University of Southern California Dissertations and Theses

A Money Supply Model: Jordan

(USC Thesis Other)

A Money Supply Model: Jordan

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 70- 25,049 MARTO, Michel Isa, 1940- A MONEY SUPPLY MODEL: JORDAN. University of Southern California, Ph.D., 1970 Economics, general University Microfilms, A XEROX Company, Ann Arbor, Michigan 'X' Copyright by MICHEL ISA MARTO 1970 Copyright by MICHEL ISA MARTO 1970 A MONEY SUPPLY MODEL JORDAN ky Michel Isa Marto 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) June 1970 UNIVERSITY O F S O U T H E R N CA LIFO R N IA TH E G RA DUATE SC H O O L U N IV ER SITY PA RK L O S A N G E LE S, C A L IFO R N IA 9 0 0 0 7 This dissertation, written by Michel Isa Marto under the direction of Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments of the degree of D O C T O R O F P H I L O S O P H Y f 'Tv Dean D aie DISSERTATION COMMITTEE / ACKNOWLEDGEMENT In the course of my research I have benefited greatly from the intellectual support of many persons. Special thanks must be extended to Professor Michael DePrano for the tireless effort and incisive assistance which have been of immeasurable importance to the success of this I study. I am also most grateful to Professor Jeffrey B. Nugent for his help and guidance during my four years at USC and Professor Edward S. Shaw of Stanford University for the suggestions which he contributed to this disser tation as well as the intellectual stimulation gained from his Seminar on Monetary Theory. I am appreciative of the help with computations provided by Mr. William MacReynolds• I would also like to acknowledge my debt to U.S.A.I.D. and to the Central Bank of Jordan for their financial support and cooperation which has made this study possible. Lastly I want to thank my parents for the understanding and confidence which benefited me in my work. ii TABLE OF CONTENTS CHAPTER PAGE X. INTRODUCTION ................................. 1 Purpose of the -Study . . ................. 1 Background of the S t u d y ............... . 3 Methods of Research.................. . 6 Organization of the Work................. 7 II. MONEY-SUPPLY DETERMINATION: THEORY AND APPROACHES .................... 9 Definition of M o n e y .........................10 Multiple Expansion ........................ 13 1. Phillips................................15 2. Shaw....................................17 3. Criticism............................. 21 Bank Adjustment and the Reserve Position . 23 Specification of Empirical Money- Supply Relationships .................... 26 1. Friedman-Schwartz.................... 27 2. C a g a n ..................................31 i ii CHAPTER PAGE 3. Brunner...............................32 4. Brunner-Meltzer...................... 38 5. Teigen.................................42 General Equilibrium ..................... 45 Balance of Payments.........................48 Summary.....................................50 III. ECONOMIC DEVELOPMENT AND THE EVOLUTION ! OF THE MONETARY AND BANKING SYSTEM | i j IN JORDAN, 1951-1965 ...................... 51 t i | The Jordan Currency Board ............... 51 j The Central Bank of Jordan............... 54 | The Commercial Banks ...................... 60 Money Supply................................ 63 Foreign Assets of Jordan ................. 65 Commercial Bank Deposits and Credit . . . 67 Balance of Payments........................70 j Economic Development .............. .... 77 i ! 1. Gross National Product 78 I 2. Behavior of Prices . . ...............80 CHAPTER PAGE IV. DETERMINANTS OF THE JORDANIAN MONEY SUPPLY................................ 82 The Commercial Banks and the Monetary System ....................... 83 1. Measuring the Money Supply and its Components......................83 2. The Commercial Banks and the Money Supply........................85 3. Reserve Requirements and Liquidity Ratio ................. 88 The Monetary Authorities and the Monetary System ........................ 90 1. The Central Bank of Jordan........... 90 2. The Treasury.......................... 92 The Base of the Monetary System........... 94 Consolidated Analysis of the Money Supply.............................. 97 1. The Consolidated Monetary Balance Sheet ................... 97 2. Factors Determining Commercial Bank Reserves.....................100 v CHAPTER PAGE 3. Summary...................... 102 V. BALANCE OP PAYMENTS, HIGH-POWERED MONEY, AND THE MONEY S T O C K ....................... 104 The Balance of Payments and the Money S t o c k ............. 107 1. The Average and Marginal Propen sities to Import................... 107 2. Foreign Assets and Their Composition ....................... 109 3. The Currency Ratio....................112 4. Relationship Between the Money Supply and the Factors That Determine I t ....................... 113 High-Powered Money ........................ 121 1. Factors Affecting the Money Supply . 121 a. High-powered Money...............121 b. Deposits-to-Reserves Ratio . . . 122 c. Deposits-to-Currency Ratio . . . 124 2. Construction of the Algebraic Relationship ...................... 127 W ± r CHAPTER PAGE 3. Effects of Changes in High-powered Money, Deposits-to-Reserves Ratio, and Deposits-to-Currency Ratio on the Money Supply...............128 VI. A MONEY-SUPPLY MODEL FOR JO R D A N............ 134 Preliminary Model of Money and Reserves . 136 The Complete Model..........................141 1. The Endogenous variables..........141 2. The Exogenous variables..........142 3. The Model with Income as an Exogenous Variable ............... 142 a. Alternative Hypotheses ......... 145 b. Reduced-form Solutions ......... 147 4. The Model with Income as an Endogenous Variable ....... 150 a. The Real Sphere Hypothesis . . . 150 b. The Reduced-form Solutions . . . 151 5. Implications of Alternative Hypotheses for Testing .......... 153 vii CHAPTER PAGE VII. TESTS AND ESTIMATES....................... 155 Test Methods and Criteria.............. 155 1. Single-equation Models ............ 155 2. Error T e r m s ..................... 157 3. Two-Stage Least-Squares Models . . . 160 4. Economic Models and Problems of Specification............... 160 5. Statistical Criteria ............... 161 6. Data Employed ............164 Test Results............................. 165 1. The Import Hypothesis...........166 2. The Currency-demand Hypothesis , . . 174 3. The Reserves-Demand Hypothesis . . . 180 4. The Demand-for-Time-Deposits Hypothesis......................184 VIII. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . 191 Summary................................. 191 Conclusions............................. 192 Recommendations .......................... 198 Recommendations for Further Research . . . 202 viii CHAPTER PAGE APPENDIX.............................................. 204 BIBLIOGRAPHY.................................... 205 LIST OF TABLES AND FIGURES TABLE PAGE 1 Composition of Jordanian Money Supply, 1951-1965 64 2 Foreign Assets of Jordan, 1951-1965 .... 66 3 Deposits and Commercial Bank Credit and Foreign Assets...................... 68 4 Balance of Trade........................... 71 5 Balance of Goods and Service Transactions . 72 6 International Transfer Payments ........... 74 7 The Financing of Jordan's Deficit ......... 76 8 Size and Composition of the Money Supply . . 83 9 Division of the Money Supply.............. 84 10 Consolidated Balance Sheet of Commercial Banks, December 31, 1966 ................. 87 11 Sectoral Distribution of Commercial-Bank Credit, December 31, 1966 89 12 The Balance Sheet of the Central Bank of Jordan, December 31, 1966 92 13 Treasury Balances, December 31, 1966 .... 93 " x- TABLE PAGE 14 Consolidated Monetary Balance Sheet December 31, 1966 99 15 Determinants of Commercial-Bank Reserves . . 101 16 Contribution of Proximate Determinants to Changes in the Money Stock in Jordan 1951-1966 132 17 Distribution of the Durban-Watson Test , . . 163 18 Structural Equations of Imports with Income Exogenous .......................... 168 IS Reduced-Form Solutions for Imports with Income Endogenous ........................ 170 20 Reduced-Form Solutions for Money Supply with Income Endogenous .................... 172 21 Reduced-Form Solutions for Income...........173 22 Reduced-Form Equations for C p ............... 175 23 Second-Stage Cp Estimates .................. 177 24 Second-Stage Rd Estimates .................. 181 25 Reduced-Form Solutions for R d ............... 183 26 Ordinary-Least-Squares Estimates of Rd . . . 185 27 Second-Stage Tp Estimates................186 28 Ordinary-Least-Squares Estimates of Tp . . . 190 FIGURE I PAGE Flow of Jordanian Foreign-Exchange Surplus..................................... 96 xii CHAPTER I INTRODUCTION PURPOSE OF THE STUDY The objective of this dissertation is to formu late an empirical money supply function within the insti tutional framework of the Jordanian monetary and banking system. To achieve this, a money supply model is pre sented and alternative hypotheses are tested, using a "two-stage, least-squares" estimating procedure. The money supply function can best be specified by examining the behavioral assumptions used to explain interactions among the three economic units of the monetary sector: the monetary authorities (including the Treasury), the commercial banks, and the public. The money supply at any one moment reflects the portfolio decisions of these three economic units and their inter action with the foreign sector. In deriving the money supply function, we solve all the structural equations of the model simultaneously 2 j j so as to obtain a reduced-form solution of the money ! supply. This reduced-form money supply gives the equili- i brium stock of money as a function of exogenous variables. The analysis is focused on explicating an account ing framework of the monetary system and specifying rele vant behavioral assumptions in order to isolate the imme diate determinants of the money supply. To achieve this, we deal separately with the mechanical operations of the commercial banks and of the | monetary authorities. The main emphasis is on showing i how the operations of each may affect the money supply i | and credit in the country. j We next look at the money-creating sector of the j economy viewed as a whole. Statistical constructs are introduced to enable us to identify the factors respon- r sible for changes in the money supply and bank reserves. ; We then build a preliminary model in which both | i f the money supply and the reserves of the banks can be shown to be determined by (a) the net foreign assets of Jordan, (b) the deposits-to-reserves ratio of the commer- j cial banks, and (c) the asset preference of the public. i I We later introduce a fuller model for money and j i I i reserves in which income is at first considered to be an exogenous variable and then an endogenous variable. Reduced-form solutions are presented. BACKGROUND OF THE STUDY A controversy exists concerning the determinants of the level of income. Whether one accepts the income- expenditure approach or the quantity-theory approach as an explanation of the level of aggregate income, one is in fact saying that both approaches indicate a relation ship between money and income. In both approaches the role of money is important, although in the income-expen- diture approach the influence of money is indirect, whereas in the quantity-theory approach, it is direct. The ability and desire of consumers, producers, and repre sentatives of government to spend, in relation to the productive potential of the economy, influence the move ments in a country's employment, output, and prices. If the total demand for goods and services is smaller than the economy's ability to produce them, unemployment will rise, and output will fall below the country's productive potential. In such a case, a government can attempt to promote economic well-being through both fiscal_and mone 4 tary actions. Only the monetary side is of interest to us here. The monetary actions are designed to promote total spending, consistent with the country's productive capacity, by influencing money, credit, and interest rates. Since in both the income-expenditure and quantity- theory approach, money influences employment, output, and prices, money can be regarded as a strategic economic variable; it becomes essential to understand the basic factors underlying changes in the money supply. Jordan has never had a financial crisis. The money issue has never been a center of controversy, al though money touches every phase of the economic and poli tical life of the country. The 1950's were notable in Jordan for political unrest and unsettled conditions, whereas the 1960's {until 1967) were notable for relative political stability and widespread confidence in the j economic progress and potential of the country. However, : in both decades the unemployment rate was high. ) j Prom 1949 to 1964, the responsible monetary I i authority in Jordan, the Jordan Currency Board (JCB), because of the laws governing it, was merely an automatic | money changer with no independent monetary policy. The 5 existing monetary arrangements precluded any central banking functions or monetary control. The foreign exchange surplus provided the basis and stimulus for ex pansion of the stock of money. No deliberate off setting policy could be adopted when there was an outflow of money abroad? nor could a policy be devised to sterilize the inflow of money stemming from the international economy. Establishment of the Central Bank of Jordan (CBJ), in 1964, did not radically alter this situation, and no significant monetary operations were undertaken. For these reasons, the CBJ can be looked upon as merely an extension of the JCB. The commercial banks kept a major proportion of their assets in London money markets and their portfolio position in a highly liquid form. Thus the ability of the CBJ to control the lending operations of the commercial banks was restricted by the relatively low volume of bank credit and the strong reserve position of the banks, which minimized the need to seek accommo dation with the CBJ. Changes in currency and credit were thus primarily functions of changes in the balance of payments; the more favorable the balance, the more 6 sterling would be deposited in London and the greater would be the quantity of local currency. The level of the money stock seems to be of no concern to the JCB. Although the CBJ looks at the level, it has never attempted to influence it. For a number of reasons — the most important being the laws governing the operations of the monetary authority — it has not been the behavior of the JCB or the CBJ that played a major role in influencing the money stock, but rather the behavior of the commercial banks and the public. Despite their neutrality, however, the JCB and the CBJ may have indirectly extinguished a part of the money stock by allowing the commercial banks unlimited access to portfolio investment in the London money markets. It is to investigate these factors and their influence on the money supply that this study is under taken. No previous single study of Jordan's money supply has been attempted. Methods of Research The relationships examined here are subjected to empirical testing. The work uses the statistical methods of ordinary and two-stage least-squares. Since the 7 results of investigations employing ordinary-least-squares techniques alone may yield biased results, attempts are made to derive relatively unbiased estimates of the struc tural parameters by the use of two-stage least-squares. in evaluating the estimates derived, statistical criteria are applied, including the coefficient of correlation and the standard error of the coefficients. Most data used in the study come from sources published by the JCB and the CBJ. Additional data were obtained from literature issued by the Jordan Department of Statistics and from the International Financial Statis tics published by the International Monetary Fund. Organization of the Work In Chapter II, a review is made of the literature concerning the theoretical formulation and the different approaches to money-supply determination. In Chapter III an historical and descriptive study of the Jordanian economy from 1951 to 1966 is presented to provide factual background needed for the analysis. Emphasis is placed on the evolution of the monetary and banking system, and on economic development. In Chapter IV, the accounting framework of the monetary and banking system in Jordan is 8 described so as to isolate the immediate determinants of the money supply. Chapter V deals with relationships among the surplus in the balance of payments, the high- powered money, and the money supply. The money-supply model is constructed and alternative hypotheses for the behavioral equations are presented in Chapter VI. Chap ter VII discusses the best methods and criteria to be used for evaluating the various hypotheses, and presents the test results. A summary of the complete study, to gether with conclusions and recommendations, is presented in Chapter VIII. CHAPTER II MONEY-SUPPLY DETERMINATION: THEORY AND APPROACHES During most of the history of modern economics, analysis of the money stock has failed to produce well- developed techniques of the kind that have been applied to other areas of economic analysis. On the demand side, i Hicks criticized the monetary theorists' practice of using accounting identities instead of the marginalist tools of value theory. On the supply side, things have not been much different. The money-supply function has been left to the institutionalists to deal with, on the reasoning that the money-supply function is an economic phenomenon influenced to a great extent by institutional arrangements and legal requirements. One can, therefore, expect changes in the money-supply function as institutions i J. R. Hicks, "A Suggestion for Simplifying the Theory of Money,'1 Economica, NS-2, (February 1935), pp. 1-19. 9 10 evolve through time, and also because of differences between countries. One can best study the development of money-supply theory by reviewing and analyzing the significant contri butions made to the theory. Before doing so, we need to look at the question of what specifically constitutes the money supply. DEFINITION OF MONEY 2 Johnson distinguishes four schools of thought on the appropriate definition of money. One school emphasizes the role of money as a medium of exchange, defining it as currency plus demand 3 deposits. Latane is one proponent of this definition. This formulation is also used in the statistical presen tations of monetary data by the International Monetary 2 H. G. Johnson, "Monetary Theory and Policy," American Economic Review, LII (June 1962), pp. 335-384. ^H. A. Latane, "Cash Balances and the Interest Rate: A Pragmatic Approach," Review of Economics and Statistics, XXXVI (November 1954), pp. 456-460; also Income Velocity and Interest Rates: A Pragmatic Ap proach," Review of Economics and Statistics, XLII (Novem ber 1960), pp. 445-449. 11 Fund and by many other monetary institutions throughout the world. This definition is preferred on an a priori basis for less developed countries by some: In developing countries the problems appertaining to important financial super structures hardly arise, so that for them emphasis on the control of the money supply . . . defined as above . . . should suffice for many years to come.4 On the other hand, the school represented by Friedman defines the quantity of money as "a temporary abode of purchasing power,and in its empirical work defines the quantity of money more specifically as cur rency plus total commercial banked deposits. In addition to justifying its definition on the basis of the lack of separate estimates of demand and of time deposits before 1914 for the United States, this school argues that the appropriate criterion for judging whether time deposits are sufficiently close substitutes for other items is the 4 E. E. Jucker-Fleetwood, "The Money Supply in Mature and in Developing Economies," Irish Banking Review (March 1961), p. 20. 5 M. Friedman and A. Schwartz, A Monetary History of the united States, 1867-1960, National Bureau of Economic Research (Princeton; Princeton University Press, 1963), p. 650. 12 observation whether income is more highly correlated with their sum than with each component separately. it thereby concludes from empirical results that money in the United States should include currency plus deposits of the commercial banks. A third school, opposing the position of the for mer two, consists of those who place emphasis on monetary policy rather than on monetary theory, and explains money according to a broader concept, measurable or unmeasurable. A measurable concept is exemplified by the long-estab- lished Federal Reserve Board theory that it is the total amount of credit outstanding that matters, with the quan tity of money exerting an influence only because bank 7 credit is a component of total credit. An unmeasurable concept is represented by the Radcliffe Committee's ®M. Friedman and D. Meiselman, "The Relative Sta bility of Monetary Velocity and the Investment Multiplier in the United States, 1897-1958," Commission on Money and Credit, Stabilization Policies (Englewood Cliffs, New Jerseys Prentice-Hall, 1963), pp. 181-183. *7 S. E. Harris, et al., "Controversial issues in Recent Monetary Policy: A Symposium, " Review of Economics and Statistics, XLII (August 1960), pp. 245-282. 13 g concept of the liquidity of the economy. Q Gurley and Shaw^ typify the fourth school. They are concerned with the implication for velocity of the presence of liquid assets that are closely substitutable 10 for money. Gurley argues that in the United States the behavior of the interest rates is explained by move ments in the relationship of liquid assets to the gross national product in current prices. In this work, money supply is defined as currency plus demand deposits of the public, consistent with the definition used by the Central Bank of Jordan. Multiple Expansion Studies on factors affecting the money supply usually consider total reserves of commercial banks or some other magnitude as a constraint on money-supply g Committee on the Working of the Monetary System (Chairman: The Rt. Hon. The Lord Radcliffe, G.B.E.), Report (London: 1959). g J. G. Gurley and E. S. Shaw, Money in a Theory of Finance (Washington, D.C.: The Brookings Institution, 1960). G. Gurley, Liquidity and Financial Institutions in the Postwar Economy, Study Paper 14, Joint Economic Committee, 86th Congress, 2nd Session (Washington, D.C.: Government Printing Office, 1960) . _____ 14 expansion for given reserve requirements, so that central bank actions affecting total reserves and reserve require ments set the maximum limit on the volume of commercial bank deposits and thus influence the total money supply. Other factors are also considered, such as the reserve requirement on time deposits, currency held by the public, and excess reserves held by the banks. The basic hypothesis of the multiple-expansion mechanism is that banks, finding themselves with excess reserves, will purchase assets and thereby increase the money supply. The banks do so because it is profitable. The extent to which they are able to do so is limited by the constraints imposed upon them by the public in the form of currency drains, deposit shifts between banks, and shifts between time and demand deposits. There are also constraints imposed by the central bank in the form of reserve requirements, as well as constraints that the banks impose on themselves in the form of excess reserves. A main assumption of the multiple-expansion approach is that these constraints take the form of fixed coeffi cients. In his analysis of the money supply, Fand writes: 15 In money and banking textbooks there is a simple link between bank reserves, de posits and money. In a world where banks use all their reserves, where there are no free reserves, and where both the banks and the public do not undertake any portfolio changes, there is no need to concern ourselves with the money sup ply since it is basically a matter of arithmetic. Once we get away from the simple, mechanical link between reserves, deposits and money, the supply of money has an independent existence as an econo mic variable determined by behavior and subject to analysis.H 1. Phillips The simple textbook link between bank reserves and money supply referred to by Fand is generally attri buted to Phillips, whose contribution was to show the difference between individual bank and systemwide expan- 12 sion for a given inflow of reserves. Phillips explained the way in which cash in commercial banks becomes the basis of manifold loans and deposits. This is the earliest complete treatment of the deposit expansion 13 process in a multi-bank fractional reserve system. ^D.I. Fand, "Some Implications of Money Supply Analysis," American Economic Review, LVII (May 1967), p. 380. 12C. A. Phillips, Bank Credit (New York: Mac millan, 1920). ^J. A. Schumpeter, History of Economic Analysis (New York; Oxford University Press, 1954), p. 1116. 16 He drew a sharp line of distinction between credit exten sion by an individual bank and that by banks taken in the aggregate. This was a breakthrough, because earlier accepted statements of banking theory had made no such distinction. It had long been observed that banks were able to extend credit equal to several times their reserves, and the inference had been that what was true of banks in the aggregate was true of each. The inference had been supported by the observed fact that the balance sheet of any representative bank carries loans several times the amount of the reserves held. it was thus rea soned that an addition to the reserves of an individual bank would put it in a position to increase its own loans manifold. Phillips' contribution lies in deriving depo sit and loan expansion coefficients and in specifying the amount of new loans which can be supported by new reserves. By assuming that the whole banking system comprises a single bank, carrying out the loan and depo sit business of the entire country and maintaining a reserve-deposit ratio of R, the net deposit of a given amount of cash or reserve c would enable the bank to lend, in addition to its outstanding loans, 17 1 „ v c - c (c - Rc) or --- R R He also showed that a representative bank in a system is actually able to lend an amount equal to its excess re serves and not a multiple of these reserves, because of the loss of reserves resulting from the creation of new loans. Phillips also showed that the formula of the loan and deposit expansion of the monopoly bank is valid for a multi-banking system. Although Phillips hinted that some cash might be lost as a result of rising prices and a consequent increased demand for hand-to-hand money, he did not attempt to reformulate his analysis to include the effect of this new factor on the loan and deposit expansion. This was left to later economists, who gave a more complete statement of the deposit expansion mech anism, taking into account other factors that affect the money supply. 2. Shaw 14 Shaw illustrated a formula of the loan 14 E. S. Shaw, Money, Income and Monetary Policy (Chicago: Irwin, 1950) pp. 122-143. 18 coefficient that not only included the reserve require ment ratio on demand and time deposits* but also took into account such factors as the relative desire of the public to hold their money in the form of currency, de mand deposits or time deposits, and the desire of banks to hold excess reserves. His explanation of the loan coefficient deals with the relationship between changes in earning assets, on the one hand, and variations in surplus reserves, on the other. He gives an analysis of the loan coefficient for the single bank and for a system of banks, and iso lates factors that determine the capacity of banks to buy earning assets, given the volume of surplus reserves. When bank A buys earning assets, there will be drains on its surplus reserves as follows: Ea [U“Ha) + r' (sHa) + (r + r»)Ha(l - s) (1 - where E_ = the security purchases of bank A in dollar terms H - the size of bank A. This size is measured with a the help of the primary deposit liabilities. If bank A receives 5 percent of the primary deposits originating from the purchase of securities, then its size is 0.05 and the size of the banking system is 1. A primary deposit is one that is not the direct proceeds of a bank's purchase of securities. r' = the legal reserve ratio for time deposits measured as a percentage r" = the working reserve ratio for demand deposits. Working reserves are part of the excess reserves of bank A held as a precaution against claims for payment that may arise from the bank's exist ing liabilities. The other part of the excess reserves is the surplus reserves, which are available to meet claims arising from additions to the present level of liabilities, r = the legal reserve ratio for demand deposits, mea sured as a percentage s = the public's demand for time deposits. It is a measure of the proportion of the liquid assets that the public wishes to hold in the form of time deposits. c = the public's demand for currency. It is a measure of the proportion of the liquid assets that the 20 public wishes to hold in the form of currency. Let Ka designate all the terms inside the brackets where K is the coefficient of cash drain for bank A a which tells us the amount of surplus reserves that will be lost to bank A for each dollar purchase of earning assets. The reciprocal of Ka is the bank A loan coef ficient. If the surplus reserves of A are B , then the a expression for the possible maximum purchase of securities is Ba(l/K ), and the relationship between the reserves and the purchased earning assets can be put as either Ba = E K or E « B (1/K ). “ a a a a a If we define the money supply to be equal to currency plus demand deposits, then the gain in the money supply is less than the addition to bank A's earning assets by s proportion of the security purchase. This is measured by the monetary coefficient / 1-s A , which is O)- the amount by which bank A is capable of adding to the money supply for each dollar of its surplus reserves. The monetary coefficient therefore defines the possible expansion of the money supply. The coefficients for the entire system of banks can be found by substituting H or 1 for H . The a 21 coefficient of cash drain is: K = r's + (r + r") £(l-s) The loan coefficient for the whole system is then l/K, and the monetary coefficient is / 1-s A . These coeffi- I K I cients define the relationship between surplus reserves and the change in earning assets and the money supply. This multiple-expansion process has been at the forefront of money-supply analysis since 1920. This is the orthodox analysis presented in present-day money and banking textbooks. 3. Criticism Money supply determination under the multiple- expansion process follows a simplistic mechanical expan sion formula with fixed coefficients that describe the behavior of the public's desire for currency, demand and time deposits, and the banks' demand for liquidity. This approach uses coefficients fixed by institutional con siderations rather than by behavioral relationships. Moreover, the multiple-expansion process considers the fixed marginal coefficients to be equal to the average coefficients calculated from the balance-sheet data. 22 Since the central bank controls the reserve base, most macroeconomic theories make the actual money supply exo genous. The deposit expansion in such forms as those pre sented by Phillips and Shaw have come under attack. Orr and Mellon1^ felt that such deposit expansion relies on a comparative static deterministic analysis. Since impor tant results in economic theory have been obtained through incorporation of uncertainty into decision models, they explored the potential effects of uncertainty in the cash flows of banks on the expansion of bank credit, and compared the banks' response to excess reserves under uncertainty with the response indicated by the traditional theory. However, since their assumptions regarding port folio alternatives are restrictive and it is difficult to appraise the effect quantitatively, Orr and Mellon felt that they had reached a tentative rather than a final conclusion. It was that the deposit expansion coefficient derived from their uncertainty model is lower 15 D. Orr and W. G. Mellon, "Stochastic Reserve Losses and Expansion of Bank Credit," American Economic Review. LI (September 1961), pp. 614-623. 23 than that derived from the deterministic model. Bank Adjustment and the Reserve Position Another study made for the purpose of finding factors that determine a country's money supply is that 16 of Riefler. Riefler stated that bank indebtedness is the most important factor in the determination of the money supply. According to him, banks borrow at the "Fed" (in the United States) only in necessity and repay their borrowing as soon as possible. Therefore, banks borrow mainly when open-market operations reduce their reserves and repay these funds when open-market operations increase their reserves. The open-market operations of the Fed itself influence bank borrowing, which is the most impor tant element in determining the market interest rates. The market interest rates are the main determinants of the public's supply of earning assets to the banks. Hence, through open-market operations, the Fed can control bank credit and the money supply. 16 W. W. Riefler, Money Rates and the Money Markets in the United States (Harper, 1930). Meigs went a step further. The question he tried to answer was whether the Fed1s open-market opera tions, which were intended to produce a particular rate of growth or contraction of member-bank deposits, should attempt to control the total reserves of the member banks or their free reserves (free reserves = excess reserves - borrowed reserves). His principal conclusion was that the use of total member-bank reserves as a proximate goal for open- market operations would afford more precise control over the rate of bank deposit expansion or contraction than would the use of free reserves. He developed two main arguments: (1) that the rate of change of deposits is closely related to the rate of change of total member- bank reserves, but is not closely related to the level of free reserves; and (2) that the use of free-reserve targets may produce perverse results, such as contraction of the money supply in recessions and rapid expansion of deposits during recoveries. Meigs was basically attacking the 17A. J. Meigs, Free Reserves and the Money Supply (Chicago: University of Chicago Press, 1962). 25 reserve-position approach to monetary control, which had its origin in Riefler's work. While Phillips and Shaw dealt mainly with the deposit-expansion mechanism in a multi-bank fractional reserve system, Orr, Mellon, Riefler, and Meigs dealt with the behavioral patterns and functional relations in the response of banks to Fed policy actions. Meigs made a contribution to the policy making process. The primary point of his work was that the ac tual level of free reserves was a poor indicator of tight ness and an equally poor policy target. The reason was that banks retained in their corporate consciousness a desired level of free reserves based on market rates and the discount rate. They adjusted in that direction irre spective of open-market policy. Meigs' contribution to the development of the theory of money supply was that adjustments by banks constituted an endogenous force in money-supply determination. Banks adjusted according to market rates and the discount rate. The main implication of the bank adjustment approach for the theory of money supply was that the money supply became an endogenous variable in this system. 26 Specification of Empirical Money-Supply Relationships In the 1960's there were many attempts to specify money-supply functions based on the studies referred to above. Some of the most important of these attempts, taking into consideration the behavior of such variables as the banks1 holdings of excess reserves, the public's holdings of currency, and time deposits, were carried out by Friedman-Schwartz,'1 ' 8 Cagan, Brunner,20 Brunner- 21 72 Meltzer, and Teigen. " I Q Friedman and Schwartz, op. cit. ^9P. Cagan, Determinants and Effects of Changes in the Stock of Money, 1875-1960. National Bureau of Economic Research (New York: Columbia University Press, 1965). 20 K. Brunner, "A Schema for the Supply Theory of Money," International Economic Review, II (January 1961), pp. 79-109. 21 K. Brunner and A. H. Meltzer, "Some Further Investigations of Demand and Supply Functions for Money," Journal of Finance, XIX (May 1964), pp. 240-283. 22r . L. Teigen, "Demand and Supply Functions for Money in the United States: Some structural Estimates," Econometrica, XXXII (October 1964), pp. 476-509. 27 23 Working's presentation with regard to the com modity markets (but equally applicable to the money mar ket) clearly indicated that observed quantities and ob served prices reflect the intersections of demand and supply functions. He pointed out that only under the condition that one of the curves is fixed for all the ob servations will the observed quantities and prices trace out the stationary function. This is not unfamiliar to economics now, but it seems to have made little progress with monetary economists in its day (1920's). Two empir ical investigations of money supply which are relatively free of simultaneous estimation bias are those of Teigen and Brunner-Meltzer. 1. Fr iedman-Schwar t z In analyzing the proximate determinants of the nominal stock of money in the United States, Friedman- Schwartz define money supply to include currency plus total bank deposits of the public. The determinants are high-powered money H, the deposit-to-reserve ratio D/R, 33E. J. Working, "What Do Statistical 'Demand Curves' Show?" Quarterly Journal of Economics, XLI (February 1927), pp. 212-235. 28 and the deposit-to-currency ratio D/C. H is an analyti cal concept cited as a constraint on the maximum size of the money supply. It is sometimes called "monetary base, " and can be calculated by defining either its sources or its uses. The Friedman-Schwartz exposition relies on the uses of H — total reserves of the commer cial banks R and currency held by the public C. D/R depends on legal reserve requirements, and the expectation of currency flows and interest rates. It takes into consideration the decision of banks re garding excess reserves. D/C depends on interest rates, income, and the public's preference for holding currency. Friedman-Schwartz present the following money-stock iden tity in terms of these three determinants: J2_( i + -EL.) MS = H -------- — D D R C The stock of money can therefore be regarded as an arithmetical result of manipulating H, D/R, and D/C.24 24 A similar connection between the money stock and the high-powered money, the deposit-reserve ratio, and the deposit-currency ratio is developed in Chapter V. The rationale behind the development of this equation is also presented. Finally, this equation is used to find 29 Any change in the stock of money can be attributed to a change in the three determinants. The interaction of changes in H and the two ratios determines changes in the money stock. Although the Friedman-Schwartz equati on is an identity, it nonetheless has a theoretical justification for the process of money-supply determination. H is pro vided by the monetary authorities. Both the banks and the public compete for its use. The banks need it to meet their reserve requirement and to obtain desired ex cess reserves. The public needs some of it for use as currency. Since by definition H = R + C, all the H is always claimed. 25 Hansen provides the theoretical justification for the functioning of this money-supply process, through the adjustment mechanism of the banks as they react to a discrepancy between their desired and the actual excess reserves, and also through the public as they respond the contribution of each determinant to a change in the stock of money in Jordan between 1951 and 1966. 25 A. H. Hansen, Monetary Theory and Fiscal Policy. (New York: McGraw-Hill, 1949) . 30 to a discrepancy between their desired and the actual currency holdings. As the monetary authorities increase H, the actual level of excess reserve of the banks becomes greater than the desired level, ceteris paribus. Banks will therefore acquire earning assets, increasing their deposits and reducing their actual excess reserves to the desired level. The public will also adjust their currency holdings. More reserves are needed to maintain the desired ratios because of the increase in deposits, until actual reserves become equal to desired reserves. There can also be a change in money supply even without an increase in H, if changes take place in the ratio d/r or D/C. With this framework, Friedman-Schwartz concluded that changes in H were the dominant determinant of long-term and major cyclical movements in the money supply, whereas changes in the D/R and D/C ratios exerted an important influence on movements in the money supply during financial panics. Changes in the D/C ratio con tributed significantly to movements in the money supply during mild cycles. They reached these conclusions through an examination of successive historical episodes in U.S. monetary history between 1867 and 1960. 31 2. Cagan This attempt is similar to that of Friedman- Schwartz. It was aimed at providing a broad historical analysis of monetary behavior. Cagan directed his study toward three specific tasks: (1) describing the secular and cyclical movements in the money supply and identifying the institutional channels through which they occurred; (2) looking beyond these channels to analyze the under lying factors and relationships at work; and (3) using the findings of supply factors to clarify the monetary effects on output and prices. Although Cagan uses the same money-supply framework as Friedman-Schwartz, never theless he arranges the three proximate determinants differently. Currency is expressed as a ratio of money C/M, and reserves as a ratio of deposits, R/D. His equation is expressed as follows: M = H ------------------------------ C + R _ C . R M M M D Using this framework, he presents a statistical and descriptive analysis of the economic factors account ing for the relative contribution of each proximate deter minant to secular and cyclical changes in the money supply between 1875 and 1960. He concludes that the growth in H is the most important factor affecting the long-term growth in the money supply. c/M and r/d contribute little to secular change in money supply. H is affected by changes in the gold stock, Fed operations, and Treasury operations. Cagan believes that the currency in circu lation exceeds the amount needed for retail transactions. Currency therefore serves as a store of wealth. Demand for currency depends not only on transaction uses affected by the volume of consumer expenditure and the cost of checking accounts, but also on wealth holdings affected by total private wealth and interest rates paid on sub stitutes (for example, savings deposits). The C/M ratio depends upon how the relevant demand factors affect cur rency and commercial banks differently. The r/d ratio is analyzed in terms of reserve requirements and desired excess reserves. 3. Brunner While the Keynesian revolution contributed to logical clarification and empirical specification of the demand for money function, Brunner believes that nothing has been done to the theory of the money supply. In his "Schema," Brunner talks of the uncultivated state of money-supply theory and complains that "...most of the literature presents us with an accumulation of analytical fragments which form at best only building blocks of an 26 empirically significant theory." His purpose was to extend the existing analysis and "... to construct a schema which can be used to formulate an empirical macro supply function of money within the institutional frame- 27 work of the American banking system." To Brunner the desired rate of change of the asset portfolio of a bank is related to its surplus reserves S. To build his money-supply function, he makes use of the "loss coefficient" along the lines of Shaw's analysis. The loss coefficient measures the drain in surplus reserves which occurs as a result of a one-dollar expansion of earning assets. He derives A through a detailed exam ination of "... the possible forms in which reserves can 20 be drained away whenever earning assets are increased.1 1 ° 26 Brunner, op. cit.. p. 79. 27Ibid., p. 80. 28Ibid.. p. 81. These forms include the withdrawal, in the form of currency of a portion of new deposits, the redeposit of some of the new deposits at other hanks, the spillover of newly created deposits into time deposits accounts, and the increase in required reserves and desired excess reserves to cover demand deposits left at the expanding bank. Con sideration of the factors influencing each of these separate loss components leads to an expression that defines the loss of the surplus reserves per dollar of asset expansion. The reciprocal of this measure repre sents the bank's expansion coefficient, which, when mul tiplied by the dollar amount of surplus reserves available, determines the total supply of bank funds which will be come available through the adjustment of the bank's port folio to the new desired level. Brunner first develops a supply function for a single bank and then weaves the material relating to single banks and their intercon nection into a money-supply function for the banking system as a whole. It is important to understand where the surplus reserves of a given bank come from. Brunner describes sources that can generate the surplus reserves which 35 induce the asset expansion. He classifies the "events and magnitudes" that generate these reserves into eight categories: 1. The currency flows into and out of banks which are independent of changes in portfolio and deposits. 2. The conversions between demand and time deposits of the public. 3. The transactions between the Fed and a bank, such as bank borrowing and open-market operations in volving the public. 4. The net clearing balances for a single bank. 5. The changes in reserve requirements. 6. The reallocation of the assets of a bank between balances with other banks and deposits at the Fed and vault cash. 7. The reallocation of the desired reserves holdings of other banks which cause an inflow of reserves to a given bank. 8. The changes in the desired level of excess reserves of a given bank. When these eight factors are properly specified. 36 they determine the generation of surplus reserves, s. This figures, together with ^ , describes the response of a single bank to s "... in the form of a demand for 29 earning assets designed to absorb s." Brunner proceeds from the position of the indi vidual bank "... to build a theory which oontains some general propositions concerning observable patterns of 30 individual banks." He uses the relationships earlier described to present a framework for the construction of an aggregative schema. This schema specified the rate of change of the money supply induced by the system's surplus reserve position. The money-supply relation thus derived is expressed as follows: M = a - mxc0 + m2t0 + mQ(B + L3) + m4 £ 0 + Q- ^ Q) where M = the total money supply (DD + C) mQ = a multiplier describing the reaction in M to a unit change in the monetary base (high-powered 29 Ibid., p. 88. 30 . . Ibid. 37 money) CQ “ a function of a vector of variables expressing the public's demand for currency t = a function of a vector of variables expressing the public's demand for time deposits (B+L )= the monetary base adjusted for changes in re serve requirements. B represents currency in circulation plus total reserves of banks, and L reserves liberated (or frozen) by changes in re serve requirements or "...changes in the system's average reserve requirements induced by shifts in the relative position of single banks associated 31 with the deposit redistribution mechanism." ) - a factor to represent the banks' demand com- o o ponent for Federal Reserve money in excess of required reserves. In empirical investigations, Brunner compared the correlation coefficients of (1) a simple hypothesis that acknowledged only B as a determinant of M and (2) extended hypothesis that also considered the effect of cQ, tQ and 33, Ibid., p. 92 38 3 L on M. The correlation coefficients are compared on the basis of their interpretation as measures of the "com parative predictive performance" of the hypothesis under consideration. He found out "... that the neglect of cq, 3 . . tQ and L results m a hypothesis with a poor and unre- 32 liable predictive power,1 1 whereas stable results are obtained and the hypothesis is relevant if currency demand, time deposit demand, and the cumulative effect of changes in reserve requirements are included in the explanation of the money supply. 4. Brunner-Meltzer Brunner-Meltzer develop both a linear and a non linear theory of the money supply. Only the linear hypo thesis will be dealt with here. This hypothesis is derived from Brunner's "Schema." The authors adopt high-powered money as the vari able limiting the maximum size of the money stock, calling it the monetary base B. Contrary to Priedman-Schwartz, 33 they use the source method of computing the base B. 32 Ibid., p. 98. 33 The source method for calculating the high- 39 B is controlled by the Fed through open-market operations and the discount rate, so that the Fed can offset changes in the base resulting from changes in the other sources. The factors of their money-supply function are: currency held by the public C, time deposits at commer cial banks T, and bank excess reserves (BR). C and T depend on both money wealth (C, T, and DD) and other economic factors. That part of the change in C or T which results from a change in money wealth is known as the "spillover effect." The rest of the change in C or T depends on various interest rates, the cost of checking and time deposit accounts, and nonmoney wealth. CQ and Tq are used to denote the part of C and T influenced by these variables. powered money or the monetary base takes into consider ation the following sources of H: Member-bank borrowings from the Fed Other Fed credit Gold stock Treasury currency outstanding Treasury deposits at the Fed Treasury cash holdings Other deposits and other Fed accounts In contrast, the use method includes: Member-barik reserves Currency held by the public. 40 Changes in bank holdings of excess reserves are divided into two parts. One part is a direct "spillover effect" where a change in excess reserves is induced by a change in bank total private deposits. The other part of the change in excess reserves depends on the reserve requirement, the cost of holding excess reserves (interest rates), and the cost of reserve deficiencies. This part is denoted by ERq. Their money-supply function is ex pressed as: M = mQ + m^(B + L) - ni2C0 - hi3To - m^ERQ where M = DD + C o ra^ = a money multiplier whose size depends on the aver age reserve requirements; the currency, time de posits, and excess reserve spillover effects men tioned above; and the pattern of interbank pay ments (B+L) = the extended monetary base. This includes the monetary base (B) and reserves "liberated" by reserve requirement changes and shifts in depo sits between classes of member banks, between non member and member banks, and between time and 41 demand deposits. These liberated reserves are called L, so that m^ (B+L) can be looked upon as the average response of money to changes in the extended base. m2Co = a part of currency held by the public nuT = a part of time deposits at commercial banks 3 O m^ERQ = a part of member bank demand for excess reserves These last three items represent the influence of economic factors other than the spillover effects included in the money multiplier. The m's in these expressions are also multipliers, but they have values different from that of m^. The theoretical justification underlying this money-supply process is similar to that of Friedman- Schwartz. Here the surplus excess reserves lead to a change in money supply. A change in B leads to surplus reserves and thus to a discrepancy between actual and desired excess reserves. The banks react to these sur plus reserves by eliminating them through an adjustment of their holdings of earning assets, resulting in a cor responding change in their deposits. Through this adjust ment process, money wealth and bank deposits expand, and 42 there are spillover effects into C, T, and ER. These effects are reflected in the size of m^. The money stock changes by an amount equal to multiplied by the change in the extended base. Surplus reserves can also arise from changes in CQ, Tq, and ER^ with no changes in (B+C). The resultant change in the money stock is a function of C - T , and ER times their various multipliers (mo, m , o o ° - 3 and m.) . 4' By investigating the economic forces underlying the demand of the public for C, T, and the banks' demand for excess reserves, Brunner-Meltzer concluded that move ments in the money supply are primarily determined by movements in the extended base and the public1s currency behavior. Another conclusion is that the Fed's open-mar- ket operations are a major determinant of variations in the monetary base, and hence in the money supply. With the exception of the 1930's, the interest rate had very little effect on the money supply. 5. Teiqen Teigen attempts to get around the problem of simultaneous-equations bias by estimating a supply-demand model of the monetary sector by means of two-stage least- squares. He deals with supply-demand relations influen cing the money stock, attempting an integration of these monetary relations into a simultaneous-equations model. In this model, an aggregate money-supply relation, an interest-responsive transactions demand function, and a reduced-form income equation are estimated jointly for the postwar and interwar periods. Teigen reduces his analysis of the factors affec ting the money supply down to the exogenous factors over which the Fed has some direct control, rather than assum ing that the Fed has direct control over the size of the money supply itself. By accepting institutional arrange ments and behavioral relations as given at any one moment of time, he conceives total reserves in the system as de termining a maximum attainable money stock. This is broken into two parts: (1) an exogenous element based on reserves supplied by the Fed (unborrowed reserves) and (2) an endo genous element based on reserves created by member-bank borrowing. His purpose is to focus on the relations deter mining bank borrowing, so as to explain the value of the observed money stock M, to the exogenously determined component of the total potential money stock M*. 44 To achieve this, Teigen assumes that the banks are motivated primarily by considerations of profit and risk. Therefore, the major determinant of the degree to which banks will expand the money supply by borrowing from the Fed and creating “endogenous" reserves is the differential between the cost of borrowing and the return from making loans. Teigen states that ...the expected relationship between the ratio M/M* and its proposed functional arguments is straightforward. When the return on loans rises, ceteris paribus, banks presumably will shift the compo sition of their portfolios in such a way that relatively fewer securities will be held, the margin of excess reserves kept as a buffer against unforseen contingencies will be cut down in size, and in general, an effort will be made to expand loans and deposits, causing a rise in the money stock relative to M*. The opposite ef fect will result from an increase in the cost of making loans. The money-supply function used in his empirical tests is a linear relation between the above-mentioned ratio and the differential between the cost of borrowing and the return from making loans, plus two dummy variables to account for changes felt to have occurred in the basic 34 Teigen, op. cit., p. 481. 45 structure of the money-supply mechanism over time: ...in particular, it was felt that the structure of the money supply mechanism was likely to have changed over the periods studied, and structural shift variables were added to the regressions to take account of such shifts.3^ The contribution of simultaneous-equations models to money-supply theory is the explicit recognition of the simultaneous-bias problem and the formulation of a more satisfactory link between the monetary and financial sectors, and vice versa. General Equilibrium The general-equilibrium approach is an extension of the methodology of the simultaneous-equation system to all monetary and financial markets. It is the appli cation of the Walrasian framework to the monetary and financial markets, where stocks of securities replace commodity flows and relative rates of interest replace relative prices. The essential conceptual point made by the gen- eral-equilibrium is that all markets are interrelated. 35Ibid., p. 482. 46 The demand for each asset depends upon relative rates of interest and each sector's wealth constraint. In this system, the interrelationships are of a higher order of complexity. Not only do they work through relative rates, hut the quantities demanded by one sector affect the wealth constraint of other sectors. The quantity of de mand deposits, time deposits, and currency held by the private sector affects the wealth constraints of the banking sector. The quantity of financial intermediary assets demanded by the public determines the wealth or total portfolio constraint of financial intermediaries. In general, the model takes the form of demand and supply equations for each asset as a function of competing rates on assets available to that sector and the sector's wealth constraint. D S Qj_ (ri»r2' • •• i ^ ~ ^1 (rl'r2' * rn ,w) • • * + • • • m *D *S Qjj (r^,r£, ... , w) — ... » rn,w) The general-equilibrium approach makes two main contributions to the theory of money supply. First, the methodological breakthrough of the simultaneous- 47 equation system is further advanced to include all markets. Second, the portfolio reactions of the banking sector are extended to include the portfolio reactions of the private sector, so that the significance of a given stock of reserves depends now on the reactions of all agents. No longer do banks automatically "create" a fixed multiple of assets and deposits, and no longer does the private sector automatically hold all the deposits that the bank ing sector wishes to "create." The private sector is faced with several choices, including demand deposits, time deposits, currency, financial-institutional instru ments, and other private securities. The money supply is now jointly determined by the banking sector, the pri vate sector, and the financial intermediary sector for any given central-bank posture. The money supply is now only one of many variables of equal standing. 36 A good example is the work of De Leeuw. Others ■37 who have contributed are Gurley and ShawJ' and F. de Leeuw, "A Model of Financial Behavior," frfre Brookings Quarterly Econometric Model of the United States. J. S. Dusenberry et al., editors (Chicago: Rand McNally, 1965), pp. 465-532. 37 Gurley and Shaw, op. cit. 48 38 GoIdfeld. Balance of Payments By studying the economies of countries under the 39 40 > 11 gold-exchange standard, Shenoy, Shannon, Vinelli, x Blowers and McLeod,^2 and Friedman and Schwartz42 concluded that the money supply was determined by changes in the balance of payments. Both expansion and contraction of the money supply were linked automatically with the balance of payments position. 38 S. M. Goldfeld, Commercial Bank Behavior and Economic Activity: A Structural Study of Monetary Policies in Postwar United States (Amsterdam: North-Holland, 1966). ^B. R. Shenoy, "The Currency, Banking, and Ex change System of Thailand," IMF Staff Papers, I (September 1950), pp. 289-314. 40H. A. Shannon, "Evolution of the Colonial Sterling Exchange Standard," IMF Staff Papers, I (April 1951), pp. 334-354; also "The Modern Colonial Sterling Exchange Standard," IMF Staff Papers. II (April 1952), pp. 318-362. 4*"P. VineHi, "The Currency and Exchange System of Honduras," IMF Staff Papers, I (April 1951), pp. 420- 431. 42G. A. Blowers and A. N. McLeod, "Currency Unifi cation in Libya," IMF Staff Papers. II (November 1952), pp. 439-467. ^3Friedman and Schwartz, op. cit. 49 With the exception of Friedman and Schwartz, these economists were satisfied to describe the gold-exchange standard and the automatic relationship existing between the surpluses and deficits in the balance of payments and the stock of money. Friedman and Schwartz showed that, in addition to the balance-of-payments position, one must include the behavior of the commercial banks and the public in determining the money stock, so that the link between the money supply and the balance-of-payments position is only partially automatic. The present study of the money supply will develop the reasoning behind this balance-of-payments approach to determining the money supply and incorporate it with the Friedman-Schwartz approach (Chapter V). Structural equa tions for the demand of currency by the public and the demand of reserves by the commercial banks will be developed. Simultaneous-equation techniques will be used to derive reduced-form solutions for the money-supply function (Chap ter VI). This money-supply function will relate the money supply to a number of explanatory variables. 50 Summary The evolution of money-supply theory has been characterized by; (i) substitution of behavioral relations for the fixed coefficients of the multiple-expansion pro cess; (ii) recognition of the fact that the observed money supply is a result of the intersection of the demand and supply functions; and (iii) explicit recognition of a multi-market, Wabrasian-type general-equilibrium market framework applied to the entire monetary and financial sector. The money supply was once thought to be an exo genously determined variable, as in most classical and Keynesian models. Later, however, it has come to be regarded as a variable determined by the behavior of all sectors. Money-supply analysis has correspondingly passed from a simple process to a complicated theory of economic behavior. CHAPTER III ECONOMIC DEVELOPMENT AND THE EVOLUTION OF THE MONETARY AND BANKING SYSTEM IN JORDAN, 1951-1965 We shall now present an historical and descriptive study of the Jordanian economy from 1951 to 1965 to pro vide a factual background for the analysis of the money supply. The Jordan Currency Board (JCB) Jordan, as constituted today, has existed only since 1950. It comprises the former kingdom of Trans jordan, which became an independent state in 1946, and that portion of Palestine which was in Arab hands when the hostilities with Israel ended in 1948. For 23 years from February 7, 1927, to July 1, 1950, the monetary unit operating as legal tender in Jordan was the Palestine pound issued by the Palestine Currency Board (PCB). This Board was established in 1926 51 52 with members appointed by the united Kingdom. During 1949, steps were taken to establish a new currency, the Jordan dinar; the Jordan Currency Board (JCB) was constituted and its duties defined by Jordanian legislation. By the provisions of the law establishing the JCB, the Jordan dinar was set as the monetary unit of Jordan, and the ex change value of this unit against the pound sterling was fixed at par. The new currency was issued on July 1, 1950, and the JCB exchanged Jordanian for Palestinian notes and coins at the rate of one Jordan dinar for one Palestine pound, redeeming the Palestine currency against payment in sterling by the PCB. The JCB consisted of a chairman and four members appointed by the Jordanian Council of Ministers for three years, all being eligible for reappointment. The law under which the JCB was constituted was in many ways simi lar to the law constituting the PCB. Both laws gave very limited powers to the Board, making it solely an issuing authority without power to deal with monetary and finan cial problems. The main functions of the JCB were to issue currency in Jordan against payment in advance in sterling in London, to make available sterling in London against the surrender of dinars in Jordan, and to invest it3 assets in sterling treasury bills and securities issued by governments other than that of Jordan. The JCB had to keep reserve assets sufficient to guarantee the redemption of currency in the ratio of not less than 100 percent, and also a reserve fund sufficient to cover a fall in the value of its assets. The net annual profit of the JCB was the only contribution made by the Board to the Jordan Government budget. Any expansion of the currency involved the immobilization of an equivalent amount of sterling, since the foreign assets of the JCB were held at all times in U.K. government securities. Any increase in the currency in circulation in Jordan resulted in a de facto loan by it to the United Kingdom. This monetary arrangement provided an absolute guarantee of 100 percent reserves. During the first seven years of its existence, the JCB had its offices in London and was represented by a Currency Officer in Amman. In 1957, the headquarters of the Board was transferred to Amman and an Investment Committee was set up in London to supervise its sterling assets. 54 Of the first five members appointed to the Board, three were from the United Kingdom, one of whom was the Chairman, and two were Jordanians. When the headquarters was moved to Amman, a new Board was appointed with four Jordanian members and one member from the United Kingdom. The Chairman was the Jordan Minister of Finance; the other three Jordanian members owned the three domestic commercial banks. The JCB continued its operations until October 1, 1964, when the newly established Central Bank of Jordan took over its functions. During its lifetime, the JCB had none of the functions of a monetary authority and had no powers for implementing a monetary policy. It kept itself completely detached from the economic life of the country. It exercised no influence on the money supply or credit in Jordan, either directly by its own operations or indirectly by influencing or controlling the operations of the commercial banks. The Central Bank of Jordan (CBJ) In the late 1950's, the government of Jordan decided to establish a Central Bank that would replace the JCB and assume a wide variety of monetary control. In 55 1959, the law constituting the Central Bank of Jordan (CBJ) came into being, but it was not implemented. In 1960, an amending law designed to expedite the implementation of the 1959 law was drafted, finally receiving parliamen tary approval in 1962. This legislation provided the legal framework within which monetary policy would operate. Toward the end of 1963, a Governor, a Deputy Gover nor, and five members were appointed to the CBJ. On October 1, 1964, the CBJ began operations by taking over the functions of the JCB and the Currency Control Depart ment of the Ministry of Finance, which had been respon sible for exchange control, including the issue of cur rency permits for imports. The objectives of the CBJ were to regulate the issue of currency, to keep reserves with a view to main taining monetary stability in Jordan and the external value of the Jordan dinar, to influence the credit situa tion to the country's advantage, and to act as banker to the government. The law provided the CBJ with the fol lowing methods for achieving its objectives: 1. Regulating the quantity and quality of credit and the rates of interest, so as to satisfy 56 the needs of commerce, industry, and agri culture , 2. Adopting suitable measures to deal with gen eral or local economic or financial crises, and making recommendations to the government on financial and economic policy. 3. Supervising banking institutions. 4. Managing the government's reserves of gold and foreign exchange. The law was at best vague, ambiguous, contradic tory, and badly written. No reference was made to growth or development among its objectives, in sharp contrast to the objectives of central banks in most other less developed countries. However, the CBJ had the more con ventional objectives of internal monetary stability and stable exchange rate. To serve these objectives, the CBJ was given the following powers: (1) the sole right to note issue, (2) authorization to buy and sell gold and foreign exchange and (3) freedom to engage in open market operations. Moreover, to aid in the control of credit and the super vision of commercial banks, the CBJ was vested with the 57 power to establish a minimum liquid-assets ratio and with direct control over the volume of credit for any one cus tomer and the conditions of commercial bank credit. The CBJ was legally defined as a corporate body. Responsibility for policy and administration was vested in the members of its Board. Appointments were made by the Council of Ministers. In addition to the Governor and the Deputy Governor, the first Board consisted of three bankers and two businessmen. Concerning its rela tions with the government, any regulations and instruc tions issued by the CBJ must be approved by the Minister of Finance. As fiscal agent to the government, the CBJ was responsible for administering the government accounts, and acted as a depository for government funds. As banker to the government, the CBJ could make temporary advances to the government in the event of a temporary deficiency of budget revenue. Such advances could not exceed 10 per cent of the total revenue of the current government budget. They had to be repaid by the end of the financial year in which they were made. If they were unrepaid after that date, the CBJ could not make more advances in any sub sequent year. The CBJ was empowered to deal with the 58 government's treasury bills and securities. Regarding its relations with the financial insti tutions, the law allowed the CBJ to require commercial banks to maintain minimum liquid-assets ratios. However, the law did not contain authorization to require com mercial banks to maintain a percentage of their deposit liabilities in the form of reserves at the CBJ. These liquid assets were defined to include currency, balances with the CBJ or any bank in Jordan or the united Kingdom, and treasury bills of the Jordanian and united Kingdom governments. The law empowered the CBJ to purchase, sell, and rediscount credit instruments of commercial banks. Instruments for financing agricultural and industrial production were made acceptable for longer maturities than those covering commercial transactions, a differentiation through which the CBJ could encourage the development and financing of these sectors. Although it seemed as if the CBJ had the power to pursue the monetary policy it desired, in fact it was unable tc do so. The reason was that other considerations in the law rendered the CBJ ineffective — principally, the requirement that the CBJ hold 100 percent foreign 59 reserves to cover the currency issued. The list of assets eligible for inclusion in the foreign reserves was carefully specified and included only the following: 1. Gold coin and bullion. 2. Currencies covertible into gold. 3. Sterling balances. 4. Sterling treasury bills of the government of the united Kingdom. 5. Sterling securities of, or guaranteed by, the government of the United Kingdom. The CBJ was, in reality, nothing more than a glorified currency board. Of the main instruments used by central banks to achieve their desired monetary policies — the discount rate, open-market operations, and minimum reserve requirements — the first two became ineffective, and the third was modified so that the CBJ did not have the power to control the supply of money and credit in Jordan, the raison d'etre of a central bank. Such legislation not unnaturally led to successive amendments in the law as the CBJ found itself unable to exercise its powers because of the contradictory legal 60 provisions. During 1966, changes in the existing legis lation became necessary and a new legislation was drafted. It became effective in October 1966, just two years after the CBJ had started its operations. These amendments will be discussed in Chapter IV. The Commercial Banks Until the late 1950's, commercial banking in Jordan was concentrated in the hands of foreign banks. In 1955, there were four commercial banks, three of them branches of British banks with head offices in London. The only domestic bank was the Arab Bank, which was established in Jerusalem, Palestine, in 1930 and moved to Amman in 1948. The Arab Bank operated in many Arab countries, so that for none of the banks was Jordanian business a major part of its total activities. One of the foreign banks, the Ottoman Bank, acted as fiscal agent to the government. By 1965, there were nine banks in Jordan. Pour of these were privately owned Jordanian banks, the other five being a Lebanese bank, an Iraqi bank, an Arab League bank, and two British banks. Prom 1951-1964, the freedom of operation of com mercial banks in Jordan — especially with respect to 61 their choice of investments — was not limited by any regulation, such as having to keep minimum reserves in Jordan. Under the canons of sound banking, the banks confined their lending to "self-liquidating" purposes, which actually meant the finance of foreign trade, and a large part of their portfolio investments was in the London money markets. Commercial banks were not allowed to acquire direct interests in commercial, industrial, or agricultural enterprises or to purchase equity shares in such enterprises. The financing of foreign trade was a major activity. Very little credit was given to either agriculture or industry. Banks in Jordan maintained a relatively high liquidity ratio, for the following reasons: 1. Under the JCB, there was no lender of last resort. 2. The commercial banks held relatively large government deposits, and there was uncertainty as to the utilization dates of these deposits. 3. There was no capital market in Jordan through which banks could buy or sell securities. 4. Because of the credit policy of the commercial 62 banks, there was a scarcity of creditworthy borrowers in Jordan. 5. Political unrest and unsettled conditions gripped Jordan during the 1950's. As Jordan had no internal public debt, the invest ment portfolio of its banks consisted entirely of balances with banks abroad, mostly in the United Kingdom, and of British treasury bills. The main characteristics of the commercial banking system in Jordan were: a concentration of banking services in the larger cities, a corresponding absence of these services in the rural areas, and a close relationship with the London money market. Nevertheless, with all the shortcomings of the commercial banking system and its lack of total partici pation in the economic life of Jordan, banks grew steadily. Demand, time, and savings deposits of both private and public sectors increased steadily between 1951 and 1965, and the economy became more monetized. Accompanying this growth, to be sure, were certain adverse and inhibiting effects: e.g., the: political instability of the 1950's, hoarding, illiteracy, the unavailability of banking 63 services in rural areas, and stamp duties on the use of checks. Money Supply The requirement of 100 percent sterling reserves for the currency issued precluded the JCB from providing rediscount facilities for domestic bank assets, and in volved the immobilization of an equivalent amount of ster ling whenever there was an expansion of the currency issued. The situation did not differ significantly under the CBJ law. In Jordan, over 50 percent of the money supply consisted of currency in the hands of the public (Table 1), a situation not uncommon in many less developed countries. As the economy grew, a further increase in the currency held by the private sector developed, which necessitated the immobilization of an equivalent amount of foreign exchange. This foreign exchange could be produced only by a net surplus between international receipts and pay ments. - We may conclude that the money supply was determined mainly by the balance of payments, the credit policy of the commercial banks, and the asset preference of the public. 64 TABLE 1 COMPOSITION OF JORDANIAN MONEY SUPPLY. 1951-1965 (millions of Jordan dinars) Year (Dec. 31) Currency in the Hands of the Public Demand Deposits of Public Money Supply 1951 8.36 5.91 14.27 1952 7.72 4.91 12.63 1953 8.55 5.08 13.63 1954 10.64 6.10 16.74 1955 11.13 6.36 17.49 1956 14.56 5.68 20.24 1957 14.71 7.49 22.20 1958 15.15 9.17 24.32 1959 15.27 8.97 24.24 1960 15.63 10.46 26.09 1961 16.97 11.95 28.92 1962 19.04 14.43 33.47 1963 20.40 16.45 36.85 1964 23.02 19.98 43.00 1965 26.36 23.20 49.55 Source: Currency Control Department and CBJ reports. 65 Jordan was heavily dependent on foreign trade and aid to finance its imports. The exchange reserves and, in turn, the money supply were therefore subject to fluc tuations over which the monetary authorities had very little control. The result was limitations on what mone tary policy could achieve. Foreign Assets of Jordan The foreign asset holdings of Jordan increased over the years (Table 2) in spite of large trade deficits. Official donations, loans, and private remittances from Jordanians working abroad were more than sufficient to cover the deficits. The JCB component of foreign assets (that is, prior to 1964) is equivalent to its total currency liabi lities plus the net IMF position. The CBJ foreign assets (1964-1965) include gold, foreign-exchange deposits abroad, and securities of foreign governments. The commercial- bank and government components of foreign assets comprise foreign-exchange deposits abroad and holdings of foreign governments' securities. 66 TABLE 2 FOREIGN ASSETS OF JORDAN, 1951-1965 (Millions of Jordan dinars) Year (Dec.) Central Bank of Jordan3 , Commercial Banks Government Total 1951 9.34 2.67 12.01 1952 8.80 4.21 — 13.01 1953 9.36 5.49 — 14.85 1954 12.04 5.39 — 17.43 1955 12.97 8.49 — 21.46 1956 16.82 8.36 — 25.18 1957 15.67 11.17 — 26.84 1958 16.37 14.40 — 30.76 1959 16.02 14.28 — 30.-30 1960 16.57 14.27 1.01 31.85 1961 18.30 16.08 0.52 34.90 1962 20.72 20.31 0.02 41.04 1963 22.38 12.04 3.22 37.65 1964 27.31 24.70 3.71 55.72 1965 49.94 9.54 3.75 63.23 aPrior to 1964, JCB. Source: Currency Control Department and CBJ reports. 67 Commercial Bank Deposits and Credit Until October 1964, when the CBJ started its operations, the government and semigovernment agencies had demand, time, and savings deposits with commercial banks. A striking feature of commercial bank deposits was the importance of the government and semigovernment agencies in the total (Table 3). At no time did private deposits exceed the currency held by the public, so that a large proportion of monetary transactions was carried out without recourse to the services of the banks, for the reasons enumerated above. Moreover, those who had deposits with bank branches now in Israeli territory (after the 1948 hostilities) were unable to draw on their deposits for many years. This experience, which dis couraged the growth of banking habits, recurred during the 1967 war, so the fears were not unfounded. Another reason for the neglect of bank services in Jordan was that depositors were not insured against bank failures. Many people preferred to deal in currency because they feared losses resulting from the receipt of checks whose payment was not enforced by law. Source: Currency Control Department and CBJ reports ( — 1 H - * I - * H - * H H * I-* H * H * H I - 1 I-* t- 1 H * co V O V O vo V O vo vo V O V D vo V O V O V O V O vo O N O ' O N O ' O ' O' C n C n Ln C n Ln C n C n C n C n C n O ' w N O H * o vo 00 O ' Ln O' Cl NJ t-1 M I — * M H V O Cn ho I-1 V O 00 -4 O ' O N C n Ln O' O' Cl C n ■ * • f t 4 f t f t • • • • f t f t f t f t v l C O H ©• Cl C n H * "4 C n M o- C n o VI H 1 I - 1 v l O "4 Cl H N J C n V O C n N - * N J Cl O 00 I-* 1-* O ' co V O00 C n O' Cl Cl Cl N 1 N - * H o o O ■ • f t f t • * * « f t f t • » • 4 4 Ui L O00 N O O ' ho vo Cl NJ Ln v l N J vo v l VI I-1 N O00 V O H * L n O ' O ' O ' u> O O' Ln O' t- 1 I-1 CO • O ' O ' C n -li CO ro Cl Ln O ' Cl Ld N J © • • • f t • f t • • • • • ■ f t 4 4 c j cn V O SJ O ' 00 O ' o CJ LJ 00 to ho O ' N J o O ' M l-v O O' O' o O o Cl vo H 00 M M o I -1 I-1 O O ' - v j O ' Vl cn ro o o o O O • • f t • • • • • • • 4 f t f t f t f t o O v | Cl 00 - I S ''I O ' h-1 av Cl N J t-» H Ln o N O N O v l O N CJ t— 1 Cn O n I - 1 N J Ld H M Cl O ' O ' N O N O * - ■ N 1 O O o H O « * • • • • f t * • ■ f t f t • f t f t O' O' co vo O' VO 00 JN VO Ln vo Cn o NJ VI 00 NO Ln O' NO Cn O' NJ © -t> cn 00 Cn H Cl C O M (-* ( — 1 o o O o O t-v O O © © * • • • * f t t f t f t • 4 f t ■ f t f t o O' CO NO O O' >4 Cl O' H CJ ro I - * © o u> O © 00 vl NO Cn o O' 00 00 O' Cl NJ tv O' O' CJ co Cl NO NO to H * h - * M I - * 4!* VO O' vo o vl O' M VO O n O' I-1VO 00 * • ■ a 4 f t f t f t f t • f t ■ • f t • o O ' (O o V O Ln o O ' o- L O C n o Ln N J -P - O ' VI O' vo H * vo O ' O' -p- O n C n Cl 00H* U J CJ N O M N O H * M t-* H * H* v l O' Q CO . Ln N O © O 00 O ' Cn O' CJ LJ »' , 1 * f t » * » f t f t f t • f t f t f t f t • N On 00 NO o : CO CJ o ho 00 O' K VJ O' ON Vl VO O' . O' Cn Ln O' O n O' O' 00 O Ln ro M o o o O O O o O O O O o o f t f t f t f t f t • • f t f t ■ • f t f t © O' vo Ln VO VI 00 - s j O n Ln • S ' Ln O' NJ O' M ho C N 00 4S- vo C J 4h* vjD Ln NO o VO O' Cl NJ NJ ro t - * M t - 1 I - 1 K* Cl VO cn H 00 O' Cl o © 00 Ln - p - O' LJ f t f t f t • f t f t f t f t f t f t • • f t f t • Cl NJ - v l o O' vl M vo vl vo LO vo O' H O' © vl vo Cn O' 00 O' 00 00 Ln VO LO 00 VO vo NJNJNJH M H M h - V © O' O' © O' O' • P " O' I — 1 00 00 C n o - O' NJ * 4 f t 4 4 4 4 4 4 f t f t 4 4 f t f t Cn v l vlCJ o NJNJ O' M Cl O' CJ 04 M ON O' O O 00 vl 00 © vl O' vo vo oo VO - ■ J 89 Year (December) Demand Deposits of the Private Sector Time Deposits of the Private Sector Demand Deposits of the govern ment Time Deposits of the govern ment Demand Deposits of the Public Institutions Time Deposits of the Public Institutions Total Deposits at Banks Domestic Credit to the Private Sector Domestic Credit to the Public Institutions Total Domestic Credit of Banks Foreign Assets Held by Banks < - ■ o H a § CL H- » iu n ( A o M no o U1 n H C / 5 © Q to n f * ; o 50 M D J-H Pd CJ © Id O 8 a 2! > CO CO M H CO 69 Interest rates paid by banks on deposits varied according to the nature and size of the account. There was no strict uniformity among banks with respect to rates paid on time and savings deposits. No interest was paid on demand deposits. During the period 1949 to 1965 the interest rate paid by banks on time deposits varied be tween 3.5 and 5.5 percent. The interest rate on savings account was 3.5 percent during the period 1949 to 1956, and 4 percent during the period 1956 to 1965. Because the operations of the commercial banks were not confined to Jordan, a rise in their reserves exerted little pressure for a large expansion of credit in Jordan. There was always the alternative of purchasing earning assets in the United Kingdom. Nevertheless, bank credit to the private sector increased, most of it short term and heavily concentrated in commerce (Table 3) . The charges made for advances varied between banks. The rates amounted to 6 to 8 percent between 1949 and 1956, and 7 to 9 percent between 1956 and 1965, depending on the standing of the customer and the amount of the advance. 70 Balance of Payments A look at Jordan's trade balance between 1951 and 1965 (Table 4) reveals an alarming deficit against Jordan. While exports increased from 2 to 10 million Jordan dinars, imports rose from 16 to 56 million. Jordan followed a lib eral import policy with no restrictions. Import licenses were easily obtained from the Exchange Control Department of the Ministry of Finance. The exchange rate of the Jordan dinar remained at par with the pound sterling during the period (1 Jordan dinar = 1 pound sterling — $2.80 U.S.). If we include the services (Table 5), so as to look at the current account section of the balance of payments, the deficit decreases because of foreign-exchange earnings from tourism. After 1960, tourism became increas ingly important to Jordan, and foreign-exchange earnings from this source rose from 1 million Jordan dinars in 1951 to 10 million in 1965, an amount equal to the total value of exports during that year. Table 5 shows the balance of goods and service transactions. Without international transfer payments to Jordan, in the form of grants, remittances of Jordanians working 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 71 TABLE 4 BALANCE OF TRADE (Millions of Jordan dinars) Import Export Balance 16.18 2.0 -14.18 17.15 2.11 -15.04 18.21 2.66 -15.55 18.59 3.05 -15.54 25.26 3.57 -21.69 24.61 5.16 -19.45 29.76 5.54 -24.22 33.97 3.53 -30.44 39.39 3.41 -35.98 41.43 3.95 -37.48 40.93 5.27 -35.66 43.51 5.92 -37.59 53.63 6.56 -47.07 49.38 8.73 -40.65 55.77 9.91 -45.86 Currency Control Department and CBJ reports. 72 TABLE 5 BALANCE OF GOODS AND SERVICE TRANSACTIONS (Millions of Jordan dinars) Year Total Expenditures Total Receipts Balance 1951 17.42 4.22 -13.20 1952 18.50 4.97 -13.53 1953 19.61 5.99 -13.62 1954 20.03 6.74 -13.29 1955 27.06 8.03 -19.03 1956 26.68 10.23 -16.45 1957 32.59 13.62 -18.97 1958 36.78 11.92 -24.86 1959 42.82 9.95 -32.87 1960 45.85 10.09 -35.76 1961 46.08 14.53 -31.55 1962 50.75 18.77 -31.98 1963 61.35 16.61 -44.74 1964 57.15 20.79 -36.36 1965 64.12 26.68 -37.44 Source: Currency Control Department and CBJ reports. 73 abroad, and UNRWA payments, the deficits arising in the current account would probably not have reached such ex treme levels, and expenditures would probably have been restricted, thus slowing the pace of Jordanian economic development. Total transfers to Jordan amounted to 9 million Jordan dinars in 1951, but 40 million in 1965 (Table 6). During the period under study, Jordan depended heavily on foreign grants. Prom 1951 to 1956, the United Kingdom was the most important of the donors to Jordan; since 1957 this role has passed to the United States. Most U.S. grants were allocated for government budget support; the U.K. grants had been given as a direct sub sidy to the Jordanian Army. Foreign grants from all sources (excluding the private remittances of Jordanians working abroad) amounted to 7 million Jordan dinars during 1951; in 1965 they amounted to 27 million. Private remittances, mainly from Jordanians working abroad, and to a small extent from religious and charitable institutions rose from 2 million Jordanian dinars in 1951 to over 13 million in 1965. Table 6 is a summary of the international transfer 74 TABLE 6 INTERNATIONAL TRANSFER PAYMENTS (Millions of Jordan dinars) Year From U.K. From U.S. From UNRWA Other Total Official Transfers (a) Private Remit tances (b) Total Trans: (a + 1 1951 3.55 3.36 6.91 2.03 8.94 1952 4.60 0.49 4.36 — 9.45 2.30 11.75 1953 6.43 0.97 4.86 — 12.26 2.34 14.60 1954 5.87 1.25 5.50 — 12.62 2.37 14.99 1955 8.04 2.73 4.66 — 15.43 3.37 18.80 1956 8.01 0.47 5.47 0.85 14.80 3.62 18.42 1957 1.15 6.73 4.90 3.51 16.29 4.20 20.49 1958 * • • * • * • • 23.91 3.22 27.13 1959 2.42 17.32 5.40 — 25.14 5.78 30.92 1960 2.00 18.20 5.29 0.64 26.13 7.63 33.76 1961 2.36 17.05 5.15 0.77 25.33 6.97 32.30 1962 1.50 16.71 5.26 — 23.47 10.54 34.01 1963 1.50 15.51 5.51 — 22.52 10.18 32.70 1964 1.50 15.03 5.37 4.67 26.57 13.86 40.43 1965 1.40 11.98 6.01 7.39 26.78 13.29 40.07 Source: Currency Control Department and CBJ reports. 75 payments to Jordan. It includes all official transfers to the government, UNRWA payments, and Jordanian citizens' remittances. Loans, which are excluded, will be considered in the capital accounts. International transfer payments were more than sufficient to cover the deficit on the current account during the study period except for the years 1951, 1952, 1955, 1959, 1960, and 1963 (Table 7). To complete this survey of balance of payments, it is necessary to include (1) all capital receipts resulting from deliberate acts of policy on the part of the government and from movements in private capital and official loans, and (2) the sur plus or deficit resulting from the addition of capital receipts and payments to the balance on the current account, which determines the movement in the foreign-exchange reserves of Jordan. Except for the years 1951, 1959, and 1963, there was an annual increase in the reserves throughout the period under study (Table 7, last column) . By adding these reserves to the foreign-asset holdings of Jordan for the previous year, we arrive at the total foreign-assets holdings at the end of each year (Table 2, last column). TABLE 7 THE FINANCING OF JORDAN1S DEFICIT (1+2) (Millions of Jordan dinars) m « o a a -h o w T 3 u at s > o O O. V» W 0) o » » h -h ( 2\ o mto um h g (B <u cn a <u o o e u x * 4 - 1 U Vl *H O « l 4 J T3 C PS U O *H Vi M f l ’ J i- l PS tl + *H la > a) (0 V ' 4 J Vl C O Q J O 0) 4 - 1 a> L < h ai cdV4J rH s —' to oo a) u a) i s cj s > h a « F, o n e f S V3 O a 44 Q I I 4 J I t 1-4 ( fi <9 O rt VI co CU R B 1HPL4 (0 C .C E i —i t3 o t) 4 - i i —4 4j co >, a. ij ntu at C R <u co c Vi co co Vi o j: X «vt micfl + t f l 53 H H H (V O O H O H O (a)_____________(b)__________(a+b)__________ (c)_____________(d)_________(c+d)________ (a+b+c+d) 1951 -13.20 +0.66 -12.54 8.94 0.78 9.72 -2.82 1952 -13.53 40.47 -13.06 11.75 2.31 14.06 1.00 1953 -13.62 -0.14 -13.76 14.60 1.00 15.60 1.84 1954 -13.29 -0.15 -13.44 14.99 1.03 16.02 2.58 1955 -19.03 40.11 -18.92 18.80 4.15 22.95 4.03 1956 -16.45 -0.02 -16.47 18.42 1.77 20.19 3.72 1957 -18.97 40.24 -18.73 20.49 -0.10 20.39 1.66 1958 -24.86 0.00 -24.86 27.13 1.65 28.18 3.02 1959 -32.87 40.87 -32.00 30.92 0.62 31.54 0.46 1960 -35.76 +2.13 -33.63 33.76 1.42 35.18 1.55 1961 -31.55 40.71 -30.84 32.30 1.59 33.89 3.05 1962 -31.98 +1.16 -30.82 34.01 2.95 36.96 6.14 1963 -44.74 +2.53 -42.21 32.70 6,12 38.82 -3.39 1964 -36.36 +6.01 -30.35 40.43 7.99 48.42 18.07 1965 -37.44 +0.99 -36.45 40.07 2.89 43.96 7.51 Source: Currency Control Department and CBJ reports 77 We shall now turn our attention to the economic development that took place in Jordan under the existing monetary and banking system and trade policies. Economic Development For the purpose of studying the economic develop ment in Jordan, the period 1951-1965 can be divided into two parts: 1951-1959 and 1959-1965. The justification for the division is the fact that during the first period development took place under extraordinary circumstances. As a result of the Palestine War of 1948 and the formation of Jordan from Transjordan and what was left of Palestine, the Jordanian government faced the problem of administering and providing services for a population that had trebled instantly. The population in Transjordan was around 400,000, the resident population of the portion of Palestine incorporated into Jordan was approximately 460,000 and the number of persons who entered this terri tory from what became Israel amounted to 350,000. The major economic effect of the was was inevitably a high level of unemployment. Another economic effect of the War was the dis ruption of the transport and marketing links between the 78 country and the rest of the world. During this period there was also constant political instability, which disrupted the economic life of Jordan. Economic develop ment therefore had to proceed under extremely difficult circumstances, the most difficult being the lack of an administrative infrastructure. Because of political instability, the governments were short-lived and incapable of administering and providing adequate services for the population. The political disturbances of the 1950's, culmin ating in the grave crisis of 1958, gave rise to a new period starting in 1959 which enjoyed a relatively high degree of political stability, a more efficient government apparatus, and a relatively adequate administrative infra structure. 1. Gross National Product Needless to say, there are no exact figures of Jordanian GNP. As in many less developed countries, esti mates are made and then revised in the light of new infor mation and the continual refinement of method. Nevertheless, the available data clearly indicate that development took place. 79 Between 1954 and 1955, the GNP fell as a result of the fall in income originating in the agricultural sector, owing to bad weather. The increase in GNP between 1955 and 1956 can be attributed to a large extent to high agricultural output, 1956 being the best year for agri culture in the first period. There was very little change in GNP between 1956 and 1957, but there was important changes in the structure of the economy. Agricultural production decreased, but this decline was balanced by increases in the production of the other sectors, probably stimulated by increases in the governmental and services sectors. GNP increased rapidly in 1958 and 1959 in spite of the fact that income originating in agriculture showed a further decline. The stimulus to expansion was provided largely by the increases in government expenditures, and in the expenditures of tourists in Jordan. Although the GNP rose by over 50 percent between 1954 and 1959, income originating in the agricultural sector declined both absolutely and relatively. During this period, agriculture was a highly unstable industry. A large proportion of total agricultural output derived from dry farming in areas subject to frequent droughts. 80 GNP rose by over 80 percent between 1959 and 1965. Income from agriculture showed a large increase because of additions to land under irrigation and more efficient use of land by shifts of production from grains to fruits and vegetables. Modern machinery and techniques, and the use of fertilizers and insecticides, were also intro duced. There was, furthermore, more abundant and regular rainfall during this period. Income from manufacturing doubled with the introduction of new, mostly light, indus tries. The increase in GNP may be attributed to the in crease in capital formation of both the private and the public sectors. Factors that contributed to this process were the increases in transfers to the private and public sectors from abroad, in exports of goods and services, and in commercial bank credit. 2. Behavior of Prices There is no cost-of-living index in Jordan. There is, however, a wholesale price index for Amman, weighted by general consumption ratios of (1) cereals and wheat flour, (2) other foodstuffs, (3) building materials. 81 and (4) fuel. This wholesale pricr, index is reliable, although one cannot know exactly how much it reflects the actual cost of living in Jordan, as it fails to include the consumption of manufactured goods, the cost of house rents, recreation costs, and other important items. More over, it only considers the capital city, where prices are generally high. This wholesale price index neverthe less demonstrates that prices have not characteristically risen in Jordan, but have remained relatively stable during the period of this study. CHAPTER IV DETERMINANTS OP THE JORDANIAN MONEY SUPPLY In discussing the determinants of the Jordanian money supply we shall first deal with the mechanical operations of the commercial banks, the Central Bank of Jordan, and the Treasury. Our main emphasis will be on showing how the operations carried on by each may affect the money supply and credit of Jordan. In this chapter, we shall look at the money- creating sector of the economy as a whole. To achieve this an accounting framework will be formulated, enabling us to identify the factors responsible for changes in the money supply and the bank reserves. In the next chapter we shall outline how the money supply and the reserves of the commercial banks are determined by the following in fluences: (1) the net foreign assets of Jordan, (2) the deposits-to-reserves ratio of the commercial banks, (3) the deposits-to-currency ratio of the public, (4) the Treasury 83 deposits (in Jordan vs. abroad), and (5) the local assets of the CBJ. THE COMMERCIAL BANKS AND THE MONETARY SYSTEM 1. Measuring the Money Supply and its Components For the purposes of this analysis, the money supply (Table 8) will be defined to include (1) currency held by the public and (2) demand deposits of the public. The data presented in all the tables that follow are those of December 31, 1966, in millions of Jordan dinars. TABLE 8 SIZE AND COMPOSITION OF THE MONEY SUPPLY Amount Percent 30.33 51.7 28.35 48.3 58.68 100.0 Demand deposits of the public make up less than 50 percent of the money supply. There are various reasons why they are not very attractive, the more important being the lack of banking habints on the part of the Currency held by the public Demand deposits of the public 84 public, the stamp duties on checks, the unavailability of bank branches in the rural areas and their concentra tion in the larger cities only, and the absence of legal enforcement on payment of bad checks. Contrary to the practice in countries with effec tive central banks, the size of the money supply in Jor dan is not determined by the government or the CBJ. On the one hand, the public's preference determines the division between demand deposits and the currency it wishes to hold, and, on the other, the commercial banks determine the desired deposits-to-reserves-ratio. The largest amount of the money supply is held in the cash balances of the private sector (households and businesses). Table 9 shows the division of the money supply among the main groups of holders. TABLE 9 DIVISION OF THE MONEY SUPPLY Holders of Money Supply Amount Percent Private sector 55.23 94.1 Municipalities 0.39 0.7 Public corporations 1.95 3.3 Foreign 1.11 1.9 58.68 100.0 85 At the end of 1966, then, the private sector held ‘ about 94 percent of the money supply, followed by public corporations with only 3.3 percent. 2. The Commercial Banks and the Money Supply The commercial banks in Jordan, as anywhere else, are business enterprises operating for the purpose of making profits. They are divided into two main groups: local banks and foreign banks. This division is impor tant because it influences their credit behavior and portfolio investment. Foreign banks tend to look abroad for the investment of their assets, whereas local banks use the foreign-capital markets to a lesser extent. There is a widely held belief in Jordan that it is the deposits of customers that cause the banks to give credit rather than the credit given by banks that causes the increase in deposits. The size of the money supply will not change immediately when an individual deposits currency in a bank, because the increase in deposits is equal to the decrease in currency held? whereas, when a bank gives credit, deposits will increase. However, the money supply will increase when foreign-exchange earnings arising from foreign trade or transfers are deposited by 8 6 an individual in a bank. The reserves of the banks will increase, causing the banks to increase their lending. It is still safer to state that credit given by the com" mercial banks causes deposits to increase and thus in creases the size of the money supply. Although commercial banks have the power to create money, they nevertheless limit their deposit creation, because they must remain liquid. Until the end of 1966, they were not obliged to hold legal reserves or to adhere to a legal liquidity ratio. The banks in Jordan limit their deposit creation further. Before 1966, they were allowed to hold foreign portfolio investments? after 1966, this practice became illegal. Also, they receive interest on all deposits that they hold at the CBJ. Table 10 is a consolidated balance sheet of the commercial banks operating in Jordan as of December 31, 1966. Vault cash and balances with the CBJ and with banks abroad are considered reserves of the commercial banks. We can see from the table that most of the com- mercial-barik assets consist of earning assets, such as loans and balances with the CBJ and banks abroad. 87 TABLE 10 CONSOLIDATED BALANCE SHEET OF COMMERCIAL BANKS, DECEMBER 31. 1966 (Millions of Jordan dinars) Assets Vault cash Balances with: CBJ 1.57 13.47 Banks in Jordan 0.62 Banks abroad Loans Other assets Total 7.21 38.99 8.40 70.26 Liabilities Demand deposits Time deposits Net worth Borrowed from CBJ Other liabilities 30.94 21.90 5.99 0.05 10.38 Total 70.26 aThis figure differs from that of Table 8, because it includes demand deposits of the Treasury, which are not included in the definition of the money supply. 8 8 It is important to point out that the decisions of banks about the composition of their assets may have a substantial effect on the allocation of capital funds among the various sectors of the economy, and the indus tries within each sector. Moreover, they can affect the allocation of capital funds between Jordan and abroad. Therefore, the operations of the commercial banks have very important implications for the economic development of the country. Table 11 is a sectoral distribution of commercial bank credit outstanding according to the sector of the economy. Trade and commerce make up about 52 percent of all credit granted by the banks, whereas credit to industry amounts to about 12 percent, and agricultural credit to a meager 1.6 percent of the total credit. 3. Reserve Requirements and Liquidity Ratio Historically, reserve requirements have been im posed on banks for the main purpose of protecting against insufficient liquidity. In many countries they have increasingly been used as a means for controlling the money supply and credit. TABLE 11 89 SECTORAL DISTRIBUTION OF COMMERCIAL-BANK CREDIT* DECEMBER 31, 1966 (Millions of Jordan dinars) Sector Amount Percent Trade and commerce 20.33 52.2 Industry 4.72 12.1 Municipalities and public corporations 2.58 6.6 Construction 2.37 6.1 Private individuals 2.25 5.8 Transportation 2.15 5.5 Tourism and hotels 0.99 2.5 Agriculture 0.61 1.6 Other 2.99 7.6 Total 38.99 100.0 90 Until 1968, there were no legal reserve require ments for banks in Jordan. In 1969, legal reserve requirements of 7 percent on both demand and time deposits were imposed for the first time. Furthermore, a legal liquidity ratio^ of 25 percent had been placed on the commercial banks in 1968. It should be mentioned that although no legal reserve requirements had been imposed before 1969, the commercial banks had maintained large deposits at the CBJ and at banks abroad to ensure their liquidity position. THE MONETARY AUTHORITIES AND THE MONETARY SYSTEM 1. The Central Bank of Jordan Central banks can exert influence on the mone tary system by such operations as the following: 1. issuing currency. 2. conducting transactions that directly affect ^As defined by the CBJ, the liquidity ratio - cash at hand + deposits with CBJ + balances with resident and nonresident banks + short-term bills, all divided by deposits of customers + deposits of resident and non resident banks + bills receivable + loans. 91 the size of the money supply, such as open- market transactions in government securities. 3. conducting operations that influence the reserves of the commercial banks and so in directly influence the money supply, such as open-market operations and credit to the com mercial banks. The CBJ should be the principal means by which the money supply and credit are regulated, through opera tions (2) and (3), since these are the most powerful and flexible instruments of central-bank policy and possess a very high degree of precision in timing and magnitude. Such, unfortunately, is not the case. The only activity of the CBJ has been the issuance of currency, which is a passive instrument, since it is subordinate to the wishes of the public. Currency issuance is the least important role of a central bank. This relative passivity reflects the impact of the JCB on the CBJ. A clear understanding of the operations of the CBJ can be gained by looking at its balance sheet (Table 12) . 92 TABLE 12 BALANCE SHEET OF THE CENTRAL BANK OF JORDAN. DECEMBER 31. 1966 (Millions of Jordan dinars) Assets Liabilities Gold 1.21 Currency issued 31.88 Foreign-exchange Treasury deposits 10.00 deposits abroad 20.04 Bank deposits 13.47 Securities of for eign governments 28.58 Capital and reserves 3.16 Rediscounts and Other liabilities 0.41 advances 0.09 58.92 58.92 We can see that there was no CBJ credit to the government and that the credit granted to banks was less than 0.2 percent of the total assets of the CBJ: whereas, over 99 percent of the assets were held in deposits at banks abroad and in securities of foreign governments. The only difference between the CBJ balance sheet and that of the JCB is the Treasury and bank deposits at the CBJ. 2. The Treasury The Treasury does not exert influence on the 93 monetary system except for the Treasury deposits, because there is no national debt in Jordan and the Treasury issues no monetary liabilities. Government securities and treasury bills are nonexistent in Jordan, of the CBJ were either deposited in commercial banks in Jordan or in banks abroad- However, with the establish ment of the CBJ, some Treasury balances were withdrawn from local and foreign banks and deposited in the CBJ. The withdrawal of balances from the Jordan banks caused a corresponding decrease in bank reserves. By the end of 1966, the Treasury held the balances shown in Table 13. The Treasury balances prior to the establishment TABLE 13 TREASURY BALANCES. DECEMBER 31. 1966 (Millions of Jordan dinars) Amount Percent Deposits in CBJ 10.00 65.0 Deposits in commercial banks in Jordan 2.79 18.1 Deposits in banks abroad 2.60 16.9 Total 15.39 100.0 94 THE BASE OF THE MONETARY SYSTEM Surplus in the Balance of Payments The foreign-exchange surplus originating in the balance of payments forms the base of the monetary system in Jordan. It is primarily the movement of foreign ex change into or out of Jordan that causes changes in the reserves of the banks and in the money supply, except when 2 the foreign-exchange surplus is sterilized. The surplus or deficit in foreign exchange re sulting from the balance of payments is equal to the total value of exports minus the total value of imports plus the net transfer and capital flows into Jordan. Part of this foreign-exchange surplus belongs to the public; the other part goes to the government. The public deposits part of this foreign exchange in commercial banks; the banks make the appropriate entry and create deposits for the public in Jordan dinars. The remainder of the public's foreign exchange is exchanged Through government portfolio investments in foreign money markets. 95 for Jordan coins and notes, which then form part of the public's currency holding. The government also deposits a part of its for eign exchange in banks in Jordan; the banks again make the appropriate entry and create deposits for the govern ment in Jordan dinars. The government deposits the remainder of the foreign exchange in banks abroad. The commercial banks deposit a part of their newly acquired foreign exchange in banks abroad; the other part is exchanged with the CBJ for Jordan currency, which is used to satisfy their customers' currency demand. However, after the establishment of the CBJ, the govern ment started depositing most of its balances in the CBJ, the remainder being deposited in commercial banks in Jordan and abroad. Likewise, the commercial banks, after the establishment of the CBJ, started depositing their balances in the CBJ, keeping only international working balances for financial settlements at banks abroad. Figure 1 illustrates the flow of the foreign- exchange surplus and its effect on bank reserves, bank credit, and the money supply. We see that — as a result of this surplus — bank reserves, bank credit, and the FIGURE 1 FLOW OF JORDANIAN FOREIGN-EXCHANGE SURPLUS SURPLUS IN THE BALANCE OF PAYMENTS I j Surplus of the"Public Sectorj Surplus of the Government ■f New deposits at basks in Jordan 1 New Currency Holding of the Public New Deposits at CBJ New Deposits at Banks in Jordan Foreign portfolio investments and deposits at banks Fncw reservesat banks Foreign assets at CBJ Deposits created as a result of credit Credit extended as a result of new reserves Currency withdrawal by public as a result of credit 97 money supply all increase. The opposite occurs with a loss of foreign exchange resulting from a deficit in the balance of payments. CONSOLIDATED ANALYSIS OF THE MONEY SUPPLY So far we have clarified the mechanical operations of the commercial banks, the CBJ, and the Treasury and have observed the effect of these operations on the money supply. We shall now look at the money-creating sector of the economy as a whole and set an accounting framework that will enable us to identify factors responsible for the changes in the money supply and bank reserves. 1. The Consolidated Monetary Balance Sheet In recent years, an accounting mechanism that systematically with factors affecting the money supply, and that can be used to determine the monetary effects of any transaction, has been developed. This mechanism is the consolidated balance sheet combining the various accounts of the monetary sector. By netting out the items appearing on both sides of the accounts of the CBJ and the commercial banks, we are left with a picture of 98 relationships existing between the monetary sector as a whole and the rest of the economy. In this consolidated balance sheet, certain items that exist entirely within the monetary sector, such as cash holdings of the com mercial banks and interbank deposits, are omitted. Table 14 presents such a consolidated account, derived directly from data presented in previous tables. We can see that the total assets of the monetary sector are equal to the monetary liabilities plus the nonmonetary liabilities and capital, so that the money supply is equal to the total assets of the monetary sys tem minus the nonmonetary liabilities and capital. The monetary importance of the foreign assets and credit of banks is thus clearly demonstrated. The money supply in Jordan in the period 1951 to 1966 changed in the same direction as the total monetary assets. The nonmonetary liabilities also changed in the same direction as the total monetary assets. However, the response of the money supply to a change in the mone tary assets was varied, chiefly reflecting the desire of the public to hold time deposits instead of currency and demand deposits. 99 TABLE 14 CONSOLIDATED MONETARY BALANCE SHEET DECEMBER 31, 1966 (Millions of Jordan dinars) Assets Liabilities Foreign Assets: Money Supply: Held by CBJ 58.44 Demand deposits 28.35 Held by banks 7.21 Currency with public 30.33 Held by the Subtotal 58.68 Treasury 2.60 Local assets of CBJ 0.09 Non-monetary liabilities and capital Loan and advances of banks 38.99 Time deposits 21.90 107.33 Treasury deposits 12.58 Capital and net miscellaneous 14.17 Subtotal ____48.65 107.33 100 2. Factors Determining Commercial Bank Reserves The foreign exchange deposited by the public and the government in banks in Jordan as well as the change in currency holdings of the public are the main deter mining factors of commercial-bank reserves. These re serves are in turn deposited either in the CBJ or in banks abroad, a small part (in Jordan dinars) being re tained in the vaults of the banks. It is on the basis of these reserves that the commercial banks grant credit. The relation of these factors to commercial-bank reserves is shown in Table 15. This table is construc ted on the assumption that vault cash plus deposits in the CBJ and in banks abroad constitute commercial-bank reserves. The item labeled "Sources" is known as the monetary base and consists of the reserve assets of the monetary system. The claims against these assets are (1) the currency in the hands of the public, (2) the commercial-bank reserves, and (3) the Treasury deposits in the CBJ and in banks abroad. Consequently, bank reserves increase by any of the following ways: (1) an increase in the monetary base, (2) a decrease in the cur rency held by the public, and (3) transfer of Treasury 101 TABLE 15 DETERMINANTS OF COMMERCIAL-BANK RESERVES (Millions of Jordan dinars) A. Sources 1. Net foreign assets of the CBJ 58.44 2. Foreign assets of banks 7.21 c L 3. Foreign assets of the Treasury 2.60 (deposits at banks abroad) 4. Local assets of the CBJ 0.09 68.34 B. Competing Uses 1. Currency held by the public 30.33 2. Treasury deposits at the CBJ 10.00 3. Treasury deposits at banks abroad 2.60 4. Capital and reserves of the CBJ 3.16 46.09 C. Commercial-Bank Reserves 1. Deposits at the CBJ 13.47 2. Deposits at banks abroad 7.21 3. Currency in vaults of banks 1.57 ______________ 22.25 aThis item could be eliminated from the Sources. It is retained here because the analysis in Chapter V and VI focuses on changes in tot'^1 foreign-assets holdings. deposits from the CBJ to the commercial banks. It is important to note the domination of the monetary base by foreign assets. Foreign assets-in Table 15 (A 1, 2, 3) are roughly equivalent to high-pow ered money (C + Bl)r they differ only by (A4, B2,3,4). If, however, the money supply is defined to include Treasury deposits at the CBJ (B2), then the Treasury deposits would also be included under high-powered money. In this case, the foreign assets would be closely equi valent to high-powered money. However, before the establishment of the CBJ, the determinants of commercial-bank reserves were some what different. By omitting item (4) from (A), items (2) and (4) from (B) , and item (1) from (C) in Table 15, we get the determinants of commercial-bank reserves as they would have appeared, under the JCB system. 3. Summary From Tables 14 and 15, we see that the foreign assets owned by Jordan as a whole are determinants of both money and reserves. Preferences of the public for different types of liquid assets also affect both money and reserves. The money supply is affected by the public' 103 preference for currency and demand deposits, on the one j hand, and by time deposits, on the other. Bank reserves are affected by the public's decision to hold demand deposits vis-a-vis their holdings of currency. Another determinant of the money supply is bank credit to the public. Credit, in turn, depends on the reserves available to the banks, on the banks' deposits-to-reserves ratio, and on the asset preferences of the public. CHAPTER V BALANCE OF PAYMENTS, HIGH-POWERED MONEY, AND THE MONEY STOCK In Chapter IV, we found the surplus in the Jor danian balance of payments to be the basic determinant of the high-powered money, and the foreign assets origi nating in the balance of payments to be almost equal to the high-powered money. In this chapter, we shall analyze in greater detail the relationships among the balance of payments, the high-powered money, and the money stock. The algebraic relationship between the money stock and the balance of payments will be constructed and explained. For a number of reasons, the most important being the structure of the monetary institutions, it was the behavior of the commercial banks and the public, rather than that of the JCB or the CBJ, that played a major role in influencing the money stock. Although the JCB and the CBJ were neutral in directly creating or ex tinguishing money, they may have indirectly extinguished 104 105 a part of it through their lack of action and direction, by allowing the commercial banks unlimited access to portfolio investments in the London money markets. Under a flexible exchange-rate system, the stock of money can be looked upon as an independent variable determined by domestic considerations. The stock of money is of major importance in influencing the level of money income and the price level, which in turn influences the rate of exchange between domestic and foreign currencies. Under a gold-exchange standard with fixed exchange rates'** the money stock can become a dependent variable determined mainly by external influences. The major chan nel of influence is from fixed exchange rates with for eign currencies through the balance of payments to the stock of money, and so to the level of domestic prices consistent with those exchange rates. Under this system, the stock of money must be whatever is necessary to main tain balance with foreign countries, and the amount of foreign-exchange reserves alters through international ^Jordan operated under such a system between 1957 and 1966, and now operates under a modified gold-exchange standard system. 106 trade to produce this result. If the monetary authorities; act automatically in monetizing international reserves, the stock of money becomes a dependent rather than an independent variable, and is therefore not under the direct influence of the monetary authorities. However, under the flexible exchange-rate system, or with central bank discretion in monetizing or sterilizing international reserves, this link between external conditions and the stock of money is broken. It is then possible for the stock of money to be determined by domestic considerations. The link between the balance of payments and the money stock is not really so rigid as some economists have indicated; consequently, the money stock can deviate significantly from the level dictated by external influ ences. These deviations can occur for various reasons — for example, whenever the domestic price level is inconsistent with the fixed exchange rate, or whenever 2 See H.A. Shannon, "Evolution of the Colonial Sterling Exchange Standard," IMF Staff Papers, April 1951, pp. 334-354, and "The Modern Colonial Sterling Exchange Standard," IMF Staff Papers. April 1952, pp. 318-362. Also G. A. Blowers and A. N. MacLeod, "Currency Unifi cation in Libya," IMF Staff Papers, November 1952, pp. 439-467. 107 the money stock is inconsistent with the domestic price level. We shall analyze the relationship between the balance of payments and the money stock, indicating the r . shortcomings of this functional relationship. THE BALANCE OF PAYMENTS AND THE MONEY STOCK The surplus in the balance of payments under the gold-exchange standard system results in an increase of foreign-exchange reserves, on which the commercial banks increase their deposits by a multiple amount, thus causing the stock of money to increase mechanically. Likewise, a deficit in the balance of payments causes a loss of foreign-exchange reserves, a contraction of banks deposits and thus an automatic decrease in the money stock. How ever, this mechanical link is not very complete. 1. The Average and Marginal Propensities to Import Both the average and the marginal propensities to import are relatively high in Jordan. The implications 108 ! of these high propensities are (1) that an expansion in the money stock automatically leads to an increase in the demand for imports; (2) that money used to finance imports is extinguished from the money stock, because the demand for imports reduces the reserves of the com mercial banks (not only limiting their ability to create deposits, but, as a result of the reduction of the re serves at the disposal of the commercial banks, purportedly causing a contraction in deposits); and (3) that for any given increase in the reserves of the commercial banks, the amount by which the banks can increase their deposits will be greater, the smaller the propensity to import and the smaller the desire of the public to hold money balances in the form of currency rather than deposits. These high import propensities in Jordan are unfavorable to the deposit-creating capacity of the banks, and there fore have a significant influence on the determination 3 of the money stock. See J. J. Polak and Lorette Boisonneault, "Mone tary Analysis of Income and Imports and its Statistical Application," IMF Staff Papers, April 1960, pp. 349-415? and J. Marcus Fleming and Lorette Boissonneault, "Money Supply and Imports," IMF Staff Papers, May 1961, pp. 227- 240. 109 2. Foreign Assets and Their Composition The change in the total amount of the foreign- exchange reserves of Jordan for any given year is deter mined by the balance-of-payments position for that same year. The change in foreign reserves is equal to the surplus in the balance of payments. This surplus is equal to total exports minus total imports of goods and services, plus net transfer payments and capital move ments. The total foreign reserves are the summation of all these surpluses over the years. As the public does not hold substantial amounts of foreign exchange, these reserves are owned partly by the monetary authorities, partly by the commercial banks, and partly by the govern ment. The banks in Jordan have maintained under the JCB a relatively high ratio of external reserve balances to their assets. It is the availability of reserves at the commercial banks that influences their deposit crea tion, so that reserves determine the amount and rate of change of the money stock. It is therefore meaningful to say that changes in the ownership composition of the reserves may influence deposit creation and, as a result. 110 money-stock determination. It is not just the change in total reserves that may bring about a change in the money stock; other elements of major importance are (1) how the change in reserves has been brought about and (2) how this change has affected the ownership composi tion of reserves. We must examine these elements before it becomes valid to say that a surplus in the balance of payments will cause an automatic increase in the stock of money. Many cases can be constructed to explain sur pluses or deficits in the balance of payments, and so an increase or a decrease in total reserves does not neces sarily lead to similar changes in the money stock. Below, hypothetical cases not uncommon to Jordan, are presented as evidence to support the argument. Let us assume that the Jordanian government has decided to buy armaments, creating a balance-of-payments deficit and a fall in the total reserves of the country. The action will have no effect on the money stock if the government finances its purchase through its investment 4 portfolio held in foreign exchange abroad. However, ^Table 15 of Chapter IV helps us see how the items change in response to these hypothetical operations. the money stock will contract if the government finances its purchase by using funds it has in the commercial banks. The loss of reserves by the banks will force them to contract their deposits, if they desire to maintain a constant reserve-to-deposit ratio. The effect of the balance-of-payments deficit on the money stock, in this case, depends on whether the transaction passed through or bypassed the banks, and on the desired reserves-to- deposits ratio of the banks. The same is also applicable if the government achieves surpluses in its annual fiscal budgets, leading it to accumulate foreign reserves. If the budget sur plus in any given year has been achieved through an in crease in the rate of taxation, or through an improved system of tax estimation and collection, the money stock will decrease. If the surplus is realized through for eign aid, there will be no change in the money stock, given that the surplus is held in external reserves; whereas, there may be a multiple increase in deposits, if the surplus resulting from foreign aid is deposited in domestic banks and the banks increase their deposit creation as a result. 112 ] 3. The Currency Ratio The ratio of currency to deposits in less developed 5 countries is relatively high. Such a high currency ratxo reflects the public's preference for currency over depo sits. This currency ratio is an important factor affecting the money stock. Currency is effectively 100 percent reserve money, while deposits are fractional reserve money. Therefore, the commercial banks can increase their depo sits by a larger amount, for a given quantity of reserves, with a low currency ratio than with a high currency ratio. The significance of the currency ratio in affec ting the money stock has been recognized by economists. Brunner-Meltzer argue that historical evidence supports their theory, which gives "a substantial role to currency g behavior in the money supply process." There have also 5 See Joachim Ahrensdorf and S. Kanesathasan, "Variations in the Money Multiplier and Their Implications for Central Banking," IMF Staff Papers, November 1960, pp. 126-149? and J. Daniel Khazzoom, The Currency Ratio in Developing Countries (New York: Praeger, 1966). g Karl Brunner and Allan H. Meltzer, "An Alterna tive Approach to the Monetary Mechanism," Part III of Analysis of Federal Reserve Policy-making, A Staff Analy sis Prepared for the U.S. Congress, House Committee on Banking and Currency 88th Congress, 2nd Session, 1964, p. 56. 113 i been attempts to quantify the determinants of the public’s: demand for currency. The most rigorous investigation has 7 been conducted by Cagan. He considers the reasons for holding currency, and also estimates a long-run demand function. Among the reasons for holding currency, Cagan identifies the cost of currency, the cost of currency sub stitutes, real income, interest rates, and special factors such as war. 4. Relationship Between the Money Supply and the Factors That Determine it Before the establishment of the CBJ in 1964, cur rency came into existence in Jordan only when an equiva lent amount of foreign exchange was surrendered to the monetary authorities. Consequently, it was the supply of foreign exchange to the JCB that determined the amount of currency in circulation. This supply of foreign ex change had its origin in the balance-of-payments surplus, making the currency in circulation dependent on external 7 Phillip Cagan, The Demand for Currency Relative to Total Money Supply, NBER, Occasional Paper 62,_New York, 1958; and Determinants and Effects of Changes in the stock of Money 1875-1960, NBER (New York: Columbia University Press 1965). 114 variables. With a surplus in the balance of payments, a part of the resultant increase in external reserves accrued to the commercial banks, which kept some of it in foreign reserves and surrendered the remainder to the monetary authorities for the purpose of obtaining local currency to satisfy public demand for this kind of asset holding. However, after 1964, when commercial banks started depositing reserves at the CBJ, currency also came into existence when individuals withdrew currency from their demand deposits, and the commercial banks obtained currency from the CBJ by drawing down their reserve deposits. The surpluses and deficits in the balance of payments did not automatically lead to multiple expansion and contraction of deposits by the commercial banks. The reaction of the banks to changes in reserves were incon sistent. We cannot say, therefore, that the stock of money and its rate of change varied automatically with Q the balance of payments — which is what Shannon,0 Q Shannon, op. ext. 115 9 Blowers and MacLeod, and others have claimed for coun tries under the system of the gold-exchange standard. The relationship between the money stock and the balance of payments can be expressed as follows: N MS = f ( Z S) i=l where MS = the money stock S = surplus in the balance of payments i — time period An algebraic relationship between the money stock and the balance of payments underlying this equation is constructed below. Let d = reserve ratio of the commercial banks n = currency ratio of the public S =» surplus in the balance of payments D = deposits at the commercial banks AD = change in deposits at the commercial banks C = currency holdings of the public AC = change in currency holdings of the public MS - money supply 9 Blowers and MacLeod, op. cit. j 116 AMS - change in the money supply A Fb “ change in the foreign-asset holdings of the commercial banks AF = change in the foreign-asset holdings of the m JCB V.1 Allocation of the surplus in the balance of payments: s = AF. + A f„ 1j m V.2 Supply of currency to the public resulting from the surrender of foreign exchange under the JCB system: C “ m V.3 The change in currency holdings: C = nAD V.4 Deposit creation by the commercial banks, under the JCB: d(AD) = A f b or AD = AFb d V.5 Definition of the change in the money supply: AC + AD = AMS V.7 V. 8 V.9 V.10 v.n V. 12 117 By equations V.3 and V.5, and substituting: nAd + AD = A MS A d = l a ms n + 1 By equations V.2 and V.3: n A D = A F m Combining equations V. 1 and V.7 to express AF^: 4 Fb - s - A Fm = S - n A D Combining equations V.4 and V.8 to express A D: AD = 1 S d + n Combining equations V.6 and V.9 to express the change in the money supply: AMS a n + i S n + d The money supply is expressed as: MS = n + 1 JEL- s n+d i— l The change in the currency holdings can be ex pressed as: Ac =n4 D = n s d +n 118 Since Ac = , equation V.12 also expresses m the change in the foreign-asset holdings of the monetary authorities. V.13 Finally, the change in the foreign-asset holdings of the commercial banks is expressed as: AF = d S d +n This model represents the operations that took place under the JCB system. The only reason that the commercial banks surrendered foreign reserves to the JCB was to obtain currency. The commercial banks, in fact, were under no obligation to keep deposits at the JCB, as no reserve requirements were imposed on them. The com mercial banks were also allowed to invest their foreign assets in the London money markets. Therefore, they had no reason to surrender foreign reserves except to obtain currency. This practice was consistent with the opera tions of the JCB, whose functions were simply the issue and redemption of currency. However, with the establishment of the CBJ, the commercial banks were asked to surrender (within a speci fied period of time) their foreign assets to the CBJ. 119 They were permitted to hold foreign assets abroad only as international working balances. As a result, equations V.l, V.2, V.4, and V.7 in the model have to be expressed differently under the CBJ. The main changes are the following: V.l S = A f m This is the case where commercial banks surrender all their foreign assets to the CBJ and obtain reserve depo sits (R) at the CBJ. Equation V.2 is now: V. 2 A R + A C = A F m where is equal to the change in the commercial banks1 deposits at the CBJ. Equation V.4 is now: V.4 A D = A r d And equation V.7 is now: V.7 A R +nAD = AFm The money supply equations V.10 and V.ll, as well as the currency equation V.12 come out exactly the same as before. Although simplification is essential in economic analysis, especially when used for testing purposes and therefore to be expected, we feel an obligation to state explicitly that this money-stock relationship is very simplified. The following are the major ways in which 120 the hypothetical formulation has been made less complex: 1. A single equation is used. 2. The money supply is specified by a single explanatory variable. 3. A linear# rather than a non-linear, model is used. 4. The currency ratio of the public is assumed to remain constant. 5. The relationship between the variables estab lishes itself instantly, with no time lags. 6. The reserve ratio of the commercial banks is assumed to be constant during the period. 7. The time deposits of the public are not con sidered. 8. The demand and time deposits of the govern ment are not considered. 9. The effect of the foreign-asset holdings of the government is neglected. 10. The surplus in the balance of payments is considered to be the cause of changes in the money supply (however, with imports being endogenous; S cannot really be a cause). 121 11. Income is absent. Our model in Chapter VI will attempt to deal with items 1, 2, 4, 7, 8, 9, 10 and 11 from this list. It will not attempt to deal with items 3, 5, and 6, which could be treated in some future research. HIGH-POWERED MONEY 1. Factors Affecting the Money Supply We have isolated, and shall discuss below, three main factors that can be regarded as determinants of the Jordanian money supply. These three factors correspond to economically meaningful categories of influences which are to a great extent independent of one another. We can use the model presented above in the proceeding sec tion to derive the separate effect of each of the factors, and also to estimate the fraction of the percentage change in the money supply which can be attributed to each, a. High-powered Money The first factor is the surpluses in the balance of payments. These surpluses are essentially equal to high-powered money. The differences between balance-of-payments surpluses and high-powered money are 122 attributable to: (1) the exclusion of the local assets of the CBJ (which are minimal), (2) the capital and re serves of the CBJ (minor compared with the assets of the CBJ), (3) the Treasury deposits at banks abroad (small and decreasing), and (4) the Treasury deposits at the CBJ. Because the differences are minor, we can use H instead of S in the equations. The causation runs from S to H, S being basically the sources of H, and the cur rency holdings of the public and the reserves of the banks being the uses H. b. Deposits-to-Reserves Ratio The second factor which reflects the behavior of the commercial banks, is the amount of deposits that the banks create relative to the amount of reserves (R) that they hold. Under the JCB, commercial-bank reserves consisted entirely of cash and of foreign-exchange re serves held abroad. As no legal reserves were required in Jordan, the amount of reserves that the banks wished to hold relative to their deposits depended on their desire to hold "precautionary reserves." This desire reflected what each bank thought it wise and safe to hold to assure the ability to honor demands for redemption of 123 its liabilities. This is the age-old problem facing the banks in striking a balance among profitability, liquidity, and risk. One can assume that in the short run each bank has its reserves well under control and tends to look upon its deposits as given. We should be careful to point out that the situa tion facing the banking system as a whole is not the same as the situation facing the individual bank, but rather very much the opposite. An individual bank feels that it can decide how much reserves it wishes to hold: if the bank decides to increase its reserves, it can do so by selling some of its assets. But by following this course, the bank is acquiring reserves at the expense of other banks, because the total amount that all banks to gether can acquire is constrained by the total amount available in the country. If all the banks decided to increase their reserves by selling assets, there would be a decrease in deposits through the deposit-contraction process. The absolute holding of reserves by all the banks might or might not increase? however, in trying to achieve this increase in reserves, the ratio d/r would fall. Sooner or later the fall in the D/R ratio would 124 satisfy the desire of the banks for liquidity, whether or not the absolute amount of reserves rose. Prom this analysis we can conclude that it is not D or R alone that the commercial banks control, but rather the ratio D/R. However, under a number of circumstances, the actual and the desired ratios may be temporarily different. c. Deposits-to-Currencv Ratio The third factor which reflects the behavior of the public with respect to its money holdings is the ratio between the public's holdings of deposits and its holdings of currency (D/C) . The decision by the public to divide its money holdings between currency and deposits can be affected by various considerations, the most impor tant of which are: (1) the relative usefulness of cash, 10 (2) the cost of holding currency, and (3) income. There are also indirect factors such as the literacy rate, the availability of banking services in various parts of the country, the cost of the use of deposits for effecting payments, and the banking habits of the public. •^Cagan, op. cit., p. 4. 125 The commercial banks can influence the public1s D/C ratio by providing better and cheaper services and extending them by opening branches in all parts of the country? the banks have no other influence on this ratio. In Jordan, currency is used for transactions purposes and as a store of wealth. People hold it for convenience and for the safety it provides. Although there have been no bank failures (with the exception of the freezing of customers' deposits as a result of the Palestine War), neither have there been sigfns of emerging banking habits on the part of small savers. We can make the same analysis for the individual vis-a-vis the whole public as we have made above for the single bank vis-a-vis the entire banking system. What the public controls is not currency or deposits alone, but rather the D/C ratio. The individual moneyholder feels that he can control the absolute amount of currency and deposits: whenever he decides to exchange deposits for currency, he can do so easily at the bank. However, as a result of the increase in his currency holding, the reserves of the bank he deals with decrease. The bank is therefore forced into contracting its deposits to achieve 126 its desired D/r ratio. Thus, given H and the behavior of the banks D/R ratio, any decision to convert deposits into currency (a reduction of the individual's D/C ratio) will result in a decrease in the money stock because the increase in currency is more than offset by the decrease in deposits. The money stock is therefore determined by three factors: H, D/R, and D/C. H reflects the operation of the gold-exchange system, but not the behavior of the monetary authorities. These authorities have no power to increase H; because under this system (JCB) they are not allowed to issue fiduciary H; whereas, in the United States, H reflects the behavior of the monetary authori ties and they can increase it at will.^ The behavior of the commercial banks is reflected by D/R and that of the public by D/C. These two factors are therefore under the control of two different economic units. (Although various influences affect each ratio independently there are still some links between the ratios, as shown ^^With amendment of the CBJ law in 1966, the mone tary authorities were able to influence H within narrow limits. 127 above.) It is assumed in this Chapter that the average deposit-to-reserves ratio and the average deposit-to-cur- rency ratio are equal to their marginal ratios, and that the factors are independent of each other. 2. Construction of the Algebraic Relationship V.14 The definition of the money stock: MS = C + D V.15 The definition of the high-powered money: H = C + R V.16 Dividing equation V.14 by V.15, then dividing both numerator and denominator by C and multiply ing them by D/R, we have: MS = ( d/r)_(1 +. P/C? H (D/R) + (D/C) 12 V.17 Our money-stock function can be expressed as: MS = H / (D/R) (1 + D/C) \ ^ (D/R) + (D/C) J 12 This equation is merely a transformation of Equation V.12. We use it for simplicity of calculations. H replaces S, as they are almost equal. 128 3. Effects of Chancres in High-powered Money, Deposits-to-Reserves Ratio, and Deposits- to-Currency Ratio on the Money Supply Equation V.17 deals with the levels of the varia bles. It is, however, preferable to deal with the dif ferences, in order to determine the change in the money stock attributable to changes in the three factors. More over, it is necessary to determine the fraction of any given change attributable to the change in each one of the three factors. By taking the logarithms, we represent the absolute changes in this form: V. 18 log MS = log H + log + log (1+/? ) - log ) However, for continuous changes we write the total deri vatives of log MS as: Let ot represent D/R, equation V.17, so that13 represent D/C in 13 . In relation to our earlier symbols, of course, is 1/d and ft is 1/n. 129 V. 19 d log MS _ c)locr MS dH log MS do£ dt J H dt } ot dt + ^loq ms d/9 }/3 dt which can be expressed as: V. 20 d log MS _ 1 dH | ft_____ d dt H dt oi(ck+/? ) dt + ot-1 dft <1-+-/S) <<*+/3) dt It is inferred from equation V.20 that the left-hand side represents the rate of change in the money stock, and that the three terms on the right-hand side provide the contribution to that rate by the rate of change in high- powered money (H), in the deposits-to-currency ratio (D/C), and in the deposits-to-reserves ratio ( d/r). From equation V.18 we find that the change in MS attributable to H can easily be calculated, because H enters into this equation as a separable item, and ^ and <x£ do not enter into the partial derivative of log MS with respect to H. Nevertheless, a new problem is created by equation V.18, because and oC are not separate items but enter jointly into the final term of log MS. vative of log MS with respect to 9L , and the same is is a variable in the partial deri- 130 true for that of ^ with respect to . Therefore# we have an interaction between the effects of changes in and oi , with the result that there is no unique way to determine the effect of each of the three factors separately on the change in log MS. This is an index- number problem, which could be solved by considering time points — for example, t and t. .. The change in log MS i i+l between t. and t. , is 1 i+l V.21 log MS = log MSi+1 - log MSi A solution can be provided by considering the changes in MS where each of the factors changes in turn, while the other factors remain at their initial values. We get the following equations: V.22 Effect of H: H = log Hi+1 - log H± V.23 Effect of D/R: D/R = log<D/R)i+1 - log(D/R)i -log (D/R) ±+1+ (D/C) + log (D/R)i+(D/C)i V.24 Effect of D/C D/C a log 1 + {D/C)i+1 - log 1 + (D/C) ^ -log (D/R). + (d/C) . + log (D/R). + ( d/c) , x i+l x i V. 25 Interaction = -log (D/R)+ (D/C) i+^ +log <D/C)i+1 + (d/r) ^ +log (d/r) + (D/c)i+1 -log (D/R) . + (D/C) . X X Equation V.22 shows the separate effect of H, and equations V.23 and V.24 show the separate effects of D/R and d/c , respectively. But the summation of equations V.22, v.23, and V.24 is not equal to equation V. 21. The difference between the sums is equal to equation V.25. When we divide equations V.22, V.23, V.24 and V.25 by equation V.21, we get an estimate of the frac tion of the percentage change in MS which could be attri buted to each factor. Equation V.25 shows the change in MS that cannot be attributed to D/C or D/R separately. Calculations are presented in Table 16 for Jordan, 1951- 1966. 132 TABLE 16 CONTRIBUTION OF PROXIMATE DETERMINANTS TO CHANGES IN THE MONEY STOCK IN JORDAN* 1951-1966 Change in Money Stock that would have been Proximate produced by the indicated Determinant, if Determinant it alone had changed Rate of Change Per Year (Percent) 1. High-powered Money 10.35 2. Deposits-to-Reserves Ratio -0.53 3. Deposits-to-Currency Ratio 0.21 4. Interaction of Ratios -0.03 5. All 10.00 Fraction of Total Change 1. High-powered Money 1.035 2. Deposits-to-Reserves Ratio -0.053 3. Deposits-to-Currency Ratio 0.021 4. Interaction of Ratios -0.003 5. All 1.00 133 The motivation for discussing the money stock in terms of these three factors is that each of them repre sents the behavior of a sector of the economy. In coun tries whose monetary authorities set monetary policies, it is assumed that if the level of H is determined by the authorities, and if the money multiplier (specifi cally they would also be able to determine the level of the money stock CHAPTER VI A MONEY-SUPPLY MODEL FOR JORDAN In Chapter V, we presented the relationship be tween high-powered money and the money stock. However# we introduced a large number of simplifications, in this chapter# we will discuss many of these simplifica tions, in the interest of developing a more complete model of the money supply. We shall start by describing a preliminary model for money and reserves. We shall then introduce the com plete model# and present alternative hypotheses for the behavioral equations of the complete model. Income will at first be considered an exogenous variable in the com plete model; later it will be considered an endogenous variable. Reduced-form solutions will be presented for the model# taking into consideration alternative behavioral hypotheses. A brief comment about models is appropriate at 134 this stage. All models are abstracts of reality. No model can or should be an exact description of the real world. Because we live in a complicated world, we try to impose order and simplicity by seeking out relationships among only a few very important variables. Abstractions come about through necessity; although they do not des cribe the real world exactly, they nonetheless facilitate our understanding of the most important relationships con stituting our economic environment. It is only through an understanding of these relationships that we become capable of forecasting and making policy recommendations with some degree of confidence. Our complete model will have to be limited, be cause certain structural equations are either unavailable, owing to a lack of data, or inappropriate to the Jordanian Monetary sector. For example, no structural equations exist which define demand and supply conditions for the financial markets. 136 PRELIMINARY MODEL OF MONEY AND RESERVES By looking at the major items appearing in the tables of (1) the consolidated monetary balance sheet and (2) the determinants of commercial-bank reserves in Chapter IV, we see that the foreign assets owned by Jordan as a whole are determinants of both money and re serves. The preference of the public for different types of liquid assets also affects both money and reserves. The money supply is affected through the public's rela tive preferences for currency, demand deposits, and time deposits. Bank reserves are affected through the pub lic's decision to hold demand deposits vis-a-vis its holdings of currency. Another determinant of the money supply is bank credit to the public. Credit, in turn, depends on the reserves available to the banks, on the banks' deposits-to-reserves ratio, and on the asset pre ference of the public. It is possible to construct an integrated equili brium analysis of the money supply and bank reserves as 137 a whole,1 as an extension of the simple expansion coef ficient explained in Chapter II. The balance sheet iden tities from Tables 14 and 15 of Chapter IV can be reduced to the following: VI.1 Determinants of the money supply: Foreign assets of Jordan (FA) + local assets of the CBJ (LA) + bank credit to their cus tomers (BC) = demand deposits (D) + time de posits (T) + currency held by the public (c) + capital and net miscellaneous (K). VI.2 Determinants of bank reserves: Foreign assets of Jordan (FA) + local assets of the CBJ (LA) = bank reserves (Rs) + cur rency held by the public (C) + Treasury de posits at the CBJ and at banks abroad + capital and reserves of the CBJ (TD). We now introduce an equilibrium equation in which the desired reserves (Rd) depend on the demand deposits This framework is adopted from Paul B. Trescott. Money, Banking and Economic Welfare. 2nd ed. (New York: McGraw-Hill, 1965), Part Two, pp. 55-121, as a preliminary step toward building a complete model. 138 (D) and the time deposits (T). VI.3 Desired reserves: Rd = dD + tT 2 where d and t are behavioral parameters. VI.4 And in equilibrium: Rd = Rs However, the system is not yet complete. We need one more equation, since the asset preference of the public is a significant influence. If we assume that time deposits are independent of the other items in these equations, and that the public holds currency in a fixed proportion to its holdings of demand deposits, then the equation becomes: VI.5 Asset preference of the public: C — nD where n is the desired currency ratio. If we are now given the amounts of the foreign 2 The reason for d and t being behavioral para meters rather than legal reserve requirements is that during the period 1951-1966 there were no legal reserve requirements in Jordan. Consequently, we cannot speak of "excess reserves" in the usual sense. However, it is possible for a disequilibrium to exist in which actual reserves differ from desired reserves. assets of Jordan (FA), the local assets of the CBJ (LA), the time deposits (T), the capital and net miscellaneous (K), the Treasury deposits at the CBJ and abroad and the capital reserves of the CBJ (TD), as well as the values of d, t, and n, we can easily determine the money supply and the other elements of the monetary situation, as shown below* VI.1 FA+LA+BC-D+T+C+K VI.2 FA + LA - RS + C + TD VI.3 Rd = dD + tT VI.4 Rd = Rs VI.5 C = nD The unknowns are: Rs, Rd, D, C, and BC. Because we know the values of FA, LA, T, K, TD, d, t, and n, we can solve this system by substituting in the respective equations the values known to us. The reduced-form solu tions for D, C, Rs, and Rd are: 140 If we define the money supply ( ms) as currency held by the public (C) + demand deposits (D), then the reduced-form solution of MS will be: MS = 1+n (FA + LA - tT - TD) n+d The high-powered money (H) is defined as the reserves held by commercial banks + the currency held by the public. The bank reserves are equal to the balances held by the banks in the CBJ and in banks abroad + the cash in the vaults of the banks. Therefore, the reduced- form solution for H will be: H = R + C = FA + LA-TD However, because LA and TD are small, H and FA tend toward equality. One of the uses to which such a model can be put is in estimating the way in which changes in the assump tions of the behavioral equations alter the equilibrium position. The model is, moreover, a useful device for evaluating the relative magnitude of various changes in the data. 141 THE COMPLETE MODEL This model differs from the preliminary one by the introduction of three more behavioral relationships: (1) the demand for currency by the public, (2) the demand for time deposits by the public, and (3) the demand for imports. (Alternative hypotheses are given for each.) These three behavioral relationships, together with the demand for reserves by the commercial banks, will be tested within the framework of the complete model. Also, some new explanatory variables and new definitional equa tions have been added. Some variables which were exogen ous in the previous model are now endogenous. The endogenous and exogenous variables used in constructing the complete model may be defined as follows: 1. The Endogenous Variables: Cp = Currency held by the public M = Imports S = Surplus in the balance of payments Tp = Time deposits of the public Dp = Demand deposits of the public Rs = Supply of reserves Rd = Demand for reserves 142 PA = Total foreign assets of Jordan 2. The Exogenous Variables: Y = Income DU = Duty Rates X = Exports TR - Transfers W a X + TR Dg = Demand deposits of the Treasury Tg = Time deposits of the Treasury LA = Local assets of the CBJ TD = Treasury deposits at the CBJ and at banks abroad, and capital and reserves of the CBJ L = LA - TD FAt-1 - Foreign assets of the previous period 3. The Model with Income as an Exogenous Variable Money supply as used in this model will be defined to include currency held by the public and demand deposits of the public at commercial banks. The structure of the model necessitates the division of this concept of money into its component parts. We shall, therefore, consider separately the demand of the public for currency and for 143 demand deposits. As time deposits enter into the model, we shall also consider determinants of the public's demand for time deposits. Currency is held in Jordan for transactions pur poses and as a store of value. In the case of Jordan, we hypothesize initially that these are best represented by income, so that demand for currency may be specified as a linear relation with income; VI.6 Cp = a + bY Because a large proportion of Jordan's consump tion and investment goods are imported, imports rise as income increases. Although Jordanian tariffs have been, and continue to be, designed primarily to raise revenue, increases in tariff rates (DU) should have a negative effect on imports. Our initial linear import function will be as follows: VI.7 M = h + gY = jDU The money supply framework described in Chapter IV demonstrates the importance of explicit consideration of commercial banks' time deposits within any study of the determinants of the money stock. Likewise, the very high rate of growth of the stock of time deposits over the 144 period covered by our study increases the importance of such an analysis. For Jordan, we postulate a direct re lationship between the demand deposits of the public and its holdings of time deposits, unlike currency and demand deposits, time deposits held at the commercial banks yield a nominal return greater than zero. Because this return has seldom changed during the period of our study, we can neglect it in our equation: VI.8 Tp = i + kDp These initial equations for Cp, M, and Tp are open to question, we shall present alternative hypotheses later in this chapter. There were no legal reserve requirements in Jordan during the period of our study; therefore, we cannot speak of "required reserves" and "excess reserves." However, it is postulated that deposit liabilities were the main influences on the demand for reserves. VI.9 Rd = 1 + dDp + pDg + tTp + qTg And in equilibrium; VI.10 Rd = Rs Rs is determined as in Chapter IV: VI.11 Rs = FA - Cp + LA - TD RS = RA = Cp + L 145 By definition# we have: VI.12 S = X - M + TR S = W - M and VI.13 FA = FA. n + S t-1 Also by definition: VI.14 MS = Cp + Dp Our behavioral equations are VI.6, VI.7, VI.8, and VI.9. The rest are definitional equations# with the exception of VI.10, which is an equilibrium equation. This model is different from money-supply models found in textbooks and academic journals# because it does not take into consideration reserve requirements# for these were nonexistent in Jordan during the period under study. Interest rates are also left out# because they seldom changed. Legal reserve requirements and interest rates are important in other models# but they can be ignored in the model of Jordanian money supply, a. Alternative Hypotheses The behavioral equations concerning Cp, M, Tp# and Rd can be expressed differently when alternative hypotheses concerning their behavior are postulated. 146 We may specify the Cp equation as a linear rela tion between Dp and Cp: Cp = a + nDp We may also, as a second alternative, postulate Cp to be a function of both Y and Dp: Cp = a + bY + nDp Similarly, imports may have a linear relationship with income, so that (as a first alternative) M = h + gY Imports may be affected by exports and transfers (W) as well as income, so that a second alternative hypothesis for imports is possible:^ M - h + gY1 + fW Still other import hypotheses are as follows: M = h - jDU M = h + fW M = h + gY1 + fW - jDU M = h - jDU + fW We now turn to alternative hypotheses concerning Tp. Tp can be postulated to have a direct linear 3 Where Y^ = Y - X, and W = X + TR. 147 relationship with Dp and Cp: Tp — i + kDp + sCp Tp may also be expressed as a function of Cp and Dp com bined (MS) , so that Tp = i + zMS And it may be affected by income as well: Tp = i + zMS + Y Tp - i + kDp + sCp + WY Our final alternative hypotheses concern Rd. These can be postulated as follows: Rd = 1 + d(Dp + Dg) + t (Tp + Tg) Rd = 1 + p(Dp + Tp) + q (Dg + Tg) By combining the deposits in this manner, we alleviate the multicollinearity problem. These, then, are the various alternative hypo theses for Cp, M, Tp, and Rd. They have been specifically postulated here so that we may perform tests leading to the most appropriate behavioral equations for the money- supply model for Jordan. b. Reduced-form Solutions The solutions are presented in reduced form below for clues as to whether they can provide a means 148 of distinguishing between the various hypotheses postu lated above. The model, as presented in its original form, has the following reduced-form solutions: VI.15 Cp = a + bY VI.16 M = h + gY - jDU VI.17 S=W-h-gY + jDU VI.18 PA = FAt-1 + W - h - gY + jDU VI.19 Dp = 1 (FAt-1 + W + L) - b+q Y - p Dg d+tk d+tk d+tk - q Tg + i DU - a+h+l+ni m+nk d+tk d+tk VI.20 Tp = k (FA. . + W + L) - bk+qk Y - pk DG d+tk d+tk d+tk - qk Tg + jk DU - ka+kh+kl-di d+tk d+tk d+tk VI.21 Rd « (FAt-l + w + L) “ (k+g)Y + jDU - (a + h) VI.22 Rs = (FA^j^ + W + L) - (b+g)Y + jDU - (a + h) The reduced form money-supply function is therefore: MS = f ^(FAt-1 + W + L) , Y, Dg, Tg, Duj The explanatory variables of the money supply will have the following signs: ^ + W + L) and DU positive signs; Y, Dg, and Tg, negative signs. The ex pression (FA^_2. + W + L) has a positive sign because as the foreign assets, transfers, exports, and local assets 149 of the CBJ increase, money supply can be expected to increase. Likewise, as the tariff rates (DU) increase, imports can be expected to decrease. However, as income increases, imports and the public's demand for currency will increase, causing bank reserves, and therefore de posits, to decrease. Thus, Y has a negative sign. The same is true for Dg and Tg: as these increase, the com mercial banks will want to hold more reserves, causing the private deposits to decrease. If we now introduce into the model the first al ternative hypothesis of Cp in place of the initial Cp behavioral equation, we read a new reduced-form solution for Cp. Nonetheless, the reduced-form solution of Dp is expressed by the same explanatory variables as in the original model, although the reduced-form coefficients are different. Likewise, if the second alternative for Cp is used, the reduced-form solutions for Cp, Dp, and MS will be expressed by the same exogenous variables, although their coefficients will be different. However, if we use the first alternative for M, the Cp, Dp, and MS in their reduced forms will be expressed by all the exogenous variables with the exception of DU, which drops 150 out. By using the various alternative hypotheses, we reach a number of reduced-form solutions of Cp, Dp, and MS — depending on the hypotheses used in the model. Some exogenous variables will be excluded, and some will either be aggregated or disaggregated. 4. The Model with Income as an Endogenous Variable We now treat income as an endogenous variable. To do this, we add one more equation for income to the original model. a. The Real Sphere Hypothesis We use a classical rather than a Keynesian money-income hypothesis, because interest rates were fixed and seldom changed throughout the period of our study, and also because of the absence of a capital mar ket in Jordan. The demand for money is dependent on the preferences to hold currency for transactions purposes and as a store of value. If the "store of value" or "asset" demand for money depends on interest rates, and if interest rates do not change, the amount of money demanded will tend to be constant. This effect reduces even a Keynesian money-demand equation to a constant- 151 velocity (or constant-marginal-velocity) case. For this reason, we use the simplified assumption of a con stant velocity. As the quantity of money increases, the increase would be spent on the purchase of commodities, thereby increasing income — with no effect on the inter est rates, since they are fixed. Our behavioral income equation will be Y = v(Cp + Dp) b. The Reduced-form Solutions The reduced-form solutions for Cp and Dp, when income is endogenous and added to the original model are: Cp = __________ bv_______ (FAt_1+W+L) + ________byj________ DU d-dvb+gv+bv+tk-tkvb d-dvb+gv+bv+tk-tkvb -___________bvp_____Dg - _______ bvq__________Tg d-d vb+g v+bv+tk-tkvb d-dvb+gv+bv+tk-tkvb -h+a+l+ti-vbh-vbl-vbti+gva . bv + a d-dvb+gv+bv+tk-tkvb 1-vb 1-vb Cp = ________ 1-vb_______ (FAt_1+W+L) + _________j (1-vb) DU d-d vb+gv+bv+tk-tkvb d-dvb+gv+bv+tk-tkvb - _______ p(l-vb) Dg - q (1-vb)____ Tg d-dvb+gv+bv+tk-tkvb d-dvb+gv+bv+tk-tkvb - h+a+l+ti+vbh-vbl-vbti+qva d-dvb+gv+bv+tk-tkvb The only difference between these reduced forms 152 of Cp and Dp and the previous ones is the disappearance of income as an explanatory variable in the determination of Cp and Dp, and hence of MS. The reduced-form money- supply function will therefore be: MS = f £(FAt The expression (FA^-j+W+J^ will have a positive sign; DU, Dg, and Tg will have negative signs, as demonstrated earlier. The reduced-form solutions of Cp and Dp are ex pressed by the same explanatory variables, regardless of which alternative hypotheses for Cp and Tp are ,sed. Most complications arise by using the alternative import hypotheses. In some equations, DU as an explanatory vari able is dropped out; in others, (FAt_1+W+L) is disaggre gated into (FA^-j+L) and (W) . The only change arising from using the alternative demand-for-reserves hypotheses is that Tg and Dg appear in the reduced form either as an aggregated variable or as separate variables. It is naturally noteworthy that when income is treated in the model as an exogenous variable in its ori ginal form, the reduced forms of Cp, S, and FA do not include all the exogenous variables, whereas in the sub- -1 + W + L) , DU, Dg, Tg 153 sequent models they do, whether income is endogenous or exogenous. 5. Implications of Alternative Hypotheses for Testing The money-supply tests cannot perform the job of discrimination when the reduced-form solutions of the alternative hypotheses have the same implications. If an alternative hypothesis does not introduce or eliminate exogenous variables, or does not suggest other crucial differences, then the first-stage tests of the two-stage, least-squares method cannot discriminate. On the other hand, if an alternative hypothesis introduces or eliminates exogenous variables or suggests other crucial differences, then the reduced-form solutions will be different and the tests can therefore discriminate. However, whether or not a first-stage test is effective, we can always per form a second-stage test on the alternative hypotheses. In the first stage, we apply the ordinary-least- squares estimation to the reduced forms of our system of equations, in the second stage, however, we use the estimated values of the first stage to estimate the coef ficients of the behavioral equations. Although the or dinary least-squares method is easier to apply, we use 154 the two-stage, least-squares technique to eliminate the biases of the ordinary-least-squares method. It is interesting to note that the only change in the money-supply reduced form when income is treated in the original model as endogenous instead of exogenous, is the elimination of income as an explanatory variable. It should be clear, however, that we are not testing endogenous versus exogenous income. When income is endo genous, we have a more complete model, and the tests are more careful tests of the various monetary hypotheses. We now turn to the test methods and to the empiri cal results of the estimation of the structural equations of the model. CHAPTER VII TESTS AND ESTIMATES TEST METHODS AND CRITERIA Economic hypotheses attempt to define relation ships existing among economic variables. Their purpose is to assist us toward a better understanding of the econ omic world, so that we may design economic policies that will accelerate economic welfare. But these hypotheses have to be checked against data obtained from the real world. If empirical data verify the relationship proposed by the hypothesis, we may accept the hypothesis; otherwise, we must reject it. To provide guidance for economic policymaking, we also need to measure quantitative relationships between econo mic variables. 1. Single-equation Models The relationships formulated in our model which 155 156 we wish to investigate will be subjected to empirical testing. This will be done both with a "single-equation, least-squares" method and with a "two-stage, least-squares" method. 1 The single-equation techniques must be scrutinized carefully, because they are subject to the criticism that they may yield biased results. For these techniques to have any validity for testing estimates, a one-way inter action of the explanatory variables on the dependent vari ables must exist. Frequently, relations between variables do not really operate in one direction only? there is interdependence and mutual interaction. Thus, the results obtained by using these techniques must be looked upon as preliminary. The single-equation least-squares technique is inappropriate for estimating the model and would not yield the best linear, unbiased estimates of the parameters for reasons discussed below. For the problems faced in these methods, see C. Christ, Econometric Models and Methods (New York: Wiley, 1966? A. S. Goldberger, Econometric Theory (New York: Wiley, 1964) ? J. Johnston, Econometric Methods (New York: McGraw-Hill, 1963)? L.R. Klein, An Introduction to Econo metrics (Englewood Cliffs, N.J.: Prentice-Hall, 1962)? and E. Malinvaud, Statistical Methods of Econometrics, (Chicago: Rand McNally, 1966). 157 The relationships postulated in our study can be considered to include the factors in the economy which most influence the money supply. They are not exact specifications of the observed facts of reality, because it is, of course, impossible to include all the relevant factors. For this reason, we include error terms to represent such unspecified influences, in addition to the variables mentioned in Chapter VI. 2. Error Terms To enable us to derive meaningful results from the statistical tests, the distribution of the error term, u^, must be specified. Under the least-squares method, a number of speci fic assumptions are required for satisfactory results. We shall list these assumptions because they will help in the interpretation of the estimates and will demon strate the danger of using this technique on interdepen dent relations. They will also provide a basis upon which to judge the best estimation procedure to employ. Least-squares regression yield the best linear unbiased estimates of the parameters of a model only if certain assumptions concerning the distribution of the 158 2 random term, u^, are satisfied in each equation. These assumptions are as follows: 1. The observations on the independent variables in the regression equation are either fixed numbers (non stochastic constants) or random variables distributed independently of the error terms. 2. The error term is distributed independently with zero expectation — that is, E (uT) =0. 3. The distribution of the error term has a con stant finite variance (homoscedasticity), and successive values of this term are pairwise uncorrelated — that is, 2 E(uu') = G * I, where I is the unit matrix. Inclusion of the error term in the mo^lel and specification of these assumptions about its distribution serve to define the distribution of each of the dependent variables corresponding to each set of values of the ex planatory variables. If all of the assumptions are satisfied, we can expect to derive consistent estimates of the parameters by the least-squares method. However, violation of the 2 Malinvaud, op. cit., pp. 75-78. 159 assumptions can introduce problems. For example, multi- collinearity may arise in studies using time series of the "levels" of the variables.^ This problem appears when some of the explanatory variables are so highly cor related that it becomes difficult to measure their separate influence or even to estimate their relative effects. Because the variance of the estimation is larger, the significance or efficiency of the estimation is lost. When we find multicollinearity between two explanatory variables, we must either omit one variable from the re lation or replace the two of them with a variable which is a combination of both. In the light of these assumptions, it is obvious that the single-equation, least-sqv.ares technique would be inappropriate to the estimation of the model and would not yield the best linear unbiased estimates of the para meters. For, even if the other assumptions were satis fied, our placing the relationships within the framework of a simultaneous-equation system violates assumption (1) . ^D. E. Farrar and R. R. Glauber, "Multicollin earity in Regression Analysis: The Problem Revisited," Review of Economics and Statistics, Vol. 49 (1967) , pp. 92-107. 160 That is, some of the explanatory variables in the model are not distributed independently of the error terms in the equations as they appear. 3. Two-Stage Least-Squares Models To minimize bias in the estimates, therefore, we must employ an estimating technique that explicitly con siders the simultaneous nature of the relationships in cluded within the model. One such technique which cor rects for the interaction of the explanatory variables and the error term in each equation, and which, in effect, purges these dependent variables of the stochastic com ponent associated with the error term, is the method of 4 two-stage least-squares. This is not the only possible method of solving the problems caused by the simultaneous nature of the relationships but it is at present the most practical one. 4. Economic Models and Problems of Specification In building the equations, we have been careful to make the model a good description of the real system. 4 Johnston, op. cit., p. 236. 161 We must ask ourselves whether the model includes all the explanatory variables that are significant in explaining the values, or the change of values, of the dependent variables. Whenever one specifies economic relationships by mathematical formulation, one encounters uncertainties about which variables should be included in the equation, what form they should assume, and whether single equations may be analyzed independently. If we fail to include some of the significant explanatory variables, the error term u may not be random. So, if we assume the relationship of an explanatory variable to be linear when in fact it is not, we would get an incorrect indication from the correlation coefficient. The problem with fitting linear form to a nonlinear relationship is that it may cause us to reject a correct hypothesis. 5. Statistical Criteria In our computations, we shall employ the single equation, least-squares and the two-stage, least-squares techniques. In evaluating the resulting estimates, we shall employ certain statistical criteria, including the coefficient of correlation (R) and the standard error of the coefficient. To test for the presence of autocorrelated dis 5 turbances, we shall use the Durban-Watson statistic, which is given by Although exact significance levels are not available, tables have been prepared which give the upper and lower bounds (d and d ) of the statistic for several observa- n L tions and the number of independent variables. Table 17 shows the values of d and d at the 5 percent signi- L u ficance level (for the test against the single alterna tive of positive autoregression) as a function of m* (the total number of exogenous variables, excluding the con stant dummy variable) and of n (the number of observa- 6 tions). Correlation in Least Squares Regression," Biometrika. December 1950 and June 1951. n 2 5J. Durban and G. S. Watson, "Testing for Serial ^Malinvaud, op. cit., p. 424 163 TABLE 17 DISTRIBUTION OF THE DURBAN-WATSON TEST n m1=1 m'=2 m 1=3 m'=4 m 1 =5 d d d d.. dT d d d L u l u L u l u dT d L u 15 1.08 1.36 0.95 1.54 0.82 1.75 0.69 1.97 0.56 2.21 To use this table, we compare the obtained value of d with d_ and d . If d<dTl the probability of a L i U - * - < value as small as d is less than ©C- We reject the hypo- ] j thesis of independence. If d > du, the probability of a I value as small as d is greater than oC. . We do not reject ; the hypothesis of independence if d^d^. The tables j leave us in some doubt, and the test is inconclusive. Malinvaud states: j It is possible that the hypothesis of indepen dence should be rejected at the significance level. But this could only be ascertained by studying the distribution of d with the sequences of values actually taken by the exo genous variables Zjt. In practice we are generally content in this case to state that the value of d lies in the indeterminant zone of the tables. n Malinvaud, op. cit. 1 164 The Durban Watson test is also applicable to the hypothesis of independence, against the alternative of negative correlation. We ask if the value of d is excep tionally large in the case of independence — that is, if the value of (4-d) is exceptionally small. To test in dependence, we need only to look at the position of (4-d) relative to d„ and d . L u 6. Data Employed In our study, we have dealt with broad, aggregate economic classes. If we were to choose different repre sentative indexes, we would obtain different quantitative results. There are many types of monies and near-monies, and they are held for many different reasons. The re serves of some commercial banks may have a far different effect on the money stock from the reserves of other com mercial banks. However, we cannot allow for these fac tors, lest we become lost in the web of interrelations. The available statistical data are reasonable satisfac tory for testing our equations. We employ annual figures for variables. All the "JD valued" variables are mea sured in millions of Jordan dinars. The measure employed in our estimates is the level of the variable. The data 165 are presented in the Appendix. The money stock in Jordan is defined in the nar row sense to include both currency in the hands of the public (but not currency in bank vaults) and the demand deposits of the public. The public, as we use the term, includes all holders of Jordanian currency or deposits located in Jordan, with the exception of the commercial banks and the government. It includes individuals, corporations, partnerships, municipalities, and credit agencies. Non bank foreigners who are holders of Jordanian currency and deposits are also included in the estimate of the money stock. (In principle, of course, the holdings of foreigners should be excluded, but data are not available for this.) The money stock is the money stock at the end of the year (December 31) . We shall choose 1951 as the starting point for our series. (Jordan, as known today, originated in 1949.) TEST RESULTS We shall now present the empirical results of the estimation of the structural equations and their 166 reduced-form solutions. The purpose of this section is to make decisions concerning the alternative hypotheses under question — namely (1) the demand for currency by the public, (2) the demand for imports, (3) the demand for reserves by the commercial banks, and (4) the demand for time deposits by the public. The introduction of alternative hypotheses in the model leads to different reduced-form solutions. Many tests become necessary for choosing among the alternative hypotheses. However, most complications are due to the import equations, so we shall turn first to the import hypothesis. 1. The Import Hypothesis To make decisions concerning the import equations, we shall first look at the model with income exogenous, and then with income endogenous. When income is treated as an exogenous variable, the reduced-form solutions for import reached by using the various import hypotheses are the same as the structural equations for imports. The results obtained with the use of the ordinary-least-squares method can therefore help us in selecting the import hypothesis to be used in the final model. In certain 167 equations, Y^ is used in place of Y; it denotes income when exports have been excluded from Y. We find it mean ingful not to include exports twice in the same equation, since exports are already a part of W. Table 18 presents the results of the tests. Y in equations (1) and (4) has a meaningful coef ficient (very small standard error); in equations (5) and (6), Y^ has a large standard error relative to its coefficient. In equations (3) , (5), (6) , and (7) W has significant coefficients; however, the introduction of W in equations (5) and (6) does not improve R by much. The results of equations (1) and (3) seem to be the most successful from the standpoint of the stability of the 2 coefficients and the value of R . When DU is used in 2 equation (2) we get an extremely low value of R and a sign that is not expected from our postulated theory. The addition of DU to equation (4) leads only to a more 2 complicated model and does not improve R . In both equa tions (6) and (7), the coefficient of DU is smaller than its standard error. We can conclude that DU does not help us in any way, and only complicates our model. Therefore, the two most useful equations are (1) and (3). 168 TABLE 18 STRUCTURAL EQUATIONS OF IMPORTS WITH INCOME EXOGENOUS 1. M = -0.3049 + 0.39117 Y (0.02497) (DW = 2.44790; R2 = 0.95) 2. M = -16.7934 + 4.12985 DU (1.63467) (DW = 0.36789; R2 == 0.26) 3. M = 2.7341 + 0.85880 W (0.05030) (DW = 2.61640; R2 = 0.95) 4. M = 6.9052 + 0.41553 Y - 0.70483 DU (0.03177) (0.58306) (DW = 2.48541; R2 = 0.94) 5. M = 0.6131 + 0.15951 + 0.58322 W (DW ) (0.12595) (0.22311) - 2.59119; R2 = 0.95) 6. M = 5.2911 + 0.18629 Y ± + 0.57053 W - 0.47070 DU (0.13067) (0.22555) (0.53554) (DW = 2.67462; R2 = 0.95) 7. M = 5.8641 = 0.29266 DU + 0.87968 W (0.54107) (0.06446) (DW = 2.71885; R2 = 0.95) 169 When income is endogenous, the use of the alter native import hypotheses leads us to a number of reduced- form solutions for M, as shown in Table 19. Import hypo thesis (1) of Table 18 implies reduced forms (3) and (4) of Table 19? hypotheses (3) and (5) of Table 18 imply reduced forms (5) and (6) of Table 19; and hypotheses (2) and (4) imply reduced forms (1) and (2). The reduced- form equations (7) and (8) are implied by import hypo- Q theses (6) and {7) of Table 18. We can see that the reduced forms implied by each of the alternative import hypotheses seem to give good results. However, although W and DU are significant, they lead to a more complicated model and do not signi- 2 ficantly improve R For example, the reduced forms implied by hypotheses (3) and (5) complicate the model and decrease the degrees of freedom. The same is true for hypotheses (2) and (4). Likewise, for hypotheses (6) and (7), we reach a complicated model and reduce our de- 2 grees of freedom without significantly improving R . O The hypotheses concerning Rd and Cp also cause differences in the reduced-form equations; they will be discussed later in this chapter. TABLE 19 REDUCED-FORM SOLUTIONS FOR IMPORTS WITH INCOME ENDOGENOUS 1. M = 13.4925 + 0.69196(W+L+FAt_1) - 0.45908Dg - 0.76895Tg - 1.2468DU DW = 1.37 (0.02387) (0.32776) (0.04486) (0.30754) R2 = 0.99 2. M = 14.9696 + 0.69477(W+L+FAt_x) - 0.72829 (Tg+Dg) - 1.31141 Du DW = 1.54 (0.02349) (0.13669) (0.29686) R2 = 0.97 3. M = -2.0143 + 0.61213(W+L+FAt-1) + 0.24194 Dg - 0.39037 Tg DW = 1.68 (0.02041) “ (0.42103) (0.16747) R1 = 0.98 4. M — -0.5632 + 0.60926(W+L+FAt„i) " 0.26033(Tg+Dg) DW = 2.11 (0.02073) (0.13451) R2 = 0.98 5. M = -3.5188 + 0.92908(L+FAt-1) + 0.1345lDg - 0.37853Tg + 0.46238W DW = 1.44 (0.14137) ~ (0.36651) (0.14465) (0.06858) R2 = 0.98 6. M = -2.4523 + 0.94660 (L+FA^) - 0.27379 (Tg+Dg) + 0.45073 W DW - 1.74 (0.14275) ~ (0.11551) (0.06890) R = 0.98 7. M = 10.2979 + 0.83995(L+FAt-1) - 0.40346 Dg + 0.70356 Tg DW = 1.13 (0.10735) ~ (0.31639) (0.14628) R2 = 0.99 + 0.60349 W - 1.05124 Du (0.06667) (0.32555) 8. M = * 11.6840 + 0.84470(L+FAt.x)-0.66331(Tg+Dg)+0.60510W-1.11117DU D^ = 1.27 (0.10643) (0.13848) (0.06615)(0.31653) R = 0.99 -j o 171 Therefore, our hypothesis (1) which postulates that im ports are specified by income and implies reduced form equations (3) and (4), seems to be the most appropriate to our model. Xn equation (4), the variable (Tg+Dg) has a negative sign; this is expected, as shown in the solution in Chapter VI. As (Tg+Dg) increases, the com mercial banks' demand for reserves increases, because the government's deposits are very volatile. This brings about a reduction in the demand-deposit expansion of the commercial banks, and thus causes the money supply to contract. A decrease in the money supply has a negative effect on income, which in turn causes imports to decline. Equation (1) from Table 18 and Equation (4) from Table 19 come from the same model, as noted earlier. The evidence from this model, whether income is endogenous or exogenous, leads us to believe that imports are best explained by income alone. For additional evidence of this hypothesis, we look at the reduced-form solutions of both money supply and income for the presence of signi ficant equations when this import hypothesis is used in the model. Tables 20 and 21 present the empirical results of the estimation of the reduced forms. c TABLE 20 1. MS = 2. MS = 3. MS - 4. MS = 5. MS - 6. MS = 7. .MS = 8. MS = REDUCED-FORM SOLUTIONS FOR MONEY SUPPLY WITH INCOME ENDOGENOUS -9.1735 + 0.40504(W+L+FAfc_1) - 0.53065 Dg - 0.28366Tg + 0.82212DU (0.02065) (0.28356) (0.12533) (0.26607) -10.3508 + 0.40280(W+L+FAt-1) - 0.31606(Tg+Dg) + 0.87367DU (0.02022) (0.11763) (0.25548) 1.0511 + 0.45767 (W+L+FA^_-L) - 0.99288 Dg - 0.53327 Tg (0.01528) ” (0.31522) (0.12538) -0.0036 + 0.45967(W+L+FAt_1) - 0.62780 (Tg+Dg) (0.01547) (0.10037) 2.1536 + 0.22542(L+FAt_x) -0.91422Dg -0.54195Tg + 0.56740 W (0.10687) ” (0.27707) (0.10935) (0.05184) 1.3799 + 0.21271(L+FAt_x) - 0.61794(Tg+Dg) + 0.57586 W (0.10744) (0.08694) (0.05186) (0.09348) (0.27550) (0.12738) (0.05805)(0.28348) -7.6045 + 0.27747(L+FAt_L) - 0.37038(Tg+Dg) + 0.47774W +0.70620DU (0.09208) (0.11981) (0.04724) (0.27386) DW = 1.57 R2 = 0.99 DW = 1.64 R 0.99 DW = 2.11 R2 S3 0.98 DW = 2.46 DW — 2.33 R2 — 0.98 DW 2.51 R2 s s 0.98 DW z z 1.80 R2 0.99 DW 25 1.81 H* -J to TABLE 21 REDUCED-FORM SOLUTIONS FOR INCOME 1. Y = -21.4792 + 1.36111(W+L+FA ) -0.19019Dg + 0.2261lTg + 2.13823DU (0.13406) (1.84063) (0.81352) (1.72707) 2. Y = -23.4637 + 1.35733(W+L+FAt-1) + 0.17149(Tg+Dg) + 2.22512 Du (0.12753) (0.74209) (1.61172) 3. Y = 5.1138 + 1.49801(W+L+FAt_1) - 1.39239Dg - 0.42312Tg (0.07746) (0.59799) (0.63563) 4. Y = 2.8894 + 1.50240(W+L+FAt-1) - 0.62246(Tg+Dg) (0.07474) (0.48509) 5. Y = 7.7217 +0.94859(L+FA^) -1.20633Dg -0.44364Tg +1.75759 W (0.62744) (1.62664) (0.64199) (0.30435) 6. Y = 6.13565 + 0.92255 (L+FA^) - 0.59932 (Tg+Dg) + 1.77491 W (0.60153) (0.48672) (0.29033) 7. Y = -15.8580 +1.10070(L+FAt_1)-0.28806Dg+0.11106Tg+1.51659W+1.79405Du (0.65475) (0.92970) (1.89220) (0.40662)(1.98556) 8. Y = -17,7015 +1.09438(L+FAt_i)+0.05753(Tg+Dg) +1.51458W +1.87376DU (0.62479) (1.92970) (0.38835) (1.85814) DW = 2.36 R2 = s 0.97 DW 2.33 R2 ss 0.97 DW = 2.45 R2 0.96 DW 2.46 R2 0.97 DW 2.72 R2 = 0.96 DW 2.70 R2 = 0.97 DW = 2.52 R2 = 0.96 DW — 2.49 R2 — 0.97 H -J 174 Although in Table 20, all of the money-supply equations have significant results, the addition of vari ables beyond that in reduced forms (3) and (4) implied by the import hypothesis (imports as a function of income) does not provide much additional explanatory power. The model becomes more complicated and there is a reduction in the degrees of freedom. In Table 21, the only equation with all coeffi cients significant is (4). The signs are as expected, and the D-W statistic indicates that the hypothesis of independence of residuals is not rejected. Equation (4) is implied by the same import hypothesis (imports as a function of income). We shall therefore use this import hypothesis in the model, and in the tests for the Cp, Rd, and Tp equations. 2. The Currency-demand Hypothesis We now proceed to find the demand-for-currency (Cp) hypothesis that best specifies the behavior of the public in Jordan. By introducing the various Cp hypo theses in the model, we get only three different reduced- form solutions for Cp when income is exogenous, and two when incou’ /a is endogenous (Table 22). When Cp is a 175 TABLE 22 REDUCED-FORM EQUATIONS FOR Cp 1. Cp = 1.7371 + 0.13869 Y (0.0074) R2 = 0.958 DW = 1.017 2. Cp = 2.0762 + 0.16888(W+L+FA. .) + 0.04041 Y (0.04470) (0.03633) - 0.19014 (Tg+Dg) (0.06146) R2 = 0.974 DW = 1.31 3. Cp = 2.0403 + 0.16858 (W+L+FAfc_1) + 0.04071 Y (0.04693) " (0.03826) - 0.17832 Dg - 0.19307 Tg (0.20465) (0.08027) R2 = 0.972 DW = 1.32 4. Cp = 2.3540 + 0.21752(W+L+FAt_!) - 0.20386 (Tg+Dg) (0.00936) (0.06076) r 2 = 0.974 DW = 1.758 5. Cp = 2.3963 + 0.21743(W+L+FAt_!) - 0.21853 Dg (0.00980) (0.20224) - 0.20006 Tg (0.08045) R2 = 0.972 DW = 1.74 176 function of both variables Dp and Y, we get the reduced forms (2) and (3); and, under the hypothesis that Cp is a function of Dp, when income is endogenous, we have the reduced forms (4) and (5). From the reduced-form equa tions below, we can see that may well be ignored, as it complicates the model and reduces the degrees of free- dom without improving R significantly. If we now look at the second-stage tests of the Cp hypotheses, when income is endogenous and exogenous, we get the results presented in Table 23. When income is exogenous and (Tg+Dg) is used as one variable, we have equation (1), for the case where Cp is a function of Dp only. Under the same conditions but with Cp as a function of both Dp and Y, we get equation (2). When income is again treated as an exogenous variable but Dg and Tg are used as separate variables, we get equation (3) if Cp is a function of Dp, and equation (4) is Cp is a function of both Cp and Y. The results in equation (5) are obtained when income is endogenous, and (Tg+Dg) is used as one variable, for the case where Cp is a function of Cp. Equation (6) presents the results under the same conditions but with 177 TABLE 23 SECOND-STAGE Cp ESTIMATES 1. Cp = 5.6586 + 0.90678 Dp (0.04752) 2 R = 0.96 DW = 1.69 2. Cp = 3.5624 + 0.47353 Dp + 0.06841 Y (0.15849) (0.02426) R2 = 0.97 DW = 1.29 3. Cp = 5.7092 + 0.90239 Dp (.05041) R2 =0.955 DW = 1.27 4. Cp = 3.4215 + 0.43652 Dp + 0.073894 Y (0.15794) (0.02463) R2 = 0.97 DW = 1.08 5. Cp = 5.6611 + 0.90656 Dp (0.04888) R2 =0.96 DW = 1.86 6. Cp = 2.9281 + 0.35151 Dp + 0.8810 Y (0.18407) (0.02857) R2 = 0.974 DW = 1.76 7. Cp = 1.4509 + 0.14145 Y (0.00652) R2 = 0.97 DW = 1.26 8. Cp = 5.7142 + 0.90196 Dp (0.05156) R2 = 0.95 DW = 1.38 178 TABLE 23 (continued) 9. Cp = 2.779 + 0.31269 Dp + 0.09386 Y (0.18510) (0.02880) 2 R = 0.97 DW = 1.49 10. Cp = 1.45888 + 0.14138 Y (0.00657) R2 = 0.97 DW = 1.23 Cp as a function of both Dp and Y, and equation (7) with Cp as a function of income. When income is again endo genous and Dg and Tg are used as separate variables, we get equation (8) for the case where Cp is a function of Dp; equation (9) where Cp is a function of both Dp and Y; and equation (10) where Cp is a function of income only. Prom these results, we may conclude that the three Cp hypotheses perform equally well, and that the Dp and Y coefficients are meaningful. However, we notice that by including both Dp and Y as explanatory variables in one equation, we get no better results than by using Y or Dp alone. Although both Dp and Y used alone can satisfactorily specify Cp, we believe that Dp is a better choice than Y as the explanatory variable. The reason is that Dp statistics are more dependable. They are care fully collected each month from the nine commercial banks operating in Jordan. The income statistics, in contrast, are estimated once a year and come from thousands of sources. 180 3. The Reserves-Demand Hypothesis The second-stage estimates of the various reserves-demand (Rd) hypotheses are presented in Table 24. The first-stage equations use the preferred import and currency equations discussed above. When income is exo genous we get the estimates in equations (1), (2) , and (3). The estimates in equations (4) and (5) refer to the RD hypotheses when income is endogenous. We obtain the most significant coefficients for the Rd equations, whether income is exogenous or endo genous, when Tg and Dg are combined into a single variable, and Tp and Dp are combined into another. The coefficients in equations (1) and (4) are superior to those of (2), (3), and (5) with respect to the standard error of the coefficients. Very bad results (standard error larger than the value of the coefficients) are obtained in equation (2). Multicollinearity was probably the reason for the results here; poor results were also obtained with ordinary least squares, and further work with this hypothesis was dropped. The total deposits also explained well, but it was clear that disaggregation into two classes of deposits explained best. The only question left was whether to use total 181 TABLE 24 SECOND-STAGE Rd ESTIMATES 1. Rd = 1.3918 + 0.76892 (Tg+Dg) + 0.36670 (Tp+Dp) (0.090505) (0.03305) R2 = 0.93 DW = 2.54 2. Rd « -1.51 + 1.22928 Dp + 0.55402 Dg + 0.78828 Tg (2.11187) (0.43422) (0.19862) - 0.55686 Tp (2.23816) R2 = 0.93 DW = 2.04 3. Rd = 6.2080 + 0.90503 (Tg+Tp) - 0.13792 (Dp+Dg) (0.13172) (0.14922) 2 R = 0.93 DW = 1.90 4. Rd = 1.3848 + 0.76885 (Tg+Dg) + 0.36711 (Tp+Dp) (0.09524) (0.03316) 2 R = 0.93 DW = 2.50 5. Rd = 6.0976 + 0.89496 (Tp+Tg) - 0.12720 (Dp+Dg) (0.13116) (0.14889) R2 = 0.93 DW = 1.80 182 government deposits and total private deposits, or total time deposits and total demand deposits. The results in Table 24 suggest the former. Let us not look at the reduced-form solutions of Rd (Table 25). When Tg and Dg are combined into a single variable, we have reduced-form equation (1) if income is exogenous, and equation (3) if income is endo genous. However, when Tg and Dg are separate explanatory variables (or when we use total time deposits and total demand deposits), we obtain equation (2) if income is exo genous, and equation (4) if income is endogenous. We notice that by combining Tp and Dp into one variable, and Tg and Dg into another, we get highly successful results. The results obtained by the combined variable (Tg+Dg) far surpass those obtained when the terms are used as separate explanatory variables. In equations (2) and (4) of Table 25, we even get the wrong sign for Dg; moreover, the standard error is larger than the Dg coefficient. Further evidence can be obtained by looking at the reduced-form solutions of money supply, income, and inports. All the tests clearly indicate that superior TABLE 25 REDUCED-FORM SOLUTIONS FOR Rd X. Rd = -1.7848 + 0.14808(W+L+FAt.i) + 0.02009Y. + 0.46444(Tg+Dg) DW = 2.52 (0.07771) (0.06316) (0.10685) R2 = 0.92 2. Rd = -0.1935 + 0.16108 (W+IH-FA^) + 0.006824Y. -0.05932Dg +0.49432Tg DW = 2.04 (0.07214) ” - (0.05881) (0.31460) (0.12340) R2 = 0.93 3. Rd = -1.6467 + 0.17226{W+L+FA^) + 0: .45758 (Tg+Dg) DW « 2.50 (0.01656) (0.10099) R2 = 0.93 4. Rd = -0.1338 + 0.16927(W+L+FAt-1) - 0.06606 Dg + 0.59315 Tg DW = 2.03 (0.01436) “ (0.29621) (0.11782) R2 = 0.94 CD OJ 184 results are obtained by combining Tg and Dg into one explanatory variable. Additional support can also be obtained by inves tigating the results obtained in earlier tables, while testing for the import and demand-for-currency hypotheses. Still further evidence car be had from the results of the ordinary-least-squares estimates of the Rd behavioral equations. The latter results are presented in Table 26. It is clear that we get superior results by the combined variable (Dg and Tg). We therefore believe that Rd is best specified by the two explanatory variables (Tg+Dg) and (Tp+Dp). By expressing our function in this form, we also lessen the problem of multicollinearity. 4. The Demand-for-Time-Deposits Hypothesis The second-stage tests of the demand-for-time- deposits (Tp) hypotheses give the results presented in Table 27. The first five equations present the Tp es timates when income is exogenous, and the rest when income is endogenous. The best results are obtained when time deposits are either a function of Dp alone or MS alone. However, since Dp is a component of MS, and both explain'equally 185 TABLE 26 ORDINARY-LEAST-SQUARES ESTIMATES OF Rd 1. Rd = 5.1241 - 0.26818 Dp + 1.03213 Tp + 0.34130 Dg (0.36070) (0.38125) (0.22337) + 0.90147 Tg (0.08612) R2 = 0.9687 DW = 1.745 2. Rd = 1.3249 + 0.76822(Tg+Dg) + 0.37060 (Tp+Dp) (0.07062) (0.02425) R2 = 0.96 DW = 1.92 3. Rd = 6.1460 - 0.13605 (Dp+Dg) + 0.90277 (Tp+Tg) (0.11173) (0.09853) R2 = 0.959 DW = 1.338 186 TABLE 27 SECOND-STAGE Tp ESTIMATES 1. Tp = -3.9293 + 0.93787 Dp (0.04460) 2 R = 0.967 DW = 1.625 2. Tp = -2.9995 + 1.08687 Dp - 0.16431 Cp (0.35346) (0.38660) R2 = 0.965 DW « 1.885 3. Tp = 0.0168 + 1.44825 Dp - 1.16145 Cp + 0.08571 Y (0.40445) (0.72618) (0.05392) R2 “ 0.97 DW = 2.49 4. Tp = -6.6422 + 0.48932 MS (0.02562) 2 R = 0.96 DW = 0.904 5. Tp = -6.5000 + 0.53287 MS - 0.01299 Y (0.13573) (0.03970) R2 = 0.96 DW = 1.02 6. Tp = -3.9184 + 0.93693 Dp (0.04715) R2 = 0.96 DW ~ 1.56 7. Tp = -2.9365 + 1.09417 Dp + 0.17344 Cp (0.37381) (0.40884) 2 R = 0.96 DW = 1.80 8. Tp = -6.6293 + 0.48885 MS (0.02685) R2 = 0.96 DW = 1.80 187 9. Tp = R TABLE 27 (continued) -5.6829 + 0.76448 MS - 0.08264 Y (0.17171) (0.05091) 2 = 0.96 DW = 1.80 well, it appears preferable to use Dp. When we look at the ordinary-least-squares estimates of the alternative Tp hypotheses. Dp also explains well. The results are presented in Table 28. The Dp coefficient in Equation (1) is very significant. 18S TABLE 28 ORDINARY-LEAST-SQUARES ESTIMATES OF Tp 1. Tp = -3.8607 + 0.93193 (0.03118) R2 = 0.98 2. Tp = -4.7321 + 0.80296 (0.12193) 2 R = 0.98 3. Tp = -4.7359 + 0.80555 (0.13032) R2 = 0.98 4. Tp = -6.6664 + 0.49019 (0.01938) 2 R = 0.98 5. Tp = -6.3554 + 0.57752 (0.09233) R2 = 0.977 Dp DW = 1.88 Dp + 0.14636 Cp (0.13384) DW « 1.67 Dp + 0.15937 Cp - 0.00228 Y (0.20416) (0.02610) DW - 1.90 MS DW = 1.22 MS - 0.02629 Y (0.02717) DW = 1.55 CHAPTER VIII SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS SUMMARY The purpose of this study, as stated in Chapter I, has been to formulate an empirical money-supply function within the institutional framework of the Jordanian mone tary and banking system. To achieve this, a model has been developed. It is hoped that this model will shed light on various monetary issues in Jordan. Through the specification of commercial banks' demand for reserves, we sought to gain an insight into the influence of bank \ behavior upon the determination of the money stock. Also, through the specification of public demand for currency, time deposits, and imports, we sought insights into the influence of the public upon the determination of the money stock. We started by presenting an accounting framework of the monetary and banking system in Jordan, and isola ting the main determinants of the money stock. We then 190 191 ; t proceeded to build a complete model by introducing ad- i i j ditional behavioral equations and identities. To derive the money-supply function, we solved the structural equations of the model simultaneously and obtained reduced-form solutions for the money supply. The solution indicated that the Jordanian money supply is a function of lagged foreign assets, exports, trans fers from abroad in the form of private remittances and foreign aid to the government, the local assets of the Central Bank of Jordan, and the demand and time deposits of the Treasury. The complete model was estimated by a two-stage least-squares technique, and the results were compared with those derived by estimating each of the structural equations by a single-equation, ordinary-least-squares technique. i i CONCLUSIONS In Chapter VII, we investigated various alterna- * i tive hypotheses concerning public demand for imports, currency, and time deposits, and commercial banks' demand: for reserves. These alternative hypotheses led to dif ferent solutions, with most complications arising from 192 the import equations. We tested the solutions and struc tural equations, so that decisions could be made among the alternative hypotheses to be used in the final model. We concluded that imports were best specified by income, and that the demand for currency and time deposits by the public was best explained by the demand deposits of the public. The demand for reserves was best specified by the demand and time deposits of the public and of the Treasury. The best results were obtained by combining the demand and time deposits of the public into a single explanatory variable, and the demand and time deposits of the Treasury into a second such variable. The resulting relationships are summarized below, where income is treated as an exogenous variable. They are estimated by the two-stage, least-squares technique, j except for the imports equation, which is already a reduced-form equation. 1. Cp = 5.6886 + 0.90678 Dp (0.04752) 2 R = 0.96 DW = 1.69 2. M = -0.3049 + 0.39117 Y (0.02497) 8 R2 = 0.94 DW = 2.45 193 3. Tp = -3.9293 + 0.93787 Dp (0.04460) R2 = 0.97 DW = 1.62 4. Rd = 1.3918 + 0.76892 (Tg+Dg) + 0.36670 (Tp+Dp) (0.09505) (0.03305) 2 R = 0.93 DW = 2.54 5. Rd = Rs 6. Rs = FA + L - Cp 7. FA = FAt-1 + S 8. S = W - M 9. MS = Cp + Dp The reduced-form equations of the model, with income exogenous, are as follows: Cp = 2.0762 + 0.16888(W+L+PAt_1) + 0.04041 Y. (0.04470) (0.03633) - 0.19014 (Tg+Dg) (0.06146) R2 = 0.974 DW = 1.31 Rd = -1.7848 + 0.14808(W+L+FA ) + 0.02009 Y (0.07771) t_1 (0.06316) + 0.46440 (Tg+Dg) (0.10685) R2 = 0.924 DW = 2.52 194 . ® Tp = -6.3255 + 0.15709(W+L+FA ) + 0.05836 (0.05544) (0.04506) - 0.40869 (Tg+Dg) (0.07623) 2 R = 0.96 DW = 2.02 Dp = -2.6250 + 0.19541(W+L+FA ) + 0.03891 Y. (0.05731) 1 (0.04658) - 0.41073 (Tg+Dg) (0.07623) R2 = 0.965 DW = 2.098 MS = -0.5488 + 0.36429(W+L+FA ) + 0.07932 Y, (0.07226) 1 (0.05873) - 0.60087 (Tg+Dg) (0.09935) R2 = 0.98) DW = 2.497 S = -0.6131 + 0.41678 W - 0.15951 Y (0.22311) (0.12595) 1 R2 = 0.34 DW — 2.59 When income is treated as an endogenous variable, we derive the following two-stage, least-squares esti mates of the equations. 1. Cp = 5.6611 + 0.90656 Dp (0.04888) R2 - 0.96 DW •= 1.86 2. M = -1.7686 + 0.40529 Y (0.01277) R2 = 0.99 DW = 2.15 ' ' , L . A 't 3. Tp = -3.9184 + 0.93693 Dp (0.04715) ; 2 R = 0.96 DW = 1.56 j 4. Rd - 1.3848 + 0.76885(Tg+Dg) + 0.36711(Tp+Dp) i (0.09524) (0.03316) l ! R2 = 0.93 DW = 2.50 The test results of the reduced-form solutions, with income endogenous, are presented below: Cp = 2.3540 + 0.21752(W+L+FA ) - 0.20386 (Tg+Dg) (0.00936) (0.06076) 2 R = 0.97 DW = 1.757 M = -0.5632 + 0.60926 (W+L+FA ) - 0.26033 (Tg+Dg) (0.02073) (0.13451) R2 = 0.98 DW = 2.11 Rd “ -1.6467 + 0.17226(W+L+FA ) + 0.45758 (Tg+Dg) (0.01556) (0.10099) R2 = 0.93 DW = 2.50 Tp = -5.9244 + 0.22732(W+L+FA .) - 0.42850 (Tg+Eg) (0.01180) “ (0.07660) R2 = 0.96 DW = 1.80 MS = -0.0036 + 0.45976(W+L+FA. ,) - 0.62780 (Tg+Dg) (0.01547) (0.10037) R2 = 0.98 DW = 2.46 . j V.V i Y = 2.8894 + 1.50240(W+L+FA ,) - 0.62246 (Tg+Dg) (0.07474) (0.48509) '^;ivv2' , . i 196 S = -3.2464 + 0.07103(W+L+FA .) + 0.23318 (Tg+Dg) (0.03932) (0.25521) | R2 = 0.21 DW = 2.12 r I We see that the reduced-form estimates of the i model for the various endogenous variables — whether income is treated as exogenous or endogenous — are very similar. We also see that the estimated values j 1 of the coefficients in the Cp, M, Rd, and Tp functions i are similar, whether income is endogenous or exogenous? so that it does not matter greatly whether income is treated as exogenous or endogenous in the model. However, ! simplicity (and distrust of the income statistics) i j would suggest reliance on the model in the case where I I income is endogenous. j Finally, we observe that we have a stable money- | supply function dependent on very few variables. These i j i variables are basically the lagged foreign assets, the i i | transfers and exports, the local assets of the CBJ and i I I the Treasury deposits. It is these variables that exert { control over the money stock. Although the CBJ cannot I ; j j control exports, transfers, and the volume of the foreign i 1 ! ; assets of Jordan, nonetheless it can control local assets, ( _ __ . . . ___ u u t i " 197 1 which have the same effect on the money supply as the I other explanatory variables. It is therefore capable I ; of controlling the money stock. Tariff rates can be ! i excluded from the model, since exports and transfers i j ; seem to be better predictors of the money supply. RECOMMENDATIONS j it was not the behavior of the monetary authori- i ties alone that controlled the money supply in Jordan I during the period of our study, 1951-1965. The money i supply was to some degree influenced by the behavior of both the commercial banks and the public. This was very : unfortunate, because of the unique importance of the I j I money supply in a country's economy. Many economists . contend that the money supply is the primary cause of changes in the price level and the volume of the national | income. Monetary theory emphasizes the channels through i j which the policies of the monetary authorities are trans- i I 1 mitted in the economy. For the money-supply theorist, i the money supply is the main channel of transmission. Under the Jordan Currency Board legislation, the f monetary authorities had no way of influencing the money i ! j stock. The establishment of the CBJ should have given 198 the monetary authorities control over the money supply, but this did not take place. The reason was that the CBJ legislation excluded any influence by the monetary authorities over the money stock except within very narrow limits, without controlling the money supply, the monetary authorities therefore could not significantly influence the economy. The superstructure of the money supply in Jordan was tied to the gold-exchange standard, and therefore to the foreign-exchange reserves. Money supply was thus made subservient to the dictates of the gold exchange system, as the ebb and flow of foreign-exchange reserves into and out of Jordan affected the money supply. It is possible to state that the monetary authorities in Jordan have completely surrendered their national mone tary sovereignity, and have submitted to the wishes of the automatic mechanism of the gold-exchange standard. The rigid link, imposed by the requirement of 100 percent foreign-exchange reserves, between Jordan's balance of payments and its money supply may subject the economy to undue stress when sudden economic disturbances occur. With an adverse balance of payments, a reduction in income may be expected. To guard against this pos- I sibility, a certain amount of purchasing power needs to i | be injected into the economy. A 100 percent currency j reserve system based on foreign assets precludes the I ! use of internal monetary expansion for this purpose. However, from Chapter VII, we see that the coef ficients of the money-supply function are stable, and j that both foreign assets and local assets have the same effect on the money supply. The money supply can there fore be influenced through local assets instead of for- I eign assets. A major difference between a 100 percent i i reserve system and a partial system is that the former | requires that any additional savings the community chooses j to hold in the form of currency must be invested in for- j eign-exchange reserves, while the latter permits them j to be invested in productive physical capital internally, I with implications for economic development. For this purpose, the list of assets that the CBJ is allowed to % hold should be changed. Local assets should become I • eligible for inclusion in the list, so that the CBJ will i ! be able to control the money supply and therefore influence \ j the economy. Jordan's domestic policies should be adjusted to its own needs and not to its foreign reserves. Jordan should have adequate liquidity for the full play of the expansionist process in the economy. Although Jordan can claim that it has achieved price stability, this point is not at issue. For develop ing countries, the question is not whether prices are stable or increasing; it is whether important develop ment projects can easily find adequate financing. The use of central-bank credit can therefore be significant. If, during the process of undertaking a program of econo mic development, prices rise, it will be a small penalty to pay for development. However, this need not be the case in Jordan, where local assets will not be any more inflationary than foreign assets — as can be inferred j ! i from our stable money-supply relationship. Many projects j i in Jordan have been shelved because no foreign loans j could be secured, at a time when Jordan's foreign assets j 1 were sitting idle in London money markets. Nothing was i I done to alleviate the problem of unemployment. j The CBJ must face the age-old question of which theory, or combination of theories, are likely to yield the best results. Its answer to this question will 201 determine to a great extent the kinds of guidelines needed in the formulation and implementation of monetary r> policy. RECOMMENDATIONS FOR FURTHER RESEARCH It is difficult to recommend further research on the money supply in Jordan, where no specific research other than the present study exists. A few studies touch on monetary aspects of the Jordanian economy. They may describe the system and give a historical account of it. No previous attempts have been made to derive behavioral monetary equations for Jordan. Nonetheless, further research may be recommended on the behavior of the public*3 demand for currency and for demand and time deposits. Research may also be recommended on the behavior of the commercial banks' demand for reserves, so that the determinants of the loan coefficients of the commercial banks may be found, and thus the rela- | tionship clarified between bank reserves and the money ' K - J supply. There is a need for empirical work on the beha- o vior of the banks and the public. In Chapter VI, we did not attempt to deal with 202 simplifications of the hypothetical formulation of the model. These simplifications concern the use of linear rather than nonlinear models and static rather than dynamic models (other than lagged foreign assets). We therefore feel the need for further research concern ing these simplifications. An attempt should be made to design a nonlinear dynamic model of the relationships discussed in this study. VO o n o n J* £ Currency in the hands of the nubile n Demand Deposits of p' the public H S' m Time Deposits of the § public Demand Deposits of the government Time Deposits of the government Lagged Foreign Assets Tariff Rates H H I —1 H I — 1 VD \ Q VO \ f l VO o n o n cjn o n on VJl 4 P C 0 r o H r o r o r o l— 1 H O M j O o v o a \ c o o o o v o \ j i r o -p- q p - q ro M H H M b d v o O N P ~ H r o v o q p p * v o o c d v j i c o vji M H H v o Jp H v o ON VJl VO H v n C \ P " r > CD —q 0 3 H U P P 'V J l * * * • * c o v ji v o -0 p " O OVHH H H H O H H O On O O - 4 CO 0 3 - q o r o r o - q VJl CO 4 P CO CO v n - q H P H • • * i • H I O V O V O O ) i'J VJl P O VJl H H H I —1 H V O P H I O U JP V O ON VJl o H VO 0 3 VJl ON H H 1 — 1 H H H H H H H K vo vo vo vo V O V O vo V O V O V O n> O N VJl VJl VJl VJl VJl vn vji VJl VJl P o vo o s — q O N V Jl qp C O ro H1 fi H H H H H H H vji vji vji P " P " H o cd— q C O o t * • t 4 p « s p p tr O n ro H - 4 vji H O n vji — q C O b o — 1 VJl 1 — 1 O N C O P'VJl ro O N H o 03 vo — q vji O N O N VJl P'VJl to * * * * p p p p.. p p « P V O H P O V C O H vo vo vo o n — q - q vo 0 0 O N O 0 0 H H S C O H' G H g H P " 4="b0 CO ro co H H o O H3 H* ■ • » • p p p p p p Hd o td o n — q o n vji — q o p o - q - q p HJ - q O qp O H C D C O 03 O O C O w o 0 H i to p - c o PO CO VJl P - c o C O ro o > p p p p p * p p p p O O £3 C O O N O CO c o 0 3 ro ro q p ro CR > O P H p - O O o C O vo H 0 3 P pn P 1 P Hd p. P H ' P 3 cH — q O n — q vji ro o o o O o H 3 P • • • • p p p p p p OR Hi > p - - q o n h O N CO ro h H VJl P tn VJl C O H VJl O N H ro co H H ’ co l o ro ro ro H H H H H ‘ ’d O O O N VJl H - q p - C O ro q p > • • • • « * * * * • c+ co — q cd H p - P 0 3 0 o 03 1 O O N p - C D O N C O VJl H t-1CO H h-1 H H H H H H I-1 H ro ro h ro CO ro P - C O ro CO Q ■ • • • p p p • • C h 0 3 - q v o o — q — q ro C O C O P"VO P " C O C D H — 3 O ro -q £ M X 203 ■ ) c 20** APPENDIX ■ ‘ t SUMMARY OF DATA - Part B (Millions of Jordan dinars) Year Yi MS FA s X M TR 1951 1 * 2.62 ll*.27 12.01 -2.82 if. 88 17.1+2 9.72 1952 1*8,81* 12.63 13.01 1.00 5.86 18.50 13.61* 1953 1*2.07 13.63 lU.85 1.81* 5.83 19.61 15.62 195^ 57.72 16.71* 17.1*3 2.58 5.18 20.03 17.^3 1955 1*9.98 17.1*9 21. 1*6 1*.03 9.82 27.06 21.27 1956 72.1*1 20. 21+ 25.18 3.72 9.79 26.68 20.61 1957 70.32 22.20 26.81* 1.66 13.78 32.59 20.1*7 1958 80.51 2l*.32 30.76 3.92 11.99 36.78 28.71 1959 88.27 2l*.2l* 30.30 -0.1*6 10.33 1 * 2.82 31.53 I960 93.98 26.09 31.85 1.55 11.72 1*5.85 35.68 1961 111.70 28.92 31+.90 3.05 15.1*0 1 * 6.08 33.73 1962 110.98 33.1*7 1 * 1 . OU 6.11* 19.82 50.75 37.07 1963 113.86 36.85 37.65 -3.39 23.71* 61.35 31+.22 196U 132.0U 1*3.00 55.72 18.07 28.56 57.15 1*6.66 1965 131.53 1*9.55 63.23 7.51 28.97 6 1 * . 12 1*2.66 1966 153.35 58.68 69.76 6.53 32.25 77.13 51. 1*1 u 1 0 rH t o o j -P 0 pq d > ■H t o C O • p SB. 3 > 3 0 p c d H H t o ft C J ft t o d X t o & < ■H Pi Q J P d O t o 1 m w d t o t o 0 0 b0 d 0 P -p a a -H H O u d O X 0 0 * ft d s a CO 0 0 O 0 d u & d H 0 0 ft A 1 o 205 APPENDIX SUMMARY OF DATA - Part C (Millions of Jordan dinars) Year w Rs=Rd LA TD L Y 1951 i4.6o 3.65 0.00 0.00 0.00 47.50 1952 19.50 5.25 0.00 o.o4 -0.04 54.70 1953 21.45 6.26 0.00 o.o4 -0.04 47.90 1954 22.61 6-75 0.00 o.o4 -o.o4 62.90 1955 31.09 10.29 0.00 o.o4 -0.04 59.80 1956 30.40 10.58 0.00 o.o4 -0.04 82.20 1957 34.25 12.09 0.00 o.o4 -o.o4 84.10 1958 40.70 15.57 0.00 o.o4 -0.04 92.50 1959 42.36 14.99 0.00 0.04 -o.o4 99.10 i960 47.40 15.04 0.00 1.18 -1.18 105.70 1961 49.13 17.08 0.00 0.85 -0.85 127.10 1962 56.89 21.50 0.00 0.50 -0.50 130.80 1963 57.96 13.40 0.00 3.85 -3.85 137.60 1964 75.22 26.46 0.00 6.24 -6.26 160.60 1965 71.63 19.03 0.00 17.85 -17.85 180.50 1966 83.66 . 22.30 0.09 17.22 -17.13 . 185.60 & 0 •p to p £ p *rl 0 to 0 n 3 CD Pi p a Q J n •p 0 1 in 0) £ 0 ■ H •P M >» 1 oJ + 0) aj p t> u H hi p CD LA +) C O 0 tt) cd pq c d m p I I m 0 0 O J ii 0 0) 0 p P w PI E H Pi O 206 BIBLIOGRAPHY ARTICLES Ahrensdorf, J. and Kanesathasan, S. "Variations in the Money Multiplier and their implications for Cen tral Banking," IMP Staff Papers, VIII (November, 1960), 126-129. Ando, A. and Modigliani, P. "The Relative Stability of Monetary Velocity and the investment Multiplier," American Economic Review, LV (September, 1965), 693-728. Angell, James W. and Picek, Karel P. "The Expansion of Bank Credit," Journal of Political Economy, XLI (February, 1933), 1-32 and (April, 1933), 152-193. Black, R. P. "The Impact of Member Bank Reserves upon the Money Supply," Southern Economic Journal, XXIX (January, 1963) , 199-210. Blowers, G. A. and MacLeod, A. N. "Currency Unification in Libya," IMF Staff Papers, II (November, 1952), 439-467. Brunner, Karl. "A Schema for the Supply Theory of Money," International Economic Review, II (January, 1961) 79-109. Brunner, K. and Meltzer, A. E. "Predicting Velocity Im plications for Theory and Policy," Journal of Finance, XVIII (May, 1963), 319-354. Brunner, Karl and Meltzer, Allan H. "Some Further Inves tigations of Demand and Supply Functions for Mon ey, ” Journal of Finance, XIX (May, 1964), 240-283. o Cagan, P. "The Demand for Currency Relative to the Total Money Supply," Journal of Political Economy, LXVI (August, 1958), 303-328. O 207 Cochrane, D. and Orcutt, G. H. "Application of Least Squares Regressions to Relationships Containing Autocorrelated Error Terms," Journal of American Statistical Association, XLIV (March, 1949), 32-61. DeLeeuw, F. and Gramlich, E. "The Federal Reserve - MIT Econometric Model," Federal Reserve Bulletin, LIV (January, 1968), 11-40. DePrano, M. and Mayer, T. "Tests of the Relative Impor tance of Autonomous Expenditures and Money," American Economic Review, LV (September, 1965), 729-752. Dewald, William G. "Free Reserves, Total Reserves and Monetary Control," Journal of Political Economy, LXXI (April, 1963), 141-153. Drake, P. J. "A Note on the Balance of payments and Money Supplies on a Colonial Monetary Economy," Social and Economic Studies, XIII (June, 1964), 311-312. Durban, J. and Watson, G. S. "Testing for Serial Corre lation in Least Squares Regression," Biometrika (December, 1950), and (June, 1951). Fand, David I. "Some Implications of Money Supply Analy sis, " American Economic Review, LVII (May, 1967), 380-400. Fand, David I. and Tower, J. "An Analysis of the Money Supply Process in Canada," Canadian Journal of Economics, I (May, 1968), 387-404. Farrar, D. E. and Glauber, R. R. "Multicollinearity in Regression Analysis: The Problem Revisited," Review of Economics and Statistics. XLIX (Febru ary, 1967), 92-107. Fellner, W. and Dillard, D. "Keynesian Economics after Twenty Years," American Economic Review Proceed ings, XLVII (May, 1957), 67-86. Fleming, Marcus J. and Boissonneault, L. “Money Supply 208 and Imports," IMF Staff Papers, VII (May, 1961) 227-240. Friedman, M. "Vault Cash and Free Reserves," Journal of Political Economy, LXIX (August, 1961), 181-182. Friedman, M. and Schwartz, A. "Money and Business Cycles," Review of Economics and Statistics, LXV (Febru ary, 1963), 32-78. Gramley, L. and Chase, S. "Time Deposits in Monetary Analysis," Federal Reserve Bulletin (October, 1965), 1380-1406. Harris, S. E. "Controversial Issues in Recent Monetary Policy: A Symposium," Review of Economics and Statistics, XLII (August 1960), 245-282. Hicks, J. R. "A Suggestion for Simplifying the Theory of Money," Economica, NS-2 (February, 1935), 1-19. Johnson, H. G. "Monetary Theory and Policy," American Economic Review, LII (June, 1962), 335-384. Jucker-Fleetwood, Erin E. "The Money Supply in Mature and Developing Economies," Irish Banking Review, (March, 1961). Latane, H. A. "Cash Balance and the Interest Rate: A Pragmatic Approach," Review of Economics and Statistics, XXXVI (November, 1954), 456-460. Latane, H. A. "Income Velocity and Interest Rates: A Pragmatic Approach," Review of Economics and Sta tistics, XLII (November, 1960), 445-459. Meltzer, A. H. "The Demand for Money: The Evidence from the Time Series," Journal of Political Economy, LXXI (June, 1963), 219-246. Moulton, H. G. "Surplus in Commercial Banking," Journal of Political Economy, XXV (December, 1917), 1003- 1018. 209 Orr, D. Polak, Polak, Shannon Shannon Teigen, Tobin, Tobin, Tobin, Tolley, Working and Mellon, W. G. "Stochastic Reserve Losses and Expansion of Bank Credit," American Economic Review, LI (September, 1961), 614-623. r. J. and Boissonneault, L. "Monetary Analysis of Income and imports and its Statistical Applica tion, " IMF Staff Papers, VII (April, 1960), 349-415. r. J. and White, H. H. "The Effects of Income Expansion on the Quantity of Money," IMF Staff Papers, IV (August, 1955), 398-433. H. A. "Evolution of the Colonial Sterling Ex change Standard," IMF Staff Papers, I (April, Q 1951), 334-354. H. A. "The Modern Colonial Exchange Standard," IMF Staff Papers. II (April, 1952), 318-362. Ronald L. "Demand and Supply Functions for Money in the United States: Some Structural Estimates," Econometrica, XXXII (October, 1964), 476-509. r. "Liquidity Dreference and Monetary Policy," Review of Economics and Statistics, XXIX (May, 1947), 124-131. r. "A Dynamic Aggregative Model," Journal of Poli tical Economy, LXIII (April, 1955), 103-115. "Money, capital and Other Stores of Value," American Economic Review Proceedings, LI (May, 1961), 26-37. G. S. "Providing for Growth of the Money Supply," Journal of Political Economy. LXV (December, 1957), 465-485. Elmer, "What do Statistical 'Demand Curves' Show?" Quarterly Journal of Economics. XLI (February, 1927), 212-235. 210 BOOKS Agger, E. E. Organized Banking. New York: Henry Holt and Co., 1918. Anderson, R. L. and Bancroft, T. A. Statistical Theory in Research. New York: McGraw-Hill, 1952. Burgess, W. R. The Reserve Banks and the Money Market. New York: Harper and Brothers, 1936. Cagan, Phillip. The Demand for Currency Relative to Total Money Supply. National Bureau of Economic Research, Occasional Paper 62, New York, 1958. Cagan, Phillip. Determinants and Effects of Changes in the Stock of Money, 1875-1960. National Bureau of Economic Research, New York: Columbia Univ ersity Press, 1965. Carson, D. (ed.) Banking and Monetary Studies. Homewood, 111.: Richard D. Irwin, 1963. Christ, C. Econometric Models and Methods. New York: Wiley, 1966. Currie, Lauchlin B. The Supply and Control of Money in The United States. Cambridge: Harvard Univer sity Press, 2nd ed., rev., 1935. de Leeuw, Frank. "A Model of Financial Behavior," The Brookings Quarterly Econometric Model of the United States. James S. Duesenberry, et. al., editors. Chicago: Rand McNally and Co., 1965, 465-532. Duesenberry, J. S. et al (eds). The Brookings Quarterly Econometric Model of the United States. Chicago: Rand-McNally, 1965. Feige, Edgar L. The Demand for Liquid Assets: A Tem poral Cross-Section Analysis. Englewood Cliffs, New Jersey: Prentice-Hall, 19'64. 211 Friedman, M. (ed.) Studies in the Quantity Theory of Money. Chicago: University of Chicago Press, 1958. Friedman, M. A.Program for Monetary Stability. New York: Fordham University Press, 1960. Friedman, M. Dollars and Deficits. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1968. Friedman, M. and Meiselman, D. "The Relative Stability of Monetary Velocity and the Investment Multiplier in the United States, 1897-1958," Commission on Money and Credit, Stabilization Policies. Engle wood Cliffs, New Jersey: Prentice-Hall, Inc., 1963, 165-268. Friedman, M. and Schwartz, A. A Monetary History of the United States, 1867-1960. National Bureau of Economic Research. Princeton: Princeton University Press, 1963. Goldberger, A. S. Econometric Theory. New York: Wiley, 1964. Goldfeld, Stephen M. Commercial Bank Behavior and Econo mic Activity: A Structural Study of Monetary Policies in Postwar United States. Amsterdam: North-Holland Publishing Co., 1966. Gurley, J. G. Liquidity and Financial Institutions in Postwar Economy, study Paper 14, Joint Economic Committee, 86th Congress, 2nd Session, Washing ton, 1960. Gurley, J. G. and Shaw, E. S. Money in a Theory of Fi nance. Washington: The Brookings Institution, 1960. Hansen, A. H. Monetary Theory and Fiscal Policy. New York: McGraw-Hill, 1949. Hicks, J. R. Value and Capital. London: Oxford Univer sity Press, 1939. 0 2 1 2 Johnston, Harry G. Essays in Monetary Economics. Cam bridge: Harvard University Press, 1967. Keynes, John M. A Treatise on Money. Vol. II. London: Harcourt, Brace and Co., 1930. Keynes, John M. The General Theory of Employment, Inter est and Money. New York: Harcourt, Brace 'and World, Inc., 1936. Khazzom, Daniel J. The Currency Ratio in Developing Countries. New York: Praeger, 1966. J Klein, L. R. An Introduction to Econometrics/ Englewood Cliffs, New jersey: Prentice-Hall, 1962. ; ^ Lutz, P. A. and Mints, L. W. (eds.) : Readings in Monetary Theory. Somewood, 111.: Richard D. Irwin, Inc., 1951. v ,u . - n Malinvaud, E. Statistical Methods of Econometrics. Chi cago: Rand McNally and Co., ®1966. Mayer, T. Monetary Policy in the Unitecl °States. New York: Random House, 1968. c o Meigs, A. J. Free Reserves and the Money Supply. Chi cago: TljeDUniversity of Chicago Press, 1962. 0 * * Morrison, G. Liquidity Preferences of Commercial Banks. C^cago: University of Chicago Press^ 1966. 0 ■ ; o Pat inkin, ifon. Money, interest, and Priced. 2nd ed./ New York: Harpqy and Row, 1965. " ° n 0 Phillips, C. A. Bank Credit:. New York: The MacMillan ' Co., 1920G © ( r -, 0 © 0 Riefler,) W. W. Money Rates and Money Markets in the United States. New YcSrk: Harper and Brothers, 1930- n o 0 Q , (! ) G IM Schumpeter, J. A. History of Economic Analysis. New York: Oxford University Press, 1954. Shaw, Edward S. Money, Income and Monetary Policy. Chi cago: Richard D. Irwin, Inc., 1950. Suits, Daniel B. The Theory and Application of Econome tric Model3 . Athena: Center of Economic Research 1963. Thorn, Richard S. (ed.) Monetary Theory and Policy. New York: Random House, 19(56. Trescott, Paul B. Money, Banking and Economic Welfare. New York: McGraw-Hill, 2nd ed., 1965. Turner, Robert C. Member-Bank Borrowing. Columbus: The Ohio State University, 1938. GOVERNMENT PUBLICATIONS Committee on the Working of the Monetary System (Chair man; The Rt. Hon;. The Lord ( Radcliffe, G.B.E.) Report, London, 1§59. , * } r Central Bank of Jordan. Annual ’ Report. Amman, relevant issues. Currency Control Department. Balance of Payments. Amman, relevant issues. ‘Currency Control Department. Banking and Foreign Exchange Statistics!, -Amman, relevant issues. Department of.Statistics. Statistical Yearbook. Amman, relevant issues. Federal Reserve Balletin. Washington, D.C.: Board of Governors of the Federal Reserve System. 214 Jordan Currency Board. Annual Report. Amman, relevant issues. * United Nations. Statistical Yearbook. New York, rele vant issues. United Nations. Yearbook of International Trade Statistics, relevant issues.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Interaction Among Financial Intermediaries In The Money And Capital Markets: A Theoretical And Empirical Study

PDF

An Econometric Analysis Of The Influence Of Money Supply And And Money Demand Relations In The Determination Of National Income

PDF

The Effect Of Wealth On Consumption: Empirical Tests For Significance Andlags

PDF

The Role Of The Manager Of The Federal Reserve System Open Market Account: 1951-1961

PDF

A Dynamic Model For Predicting The Effects Of Fiscal Policy Measures

PDF

Planning For Stability And Self-Reliance: An Evaluation Of Policy Approaches For India

PDF

The Economic Impact Of Savings And Loan Associations On Residential Construction

PDF

The Pricing Of Initial Public Offering And Privatizations: An Influence-Seeking Theory And An Empirical Investigation With New Issues In The Nigerian Equity Market

PDF

Some Effects Of Monetary And Fiscal Policy On The Distribution Of Wealth

PDF

Contributions To The Theory Of Industrial Location: Urban Decentralization

PDF

A Comparison Of Employment Performance In The Manufacturing Sector Of Less Developed Countries: Empirical Tests Of Alternative Hypotheses

PDF

The Economics Of Sugar Quotas

PDF

The Optimal Supply Of Money And Optimal Monetary Policies

PDF

Diversified expectations in the foreign exchange market

PDF

Export Instability And Economic Development: A Statistical Verification

PDF

The Welfare Cost Of Protection: A Parametric Simulation

PDF

A Theoretical Analysis Of The Dynamic Aspects Of Economic Integration Of Less Developed Countries

PDF

Applications Of Stochastic Processes To Indian Agriculture

PDF

Libya's Petroleum Policy

PDF

Approaches To Economic Policy For Stabilization And Growth

Asset Metadata

Creator Marto, Michel Isa (author)

Core Title A Money Supply Model: Jordan

Contributor Digitized by ProQuest (provenance)

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

Advisor DePrano, Michael E. (committee chair), MacReynolds, William (committee member), Nugent, Jeffrey B. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-428877

Unique identifier UC11361809

Identifier 7025049.pdf (filename),usctheses-c18-428877 (legacy record id)

Legacy Identifier 7025049.pdf

Dmrecord 428877

Document Type Dissertation

Rights Marto, Michel Isa

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