, has the same influence on interaction as any other
unit. The probable frequency of interaction generated by
an individual at a given location is inversely proportion
al to the difficulty of reaching that location. The
friction against this movement over space is directly
proportional to the intervening distance.
The gravity model concept is based on Carey's
original nineteenth century formulation of the law of
molecular gravitation as a law of social interaction.13
' Even though it has been observed that people behave
i differently in groups from individually, an assumption
^3See H.C. Carey, Principles of Social Science,
(Philadelphia: J.B.L. Lippincott and Co., 1858).
remains that human beings can make decisions which mole
cules (presumably) cannot do. Nonetheless, reasonable
hypotheses pertaining to consumer behavior in retail
markets can be formulated with the aid of gravity models.
Earlier formulations of gravity models postulated
human interaction as conforming to centrifugal and centri
petal forces in the following manner:
Iij = f_(PL_p-it
f (Dij)
iMj = -
Dij
where iMj is migration from source j to center of absorp
tion i, f (Pi) indicates that migration is some function
of population at i, Pi, Pj are the populations of i and j,
and Dij is the intervening distance between i and j. This
formulation^ states that interaction between a source
and a center of absorption (Iij) occurs in direct pro
portion with the size of population at the center, and
inversely with the distance between i and j.
In the 1920's, E.C. Young^ added greater sophisti
cation to the gravity model formulation. His mathematical
14see E.G. Ravenstein, "The Laws of Migration,"
Journal of the Royal Statistical Society.
(June, 1885), 167-235.
l5See E.C. Young, The Movement of Farm Population.
(Ithaca: Cornell University, Agricultural Experiment
Station Bulletin No. 426, 1924).
144
statement was that
xMj = k -
Di
where Zi is the force of attraction at i, and k is a
constant of proportionality. Young's formula states that
the relative volume of migration to a given destination
from each of several sources varies directly with the
"force of attraction"of the destination, and inversely
with the square of the intervening distances. Young felt
that spatial interaction was some exponential inverse
function of distance. The numerical value that should be
assigned to this distance exponent, for various types of
spatial interaction, has been the subject of much contro
versy, as will be seen later.
Gravity model formulation can be generalized to
account for the spatial interaction of an area with any
number of other areas in, say, a metropolitan area. This
more generalized gravity formula would appear as
Ei = k (—— + + . . .+ +. . .+
^Dil Di2 Di3 Dij DinJ
or
n__
Ei = k ^ . —- U
j = 1
In this case, the total energy of interaction of a given
area i (Ei) equals the sum of the energy of its inter—
action with each of n other areas comprising a metro
politan area.
When applied to the study of retail trade areas,
gravity models are often referred to as market potential
models. They can be formulated in terms of describing
potential interaction in the following manner:
iVj = k p.l
Dij
where iV is total population potential at i. At i, the
potential influence with respect to an individual at i,
from population at j, is greater as population at j is
larger and less as distance between i and j increases.
Voorhees says that,
...accessibility is defined as the potential
of opportunities for interaction...it is a
measure of the intensity of the possibility
of interaction rather than just a measure of
the ease of interaction...accessibility is
a generalization of the population-over-
distance relationship or "population
potential" concept...accessibility is a
measurement of the spatial distribution
of activities about a point, adjusted
for the ability and desire of people
or firms to overcome spatial separation.16
In the case of a retail shopping center, this type
of model could measure the access that a proposed center
might have to all possible customers in a surrounding area.
■^Alan M. Voorhees, "Land Use and Traffic Models,"
Journal of the American Institute of Planners,
(May, 1959), 73.
The actual number of customers shopping at the center
would be diminished by various factors, e.g., competition
from other shopping centers.
Retail market potential.— One of the earliest
applications of the gravity or market potential concept
to a study of retail trade areas was made by William J.
17
Reilly. Reilly's "law of retail gravitation" states
that a town attracts retail trade from an individual in
its surrounding territory in direct proportion to its
population size and in inverse proportion to the square
of an individual's distance from the town.
A hypothetical example will help to indicate the
manner in which Reilly's law determines the proportionate
retail trade shares that two towns obtain from an inter
mediate trade area. In its usual mathematical formula
tion, Reilly's law states that
where Ba and Bb are the proportions of trade attracted
from an intermediate area by town a or b, Pa and Pb are
the towns' respective populations, and Da or Db is the
distance from the intermediate area to town a or b. The
Ba
Bb
^See W.J. Reilly, Methods for the Study of
Retail Relationships, (Austin: University of Texas,
Bureau of Business Research, Monograph No. 4, 1929).
147
place where Ba equals Bb is a point of equilibrium and
represents the 0.5 probability position between towns
a and b. Bb is the breaking point (in the intermediate
area) between towns a and b in terms of miles (or driving
time) from town b.
Suppose that
This means that town a attracts 0.915 times as much
retail trade from the intermediate area as town b.
Relative percentages are derived by dividing 0.915 by
1.915. This yields 48 per cent of the retail trade to
town a and 52 per cent to town b. These percentage values
are valid, of course, only to the extent that the factors,
and their postulated relationships, in the model represent
the actual factors controlling consumer decisions.
The boundary line of a trade area determined by
Reilly's formula is a locus of points where consumers
probably will divide their trade between two competing
towns on a 50-50 basis.
As Converse says,
The boundary of its area is the 50-
per cent line. People living on this
line divide their trade equally between
the two competing trade centers. People
inside the area trade more in the trading
center than they do in competing trading
centers. In other words, the trade
moving in opposite directions across
the boundary line is equal.18
Reilly's formula implies, then, that there is a locus of
equilibrium points between competing centers where com
petitive influence is equal, and, that this locus of
points divides the trading areas of competing retail
shopping centers. In practice, any line drawn around a
shopping center on the basis of Reilly's formula would
represent an ameoba-like shape connecting all points
where business is divided on a 50-50 basis between
centers.
There have been numerous refinements of Reilly's
original formulation of the law of retail gravitation,
as it applied to determining the trade areas of competing
towns. The same principle is now applied to the analysis
of retail shopping center trade areas within the context
of a larger metropolitan area. Square feet of selling
space in retail centers and driving times from various
neighborhoods to the centers have been substituted for
population and miles. Different distance exponents are
sometimes used in place of Reilly's squaring of the
distance factor.
l®Paul D. Converse, Harvey W. Huegy, and
Robert V. Mitchell, Elements of Marketing,
(Englewood Cliffs, N.Y.: Prentice-Hall, Inc., 1958),
p. 30.
149
The probability approach to retail market potential
analysis.— There is good reason to seriously
question some of the underlying assumptions of the Reilly
approach to retail market potential determination. To
begin with, it is assumed that, other things being equal,
consumers will shop at the biggest retail facility to
which they have easiest access. It is supposed that
consumers make definite choices as to a particular center
which meets their needs, and then are "attracted" largely
to that center alone. This assumes consumers will always
choose one particular (most desirable) alternative with
probability 1. All other alternative retail centers
have zero utility to consumers who have made a selection.
It is supposed that,
Since customers bear the burden of costs
of movement— economic, temporal, or
psychic— to the retail center, the actual
locations of the retail centers are
influenced by the intricate patterns of
consumer movements for retail goods.
Generally, there is a desire to minimize
these costs of movement on the part of
the consumer. This overall tendency
has persuaded market analysts to assume
that a consumer, confronted with a choice
among several alternative shopping centers,
will inexorably choose the nearest center.19
^9T. R. Lakshmanan and Walter G. Hansen,
"A Retail Market Potential Model,"
Journal of the American Institute of Planners,
(May, 1965), 135.
On the basis of this type of assumption as to consumer
behavior, a trade area could be delineated which encloses
consumer expenditures in a closed market area around a
shopping center or centers.
Rather than a situation in which consumers select
among alternatives,
Empirical studies have demonstrated,
that there is, instead, a continium
of market orientation of consumers
to shopping centers...shoppers engage
in an information-seeking process,
which, over a period of time, tends
to attract them to different centers
in some constant proportion.20
When a consumer is faced with, say, four shopping center
choices, Huff feels that he perceives relative utilities
associated with each shopping center. Instead of visiting
only one center, a consumer might view the relative utili
ties associated with each of these four alternatives in
proportion 1 : 2 : 3 : 4. In this choice situation, a
consumer would tend to choose the last shopping center
about 4/ (1 + 2 + 3 + 4) =40 per cent of the time. Huff
concludes that
... if a consumer is confronted with the
same choice situation a number of times,
he will tend to choose among such alter
natives in some constant proportion:
hence, a 'relative utility' can be
assigned to each.21
2°Ibid.
^David L. Huff, Determination of Intra-Urban Retail
Trade Areas, (Los Angeles: U.C.L.A., Real Estate
kesearch program, 19fo2), p.16.
Empirical evidence indicates that consumer selection
of shopping centers should be expected to conform to some
sort of proportional ranking based on rules of probabil
ity. A probability formula might explain spatial inter
action in a metropolitan area thusly,
Tij = k 2L.ll
P
■where Tij is the expected number of trips (according to
the rules of probability) by individuals in i to j, P is
total population in the metropolitan area, Pi and Pj are
the populations of sub-areas i and j, and k is a constant
of proportionality expressing the average number of trips
per capita in the metropolitan area to j (T/P). If Pj is
100,000 and P is 1,000,000, then for the typical individ
ual in i, ten per cent of his trips will be to j, i.e.,
Pj/P.
Rather than attempting to define a breaking point
to delimit a retail trade area, Huff feels that a trading
area should be represented as "...a series of radial
zones in which the proportion of consumers patronizing
a given retail facility varies from zone to z o n e . "22
Instead of having a single fixed boundry circumscribing
market potential, Huff's trade area would appear much
as a topographical contour map, with many equiprobability
22Ibid.. 11.
contours circumscribing a particular retail center.
Each contour line would connect points which define
radial zones in which consumer attendance at the center
had the same or similar probability.
Huff's probability model to determine the market
potential of a retail shopping center permits thorough
and precise micro-analysis of community retail trade
activity. Complete development of Huff's model is as
follows:
where P (Cij) is the probability of a consumer at resi
dential location i traveling to shopping center j, Sj is
the square feet of selling space for a given class of
goods sold by shopping center j, Tij is the travel from
i to j, and a is the parameter estimated empirically to
reflect the effect of travel time on different kinds of
shopping trips.
where E (Cij) is the expected number of consumers from
each i which will visit j, and Ci is the actual number
of consumers residing in each i.
si
. .a
Ci (for all i's)
r
153
E (Ai j) = n ----- • ci fii^ (for all i's)
2 TZ
j = 1 \Ti j /
where E (Aij) is the expected annual sales potential of j
for a given product class, and Bik is the annual amount
budgeted by consumers for product class k.
where E (Sijm) is the expected number of trips from each
i to j for purchases in a given product class during
time period m, and Sm is the average number of trips
occuring during m.
Assume that: (1) the probability of consumer visits
to j from i is 0.8; (2) the number of households in i is
1,000; (3) the average number of trips per month for
clothing purchases from i to j is 2; and (4) the annual
expenditure for clothing from i at j is $1,200. Then,
Sj
Ci . Sm (for all i's)
(.8) (1,000) (2) = 1,600 shopping trips
per month from i to j
and
Si.
Sm (for all i's)
154
where E (Dijm) is expected dollar sales made at j from i
for k during m, AS is the annual number of trips made
for purchases of k, and Bik/AS is the amount spent per
trip. So,
E (pijm) = (.8) (1, 000) (^1a!00) 2 = $80,000 expected
' ' \ / monthly
expenditure for
clothing by all
households from
i at j
From empirical evidence, Huff calculated a distance ex
ponent of 3.191 for clothing trips. This is a fairly
high value and indicates that the friction of distance
is significant for clothing purchase trips. An exponent
value of 2.723 calculated for furniture purchase trips
indicates that intervening distance does not deter con
sumers as much on this kind of shopping trip.
This type of model permits a detailed micro-ana
lysis of an entire retail trade area or areas, and deter
mination of a shopping center's (or a municipality's)
retail market potential. And, more realistically,
"It implies that there is no trade area boundry but a
shopping interaction between all zones, though this may
fall off sharply with distance."2^
2^Lakshmanan and Hansen,
Journal of the American Institute of Planners.
(May, 1965), 136.
Determination of retail trade area potentials.— In
relation to the provision of local retail facilities, the
major purpose of a retail trade area study is to determine
the amount and location of potential consumer demand. The
kind of trade area under discussion must be clearly de
fined because, as Simmons notes, "The term 'trade area'
means many things to many people."24
While Huff's probability contour approach offers a
: more sophisticated way to view the configuration of a
retail trade area, there must be a more definite deter
mination of the geographic area in which consumer shopping
habits will be studied in detail. Also, for a shopping
center consisting of many individual retail stores, there
must be a determination of which stores constitute con
trolling attractive forces for various categories of
consumer purchase. The extent of the trade areas for
these selected stores would represent an overall trade
: area for a shopping center. In similar fashion, the
combined trade areas of two or more shopping centers
would constitute an overall trade area for the municipal-
i
ity within which they were located.
| 24James Simmons, The Changing Pattern of Retail
j Location.
j (Chicago: University of Chicago Press, 1964), p.49.
There are many arbitrary rules as to how a trade
area should be determined and divided up for purposes of
mapping and analysis. A primary trading area can be
defined, which includes that portion of the total trade
area providing the greatest density of customers in rela
tion to total population. This primary trading area might
account for, perhaps, 45 to 75 per cent of a shopping
center's customers. Some researchers feel that the
customer/population ratio in the primary zone should be
at least twice that of adjoining portions of the trade
area. A secondary trading area can also be defined,
including that portion of the total trade area accounting
for 20 to 40 per cent of a center's customers. Finally,
a fringe trading area can be outlined which contains 5
to 15 per cent of a center's customers.
The diminishing "pull" of a shopping center is
apparent as one proceeds from the primary, through the
secondary, and to the fringe portions of a trade area.
This arbitrary division of a trade area permits delinea
tion of specific geographic areas, which can be analyzed
more or less intensively, depending on the customer/
population ratio. Obviously, the primary trading area
has greatest relative significance in terms of retail
sales potential, and should receive the most intense study.
On the other hand, the fringe area may have little signif
icance, and, therefore, should receive only cursory
attention.
Another way of determining a trade area involves
defining its boundary line in such a manner that it in
cludes the entire area from which a shopping center draws
any significant (say, above 5 per cent of the population)
amount of business. Drawn in this way, trade area bounda
ries of competing retail centers will overlap. Trade area
boundary lines could be drawn on the basis of Reilly's
formula, so that the area included is only that from
which a center draws (or should draw) more than 50 per
cent of the business. Drawn in this manner, trade area
boundaries of competing retail centers will not overlap—
they are mutually exclusive.
Mutually exclusive trade areas are not as analytic
ally useful or realistic as overlapping ones. Applebaum
feels that
A correctly defined trade area will
encompass the concentrated population
contributing approximately 85 per cent
of the sales volume of the [department
store) branch unit after it opens.
The remaining 15 per cent will come
from widely scattered areas,...25
Shopping centers often draw relatively large proportion of
2^William Applebaum, Store Location and Development
Studies. (Worcester, Mass.: Clark University Press,
1961), p. 34.
158
their customers from areas which lie within the dominant
range of their competitors. The reason for this is at
least partly explained by Huff's hypothesis of consumer
behavior: that consumers divide their shopping allegiance
among competing retail stores and centers in an urban area
in some constant proportion; and that they do not always
exclusively choose the nearest shopping center in terms
of time-distance.
In spite of the significant amount of overlap of
retail trade areas in an urban area, the proportion of
the population regularly shopping at a retail center
rapidly diminishes as distance (or driving time) between
center and consumer increases. In fact, the amount of
trade area overlap occurring depends on the minimum pro
portion of population shopping at the center, which is
considered analytically significant. Usually, a trade
area boundary would not be extended beyond the point where
the center was drawing less than 5 per cent of the popula
tion. Generally speaking, an analyst^". .. sets the limits
of the trade area at the point where he believes business
volume coming to the center from the outer limits of the
trade area will be so insignificant as to be immeasurable.1 1
The trade area, then, is still delineated within the
^Richard l . Nelson, The Selection of Retail
Locations, (New York: F.W. Dodge Corp., 1958), p. 189.
I
limits of probability, i.e., it is not extended beyond
the probable drawing range of an existing or proposed
shopping center. Trade area size can be described in
terms of driving time isochrons, i.e., lines connecting
points of equal driving time around a shopping center.
The trade area is divided into concentric zones, each one
representing the market range of one or a group of stores
located in the center. The primary trading area might lie
within the 15 minute isochron, the secondary trading area
within the 20 minute isochron, and so on, out to the
driving time limit of 25 or 30 minutes, beyond which
competition and other factors reduce customer draw to an
insignificant level.
Since the probability of attendance at a retail
center declines as driving time increases, graduated
percentage discounts are made for the total number of
consuming units residing in the more distant isochron
zones. This time-distance approach to trade area mapping
includes the area lying within an arbitrary maximum driv
ing time in all directions from a shopping center.
In the case of a large shopping center with retail
stores of different type and size, a decision must be
made as to which store trading areas are to be measured.
Of course, an individual store-by-store analysis of an
existing or proposed shopping center could be conducted.
This approach may have relevance, because
While the limits of a shopping center are
delineated by the drawing power of its
most powerful store, every store in the
shopping center will draw some business
from all parts of the trading area, the
intensity of draw for each store dimin
ishing at a different rate with increasing
distance.27
Resources are normally not adequate to permit so
intensive an analysis of all retail components in a large
shopping center. Actually, such detailed study is not
necessary. Applebaum indicates that "Since there are so
many retail units in a shopping center, only the major
tenants are identified on the map and subsequently analyzed
on a separate b a s i s . "28 Because the largest purveyors of
shopping goods, e.g., department stores, in a retail
center determine the maximum geographic extent of a
trade area, and are also responsible for much of the in
cidental convenience goods shopping, detailed study of
their (and other selected) trading areas will yield
sufficient knowledge about a center's market potential.
A curve depicting the geographic distribution of
department store sales in an outlying shopping center
typically conforms to a bell-shape when plotted on a
graph; with per cent of per capita expenditures being
made at the department store on the vertical axis, and
27Applebaum, Store Location..., 88.
28Ibid.. 16.
consumer distance from the store on the horizontal axis.
This bell-shaped per capita sales curve for a department
store in a regional shopping center
...tends to be relatively flat at the
top, falls off rapidly in the vicinity
of the 10-12 minute driving time zone,
and then flattens out with a gradual
slope to about 20 to 25 minutes
driving time.29
There are numerous factors which determine what
trade area boundries are now and will likely be in the
future. Changes in: population density; income levels;
competition; transportation systems; travel barriers;
etc., can all affect trade area boundries and potentials.
In general, these factors all affect a retail shopping
center's accessibility to its market potential. The
degree of accessibility of consumers to a shopping center
is of prime importance in deriving the geographic size
of a trade area, and the proportion of the population
likely to shop there.
In conclusion, it should be stressed that trade
area boundries have no great significance in themselves.
As Nelson puts it.
There is absolutely nothing which happens on
one side of the line that dees not happen on
the other side. The line is merely an
arbitrary boundry to exclude from the area
29Ibid.. 34.
162
of research those places where the
amount of business which might come
to the center fades out the point
of becoming immeasurable. 0
The criterion for locating a trade area boundary line is
immeasurability, and not an assumption that people will
behave differently.
Applying trade area analysis to retail location
problems.— Study of retail market potential to aid
in determination of size and location of retail facilities
at the local level is becoming more widespread and
sophisticated. Various formulations of Reilly's law of
retail gravitation have been used for years to determine
how new retail developments will share a market area with
existing developments, how large these new developments
should be, and how large the component stores in a shopp
ing center complex should be.
In discussing shopping center location studies in
Los Angeles a number of years back, a local planner ex
plained that "...we locate these centers by first ana
lyzing purchasing power by small civil divisions."
Competition and time-distance discounts are then applied
to this surrounding purchasing power. "We then locate
these regional shopping areas, properly overlapped, so
30
Nelson, The Selection.... 190.
as to surround completely a metropolitan center." This
permits reduced advertising and transportation costs.
"The net result is an eventual reduction in the cost of
distributing goods,..."31
While theorists are continually striving to discover
fundamental relationships to help explain urban and metro
politan structure, practical urban planners are faced
with the necessity of quantifying urban theories and of
providing specific answers to problems of urban retail
development. To deal with some of these planning
problems, Carrothers feels that market potential concepts
can
...provide a basis with which data may be
applied to aid in solving such specific
problems as market analysis for shopping
center location, population and migration
forecasting, traffic flow analysis, and
allocation of land for residential,
business, industrial and other u s e s .32
As Baker and Punaro point out, in retail trade
studies, "The analyst's basic purpose is always to
discover how many stores, of what type and what size,
3-l-Kenneth C. Welch, "Regional Shopping Centers:
Experience in Los Angeles County,"
Journal of the American Institute of Planners, (Fall, 1948)
32Qerald A. P. Carrothers, "An Historical Review
of Gravity and Potential Models of Human Interaction,"
Journal of the American Institute of Planners, (May, 1956)
94.
can reasonably expect to operate profitably in the pro
posed shopping center."33 The calculation and mapping of
trade area purchasing power provide insight into a center1s
present and future sales potential. One important result
of trade area interviewing, and the information derived
on trade area compositions and characteristics, is the
compilation of information files for comparative study
purposes. Studies of existing shopping centers and their
trade areas yield information for reference purposes when
studies are undertaken for locating and determining the
size and composition of proposed centers.
Measuring and Evaluating Local Expenditures
Probability analysis of consumer behavior.— This
section is primarily concerned with measuring and evalua
ting consumer propensities to spend on various retail
goods and services offered, and the geographic pattern
of these expenditures; after basic trading areas have
been established (by gravity model and/or consumer field
survey) for the relevant retail facilities or proposed
sites. Before proceeding, however, some basic assumptions,
relating to the manner in which consumers perceive the
3 -^Geoffrey Baker and Bruno Funaro,
Shopping Centers— Design and Operation. (New York:
Reinhold, 1951), p.17.
utility of a shopping to themselves, are offered.
A clear detailing of the writer's assumptions about
consumer shopping attitudes should precede any discussion
of expenditure patterns and propensities because, as
Applebaum points out,
The factor of population is paramount in
evaluating store sites. Consideration
is given to number, composition, density,
growth, income, expenditures, and buying
habits. These population characteristics,
determined on the basis of census and
survey data, can be translated into
total market potential.34
Selection methods employed by consumers in choosing a
shopping center at which to make retail purchases, are an
underlying factor which must be borne in mind throughout
a trade area analysis.
Empirical evidence indicates two controlling vari
ables influencing consumer choice of a shopping center.
These two variables include: (1) the number of items of
the kind a consumer desires that are carried by various
shopping centers; and (2) the travel time involved in
getting from a consumer's residence (or other travel base)
to alternative shopping centers. Since shopping center
size, in terms of square feet of enclosed selling space,
and the variety of the merchandise offering are virtually
always closely correlated, size of a center can be termed
as controlling variable.
34Applebaum, Store Location..., 68.
These two variables are the most important deter
minants of the utility of a shopping center to consumers
in a trade area. Utility is usually postulated as a
positive linear function of shopping center size, and as
an inverse parabolic function of travel time.
Even though the effect of a shopping center1s size
on its utility to consumers is assumed to be a single
proportionality function, it may be curvilinear in form.
That is, the incremental increases in the
number of merchandise offerings may not
necessarily bring about a corresponding
per unit increase in the perceived
utility obtained from greater o f f e r i n g s . 35
This is based on the notion that as shopping centers grow
in size, it becomes more difficult for consumers to make
selections among merchandise offerings, and marginal
utility declines. For purposes of analysis, though, a
linear function offers a close approximation of actual
consumer utility perceptions.
Anticipated transportation costs, effort involved
in preparing for and making a trip, and other foregone
opportunities, all tend to detract from a shopping
center's utility, the further it is from a consumer's
travel base. Disutilities associated with a particular
shopping trip rise in markedly greater proportion than
travel time, as distance between consumer and center
^Huff, Determination. . ., 17.
increases. Costs of foregone opportunities rise suf
ficiently to act as a check against further time losses
to a consumer. Therefore, a shopping center's value to
a consumer is inversely related to the effort and expense
involved in reaching the center.
According to neo-classical utility theory, other
things being equal, rational consumers ought to shop at
the largest shopping center to which they have easiest
access. For this and similar reasons, Hoyt and other
market researchers have emphasized the importance of
knowing what major department store is going into any
proposed regional center.36 The assumption appears to
be that a large, well known, store will draw many more
customers on a regular basis than a lesser known or
smaller store.
It is correct to assume that a larger shopping
center will probably draw more customers, from a wider
area, and more frequently than a smaller shopping center.
It is wrong to assume, however, that consumers develop
rather exclusive allegiance to one shopping center of
given size, and virtually ignore all other competitive
centers in a metropolitan area.
-a c
JOSee Homer Hoyt, "A Re-Examination of the Shopping
Center Market," Urban Land Institute,
(Technical Bulletin No. 33, September,1958).
In this study, it is assumed that the number of
times a consumer visits a shopping center during a given
time period is in direct proportion to the size (in square
feet of enclosed selling space) of the shopping center
relative to competing centers. Huff asserts that,
...the consumer is not able to discriminate
among choices perfectly and, as a consequence,
is incapable of maximizing in the sense of
choosing one alternative exclusively...
when the perceived differences among
alternative choices are small,...37
The insignificance of alternatives among retail purchase
choices makes it difficult to discriminate among them.
Therefore, a consumer tends to choose somewhat randomly,
and, since he is uncertain as to the true "pay-offs"
associated with various alternatives, he tends to check
on his intuitive beliefs by also choosing from other
alternatives. In other words, a consumer does not select
the shopping center with greatest absolute utility for
him, to the exclusion of competing centers of similar,
greater, or smaller size. A consumer, instead, tends to
visit competing shopping centers roughly in proportion
to their relative utilities to him.
Consumer purchase categories.— Consumer purchases
can be categorized or classified in ways too numerous to
■^Huff, Determination..., 16.
r 169
mention here. To simplify the analytical concept of a
trade area, however, two basic categories of consumer
purchase are usually defined. People make expenditures
for either convenience goods or shoppers goods.
Even though consumers are, in a sense, free to
purchase wherever they please, their places of purchase
tend to fall into a definite pattern. This pattern is
based on differences between goods and differences in
distance, time, cost, and effort to reach the stores and
i retail centers that sell the desired goods.
As Converse says,
Some purchases are considered more important
than others. People will devote more time
and attention— that is, shop around more—
for articles that absorb a substantial
portion of the family income,...than for
small articles bought frequently,...
Thus we have two broad classification of
goods: shopping and convenience.38
Both shoppers and convenience goods fall, of course,
under the general heading of consumption goods, which are
those goods "...sold through retail channels to household
consumers, regardless of whether or not they are in form
for final consumption."39 According to data presented
by Converse, shopping goods purchases have risen in their
I proportionate share of retail sales, from 41 per cent in
1939 to 46 per cent in 1954.^0
38Converse, Huegy, and Mitchell, Elements.....28.
39Ibid., 128.
4°Ibid.
170
The relevance of categorizing consumer purchases
into shopping and convenience groups, lies in the reality
that a single large shopping center will have at least
two basic trade areas. One trade area will be for con
venience goods and will be of the same size and shape as
if the larger stores were absent (this last statement is
not always true and will be modified below). The second
trade area will be for shopping goods and will extend
much farther out than the convenience goods trade area.
These different trade areas for shopping and con
venience goods are discussed in terms of central place
functions by Simmons.4! Convenience goods only are sold
in neighborhood shopping centers (lower in the hierarchy
of urban retail places), while shoppers' goods are sold
exclusively in community and regional shopping centers.
But, convenience goods may also be found in the larger
centers, and have larger than expected trade areas be
cause their purchase is included in a multiple-purpose
shopping trip. The convenience goods purchase, in other
words, is incidental to a trip for shopping goods in a
larger center.
The predominant store type in a large shopping
center that sells mostly shopping goods is the department
store. Since department stores sell practically every
4-*-Simmons, The Changing Pattern. . . . 25.
I
i line of merchandise offered by the specialty and limited
line variety stores typically comprising a shopping
center, their trade areas are often assumed to encompass
an area that includes the trade areas of all the other
stores in a retail center.
This study concentrates mainly on department store
trade areas and sales operation. This attention to
shopping goods purchases is not intended to imply that
study of convenience goods trade areas is not important.
Rather, it stems from the fact that this study is not
concentrating primarily on small neighborhood shopping
centers that sell only convenience items. Since conven
ience goods trade areas are more predictable and easily
determined, and are included within the larger shopping
goods trade areas, there appears to be no good reason to
give them explicit treatment.
Family budget and retail sales information.— At the
heart of a retail market potential study is the need for
sufficient accurate information on local income, expend
iture, and retail sales levels and compositions. Data
pertaining to these economic variables must be generated
for a local area, e.g., a municipality, and its surround
ing retail trade area. This necessity for specific de
lineation of geographic source of income (for an economic
172
base study) and incidence of retail expenditure or sale
(for a trade area study), presumes the use of consumer
field surveys to generate accurate data.
The economic and social characteristics of the
population residing in the trade area must be determined.
The number and composition of families and other spending
units in the trade area should be ascertained. Total
disposable personal income accruing to individuals and
families, current consumption expenditures, total retail
expenditures, retail expenditures by product class or
store type, and the place of expenditure should all be
estimated with reasonable accuracy.
Income and expenditure data can be tabulated by
various social, family, and economic characteristics,
and even cross-tabulated in more complex fashion. The
level of family income, type of employment, number of
children, length of residence in the community, and so
on, all affect the rate and nature of local retail
purchase.
The following data, presented in Table 1, indicate
the distribution of family income, when it is budgeted
for consumption purposes.
Family budget information, as in Table 1, derived
from previous consumer surveys, can be applied to a trade
area population to estimate likely expenditure propensi-
TABLE 1
DISTRIBUTION OF CONSUMER EXPENDITURES
BY INCOME LEVEL NATIONALLY, 1958
Gross family income
level
2
or
less 2-2.9 3-3.9 4-4.9 5-5.9 6-6.9 7-7.9
8
or
more
Per Cent saved
and taxed 23.7 24.3 26.6 26.7 30.6 30.7 32.3 33.2
Per Cent consumed 76.3 75.7 73.4 73.3 69.4 69.3 67.7 66.8
Type of Goods
or Store Per Cent of Gross Income Expended
Convenience 48.1 46.1 44.4 44.5 40.2 39.1 36.5 35.5
Shopping 20.7 21.7 20.7 20.8 21.3 22.5 23.5 23.5
Department Store 5.0 6.0 6.8 7.4 7.6 8.3 8.0 7.8
General Purchases 7.5 7.9 8.3 8.2 8.0 7.8 7.7 7.9
Total 76.3 75.7 73.4 73.3 69.4 69.3 67.7 66.8
Source: Richard L. Nelson, The Selection of Retail
Locations, (New York: F.W. Dodge & Co., 1958), p.342.
ties of families at different income levels. An average
family might be expected, for example, to spend around
7 to 8 per cent of its total income in local department
stores. The geographic pattern of department store
spending must, of course, be estimated separately.
Table 2 presents historical data on the way dis
posable personal income has been distributed among retail
purchases nationally between 1929 and 1954.
It can be interpreted from the information in
Table 2, that, while retail stores as a whole have been
gaining in their proportionate share of consumer expend
itures, the distribution of retail expenditures among
different categories has been changing somewhat. Pur
chases in the General Merchandise Group, for example,
have declined from almost 11 per cent of disposable
personal income in 1939, to only 7 per cent in 1954.
With the use of historical data such as these, changes
in retail expenditure propensities and patterns can at
least be roughly anticipated.
A considerably more complete series of historical
data is presented in Table 3. These indicate the pro
portion of personal consumption expenditures made annual
ly in, respectively: all retail stores; general merchan
dise, apparel, and furniture and appliance stores
combined; general merchandise stores alone; and department
TABLE 2
DISTRIBUTION OF DISPOSABLE PERSONAL INCOME
AMONG SELECTED RETAIL PURCHASES,
1929-1954
YEARS 1929 1939 1948 1954
DPI
(billions of dollars) 83.1 70.4 187.6 254.8
Retail Sales
(billions of dollars) 48.3 42.0 128.8 170.0
Retail Category Percent of DPI Expended
General Merchandise 10.8 9.2 8.4 7.0
Apparel 5.1 4.6 5.2 4.3
Furniture & Appliances 3.3 2.5
f
3.5 3.4
Total Retail 58.1 59.7 68.5 67.2
Source: U. S. Department of Commerce,
Survey of Current Business, (Washington, D.C.:
U. S. Government Printing Office, April, 1965).
175
n
i
Year
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
I 1960
i
i
176
TABLE 3
DISTRIBUTION OF CONSUMER EXPENDITURES
AMONG SELECTED RETAIL STORE GROUPS, 1935-1960
All General
Retail G.A.F.a Merchandise Department
______Stores_______Stores________Stores_______Stores
(Per Cent of personal consumption expenditures)
58.3 17.2 10.1 5.7
61.1 17.7 10.2 5.9
62.6 17.5 10.0 5.8
59.0 16.4 9.4 5.6
62.1 17.0 9.6 5.8
64.5 17.1 9.6 5.7
67.5 17.9 9.8 6.0
63.8 18.6 10.3 6.0
62.9 18.3 10.1 5.9
63.9 18.3 10.1 5.9
64.1 18.2 9.7 5.8
69.7 19.3 10.0 6.3
72.3 19.2 9.7 6.1
73.2 19.0 9.6 5.9
72.1 17.9 9.0 5.5
73.7 17.8 8.8 5.4
74.6 17.6 8.7 5.4
73.8 17.4 8.5 5.3
72.7 16.5 8.2 5.0
71.1 16.0 7.9 4.8
71.6 15.9 7.8 4.7
70.3 15.9 7.7 4.7
70.1 15.4 7.4 -
68.3 15.2 7.4 4.8
68.8 15.2 7.5 4.9
66.8 14.6 7.2 4.8
Source: Malcolm P. McNair, The American -
Department Store 1920-1960, (Boston:
Harvard University Graduate School of
Business Administration, Research Bulletin
No. 166, 1963), p.12, and U. S. Department
of Commerce, Survey of Current Business.
March, 1961.
aCombined total of General Merchandise,
Apparel, and Furniture and Appliance Groups.
stores (which constitute the largest element in the general
merchandise store group).
The Table 3 percentage distribution figures indicate
that the G.A.F., General Merchandise, and Department Stores
have been slipping in their shares of total consumption
expenditures since the Second World War. This slippage
has been occuring concurrently with a relatively strong
position for all retail stores combined. The implication
in this situation is that there has been some redistribu
tion of consumer expenditure allocations to other types
of retail stores. The new super-market and super-drug
stores, for example, combine many variety and specialty
lines now with their traditional merchandise offerings,
and these new store types have had some impact on tradi
tional general merchandise and apparel stores.
Even though family budget and retail sales informa
tion, collected at the national or state level, are ex
tremely useful in providing some notion of the order of
magnitude of various retail expenditure allocations;
local data, generated by field survey, are usually a pre
requisite for accurate determination of the level and
composition of market potential in a given retail trade
area.
A study conducted in Norwalk, California in 1960
is a good example of the need for careful local
analysis.^2 Total income accruing to Norwalk families in
I 1960 was estimated at $123 million. Retail expenditures
of residents (made both inside and outside the city of
Norwalk) were estimated at $86 million for 1960. Resident
expenditures made in Department Stores in 1960 were about
$9.5 million. These expenditure figures for Norwalk resi
dents indicate that about 70 per cent of their personal
income is spent on retail purchases, and that about 7.7
per cent of that income is spent in Department Stores.
Use of the more recent national average figures from
Table 3, to estimate Department Store expenditure alloca
tions for Norwalk residents, would have led to an almost
3 per cent underestimation of local Department Store
market potential.
The retail expenditure allocation figures for the
State of California in Table 4, illustrate further that
temporal and geographic variations in these magnitudes
cast doubt on the validity of indiscriminate application
of either state or national averages to local situations.
Comparison of the percentage of disposable personal
income spent in Department Stores on the state level, with
that spent on the local level in Norwalk, indicates that
4^See william T. Kopp and Louis Salomone, Jr.,
The Market Potential for a Regional Shopping Center in the
City of Norwalk,(South Pasadena, Calif.: Southern Calif
ornia Laboratory of the Stanford Research Institute,
November, 1960).
TABLE 4
DISTRIBUTION OF DISPOSABLE PERSONAL INCOME IN CALIFORNIA
AMONG EXPENDITURES IN SELECTED RETAIL STORE CATEGORIES
1963
CATEGORY TOTAL
~ TER------
HOUSEHOLD
(billions
of $)
(%) (thousands
of $)
( ° / o )
DPI 45.2 100.0 8.0 100.0
Total Retail Sales 26.8 59.3 4.8 59.3
General Merchandise 3.6 8.1 0.6 8.1
Department Stores 2.8 6.2 0.5 6.2
Source: Sales Management Magazine,
Survey of Buying Power, 1964.
179
application of the state-wide figure of Table 4 to the
Norwalk trade area would again lead to an underestimate
of the propensity of local residents to spend in depart
ment stores.
Investigations of family budgets can be carried to
a very great fineness of detail if resources, time, and
the analytical requirements of the study demand it. In
the case of retail trade area studies, it is often
necessary to quantify only certain significant categories
of spending propensities and patterns. The present study
will be concerned mainly with propensities to spend on
shopping goods in department stores, and the geographic
patterning of these expenditures.
Micro-analysis of retail shopping behavior.— As in
economic base studies, application of national or state
averages of family budget data to trade area studies,
ignores often significant differences in local income and
expenditure propensities and patterns. Differences of a
few percentage points between local consumer and national
or state-wide consumer allocations of retail expenditures
among product classes and store groups, may translate
into differences of several hundreds of thousands (or
millions) of dollars of retail sales potential.
An even more difficult phenomenon to estimate
independently of local observation is the geographic
pattern of retail spending. Even with the use of trade
area studies conducted in communities similar to one
being analyzed, anticipation of all unique local factors
and forces affecting spending patterns would be difficult.
The underlying parameters determining and affecting
shopping trip patterns vary markedly from one locality
to another. For this reason, close study must be made
in each instance: of local road and public transportation
systems; income levels; the location and nature of com
peting retail facilities; and social, economic, and
aesthetic attitudes affecting shopping desires of
consumers.
Micro-analysis of a trade area consists of detailed
study of consumer spending propensities and patterns in
selected small segments of the trade area. These small
segments might be city blocks, census tracts (i.e.,
homogeneous "neighborhoods"), or other arbitrary small
segments. For trade areas of larger regional shopping
centers, the blocks and tracts are usually considered
too small to serve as geographic units of analysis.
Nelson's suggestion of dividing large regional trade
areas into octant segments is felt to be more suitable.^3
^3Nelson, The Selection..., 193, 194.
Once the general extent of a trade area has been
determined, either by application of a gravity model
formula or license plate surveys and consumer question
ing at retail centers, segmentation for micro-analysis
can be effected. Equiprobability contours in the form
of concentric rings can be drawn around the shopping
center site. These contours enclose concentric zones
inside of which the probability that consumers will
regularly visit the center to shop declines as driving
time from the center increases. Each concentric zone
lying farther out from the center would contain some
number of consumers, for whom driving time to the center
is greater than for consumers residing in closer zones,
and for whom probability of regular shopping trips to
the center is diminished.
Octant segments are formed in a roughly circular
trade area by drawing lines through the center and the
concentric equiprobability (or isochron) contours, to
the outer limits of the trade area. Pour lines are
drawn: north to south; east to west; northwest to south
east; and northeast to southwest. As one proceeds farther
out along or between any of these radials, the number of
consumers visiting the shopping center tends to decline
in all octant segments in a similar manner.
The sectioning or segmenting resulting from inter-
section of the concentric circles and radials, creates
very small areas for intensive analysis close to the
shopping center. The segments grow progressively larger
the farther one gets from the center, and less accurate
analysis occurs. In other words, nearer the center,
where the proportion of families shopping there is very
high, intensive micro-analysis occurs, and farther out,
where the proportion of families shopping there is much
lower, a less intensive analysis is conducted.
Table 5 shows how the calculations are carried out
to estimate expected retail expenditures at a shopping
center from a portion of its trade area.
The expenditure figures in the last column of
Table 5 indicate that the shopping center should capture
about 44 per cent of all retail expenditures made by
families living within a one mile radius, and only about
24 per cent of the expenditures made by families in the
one-to-two mile radial zone. Percentage assignments of
business going to the center, which are applied to trade
area gross income and expenditure figures, are derived
from expressed shopping location preferences of consumers
in the various segments. Nelson emphasizes that family
questionnaire interviews conducted by interviewers in
the field are essential to the accuracy of this approach.
As far as actual selection of families to be inter-
TABLE 5
SHOPPING CENTER RETAIL EXPENDITURE POTENTIAL
Assignment
Total Income
of Prospective
Customers
Total
Retail
Expend
iture
Retail
Expend
iture!
at
Shopp
ing
Center
Micro-
analytic
Seqments
Total
Popula
tion
Number
of
Families
Average
Family
Income
Total
Gross
Income
Conven
ience
Goods
Shopp
ing
Goods
Conven-Shopp-
ience ing
Goods Goods
Within a
1 mile
radius 10.5 3.0 6.5 17.2 60 45 10.0 7.7 11.9 5.2
Within a
2 mile
radius 93.3 26.6 6.0 144.2 30 35 40.8 45.0 100.1 24.4
Total 103.7 29.6 6.3 161.4 45 40 50.7 52.7 112.0 29.6
Source: R. L. Nelson, Selection of Retail
Locations, p. 196.
o o
viewed is concerned, a comprehensive geographically
stratified random sample of all families in the trade
area is probably best. This approach yields geographic
sets of families accounting for any uneveness of spatial
distribution within the trade area. In the case of a
very extensive regional trading area, a variation of
this area sampling approach might be followed. Perhaps
after a complete random sampling of the closest in portion
of a trade area, say the first mile out, a selection
could then be made of five to ten other segments farther
out, from which random samples could be drawn. The
selected segments should be:
(1) representative or typical of several
other segments so that interpolation
can be made;
(2) at critical points where dividing lines
between competing shopping centers are
indicated; and
(3) selected to fit into before and after
(comparative static) studies of other
shopping center trade areas.
According to Nelson's data, there is no dramatic
variation in allocations of family income to various
I
classes of product purchase between the East, West,
A
! 186
South, and Mid-West regions of the United States.44
There are expected variations for certain product classes,
but these usually amount to differences of no more than
a few percentage points. In other words, Nelson feels
that national average shopping habits are typical of
regional average shopping habits. The same orders of
magnitude of consumer purchase are believed to prevail
regionally as prevail nationally.
These conclusions still do not obviate the necessity
for careful and detailed local consumer expenditure
studies. State, or multi-state regional, averages
suppress a multitude of local economic, social, and
physical peculiarities, which may materially affect rates
and distributions of consumer retail purchase. Local
conditions should be analyzed thoroughly to accurately
estimate retail market potential. Regional or national
averages may be useful as guidelines against which to
check local survey results, but they certainly should
not be substituted for the latter.
Limitations of Trade Area Analysis
; Micro-analytic orientation.— The fact that retail
market potential analysis is a micro technique, offers
both advantages in its application to land use planning
44Ibid., 221.
187
problems and limitations to its usefulness. On the one
hand, detailed study of local areas of retail demand is
possible with trade are analysis. Fairly precise spatial
distributions of this demand can be discerned, and retail
shopping facilities can be located to maximize access
ibility to their market potential. Given a determination
of potential expenditure (and, therefore, sales) levels,
more precise allocations of land space can be made for
retail use, given the productivity of retail space
utilization.
On the other hand, retail market potential analysis
does not provide an adequate device for conducting fore
casts and/or predictions of overall levels of local
(i.e., municipal, in most instances) economic activity.
At the geographic level at which most intra-urban retail
market potential studies are conducted, it is impossible
to construct a theoretical framework postulating function
al economic relationships among significant variables
which, in turn, determine aggregate levels of local eco
nomic activity. This shortcoming is discussed in greater
detail below.
Retail market potential analysis is, in comparison
with economic base multiplier analysis, a micro-demand
approach to the study of small geographic areas. In
addition, trade area studies contain definite retail
space supply implications which are not inherent in base
studies. Retail space productivity analysis, which
specifies the supply requirements implied in a market
potential study, is discussed in the following chapter.
Relationship to other studies.— While a market
potential study is valuable to local analysis by itself,
its usefulness is enhanced greatly when it becomes a
component in a more comprehensive community economic ana
lysis. This is so because of the highly disaggregated
level of analysis at which intra-urban retail trade area
studies are conducted.
The notion of the gravity model, as originally
developed pertains to relatively large masses of retail
or other activity (e.g., inter-metropolitan potential
studies) composed of a multitude of individual units.
Isard says that,
Within such a mass it is reasonable
to assume that the irregularities,
peculiarities, and idiosyncrasies
of any individual unit or small
subgroup of units are cancelled
or averaged out. In such a
situation the ceteris paribus
clause is valid to some extent
at least.45
When shopping trips are disaggregated by neighbor-
45Walter Isard, Methods of Regional Analysis,
(Cambridge: M.I.T. Press, 1960), p.513.
189
hood of origin, class of retail purchase, and retail
store or center visited,
...the peculiarities of each category
tend to become more manifest and
dominant; and the extent to which
the gravity model describes or
explains any regular 'falling-off'
effect tends to decrease.46
If the mass being studied is only a disaggregated segment,;
e.g., the retail stores and centers in a municipality,
of a much larger metropolitan activity mass, then many
random factors are likely to fee operating which cannot
be accounted for on a probability basis.
Mathematical market potential models based on
probability are, as Voorhees points out, inherently weak
in dealing with small areas.47 They are most effective
when needs are studied on a broad (metropolitan) scale.
If market potential analysis is conducted on a sub
metropolitan scale, there must be comprehensive research
in the form of consumer field surveys.
Under these conditions, retail market potential
studies offer little usefulness in projecting economic
activity levels on a municipal scale. Trade area ana
lysis lacks theoretical content. It has been used mainly
as a tool of analysis, and little has been done to ex-
46Ibid.
47Voorhees, Journal of the American Institute of
Planners, (May, 1959), 58.
190
amine behavioral implications associated with its use.
It is true that Loeschian market area theory has explain
ed aspects of the functional economic nature of geograph
ic areas of demand. It is equally true that market area
theory lacks an explanation of the way economic activity
levels change over time in small geographic areas.
Economic base theory, on the other hand, does
provide an explanation of local economic growth. By
assessing future possibilities of local exports, and
calculating local income generation through the local
income multiplier process, projections of future income
and activity levels can be made for an area. The ana
lytical utility of both base and trade area studies are
obviously enhanced by synthesis. Specific location of
market potential can be obtained from a trade area ana
lysis (with implications for space allocations) conducted
within the theoretical framework of an economic base
study and forecast.
Isard has worked very hard at synthesizing various
approaches to urban and regional analysis. He feels,
with regard to the market potential model, that
Solutions to the problems of defining
relevant mass and distance, assigning
appropriate weights and exponents,
establishing a basis of comparison
with other research results, and
broadening its probability framework,
etc., require extensive systematic
191
probing. Only with such research can
we extend basic knowledge of the forces
underlying gravity models, construct
more meaningful and useful descriptive
and projective models, and desirably
achieve a firm synthesis with other
techniques of analysis and bodies of
theory.48
A trade area study expands the scope of analysis to in
clude a wider geographic area than that enclosed by
municipal boundries, and yields a micro-analysis of the
"export" area served by retail facilities in a municipal
ity; while an economic base study elevates the level of
analysis to encompass economic activity levels for the
municipality— including retail export levels.
Time-space limitations: changes in the
parameters of spatial interaction.— At different
times and in different places economic and social factors
and forces may be operating in different relationship to
one another. The parameters determining the spatial
distribution and movement of people, income, and expend
itures are usually altered when temporal or spatial
dynamic changes occur. Since a market potential model
is a mathematical statement of observed relationships
existing at a given time and place, it can easily become
outdated when it no longer explains changed relationships
48Isard, Methods.... 566.
192
at a later time or different place.
Trade area configurations usually change signifi
cantly only when a dramatic alteration occurs in one or
more of the underlying parameters of spatial interaction,
e.g., a new freeway might drastically reduce driving
times and increase accessibility to a shopping center.
A less significant change in the trade area might have
little or no effect on its shape and size. For example,
if a retail store were replaced by another store of the
same type with improved merchandising facilities only,
the trading area will remain unchanged in size, although
marXet penetration within existing boundaries may improve.
On the other hand, Applebaum notes that,
Where the functions of a store change
significantly, as reflected by changes
in the merchandise mix and in service,
then trading area boundaries will c h a n g e . ^
A great variety of factors may change, and to any degree,
to cause changes in the size, shape, and internal nature
of a trade area.
Dynamic change should be expected in retail trade
areas, however, because
Store trading areas are phenomena that are
fixed in space for recognizable periods of
time. Their boundaries are zones, not lines;
the boundaries are flexible, not rigid. These
^Applebaum and Cohen, Journal of Retailing.
(Fall, 1961), 56.
193
boundaries change with changes in
shopping-habit patterns and with
changes in merchandising practices.^®
Once the effects of certain parametric changes are under
stood, it is possible to anticipate future trade area
configurations. In 1946, for example, Converse conducted
a comparative static analysis of retail trade movements
in Illinois. His study was mainly of conditions as they
existed before, and after, the Second World War. His
results showed that dramatic changes in trade area con
figurations and relationships had occurred during the
war, because of shifts in population, changes in highway
transportation, income levels, etc. With fairly accurate
estimation of future change in the more important para
meters, and some knowledge of their effects, trade area
change could be anticipated to some extent.
Distance exponents.— There is probably no other
single element in market potential models and analysis
as controversial as the distance factor. How restrictive
is distance in discouraging various trip decisions?
Reilly postulated distance as an exponential function— he
squared the distance factor in his trade area formula.
Earlier gravity formulas had left the distance factor with
no exponent.
50Ibid., 16.
Since distance (or driving time) appears in the
denomenator of a market potential formula, and is divided
into the measure of mass, it represents a restrictive
force opposing the attractive force of the mass (i.e.,
retail shopping center, etc.). Because the attractive
force of the mass is usually postulated as a simple pro
portionality function, the higher the exponential value
of the distance factor, the more restrictive a force it
represents.
Some market analysts feel that the restrictive
force of distance is definitely an exponential function
that should be raised to some power for various types of
trips. Voorhees has noted that intra-urban trip studies
...indicate decreases in the exponent
as trips become more important, i.e.,
school trips 2 . 0 +, shopping trips
2.0, social trips 1.1, work
trips 0 . 9.51
Huff notes, more generally, that
Studies have shown that the exponent
has ranged from 1 . 5 to over 3,
depending on the trip type being
analyzed as well as the geographical
setting involved.52
When the frequency of trips from, say, a residential
neighborhood to a given shopping center, are placed along
a vertical logarthmic scale, and the intervening distance
51Voorhees, Journal of the American Institute of
Planners, (May, 1959), 74.
52fjuff, Determination. . . , 12.
195
is placed along a horizontal arithmetic scale, the numer
ical value of the slope of the regression line fitted to
a scatter of data plotted to these coordinates, equals
the distance exponent. The steeper the slope of the re
gression line, the higher the value of the exponent, and
vice versa. Trips for convenience items, for example,
have a steeper line than trips for shoppers' items.
Upper and lower limits can usually be established for
various trip categories, but much variation occurs within
the range, so that local empirical observations are
necessary for accurate calculation of the distance expon
ent.
Some problem of accurate intra-urban retail
trade area determination.— A moment's reflection
should be taken on the problem of whether or not a trade
area resembling reality is actually being determined for
a retail facility or group of facilities. Complex con
ditions of competition and dominance and sub-dominance
of centers of retail activity within a metropolitan area,
make it extremely difficult to discern whether or not
correct assumptions are being made about weighting factors
distance exponents, and so forth in market potential
formulas.
Loesch says, that for any two towns it is being
196
used to analyze, Reilly's law of retail gravitation
"...assumes that each town is large enough to exclude
the other from its local market. That is to say, the
pair must not be too u n e q u a l . ”53 If one town is much
larger, the second town might be included in the first
town's trade area. In an intra-urban study, the decision
as to whether one group of retail facilities is or is not
included in the trade area of another dominant retail
center, is much more complicated than in the simpler
inter-urban case involving only two cities.
It may be unrealistic to apply market potential
formulas in metropolitan areas without engaging also in
extensive field research. Applebaum notes, that for
intra-urban studies,
...it is extremely difficult to employ
gravitational models. This is because
competition is to be found in so many
different directions at so many different
scales. Also, driving-time distinctions
may be only a matter of moments.54
The relative differences in attractive force (or utility
to the consumer) of different retail centers may be so
slight, and the relative differences in the restrictive
force of distance so subtle, that no gravity model could
account for these differences in a meaningful way. Under
S^Loesch, The Economics..., 411.
54Applebaum, Store Location..., 79.
these conditions, field shopper surveys are necessary to
determine home addresses of consumers visiting a shopping
center or department store. In other words, an empirical,
as opposed to a formula, approach is required to properly
delineate a trade area.
In an empirical study of competing shopping center
trade areas, Huff found that his potential model as con
sistently predicting a smaller number of consumers going
to a particular shopping center than were observed going.
Predicted visits to the center were anywhere from 10 to
58 per cent of the observed, except in one instance where
they were 114 per cent of observed. Huff traced the
cause of these discrepancies to the fact that a competing
shopping center, located nearby, was getting a much great
er proportion of predicted consumers than were actually
observed visiting the center, because of its close proxi
mity . ^ There is a constant need for change in potential
model parameters to make them realistic. For detailed
studies, they should only be used in conjunction with
empirical observation.
Calculation of breaking points between retail
centers is entirely inadequate in intra-urban analysis,
because this procedure merely outlines large areas within
which each center is presumed to be dominant. But, as
5^Huff, Determination..., 29.
198
Huff has shown, retail trade areas actually represent a
series of radial zones in which the proportion of con
sumers patronizing a given retail facility varies from
zone to zone.
If an analyst were to use a Reilly-type breaking
point formula to delimit the trade area of a proposed
shopping center, located in an area including a pre
ponderance of light industries and non-competing com
plementary commercial activities, the resulting evidence
is likely to be highly distorted. A cross-section field
survey of residential areas should be conducted to deter
mine the proportion of households likely to patronize the
proposed center as well as competing centers. The result
ing trade area will probably be much different and more
realistic than a Reilly gravity model formula would
indicate.
CHAPTER V
SPACE REQUIREMENTS ANALYSIS
This chapter is intended to provide an overview of
the development, content, and deficiencies of retail
space-input analysis. Space requirements studies con
stitute the micro-supply analytical component of the
model presented in Chapter VI.
Since space requirements analysis owes its theore
tical content to input-output analysis, a thorough
development of the concept of input-output is a pre
requisite for treatment of retail space-input studies
themselves.
This chapter also extends to a broader discussion
of retail operation and market structure. This extension
is necessary in order to provide sufficient perspective
on retail space productivity as it relates to the
establishment of "standards" or "norms" of efficiency
in the use of retail space.
Historic Review
The concept of input-output.— The space require
ments approach to providing micro-analysis of local retail
trade supply needs does not have an extensive literature,
as do the economic base and market potential approaches.
This is because space requirements analysis for retail
trade, as presented in this study, actually represents
a modification of yet another analytical approach— the
input-output study.
There is much difference between a retail space
requirements study (conducted in conjunction with an
economic base-market potential analysis), and implementa
tion of a traditional input-output study. In spite of
this difference, however, input-output analysis still
provides the conceptual framework encompassing the space
requirements approach.
Input-output analysis in the United States origina
ted from the work of Wasily W. Leontieff. "A pioneer
in input-output analyses, Professor W. W. Leontieff of
Harvard published the first national table before World
War II."1 He and his associates have since contributed
much to the development and application of the input-
output approach to studies of national and regional eco
nomic structure.2
A
Input-output analysis explicitly sets out the
lAbe Gottlieb, "Planning .lements of an Inter
industry Analysis," Journal of the American Institute
of Planners, (Fall, 1956), 233.
2See Leontieff's definitive work, The Structure
of American Economy, (New York: Oxford University Press,
1951) .
interrelationships between industries and sectors within
an area, and, if desired, between areas. These inter
industry and interarea technical relationships can be
studied to virtually any degree of fineness of detail,
given the purposes of analysis and resources available
to implement the study.
With an input-output study, Dorfman says that
Whatever the number of sectors, each
sector is considered to buy its inputs
from other sectors, and sell its output
to other sectors and, perhaps, to an
"autonomous sector" which has no output.
The autonomous sector, if there is one,
represents final demand and is unexplained
within the model.3
Normally, final demand sectors are not determined within
an input-output model. Consumer demand for local retail
goods and services, for example, is not usually explained
by an input-output model. With a given change in retail
demand, this type of model would trace through local
interindustry or intermediate demand effects. In other
words, it conceptualizes the interdependence that exists
between all of the different industries in an area (e.g.,
a municipality) , in terms of the product flows linking
these industries.
In the course of the conduct of an input-output
^Robert Dorfman, "The Nature and Significance of
Input-Output," Review of Economics and Statistics,
(May, 1954), 123.
! 202
j
study, an input-output table is usually constructed. An
i
input-output table lists industries selling their outputs
along the left side, and industries purchasing inputs
across the top. The main body of the table, then,
presents industry outputs in rows across the table, and
industry inputs in columns down the table. Table 6
provides a simplified example of an input-output table
for a hypothetical area.
Suppose that one needs to know how much of the
Trade industry's output is allocated to the Construction
industry. Read down the Industry Producing column to
Trade, then read across the row of numbers to the one
below the Construction column. Table 6 indicates that
$2.5 million dollars' worth of Trade's output was purchas
ed by the Construction industry. Reading down the column
of figures below Construction, this same $2.5 million
value would be interpreted as a Construction industry
input from the Trade industry, rather than a Trade indus
try output. It can be seen from this, that an input-
j
i output study emphasizes the consistency and simultaneity
of relationships involved. In this case, one industry's
sale is another industry's purchase.
I
| Table 6 can also be converted into a table of
input or production coefficients. This can be accomplish
ed by dividing each input value in every industry column
TABLE 6
INPUT-OUTPUT TABLE
Industry
Producing
Industry Purchasing
Mfg Trade Services Construction
House
holds
Total
Output3
(millions of dollars)
Manuf a c tur ing 28.3 0.9 2.7 1.1 43.1 —
Trade 1.6 0.2 0.5 2.5 27.1 —
Services 1.7 3.1 0.8 1.0 7.5 —
Construction 0.3 0.2 0.1 0.1 0.2 —
Households 25.8 26.2 9.0 11.5 2.1 —
Total Inputs3 — — — — — —
Source: Abe Gottlieb,
Interindustry Analysis,
"Planning Elements of an
" Journal of the American
Institute of Planners, (Fall, 1956), 232.
aTotals for inputs and outputs are omitted
because some sectors in the original table have
been eliminated to compress it for easier
presentation.
203
| 204
by the total input values at the bottom of the table.
For example, the Manufacturing coefficient for the Trade
industry is derived by dividing the $0.9 million figure
at the top of the Trade column by total input value
(not shown) for the Trade industry. When this operation
has been completed for each input value in each column
in Table 6, a production coefficient table is derived.
These coefficients are constants of input for any
given industry, that indicate the production require
ments imposed on other industries in the economic system
when a rise in output occurs in that industry. A rise in
consumption demand in the household industry (or demand
sector), for example, would require an increase in Trade
industry output which, in turn, would require proportion
ately greater inputs from all the other industries shown
in Table 6 to produce that higher output. Ghosh defines
an input-output model as "...a matrix of injections
showing the demands placed on the producing sectors by
the non-producing sectors,..."4 In this example, the
i household sector is non-producing and is referred to as
I
a "final demand" sector.
An input-output table offers a detailed presenta-
| tion of the production and distribution characteristics
i
j — — ■ ■ ■ ■ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
i 4A. Ghosh, Experiments with Input-Output Models,
(Cambridge: Cambridge University Press, 1964), p.4.
of individual industries. It also indicates the nature
of interrelationships among these industries themselves,
and among these industries and other economic sectors,
e.g., household or consumer final demand.
Isard says that,
It becomes sensible to distinguish between
(1) these sectors which may be termed the
final demand, or bill of goods, or exogenous
sectors, and (2) other sectors which may be
termed endogenous or processing sectors
whose outputs we judge can be reliably
approximated by a set of coefficients.5
The endogenous or intermediate demand sectors constitute
the structural matrix of an input-output table. The
columns and rows referring to the final demand sectors
(e.g., households in Table 6), are usually either set at
the far right and bottom of the table, or are eliminated
entirely from the presentation.
As Ghosh says, "Final demand in the home market
consists of household expenditure on consumers' goods and
services,..."6 He continues, by noting, "It was found
for a certain number of industries the component of house
hold consumption was by far the most significant part of
final demand..."7 This is overwhelmingly the case with
^Walter Isard, Methods of Regional Analysis,
(Cambridge: M.I.T. Press, 1960), p.335.
^Ghosh, Experiments..., 8.
7lbid.
retail trade. The most significant portion of retail
trade output goes directly to the household consumption
demand sector.
The input-output relationship for retail trade can
be stated with the following mathematical expression:
X - xD - = Y
which simply states that the gross output of the retail
trade industry (X) less sales to manufacturing (xQ) and
itself (x^) equals final demand (Y) for retail commodit
ies by household consumers. The household sector is
exogenous and (in this example) accounts for all final
demand, while the manufacturing and retail trade industr
ies are endogenous and constitute intermediate demand.
Future intermediate demand requirements of manu
facturing and retail trade can be predicted from the
production coefficients of the input-output model.
Estimates of final demand (to which retail output
responds) "...may be based on demographic analysis and
best judgement concerning the consumption pattern of the
projected population in the future year."® Expected
future levels of total retail output or sales (based on
consumer income and expenditure forecasts) are multiplied
by the coefficients to determine absolute amounts of
®Isard, Methods.... 364.
207
future input requirements. The coefficients are, of
course, expected to remain constant for reasonable periods
of time.
The input-output concept and retail space
requirements analysis.— Inputs (purchases) by the
household sector represent consumer expenditures, i.e.,
final demand as opposed to the intermediate demands of
industries purchasing from other industries. The only
"output" the household sector is conceived as having is
labor services, which represent inputs purchased by other
industries to produce their outputs. The sale of labor
services represents the financial source providing income
for consumption spending on retail inputs.
The retail trade industry also purchases inputs
from various other industries in an area (and other areas
if the scope of the analysis is interarea). Some of
these inputs are purchased from the household sector as
labor services, and many other types of inputs are
purchased from intermediate industrial sources.
The major difference between input-output and
space requirements analyses is that the latter does not
provide as comprehensive a scope of study. Space require
ments analysis does not study in detail the entire input
structure of local retail trade but, instead, focuses
208
attention on that part of the retail input structure
pertaining to physical space requirements, measured in
terms of square feet of enclosed selling space. That is,
retail trade space requirements for given expected out
put levels (i.e., sales to the household sector) are to
be determined and analyzed.
There are several linkages that have to be estab
lished and determined. First, there must be some ade
quate forecast of local income levels. Second, pro
portions of these expected income levels which will be
allocated to retail expenditure must be determined.
Third, expected retail output or sales levels, arising
in response to consumer retail purchase allocations, have
to be ascertained. Fourth, physical space input require
ments to economically attain expected retail sales levels
have to be specified.
Future income levels can be forecast from an eco
nomic base and income multiplier analysis. Future ex
penditure propensities and patterns can be determined
from a trade area consumer survey. Retail sales levels
can be ascertained from a market potential analysis
accounting for existing and probable future competition.
Retail space requirements can be linked to expected sales
levels via a set of space coefficients.
These retail space coefficients would largely be
209
determined from a study of retail space productivity.
Space productivity is measured in terms of dollars of
sales per square foot of selling space per year. It is
assumed that overall efficiency of a retail operation is
reflected in the productiveness of its space utilization.
The validity of these space coefficients for
predictive purposes depends, as in other input-output
studies, on the assumption that technological conditions
in retail trade will remain stable or follow existing
trends for reasonable periods of time. Empirical evi
dence, offered later, indicates that retail operations
enjoy certain economies of scale, i.e., productivity of
retail space increases with scale of operation. This
implies that adjustments may need to be made for coef
ficients to accurately predict future space needs.
A study conducted by the Stanford Research Institute
for the City of Norwalk,^ offers one of the few examples
where a market potential study for a regional shopping
center has been linked with a form of space requirements
analysis. The Stanford Research study tried to determine
trade area sales potential for the proposed center under
^See William T. Kopp and Louis Salomone, Jr.,
The Market Potential for a Regional Shopping Center
in the City of Norwalk, (South Pasadena: Southern
California Labs of Stanford Research Institute,
November, 1960).
three different sets of conditions. The first set of
conditions assumed that the center had what was termed
a class three department store, which was a smaller store
capable of generating only about $70 in sales per square
foot of selling space a year. Subsequently, class two
and one department stores, generating, respectively,
$110 and $140 sales per square foot, were postulated for
the center.
The assumed department store size in each case
affected the size of the trade area and market potential
of the center. However, a thorough consumer survey of
the trade area population was not undertaken, and fairly
automatic a priori assumptions were made as to the impact
of a larger department store on market potential. Table 7
presents the findings of the Stanford Research study.
Each column of figures offers income and sales potentials,
and space requirements, for the proposed Norwalk regional
shopping center, under each of the three postulated sets
of department store size conditions.
From Table 7 it can be seen, that as department
store size and productivity increase, roughly proportion
ate increases are assumed in trade area size also, in
terms of rises in population, income, and sales potential.
With this approach, virtually any size department store
and shopping center can be justified, from the assumption
TABLE 7
NORWALK REGIONAL SHOPPING CENTER
SALES POTENTIAL AND SPACE REQUIREMENTS
Sales Potential and
Space Recruirements
Class One
Dept. Store
Class Two
Dept. Store
Class Three
Dept. Store
Trade area population 214,605 170,665 114,270
Trade area personal income $310,500,000 $244,000,000 $160,600, 000
Total retail sales potential 215,700,000 169,300,000 111,400,000
Total dept, store sales potential 24,100,000 18,900,000 12,400,000
Existing dept, store sales 5,500,000 3,900,000 3,500,000
Net dept, store sales potential 18,600,000 15,000,000 8,900,000
Sq. ft. of dept, store sales space 133,000 136,000 127,000
Sq. ft. of dept, store gross space
Sq. ft. of regional shopping center
200,000 204,000 190,000
space 433,000 442,000 411,000
Acres of parking space 35 35 33
Total center acreage 45 45 43
Total center sales potential $ 35,200,000 $ 31,900,000 $ 24,700,000
Source: William T. Kopp and Louis Salomone, Jr.,
The Market Potential for a Regional Shopping Center
in the City of Norwalk, (South Pasadena: Southern
California Labs of Stanford Research Institute,
November, 1960), pp.43, 50 and 57.
211
212
that a roughly proportionate increase in market potential
will result from a given assumed increase in store size.
The importance of the Stanford Research study lies in
its recognition of the value of using retail space
productivity data to establish standards of space use
from which space requirements can be derived. But the
analysis should include a more thorough consumer survey
of the area, to objectively determine the type and size
of department store which could be supported by existing
market potential.
Retail space input coefficients.— The coefficients
discussed above are constants of retail space input. It
is assumed that these coefficients remain stable in value
for periods long enough to permit prediction of space
requirements from income and expenditure forecasts. In
the event that these coefficients change because of scale
economies, it is assumed that these changes in value can
be anticipated from empirical data on retail space
productivity.
Much concern has been expressed over the lack of
constancy through time, and among locations, of production
coefficients. Tiebout notes that, often in input-output
studies,
...the production functions for various
213
industries are assumed to be uniform
throughout the whole country. Yet
a mere examination of Northern versus
Southern fuel bills is enough to
indicate that this is not the case.^O
There is a definite need for development of regional or
local production coefficients, since national averages
are often grossly inaccurate when applied to a particular
location. Studies to determine local or regional pro
ductivity levels in the use of retail space would be
highly valuable.
Even greater sophistication can be attained if
distinction is made between types of coefficients to be
used in different demand situations. The three basic
coefficient types are: (1) average coefficients, deter
mined empirically from the Census of Manufactures for
presently operating facilities; (2) best coefficients,
which represent the coefficients that would hold if the
newest planned facilities were in operation; and (3)
worst coefficients, which represent the coefficients of
marginal firms.
If demand levels are stable and no new facilities
are required in the near future, average coefficients
would be appropriate. If demand levels are rapidly
•^Charles M. Tiebout, "Interregional Input-Output
Models: An Appraisal," Southern Economic Journal
(October, 1957), 143.
214
rising and new facilities are anticipated, best coef
ficients should be used. If demand levels are falling,
older or marginal operations with worst coefficients are
most likely to be affected.
While noting that purchase-sale (input-output)
relationships are changeable, Gottlieb also feels that
they are fixed for periods of time long enough to permit
analysis and forecasting. He says that,
Obviously, the dollar value of purchases
and sales of one industry vis-a-vis any
other is not fixed or immutable. Shifts
in consumer preferences, new technological
processes and demand for new consumer goods
and services,...will tend to affect the
extent to which industrial activities are
related to each other in a "buyer-seller"
nexus. These factors, however, are not
generally subject to rapid changes so that
the industrial relationships expressed in
_ valid for a consider-
Perhaps Gottlieb is a bit too sanguine in his assumption
that input values are valid for a "considerable" period
of time. Coefficients may have a greater degree of
constancy in some of the intermediate demand sectors
which are farther removed from final demand. Retail
trade is relatively volatile in terms of changes in
output levels, because of its sensitivity to changes in
the levels of family income and consumption expenditures.
^Gottlieb, Journal of the American Institute
of Planners, (Fall, 1956), 231.
215
Historical data also indicate that the productivity of
retail space use has increased noticeably during the past
thirty years. So, a need for careful local study of
technical and demand conditions in retail trade is
evident.
Limitations of the input-output approach.— If the
input-output approach emphasizes detailed study of the
structure of economic relations, and internal consistency,
why should it be modified and some of these apparent
advantages lost? Mainly, it is because input-output
analysis is best suited for larger regions, and for
industrial and manufacturing activities where there exist
complex technical input-output relationships that should
be measured accurately because of the technological link
age between these industries, and the latter's signifi
cance in the provision of estimated final output.
In local (e.g., municipal) retail and consumer
studies, incomes, expenditure patterns, and resulting
local retail sales levels are the variables of interest.
Income increases arise from unexplained increases in
export demand and via the local multiplier process.
These changes are dependent on income flows and geo
graphic spending patterns, not technical relationships.
On the other hand, there are numerous reasons why
216
input-output analysis could easily be modified and
fused with economic base analysis. While some writers
emphasize the differences between these two approaches,
Tiebout points out their similarity. He says, in fact,
that,
...at both a conceptual level and at an
operational level there is no basic
difference between these two models;
and...at a less than purely formal
level there are advantages in using
the foreign trade multiplier approach.12
Consider the following example. A local income
multiplier model could state that every dollar of export
sales creates, say, $0.80 worth of local income and
$0.20 worth of imports from outside business firms. If
residents have a propensity to consume locally of .75,
then $0.60 will be spent locally, i.e., $0.80 x .75.
If, in turn, local retail sales have a propensity to
create local income of .20, then $0.12, i.e., $0.60 x
.20, becomes local income again. The remaining $0.48
goes for imports of local retail firms. These coef
ficients could just as well be placed in an input-output
grid, to show that for each dollar of export output,
$0.80 worth of input is required from households, and
$0.20 worth of input is required from imports.
12Tiebout, "Input-Output and Foreign Trade
Multiplier Models in Urban Research," Journal of the
American Institute of Planners, (August, 1957), 128.
217
These multiplier processes can be traced through
the linked industry effects of economic base studies, or
the indirect industry effects of input-output studies.
The underlying technical assumption of input-output
studies is a linear production function, i.e., constant
coefficients. An income multiplier analysis does not
imply these, but, rather, deals with consumption
functions.
Isard defines explicitly two types of multipliers.
The first is based on the linked or interindustry effect
and
...is determined by the extent to which
the final export products contain or
utilize intermediate products locally
manufactured;... The second type of
multiplier is the Keynesian-type
multiplier dependent on changes in
local income flows and determined
by the consumption habits of employees
of the export industry, of the inter
mediate industry, and of the service
industry;... 13
An input-output study analyzes the multiplier based on
the interindustry effect,.while the economic base income
multiplier study measures the effects of local consumption
habits. Tiebout notes further, that,
For large communities and, even more,
for smaller areas, the propensity to
consume local goods and services is a
vital factor in the local income
l-^Isard, Methods. . . . 143.
218
multiplier. In considering income changes,
for all areas but the largest, the linked
industries or indirect effect is not
large when compared with the consumption
function multiplier effect. Where this
is the case, a foreign trade model yields
all the information that an input-output
model yields.14
Application of an unmodified input-output model
to the study of a small geographic area with relatively
simple economic structure, is not warranted analytically.
Instead, the input-output concept should be drawn on to
provide a modified kind of input requirements analysis,
to be used as an adjunt to economic base and market
potential study. Because of the parallelism among the
elements of economic base and input-output models, as
indicated above, a combining of these two analytical
approaches is neither technically strenuous, nor
spurious in result. The combining mainly involves
changing the input-output emphasis from technical to
spending relations, while adding a technical supply ele
ment to the macro-demand emphasis of a base study.
Measuring the Productivity of Retail Space
The unit of measure.— The two most common methods
of measuring the productivity of retail operation, are
14Tiebout, The Community Economic Base Study,
(New York: Committee for Economic Development,
Supplementary Paper No. 16, 1962), p.11.
219
in terms of dollar sales per employee and per square
foot of floor area. Since this study is concerned with
the spatial dimension of retail productivity, and not
labor productivity, dollar sales per square foot of
floor area are the relevant unit of measure.
Furthermore, the unit of measure should be refined
to dollar sales per square foot of net sales area. Net
sales area is equal to gross floor area minus office,
service, and storage space. It consists of the floor
space actually used in selling merchandise to customers.
The use of net sales area is better for comparative
purposes, because various retailers in the same category
differ greatly in the amounts of service and storage
space they require. Even a store chain will have widely
varying policies on gross floor area for different units,
depending on the location and adequacy of local ware
housing facilities, type of local trade, and so forth.
Types of retail establishments measured.— The scope
of interest in retail establishment types in this study
is limited primarily to those types of stores usually
found in regional planned shopping centers. This means
that apparel, home furnishing and appliance, variety,
specialty, and full-line department stores are of prime
interest.
220
Space productivity data is available, in varying
degrees of completeness, for most of the above mentioned
store types. However, space productivity data has been
compiled historically only for department and specialty
stores. That is, only the department and specialty store
figures exist in the form of complete time series' of
trendable data.-^ other store type productivity data
are occasionally available for previous years, but not in
a continuing series compiled on an annual basis.
Department store space productivity figures are
also compiled and tabulated by many more characteristics,
e.g., sales volume, age, location, profitability, etc.,
than are data for any of the other store types usually
found in regional shopping centers. For these reasons,
and because department stores are representative of many
store types as a result of their merchandise mix, this
study concentrates on a thorough measurement and analysis
of department store space productivity. Since such a
relatively large proportion of a shopping center’s space
is occupied by the main department store, an accurate
•^See Malcolm P. McNair, Operating Results of
Department and Specialty Stores. 1953-1962, (Boston:
Harvard University, Bureau of Business Research Bulletins
Nos. 156-165), and the Controller's Congress,
Merchandising and Operating Results of Department Stores
and Specialty Stores, (New York: National Retail
Merchants Association, 1950-1965).
analysis of department store space productivity is
important. The same measurement and analytical approach
is applicable to the other store types, to the extent
that space productivity data can be made available for
them.
Measurement of retail space productivity.— Virtua1ly
all of the department store data has been generated from
two primary sources— McNair's Harvard study series and
the Retail Merchants' Association's Merchandising and
Operating Results series. Department store space product
ivity data from the MOR series are presented in Table 8,
by sales volume class for the period from 1928 through
1963. These dollar per square foot figures are median
values for each sales volume class and represent "typical"
productivity levels.
The data in Table 8 indicate that the dollar sales
volume capacity of department store selling space tends
very definitely to rise with increases in size of opera
tion. This implies an increasingly efficient utilization
of department store space as scale of operation increases.
Moreover, there has been a steady rise in space product
ivity over time, after allowance has been made for short
term drops and rises in sales volumes because of sharp
downturns and upturns in consumer demand, e.g., the 1930's
222
TABLE 8
DEPARTMENT STORE SPACE PRODUCTIVITY 1928-1963
Sales Volume Category
$1-2 $2-5 $5-10 $10+ — $10-20 $20-50 $50+
Year million million million million million million mxllion
(Dollars per square foot of selling space)
1928 22 34 39 51
1929 28 30 48 50
1930 25 30 46 54
1931 23 31 37 40
1932 20 27 31 36
1933 20 26 29 35
1934 25 28 37 37
1935 23 28 34 37
1936 25 29 34 41
1937 25 29 34 41
1938 23 28 35 39
1939
— — — —
1940 22 28 38 41
1941 25 31 41 48
1942 29 38 45 53
1943 34 45 55 60
1944 39 47 62 67
1945 36 50 59 76
1946 66 63 67 89
1947 55 60 73 91
1948 47 58 76 93
1949 51 59 70 79 91 113
1950 54 62 70 79 97 104
1951 61 59 74 75 89 111
1952 70 60 76 76 88 105
1953 69 71 75 78 91 104
1954 52 68 66 75 86 114
1955 57 68 71 78 93 114
1956 52 70 73 77 88 103
1957 55 65 73 75 83 105
1958 54 65 70 72 81 105
1959 61 66 75 77 75 104
1960 58 63 65 75 72 98
1961 57 58 68 72 73 94
1962 56 58 64 69 70 90
1963 63 60 72 67 70 85
Source: Controller's Congress, Merchandising
and Operating Results of Department Stores and
Specialty Stores, (New York: National Retail
Merchants Association, 1929-1964).
and immediate post-World War II period, respectively.
Economies of large scale operation, coupled with
the post-war technological advances in retail trade, main
ly in the form of planned shopping centers, have caused a
significant trend toward consolidation of retail operation
into larger, more centrally located shopping center units.
Average retail store size has doubled since 1945. In
creasing sales to store ratios are an indication of in
creasing scale of retail establishments. There has also
been a decrease in the number of stores per 1,000 persons
in recent years in the United States. Jefferys and Knee
note that empirical evidence for Europe shows a decided
trend toward larger average size for retail establish
ments during the past thirty years. This trend has been
even more pronounced in the United S t a t e s . 16
Within a department store, productivity of space
use varies greatly among departments. The department
with highest sales capacity per square foot of selling
space is usually the women's hosiery department, which
can generate around $335 per square foot in sales in a
year. The lowest per square foot sales generator is
usually the furniture department with perhaps $25 per
James B. Jefferys and Alfred Knee,
Retail Trade in Britain 1850-1950, (Cambridge: Cambridge
University Press, 1954), p.28.
square foot a year. Men's and women's clothing depart
ments have sales capacities falling somewhere between
these two extremes. A men's clothing department might
generate sales of $80 to $140 per square foot a year,
while a women's clothing department might expect some
thing between $60 to $125 per square foot a year.
Just as there are inter-departmental, intra-store
space productivity differentials, there are also inter
store or inter-shopping center differences. An average
space productivity figure for an entire department store
suppresses a wide range of variation in space use
capacities. In similar fashion, median space productivity
data, such as those presented in Table 8, do not reveal
the productivity range of retail operations. A median
figure simply indicates the productivity level of those
stores in the "dead center" of a distribution. To
establish space requirement standards for a new shopping
center, it is more important to know what the upper range
productivity levels area. What is the sales volume
capacity per square foot in stores which are the most
efficient in space utilization?
McNair's studies have included data on middle range
(first and third quartile) values, as well as median
figures. Table 9 presents, for purposes of comparison,
productivity figures for the low and high ends of this
TABLE 9
RANGE OF DOLLAR SALES PER SQUARE FOOT OF DEPARTMENT STORE SPACE
VOLUME CLASS 1959 1960 1961 1962
$1-2 Million low middle hioh low middle hiqh low middle hicrh low middle hicrh
Gross space 30 39 47 29 37 45 35 40 45 29 37 39
Sales space 50 61 71 44 58 71 47 58 66 44 56 67
Sales/Gross (%) 58 67 76 59 64 73 58 65 73 60 64 69
$2-5 Million
Gross space 32 41 48 32 40 48 31 36 44 30 35 43
Sales space 55 66 75 54 63 75 51 58 66 52 58 69
Sales/Gross (%) 56 64 71 56 63 69 58 65 71 58 64 73
$5-10 Million
Gross space 38 46 58 32 40 46 33 44 50 32 38 44
Sales space 60 75 99 58 65 70 60 67 88 51 64 82
Sales/Gross (%) 54 58 63 54 57 66 54 59 68 52 56 62
225
TABLE 9— Continued
VOLUME CLASS 1959 1960 1961 1962
$10-20 Million low middle high low middle high low middle high low middle high
Gross space 33 42 50 34 40 48 32 38 42 30 35 40
Sales space 67 77 89 64 75 80 62 73 80 61 69 78
Sales/Gross (%) 50 55 58 50 54 61 49 52 58 49 51 54
$20-50 Million
Gross space 35 41 45 34 38 42 35 38 43 32 37 41
Sales space 65 75 80 66 72 78 66 73 78 62 70 75
Sales/Gross (%) 48 52 60 50 52 54 49 52 54 48 52 56
$50+ Million
Gross space 45 52 65 40 48 63 41 45 49 38 44 48
Sales space 95 104 117 85 98 112 81 95 105 81 90 100
Sales/Gross (%) 47 51 56 40 49 54 44 49 58 44 49 54
Source: Malcolm P. McNair, Operating Results of
Department and Specialty Stores, (Boston: Harvard
University, Bureau of Business Research Bulletins
Nos. 162-165, 1960-1963).
226
H 227
middle range, as well as for the middle. These data
include values for both gross and net selling space, and
percentages of gross space devoted to selling.
Prom the data in Table 9, it can be seen that the
rise in space productivity with increase in scale of
operation, is more readily apparent from the sales per
square foot of selling space data, than from gross space
data. The sales per square foot of gross floor area are
not as sensitive to scale economies because of the varied
policies in department stores as to the percentage of
total space devoted to non-selling activities. Smaller
department stores tend to devote a larger percentage of
their gross floor to selling activities. These smaller
operations, say, under $10 million sales per year, devote
around 55 to 75 per cent of their total space to selling.
The main reason for this is that smaller stores must draw
on external economies available to them, e.g., central
warehousing facilities of wholesale distributors.
Larger department stores, in the over $10 million
sales range, are located either in the central business
district or in large and widely scattered regional shopp-
ing centers. These larger operations devote only about
40 to 60 per cent of their total space to selling activi
ties. This is because large department stores often
internalize many of the external facilities that smaller
i
stores draw on freely. Relatively large amounts of space
are devoted to non-selling activities in a department
store in a regional center, because a large scale opera
tion requires more storage and scheduling facilities to
maintain a smooth flow of merchandise to the selling
departments.
The data in Table 9 show the large spread between
the productivity values for the low and high ends of the
middle range. There is wide variation in the efficiency
of retail space use, even among department stores in the
same size category. There is significant difference,
too, between median sales capacity figures and the high
values. For this reason, the high end values are probably
much more relevant when productivity standards are being
set for proposed shopping center facilities.
For general reference. Table 10 presents sales
capacities per square foot of net sales area that are
considered to be adequate for use in planning new shopp
ing center facilities. These figures were offered by a
prominent market analyst in 1958.
These productivity standards offered by Nelson
should not be viewed as fixed standards to be rigidly
adhered to. As previous data have indicated, careful
study must be made of each retail location and develop
ment decision. There can easily be wide variation among
229
TABLE 10
SALES CAPACITIES
PER SQUARE FOOT OF SELLING SPACE
FOR PROPOSED SHOPPING CENTER FACILITIES
Store Cateqory
Dollar Sales per
Square Foot of
Selling Space
per Year
Convenience Stores
Drugs 150
Grocery 200
Other food 175
Liquor 150
Hardware 45
Shopper s' S tor e s
Department stores 65 - 100
Family and other apparel 65
Mens' clothing 60
Mens' shoes 70
Womens' shoes 75
Womens' apparel 75
Variety stores 50
Jewelry 75
Furniture and
household furnishings 50
Household appliances 150
Eating and drinking places 80
General Purchases
Auto parts and accessories 60
Source: Richard L. Nelson, The Selection
of Retail Locations. (New York: F. W. Dodge
& Co., 1958), p.224.
i
applicable space productivity standards for different
market situations.
Other retail operating characteristics and
determination of space productivity standards.— Be
fore moving on, a couple of other refinements in setting
space capacity standards should be noted. McNair attempts
to set "goal" productivity figures to compare with
"typical" (i.e., median) performance figures. The goal
figures are not upper quartile values but, rather, are
values for department stores whose pretax earnings were
in the top quarter of the array for that figure in
McNair's sample of stores. McNair says that,
The comparisons were more clear-cut
with respect to sales per employee
and sales per square foot, both of
which revealed 'goal' figures higher
than typical figures except in the
case of the $1 to $2 million class.17
McNair uses dollar sales per square foot of gross
area but, nonetheless, his data indicate significant
differences between the space productivity of the most
profitable stores and the others.
Another important factor in establishing standards
of space efficiency is the difference between main and
branch department store productivity figures. In recent
17
McNair, Operating Results...1962, 38.
231
years sales per square foot in all volume classes have
been about 20 per cent higher for main than for branch
stores. This stems largely from the fact that suburban
trade areas are often quite a bit less densely populated
than central business district trade areas, and are much
more extensive in geographic terms. Since most new large
department stores are located in outlying suburban shopp
ing centers, this factor should be taken into considera
tion.
Retail Operation
Retail shopping center and store types.— Even though
most retail activities have many economic characteristics
in common, there are meaningful differences between
various types of retail operation. Regional planned
shopping centers, and the types of retail stores found
in them, are of main interest in this study. To aid in
visualizing the position of regional planned shopping
centers on the spectrum of shopping center types,
Tables 11 and 12 offer comparative information on the
type and location of centers, and type and composition
of centers, respectively.
As indicated by Table 11, shopping areas can be
either unplanned or planned. The term "planned" or
"controlled" shopping center usually refers to a group
TABLE 11
RETAIL CENTER CLASSIFICATIONS AND LOCATIONS
Unplanned
Business Centers
Planned
Business Centers
1) Central business district
(CBD) centers.
2) String store adjoining CBD.
3) Secondary business district
(SBD) serving other parts of
central city or suburbs.
4) String stores adjoining SBD.
5) Neighborhood stores in small
clusters or isolation.
6) Outlying highway stores in
strings or isolation.
1) CBD planned center from
urban renewal.
2) Regional planned shopping center
in strong competition with CBD.
3) Community planned shopping center
in competition with SBD's and
string developments.
4) Neighborhood planned shopping
centers in competition with
neighborhood stores.
5) Outlying planned shopping center
drawing from passing highway
traffic.
Source: Eugene J. Kelly, Shopping Centers—
Locating Controlled Regional Centers,
(Washington, D.C.: Eno Foundation for Highway
Traffic Control, 1956).
232
TABLE 12
COMPOSITION AND CHARACTER OF MAJOR SHOPPING CENTER TYPES
Center
Type
General
Character
Source of
Customers
Store
Types
Goods
Sold
Retail CBD Business and
recreational
heart of a
metropolis.
Come from all
parts of the
city. Sites
are accessible
to most customers
Largest in floor Shopping goods
space and volume, emphasis. Area is
Multi-story de- center of apparel,
partment stores, furniture and other
.Leading specialty department store
shops. lines.
Controlled
Regional
Shopping
Center Overall unity Draws from Attempt to One or two depart
obvious from families in 30 duplicate ment store branches
integrated land minute driving shopping and satellite stores
scaping and radius. facilities of offer widest range
design. Harmon Customers come CBD with mini of merchandise
ious effect is from several mum overlapping. outside of CBD.
object. May suburban One stop shopp
serve as area's communities. ing in suburbs.
civic and Pull varies with
cultural center. effectiveness of
CBD and competing
centers.
233
Source: Eugene J. Kelly, Shopping Centers— Locating Controlled
Regional Centers, (Washington, D.C.: Eno Foundation for Highway Traffic Control,1956).
of stores planned as a unit and under single ownership.
Individual stores in the center are leased to various
retail operators by the shopping center developer.
McKeever defines a planned center as a
...group of commercial establishments
planned, developed and managed as a
unit, with off-street parking provided
on the property, and related in location,
size and type of shops to the trade area
that the unit serves— generally, in an
outlying, suburban area.18
Both Table 11 and Table 12 indicate that regional
planned retail centers compete very strongly with retail
facilities located in CBD*s. Regional shopping centers
are an attempt to duplicate downtown shopping goods
offerings at various locations around a large metropolitan
area. Even though there may be an attempt to minimize
trade area overlap between CBD and given regional retail
facilities, as indicated in Table 12, the total effect
of ringing the CBD with outlying regional centers is to
siphon off most of the CBD1s original market potential.
Large regional centers even offer competition to nearby
convenience facilities because of their one-stop shopping
appeal.
18
Ross J. McKeever, Shopping Centers: Principles
and Policies, (Washington, D.C.: Urban Land Institute,
Technical Bulletin No. 20, 1953), p.6.
235 ’
McKeever offers a comprehensive functional catalog
ing of shopping center types. He arranges them in a
hierarchical manner according to the extensiveness of
their trade functions.
Neighborhood centers— These centers provide
for sale of daily living needs (convenience
goods)— foods, drugs, sundries, personal
services; include 10 to 15 stores; require
at least 1,000 families for support and
need 5 to 10 acres for site area.
Community or Suburban Centers— (sometimes
District Centers). These centers, in
addition to convenience goods and personal
services, provide for sale of soft lines
(apparel, etc.) and hard lines (hardware,
equipment, etc.); include 20 to 40 stores
(a junior department store may be included);
require at least 5,000 families to draw
from and a site area of 10 to 25 acres or
more.
Regional Centers— These centers have at
least one department store for their core
and offer shopping goods in full depth
and variety including fashion items and
house furnishings; include 50 to 100 stores;
require more than 100,000 people (250,000
people, if very large) for support; require
35 acres as a minimum site area (areas over
80 acres begin to get unmanageable for
pedestrian distances). The shopping
provided for in these centers is often
called "one-stop shopping"— meaning one
stop of the car to let the customer do
all kinds of buying in one p l a c e . . . 19
Each center of higher rank in the retail hierarchy takes
on new and more extensive trading functions, while still
including the functions of lower rank centers.
l^Ibid.
i
Converse notes that "The regional center has stores
selling all kinds of fashion goods...and usually also
many kinds of convenient goods."20 Regional shopping
centers compete on most retail trading levels. Shopping
goods competition is, however, the most significant.
A typical tenant mix for a moderate size regional
shopping center is presented in Table 13. As seen from
the stores represented in Table 13, a regional shopping
center provides a variety and depth of shopping goods
comparable to CBD retail facilities. These include
general merchandise, apparel, and home furnishings, as
well as a variety of services. Recreational facilities
may also be included. One or more department stores are
the principal tenants in a regional center. Normally,
each full-line department store will have a gross enclosed
floor area of at least 100,000 square feet.
It is obvious that the department store dominates
a regional center, in terms of its proportionate share
of total gross square footage. In the hypothetical
center of Table 13, the department store alone accounts
for over one half of total gross floor area. The depart
ment, variety, and apparel stores together constitute the
2®Paul D. Converse, Harvey W. Huegy, and
Robert V. Mitchell, Elements of Marketing.
(Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1958),
p. 32.
TABLE 13
TYPICAL TENANT COMPOSITION
OF A REGIONAL SHOPPING CENTER
Tenant
Category
Tenant
Classification
Square
feet
of
gross
area
Annual
dollar
sales
per sq.
foot of
gross
area
Apparel and
shoes Ladies 1 wear 6, 700 54
Mens' wear 5,500 63
Family shoes 4,600 48
Ladies' specialty 4,600 53
Ladies' shoes 5,900 48
Childrens' wear 4,000 38
General
Merchandise Variety store 26,000 29
Department store 168,200 59
Food and
food
service Supermarket 23,900 107
Restaurant 4,500 76
Candy, nuts 1,200 62
Bakery 2, 000 66
Services Beauty shop 1, 600 46
Barber shop 850 53
Cleaners 1, 700 29
Other Retail Drugs 9,200 62
Jewelry 2, 300 65
Cards and gifts 1, 700 39
Bank 5,800 —
Source: Urban Land Institute, The
Dollars and Cents of Shopping Centers
Part Two: Tenant Characteristics,
(Washington,D.C.: Urban Land Institute, 1962),
p.16.
predominant "drawing card" of a regional center. These
are the stores that attract consumers to a regional
center to make selective purchases requiring relatively
large amounts of shopping time.
Proper proportioning of store types in a regional
center is important in attracting customers on a continu
ing basis. While Hoyt has emphasized the need for a
large well-known department store to serve as major
tenant in any proposed regional center, he also warns
that,
The department store should not be
top-heavy and its area normally
should not exceed half of the total
store area of the center. If there
are too few other stores in fashion
goods lines besides the department
store, shoppers will not patronize
them extensively. The department
and apparel stores should occupy
70 per cent or more of the total
building area in a regional shopping
center.21
The department-variety-specialty-apparel store complex
within a regional center is comprised of store types
whose individual market potentials are each greater when
they are parts of a center complex, than when they oper
ate in isolation. This cumulative force of attraction
can be enhanced, and its effect maximized, only when the
2lHomer Hoyt, A Re-Examination of the Shopping
Center Market, (Washington, D.C.: Urban Land Institute,
Technical Bulletin No. 33, 1958), p.9.
239
component stores of a shopping center complex are re
presented in proper proportion.
The changing character of retail store types.— Store
types can pose a problem for retail analysis. Often,
analysts use the pure store types developed by the Depart
ment of Commerce for census purposes. It is difficult
to avoid use of pure store types, if sales volume and
space estimates are to be made on a basis where compari
sons from one study to another can be made, and general
izations drawn. However, many newer stores contain a
combination of these pure types, such as the new super
drug stores which carry many variety, food, and liquor
items in addition to the traditional drugs and fountain
service.
In the United States, Converse says that,
The tendency for 'scrambled' lines in
retail outlets appears to be a continuing
one; by 'scrambling' we mean the addition
of lines of merchandise by stores not
formerly stocking that line, or the
addition of lines not usually thought of
as being standard for stores of that kind.22
Jefferys and Knee note that in Europe there also is
"...a trend away from specialist retailing to mixed
merchandising."23 The larger stores are each carrying
22converse, Huegy, and Mitchell, Elements..., 313.
23Jefferys and Knee, Retail Trade..., 48.
i
240
a greater number and variety of items, within limits.
In other words, previously identifiable store types
are still recognizable.
Most retail stores are getting larger and the
lines of demarcation between types of stores are becoming
progressively blurred. In some instances, as Silverman
says,
...apparel and variety chains are turning
into junior and even full-fledged department
stores. Indeed, it sometimes seems that
every retailer is trying to become a
department store.24
Because of this phenomenon it is not always a simple
task to establish space productivity standards for vari
ous distinct types of retail operation. This is one
reason why the present study concentrates on department
store space productivity standards. Aside from data
availability, the department store is representative of
the numerous specialty, variety, and apparel store types
which populate a regional shopping center.
Department stores.— Large departmentalized retail
stores have been around a long time. The first depart
ment stores appeared around the middle of the nineteenth
century in Paris, New York City, Boston, and London.
24charles E. Silverman, "The Department Stores
are Waking Up," Fortune, (July, 1962), 252.
i
They evolved from general and apparel stores.
Large departmentalized retailing was made possible
by several developments. Urbanization meant large con
centrations of population in small areas. Industrial
development led to an increasing flow of manufactured
goods. Organizational improvements enlarged the scope
of owners and managers to the point where they could
supervise larger numbers of employees. Improved urban
transportation aided development of centralized shopping
districts.
Since their inception, department stores have be
come a very important retail type because, as Converse
points out, "...they do about 6 per cent of the retail
business in the United States."^ Table 14 compares
department stores' proportionate share of total retail
sales for the United States, California, the Los Angeles-
Long Beach metropolitan area, and the City of Los
Angeles, during the last three retail census years of
1954, 1958, and 1963.
In Table 14 it can be seen that as geographic size
of area decreases, department store sales as a percentage
of the total rise. The disproportionately large amounts
accounted for by department store sales in, say, the
25Converse, Huegy, and Mitchell, Elements..., 331.
242
TABLE 14
DEPARTMENT STORE SALES
AS A PER CENT OF TOTAL RETAIL SALES
Area
1954
Per Cent
Years
1958
Per Cent
1963
Per Cent
United States 6.2 6.7
California 9.0 9.4 10.1
Los Angeles-Long Beach
Metropolitan Area 7.7 7.8 11.5
City of Los Angeles 10.7 8.8 12.5
Source: U. S. Bureau of Census,
Census of Business: 1954, 1958, 1963.
Volume II, Retail Trade— Area Statistics.
City of Los Angeles, arise because local department stores
are serving trading areas that in many instances, lie, to
a great extent, beyond Los Angeles city boundaries. In
other words, smaller geographic areas are not perfect
microcosms of the national economy, i.e., economically
balanced and serving largely their own needs. As for
the State of California, the higher share for department
stores may be largely because of the higher than national
average incomes of California residents. Studies in
dicate that the higher the income level the higher the
rate of department store purchase, other things equal.
243
The truly unique feature of a department store
as a retailing type, is its relatively large scale of
operation. A department store must be large enough to
operate efficiently on a departmentalized basis of
merchandising; which is appropriate only to relatively
large scale enterprises. A large operating scale leads
to many economies yielding a competitive advantage to
department stores. Integration of functions, purchasing
power, expert buyers, etc., enable department stores to
secure quality merchandise in wide assortments at low
prices. Employment of specialists enables each to do a
good job in his merchandising specialty. The variety of
goods in various departments often induces consumers to
make several purchases of different kinds during a single
shopping trip to a particular store.
For these reasons department stores serve as
"anchor" establishments in regional shopping centers.
Normally, shopping center developers attempt,
...to develop shopping centers with a
department store as its chief focal
point, thus establishing the department
store as a big 'drawing card' for
shopping goods customers.26
This is not to imply that department stores lack
26Robert D. Entenberg, The Changing Competitive
Position of Department Stores in the United States
! by Merchandise Lines, (Pittsburgh: University of
Pittsburgh Press, 1957), p.102.
244
important competitive disadvantages. They do not.
Converse points out that,
The high operating expenses of the
department store are an important
limitation. The expense ratio in
turn requires a relatively high
margin, which is a handicap in
meeting price competition of more
agile competitors. Fairly rigid
store-wide margin requirements
existing in many department stores
are also a handicap in meeting
competition in specific departments,...27
This means that department stores must achieve the most
productive use of their physical plant possible, since
their expenses per square foot of floor space tend to
level off after a certain size, i.e., the impact of scale
economies begins to be felt. Because of the leveling off
of operating expenses and the rise in sales capacity per
unit of floor space as scale of operation increases,
department stores are able to overcome their competitive
cost disadvantages to a large extent.
Factors determining the scale of retail operation.—
Given the cost structure of retail operation, there is,
theoretically, some minimum or threshold level that
market demand must attain for particular classes of
goods before retail operation commences. This threshold
level represents the minimum operating size, below which
21
*'Converse, Huegy, and Mxtchell, Elements..., 338.
n
245
retail facilities will not remain in existence in the
long-run.
Given the demand preferences of consumers in a
trade area, the maximum profit scale of retail operation
is, theoretically, determined at the point where the
marginal cost of selling a unit of product is equal to
the marginal revenue derived from its sale. The size of
the trade area for a given shopping center is, of course,
related to some extent to the size of the center and its
store and merchandise composition (it is assumed here
that only regional centers of varying size are being
considered, and not centers of different rank in the
hierarchy). The larger and more diverse is a retail
center of a given class, the greater are the potential
expenditures it can expect from its surrounding trade
area, up to p limit.
While department stores and other shopping center
facilities may enjoy economies of large scale operation,
there is also the declining utility to consumers of
ever-larger centers to be considered. Huff says that
empirical evidence indicates the marginal utility of
shopping centers to consumers
...increases rather slowly for the
smaller size centers, rises more
markedly at some larger size level,
and then drops off again with
successively larger size c e n t e r s . 28
Since Huff is discussing centers of different rank in
the hierarchy (i.e., neighborhood, community, and regional
centers) as size increases, his evidence indicates that
regional centers would be exclusively in that range of
the marginal utility curve where utility increments from
increased size were rapidly diminishing.
As the utility of larger centers to consumers de
clines, given size increases will yield smaller and
smaller percentage increases in market potential. Search
ing and time costs to a consumer looking for a shopping
good within a center increase as size increases, and
competition from other centers limits market potential
levels. Lakshmanan and Hansen point out that,
"Conceivably, these costs could increase in a system of
very large centers to a point where consumers may prefer
smaller centers."29
Without precise information on the shapes and
levels of retail cost and revenue functions, specifica
tions of minimum and maximum operating levels, that are
2^David l. Huff, Determination of Intra-Urban
Retail Trade Areas, (Los Angeles: U.C.L.A., Real Estate
Research Program, 1962), p.31.
29t. R. Lakshmanan and Walter G. Hansen, "A
Retail Market Potential Model," Journal of the American
Institute of Planners, (May, 1965), 138.
247
economically feasible, must be related to observed trends
in retailing. For example, as suggested earlier by
McNair, goal productivity levels might be set for given
types of retail operation. These goal levels would be
derived from dollar sales capacity per square foot of
stores making the most efficient use of their physical
plant. Or, the goal values might be modified by using
the most profitable firms as standards of operating
efficiency (rather than merely the most productive ones,
regardless of profitability).
Using this empirical approach, some orders of
magnitude can be established for retail space product
ivity, pertaining to stores and centers of various size
and type. The sales capacities per unit of floor space
required to support a viable retail operation can be
obtained and applied to proposed shopping center develop
ments. If a shopping center were assumed at a certain
size and composition, and a field study and market
potential model analysis indicated that the center would
not achieve a viable minimum overall sales level, say,
one hundred dollars per square foot, then some alteration
in the center's proposed size and tenant mix might be in
order. "Thus, this criterion helps to determine the
scale at which a shopping center is viable."30
30Ibid.. 139.
r
248
Market Structure and Retail Location
Retail development and location decisions;
market potential and rent paying capacity.— The
factor of overriding importance in retail store and center
location decisions is access to potential customers.
Private shopping center developers are primarily interest
ed in locating their projects at places where they maxi
mize their access to a given population in the surrounding
area. If land rent is relatively high at the location of
maximum market access, it should be remembered that a
retail center's sales' capacity per unit of space util
ized is greater at this location and, therefore, its
rent paying capacity per unit of space will also be
greater.
Retail development and location decisions can be
quite complex, because not all stores in the same shopp
ing center sell to the same trade area. Two or more
stores may even divide up a larger trade area and each
perhaps sell exclusively to certain segments of it. In
terms of establishing minimum market potential levels,
Simmons feels that,"This bears out the necessity of
considering the threshold supporting population of a
store to be the minimum amount of access to a certain
set of people."3^
The number of retail establishments in a market
area selling a particular set of goods, say, mens'
apparel, equals the total purchasing power of the area,
available for mens' apparel purchases, divided by the
threshold requirement for mens' apparel stores. Any
remainder of purchasing power would be divided up among
existing stores as excess profit. "The maximum number
of establishments of a business type in a central place
can never be greater than the number of thresholds units
contained in the hinterland of that central p l a c e . " 3 2
The stores with lowest thresholds will be widely dis
persed, and perhaps grouped together with other conven
ience stores in neighborhood centers; while the stores
with highest thresholds will be centrally located in
regional centers.
Location and growth patterns for new and existing
shopping centers will, of course, be dynamic, shifting
as threshold requirements for a center change and as
population and income structure change. As Simmons says,
Centers will expand and diminish, and
lose or gain rank in the hierarchy as
the composition and range of their trade
area varies (trade area may be defined as
31James Simmons, The Changing Pattern of Retail
Trade, (Chicago: University of Chicago Press, 1962), p.49.
32Ibid.. 29.
the space to which they are more
accessible than any other center of
at least the same order).33
Retail shopping center development decisions, then,
are based on observations of: relative rates of popula
tion growth and redistribution; changing income levels
and expenditure patterns; and competitive environments
surrounding proposed sites. Differential investment
returns are then estimated for various potential sites.
Returns on investment depend on relative land costs
(rents) for different sites, improvements required, taxes
additional advertising costs (which depend on location),
and so forth. The decision as to which site should be
developed, depends on an expected minimum rate of return
at a particular location, relative to returns available
at known alternate locations.
Emphasizing the variability of retail sales levels
with changing location, Alonso points out that the costs
of friction— rent and transport costs (i.e., travel
costs to consumers in reaching a retail location)— are
not sufficient to determine retail location.
The view of the firm as locating so as to
minimize its costs of friction is acceptable
in the cases of agriculture and of a manu
facturing firm in a market of perfect
competition, but it is not acceptable for
the case of a retail firm. The volume of
sales of a retail store will vary with its
33Ibid., 167.
location, and the firm must weigh the costs
of friction against this factor. The costs
of friction are not, by themselves, sufficient
to determine location, unless changes in the
volume of sales are themselves somehow
considered to be "costs of friction."
More generally, the minimizing of certain
costs, such as the costs of friction, is
a valid criterion for location only in
cases where revenue and all other costs
are constant...34
For retail establishments, the sum of the two items (the
costs of friction) is not a constant, but, rather, it
varies with the site. Once threshold purchasing power
is attained, a store's location is governed to varying
degrees by accessibility to this purchasing power. From
knowledge of its customers each firm, consciously or not
defines a surface of rent paying capacity over the city,
on the basis of the sales it can expect at various place
Making comparisons with actual land rents across the
city, a retail firm will select locations where its rent
paying capacity is competitive with or higher than any
other land use.
Again, minimum sales' revenue levels for different
locations can be expressed in terms of dollar sales
volume per square foot of sales space. Comparison with
established standards of space productivity will permit
determination of whether or not a viable operation is
■^^William Alonso, Location and Land Use,
(Cambridge: Harvard University Press, 1964), pp.7,8.
feasible. If market potential at a given location is
not sufficient to support efficient utilization of the
physical retail plant required to supply it, then an
alteration in either size or location is implied.
Recent trends in retail location patterns.— Between
1920 and I960, the proportion of total population in the
United States residing within metropolitan areas in
creased dramatically from 51 per cent to 70 per cent.
Before the Second World War most people depended on mass
transit for intra-urban travel, and shopped downtown at
the hub of the transportation system which radiated out
in all directions.
After 1945, the proportion of people in metro
politan areas was increasing rapidly, but their dis
tribution within these areas was altering greatly. A
movement to the suburbs was pronounced with central city
population and retail sales declining— even in absolute
terms.
Transportation by private automobile largely re
placed public transportation systems in many areas.
Large suburban shopping centers arose during the post
war period, to accommodate shoppers arriving by car, for
both shopping and convenience goods. This vast intra
metropolitan redistribution of population profoundly
253
affected the retail trade industry. McNair notes that,
...department store suburban branches,
only a sporadic development prior to
World War II, began to mushroom in the
1950's as department stores increasingly
became multi-unit organizations... the
bold concept of planned shopping
centers, regional, community, and
neighborhood, took hold of the
imagination of real estate developers,
architects, and merchants alike;...35
Decentralization of population in urban areas has been
accompanied by a major dispersal of retailing activity.
The burgeoning outlying shopping centers plus many of
the established suburban retail areas are increasing
their proportion of total retail sales.
Increasing size of retail stores and centers, on
the average, indicates their growing dependence on wider
trade areas made possible by automobile transportation.
Consumption propensities have also shifted heavily
toward shopping goods in response to rising incomes.
For these reasons, planned regional shopping centers have
had their roles sharply defined. They carry only a mini
mum amount of convenience goods, isolate themselves from
office and other activities, and become specifically
retailing locations.
With the movement of retail trade away from central
shopping districts, many department stores have establish
ed branches in new or established suburban shopping
^5McNair, Operating Results... 1962,
centers. In many instances, department stores have
taken the lead in developing new centers. Additional
downtown sales volume can often only be obtained by
greatly increased operating costs, which may more than
outweigh the advantages of achieving optimum size for
maximum efficiency and productivity.
To quickly increase the ratio of sales volume to
operating costs has required moving to the suburbs where
the sales potential is geographically situated. In re
cent years, branch department stores, developed in local
suburban trading areas, have represented the major means
of department store geographical expansion. As Entenberg
states it,
Department store geographical expansion
into suburban areas with branch stores...
has become the chief means of department
store growth in the United States.36
Most branch department stores are located in the
same metropolitan area as their main (downtown) store.
About one half of the branches are being located in un
planned commercial areas, and one half in planned shopp
ing centers. Brown and May indicate that
The younger, the larger, and the
'most profitable' branches were
more frequently constructed in
3^Entenberg, The Changing Competitive
Position..., 100.
planned shopping centers, particularly
regional centers, while the older,
smaller branches were more frequently
located in unplanned central or other
commercial areas.37
Table 15 indicates the distribution of the most
profitable branch department stores among unplanned and
planned shopping centers. Brown and May define these
most profitable branches shown in Table 15 as those
reporting a net operating profit of over five per cent— a
breaking point approximately at the top of the middle
range (third quartile) in their sampling of stores. The
percentages in Table 15 show that a greater proportion
of the most profitable department store operations have
located in planned shopping centers than all branch
stores, on the average. The fact that there are more
profitable branch operations in planned shopping centers
than elsewhere, provides some indication that these
branches are probably both attaining higher sales levels
and making more efficient and productive use of their
physical plant.
Since shopping center development represents the
most significant form of retail store expansion, the
analytical procedures for economically allocating retail
^Milton P. Brown, and Eleanor G. May,
Operating Results of Multi-Unit Department Stores.
(Boston: Harvard University, Bureau of Business
Research Bulletin No. 159, 1961), p.12.
256
TABLE 15
DISTRIBUTION OP THE LOCATIONS OF THE MOST PROFITABLE
| DEPARTMENT STORES AMONG UNPLANNED AND PLANNED SHOPPING
CENTERS
Per Cent of Per Cent
most profitable of all
Type of branches branches
Store Location locatinq locating
Unplanned areas (total) 43.8 48.3
CBD 18.9 27.3
Other unplanned 24.9 21.0
Planned shopping centers (total) 54.1 48.2
Neighborhood or community 21.6 18.9
Regional 32.5 29.3
No answer 2.7 3.5
Total 100.0 100.0
Source: Milton P. Brown and Eleanor
A. May, Operating Results of Multi-
Unit Department Stores, (Boston:
Harvard University, Bureau of Business
Research Bulletin No. 159, 1961), p.65.
space, presented in the following chapter, are developed
primarily for application at the municipal (small geo
graphic) scale of economic analysis and land use planning.
CHAPTER VI
RETAIL SPACE ALLOCATION
This chapter is intended to supply a background
discussion of local land use planning analytical require
ments, and a combining of the analytical components
presented in Chapters III through V, to meet these ana
lytical needs.
Fairly detailed presentation is made concerning
the geographic scale of planning viewpoints, and types
of allocative decisions occuring in the local land use
planning process.
A retail space allocation model is then constructed
and applied to a municipal retail land use allocation
problem. In construction of the model, macro-demand,
micro-demand, and space-input components are each
systematically treated, and drawn together to provide
a model of retail space allocation on a municipal scale.
Finally, data requirements of the model are
discussed briefly. Attention is given to both secondary
data and the generation of primary data from field
surveys. Examples of questionnaire design are presented
in an Appendix to the study.
258
Local Public Planning Needs
The geographic scale of local government planning
decisions.— Since economic analysis for planning
decisions can be conducted at virtually any geographic
level (national, regional or state, metropolitan, and
local or city), it is important to specify the level,
and relate the analysis to that level's planning
problems.
National economic analysis and planning has largely
ignored the spatial dimension of planning problems.
That is, economic analysis at the national level has
tended to concentrate on levels of economic activity,
their determination, and the way they change through
time, while abstracting from the spatial distribution
of economic activity. More recently, Loesch, Isard,
and others, have attempted to redress this imbalance by
emphasizing the spatial dimension in economic and social
analysis.
Much of Isard's analysis, for example, is aimed
at regional planning problems. Perhaps, theoretically,
' it appears valid to approach spatial planning problems
at the regional level. Isard may feel that an integrat
ed resource region is a more rational unit of study than
a fragment of a larger economic entity. In this sense
he is probably correct. However, much of the economic
information generated by a regional study is not directly
useful at the local or sub-regional level. Conclusions
about the nature and rate of regional economic growth
may be necessary information for the local planning
process? but they may not provide sufficient data on
which to base specific allocations of land area to part
icular uses within a community.
A regional economic location study might indicate,
for example, that Southern California is likely to re
ceive a large proportion of future government and com
mercial aircraft construction activity, because of an
existing qualified labor supply, other economies external
to aircraft plants, and favorable climatic conditions.
From a comparative analysis of alternative locations, a
forecast of regional economic growth could be made for
Southern California's aircraft industry.
This regional analysis indicates that aircraft
construction is oriented to the Southern California region.
Southern California, depending on how it is defined,
could extend all the way from the south end of the San
Joaquin Valley to the Mexican border, and from the Pacific
Ocean to the Arizona border. Southern California could
contain an area of around a quarter of a million square
miles. In fact, aircraft activity is scattered through-
260
out this region, with notable concentrations at several
locations.
The point is this: even though regional economic
location analysis may indicate an orientation of activity
to a particular region, it usually involves no determina
tion of precisely where within the region that the loca
tions will occur. From the viewpoint of regional eco
nomic analysis, many location decisions occurring at a
sub-regional level appear to be taking place on a random
basis, i.e., they cannot be accounted for in a predict
able manner. This, by the way accounts in part for the
emphasis on probability analysis to allocate economic
activity spatially at a local level.
Differential land rent levels within a metropolitan
region, based on differences in access to market potential
(consumers), tend to allocate selected sites to various
retail uses of varying scope and intensity. However,
access differentials may be virtually nonexistent in some
urban areas with highly developed freeway and mass
transit systems. Access differentials in urban areas are
often so subtle, that shopping center developers may base
their location decisions on factors not entirely account
ed for by economic studies.
In a highly developed urban area, of course, there
are likely to be relatively few suitable sites available
on which to locate a planned regional shopping center.
Therefore, given the existing spatial distribution of
consumers in the surrounding area, there will actually
probably be only a handful of suitable sites. Since re
tail development decisions are made on a local level,
the problem becomes one of determining the size and
nature of a community's trade area, and its overall
growth potential jas segment of the larger region of
which it is a part.
For purposes of governmental land use planning
decisions, the relevant decision making unit in this
study is the municipality. As Delafons says, "Planning
and land-use control are carried out by municipalities,
incorporated units of local government."! None of the
standard metropolitan areas in the United States, as
defined by the Bureau of Census, is governed by a single,
multi-purpose political authority with the power to zone
and allocate land to different uses. Virtually every
where, the municipality is the political decision making
unit with regard to land use planning.
Regardless of theoretical suitability, economic
planning does occur on a municipal scale. For this
^-John Delafons, Land-Use Controls in the
United States, (Cambridge: Joint Center for Urban
Studies of M.I.T. and Harvard University, 1962),
p.10.
262
reason, suitable analytical procedures useful to the
local land use allocation process should be provided.
The combination of economic base, trade area, and space
requirements analyses offered in this chapter, is an
attempt to provide an analytical procedure effective in
determining present and future space needs for retail use
within a community.
If a municipality appears as a probable location
for a regional shopping center, thorough knowledge of
local (i.e., internal) economic conditions is a pre
requisite for rational planning decisions. Knowledge of
the municipality's economic relationships with the rest
of its metropolitan area are also necessary. In other
words, analytical procedures adaptable to a micro-geo
graphic study scale are called for in local retail space
allocation decisions.
The viewpoint of municipal land use planning.—
Local planning agencies have definite responsibilities
in certain areas in order to enhance the welfare of
local residents. Land use planning, or, the allocation
of land to various categories of uses, is of prime
significance in this local welfare maximizing process.
This is one of the few areas where municipal planning
agencies can exercise any influence. Income and employ
ment levels are determined on national and regional
levels. However, given the level of economic activity,
local planners can make a determination of the pattern
of land uses evolving within their communities. They
also can specify in what proportion various categories
of land use shall be represented, depending on the pre
ferences of local residents and businessmen.
Harris feel that,
...the correct allocation of land uses
in the present and the appropriate
influence of their development over
the years ahead is a most suitable
economizing role of the city planner.2
Municipal planning is concerned with the disposition of
a limited resource, land, on which will be placed a
heavy concentration of fixed capital. Municipal planning
attempts to evaluate and guide the most economic uses of
this land, both for the present and future and in both
monetary and welfare terms. This land allocation,
Harris notes,
...implies a detailed knowledge of land
requirements for various activities, in
addition to the prediction of activity
levels...^
Municipal land use planning is a process involving
continuous study of the local scene as it is affected by
^Britton Harris, "Projecting Industrial Growth
of Metropolitan Regions," Regional Science Association
Papers, (1956), 240.
3"lbid.
population growth and migrations, scientific and technical
developments, changes in national and regional income
levels and distributions, changes in local preferences
and values, shifts in industrial and commercial location,
and so forth. Many of these factors are difficult to
assess by themselves, let alone their effect locally.
Nonetheless, their local impact must be assessed in some
satisfactory manner if rational land use plans are to be
made.
Municipal government planning and allocation of
land to retail use.— There is often a question as
to what extent municipal government authorities should,
or can, determine the locations and amounts of land in
a community for various uses. In the case of retail
trade, local authorities may have the power to determine
the amount of retail activity that can locate in a
community, to serve both local and regional needs. If
consumer needs are to be adequately met, local planning
authorities must have sufficient information to make
economically sound retail space allocation decisions.
Since local government does not engage in retail
development and operation itself, local planning for
retail facilities involves some sort of coordination and
cooperation between municipal planners, who are concerned
265
with protecting and improving community interests, and
private retail developers and operators, who are con
cerned with the success of their individual business
ventures. Some writers feel that urban planning should
provide a framework within which private retail develop
ers could more rationally pursue their goals. As
Woodbury says, local planning agencies can provide others
with "...reliable information; analyses and estimates of
the future of the community and its economy; coordination;
professional collaboration and encouragement."4
Woodbury's suggested role for municipal planning
agencies appears to be a rather passive one. There are
many who feel that land use must be dictated by the
prevailing market situation on the basis of supply and
demand, to allow for operation of "natural" (bid and ask)
forces that set prices for individual land sites. The
best use of land, then, is realized under conditions
which promote the highest investment returns. From this
viewpoint, Wilhelm notes, "Zoning classifications are
expected to reflect economically relevant cognitive data
because such data inevitably determine the pattern of
^Coleman Woodbury, "Land Economics Research for
Urban and Regional Planning," Land Economics Research,
(New York: Praeger, 1964), ed. Joseph Ackerman,
Marion Clawson and Marshall Harris, p.73.
266
land-use development."5 A zoning classification that
failed to allow utilization in accordance with highest
investment returns possible under interpreted economic
conditions would be undesirable.
So far as controlling private development is con
cerned, the job of municipal planning is largely that
of anticipating the trend of private development, making
adequate provision for all acceptable uses, and setting
standards for new development which reflect local wishes.
Wilhelm expresses the nature of the conflict between a
passive and a more positive role for local government
in the land use planning process, when he says that,
...the economic orientation does not
perceive zoning as a predictive instrument;
rather, zoning, according to this perspective,
merely mirrors those economic factors that
determine land use in the time yet to come.
On the other hand, because the protectionists
maintain as part of their value orientation
that zoning must determine— not reflect—
land valuations, they endeavor to make
zoning itself a predictive device to foretell
the course of land use over space.6
In fact, of course, a municipal master plan of
land use designates certain economic conditions: place
ment of streets and other transportation facilities;
location and sequencing of public capital improvement
5Sidney M. Wilhelm, Urban Zoning and Land-Use
Theory, (Glencoe: The Free Press of Glencoe, 1962), p.95.
6Ibid., 158.
projects; and designation of certain land uses in certain
areas of the city. A master plan requires these items
by imposition and land use must conform to these deter
minants. To the extent that the municipal zoning process
insures that land use actually adheres to such planning,
a master plan becomes a predictive device. Wilhelm
points out "...that zoning as an effort to exert control
over the distribution of land use is an integral part
of urban master planning."7 He also emphasizes, however,
that "...both the notion that zoning must reflect eco
nomic conditions rather than control them, and the belief
in the utilitarian doctrine of private ownership, inter
vene to prevent reliable prediction."® In other words,
a great deal of "flexibility" is often called for in
municipal master plans of land use.
The problem of controlling local allocations of
land to various uses boils down to Delafons' observation
that, "Although every municipality has the power to
control land use and private development, the decision
may be to do without it. There is no obligation to
exercise control..."® While it is not mandatory that
municipal planning authorities exert control over retail
7Ibid., 2.
8Ibid., 160.
^Delafons, Land-Use..., 12.
development and land use; they should be interested in
■whether or not the natural pattern of emerging retail
structure in their community is the most economic and
satisfactory solution for all residents. Or, is there
a better alternative? The municipal land use planning
process really involves the decision of how to balance
reality with the need for land use control and planning.
If municipal planning authorities abdictate their
decision-making role in the retail development process,
social costs may arise from uncontrolled planned and
other shopping center development. Uncontrolled retail
development can lead to destructive competition for
existing facilities and, perhaps, to inadequate provision
of convenience goods for local residents, and an excess
ive amount of shopping goods for consumers at the regional
level.
In part, the reticence of municipal planning
authorities to engage in firm retail space allocation
decisions has been because these authorities lack suf
ficient objective information on which to base such
decisions. The municipal planning process requires its
own independent sources of comprehensive, complete,
accurate, and up-to-date information; collected, process
ed, and analyzed on a continuing basis. These data should
relate to: the overall economic position and growth
potential of a community; the specific market potentials
for existing and proposed retail facilities located in
a community; and the nature of the market structure and
technical operating characteristics of local retail
facilities.
The emphasis on up-to-date information should be
strong. Trade areas may shift radically over time and
technological changes occur continually. This is why a
local planning agency must have a continuing program of
data collection to fulfill even minimum analytical needs.
As Simmons points out, "Planning for the future can be
done only with the most up-to-date data and with the
realization that all facilities will be outdated within
a few years."I®
The municipal planning process needs specific
information at the individual shopping center or retail
store location level. There must be criteria to evaluate
probable investment decisions by individual retail develop
ers. Nonetheless, the evaluation problem of the proposed
model, offered in this chapter, is broader in scope than
that of an individual retail developer interested in a
specific location. It involves the entire existing and
probable future retail pattern in a municipal planning
■^Ojames Simmons, The Changing Pattern of Retail
Location, (Chicago: University of Chicago Press, 1964),
p. 169.
270
area (including a municipality's retail trade area).
The problem is; are the spatial relationships of all
retail facilities in a community satisfactory in terms
of sufficiently serving consumer needs, and efficient
allocation of terrestrial space to retail use?
Certain recent developments in retail technology
may even facilitate retail space allocation decisions in
the municipal planning process. The dominance of planned
regional shopping centers offers local planners new
possibilities. Simmons feels that, "Although less flex
ible, once built, the unified planned center, designed
for a specific retailing role makes possible a tighter
control over the retail landscape. Even though the
present combined analysis concentrates on planned region
al shopping centers, the approach implies that local
retail space allocations can be made for any level in
the retail hierarchy. Smaller neighborhood and community
centers should provide complementary clusters of retail
facilities with less extensive trading functions, and
their sizes and locations would be determined in a manner
similar to that for regional centers.
A Model of Retail Space Allocation
The planning problem.— A hypothetical planning
1]-Ibid.
situation will be set up in this section to illustrate
the analytical approach to community economic analysis,
offered by this study. The planning body involved in
our hypothetical example is a municipality. The Specific
planning problem encountered in the example consists of
determining the need for, and the present and future
space requirements of, a regional planned shopping center.
It is assumed that the subject municipality is so situa
ted within its metropolitan area that it provides a
favorable potential location for a regional shopping
center.
Much of the data included in this hypothetical
planning problem are drawn from other empirical studies
undertaken by the writer and others. The economic and
social data describing the subject municipality, for
example, are based on information derived from an eco
nomic base and income multiplier study of Redondo Beach.
This is done in order that the economic magnitudes and
propensities, and shopping behavior patterns of residents
will correspond to actual conditions in typical com
munities to which the analytical approach is applicable.
12
See Philip H. Friedly, "An Economic Base and
Income Multiplier Study of Redondo Beach, California,"
Unpublished Masters Thesis, University of
Southern California, June, 1964.
272 ■
The economic variables and functional relationships
of the market potential analysis are purely hypothetical,
i.e., not drawn directly from any particular trade area
study, but are, nonetheless, based on reasonable assump
tions as indicated by marketing studies. The retail
space productivity data are, of course, derived directly
from the exhaustive empirical studies of Malcolm McNair
and the National Retail Merchants Association.
Redondo Beach is not an isolated economic entity,
but, rather, is embedded in a metropolitan matrix. In
this situation it is difficult to visualize our munici
pality as any kind of functioning economic unit. Since
Redondo Beach is a residential community with only retail
trade as a significant industrial activity, it would
appear that export of labor services to the rest of the
metropolitan area is the major economic function.
If Redondo Beach is not to be viewed simply as an
insignificant segment of the metropolitan labor supply
area, then it must be viewed as a retail trade center.
This latter viewpoint of Redondo Beach as an economic
entity, encompasses considerably more area than is
contained by its municipal boundries. For, as Kelley
points out,
Marketing managers have long recognized
that markets do not follow strictly
political lines. For many marketing
purposes the concept of the metropolitan
area has more relevance than arbitrary
political divisions. In locating marketing
agencies it is important to consider as a
market unit the entire population and area
around the site, rather than merely the
political subdivision in which the tract
is situated.13
Structurally, a controlled regional shopping center is
located in relation to its trade area so as to attract
customers from the immediately surrounding areas (e.g.,
from within the municipality in which it is located),
and from the entire region or quadrant of the metropolitan
area in which it is located.
The geographic areas of study for retail trade
centers must, as Berry says, be large enough to be units
"...within which demands and supplies are equilibrated...
Such units are retail business centers and the market
areas sei*ved by these centers. "14 it is the municipality
and that segment of the metropolitan area to which it
exports retail goods and services that serves as the
study area.
The macro-demand analysis.— In terms of the proposed
l^Eugene J. Kelley, Shopping Centers— Locating
Controlled Regional Centers. (Washington, D.C.: Eno
Foundation for Highway Traffic Control, 1956), p. 41.
14Brian J. L. Berry, "The Retail Component of the
Urban Model," Journal of the American Institute of
Planners, (May, 1965), 151.
analysis, it is the job of the economic base study to
determine and forecast total income levels in the plann
ing area. In the case of Redondo Beach, this involves
analyzing and forecasting income potential for a city of
about 50,000 population, which is situated within a
metropolitan area of about six million people.
Determining present and future income levels in
a place like Redondo Beach involves ascertaining present,
and forecasting future, levels of population in the city
and in its retail trading area. The most important
source of new income for Redondo Beach is in the form
of new residents, who arrive there in response to ex
ogenous regional growth factors. Because most Redondo
Beach residents earn their income from elsewhere in the
metropolitan area, their export of labor services pro
vides the city's main source of basic income. The other
important source of basic income in Redondo Beach, the
export of retail goods and services, is derived directly
from present and future income potentials of its retail
trading area.
Population forecasts for both Redondo Beach and
its retail trade area will be based on the same exogenous
regional economic factors. To estimate future population
growth potential for a Montgomery Ward's store in the
Port Worth-Dallas area, for example, use was made of
estimates of future basic manufacturing activity in that
area which, it was assumed, was the prime mover of
population growth locally. Fort Worth is a trading and
manufacturing center with a relatively high proportion
of the labor force in manufacturing. Since World War II,
...manufacturing has been dominated by
the aircraft industry, which accounts
for almost one-half of total manufacturing
employment. The recent high rate of
population increase is directly related
to the growth of manufacturing and the
same will be true of future population
growth.15
The Fort Worth area was divided into six geographic
areas and population growth charted for each. An attempt
was made to make intra-urban assignments to each of
these sub-areas from the overall Fort Worth metropolitan
area population forecast. Mainly, the assignments
amounted to assuming approximately equal growth rates
for all the areas except the CBD, which was expected to
lag behind. Two of the geographic areas closer to
Dallas were expected to lead slightly.
This population forecasting method makes an overall
population forecast of the metropolitan area and then
distributes it to small sub-areas within the metropolitan
area. If these zonal population projections are made
-*-^Wm. Applebaum, Store Location and Development
Studies. (Worcester: Clark University, 1961),
p. 24.
from mathematical models, the following assumptions are
usually made: (1) that existing trends in residential
site selection will continue; and (2) that existing
municipal government policies relating to land use
and zoning, capital projects, and so forth, will continue
as in the past. So, the distribution of consumers
assumed as inputs to the market potential forecast will
reflect the probable effect of existing policies. It
is very difficult to conduct such a forecast in any
other manner.
In community economic forecasts, Bassie notes
"...that methods must be adapted to the requirements of
the specific local situation. Beyond that, generaliza
tion ceases, and the ingenuity of the local forecaster
takes up the burden."16 He further points out that,
"The various aspects of the community, its surrounding
area, and its inhabitants which afford opportunities for
efficient industrial operations are its economic re-
s o u r c e s . " 1 7 The economic potential of each local area
must be analyzed in terms of its own objective situation
relative to the regional growth occurring around it.
Voorhees says that small area projections of
■ I f i
- * - 0Lewis Bassie, Economic Forecasting.
(New York: McGraw-Hill Book Company, Inc., 1958),
p. 558.
17Ibid.. 163.
277
future population growth are
...done by distributing total metropolitan
growth to the individual zones on the basis
of vacant developable land; i.e., if Zone A
contained 10 per cent of the vacant land in
the metropolitan area, 10 per cent of the
expected growth was assigned to Zone A.18
Objective factors such as buildable vacant land sites
and population density must be noted for a city and its
retail trade area, if realistic assessments of growth
are to be made. Normally, in a large metropolitan
center, a systematic decline in net residential density
is observable as distance from the CBD increases.
Assumptions can be made, for example, that the rate of
future population growth (i.e., density increases) will
be slower in zones with larger proportions of their land
area already developed, than in more sparsely settled
and developed zones.
Probably over ninety per cent of the buildable
land area in Redondo Beach has been developed. The
population density per square mile is relatively high,
as can be seen from the data in Table 16.
The population density per square mile is quite
high relative to the rest of the Los Angeles area, in
both Redondo Beach and most of its neighboring communities
■^Alan M. Voorhees (ed.) "Land Use and Traffic
Models: A Progress Report," Journal of the American
Institute of Planners. (Special Issue, May, 1959), 75.
TABLE 16
IN REDONDO
POPULATION DENSITIES
BEACH AND SELECTED ADJACENT AREAS
Areas
Square
Miles of
Area 1950a 1960a
Per Cent
Increase
1950-1960
Redondo Beach 6.0 4,204 7,371 86.3
Los Angeles County 4,060.0 1,023 1,487 45.3
South Bay Areab n.a. n.a. n.a. n.a.
Torrance 20.0 1,112 5, 050 354.1
Manhattan Beach 3.9 4,444 8,701 95.8
Source: U. S. Census of Population, 1950 and 1960.
aDensity figures are in thousands of persons
per square mile.
Whe South Bay Area is that segment of the
Los Angeles metropolitan area that surrounds Redondo
Beach, and constitutes the bulk of the export market
for the latter's regional shopping center which opened
in 1957. The South Bay Area consists of the cities of
Redondo Beach, Manhattan Beach, Hermosa Beach, El Segundo,
Torrance, Hawthorne, Gardena, Lawndale, Palos Verdes
Estates, Palos Verdes, Rolling Hills, and Rolling Hills Estates,
n.a. = data not available for the entire area.
278
279
in the South Bay Area. Relatively high rates of popula
tion growth are indicated for the South Bay communities
over the past decade, too. These high density figures
and tremendous growth rate figures in Table 16, coupled
with the lack of vacant buildable land, give indication
that a reduction in future population growth rates are
likely for Redondo Beach and parts of its retail trade
area.
Plotting straight line regression curves extra
polating the growth trends of the past ten years for
Redondo Beach and its retail trade area would not be
reasonable. Straight line regression curves imply, as
Isard points out, that growth
...will continue indefinitely into the
future, and that population will approach
an infinite size. Although it is con
ceivable that growth may continue for an
indefinite time into the future, it is
inconceivable that population size could
reach infinity, especially within a
politically subordinate area of limited
physical size.19
Linear extrapolation of historical population data
implies that the character of the most recent period of
development is expected to continue into the future.
This does not appear to be the case with Redondo Beach
and its surrounding trade area. The type of growth
l^Walter Isard, Methods of Regional Analysis.
(Cambridge: M.I.T. Press, 1960), p. 10.
280
curve expressing probable future population increase
for this area might be one showing a declining percent
age rate of growth, perhaps even declining absolute
increments of population growth in the not too distant
future.
To ascertain the approximate amount of population
growth pressure likely to be exerted on Redondo Beach
and its trade area in the future, the economic potential
of the area should be analyzed. Table 17 presents the
distribution of the Redondo Beach labor force among
occupations and industrial categories in 1962. It can
be seen that almost sixty per cent of the Redondo Beach
labor force are occupied as craftsmen, foremen, labor
ers, or service workers. Also, nearly fifty per cent
of the Redondo Beach labor force work in the construc
tion and manufacturing industries. Detailed study of
the growth potential of these two industrial groupings
should be undertaken to determine basic income growth
possibilities. Other relatively important occupational
and industrial categories should be studied in detail
too. The potentials for economic expansion in these
basic income sources in the Los Angeles metropolitan
area, and especially in the South Bay Area, should be
explored.
National and regional economic forecasts should
281
TABLE 17
OCCUPATIONAL AND INDUSTRIAL DISTRIBUTION
OF THE REDONDO BEACH LABOR FORCE IN 1962
Occupation
Per Cent
Employed Industry
Per Cent
Employed
Professional,
technical 18.8 Construction 15.4
Manager, Manufacturing 34.4
proprietor,
official 14.4 Retail, wholesale 17.4
Clerical 3.7 Government,
education 20.3
Sales 4.1
Business,
Craftsman, professions 10.6
foreman 36.2
Real estate,
Laborer, finance,
service 22.8 insurance 1.9
Source: Planning Research Corporation
Field Survey, January 1963.
be obtained for the selected industries. Intra-metro
politan location trends of these industries should also
be studied. Assuming that the occupational and indus
trial distribution of the trade area's labor force is
similar to that for Redondo Beach, the same line of
analysis should be followed to determine income growth
potential in the trade area.
In 1960, the U. S. Census indicated a mean family
income figure in Redondo Beach of $7,518. In 1963, field
survey data indicated a total income accruing to Redondo
282
Beach families from all sources of $121 million. This
averaged out to about $7,318 per family for the 15,666
families residing in Redondo Beach in 1963. About 81
per cent, or $98 million, of this total family income
was basic. The other $23 million was earned locally
by residents, mainly through employment in Redondo
Beach retail establishments.
The other major source of basic income for
Redondo Beach, not accounted for in the family survey
data, are retail exports. Table 18 presents a balance
of payments index for retail trade in Redondo Beach, on
a per capita basis.
It is obvious from Table 18, that retail trade
provides the major source of basic income to Redondo
Beach, aside from the sale of labor services by resi
dents. In 1957 a regional planned shopping center
opened for business in Redondo Beach. However, the
major department store branch did not open until early
1959. The dramatic increase in retail exporting activ
ity between 1958 and 1959 is undoubtedly because of
the initiation of the major department store's operation
in the regional center.
Comparative population growth rates for Redondo
Beach and its surrounding trading area are offered in
Table 19.
TABLE 18
PER CAPITA BALANCE OF PAYMENTS INDEX
FOR THE REDONDO BEACH ECONOMY
SECTOR 1958 1959 1960 1961
Retail Trade 117 153 148 165
Services 108 96 93 100
Manufacturing, wholesale 28 25 25 29
All sectors 88 100 107 119
Source: Trade Outlets and Taxable Retail
Sales in California.
Note: An index value of 100 indicates a
zero balance, a value of less than
100 indicates a negative trade balance, i.e.,
imports, and a value greater than 100 indicates
a positive trade balance, i.e., exports.
283
TABLE 19
POPULATION GROWTH IN REDONDO BEACH
AND
ITS SURROUNDING RETAIL TRADE AREA
IN THE LOS ANGELES REGION FROM 1940 to 1962
Area 1940
Population3
1950 1960 ' 1962
Per Cent
1940-
1950
Increase
1950-
1960
Redondo Beach 13 25 47 50 92.7 86.3
Los Angeles County 2, 786 4,152 6,039 6,395 49.0 45.5
South Bay Area 73 188 319 343 157.0 70.0
Torrance 10 22 101 113 123.5 354.1
Manhattan Beach 6 17 34 35 170.9 95.8
Hermosa Beach 7 12 16 17 64.3 36.3
Sources: City and County Data Book, U. S. Government
Printing Office, Washington 25, D.C., 1962? U. S. Census of Population,
1950 and 1960; California Statistical Abstract, Economic Development
Agency, Sacramento, 1962; and Population and Dwelling Units, Los Angeles
County Regional Planning Commission, Population Research Section, Quarterly.
aAll population figures are to the nearest thousand.
284
In the decade 1950-1960, Redondo Beach population
growth rate was lower than for the 1940-1950 decade.
The great increase in residential density per square
mile during this period, from 4,204 to 7,371, indicates
that Redondo Beach is filling up and may he reaching its
population holding capacity, given its present low
density zoning practices. The Los Angeles Regional
Planning Commission forecasts a population of only 60,000
for Redondo Beach by 1980. This represents a population
increase of about twenty per cent over two decades, when
the previous decade alone had experienced an 86 per cent
increase.
Most of the undeveloped sections of the South Bay
Area were being rapidly filled in during 1950 to 1960.
A similar drop in this area's growth rate is also notable
between the 1940-1950 and 1950-1960 periods, as seen in
Table 19. This declining rate of expected population
increase is reflected in the Los Angeles Regional Plann
ing Commission's 1980 forecast of about 444,000 people
in the South Bay Area. This represents only about a
thirty per cent increase over two decades, as compared
with a seventy per cent rise in population from 1950 to
1960. The prognosis for both Redondo Beach and its
retail trade area is slower future growth.
In 1962 disposable personal income accruing to
286
family or household units in Redondo Beach was about
$114 million dollars. This figure averaged out to around
$7,260 per family. Since the per family income figure
for Redondo Beach is nearly identical with the same
figure for the South Bay Area, the Redondo Beach data
will be generalized to apply to the retail trading area.
In recent years, disposable personal income per family
has been rising at a rate of about five per cent each
year in Redondo Beach and its retail trading area.
If, after careful analysis, it is estimated that,
between 1960 and 1980, 3,000 new families earning their
incomes from elsewhere in the metropolitan area will
arrive in Redondo Beach, and that nearly 39,000 new
families will arrive in the retail trade area located
outside of Redondo Beach; what are the future income
and market potential implications for Redondo Beach and
its retail facilities? The implications of the trade
area population forecast for Redondo Beach retail exports
will not be clear until the micro-demand analysis has
been completed.
However, the total income implications for Redondo
Beach from the 3,000 families who will be earning basic
income, can be ascertained by use of a community income
multiplier model. First, though, let us view the
complete set of forecast data for Redondo Beach and its
287
retail trade area (i.e., the South Bay Area) in Table 20.
TABLE 20
POPULATION AND INCOME FORECASTS FOR
REDONDO BEACH AND SOUTH BAY AREA TO 1980
i
Population
and Redondo
" "
Beach South Bay Area
Income 1960 1980 1960 1980
Population 47,000 60,000 319,000 444,000
Total families 15,000 18,750 100,000 138,750
Income per
family ($)a 7, 260 14,500 7,260 14,500
Basic employment
families^ 12,000 15,000
— ----
Basic income
(millions $) 87 218
— — ----
Total income
(millions $) 107c 268c 726 2, 012
Sources: Planning Research Corporation Field
Survey, January, 1963; and Sales Management Magazine,
Survey of Buying Power, 1958-1963.
aThis is disposable personal income available for
savings and consumption expenditures, and is somewhat
less than the family income accruing from all sources
of $7,318, mentioned earlier and derived from field
survey.
1.
DThe field survey results indicated that about
eighty per cent of resident families exported their
labor services.
p
The total income figures for Redondo Beach
were calculated by multiplying basic income by
the local income multiplier value of 1 . 23,
which is derived from the multiplier formula
below in the text.
288
The population projections to 1980 in Table 20
indicate that, if exogenous regional and national growth
factors providing jobs for Redondo Beach and South Bay
residents continue along past trends, about 3,000 new
families should move into Redondo Beach between 1960
and 1980, and 38,750 families should move into the South
Bay Area retail trade area surrounding Redondo Beach.
In response to the needs of the 3,000 families whose
wage-earners are employed outside Redondo Beach, 750
other families will also enter the community during
1960 to 1980, to accept local employment within the city.
In 1960, basic income from export of labor services
in Redondo Beach amounted to $87 million. By concentra
ting on forecasting the exogenous growth factors pro
viding jobs in the metropolitan area, total income fore
casts can be generated from a community income multiplier
formula. The following formulation was used to calculate
a basic income multiplier value for Redondo Beach:
dY = dYx . ■ ■ = = kdYx
x 1 - (ch) 1 - b x
" 1 1 - (.36 x .52) " 1 - .19 =.81 = 1*23
where c is the propensity to consume locally; h is the
propensity of a dollar of local retail sales to create
local income; b = (ch); dY is a change in total income;
dYx is a change in export income,* and k is the multi
plier. The multiplier value for Redondo Beach is 1.23.
In 1960, the total Redondo Beach income figure
of $107 million is derived by multiplying the $87 million
of basic income by the multiplier value of 1.23. To
derive a total income forecast for 1980, a forecast
has to be made of basic income. When basic income of
$218 million is multiplied by 1.23, it yields the fore
cast figure of $268 million of total income for Redondo
Beach in 1980.
The population and income figures for the retail
trade area are overall growth projections, although
they are based on any important exogenous regional or
national growth factors affecting economic potential.
The trade area forecasts provide estimates of total
income potential for local retail export expenditures.
All retail establishments in Redondo Beach in
1960 generated total sales of around $108 million.
The total propensity to consume out of current income
of Redondo Beach residents, as indicated by field survey,
is .86. This means total consumption expenditures of
$92 million. Redondo Beach residents have a propensity
to import of .5, so, $54 million was spent outside of
the city. The remainder, $38 million, was spent locally
by residents. If 68 per cent of these consumption
290
expenditures were allocated to retail purchase, as
indicated by market studies, then local residents spent
about $26 million in retail establishments in Redondo
Beach in 1960.
Since Redondo Beach retail stores and centers
generated over $108 million in sales during 1960, this
means that $82 million in retail sales were made to
non-residents. Over three-fourths of Redondo Beach
retail output or sales is basic in nature, i.e., export
ed to other parts of the metropolitan area.
Information pertaining to the demand source of
retail exports awaits the market potential analysis of
the retail trading area surrounding Redondo Beach. The
micro-demand analytical segment of the model is present
ed in the following section.
The micro-demand analysis.— While an economic
base study is important to considerations of potential
growth, it is not sufficient to accurately allocate
space to retail use at the municipal level because, as
Applebaum says,
Even completely accurate assessments of the
economic prospects of an area cannot always
indicate the desirability of such areas for
store sites. Instead, they must be weighed
against shopping habits and the quality of
available stores.
2®Applebaum, Store Location.... 78.
291
That is, a micro-analysis of the local market area
must be undertaken.
There is a question in market potential analysis
of whether demand is being estimated for a preconceived
grouping of retail outlets, or, whether a demand area
is being objectively measured and retail outlets fitted
to its needs. Simmons says that a market potential
analysis could start
...with a given set of business centers
and derive trade area populations and
income for each of them, allowing overlap
of trade areas for centers at different
levels in the hierarchy, and determine
population per retail unit rather than
retail space per population unit...A
method of constructing a retail structure
which investigates the requirements of a
given conformation and fits it into the
trade area seems as viable as starting
with the trade area and determining the
varieties of business groupings required
to serve it. In either case the problem
is to show how the requirements change
over time.21
Simmons is correct that it matters little which approach
is followed? as long as the analysis is conducted
objectively. The emphasis in this study is on deter
mining the nature and extent of market potential, and
then relating either existing or proposed retail facili
ties to consumer needs. It is assumed that the general
extent, density, and character of demand in a market
2-^-Simmons, The Changing Pattern. . . . 160, 161.
292
area will suggest the shopping center types best suited
to serve that trade area configuration, given the state
of retail technology.
The structure of the retail sector in Redondo
Beach is fairly simple. There are three noticeable
clusters and one "strip" section of retail activity in
the city. There is an unplanned shopping district of
neighborhood size at the extreme south end of the city.
There is a shopping district in the center of town, i.e.,
"downtown" Redondo Beach. And there is a "strip" or
"string" commercial district ranging for two or three
miles along a major thoroughfare in the north-central
part of the city.
Prior to 1957 these were the only significant
retail areas that Redondo Beach contained. On balance,
they provided virtually no retail export income. They
were, in other words, community and neighborhood retail
centers serving a local consumer market located largely
within the municipal boundries of Redondo Beach. In
1957, however, the retail structure of Redondo Beach
was dramatically altered by the opening of a new planned
regional shopping center just inside its eastern-most
boundry: the South Bay Center.
With the opening of a new regional center,
Redondo Beach became an exporter of retail goods and
293
services. More non-residents came into Redondo Beach
from the surrounding South Bay Area of Los Angeles to
shop, and the rate of local retail purchase by residents
probably made a discrete upward movement.
The micro-demand analysis will attempt to ascertain
present and future market potential for a regional center
in Redondo Beach. Data used in this portion of the
combined analysis are hypothetical, because field survey
results from the Redondo Beach study^2 did not yield
information pertaining to a trade area analysis of this
type. Numerical values used are, however, reasonable
figures based on empirical marketing evidence and
theoretical potential models. In any event, the emphasis
here is on methodology rather than empirical verification.
The surrounding South Bay Area of the Los Angeles
metropolitan region comprises the primary potential
trading area for a regional shopping center located in
Redondo Beach. In Table 21, the various communities
making up segments of this trade area, and their
distances from Redondo Beach, are presented.
Distances and driving times are measured from
population centers to South Bay Center. Driving time in
minutes is taken to be the most crucial factor in deter
mining the probability of consumer attendance at the
22See Friedly, Masters Thesis.
294
TABLE 21
PROXIMITY OF TRADE AREA SEGMENTS TO
REDONDO BEACH REGIONAL SHOPPING CENTER
Distance
Shopping
from South
Center
Bay
Population Center
of
Trade Area Seqment
Road
Miles
Air
Miles
Driving
Time
(minutes)
Segment 1
(El Segundo) 7 5 21
Segment 2
(Hawthorne) 6 6 18
Segment 3
(Gardena) 6 6 18
Segment 4
(Torrance) 4 4 12
Segment 5
(Rolling Hills) 6 5 18
Segment 6
(Palos Verdes Estates) 5 4 15
Segment 7
(Hermosa Beach) 3 2 9
Segment 8
(Manhattan Beach) 4 3 12
Segment 9
(Lawndale) 5 5 15
Source: Southwest Portion of Los Angeles
County. Security First National Bank,
Research Department, April, 1963.
295
center, other factors (e.g., competition) being equal.
Looking at Table 21, it would appear likely that the
probability of consumer attendance at South Bay Center
would be considerably lower in Segment 1 of the trade
area, than in Segment 7.
The nature of the South Bay Center is revealed
by the data in Table 22. The postulated overall size
of the center and the size of the major department store
clearly indicate that it is a regional shopping center.
Marketing studies have implied that the major department
store in a regional center should contain at least
125,000 square feet of enclosed selling space. The
South Bay Center department store has been assumed to
have 225,000 square feet of selling space. The assumed
annual sales levels of both the center and its depart
ment store indicate that more than adequate access to
the trade area market potential has been achieved, in
terms of sales space productivity.
Table 23 gives an idea of the geographic dis
tribution of population, family units, and disposable
personal income accruing to families, in South Bay
Center's retail trade area. Segment 4 offers by far
the greatest gross income potential to the center of
any portion of the trade area. In terms of net income
potential, Segment 4 remains most significant. The
295a-
TABLE 22
CHARACTERISTICS OF
REDONDO BEACH REGIONAL SHOPPING CENTER
Characteristics
South Bay Regional
Shopping Center
Gross size of center (sq. ft.) 700,000
Net sales space of center (sq. ft.) 500,000
Gross size of major department
store (sq. ft.) 300,000
Net sales space of major
department store (sq. ft.) 225,000
Number of tenants 40
Acreage 55
Number of parking spaces 5,800
Annual sales of center
(millions $) 66.5
Annual sales of major department
store (millions $) 28.3
Annual sales per sq. ft. of
center selling space ($) 132
Annual sales per sq. ft. of
department store selling
space ($) 126
Source: Planning Research Corporation
Field Survey, January, 1963.
TABLE 23
REDONDO BEACH REGIONAL SHOPPING CENTER MARKET POTENTIAL BY TRADE AREA SEGMENTS
Trade Area
Segment
Total
Population
Total
Families
Gross DPI
(millions
$>
Proportion
shopping at
South Bay
Center9
(probabil
ity fact
or)
Net income
potential
for South
Bay Center
(millions
$)
1 14,000 4,400 32 .5 16.0
2 33,000 10,300 75 .6 45.0
3 36,000 11,300 82 .6 49.2
4 101,000 31,600 229 .5 114.5
5 6,000 1,900 14 .4 5.6
6 10,000 3,100 23 .5 11.5
7 16,000 5,000 36 .9 32.4
8 34,000 10,600 77 .8 61.6
9 22,000 7,000 51 .7 35.7
Total^
272,000 85,200 619 - 371.5
Sources: Population and Dwelling Units. Los Angeles County-
Regional Planning Commission, Population Research Section, Quarterly.
aThese hypothetical probability factors are based on reasonable
assumptions about driving times and existing retail competition.
^This total excludes Redondo Beach since the export trade
area is being considered separately.
296
297
uneveness in family and income distribution among the
various segments of the trade area points up the
necessity of careful geographic stratification in
sampling for a micro-demand analysis. The sample, in
other words, must be truly representative in terms of
the actual geographic distribution of potential customers
and income. Segment 1, for example, has few families
relative to Segments 2, 3, and 4. The reason for this
is that Segment 1 (the city of El Segundo) consists
largely of an oil refinery and tank farm, rather than
residential areas. A disproportionately large sampling
of this trade area segment would distort the estimate of
total income potential.
The probability factors in Table 23 indicate
the proportion of families in each trade area segment
that make purchases at the South Bay Center on some
regular basis. Since income is averaged for an entire
segment, the same proportionate share of gross segment
income is allocated to the center as for net income
potential.
The probability factors are derived on the basis
of field survey information obtained from consuming
units in the various trade area segments. The geographic
shopping patterns of trade area residents are ascertained,
and the attractive "pull" of South Bay Center is establi
298
shed. Exponential values of distance exponents for
various types of shopping trips can be calculated from
comparison of actual with expected trip f r e q u e n c i e s . ^ 3
With information about shopping center size and the
distance factor, market potential models of the probabil
ity type can be utilized to estimate future income
potential for a center, assuming no drastic change in
the parameters of spatial inter-action.
The probability factors applied to gross income
potential in Table 23 were estimated on the basis of
driving time from the population center of each segment
to South Bay Center, and existing retail competition in
unplanned shopping districts and planned shopping centers
in the surrounding communities. For example, the
probability factor of Segment 4 would have been .8, if
distance alone were being accounted for (the same factor
as Segment 8 because driving times are equal), however,
it was arbitrarily reduced to .5 to account for com
petition from another regional shopping center located
in Segment 4 of the trade area.
In Table 24, expected annual export sales of
South Bay Center to its trade area outside of Redondo
Beach are derived from consumer spending propensities.
^See p. 194 of Chapter IV for the procedure
to calculate the distance exponent.
TABLE 24
EXPECTED ANNUAL EXPORT SALES POTENTIAL
FOR THE REDONDO BEACH REGIONAL SHOPPING CENTER IN 1960
Trade
Area
Segment
Net trade area
income potential
of South Bay Center
(millions $)
Proportion of income spent at South Bay Center
At major
department store
At remaining
stores
(factor)a (millions $) (factor)3 (millions $)
1 16.0 .05 0.8 .09 1.4
2 45.0 .07 3.1 .11 5.0
3 49.2 .06 3.0 .10 4.9
4 114.5 .03 3.4 .07 8.0
5 5.6 .04 0.2 .08 0.4
6 11.5
.06 0.7 .10 1.2
7 32.4 .07 2.3 .11 3.6
8 61.6 .07 4.3 .11 6.8
9 35.7 .07 2.5 .11 3.9
Total 371.5 - 20.3
- 35.2
Total export sales potential 55.2
Source: Sales Management Magazine, Survey of Buying Power, 1963;
and U. S. Census of Business, Retail Trade in California, 1963.
aThese factors are calculated on the basis of marketing data
for the Los Angeles area, indicating consumer expenditure allocations
among retail purchases in department and other stores.
299
300
These spending propensities are expressed as
factors, by which net potential trade are income is
multiplied to derive expenditures, and thereby sales,
at the center.
The expenditure propensities in Table 24 are
calculated on the basis of consumer family budget data.
Consuming (usually family) units allocate their disposable
personal income among retail purchases in fairly constant
proportions over reasonable periods of time. Expenditure
allocations are often conducted on the basis of type
of product purchase, i.e., clothing purchases, furniture
purchases, etc., each account for a certain proportion
of budgeted family retail expenditures. In this manner,
various types of shopping trips can be accounted for,
and distance exponents can be calculated for each, for
use in market potential model predictions.
In the present study, propensities are calculated
on the basis of expenditure allocations among store
types, rather than among product classes per se.
Empirical evidence indicates that, in communities in
the Los Angeles metropolitan region, about seven per
cent of family disposable income is allocated to retail
purchases in department stores, while another sixty per
cent goes for retail purchases in other types of stores.
The proportions of these expenditure allocations
being made at South Bay Center1s department and other
stores, were estimated on the basis of prevailing retail
competition. The probability of consumers visiting the
center on a fairly regular basis has been calculated.
Now it must be ascertained how much of this net potential
income will result in actual expenditures or sales at
South Bay Center during a given time period.
If consumers are neither coming to the center
from a trade area segment in which there is major shopp
ing center or department store competition, nor passing
near or through such competition, it is assumed that
they make virtually all of their department store pur
chases at South Bay Center. Also, it is assumed that
they make a relatively high percentage of their other
retail store purchases there. The more intense is
retail competition, the smaller is the percentage of
family disposable income assumed to be spent at South
Bay Center.
The remaining element of market potential to
estimate for South Bay Center is that emanating from
within Redondo Beach. As noted previously, Redondo Beach
residents spend about $26 million annually in local re
tail establishments. Field survey data from the Redondo
Beach economic base study indicate that over eighty per
cent of the local residents visit South Bay Center one
302
or more times each month. A conservative estimate could
be made that, perhaps, three-fourths of the residents
shop there on a regular basis, and that they make
virtually all of their local department store purchases
there.
Let us assume that those residents shopping
regularly at South Bay Center make fifty per cent of
their retail purchases there. This assumption yields a
total of $9.8 million spent annually at South Bay Center
by residents, of which $6.3 million was spent at the
department store. Adding this figure to the export sales
potential of Table 24, yields a grand total expected
annual sales potential of $65.0 million for South Bay
Center, including a department store sales potential of
$26.6 million.
A forecast of expected 1980 sales levels of
Redondo Beach regional shopping center facilities is
presented in Table 25. The forecast assumes slower rates,
and in some cases smaller absolute amounts, of population
growth in the South Bay area. The forecast also makes
an assumption that the parameters of spatial interaction
(e.g., retail competition, transportation facilities,
etc.) remain fairly constant, thus instituting little
change in consumer shopping behavior in Redondo Beach
and its trade area. This assumption of parametric con-
TABLE 25
FORECAST OF ANNUAL SALES POTENTIAL FOR
REDONDO BEACH REGIONAL SHOPPING CENTER FACILITIES TO 1980
Trade
Area
Seqment
Net trade
area income
potential of
South Bay
Center
(millions $)
Proportion of income spent at South Bay Center
Major
department store Remaining stores Total center
factor (millions $) factor (millions $) factor (millions $)
1 45.7 .05 2.3 .09 4.1 .14 6.4
2 114.0 .07 8.0 .11 12.5 .18 20.5
3 116.6 . 06 7.0 .10 11.7 .16 18.7
4 362.5 .03 10.9 .07 25.4 .10 36.3
5 19.7 .04 0.8 .08 1.6 .12 2.4
6 40.6 .06 2.4 .10 4.1 .16 6.5
7 82.3 .07 5.8 .11 9.1 .18 14.9
8 155.4 .07 10.9 .11 17.1 .18 28.0
9 85.3 .07 6.0 .11 9.4 .18 15.4
Total export 1,022.1 - 54.1 - 95.0
- 149.1
Redondo
Beach 134.0 .07 9.4 .11 14.7 .18 24.1
Grand Total. 1,156.1
-
63.5 -
109.7
- 173.2
Source: Population and Dwelling Units,
Los Angeles County Regional Planning Commission,
Population Research Section, Population
Forecasts to 1980.
303
304
stancy is reflected in the unchanged (from Table 24)
factors allocating consumer expenditures, emanating
from the different trade area segments, to South Bay
Center.
In an actual trade area study, an assumption of
no parametric change over a twenty year forecast would
be unwise. Careful anticipatory analysis should be
undertaken to discern trends of change in any of the
parameters affecting consumer behavior in the trade
area, or even causing an enlargement of the trade area
itself. For example, a new freeway opening adjacent to
South Bay Center a couple of years ago has, in all
probability, extended the center's trade area along a
corridor to the north through West Los Angeles and,
perhaps, even to the San Fernando Valley. These types
of change must be anticipated and accounted for as fully
as possible if realistic sales forecasts are to be
achieved.
The micro-space-input analysis.— Once income and
retail sales levels have been determined and forecasted
for the community, technical space productivity conditions
in the local retail trade industry have to be studied to
determine space-inputs required to supply the retail
output implied in the income and sales forecasts.
Standards of retail space utilization efficiency are
derived from empirical data collected for different
types of stores. These data indicate the dollar sales
capacity of a particular retail store type per square
foot of enclosed space devoted exclusively to selling.
As noted in Chapter V, historical data trends are
available only for department and specialty stores. The
McNair and Retail Merchants Association figures are the
ones used in this study. For other store types, hypo
thetical figures are supplied, based on miscellaneous
marketing study references to space productivity, for
purposes of exposition.
Because the suburban shopping center under dis
cussion is of fairly recent vintage, and because fore
casts of future requirements are being made, the product
ivity figures used represent space productivity averages
for department stores in the upper one-fourth of the
distribution for all stores in the sample. In other
words, these are "best" industry standards of space use,
and not just "typical" or median standards.
Table 26 presents the space capacity figures that
should prevail for Redondo Beach regional retail
facilities. The per square foot sales volumes in
Table 26 are those which should prevail in a regional
shopping center that has attained a maximum sales volume,
306
TABLE 26
RETAIL SPACE PRODUCTIVITY STANDARDS FOR
REDONDO BEACH REGIONAL SHOPPING FACILITIES
Store type
Annual dollar
sales per
sq. ft. of
selling space
Department store 110
Variety store 105
Women's apparel store 76
Men's apparel store 84
Shoe store 75
Other apparel store 72
Furniture store 51
Appliance store 151
Drug store 149
Other stores 76
Total center 95
Sources: Malcolm P. McNair, Operating
Results of Department and Specialty Stores, 1953-1962,
(Harvard University, Bureau of Business Research
Bulletins Nos. 156-165); and Controllers' Congress,
Merchandising and Operating Results of Department and
Specialty Stores, 1950-1964, (New York: National Retail
Merchants Association).
307
say, $100 million per year for the center including $50
million for the department store, and is enjoying most
of the economies of large scale operation.
These retail space productivity figures can then
be used to develop retail space-input coefficients which
can be applied directly to anticipated sales volumes to
obtain space requirements. Where data availability
permits, observable trends in these technical product
ivity figures should be accounted for in anticipating
future retail space requirements. For example, trends
of increase in scale of retail operation which might
affect space productivity levels over time should be
noted when predicting future space needs.
The proposed model.— At this point, the entire
analytical procedure can be brought together, and a set
of retail space-input coefficients developed. Using an
economic base and income multiplier approach, income
levels were forecasted to 1980 for Redondo Beach and its
surrounding retail trade area. Then, with the aid of a
trade area study, income and sales potentials were deter
mined for regional shopping facilities located in
Redondo Beach. Finally, a space requirements analysis
was conducted to ascertain applicable standards of
retail space use.
308
Retail space-input coefficients can now be
calculated and applied directly to projected retail
sales (output) levels. In 1960 it was assumed that the
South Bay Center department store attained a sales level
of $28.3 million. Table 27 indicates that a department
store serving a regional trade area should be capable
of attaining a sales capacity of $110 per square foot
of selling space. Dividing the department store sales
level by the productivity figure, yields a space-input
requirement of 257,000 square feet of enclosed sales
space to supply that retail output level. In turn,
dividing the space-input figure by the department store
sales level yields a coefficient value of .009. Assum
ing that space productivity standards of department
stores do not change radically during the forecast
period, this space-input coefficient can be applied
directly to present and future sales levels to derive
space requirements.
Table 27 offers an example of the manner in which
space-input requirements can be determined at a municipal
level for regional shopping facilities. The womens'
apparel store space coefficient of .013, for example,
when applied to that store's 1960 and 1980 sales levels
of, respectively, $3.7 and $9.3 million, yields space-
inputs of 49 thousand square feet in 1960 and 121 thousand
TABLE 27
SPACE-INPUT REQUIREMENTS
IN REDONDO
OF REGIONAL SHOPPING FACILITIES
BEACH 1960 and 1980
Regional
shopping
facilities
Sales level
Space
Coefficient
Space-input
1960b 1980b
(thousands of sq. ft.
of selling space)
I9603 1980a
(millions $)
Department store 28.3 63.5 .009 257 572
Variety store 9.1 21.0 .009 87 189
Women's apparel store 3.7 9.3 .013 49 121
Mens' apparel store 2.4 8.7 .012 29 104
Shoe store 1.1 4.1 .014 15 57
Other apparel store 1.7 6.8 .014 24 95
Furniture store 3.4 12.2 .020 67 144
Appliance store 4.5 17.3 .007 30 121
Drug store 3.6 12.4 .007 24 87
Other stores 8.7 22.9 .013 114 298
Total facilities 66.5 173.2 - 596 1, 788
Source: Same as Tables 16-26.
aThe assumed actual sales levels for the center from Table 22
are used here, rather than the assumed potential sales levels arrived at
in Table 24.
T.
Multiplying sales level by coefficient may not yield exactly
the same figure as appears in the space-input columns, because the
coefficients are carried to only three decimal places to simplify
calculation, and the space-inputs are rounded to nearest thousand.
309
square feet in 1980. Indicated space-inputs for all
Redondo Beach regional shopping facilities could be
expected to rise from 596 thousand square feet of
enclosed selling space in I960, to one million seven
hundred eighty eight thousand square feet by 1980.
The difference between assumed actual regional
center sales in 1960 of $66.5, from Table 22, and
assumed potential 1960 sales of $65.0, from the hypo
thetical field survey, indicates that the center might
be drawing business from sections of the metropolitan
area not included in the survey area. The survey area
could have been extended, as noted above, by a parameter
change, such as the freeway opening adjacent to the
center; assuming that there is only a small margin of
error in the survey results. In the case of an exist
ing center, license plate surveys or shopper interviews
should be conducted at the center at regular intervals
to determine prospective new areas to sample. In the
case of a proposed center, pilot surveys should be
conducted in every area where there appears a reason
able probability of some significant level of consumer
attendance at the center.
At its 1960 sales level, the South Bay Center is
assumed to have been generating sales of about $132 per
square foot of selling space. The department store is
assumed to have been generating about $126 per square
foot of selling space. In comparison with the standard
productivity figures in Table 26, of $95 and $110 per
square foot of selling space, there is indication of
over-utilized, i.e., insufficient, retail space to serve
regional retail trade needs. Table 22 shows a space-
input at South Bay Center of only 500 thousand square
feet of selling space, while Table 27 indicates that
596 thousand square feet are required to adequately
serve regional retail needs, given the present state of
retail technology.
The projected space-input requirement for regional
retail facilities, of one million seven hundred eighty
eight thousand square feet of selling space in 1980,
warrants careful attention. It can imply many things
in terms of physical retail plant. To begin with, at
the present state of retail technology, and with current
consumer shopping preferences, a single shopping center
of this size is not feasible. The implication is that
there would be at least two, and probably three, separ
ate regional centers needed to handle the indicated
sales output.
There is further need, of course, to analyze site
availability and automobile and other transportation
access within Redondo Beach, to determine whether or
312
not it would be feasible or desirable to handle a
retail space-input of the projected magnitude within
the municipal boundries.
To determine total land site requirements within
the city for regional retail facilities, currently
acceptable ratios of gross enclosed space and parking
area to enclosed selling space can be used. Hoyt says
that
In the regional suburban center, the
selling area is usually a higher
proportion of the gross building area
than in a downtown store, because the
downtown store maintains executive
offices, buyer's offices, advertising
department and warehouse space. We
estimate that 70 to 80 per cent of the
gross area in new suburban stores will
be used for selling.24
It is assumed in this study that selling area can be
expanded by forty per cent to yield gross enclosed area.
A parking index indicates the number of parking
spaces per 1,000 square feet of gross floor space.
There are numerous factors that determine what a good
parking index might be for an individual shopping
center of given type and size. Some of these factors
are: availability of public transportation; amount of
^Horner Hoyt, "A Re-Examination of the Shopping
Center Market," Technical Bulletin No. 33,
Urban Land Institute, Washington, D.C., 1958.
p. 9.
313
"walk-in” trade; and location of the center in terms
of its trade area. Each situation has to be analyzed
on the basis of its individual characteristics. Six
to seven spaces per 1,000 square feet of gross floor
area are the most frequently quoted figures applicable
to regional shopping centers. Let us assume that seven
parking spaces per 1,000 square feet are required.
Total municipal land allocations to regional
retail use can now be made. Table 28 indicates total
land requirements according to income, sales, and
space-input data developed and projected by the model.
Between 1960 and 1980 land allocations to regional retail
use should increase threefold if proper levels of retail
supply are to be maintained.
TABLE 28
LAND ALLOCATION REQUIREMENTS FOR
REGIONAL RETAIL FACILITIES IN REDONDO BEACH I960 and 1980
Regional retail use 1960 1980
Gross enclosed space
(sq. ft.) 834,000 2,503,000
Parking space
(sq. ft.) 1,795,000 4,906,000
Total land area
(sq. ft.) 2,629,000 7,409,000
(53 acres) (150 acres
Sources: See Tables 16 through 26.
j
r
314
Data Requirements of the Model
Secondary data sources.— There are numerous
governmental and private data reporting sources that
are useful in conducting economic base, trade area, and
retail industry studies. There are severe limitations
on the degree of usefulness of most of these sources,
however, especially in relation to economic base and
trade area analyses. These limitations are pointed
out below with the discussion of primary data collection.
The U. S. Census of Population provides basic
population and income information by cities and metro
politan areas. Reasonably accurate income allocations
can be made by census tracts within cities, and even
by blocks, by using average income figures and popula
tion distributions. Trade areas may either coincide
with census tract boundries, or, rough allocations of
population and income can be made, if the approximate
proportion of the tract included in a trade area is
known. The infrequent reporting intervals on census
population and income data cause it to become inadequate
for small area analysis after a few years. This in
frequency necessitates updating and other adjustments.
In between census years, municipal data are use
ful in updating population estimates. New telephones,
light and water meters, and building permits aid in
estimating new population in-migrations. However,
they do not provide information on incomes.
The retail trade section of the U. S. Census of
Business is published on a more frequent basis, about
every four years, and provides complete information on
retail sales levels by business group and store types
for cities, counties, and metropolitan areas. Compari
son of per capita sales levels of small cities with
those for benchmark areas, such as counties or metro
politan areas, indicates whether a community is just
serving its own needs or whether some exporting activity
is occurring. Census data also permit income estimates
to be made from figures for value added by local
businesses. Applebaum also points out that
Per capita expenditures for department
store-type merchandise are determined on
the basis of Census of Business reports
of department store sales adjusted by
Federal Reserve Board index information
for year subsequent to the Census. There
are many pitfalls inherent in using these
kinds of data, but they are about the only
reasonably reliable data available.25
Various issues of the Survey of Current Business,
published monthly by the Office of Business Economics,
contain data on disposable personal income, retail
sales, and consumer retail expenditures.
^ A p p l e b a u m , Store Location..., 34, 35.
316
The Bureau of Labor Statistics has, on occasion,
conducted surveys of consumer expenditures. The Bureau
makes yearly surveys in a few selected cities. In
1951, the Bureau made a national study, in cooperation
with the Wharton School at the University of Pennsyl
vania, and published the seventeen volume Study of
Consumer Expenditures, Incomes, and Savings. This
invaluable and monumental work is now becoming out
dated. It was a "once only" effort, which really needs
to be conducted on a continuing basis and published
at regular intervals. This study did, however, provide
precise and accurate information on the allocations of
disposable income among various categories of current
consumption expenditure.
An excellent private source of market data is
the "Survey of Buying Power," published annually by
Sales Management magazine. It contains up-to-date
estimates of population, number of families, incomes,
and retail sales by states, counties, and cities.
Census data have verified that the Sales Management
estimates are remarkably accurate.
The University of Michigan Survey Research
Center's Survey of Consumer Finances series contains
valuable information pertaining to family budgetary
and expenditure processes. This data source offers
much information about consumer motivations and deter
minants of consumer expenditure propensities.
Location quotient, minimum requirements, market
potential (gravity), and geographic allocation techniques
can be applied to population, income, sales, and expend-
iture data from any of the above secondary sources.
The results will vary in degree of accuracy, but for
small areas there is usually a large margin of error.
Nonetheless, for quick surveys to derive an outline of
the economic terrain, secondary data sources and in
direct measurement techniques are useful.
Generation of data from primary sources.— Economic
base and trade area studies require data that is broken
down by areal groupings other than those of the above
secondary data reporting sources. Converse says that
A serious limitation on the use of Census
data which should be kept in mind is that
the figures given are for political
divisions, such as counties and cities,
whereas retail trade crosses political
boundries freely.26
An economic base study requires knowledge of the
geographic source of demand for local products and
services. Market potential analysis requires knowledge
^P. D. Converse, "A Study of Retail Trade
Areas in East Central Illinois," University of Illinois,
Bulletin, Business Studies No. 2, 1943, 11.
of the location and nature of shopping goods demand by
small area. For these reasons, Loesch suggests
"...the direct questioning of consumers in areas of
retail trade."2^
In Redondo Beach, two sectors of the economy would
be interviewed: the retail sector; and the household
sector. A fairly complete census of retailers would
be conducted in a community the size of Redondo Beach.
In the Appendix, Exhibit 1 presents a field questionn
aire form used in an economic study of Redondo Beach
in 1963.
The important information generated in this
questionnaire pertains to: (1) geographic destination
of sales (i.e., inside or outside the city, and, if
outside, what segment of the trade area); (2) type of
store and which center or district located in; (3)
square feet of enclosed floor area; and (4) residential
location of employees. Other information generated
is for lesser analytical and descriptive purposes.
The gravitational attraction of Redondo Beach
retail facilities is ascertained when knowledge is
obtained about the geographic distribution of customer
trip origins.
2^August Loesch, The Economics of Location,
(New Haven: Yale University Press, 1954), p. 395.
Exhibit 2 in the Appendix presents a home survey
questionnaire used in the Redondo Beach economic study.
This questionnaire was intended for use only within
Redondo Beach for an economic base study. However,
with slight modifications a similar form could easily
be utilized in surveying the trade area outside Redondo
Beach.
The important data generated in the household
survey relate to: (1) amount and geographic origin of
income; (2) socio-economic characteristics determining
spending propensities, e.g., employment, number of
children, length of residence in the area, etc.; (3)
detailed information about types and locations of retail
expenditures, and amounts spent; and (4) miscellaneous
shopping attitudes that might affect the choice of a
shopping center.
Field survey questions must be structured so that
respondents clearly understand them and give meaningful
answers to them. For these reasons, the questions on
amounts and places of retail expenditure were broken
down into segments of expenditure such as durables,
nondurables, etc. Similarly, the questions on income
and savings were stated in terms of one-thousand-dollar
intervals rather than a request for an exact income or
savings figure.
320
Generally speaking, for accuracy a household
sample must be carefully selected, questionnaires so
worded as to elicit desired (objective) information,
and interviewers well trained and competent. Converse
feels that "The technique of questionnaire surveys has
been developed to such an extent that a high degree of
accuracy can be attained."28 The writer, as stated
previously, feels that retail and household field
surveys are a necessity for meaningful results with
the proposed combined analysis.
28Converse, University of Illinois Bulletin,
Business Studies No. 2, 1943, 11.
CHAPTER VII
EVALUATION OF THE MODEL AS PLANNING ANALYSIS
The Model and Social Cost
The concept and measurement of retail market
saturation.— The purpose of this section is to
discuss briefly social costs arising from overexpansion
and misplacement of retail facilities. These are costs
that could be reduced or avoided through use of the
proposed allocation model approach of determining
retail space requirements.
The concept of market saturation is often employed
to determine whether or not a trade area is adequately
served by its existing retail facilities. Applebaum
and Cohen define market saturation by stating that,
When the market area is served by stores
so located, physically planned, and
merchandized that customer needs are
adequately met and use of store
facilities is efficient, then the
market is in balance— or 'saturated'.
A saturated retail market, then, is one in which con
sumer demands are adequately met by retail facilities
•1-William Applebaum and Saul B. Cohen,
"Store Trading Areas in a Changing Market, Part I,"
Journal of Retailing, (Fall, 1961), 14.
displaying standard levels of productivity.
A retail market would be over-saturated or over
stored if there were more facilities available to meet
consumer demand than were required. A market would be
under-saturated or under-stored if the opposite situation
were true. The question now is, by what criteria is a
retail market judged to be over-, under-, or adequately
saturated?
There are numerous yardsticks by which to measure,
but Applebaum and Cohen suggest
Sales per square foot of selling area.
This, in the opinion of the authors,
does measure productivity and degree
of efficiency of use of space.
Therefore it is suggested as the
basic measure of saturation.2
There are space capacity norms of selling area producti
vity for different retail store types, as indicated in
the preceding chapter, so that comparisons of "expected"
and actual per unit sales could be made to ascertain
degree of market saturation.
McNair's data for retail space indicate a de
clining effectiveness in its utilization over the past
five years. McNair says that, "No doubt this situation
2Applebaum and Cohen, "Store Trading Areas...,
Part II," Journal of Retailinq, (Winter, 1961-1962).
33.
323
is connected with the present rapid expansion of retail
store space, the 'over-storing' that seems to be pre
valent in many cities t o d a y . McNair feels also
that department stores have been turning in a mediocre
profit performance in recent years, and to improve it
they must raise their sales per unit of space well
above prevailing rates. This may imply a slower rate
of store space expansion in some locations.
A given department store, for example, may appear
insufficient to serve a trade area because its sales
per square foot are higher than the expected level.
But, on the other hand, it may have what is considered
by management to be just an adequate share of the
total marhet, given existing competition, in other
words, if a similar department store were to open in
the trade area, neither might then be able to obtain
a sufficient marlcet share to operate profitably or
obtain an adequate return on investment.
Municipal government planners tend to emphasize
consumer service in the trade area, while private
retail developers and managers constrain their activity
to conform to certain profit and return criteria that
they deem minimal for operation. In the local planning
JMalcolm P. McNair, Operating Results of
Department and Specialty Stores. 1961, (Boston:
Harvard University, Bureau of Business Research Bulletin
No. 164, 1962), p. 14.
process there must be a satisfactory compromise between
these two viewpoints of retail service.
Normally, a trade area should be considered under
stored if: (1) sales per square foot of selling space
are unusually high by expected capacity standards;
(2) customers must travel farther to shop than they
desire; and (3) shopping conditions are too crowded
or slow by normal standards. When the reverse is
true, a trade area should be considered over-stored.
Generally speaking, if retail establishments
or shopping centers as a whole are achieving very high
sales per square foot for shopping goods, Lakshmanan
and Hansen feel that,
...consumers in these and surrounding
districts are poorly served. High
sales per square foot in such areas
would result either in an increase in
the size of the centers or in the
development of new stores nearby to
serve the consumers.^
On the other hand, too many stores may enter a
trade area in the heat of competition. Retail develop
ers have many motives for opening new stores. They
may be willing to lower their yardsticks for market
saturation and open a new store for any one or more
4t. R. Lakshmanan and Walter G. Hansen,
"A Retail Market Potential Model," Journal of the
American Institute of Planners, (May, 1965), 139.
r
325
of the following reasons: (1) to gain better purchas
ing terms from wholesalers; (2) to provide or better
use warehouse facilities; (3) to provide more retail
outlets for their own manufactured goods; (4) to reduce
relative costs of advertising or store supervision;
and (5) to anticipate future store space requirements
for handling additional lines of goods.
One market researcher has constructed an Index
of Retail Saturation (IRS) to measure the degree of
trade area saturation. His index is set out in the
following formula:
where IRS^ is the index for area one, is the number
of consumers in area one, RE^ is the retail expenditures
per consumer in area one, and RF^ represents the retail
facilities in area one.
Consider the example of a supermarket. Suppose
that there are 100 thousand consumers in a trade area
spending an average of $5.50 per week in food stores,
and that there are fifteen markets with a total of
144 thousand square feet of selling space. Then
RFj
= 100.000 x $5.50 = $550.000
144,000 144,000
= $3.82 per
square
foot
The figure of $3.82 per square foot is measured against
the number of dollars per square foot required to
break-even (or maximize profits by equating marginal
cost with marginal revenue, or price, if pure-perfect
competition prevails), thus providing a measure of
retail saturation in the trade area.5
Social and private costs arising from improper
retail land use planning.— Many detrimental
effects are caused by uncontrolled expansion of stores
in a trade area. As Entenberg points out,
Many department stores have opened in
uncontrolled shopping areas or on too
small scale and have been quickly choked
off by haphazard traffic facilities,
often due to the opening of all manner
of retail, service, and other establish
ments . 6
In these situations department store branches never
realize their full sales potentials and utilization.
These disappointing operations have usually resulted
from inadequate economic and market research and im
proper planning— bad location and depths and assort
ments of merchandise lines carried.
5 .
William D. Stevens, "The Social Responsibilities
of Marketing," Proceedings of the American Marketing
Association. (1961), 572, 573.
C
Robert D. Entenberg, The Changing Competitive
Position of Department Stores in the United States
by Merchandise Lines. (Pittsburgh: University of
Pittsburgh Press 1957), p. 103.
327
Uncontrolled entrance of retail stores and centers
into a community results in social and private costs in
the form of wasted land (space) resources, and under
utilization of capital facilities. The most prevalent
problem arising from improper retail land use alloca
tions is that of market over-saturation— too much
retail space. Excessive amounts of retail space cause
the existence of many marginal operations that clutter
the local landscape.
At the municipal level, there are usually strong
speculative pressures for commercial zoning, both in
specific locations and for a disproportionately large
amount of commercial land area generally. Shopping
center developers prefer to purchase cheap open land
presently zoned for another (lower) use, and then
subsequently apply to the municipal jurisdiction for
a zoning change or variance. The attractiveness of
proposed shopping centers often makes the zoning
changes inevitable, regardless of economic consequences
for the community.
In the 1920's many zoning ordinances were merely
predicated on maximum retail development. Property
owners and real estate promoters insisted on alloca
tion of their land for commercial use in the naive
belief that this would automatically enhance its value.
i
J
I
328
Delafons says that,
They were abetted by an influential
group of professional planners who
argued that zoning must reflect
whatever use the land seemed best
adapted for, and that if scattered
commercial development existed it
must be assumed that all land in
the vicinity was equally well adapted
for such development. The result was
vast over-zoning.7
At times in the past commercial zoning in the U. S.
cities has been up to five times the area in use at the
time. Often, a much greater density of retail develop
ment is permitted than is potentially feasible. This
leads to both an extensive waste of land resources
and an intensive waste of capital and labor resources
in a community.
Houston provides an example of the kind of retail
development that occurs without any land use planning.
There are virtually no zoning ordinances at all in the
city of Houston. Delafons notes in the case of Houston
that,
As always, the really intractable use is
the low-status commercial— the shoe-string
operations that clutter up every American
city. There certainly seem to be more of
these in Houston than anywhere else,
stretching out along every approach to the
city in an endless string..
7John Delafons, Land-Use Controls in the United
States, (Cambridge: Joint Center for Urban Studies
of M.I.T. and Harvard University, 1962), p. 27.
8Ibid.,82.
329
Houston provides strong evidence that failure to
anticipate current and future retail needs, and to
allocate specified quantities of land area for retail
use in economically justified and aesthetically desira
ble locations, leads to a state of retail competition
that is -wasteful of land and other resources and
socially undesirable.
Retail land use planning cannot be effectively
carried out unless quantitative and locational space
requirements are known in advance. In Britain,
Burns says that
...we are too often content with guessing
a minimum number and then saying that we
should extend the facilities as the demand
builds up; a method which sounds eminently
practical but which often results in a
waste of resources...9
Adding to retail space on an ad hoc basis indicates
lack of knowledge about a community's total retail
space needs on the part of municipal government plann
ing and zoning officials.
Sufficient land area should be zoned in properly
designated sections of a city to serve all the com
munity's retail needs, and surrounding land should
then be zoned for other compatible uses so that a
^Wilfred Burns, British Shopping Centres
(London; Leonard Hill Limited, 1959), p. 49.
1
I
330
ruinous proliferation of small retail stores cannot
occur. In the area immediately surrounding a planned
shopping center site, for instance, so-called intruder
stores are damaging to the extent that they do not bear
parking, landscaping, or development costs, but never
theless act as parasites on promotional efforts of the
center. The fringe retail growths hurt the appearance
of a center, blight nearby residential neighborhoods,
and create an often intolerable traffic situation.
Stedman says that these
Fringe pirating and promotional activities
could result in a congested, disorganized,
uncontrolled commercial slum. They could
destroy the investment value of a shopping
center.10
The main goal of municipal planning for retail
land use is to maintain an equilibrium of trade area
saturation, i.e., to insure that retail space-input
neither falls short of nor exceeds expected output
requirements. To maintain such an equilibrium, local
planning authorities must measure and anticipate changes
in: (1) population; (2) accessibility; (3) merchandise
assortment; (4) purchasing power; (5) retail technology;
(6) shopping patterns and buying propensities; and
(7) retail profitability or investment return concepts.
lOQordon H. Stedman, "The Rise of Shopping Centers,"
Journal of Retailing, (Spring, 1955), 21.
Applebaum and Cohen note that,
If all companies expand store facilities
beyond the needs of the market, then a
situation is created where all have low
ered their productivity in terms of sales
per square foot of selling area.H
This kind of costly under-utilization of retail space
can arise because anticipated population increases fail
to materialize, or basic transportation changes open
the trade area to outside competition, or consumer
shopping habits alter radically, etc. There is definite
need for adequate anticipations data on all these and
other possible future contingencies. The collection
of these data should be a continuing process on a
community scale. The proposed model provides the tools
of analysis and the municipal government land use plann
ing process serves as their matrix for application.
Relationship of the Model
to the Municipal Planning Process
Some inadequacies of municipal land use planning.—
Since the model is proposed as an analytical input to
the municipal government land use planning process to
firm up land allocation decisions? some areas of vague
ness and weakness in the relationship between local
11Applebaum and Cohen, Journal of Retailing,
(Winter, 1961-1962), 36.
332
planning controls and private development decisions
should be briefly explored. Delafons defines the
problem generally, when he points out that land use
planning in the United States, "...has never been used
effectively to control the programming of private
development,..."^-2 He feels that it could provide a
framework within which rational local retail develop
ment decisions could be made.
There may be numerous idealogical reasons why
local governmental units have generally failed to assert
planning controls in the interest of the entire com
munity. Within the scope of this study, however, the
major reason appears as a gap in the analytical and
informational input structure of the local land use
allocation process. In other words, land use planners
have not had available to them usable analytical pro
cedures to derive data on which to base land allocation
and zoning decisions.
After the war, municipalities were not prepared
to cope with the development of a new retailing innova
tion— the regional planned shopping center. Horwood
says that,
They could not be anticipated by the
municipality, and their relationship
- * - 2Delafons, Land-Use Controls.... 83.
333
to whatever municipal planning existed
had to be studied on an ad hoc basis.^
This amounted to complete acquiescence to private retail
planning decisions pertaining to local retail needs.
Wilhelm, as an example of the helpless attitude
of local planning officials, presents recent comments
by Austin, Texas zoning commission members. One member
was asked whether or not he felt there was an excessive
amount of land in Austin zones for commercial use.
His reply:
Too excessive. We should re-examine all
1 C‘ zoned areas and re-zone 'C areas
where it is clear that the property will
not be used commercially or is not the
best use for it. Of course, this does
not supply an answer to future commercial
use. We are without a means to prevent
this; we have no way of knowing how to
show too much 'C' zoning.14
There have been numerous past attempts at devising
measurement standards by which to judge sufficiency of
retail service. In 1942, Hall noted that the ratio of
commercial land in use per one thousand people varied
only between 2.15 and 3.15 acres for all but four out
of thirty sub-areas in Los Angeles County. By Hall's
■^Edgar M. Horwood, "Public Policy and the
Outlying Shopping Center," Journal of the American
Institute of Planners, (Pall, 1958), 217.
^Sidney Wilhelm, Urban Zoning and Land-Use
Theory,(Glencoe: The Free Press of Glencoe, 1962),
p. 153.
own admission,
This seemed to be still too wide a range
for use by the Commission as a basis for
zoning, however, and some further break
down of the figures was needed to show in
greater detail how this ratio varied in
different parts of the region.15
Even with geographic refinements, this acres-to-people
ratio is too crude a device for use in either ascertain
ing present adequate retail space levels or projecting
future space-inputs. It tells us nothing of incomes,
expenditures, shopping behavior, etc.
In the late 1950's, the Cuyahoga County Regional
Planning Commission conducted a comprehensive survey
and study of suburban shopping centers in the Cleveland
area.16 The Cleveland study attempted to find a re
lationship between store area and families, i.e., to
derive square feet of selling space required per family
Specific amounts of store space required were derived
from the average square feet per family that were
deemed adequate to serve a family's needs. Accordingly
the total store area devoted to different store types
in a shopping center should be in almost direct pro-
l^Bryant Hall, "How Much Land for Commerce?"
Journal of the American Institute of Planners,
(January-February, 1942), 3.
16see Cuyahoga County Regional Planning Commission,
Suburban Business Centers, Third Report on Land Use
in Cuyahoga County, 1959.
335
portion to that store type's share of net supporting
families in the center's trade area. The Cleveland
study indicated that the square feet of sales space-to-
family ratio should be a minimum of 25 square feet and
a maximum of 35 square feet. Thirty square feet per
family is the average for the entire United States.
While the method of measuring retail space needs
is considerably more refined in the Cleveland study,
it is still inadequate. A more explicit series of
relationships should be established; i.e., those lead
ing directly from trade area family incomes, to expend
itures, to retail sales or output, and, finally, to
retail space-input. The Cleveland study only indirect
ly establishes space requirements, based on family
spending habits. The proposed model provides a more
precise quantitative determination of retail space-
inputs.
The analytical gap in the land use planning process
described above, has resulted in a vague and unspecified
relationship between private retail development de
cisions and local government planning controls.
Horwood feels that the local land use planning process
is unable to cope with retail development in a meaning
ful way. He wants to know:
To what extent is the developer to take the
existing public "plan" as a fixed condition
in assessing his own opportunity area? How
much time, energy, and resources should he
expect to spend in changing it? The net
result is, in the present stage of planning
development, an open-ended system. Neither
the developer nor the public body can get
much guidance from the comprehensive plan.^7
Municipal land use planning can never provide firm
guidelines for local retail growth unless sound eco
nomic analysis specifying community needs is forth
coming .
In areas of rapid urban growth, where land use
patterns have not yet been clearly determined, zoning
maps have traditionally followed the private develop
ment of land, striving to firm up emerging use boundries
and preventing the more obvious infractions of land use
compatibility. Often, it is felt that zoning regula
tions have to be flexible because of the difficulty in
determining, precisely, future retail requirements.
While it is true that space-input estimates should be
revised through feedback process each time that new
income and expenditure estimates are made, this does
not imply that firm allocations of land to retail use
cannot be made. Land use allocations that follow
market demands on an ad hoc basis are quite different
-^Horwood, Journal of the American Institute
of Planners, (Pall, 1958), 217.
337
than planned allocations that are revised periodically
on the basis of sound analytical judgement.
Public planning for private retail development.—
In the United States, economic base studies are usually
conducted by or for a local government authority, and
trade area or marketing analyses are conducted by or
for individual retail enterprises— usually large organ
izations catering to a wide range of consumer needs.
Burns emphasizes that,
The same method could, however, equally
be used for calculating the itemized
floor area for a group of small shops.
The analysis would not then be done
for the private entrepreneur, but
would be done by the local authority
as developing authority, or as
planning authority.18
A municipality could easily incorporate trade
area and retail analysis into a macro-demand survey of
local economic potential. Public study could be made
of market potential and internal retail site availability.
Horwood says that,
Presumably this would follow a pattern
similar to a market analysis made by or
for a developer, though...the public
study might be more sensitive to the
social costs of one site above another...1^
-*-®Burns, British Shopping Centres, 50.
l^Horwood, Journal of the American Institute
of Planners, (Fall, 1958), 218.
338
Municipal planning authorities will be concerned with
more comprehensive questions, such as, how many and
what type of stores and centers should be permitted
within the jurisdiction? They will not be concerned
with market potential as it relates to one specific
investment decision, but, rather, as it relates to the
amount and rate of total retail development in the
community over the next few years.
The planning efforts of the City of Norwalk pro
vide a case in point. Municipal planning authorities
in Norwalk attempted to provide an analysis of regional
retail potential that would serve as an informational
framework within which local retail developers could
make investment decisions. Kopp states that:
...the City asked Stanford Research Institute
to study the market for a regional shopping
center in Norwalk. The City wishes to provide
potential developers and retail tenants with
information on the market support that would
be potentially available to a facility of
this type. In addition, the economic
information that would be developed on
the City of Norwalk and its contiguous
areas during the course of the study could
be used as a basis for other planning
decisions.20
^William t. Kopp and Louis Salomone, Jr.,
The Market Potential for A Regional Shopping Center
in the City of Norwalk. (South Pasadena: Southern
California Labs of Stanford Research Institute,
November, 1960), p. 1.
339
The main deficiency of the Norwalk study is that
it represents a "one-shot" analysis intended to ascer
tain regional retail location possibilities in the city
at the present time. This type of analysis, or at
least certain data collection and tabulation elements
of it, is probably best conducted by local planning
staff members so that it can be carried out on a con
tinuing basis. A somewhat greater time perspective is
required to derive objective planning decisions on
rates of local retail development at all trading levels.
The final decision as to whether or not new
shopping center development will be initiated in a
given trade area demand and competition situation de
pends, of course, on the policies of the retail entre
preneurs involved. Some may be satisfied by a lower
volume of sales and feel that an opening in the market
exists when, say, twenty per cent of a market demand
area is not sufficiently served by existing retail
facilities. Entrepreneurs with higher volume require
ments will not be satisfied until a greater sales
potential exists.
In spite of these differential market entry levels
for various retail developers; income, expenditure, and
sales forecasts can still serve as excellent predictors
of retail development opportunities in a particular
340
locality because, generally speaking:
Retail patterns are endogenous, generated
by a prior spatial allocation of consumers
and their characteristics...Changes in
numbers of consumers, therefore, lead to
retail growth in certain areas, generating
new space demands and consequent problems
of allocation of this space relative to
other uses,...21
When market potential and retail space needs are being
estimated and forecasted on a continuing basis, minor
discrepancies between expected and actual timing of
private retail investment are not critical. The real
problem arises when gross error in planning judgement
occurs, and greatly excessive amounts of land are zoned
commercial and, perhaps, improved with unsightly margin
al operations.
The public or local governmental element in the
retail development process, as seen in this study, is
not a passive factor merely fulfilling a data gathering
and information dispensing role. A municipal master
plan of land use, adhered to on the basis of sound
judgement founded on relevant economic analysis, does
itself serve as a predictive device and determinant of
the future configuration of community land use.
21
Brian J. L. Berry, "The Retail Component of
the Urban Model," Journal of the American Institute
of Planners, (May, 1965), 151.
341
Applebaum and Cohen feel that:
Government and political influences can
be put to positive use. They can serve
to induce or anticipate change. In the
United States, where zoning and licensing
may prevent builders from bringing changes
onto the scene, these can also serve to
encourage the development of new and
needed shopping facilities.22
By sufficient anticipation of future market potential
and probable retail development requirements, a
community places itself in the position of being able
to exert some influence on the desired nature and
location of its retail facilities.
22Applebaum and Cohen, Journal of Retailing,
(Winter, 1961-1962), 43.
CHAPTER VIII
SUMMARY AND CONCLUSIONS OF THE STUDY
Summary
Methods of economic analysis pertaining to
municipal retail land use allocations.— This study
has presented three forms of urban economic analysis
that can be applied to study of intra-metropolitan
urban planning problems. The three analytical approach
es to local area analysis discussed in Chapters III,
IV, and V should be sufficient for both an overall
study of local economic potential, and to serve as the
basis for specific planning decisions within a community.
Economic base analysis provides information from
which an outline of a community's trading relations
with the remainder of its metropolitan area can be
obtained. The overall economic potential and position
of a community relative to its metropolitan setting is
ascertained. It is the economic base study which in
volves analysis extending significantly beyond local
boundries. Observations and forecasts are made of
metropolitan, regional, and, perhaps, national, economic
trends and forces that affect local income and employ
ment levels. Economic base analysis is primarily con-
343
cerned with community economic activity levels and
sources of demand for local products and services.
Trade area analysis provides information from
which knowledge of the extent and nature of community1s
economic relations with the immediately surrounding
segment of the metropolitan area can be obtained.
Detailed study of shopping behavior within a community
and that portion of the surrounding territory that its
retail facilities serve is conducted. Trade area ana
lysis concentrates on a micro-geographic study of a
community's retail demand area.
Retail space requirements analysis provides in
formation relating to standards of efficiency of space
utilization. Detailed study is undertaken of the retail
facilities within a community. The economics of local
retail supply are analyzed. Site availability and
rent structure are determined for various types of
retail facilities. Space requirements analysis con
centrates on a micro-geographic study of the objective
factors of local retail space-input and location.
A model of retail space-input.— The second section
in Chapter VI offers a suggested method of combining
the three relevant analytical approaches to community
economic analysis. The result of this combination is
a model for allocating land (space) to retail use with
in a municipal jurisdiction.
The analytical components of the model are applied
to the specific problem of allocating land area for the
use of regional shopping facilities within a community
in a larger metropolitan area. The macro-demand com
ponent of the model provides estimates of present in
come levels in the community, and forecasts are made to
a target date in the future with the aid of an income
multiplier. The micro-demand component allocates
potential expenditures from local and trade area resi
dents to the community's regional shopping facilities.
Allocation factors are applied to both present and
future income potentials derived from the economic base
study and forecast. Finally, retail space productivity
standards are related to expected sales of the regional
shopping facilities to derive space-input coefficients.
These constants of retail space-input are then applied
directly to expected present and future sales levels.
Total land area requirements for regional shopping
facilities are ascertained through application of
relevant ratios of gross-to-sales space in stores, and
parking area-to-enclosed gross store space in shopping
centers. Use of these ratios permits expansion of
space-input estimates to total acreage figures for
allocation and zoning purposes.
The model and the local planning process.— This
study also attempts to relate the proposed analytical
methods to the local planning process— specifically,
to the land use planning component of that process.
In addition, the proposed model itself is evaluated in
terms of its applicability to local land use planning
analysis.
In Chapter II, municipal governmental and market
allocative processes are discussed, as they relate to
the determination of retail land use within a community.
The occurence of social costs, as they arise during the
market "planning" process, and their possible amelior
ation through governmental planning supplementation of
this process, are also treated.
Through actual impurities and imperfections in
local retail and urban land markets, malallocation of
land resources can take place. In addition, even under
pure-perfect competition, with an efficient private
allocation of land resources, social costs may still
be incurred by the community. Because of the phenomenon
of social costs, municipal government planning is a
requisite supplement to private planning decisions
made within a market framework.
346
The first section of Chapter VI more specifically
discusses municipal government analytical requirements
for retail land use allocation decisions. Chapter VII
evaluates the proposed model itself as an analytical
procedure for use in the municipal government land use
planning process.
The evaluation of the model in Chapter VII begins
with a discussion of its relationship to social cost,
in terms of planning procedures which might avoid over
saturation or under-saturation of local retail trade
areas.
A suggested method of implementing the model in
the local land use process is offered in the second
section of Chapter VII. Reference to "public planning
for private retail development" implies that municipal
governments should supply a planning framework within
which private retail developers and investors can
operate. Firm land use allocations should be made,
providing private.decision makers with an explicit and
solid relationship to the local planning process. The
retail space allocation model, of course, supplies the
local planning process with the analytical input
necessary to firm and accurate anticipation of present
and future retail space requirements.
347
Conclusions
The utility of the proposed model.— In many in-
stances, municipalities, located within larger metro
politan areas, serve as focuses for concentrations of
retail and other commercial activity. Sites within
their jurisdictional boundries become logical locations
for new planned regional shopping centers. These shopp
ing centers will serve local retail needs of the city
itself, and those of a much larger trade area extending
beyond municipal boundries.
In 1948, Bushnell, recognized the need for examin
ing the trade areas of given clusters of regional re
tail facilities, and relating these trade areas to
municipal retail land use planning needs. He gives
implicit recognition to the fact that a city may be a
retail exporter to a large segment of its metropolitan
area in his statement that:
There is also evidence that these prominent
shopping centers have considerably more
retail stores per thousand persons of the
city's own population. The wide scope of
the retail services rendered, and the large
proportion of retail sales that come from
the trading area outside the city, of
course account for this relatively larger
number of retail stores.1
Ijohn D. Bushnell, "Regional Shopping Centers:
Experience in Los Angeles County," Journal of the
American Institute of Planners, (Fall, 1948), 12.
The analysis provided by the retail space alloca
tion model fully recognizes the positions of some intra
metropolitan communities as regional retail trading
centers. The model accounts for the economic potential
emanating from the trade area and relates it to a
community's expected labor and retail export levels.
In a more general sense, the model is useful in
providing a broad factual basis for comprehensive com
munity planning. It can offer a series of successive
approximations (through a feedback process) over time
so that a realistic, practical, and balanced plan can
be devised. The plan could then account for the major
forces determining retail output and space-input, while
simultaneously conforming land use patterns to overall
community desires and goals.
The model in perspective.— The writer's final
conclusion is, of course, that the proposed model is
a valuable device for assisting in the land use planning
process at the local geographic level of analysis. No
implication is intended, however, that the proposed
analytical procedure, as presented in this study, is
equally valid at other geographic levels of analysis.
Metropolitan and regional economic growth and
development studies emphasize different variables, and
349
classify different variables as exogenous or endogenous
factors, from those in local studies. Wider resource
bases at the regional level, for example, cause a much
greater proportion of growth factors to be internal.
Exports may no longer be the critical element in
initiating and maintaining economic growth.
Also, in regional studies, economic location theory
plays an important role. Location factors that are
purely random at the local level, may be explained by
transport, labor, or resource orientations on a
regional scale.
The model is ideal for application to intra
metropolitan retail space-input and location problems.
With either slight, or, perhaps, no modification, it
could be applied on a metropolitan scale. The model
is seen by the writer, though, as being most applicable
to intra-metropolitan planning studies.
t
APPENDIX
350 1
APPENDIX
Survey Questionnaires
The retail questionnaire.— Exhibit 1 consists of a
questionnaire designed for use in a survey of retail
store proprietors and managers, and shopping center
developers and operators. Numbers in the far right
column by certain questions are for coding purposes.
The survey data can be transferred to IBM punch cards
and tabulated, sorted, and cross-sorted, thus greatly
reducing computation time.
This questionnaire was originally intended for use
in the Redondo Beach economic base study conducted by
the writer and others in 1963, however, time limitations
prevented fielding of the retail survey.
The household questionnaire.— Exhibit 2 consists
of a questionnaire designed for use in a survey of
families residing in a specified study area. Again,
numbers in the far right column are for coding purposes.
This questionnaire was fielded for the Redondo
Beach economic base study in January, 1963.
351
EXHIBIT 1
REDONDO BEACH RETAIL QUESTIONNAIRE
Name of Store:_________________ Shopping District No.__
Store Type No.__
1. Identity of respondents: Owner 1
Manager 2
Other employee 3
2. Under what name are you doing business? _________
3. How long has this store been under
the present name?_______________________ _________
4. How long have you owned/managed this store?_________
5. What is the gross square foot area? _________
6. What is the net square foot area
devoted to sales? _________
7. How many people do you employ?______________________
8. What merchandise lines do you handle? _________
01 Apparel
1 Women's
2 Men1 s
3 Family
4 Shoes
02 General Merchandise
1 Limited-price variety
2 Department store
3 Other
03 Grocery
1 No alcohol
2 Beer, wine
3 All alcohols
4 Liquor store
04 Specialty
1 Gifts, art, novelties
2 Sporting goods
3 Florist
4 Photographic
5 Music
6 Stationery, books
7 Jewelry
8 Office, store,
school supplies
04 Specialty— continued
9 Candy
10 Tobacco
11 Other
05 Drug Store
06 Household furnishings
07 Household appliances
08 Second-hand, antiques
09 Garden supplies
10 Eating and drinking place
1 No alcohol
2 Beer, wine
3 All alcohols
4 Bar
11 Service Station
12 Building materials
1 Lumber
2 Hardware
3 Plumbing, electrical
4 Paint, glass, wallpaper
13 Motor Vehicles
1 New
2 Used
3 Supplies, parts
14 Trailer, boat, cycle,
plane dir.
352
EXHIBIT 1— Continued
9. Do you provide parking? Yes 1
No 2
9.a. Individual 1
Shared 2
9.b. How many cars?_________________________ _________
10. What per cent of your business comes from
foot traffic? _________
11. From auto traffic?____________________________________
12. What per cent of your sales are from
30 day charges? _________
by installment? __________
for cash? _________
13. Do you live in the same neighborhood
as your store?
14. Do you live in Redondo Beach?
15. Do you belong to any civic or
community organizations?
16. What do you think are the good and bad aspects
of doing business in this shopping district?
Comments: ________________ _
Yes 1
No 2
Yes 1
No 2
None 0
Very poor 1
Poor 2
Average 3
Good 4
Very good 5
17. What do you think are the good and bad
features of doing business in the
community of Redondo Beach?
Comments:
Very poor 1
Poor 2
Average 3
Good 4
Very good 5
18. What per cent of your employees live
in the neighborhood?
in Redondo Beach?
outside Redondo Beach?
Where, if outside:
353
EXHIBIT 1— Continued
19. What per cent of your customers come from
the neighborhood?
Redondo Beach?
Outside Redondo Beach?
Where, if outside:_____________________
20. Have there been any retail failures
in this vicinity since 1950?
Number
21. Have you made any changes in your store
in the past?
How long ago?
21.a. What were they?_________________
22. Do you plan any improvements in the
near future?
22.a. What? ________________________
Yes 1
No 2
Yes 1
No 2
23. In what way has this shopping district
changed during the past five years?
Comments:
Rating: Deteriorated 1
Unchanged 2
Improved 3
24. In what respect do you think that this
shopping district will change during
the next few years?
Comments:
Rating: Will deteriorate 1
Will not change 2
Will improve 3
25. What advertising media
do you use?_____________________________________ _
354
EXHIBIT 2
REDONDO BEACH HOME SURVEY QUESTIONNAIRE
Hello, we are conducting a planning survey for the City
of Redondo Beach and would like to ask your help in
answering a few questions about shopping and related
matters.
Interviewer______ Tract No.______
Block No.______
House No.______
1. How many people are in this household? ___________
2. Are there any members of your family
who have an independent income and
do their own shopping for clothes,
cars, etc. If so, how many? ___________
3. Interviewer: Note number of persons
in spending unit
(answer to Q.l minus answer to Q.2)________________
All subsequent questions refer to spending
unit except for furniture, appliances,
supermarkets.
4. Age of head of household
(to nearest 10 years— do not write 0;
9 = DK/R). _________
5. Household composition (spending unit)
Married/no children.... 1
Married/pre-school children only.....2
Married/children up to 18.....3
Married/children over 18 only.... 4
Married/adult children who have left.....5
Single person.....6
Multiple single persons.....7
6. How long s ince you moved to
Redondo Beach? Less than 6 months.....1
1-2 years.....2
2-5 years.....3
5-10 years.... 4
10 years or more.....5
DK/R.... 9
7.
8.
9.
10.
11.
12.
13.
14.
15.
355
EJXHIBIT 2— Continued
How many persons (in the spending
unit) are regularly employed? _________
If none: Retired....8
Unemployed....9
Q.8 Q.9
In what city does the principal
earner work?
In what city does the other wage
earner (in the spending unit) work?
Redondo Beach 1 1
Listed cities 2 2
Other
Not appropriate 4 4
DK/R 9 9
What is the principal wage
earner's occupation?
In what kind of business?
How many years of schooling has the
head of the household completed?
Grammar 1-8 years.... . . 1
High 9-12 years.... 2
College 13-16 years..., . 3
Graduate 17 or more years.... . 4
DK/R___ 9
Do you own or rent your
home? Own... , . 1
Rent... . . 2
DK/R... . , 9
What is your monthly rent
(or mortgage payment)? Less than $75... . , 1
$75 to $99... 2
$100 to $124... 3
$125 to $149... ,4
$150 and over... , 5
No mortgage... a
DK/R... 9
How many cars in your family?
How many cars have you bought
in the last two years?
Interviewer: If none, circle code 8
in Questions 16-2 0 and skip to
Question 21.
356
EXHIBIT 2— Continued
16. Was it new or used? New only 1
Used only.....2
New and used.....3
Not appropriate.....8
DK/R.....9
17.-18. 1st car 2nd car 1st and 2nd car
Make, model, year _______ _______
Purchase price_____________ _______
Trade-in price_____________ _______ ______ ______
(or make, model)
If none, write none. Not appropriate 8 8
19.-20. Q.19 Q.20
In what city did you purchase
the car? (Write in and code)
1st car ___________ Redondo Beach 1 1
2nd car ___________ Listed cities 2 2
Other 3 3
Not appropriate 8 8
DK/R 9 9
21. About what are your monthly
expenditures for operating and
maintaining your car(s)? I
mean things like gas, oil,
lubrication, tuneups, and
minor repairs? Approximate Cost
Under $25.......1
$25 to $49........2
$50 to $74........3
$75 to $100....... 4
Over $100........5
Not appropriate........8
DK/R........9
22. What proportion of this is
spent in Redondo Beach? Per cent
0-10.........1
11-30 2
31-50.........3
51-70.........4
81-100.........5
Not appropriate.........8
DK/R.........9
357
EXHIBIT 2— Continued
23. Did you purchase any household appliances/ such as
a refrigerator, a washing machine, a television
or hi-fi set, and so forth last year (1962)?
Where Total
Item purchased price
Interviewer note:
Please total prices of
all items_________________ $__________ _______
24. Can you think of any items of household furnishings,
such as sofas, chairs, tables, bedroom suites,
etc., or carpeting and draperies that you
purchased last year (1962)?
Where Total
Item purchased price
Interviewer note:
Please total prices of
all items $__________ _
Interviewer:
Show card for Questions 25, 27 and 29.
25. How much was spent on men's clothing
in 1962? (Write in code number) _______
Not appropriate.......8
DK/R.......9
26. What proportion was spent in
Redondo Beach?........................None...... 1
1/4.......2
Most important other...................1/2...... 3
city........................ 3/4.......4
All.......5
Not appropriate.......8
DK/R.......9
27. How much was spent on women's
clothing last year?
(Write in code number)____________________ _______
Not appropriate.......8
r -
358
EXHIBIT 2— Continued
28. What proportion was spent in
Redondo Beach? None.......1
1/4.......2
Most important other 1/2...... 3
city......................... 3/4.......4
All.......5
Not appropriate.......8
DK/R.......9
29. How much was spent on children's
clothing? (Write in code number)_________________
Not appropriate.......8
DK/R.......9
30. What proportion was spent in
Redondo Beach? None...... 1
1/4.......2
Most important other 1/2...... 3
city_________________________ 3/4.......4
All.......5
Not appropriate.......8
DK/R.-......9
31. How much do you (entire household)
spend weekly at the supermarket? Under $10....1
$10 to $15___2
$16 to $20___3
$21 to $25___4
$26 to $30___5
$30 to $40___6
Over $40.... 7
DK/R 9
32. At what market do you do most
of your shopping?
(Name) (Location)
(If no one market is dominant, code 9 and ask) 9
33. How much of this is spent in
Redondo Beach? None.......1
1/4.......2
1/2.......3
3/4.......4
All.......5
Named specific store.......8
DK/R.......9
359
EXHIBIT 2— Continued
34. Where did you live before you
moved to Redondo Beach?
(Write in and code) Adjacent cities.
___________________ Listed city.
Other Los Angeles County.
Other Southern California.
Outside Southern California.
DK/R.
35. Why did you choose Redondo Beach
instead of some other area in
Southern California?
How often do you visit
(times/month; 0 = less than once; 9 = 9 or more)
36. South Bay Center (1)............................
37. Artesia Blvd.
between Inglewood and Aviation (2)..........
38. Downtown Redondo Beach
(old Redondo Beach) and
Triangle Shopping Center (3)............. ...
39. Hollywood Riviera
(South Redondo Beach, below Ave. "1" (4)..___
40. Note number (1-4) of area visited
most frequently. Probe if tied. ___
41. What features of Redondo Beach
shopping do you dislike?
42. What features do you particularly
like?
43. In general, would you rate
Redondo Beach shopping as: Excellent.
Very good.
Good.
Fair.
Poor.
Very bad.
DK/R. V O UI ^ OJ t o H v O ui 4 ^ u> t o i —1
EXHIBIT 2— Continued
360
44. What do you particularly like about living
in Redondo Beach?
45. What do you dislike about living
in Redondo Beach?
46. In general, would you say that
you and your family like
living in Redondo Beach? Like......1
Neutral......2
Dislike......3
Dfc/R..... 9
47. Has your opinion of Redondo Beach
changed in the last year or so.
Do you like living here more
or less than before? Like it more......1
Like it less......2
No change......3
Not applicable......8
DK/R......9
48. If more or less; why?
No change......8
FOR SPENDING UNIT ONLY
49. In order to help us get some idea of
the total income in Redondo Beach,
could you give us an estimate of
your total family income for 1962?
(Show card)
1 2
7 8
50. How much of this total family
income was earned outside
of Redondo Beach?
3
9
4
10
5
11
DK/R
6
12
00
None... . . .1
Up to 25 per cent... 2
26 to 50 per cent...... 3
50 to 75 per cent.... . 4
Over 75 per cent... 5
All.... . .6
DK/R...
361
EXHIBIT 2— Continued
51. Could you tell us about how much
you saved in 1962. This includes
savings accounts with banks,
savings and loan associations,
or credit unions; or into government
bonds or corporate stocks and
bonds? (Show card)
1 2 3 4 5 6
7 8 9 10 11 12
DK/R 13
None 14
FOR ENTIRE HOUSEHOLD
52. What was the total income of your
family including independent
wage earners?
1 2 3 4 5 6
7 8 9 10 11 12
DK/R 13
Not appropriate 14
INTERVIEWER NOTE:
53. Type and sex of respondent.
Male head of household......1
Female head of household......2
Housewife......3
Male single person......4
Female single person......5
Other male..... 6
Other female......7
54. Estimated age of
respondent 18 to 30.......1
30 to 45.......2
45 to 60.......3
Over 60...... 4
55. Type of dwelling:
Single family......1
Apartment up to 10 units......2
Apartment over 10 units......3
Address
(Include apartment number)
BIBLIOGRAPHY
BIBLIOGRAPHY
Books
Adams, James W. R. Modern Town and Country Planning.
Londom J. & A. Churchill Ltd., 1952.
Alonso, William. Location and Land Use. Cambridge,
Mass.: Harvard University Press, 1964.
A.P.R.R. (ed.) Town and Country Planning Textbook.
London: Architectural Press, 1950.
Aronovici, Carol. Community Building. New York:
Doubleday & Company, 1956.
Aubert-Krier, Jane. "Monopolistic and Imperfect
Competition in Retail Trade", Monopoly and
Competition and their Regulation.
(ed.) Edward H. Chamberlin. London:
MacMillan & Co. Ltd., 1954.
Baker, Geoffrey,
and
Funaro, Bruno. Shopping Centers— Design and Operation.
New York: Reinhold Publishing Corp., 1951.
Beil, Preston J. Variety Store Retailing.
New York: Variety Store Merchandiser
Publications, 1956.
Breese, Gerald William, (ed.) An Approach to Urban
Planning. Princeton, N.J.: Princeton
University Press, 1953.
Burns, Wilfred. British Shopping Centres.
London: Leonard Hill Ltd., 1959.
Chapin, F. Stuart, Jr. Urban Land Use Planning.
New York: Harper & Brothers, 1957.
and
Weiss, Shirley F. Urban Growth Dynamics.
New York: John Wiley & Sons, Inc., 1962.
363
BIBLIOGRAPHY— Continued
Converse, Paul D.,
Huegy, Harvey W.,
and
Mitchell, Robert V. Elements of Marketing.
Englewood Cliffs, N.J.: Prentice-Hall, Inc.,
1958.
Delafons, John. Land-Use Controls in the United States.
Cambridge, Mass.: Joint Center for Urban
Studies of M.I.T. and Harvard University, 1962.
Eckstein, Otto. Water-Resource Development.
Cambridge, Mass.: Harvard University Press,
1958.
Entenberg, Robert D. The Changing Competitive Position
of Department Stores in the United States by
Merchandise Lines. Pittsburgh: University of
Pittsburgh Press, 1957.
Greenhut, Melvin L. Microeconomics and the Space
Economy. Palo Alto, Calif.: Scott Foresman
and Company, 1963.
Gruen, Victor,
and
Smith, Larry. Shopping Towns U. S. A. New York:
Reinhold Publishing Corp., 1960.
Haar, Charles M. Land Use Planning. Boston: Little
Brown & Company, 1959.
Haig, Robert M. Regional Survey of New York and Its
Environs. New York: New York Regional
Planning Committee, 1927.
Holdren, Bob R. The Structure of a Retail Market and
the Market Behavior of Retail Units.
Englewood Cliffs, N.J.: Prentice-Hall, Inc.,
1960.
Hoyt, Homer. The Structure and Growth of Residential
Neighborhoods in American Cities. Washington,
D.C.: Federal Housing Administration, 1939.
364
BIBLIOGRAPHY— Continued
Isard, Walter. Methods of Regional Analysis.
Cambridge, Mass.: The M.I.T. Press, 1960.
________, Location and Space-Economy. Cambridge,
Mass.: The M.I.T. Press, 1957.
Jefferys, James B. Retail Trading in Britain 1850-1950.
Cambridge: Cambridge University Press, 1954.
and
Knee, Derek. Retailing in Europe. London: MacMillan
& Co., Ltd., 1962.
Kapp, K. W. The Social Cost of Private Enterprise.
Cambridge, Mass.: Harvard University Press,
1950.
Kelley, Eugene J. Shopping Centers— Locating Controlled
Regional Centers. New York: Eno Foundation
for Highway Traffic Control, 1956.
Lichfield, Nathaniel. The Economics of Planned
Development. London: Estates Gazette, 1956.
Loesch, August. The Economics of Location. New Haven,
Conn.: Yale University Press, 1954.
Margolis, Julius,
and
Lichfield, Nathaniel. "Benefit-Cost Analysis as a
Tool in Urban Government Decision Making",
Public Expenditure Decisions in the Urban
Economy. (ed.) Howard G. Schaller.
Baltimore: Johns Hopkins Press, 1963.
McKean, Roland N. Efficiency in Government through
Systerns Analysis. New York: John Wiley and
Sons, Inc., 1959.
Nelson, Richard L. The Selection of Retail Locations.
New York: F. W. Dodge Corp., 1958.
r
365
BIBLIOGRAPHY— Continued
Nussbaum, P. L. A History of the Economic Institutions
of Modern Europe. New York: F. S. Crofts &
Company, 1933.
Pfouts, Ralph W. (ed.). The Techniques of Urban
Economic Analysis. West Trenton, N.J.:
Chandler-Davis Publishing Co., 1960.
Reilly, W. J. The Law of Retail Gravitation.
New York: W. J. Reilly Co., 1931.
Simmons, James. The Changing Pattern of Retail
Location. Chicago: University of Chicago,
1963.
Sternlieb, George. The Future of the Downtown
Department Store. Cambridge, Mass.:
Harvard University Press, 1962.
Walker, Mabel. Business Enterprise and the City.
Princeton, N.J.: Tax Institute, Inc., 1958.
We imer, Ar thur,
and
Hoyt, Homer. Principles of Real Estate. New York:
The Roland Press, 1960.
Wharton School of Finance and Commerce
and
United States Department of Labor. Study of Consumer
Expenditures, Incomes, and Savings.
Volumes I-XVIII. Philadelphia: University
of Pennsylvania Press, 1956.
Wheaton, William L. C. "Public and Private Agents
of Change in Urban Expansion", Explorations
into Urban Structure. Philadelphia:
University of Pennsylvania Press, 1964.
Wilhelm, Sidney M. Urban Zoning and Land-Use Theory.
Glencoe, 111.: The Free Press of Glencoe,
1962.
r
366
BIBLIOGRAPHY— Continued
Wingo, Lowden, Jr. Cities and Space. Baltimore:
Johns Hopkins Press, 1963.
Woodbury, Coleman. "Land Economics Research for Urban
and Regional Planning", Land Economics Research,
(ed.) Joseph Ackerman, Marion Clawson, and
Marshall Harris.
Journal Articles and Periodicals
Adamson, W. M. "Measurement of Income in Small
Geographic Areas", Southern Economic Journal.
VIII, Number 4 (April, 1942), 479-492.
Andrews, Richard B. "Comment re Criticisms of the
Economic Base Theory", Journal of the American
Institute of Planners. XXIV, Number 1
(Spring, 1958), 37-40.
"Economic Planning for Small Areas: An
Analytical System", Land Economics.
XXIX, Number 1 (May, 1963), 143-155.
"Urban Economics: An Appraisal of Progress",
Land Economics. XXVII, Number 3 (August, 1961),
219-227.
Applebaum, William,
and
Spears, Richard F. "How to Measure a Trading Area",
Chain Store Age. (January, 1951), 36-41.
"A Technique for Constructing a Population
and Urban Land-Use Map", Economic Geography.
XXVIII, Number 3 (July, 1952), 240-243.
and
Cohen, Saul B. "Store Trading Areas in a Changing
Market— Part I", Journal of Retailing.
XXV, Number 3 (Fall, 1961), 14-25.
BIBLIOGRAPHY— Continued
367
l
and
________ . "Store Trading Areas in a Changing Market—
Part II", Journal of Retailing. XXV,
Number 4 (Winter, 1961-1962), 32-43,
Artie, Roland. "On some Methods and Problems in the
Study of Metropolitan Economies", Regional
Science Association Papers. VIII, (1962),
71-87.
Aurousseau, M. "The Distribution of Population,
a Constructive Problem", Geographical Review.
2 (1921), 572-581.
Bator, F. M. "The Anatomy of Market Failure",
Quarterly Journal of Economics. XXXVI,
Number 3 (August, 1958), 351-379
Berry, Brian J. L. "The Retail Component of the Urban
Model", Journal of the American Institute of
Planners. XXXI, Number 2 (May, 1965), 150-155.
Bloom, Max R. "Fiscal Productivity and the Pure
Theory of Urban Renewal", Land Economics.
XXX, Number 2 (May, 1962), 134-144.
Bushnell, John D. "Regional Shopping Centers:
Experience in Los Angeles County", Journal
of the American Institute of Planners.
XIV, Number 1 (Fall, 1948), 10-13.
Carrothers, Gerald A. P. "An Historical Review of
Gravity and Potential Models of Human
Interaction", Journal of the American
Institute of Planners. XXII, Number 2
(May, 1956), 94-102.
Converse, P. D. "New Laws of Retail Gravitation",
Journal of Marketing. XIV, Number 3
(November, 1949), 379-384.
368
BIBLIOGRAPHY— Continued
Crosson, Pierre R. "Further Comment on Economic Base
Theory", Land Economics. XXVI, Number 2
(May, 1960), 197-201.
Cumberland, John H. "Suggested Improvements in
Regional Income Accounting", Regional Science
Association Papers. II, (1956), 259-271.
Daly, M. C. "An Approximation to a Geographical
Multiplier", Economic Journal. XXXXX,
Number 2 (June-September, 1940), 248-258.
Davis, Otto A. "A Pure Theory of Urban Renewal",
Land Economics. XXVI, Number 2 (May, I960),
220-226.
Dorfman, Robert. "The Nature and Significance of
Input-Output", Review of Economics and
Statistics. XXXII, Number 2 (May, 1954),
121-133.
Dyckman, John W. "Planning and Decision Theory",
Journal of the American Institute of Planners.
XXVII, Number 4 (November, 1961), 335-343.
Elwood, Leon W., "Estimating Potential Volume of
Proposed Shopping Centers", The Appraisal
Journal. XVIII, Number 5 (October, 1954),
581-589.
Garrison, William L. "Estimates of the Parameters of
Spatial Interaction", Regional Science Associa
tion Papers. II, (1956), 280-290.
Gottlieb, Abe. "Planning Elements of an Interindustry
Analysis", Journal of the American Institute
of Planners. XXII, Number 3 (Fall, 1956),
230-236.
Greenhut, Melvin L. "Comments on Economic Base Theory",
Land Economics. XXV, Number 1 (February, 1959),
71-75.
369
BIBLIOGRAPHY— Continued
Hall, Bryant. "How much Land for Commerce?",
Journal of the American Institute of Planners.
IX, Number 1 (January-February, 1942), 3-11.
Hansen, W. Lee,
and
Tiebout, Charles M. "An Intersectoral Flows Analysis
of the California Economy", Review of
Economics and Statistics. XXXXI, Number 4
(November, 1963), 409-418.
Harris, Britton. "Projecting Industrial Growth
of Metropolitan Regions", Regional Science
Association Papers. II, (1956), 239-249.
________. "Plan or Projection: An Examination of
the Use of Models in Planning", Journal of
the American Institute of Planners. XXVI,
Number 4 (November , I960), 265-272.
________ . "Comments on Pfout's Test of the Base
Theory", Journal of the American Institute
of Planners. XXIV, Number 4 (November, 1958),
328-331.
Higgens, Benjamin. "Towards a Science of Community
Planning", Journal of the American Institute
of Planners. XV, Number 1 (Fall, 1949), 3-13.
Hildebrand, George,
and
Mace, Arthur, Jr. "The Employment Multiplier in an
Expanding Industrial Market: Los Angeles
County, 1940-1947", Review of Economics and
Statistics. XXVIII, Number 3 (August, 1950),
241-249.
Hoover, Robert G. "On Master Plans and Constitutions",
Journal of the American Institute of Planners.
XXVI, Number 1 (February, 1960), 5-24.
Horwood, Edgar M. "Public Policy and the Outlying
Shopping Center", Journal of the American
Institute of Planners. XXIV, Number 3
(Fall, 1958), 215-222.
370
BIBLIOGRAPHY— Continued
Howard, J. T. "Basic Research Problems of the Urban-
Metropolitan Region: Problems Related to
Planning", Regional Science Association Papers.
II (1956), 103-118.
Hoyt, Homer. "Economic Background of Cities", The
Journal of Land and Public Utility Economics.
VI, Number 2 (May, 1941), 188-195.
________. "A Method for Measuring the Value of Imports
into an Urban Community", Land Economics.
XXVI, Number 2 (May, 1961), 150-161..
________. "Homer Hoyt and the Dilemma of Urban
Economic Base Analysis: A Rejoiner",
Land Economics. XXVII, Number 1 (February,
1962), 70.
________. "The Utility of the Economic Base Method
in Calculating Urban Growth", Land Economics.
XXVI, Number 1 (February 1961), 51-58.
Huff, David L. "A Probabilistic Analysis of Shopping
Center Trade Areas", Land Economics. XXVIII,
Number 1 (February, 1963), 81-90.
________. "A Note on the Limitations of Intra-Urban
Gravity Models", Land Economics. XXVI,
Number 1 (February, 1961), 64-66.
Kurtz, Maxine. "Denver Economic Survey", Journal of
The American Institute of Planners. XIX,
Number 2 (Summer, 1954), 62-77.
Lakshmanan, T. R.,
and
Hansen, Walter G. "A Retail Market Potential Model",
Journal of the American Institute of Planners.
XXXI, Number 2 (May, 1965), 134-143.
Leven, Charles L. "An Appropriate Unit for Measuring
the Urban Economic Base", Land Economics.
XX, Number 4 (November, 1954), 327-336.
r
371
BIBLIOGRAPHY— Continued
________ . "Measuring the Economic Base", Regional
Science Association Papers. II (1956),
250-258.
________. "A Theory of Regional Social Accounting",
Regional Science Association Papers. IV
(1958), 221-237.
Lichfield, Nathaniel. "Cost-Benefit Analysis in City
Planning", Journal of the American Institute
of Planners. XXVI, Number 4 (November, 1960),
273-279.
________. "A Pure Theory of Urban Renewal: A
Further Comment", Land Economics. XXIX,
Number 1 (February, 1963), 99-103.
Lillibridge, Robert M. "Frontiers in Metropolitan
Planning and Land Policy",^Land Economics.
XV, Number 1 (February, 1950), 40-51.
\
Loesch, August. "The Nature of Economic Regions”,
Southern Economic Journal. IV, Number 2
(July, 1938), 71-78.
Lovelace, Eldridge. "Urban Land Use— 1949", Journal
of the American Institute of Planners. XV,
Number 1 (Summer, 1949), 26-32.
Mangiamele, J. F. "Economic and Physical Development:
A Proposal for Local Coordination", Land
Economics. XXIX, Number 4 (November, 1964),
437-440.
Matilla, J. M.,
and
Thompson, W. R. "The Measurement of the Economic Base
of the Metropolitan Area", Land Economics.
XX, Number 3 (August, 1955), 215-228.
Mayer, Harold M. "Urban Nodality and the Economic
Base", Journal of the American Institute of
Planners. XX, Number 3 (Summer, 1954), 117-121.
372
BIBLIOGRAPHY— Continued
Morrissett, Irving. "The Economic Structure of
American Cities", Regional Science Association
Papers. IV, (1958), 239-256.
Moyerman, Sue S.
and
Harris, Britton. "The Economics of the Base Study",
Journal of the American Institute of Planners.
XXI, Number 1 (Spring-Summer, 1955), 88-93.
Murdock, John C. "Homer Hoyt and the Dilemma of Urban
Economic Base Analysis: A Reply", Land
Economics. XXVIII, Number 1 (February, 1962),
66-69.
McGovern, P. D. "Identifying Exporting Industries",
Journal of the American Institute of Planners.
XXVII, Number 2 (May, 1961), 144-150.
North, Douglas C. "Location Theory and Regional
Economic Growth", Journal of Political Economy.
LXXI, Number 3 (June, 1955), 243-258.
"Oskaloosa versus the United States", Fortune.
(April, 1938), 55-62.
Peterson, Donald A. "Market Areas of Shopping
Districts", Journal of the American Institute
of Planners. XXIX, Number 4 (November, 1963),
297-301.
Pfouts, Ralph W. "Reply to Harris on Testing the Base
Theory", Journal of the American Institute of
Planners. XXIV, Number 4 (November, 1958),
326-329.
Roterus, Victor,
and
Calef, Wesley. "Notes on the Basic-Nonbasic Employment
Ratio", Economic Geography. XXXI, Number 1
(January, 1955), 17-20.
373
BIBLIOGRAPHY— Continued
Sasaki, Kyohei. "Military Expenditures and the
Employment Multiplier in Hawaii", Review of
Economics and Statistics. XXXXI, Number 3
(August, 1963), 298-304.
Schaaf, A. H. "Public Policies in Urban Renewal:
An Economic Analysis of Justifications and
Effects", Land Economics. XXX, Number 1
(February, 1964), 67-78.
Schussheim, Morton J. "A Pure Theory of Urban Renewal:
A Comment", Land Economics. XXVI, Number 4
(November, 1960), 395-396.
Sigafoos, Robert A. "The City— A Creator of Wealth",
Stanford Research Institute Journal. IV,
Number 4 (Fourth Quarter, 1960), 129-137.
Silverman, Charles E. "The Department Stores are
Waking Up", Fortune. July, 1962, 129-136.
i
Sirikin,', G. "The Theory of the Regional Economic Base",
Review of Economics and Statistics. XXXXI,
Number 1 (February, 1959), 61-67.
Stedman, Gordon H. "The Rise of Shopping Centers",
Journal of Retailing. XIV, Number 1
(Spring, 1955), 11-26.
Stevens, William H. "The Social Responsibilities of
Marketing", Proceedings of the American
Marketing Association. (1961), 567-575.
Stewart, Charles T., Jr. "Economic Base Dynamics",
Land Economics. XXV, Number 4 (November, 1959),
327-336.
Strohkark, Frank. "The Mechanics of Constructing a
Market Area Map", Journal of Marketing.
XXII, Number 3 (April, 1948), 493-496.
Thomas, Morgan D. "Economic Activity in Small Areas",
Land Economics. XXVI, Number 2 (May, 1960),
164-171.
374
BIBLIOGRAPHY— Continued
________. "The Economic Base and a Region's Economy",
Journal of the American Institute of Planners.
XXIII, Number 2 (May, 1957), 86-92.
Thompson, Gerald E. "An Investigation of the Local
Employment Multiplier", Review of Economics
and Statistics. XXXXI, Number 1
(February, 1959), 61-67.
Tiebout, C. M. "A Method of Determining Incomes and
their Variations in Small Regions", Regional
Science Association Papers. I (1955), F1-F12.
________. "Exports and Regional Economic Growth",
Journal of Political Economy. LXIII,
Number 2 (April, 1956), 160-169.
"Input-Output and Foreign Trade Multiplier
Models in Urban Research", Journal of the
American Institute of Planners. XXIII,
Number 3 (August, 1957), 126-130.
Ullman, Edward L.
and
Dacey, Michael F. "The Minimum Requirements Approach
to the Urban Economic Base", Regional Science
Association Papers. VI (1960), 175-194.
Voorhees, Alan M. (ed.) "Land Use and Traffic Models:
A Progress Report", Journal of the American
Institute of Planners. Special Issue,
XXV, Number 2 (May, 1959), 55-108.
Reports and Pamphlets
Applebaum, William. Store Location and Development
Studies. Worcester, Mass.: Clark University,
1962.
Berry, Brian J. L. Commercial Structure and
Commercial Blight. Chicago: University of
Chicago, 1963.
375
BIBLIOGRAPHY— Continued
Brown, Milton P.,
and
May, Eleanor G. Operating Results of Multi-Unit
Department Stores. Harvard University
Graduate School of Business Administration,
Bureau of Business Research, Bulletin Number
159, 1961.
Chapin, F. Stuart,
and
Weiss, Shirley F. Factors Influencing Land Development,
Institute for Research in Social Science
Monograph. Chapel Hills University of
North Carolina, 1962.
Controller's Congress. Merchandising and Operating
Results of Department and Specialty Stores.
New York: National Retail Merchants Associa
tion, 1950, 1952, 1958, I960, and 1964.
Converse, P. D. "Retail Trade Areas in Illinois",
University of Illinois Bulletin. Business
Studies Number 4, XXXXIII (1946), 68-74.
Fine, Isadore V. Retail and Trade Area Analysis.
University of Wisconsin Commerce Papers,
I, 6. Madison: University of Wisconsin
Press, January, 1954.
Hoyt, Homer. A Re-Examination of the Shopping Center
Market. Urban Land Institute, Technical
Bulletin Number 33. Washington, D. C., 1958.
Huff, David L. Determination of Intra-Urban Retail
Trade Areas. Los Angeles: U.C.L.A. Real
Estate Research Program, 1962.
Kopp, William T.,
and
Salomone, Louis, Jr. The Market Potential for a
Regional Shopping Center in the City of
Norwalk. South Pasadena, Calif.: Southern
California Labs of Stanford Research Institute,
November, 1962.
376
BIBLIOGRAPHY— Continued
McKeever, J. Ross. Shopping Centers: Principles
and Policies. Urban Land Institute,
Technical Bulletin Number 20. Washington,
D. C.f 1953.
________. Shopping Centers Re-Studied; Emerging
Patterns and Practical Experiences. Urban
Land Institute, Technical Bulletin Number 30.
Washington, D. C., 1957.
McNair, Malcolm P. Operating Results of Department
and Specialty Stores. Harvard University
Graduate School of Business Administration,
Business Research Bulletins Numbers 159,
160, 161, 162, 163, 164, 165, and 166.
Soldiers Field, Boston: Harvard University,
1957, 1958, 1959, I960, 1961, 1962, 1963,
and 1964.
f
________. The American Department Store 1920-1960.
Soldiers Field, Boston: Harvard University,
1963.
Palmer, Edgar Z. (ed.) The Community Economic Base
and Multiplier. University of Nebraska
Business Research Bulletin Number 63.
Lincoln: University of Nebraska Press, 1958.
Politz, Alfred. Life Study of Consumer Expenditures.
Volumes I-VII. New York: Time, Inc., 1960.
Research and Policy Committee. Guiding Metropolitan
Growth. New York: Committee for Economic
Development, August, 1960.
Sewell, W. R. D.,
and
others. Guide to Cost-Benefit Analysis. Ottawa:
Queen's Printer and Controller of Stationery,
1962.
Smith, Edward K. A Guide to Economic Base Studies
for Local Communities. Boston: Northeastern
University, 1955.
r 377
BIBLIOGRAPHY— Continued
Suburban Business Centers. Cleveland:
Cuyahoga County Regional Planning Commission.
Third Report on Land Use in Cuyahoga County,
1959.
The Dollars and Cents of Shopping Centers,
Part I: The Entire Center. Washington, D. C.:
Urban Land Institute, 1961.
The Dollars and Cents of Shopping Centers,
Part II: Tenant Characteristics. Washington,
D. C.: Urban Land Institute, 1962.
The Dollars and Cents of Shopping Centers.
Washington, D. C.: Urban Land Institute, 1963.
Tiebout, C. M. The Community Economic Base Study.
New York: Committee for Economic Development,
Supplementary Paper Number 16, 1962.
Public Documents
California, State Board of Equalization. Trade Outlets
and Taxable Retail Sales in California.
Annual Reports for 1960 and 1961, and
Quarterly Reports for 1957, 1958, and 1959.
Capitol Region Planning Board of British Columbia.
Population Forecast and Land Use Requirements.
Victoria, B.C.: Capitol Region Planning
Board, May, 1958.
Federal Housing Administration. Underwriting Manual.
Washington, D. C.: Government Printing
Office, 1938.
Federal Reserve Bank of Kansas City. "The Employment
Multiplier in Witchita", Monthly Review.
XXXVII (September, 1952).
Los Angeles County, Regional Planning Commission.
Population and Dwelling Units. Los Angeles:
Quarterly.
378
BIBLIOGRAPHY— Continued
U. S., Bureau of the Census. Census of Business,
Retail Trade, California: 1954, 1958, and
1963. Washington, D. C.: Government
Printing Office, 1965.
Unpublished Material
and
Other Sources
Bond, Harry Clay. "The Economic Base Survey as
Related to the Community's Master Plan".
Unpublished Master's thesis, The University
of Southern California, Los Angeles,
June, 1959.
Friedly, Philip Howard. "An Economic Base and Income
Multiplier Study of Redondo Beach, California"
Unpublished Master's thesis. The University
of Southern California, Los Angeles,
June, 1964.
Planning Research Corporation. Family field survey
conducted by the writer and others in
Redondo Beach, California, January, 1963.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
An economic base and income multiplier study of Redondo Beach, California
PDF
An Economic Analysis Of The Changes In Retail Store Sales And Location Inthe Central Business Districts Of One Hundred And Nine Central Cities In The United States, 1948-1958
PDF
The Theory Of Investment Programming: A Suggested Econometric Model
PDF
The Optimization Of Economic Resources For Economic Growth In Iran
PDF
Employment Distribution, Income, And City Size: A Statistical Analysis
PDF
An Analysis Of The Feasibility Of The Use Of Fiscal Policy To Stimulate Economic Growth In Iran
PDF
An International Comparison Of The Role Of Government In The Economic Development Of Developed And Emerging Economies, With Particular Reference To Government Corporations
PDF
New Port Cities In Western America And Their Effect On Economic Development
PDF
A Multidimensional Interpretation Of Comprehensive Developmental Planning
PDF
An Analysis Of Competition In The Title Insurance Industry
PDF
Effects Of Swedish Fiscal Policy And Planning On Economic Growth And Stability, 1930-1964
PDF
Planning Institutional Change For Economic Growth: A Study Of The Effect Of Public Institutions On The Rise Of Indigenous Entrepreneurs In East Africa
PDF
An Economic Analysis Of The Factor Market For Pharmacists
PDF
A Solution To The Problem Of Food Shortages In Korea
PDF
Institutional Mortgage Lending In The Los Angeles Metropolitan Area, 1953-1954 And 1957-1958
PDF
Some External Diseconomies Of Urban Growth And Crowding: Los Angeles
PDF
Private Foreign Investment As A Possible Aid For The Economic Growth Of Iraq
PDF
Developmental Planning In The United Arab Republic: An Economic Analysis And Evaluation Of The Egyptian Experience
PDF
A Case Study Of The Adaptation Of A Small National Economy'S Industry To International Competition: Austria
PDF
An Analysis Of The Teaching Of Certain Economic Topics In The California Public Junior Colleges
Asset Metadata
Creator
Friedly, Philip Howard
(author)
Core Title
Models And Determinants For The Economic Allocation Of Retail Space In Urban Planning
Degree
Doctor of Philosophy
Degree Program
Economics
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Economics, General,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Elliott, John E. (
committee chair
), Anderson, William H. (
committee member
), Grey, Arthur Leslie, Jr (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-191394
Unique identifier
UC11359668
Identifier
6605479.pdf (filename),usctheses-c18-191394 (legacy record id)
Legacy Identifier
6605479.pdf
Dmrecord
191394
Document Type
Dissertation
Rights
Friedly, Philip Howard
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA