University of Southern California Dissertations and Theses

The effect of managerial retention incentives on the relationship between financing constraints and voluntary disclosure

(USC Thesis Other)

The effect of managerial retention incentives on the relationship between financing constraints and voluntary disclosure

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

THE EFFECT OF MANAGERIAL RETENTION INCENTIVES
ON THE RELATIONSHIP BETWEEN
FINANCING CONSTRAINTS AND VOLUNTARY DISCLOSURE

by

Seung Hwan (Peter) Oh

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)

August 2018

Copyright 2018 Seung Hwan (Peter) Oh
i

Acknowledgements

I am indebt to Kenneth A. Merchant (dissertation committee chair) for his unconditional support
throughout this project. I am also grateful to Kevin J. Murphy for providing valuable insight to
this paper. Special thanks to Gerald Hoberg for his consistent feedback and suggestions. I
appreciate the encouragement provided by Sukeun Kwak and Mark Soliman, and the kind help
offered by Allison Kays, Tina Lang, Sam Lee, and Satish Sahoo. This paper has benefited from
the discussions with Bok Baik, Harry DeAngelo, Tyler DeGroot, David H. Erkens, Joonil Lee,
Clive Lennox, Shelley Li, Georgios Magkotsios, Steve R. Matsunaga, Regina Wittenberg-
Moerman, Stacey Ritter, Patrick W. Ryu, TJ Wong, Louis Yang, and workshop participants at
University of Southern California, Miami Rookie Camp, Yale University, McGill University,
University of Colorado at Boulder, and University of Illinois at Chicago. I acknowledge the
financial support from the Leventhal School of Accounting, Marshall School of Business,
Desautels Faculty of Management, and Kwanjeong Educational Foundation. I am solely
responsible for any possible errors. Lastly, this paper is dedicated to my parents who provided
endless care and love throughout my Ph.D. studies.

ii
Table of Contents

Acknowledgements i
Table of Contents ii
List of Tables iii
List of Figures iv
Abstract v
Chapter 1. Introduction 1
Chapter 2. Financing Constraints and Voluntary Disclosure: Hypothesis Development 7
2.1. Information Asymmetry and Different Types of Financing Constraints 7
2.2. The Role of Voluntary Disclosure in Mitigating Information Asymmetry 9
Chapter 3. Financing Constraints and Voluntary Disclosure: Empirical Evidence 12
3.1. Variable Description for H1 12
3.2. Empirical Methodology for H1 14
3.3. Sample Selection and Descriptive Statistics for H1 15
3.4. Primary Results for H1 19
3.5. Additional Mechanism Tests for H1 23
3.6. Robustness Tests for H1 30
Chapter 4. Moderating Effect of Retention Incentives on the Relationship between
Financing Constraints and Voluntary Disclosure: Hypothesis Development 38
4.1. The Role of Managers’ Retention Incentives and Horizon on
Frequency of Disclosure 38
4.2. The Role of Managers’ Retention Incentives and Horizon on
Accuracy of Disclosure 42
Chapter 5. Moderating Effect of Retention Incentives on the Relationship between
Financing Constraints and Voluntary Disclosure: Empirical Result 45
5.1. Variable Description for H2 45
5.2. Empirical Methodology for H2 46
5.3. Sample Selection and Descriptive Statistics for H2 46
5.4. Primary Results for H2 48
5.5. Additional Analyses and Robustness Tests for H2 51
Chapter 6. Conclusion 59
References 61
Appendix A: Variable Definitions 65
Appendix B: Validating Hoberg and Maksimovic (2015)’s Measures 68
Appendix C: Model 69
Appendix D: Details on Pay Duration Measure 70
iii
List of Tables

Table 1: Sample Selection Procedure for H1 16
Table 2: Basic Descriptive Statistics for H1 17
Table 3: Primary Result
- Financing Constraint and Management Earnings Guidance Frequency 20
Table 4: Mitigating the Endogeneity Concerns on Primary Result 22
Table 5: Effect of Financing Constraint and Management Earnings Forecast Frequency
on Post Financing 24
Table 6: What is Driving All the Differences? Real Issue vs. Intention to Issue 27
Table 7: Increase in Frequency of Disclosure Reducing
Information Asymmetry (Second Moment) – Number of Analyst Following 29
Table 8: Other Characteristics of Management Earnings Guidance 32
Table 9: Qualitative Disclosures 35
Table 10: Summary Statistics for H2 47
Table 11: Moderating Effect of Pay Duration on the Relationship between
Financing Constraint and Management Earnings Guidance Frequency 49
Table 12: Moderating Effect of Pay Duration on the Relationship between
Financing Constraint and Management Earnings Guidance Accuracy 50
Table 13: Placebo Test - Moderating Effect of Equity Pay on Financing Constraint
and Management Earnings Forecast 52
Table 14: Joint Effect of Financing Constraint, Management Earnings Forecast Frequency,
and Pay Duration on Post Financing 53
Table 15: Moderating Effect of Pay Duration on Qualitative Voluntary Disclosures 55
Table 16: Public vs. Private Debt 56
Table 17: Moderating Effect of Pay Duration on Financing Constraint and Optimistic
vs. Pessimistic Management Earnings Forecast 58
iv
List of Figures

Figure A: Times Series Trend of Number of Firms Providing Management Earnings Guidance 18
Figure B: 2 (Constrained vs. Unconstrained) by 2 (Issue vs. No-issue) Framework 26
Figure C: Basic Timeline 40
Figure D: Summary of the Hypotheses 44

v
ABSTRACT
This dissertation examines the effect of financing constraints on voluntary disclosure
behavior; specifically, the incremental impact of managers’ retention incentives on the
relationship between the two. Different types of financial constraints (i.e. equity vs. debt-
constraints) should lead to different voluntary disclosure behavior as firms with the former
are trying to raise capital through the public stock market while firms with the latter are
trying to finance debt capital through private relationships with lenders. I predict and find
that (1) equity-constrained firms indeed issue more earnings guidance than debt-
constrained firms in order to reduce the information asymmetry problem in the public
market and (2) this result is driven by managers in the equity-constrained firms who have
higher retention incentives. I also find that managers with higher retention incentives in
equity-constrained firms not only issue more frequent earnings guidance but also more
accurate guidance. Additional analysis shows that this behavior successfully reduces
financial constraints: managers who increase earnings guidance improve the information
environment so as to increase actual (equity) issuance in the following period. Moreover,
although my primary results rely on a quantitative voluntary disclosure (management
earnings guidance), I show that my main results are robust to qualitative voluntary
disclosures as well. Overall, my findings provide new evidence on how different types of
financial constraints, combined with managerial incentives, shape corporate disclosure
policies.
1
Chapter 1. Introduction
Despite the huge importance of voluntary disclosure in mitigating information asymmetry
which in turn reduces financing constraints, no study has yet provided evidence on the relationship
between the two. Two related and potentially important factors that are missing in prior studies of
voluntary disclosure are (1) a company’s degree of financial constraints and (2) its approach to
raising additional capital.
1
In addition, the managers' decision horizon that can impact the ex-ante
incentives to expend unobservable effort in improving the firms’ financial availability may also
potentially play a crucial role in shaping the voluntary disclosure behavior when the firm is under
(different types of) financing constraint. Yet, prior literature is relatively silent on this issue as
well.
2

This dissertation tries to provide an answer to these underexplored relationships by looking
into a unique setting where financially constrained firms take advantage of their voluntary
disclosure options to reduce information asymmetry problems. Specifically, this dissertation
examines the impacts of different financial constraint situations on companies’ voluntary
disclosure (earnings guidance) behavior, and particularly whether this baseline relationship is more
pronounced when managers have stronger retention incentives - i.e. are incentivized to focus on a
longer horizon.

1
Prior literature so far has examined (1) whether firms change their voluntary disclosure behavior upon actual
seasoned equity offerings (SEO) either to mitigate underpricing or to promote overpricing (ex. Lang and Lundhom,
2000; Li and Zhuang, 2012), and (2) how managers’ characteristics or incentives affect voluntary disclosure behavior
(ex. Aboody and Kasznik, 2000; Cheng and Lo, 2006; Bamber et al. 2010). Nevertheless, evidence on managers’ ex-
ante incentives (eagerness, intention, desire, or willingness) to raise capital, which stem from (different types of)
financial constraints, affecting voluntary disclosure behavior is scarce.
2
An overarching pillar of accounting research is how different types of managerial incentives affect corporate policies.
However, evidence on the incremental effect of retention incentives (i.e. decision horizon), derived from managers’
compensation contracts, on the aforementioned relationship is limited.
2
Managers’ voluntary disclosure of financial information is a key issue in financial reporting.
Not only do managers get evaluated on the accuracy of their disclosures (Lee, Matsunaga, and
Park 2012), but also information contained in guidance itself plays a critical role in allocating
capital to the whole economy (Li and Zhang, 2012). Managers’ earnings guidance, as the main
form of voluntary disclosure, has the potential to reduce the level of information asymmetry
between managers and outside investors. This can help to decrease firms’ costs of capital (Hirst,
Koonce, and Venkataraman 2008). For this reason, issuing the right amount of accurate
information at the most pertinent times can be especially useful to firms that are facing financial
constraints.
I hypothesize that equity-constrained firms will increase the frequency of management
earnings guidance compared to debt-constrained firms.
3
Unlike debt-holders who have the right to
request borrowers’ information through personal meetings (private channels), equity-holders are
more likely to rely solely on public disclosures such as mandatory SEC filings or voluntary
financial disclosures. Consequently, equity-holders are more sensitive to public information
whereas debt-holders are less sensitive as they can receive information from other sources. While
debt-constrained firms (or debt issuing firms in general) tend to be older, larger, and more stable,
equity-constrained firms are younger, smaller, and more innovative (Hoberg and Maksimovic,
2015). Given the fact that older and larger firms are generally thought to have less information
asymmetries, debt-constrained firms are less likely to suffer from information asymmetry issues
with their lenders than equity-constrained firms.
4
Thus, equity-constrained firms will want to

3
Based on the measures devised in Hoberg and Maksimovic (2015), I define equity (debt) constrained as (1) firms
that intend to issue equity to fund their positive NPV projects in the near future, but (2) are facing challenges in
actually raising the equity (debt), and (3) they are at risk of delaying the positive NPV investments if they fail to
actually raise the equity (debt).
4
Equity-constrained firms, being younger and more innovative, have shorter track records and little in the way of
proven cash-flows. Hence information asymmetry is a likely cause for their financial constraints.
3
disclose more information – i.e. issue more earnings guidance - to their potential equity holders in
order to enhance their chances of raising capital by reducing information asymmetry as much as
possible.
5

In addition, I propose that the positive association between equity constrained firms and
the frequency of management earnings forecasts will be more pronounced in firms with managers
who have higher retention incentives. Managers with high retention incentives (i.e. who must wait
a few years for their compensation to fully vest) are more likely to have a longer horizon since
leaving the firm within a short period of time is costly (Gopalan et al. 2014). If this is the case,
long term horizon managers (in the equity constrained firms) are even more likely to issue frequent
earnings guidance due to their career concerns and compensation. These managers need to remain
as CEOs for a relatively long period of time in order to maximize their pay, which is tied to long
term firm value. As such, it is beneficial for them to (1) resolve financing constraints and (2) invest
in the positive NPV projects that maximize the long term value of their firm by actively disclosing
their firms’ financial information. Moreover, not only do these managers want to make frequent
disclosures, but also they will want to make accurate disclosures as high retention incentives
induce them to put in more effort to issue high quality forecasts and secure their job positions (Lee
et al. 2012).
To test my first hypothesis, I need to (1) identify and locate firms that need outside capital
to fund their potential investment projects due to their lack of internal capital, and (2) distinguish

5
However, due to the fact that peer firms can maximize their investment opportunity by learning from the focal firms’
(equity-constrained firms’) valuation information (i.e. earnings guidance), increasing the frequency of disclosures can
be costly (Foucault and Fresard 2014). Other factors such as competition (Huang, Jennings, and Yu 2017), litigation
(Francis et al. 1994), and enhancement in monitoring (Shleifer and Vishny 1989) can also impose additional costs on
equity-constrained firms in disclosing their earnings guidance. For this reason, considerable amount of tension that
works against the positive relationship between the degree of equity-constraint and the frequency of earnings guidance
issuance exists.
4
financially constrained firms that plan to finance with equity or debt. Hoberg and Maksimovic
(2015) develop a novel approach to measure financial constraints. They use textual analysis to
compute the cosine similarity between the text in each firm’s Liquidity and Capitalization
Resources subsection of their MD&A discussion in their 10-K and the text used in the same section
of firms who explicitly disclose they are financially constrained to calculate the financing
constraints score. I use their continuous financial constraint score in my study to capture each
firm’s overall degree of financial constraints, as well as their measure of whether the financial
constraints are unique to equity or debt.
6
I also use the IBES Guidance database to calculate the
frequency and accuracy of management earnings forecasts, and the Incentive Lab database to
measure executives’ pay duration – i.e. the weighted average of the vesting periods of the
components of CEO compensation - which proxies for retention incentives and horizon.
I conduct a series of empirical tests that examine the relationship between the frequency
and accuracy of management earnings forecast and the degree of different financing constraints
using 8,796 firm-year observations from 2000 to 2013. Results support my prediction that equity-
constrained firms provide more frequent management earnings guidance than debt-constrained
firms.
7
I also find that, as predicted, the positive association between guidance frequency and
equity constraints is more pronounced under firms with high executive pay duration. Moreover,
consistent with my hypothesis, the subsample of firms with managers who have long pay durations
issue more accurate earnings forecasts as well.

6
I am grateful to Gerald Hoberg for generously sharing these data.
7
In Chapter 3 of this dissertation, I conduct series of additional tests to further provide underlying mechanisms and
robustness on my primary result. Using an external exogenous mutual fund selling shock, I mitigate endogeneity
concerns on the positive association between equity-constraints and frequency of earnings guidance. I also show that
increase in disclosure (for equity-constrained firms) is positively associated with decrease in information asymmetry
proxied by number of analyst following. Moreover, I further provide evidence that equity-constrained firms that issue
more earnings guidance provide accurate forecast as well.
5
Further tests document a positive association between the frequency of the earnings
forecasts and the degree of financing constraints is significant for both groups - i.e. firms that
actually end up raising capital and firms that do not raise capital in the next period. This also
suggests that firms that issue equity, but (1) were not financially (equity) constrained or (2) never
expressed strong ex-ante intention to raise equity in advance do not necessarily increase (change)
their earnings guidance policies. The results extend and refine the findings in Lang and Lundholm
(2000), indicating that it is the ex-ante intent of raising the capital which mainly drives the change
in the voluntary disclosure behavior more than the ex-post actual issuance.
8

Additional analysis shows that the increase in the frequency of earnings guidance in equity
constrained firms is positively associated with an increase in the amount of equity raised in the
next period. This supports the underlying theme that voluntary disclosure reduces information
asymmetry. Moreover, given that management earnings guidance is a quantitative voluntary
disclosure, I also show that my main results are robust to qualitative voluntary disclosure.
Furthermore, I also decompose debt-constrained firms into firms that are going for private loans
and public bonds, and find that debt-constrained firms that are trying to issue public bonds exhibit
earnings guidance issuance behavior similar to that of equity-constrained firms. This finding
corroborates my theory that voluntary disclosure plays a key role in mitigating information
asymmetry as public bonds should be more informationally sensitive than private loans.
This paper makes important contributions to the financing constraints, voluntary disclosure,
and managerial incentive literature in several ways. First, this paper is the first study to document

8
Lang and Lundholm (2000) find that firms that actually issue equity – i.e. SEOs – increase their disclosures before
the real issuance. I additionally claim that among the firms which intend to raise equity capital, subset of firms that
are unsuccessful in issuing equity (or that do not end up raising equity) can also increase the frequency of voluntary
disclosure as well as subset of firms that are successful in issuing equity (or that do end up issuing equity).
6
that ex-ante incentives
9
derived from financing constraints affect voluntary disclosure behavior. I
show that (1) firms that are more likely to suffer from information asymmetry issues provide more
frequent earnings forecasts and (2) this increase in supply of information leads to larger amounts
of real issuance. I corroborate that voluntary disclosure provides a crucial role in allocating capital
to the economy by reducing the level of information asymmetry. Second, using a refined measure
devised by Hoberg and Maksimovic (2015), I identify a group of firms that are more sensitive to
information asymmetry issues by distinguishing equity vs. debt-constrained firms and I show that
the results described above stem from equity-constrained firms having a higher willingness to
resolve information asymmetry problems in order to raise the capital. Lastly, I also document the
interplay between the financial constraints and managerial incentives in shaping firms’ voluntary
disclosure practices by showing that the increase in the frequency (and accuracy) of earnings
guidance in equity constrained firms is more pronounced for firms that have managers with high
retention incentives and a longer horizon.
The remainder of the dissertation is organized as following. Chapter 2 and 4 reviews related
prior literature and derives my first and second hypotheses, respectively. Chapter 3 and 5 discusses
the research design, sample selection, data description, and presents my main empirical findings
along with the results of additional sensitivity analyses for my first and second hypotheses,
respectively. Chapter 6 concludes the study.

9
Note that I am verifying that it is the intent (i.e. desire, plan) before the issuance which drives the voluntary
disclosure behavior more than the actual issuance.
7
Chapter 2. Financing Constraints and Voluntary Disclosure:
Hypothesis Development

2.1. Information Asymmetry and Different Types of Financing Constraints
Asymmetric information between managers and investors may in fact be the main cause of
financial constraints. Theory and evidence in the financial constraints literature well supports the
notion that the informational environment of the firm is a key issue. Myers and Majluf (1984)
illustrate that asymmetric information can sway managerial choice regarding the type of capital
that should be raised. In a notable extension, Krasker (1986) more specifically shows that firms
cannot raise more than an upper limit of capital to fund their positive NPV investments when
asymmetric information is extreme. Hence Krasker (1986)’s theory directly predicts that
asymmetric information can cause financial constraints.
A more refined view of the financial constraints theory in the literature suggests that the
information asymmetry problem will be more salient for firms that specifically face constraints in
the equity market. A stylized fact noted by many scholars
10
is that equity is a more informationally
sensitive security than debt. Since debt-holders are concerned about receiving their fixed claim
upon the maturity of the debt while equity-holders are residual claimants (Jensen and Meckling,
1976), debt-holders mostly (if not, only) focus on new information that would only affect their
fixed claim whereas equity-holders react to any new information revealed to the market that would
affect equity prices. In this manner, the aforementioned theory by Krasker (1986) is more
specifically about equity issuance rather than debt issuance.
11

10
For example, see Page 85 (last paragraph) of Constantinides, G., Milton, H., and Stultz, R. 2013. Handbook of the
Economics and Finance. Volume 2A: Corporate Finance, 1
st
Edition, North Holland.
11
The theory specifically predicts “equity rationing - firms with the best investment opportunities face financial
constraints and are forced to under invest”, and that a firm simply cannot raise more money than a fixed limit by
issuing more shares.
8
Krasker (1986)’s prediction has strong empirical support in the financing constraint
literature. In an article that uses computational linguistics to directly measure (1) the different types
of financial constraints and (2) the existence of asymmetric information (through disclosure related
to proprietary informational risk
12
), Hoberg and Maksimovic (2015) find that financial constraints
are indeed strongly related to informational asymmetry. By further documenting the fact that the
constraints in the equity market are fundamentally different from those in the debt-markets, they
extend earlier studies in financial constraints by separately measuring constraints in both markets.
Specifically, Hoberg and Maksinomovic (2015) find that only equity-market constraints are
positively linked to asymmetric information.
Unlike equity-market constraints, debt-market constraints likely are not due to informational
asymmetry. Hoberg and Maksimovic (2015), in fact, find that debt-market constraints are
somewhat negatively correlated with asymmetric information.
13
Accordingly, debt markets have
their own unique mechanisms for mitigating information asymmetry problem that can cause
financial frictions. Relational long-term lending (Bharath, Dahiya, Saunders, and Srinivasan 2011)
and special monitoring devices such as affiliated bankers on board (Erkens, Subramanyam, and
Zhang 2014) can be channels that convey private information and therefore reduce the asymmetry
of information. These findings, coupled with theoretical work by Myers (1977) suggest that the
debt overhang problem, rather than information asymmetry issue, is more likely to drive

12
Hoberg and Maksimovic (2015) identify firms with proprietary information risks as firms mentioning one protection
word (“protect,” “protection,” or “safeguard”) and one information phrase (“trade secret”, “trade secrets”, “proprietary
information”, or “confidential information”) in their 10-Ks. Then, they show that equity-constrained firms are highly
associated with having the proprietary information risks.
13
Although Hoberg and Maksimovic (2015) do not distinguish between public debt and private debt when considering
debt-constrained firms, I assume that debt-constrained firms in general have the intention to raise private debt.
Additional analysis shown in Chapter 5 (Section 5.8) of this dissertation also corroborates my assumption.
9
constraints in debt markets. An additional theory by Rampini and Viswanathan (2010) also
suggests that collateral constraints can drive financial constraints in the debt markets.

2.2. The Role of Voluntary Disclosure in Mitigating Information Asymmetry
Management guidance has become an increasingly important and pervasive form of
voluntary disclosure in the contemporary economy (Anilowski, Feng, and Skinner 2007) as it
represents a form of communication from managers to outside investors. Specifically,
management guidance allows investors to update their projections of a firm’s cash flow and risk
and thus more accurately estimate the value of the firm’s stock. With a lower level of information
asymmetry, firms can reduce their cost of obtaining external finance and increase their value
(Botosan 1997). Moreover, many of the motivations managers have for issuing earnings forecasts
are congruent with those of shareholders. That is, the supply of and the demand for forecasts is
assumed to be largely driven by stock-price considerations, with managers issuing forecasts (and
analysts and investors demanding them) to reduce the asymmetry in information between
managers and analysts and current or potential equity investors (Ajinkya and Gift 1984; Verrecchia
2001). Lower information asymmetry is viewed as desirable because it is associated with higher
liquidity (Diamond and Verrecchia 1991) and a lower cost of capital (Leuz and Verrecchia 2000).
The empirical results in the voluntary disclosure literature echo the theoretical predictions
stated above. Lang and Lundholm (2000) examine corporate disclosure activities around seasoned
equity offerings and find that issuing firms dramatically increase their disclosure activity
beginning six months before the offering, particularly for the categories of disclosure over which
firms have the most discretion.
14
Li and Zhuang (2012) examine whether management guidance

14
While Lang and Lundholm (2000) finds a positive relationship between SEOs (i.e. actual equity issuance) and
disclosure, I predict a positive association between ex-ante intentions to raise equity (i.e. degree of equity-constraints)
and frequency of issuing earnings guidance for both types of firms that issue and do not issue equity. This prediction
also implies that I do not expect to find any increase in frequency of issuing earnings guidance even if the firm have
actually issued equity when the firm is not equity-constrained (i.e. when the firm have not expressed any challenges
10
influences secondary equity offerings’ (SEO) underpricing and find that management guidance
reduces the magnitude of SEO underpricing by significantly reducing the information asymmetry,
and that this effect is more pronounced among smaller firms.
15
Balakrishnan et al. (2014) also find
that firms respond to an exogenous loss of public information by providing more timely and
informative earnings guidance. These responses resulted in improving the liquidity and in turn
increased firm value.
Motivated by both theoretical predictions and empirical results suggesting that asymmetric
information is a more salient issue for firms facing equity market constraints specifically, I make
the following predictions:
H1: Firms facing financial constraints specifically in the equity markets will provide more
frequent earnings guidance than firms facing constraints specifically in the debt markets.
16

However, strong tension exists with respect to my first hypothesis due to the following
arguments when cost of (voluntary) disclosure is considered in addition to the benefit of disclosure
mentioned above. Managers in general may not always desire to reduce information asymmetry
voluntarily if lower information asymmetry leads to greater monitoring (Shleifer and Vishny
1989). Moreover, managers may want to avoid disclosing their information even further if they
are facing intensive competition with their existing rivals. A number of theoretical and empirical
studies support the assertion that firms in competitive industries reduce voluntary disclosure due

or intentions of raising equity in advance). Please see Chapter 3 (Section 3.5) for further explanation on how my H1
is different from findings in Lang and Lundholm (2000). Additional analysis and results extending Lang and
Lundholm (2000) is also provided.
15
Numerous prior studies document that increase in disclosure leads to higher stock price (i.e. first moment effect of
voluntary disclosure). In addition to these studies, I provide further evidence that increase in disclosure for equity-
constrained firms leads to increase in number of analyst following, which is one of the indications of reduction in
information asymmetry (i.e. second moment effect of voluntary disclosure). See Chapter 3 (Section 3.5) for more
details.
16
In more detail, (1) I predict a positive association between financing constraints focused on equity markets and
frequency of management earnings guidance issuance, whereas (2) I predict no association between financing
constraints focused on debt markets and frequency of management earnings guidance issuance.
11
to increased proprietary costs (Dye 1985, Li 2010, Huang et al. 2017). Since equity-constrained
firms are normally high-growth firms that are more likely to mention concerns about the risk of
losing proprietary information in their 10-Ks (Hoberg and Maksimovic 2015), the opposite
prediction may prevail regarding their voluntary disclosure policies and make them disclose less.
Potential lenders may also prefer firms that already provide a large amount of information
and demand voluntary disclosure from the (potential) borrowers in order to protect themselves in
advance. Several studies document the borrowers’ response to the debt holders’ demand.
Shivakumar et al. (2011) find that credit markets react to management forecast news and that the
reactions to forecast news are stronger than to actual earnings news. Consistent with the
asymmetric payoffs to debt holders, forecast news is most relevant for firms with poor credit
ratings and this relevance is particularly strong during periods of high uncertainty. Lo (2014) finds
a similar result: borrowers, whose banking relationships are threatened by declining bank health,
increase their public disclosures (management forecasts) of forward-looking information.
Although prior studies indicate that lenders and borrowers rely on private communications rather
than public disclosures in relationship lending to resolve information asymmetries, recent evidence
indicates that borrowers must reconsider their disclosure policies when the private information
channel is threatened.
Thus, H1’s prediction of a positive (no) relationship between equity (debt) constrained firms
and the frequency of voluntary disclosure is not a foregone conclusion.

12
Chapter 3. Financing Constraints and Voluntary Disclosure:
Empirical Evidence

Having developed my first hypotheses, I now turn to a discussion of my research design.
This discussion includes a description of variables, sample selection, and descriptive statistics.

3.1. Variable Description for H1

Financing Constraints
I use the text based financial constraint measures developed in Hoberg and Maksimovic
(2015) to identify firms that need to raise outside capital - either equity or debt - to fund their
potential investment projects due to a lack of internal capital. Hoberg and Maksimovic (2015)
develop a novel approach to measure financial constraints based on textual analysis of the MD&A
section of firms’ 10-Ks. SEC regulations require firms to discuss challenges to their liquidity issues
and how these challenges impact their investment plans. Hoberg and Maksimovic (2015) use text-
extraction techniques to identify firms that disclose the risk of delaying positive NPV investments
due to financial liquidity difficulties. Although relatively few firms explicitly state that they face
financial constraints, Hoberg and Maksimovic (2015) construct a continuous measure of
constraints by calculating the overall verbal similarity of each firm’s MD&A text to that of firms
that explicitly state their constraints.
17
By computing the cosine similarity between the text in each
firm’s Liquidity and Capitalization Resources subsection in MD&A, Hoberg and Maksimovic
(2015) construct continuous constraint variables for each firm.
18

17
Hoberg and Maksimovic (2015) uses multiple list of words that identify firms do have investment projects that are
at the risk of being delayed for lack of internal funds. Few examples of wordings are: (abandon, eliminate, curtail,
postpone) and (construction, expansion, acquisition, restructuring, project, research, development, exploration,
expenditure, manufacture, renovate, growth, commercial release, business plan).
18
See Appendix B in Hoberg and Maksimovic (2015) for further information related to the variable construction
process.
13
I use these more informative
19
financial constraint scores in my study. While the general
Constrained variable measures each firm’s degree of overall financial constraints, the other two
variables (Equity_Constrained and Debt_Constrained) measure financial constraints uniquely
faced by firms intending to issue either equity or debt, respectively. Due to the fact that I need to
capture firms’ desire (i.e. intention, plan) to raise external capital, higher values on these financial
constraint variables does not need to lead to actual equity or debt issuance although positive
correlation can exist between the two concepts.
20
Specifically, the text-based financing constraint
variables are defined as following.
Constrained = Firms with higher values are more similar to (in terms of
disclosures in their 10K filings) a set of firms known to be at
risk of delaying their investments due to issues with liquidity

SEC-
Filings
Equity_Constrained = Firms with higher values are more similar to a set of firms (1)
that are at risk of delaying their investments due to liquidity
issues and (2) that indicate plans to issue equity (presumably
to address their liquidity challenges)

SEC-
Filings
Debt_Constrained = Firms with higher values are more similar to a set of firms (1)
that are at risk of delaying their investments due to liquidity
issues and (2) that indicate plans to issue debt (presumably to
address their liquidity challenges)

SEC-
Filings

Management Earnings Forecast
I use management earnings guidance information to proxy for (quantitative) voluntary
disclosure made by firms. Following the management earnings forecast literature (Lennox and
Park 2006, Baik et al. 2011, Lee et al. 2012), I construct my main dependent variables -
management earnings forecast frequency, management earnings forecast accuracy, and likelihood

19
Farre-Mensa and Ljungqvist (2016) criticizes traditional financing constraints measures such as KZ index, WW
index, and HP index after verifying that firms typically classified as constrained based on these measures in fact do
not actually behave as if they were constrained. They mainly attribute this failure of traditional financing constraints
measures to the endogenous nature of the measures. In contrast, text-based financing constraint measures constructed
in Hoberg and Maksimovic (2015) are more direct measures of constraint disclosures rather than a function of other
endogenous variables. Note that I am not claiming that text-based financial constraints measures in Hoberg and
Maksimovic (2015) is free from endogeneity issue, but relatively less endogenous compared to previous measures.
20
Results in Appendix B show that the Equity_Constrained (Debt_Constrained) variable used in this study is
positively associated with next-period equity (debt) financing on average, and this relationship is more pronounced
when the firms raise a large amount of capital.
14
of managers’ issuing earnings forecasts - using the IBES Guidance database. IBES Guidance
database provides quantitative information of management’s earnings estimations along with
comments directly from the executives about future expectations. Although the earnings forecast
information is provided both on a quarterly and an annual basis, I focus on annual management
earnings forecast information as all other key independent variables of interest – i.e. financing
constraints measures and other control variables - are constructed on an annual basis. Variables
are defined as follows:
Frequency = Number of management earnings forecast issued in year t

IBES
Guidance
Forecast_Error = Absolute value of forecast error deflated by price (i.e., |actual
earnings less management forecast|/EPS), multiplied by 100

IBES
Guidance
Disclose = Indicator variable that equals 1 if there is at least one management
earnings forecast issued in year t, and 0 otherwise

IBES
Guidance

3.2. Empirical Methodology for H1
I follow the standard empirical models in prior management earnings forecast studies
(Lennox and Park 2006, Baik et al. 2011, Lee et al. 2012) to test my hypotheses. All the regressions
include year and firm fixed effects to reduce potential biases from correlated omitted variables.
21

In addition, I cluster standard errors at the firm level to control for residual dependence in my
pooled time-series cross-sectional regression and winsorize each continuous variable at the 1
percent and 99 percent values of their distributions to reduce the influence of extreme observations.
I first test whether equity (debt) constrained firms are positively (not) associated with the
frequency of management earnings guidance, by estimating the following model:
Frequencyit = α + β1Constrainedit + ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (1)

21
The firm fixed effects serve to control for omitted firm-specific variables. As such, I hold constant any omitted
factor that is constant at the firm level across time. Thus, in order for an omitted variable to affect my results, it must
be the case that changes in any such variable is associated with time-series variation in both my measures of financing
constraints and management earnings guidance, which I view as less likely.
15
Where, Constrained indicates either Equity_Constrained or Debt_Constrained which are defined
in the previous section, and control variables are stated below in more detail. My first hypothesis,
H1, predicts that β1 is positive (not positive) for Equity_Constrained (Debt_Constrained)
indicating that firms that need to issue equity in the near future to fund their positive NPV projects
issue more management earnings guidance (than firms that needs to issue debt) to reduce the
information asymmetry in the public stock market.
I include three sets of control variables, which I define in Appendix A. First, I include firm
size (Size), market-to-book ratio (MTB), debt-equity ratio (Leverage), profitability (ROA),
incurring loss (Loss), amount of property, plant, and equipment (PP&E), and general expenses
(SG&A) to account for basic firm characteristics. Next, I include an indicator variable that equals
one if firms’ earnings have increased over the year (Increase), amount of special items in the
financial statements (Special), stability of operating environment (Earn_Vol, RET_Vol, Beta),
business complexity (Foreign_Trans), and market competition (HHI) to capture firms’ financial
characteristics and surrounding environments (Lee et al. 2012). Finally, I include managers’ tenure
(Tenure), institutional ownership (INST_OWN), percentage of independent directors
(Independent), number of analyst following (N_Analyst), and the amount of audit fees paid to
auditors (Ln_Auditfee) to control for firms’ governance characteristics and monitoring intensity
(Baik et al. 2011, Hong et al. 2016).
3.3. Sample Selection and Descriptive Statistics for H1
I retrieve information related to (1) basic financials from Compustat Global and North
America database, (2) stock returns from CRSP, (3) governance and board characteristics from
BoardEx and RiskMetrics, (4) executive compensation components from Execucomp, (5) auditors
from Audit Analytics, and (6) institutional holdings from Thomson & Reuters. Table 1 presents
the sample selection process. Out of non-financial and unregulated firms in the Execucomp
16
universe, I obtain 94,494 firm-year observations that have texted-based financing constraint
measures from the data used in Hoberg and Maksimovic (2015). I exclude 49,838 firm-year
observations that do not have management guidance information and 14,146 firm-year
observations which lack relevant financial data. Consequently, the final sample size for testing H1
is 30,510 firm-year observations.

TABLE 1
Sample Selection Procedure for H1

Table 1 presents the sample selection process. The final sample size for testing H1 is 30,510 firm-year observations.

Panel A in Table 2 provides the descriptive statistics for all the main and control variables
used in the empirical analysis. The positive mean values for ROA of 0.069 and Increase of 0.652
indicate that the firms in the total sample are generally profitable and usually experience increases
in annual earnings. However, more than 19 percent of the firm-year observations experienced a
loss and approximately 31.4 percent of the sample firms operate in foreign areas.
22

22
Since a large percentage of the total sample is firms in S&P 1500 Index, the descriptive statistics coincide with the
characteristics of large listed companies.
Selection Process Number of firm-years
Non-financial firms in Compustat Universe from 2000 to 2013

123,652)

Less: Firms without text-based financing constraints information (29,158)
94,494)

Less: Firms not included in the IBES Guidance Database (49,838)
Less: Firms without sufficient data for calculating control variables (14,146)

Final Sample for testing H1 30,510)
17
TABLE 2
Basic Descriptive Statistics for H1

Panel A – Summary Statistics

Variable N Mean STD Q1 Med Q3
Equity_Constrained 30,510 -0.032 0.125 -0.124 -0.013 0.237
Debt_Constrained 30,510 0.004 0.098 -0.087 0.006 0.163
Frequency 30,510 2.065 3.418 0.000 0.000 4.000
Forecast_Error 11,487 0.153 1.326 0.015 0.032 0.070
Disclose 30,510 0.438 0.512 0.000 0.000 1.000
∆ Equit y 30,384 -0.022 0.189 -0.061 -0.025 0.008
∆ D ebt 30,384 0.031 0.135 -0.022 -0.001 0.019
Size 30,510 6.985 1.603 5.877 6.963 7.852
MTB 30,510 2.997 3.216 1.458 2.198 3.565
Leverage 30,510 0.195 0.215 0.017 0.191 0.384
ROA 30,510 0.069 0.091 0.028 0.059 0.103
Increase 30,510 0.652 0.513 0.000 1.000 1.000
Loss 30,510 0.194 0.293 0.000 0.000 0.000
Earn_Vol 30,510 2.781 90.176 0.047 0.066 0.138
RET_Vol 30,510 0.219 0.171 0.133 0.197 0.284
Beta 30,510 1.833 0.652 0.875 1.352 1.708
PP&E 30,510 0.389 0.401 0.156 0.337 0.635
Special 30,510 -0.017 0.051 -0.020 -0.009 0.000
SG&A 30,510 0.236 0.275 0.085 0.177 0.298
Foreign_Trans 30,510 0.314 0.554 0.000 0.000 1.000
HHI 30,510 0.287 0.212 0.114 0.183 0.306
#_Analyst 30,510 1.486 0.884 0.063 1.218 1.877
Ln_Auditfee 30,510 13.574 1.475 11.013 13.127 13.995

Table 2 provides the descriptive statistics for all the main and control variables used in the empirical analysis. All
the variables are defined in the Appendix A.

Figure A presents the time series trend of the number of firms that provide management
earnings guidance. Although the number of firms providing earnings guidance decreased during
the financial crisis period, the trend becomes relatively stable in the recent period. The pairwise
correlations among the primary variables are reported in Panel B of Table 2. The main variables
of interest, Equity_Constrained and Debt_Constrained, are negatively correlated with each other,
consistent with their definitions.
18

FIGURE A
Times Series Trend of Number of Firms Providing Management Earnings Guidance

Figure A presents the time series trend of number of firms that provide management earnings guidance. Although the
number of earnings guidance providing firms decreased during the financial crisis period, the trend becomes relatively
stable in the recent period.

TABLE 2 (Continued)

Panel B - Pairwise Pearson Correlations for Primary Variables

Correlations significant at the 0.01 level are shown in bold. All variables are defined in the Appendix A.

Equity
Constrained
Debt
Constrained
Forecast
Frequency
Forecast
Error
Disclose #_Analyst
Equity
Constrained
1.0000
Debt
Constrained
-0.1342* 1.0000
Forecast
Frequency
-0.0113 0.0488* 1.0000
Forecast
Error
0.0056 0.0072 -0.0936* 1.0000
Disclose -0.0165 0.0527 0.8780* -0.0497* 1.0000
#_Analyst 0.0864* -0.0425* 0.1205* -0.0163 0.0656* 1.0000
19
3.4. Primary Results for H1
I first test whether relationships stated in my H1 are consistent with the predictions.
Subsequently, I conduct a number of additional tests to strengthen the robustness of my main
results and also further document the underlying mechanisms beyond the main effect.
Testing H1
The result of equation (1) testing H1, which predicts a positive (no) association between equity
(debt) constrained firms and frequency of earnings guidance, is presented in Table 4. Consistent
with H1, I find a positive (1.045) and significant (p < 0.01) coefficient on Equity_Constrained ( β1)
in column (1), indicating that the equity-constrained firms issue their earnings guidance more
frequently to reduce the information asymmetry between the management and potential
shareholders. In contrast, I find a negative (-0.756) and insignificant (p = 0.134) coefficient on
Debt_Constrained in column (2), suggesting that debt-constrained firms do not necessarily
increase their frequency of management earnings guidance. The sharp difference in voluntary
disclosure behavior between the equity-constrained firms and debt-constrained firms implies that
these two types of firms indeed face different levels of information asymmetry with their potential
capital providers. Thus, results presented in Table 4 strongly support my first hypothesis.
Mitigating Endogeneity Concerns
Although the primary results displayed in Table 4 are ran with firm fixed effects to control
for omitted firm-specific variables, there still lies potential endogeneity issue that can affect the
regression outcomes and its corresponding interpretations. To mitigate this concern, I utilize an
instrument – external mutual fund selling shock (MF_Shock) – and show that my primary results
are less likely to suffer from endogeneity problem (Hoberg and Maksimovic 2015). MF_Shock is
an instrumental variable devised in Edmans, Goldstein, and Jiang (2012) which captures the equity
price pressure created by mutual fund trading that is not induced by information or firm fundamentals
but rather by investor flows and market frictions.
20

TABLE 3
Primary Result - Financing Constraint and Management Earnings Guidance Frequency

VARIABLES (1) Frequency (2) Frequency (3) Frequency
Equity_Constrained 1.045*** 1.017***
(0.000) (0.000)
Debt_Constrained -0.756 -0.698
(0.134) (0.165)
Size 0.362*** 0.359*** 0.359***
(0.002) (0.002) (0.002)
MTB 0.020* 0.020* 0.020*
(0.065) (0.065) (0.063)
Leverage -0.034 -0.021 -0.013
(0.910) (0.944) (0.967)
ROA 2.286*** 2.286*** 2.283***
(0.000) (0.000) (0.000)
Increase -0.080* -0.081* -0.080*
(0.067) (0.064) (0.069)
Loss -0.170* -0.170* -0.168*
(0.054) (0.053) (0.056)
Earn_Vol 0.065 0.066 0.065
(0.302) (0.302) (0.303)
RET_Vol -1.367** -1.319* -1.327*
(0.047) (0.057) (0.055)
Beta -0.093* -0.095* -0.094*
(0.092) (0.087) (0.088)
PP&E -0.514 -0.527* -0.519
(0.106) (0.098) (0.102)
Special -1.152** -1.194** -1.154**
(0.014) (0.011) (0.014)
SG&A -0.328 -0.334 -0.315
(0.386) (0.379) (0.406)
Foreign_Trans 0.189* 0.188* 0.189*
(0.073) (0.073) (0.072)
HHI -0.233 -0.250 -0.244
(0.605) (0.581) (0.588)
#_Analyst 0.154*** 0.156*** 0.154***
(0.002) (0.002) (0.002)
Ln_Auditfee -0.126* -0.127* -0.124*
(0.067) (0.067) (0.071)

Observations 30,510 30,510 30,510
R-squared 0.805 0.804 0.805
Year FE YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

The result testing H1, which predicts positive (no) association between equity (debt) constrained firms and frequency
of earnings, is presented in Table 3. Regressions include fixed effects for year and firm. The dependent variable,
Frequency, is the number of management earnings forecast issued in year t. All control variables are defined in the
Appendix A.
21
An important feature of this MF_Shock measure is that it is constructed through using
hypothetical orders projected from their previously disclosed portfolio, capturing the expansion or
contraction of a fund’s existing positions that is mechanically induced by investor flows to and
from the fund. This particular forced mutual fund selling shock is not only more unexpected and
exogenous but also should only impact firms’ disclosure policies through the channel of equity
market liquidity.
23
Thus, unlike the other macroeconomic events (i.e. financial crisis or the
technology bust), this shock should not directly affect the real side of the firm and should satisfy
the exclusion restriction of econometric requirement of being correlated with the equity-
constraints but not directly with the voluntary disclosure.
Using both continuous and indicator variable that equals one for firm-years in the highest
decile of forced mutual fund selling as defined in Edmans et al (2012) and 0 otherwise, I run the
following equation (2) to alleviate the concerns for endogeneity in my primary result.
24

Frequencyit = α + β1Equity_Constrainedit + β2Equity_Constrained•MF_Shockit + β3MF_Shockit
+ ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (2)

MF_Shock is interacted with Equity_Constrained either by using continuous measure or indicator
variables respectively. Since equity-constrained firms are even more likely to be constrained from
unexpected external mutual fund selling shock, I expect β2 of the first interaction term to be
significantly be positive (and β4 of the second interaction term to be insignificant).

23
The forced mutual fund selling shock is uniquely a shock to equity market liquidity and not debt market liquidity.
24
See the Appendix in Edmans et al. (2012) for the further technical details on the construction of MF_Shock. I thank
Alex Edmans for sharing the mutual fund selling shock measure in his website.
22
TABLE 4
Mitigating the Endogeneity Concerns on Primary Result

Using continuous MF_Shock measure Using top decile MF_Shock dummy
VARIABLES (1) Frequency (2) Frequency (3) Frequency (4) Frequency (5) Frequency (6) Frequency

Equity_Constrained 0.177 0.163 0.518 0.521
(0.801) (0.816) (0.408) (0.405)
Equity_Constrained*MF_Shock 0.104** 0.107** 1.868** 1.820**
(0.032) (0.031) (0.018) (0.022)
Debt_Constrained -0.308 -0.220 0.064 0.141
(0.718) (0.796) (0.925) (0.835)
Debt_Constrained*MF_Shock 0.037 0.050 -1.018 -0.707
(0.667) (0.560) (0.413) (0.570)
MF_Shock 0.007* 0.002 0.007* 0.100 0.055 0.101
(0.068) (0.514) (0.058) (0.181) (0.457) (0.176)

Observations 14,695 14,695 14,695 14,695 14,695 14,695
R-squared 0.817 0.816 0.817 0.817 0.816 0.817
Controls YES YES YES YES YES YES
Year FE YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 4 displays the result of equation (2). MF_Shock is either continuous measure or indicator variable that equals one for firm-years in the highest
decile of forced mutual fund selling as defined in Edmans et al (2012) and 0 otherwise. The dependent variable, Frequency, is the number of
management earnings forecast issued in year t. Regressions include fixed effects for year and firm. All other variables are defined in the Appendix
A.
23
The results of equation (2) are shown in Table 4 - Column (1) to (3) show the results using
the continuous MF_Shock measure and column (4) to (6) show the results using the top decile
indicator variable. Consistent with my expectation I find positive and significant coefficients on
the interaction term Equity_Constrained•MF_Shock and insignificant coefficients on the second
interaction term Debt_Constrained•MF_Shock across all the columns. Hence, the concern for
endogeneity issue is slightly mitigated as my primary results are more pronounced for the firms
suffering from the exogenous mutual selling shock.

3.5. Additional Mechanism Tests for H1
In this section, I conduct a series of additional analyses to shed further light and provide
possible underlying mechanisms on my main findings: positive (no) relationship between the
degree of equity (debt) constraint and the frequency of management earnings guidance.
Earnings Guidance Frequency and the Actual Issuance
Although I document that (equity) constrained firms increase the frequency of disclosure to
improve their information environments, it is still unclear whether the enhanced amount of
voluntary disclosure effectively mitigates the information asymmetry problem and eventually
helps constrained firms to actually issue capital. To connect this missing link, I examine whether
increases in management earnings guidance in equity (debt) constrained firms indeed leads to
actual equity (debt) issuance. This will corroborate my key assumption that reducing information
asymmetry by providing more information to the market helps constrained firms raise capital.
Using the post-financing variables from Balakrishan, Core, and Verdi (2014), I run the following
equation (3) to test my prediction.
∆Financeit+1 = α + β1Constrainedit + β2Constrainedit •Frequencyit + β3Frequencyit + ΣβjControlsit
+ ΓtYear_Dummy + ŋiFirm_Dummy + εit (3)

∆Finance = Either net equity ( ∆Equity) or net debt ( ∆Debt) issuances scaled by the
lagged total assets (Balakrishnan et al. 2014)

Compustat
24
TABLE 5
Effect of Financing Constraint and Management Earnings Forecast Frequency on Post Financing

(1) (2) (3) (4)
VARIABLES ∆Equity t+1 ∆Debtt+1 ∆Equity t+1 ∆Debtt+1

Equity_Constrained -0.028 -0.030 0.024
(0.336) (0.317) (0.557)
Equity_Constrained*Frequency 0.010** 0.011** -0.016
(0.043) (0.043) (0.177)
Debt_Constrained -0.019 -0.005 -0.015
(0.677) (0.907) (0.732)
Debt_Constrained*Frequency -0.011 0.009 -0.014
(0.340) (0.253) (0.219)
Frequency 0.001* -0.001 0.001 -0.001
(0.088) (0.437) (0.128) (0.216)

Observations 30,384 30,384 30,384 30,384
R-squared 0.721 0.722 0.722 0.722
Controls YES YES YES YES
Year FE YES YES YES YES
Clustered by Firm Firm Firm Firm
Firm FE YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

The results examining the effect of financing constraint and guidance frequency on the actual issuance are presented in Table 5. The dependent
variable, ∆Equity ( ∆Debt), is net equity issuances scaled by the lagged total assets. Regressions include fixed effects for year and firm. All control
variables are defined in the Appendix A.
25
As I predict that an increase in management earnings guidance will improve the chance of
equity financing for equity constrained forms, I expect to find a positive and significant coefficient
for the interaction term, Equity_Constrained •Frequency. In contrast, I do not expect to find any
significant coefficient for Debt_Constrained •Frequency. The results of equation (3) are shown in
Table 5. Consistent with my prediction, I find a positive (0.010) and significant (p < 0.05)
coefficient on Equity_Constrained •Frequency ( β2) in column (1), indicating that the increase in
management earnings guidance in equity-constrained firms does leads to a higher chance of real
equity issuance in the next period. In contrast, I find a negative (-0.019) and insignificant (p =
0.677) coefficient on Debt_Constrained •Frequency in column (2), suggesting that debt-
constrained firms do not rely on the frequency of management earnings guidance when raising
actual debt.
Ruling out Alternative Explanations - Desire to Issue vs. Actual Issuance
A number of prior studies document a positive association between the frequency of
management earnings guidance and real capital (equity) issuance (Lang and Lundholm 2000; Li
and Zhuang 2012). For this reason, there is a possibility that my results on H1 are not driven by
ex-ante intentions (i.e. desire, eagerness, willingness) to issue equity in the near future, but merely
capture actual issuance in a noisy way. To rule out this explanation, I divide the total sample based
on the firms’ real issuance of equity (debt) in the following period (i.e. issue vs. no-issue) and
conduct the primary analyses for H1 again.
Regardless of ex-post actual issuance, I predict a positive association between ex-ante
intentions to raise equity (i.e. degree of equity-constraints) and frequency of issuing earnings
guidance for both sub-groups. This prediction also implies that I do not expect to find any increase
in frequency of issuing earnings guidance even if the firm have actually issued equity when the
firm is not equity-constrained (i.e. when they did not express their intentions or difficulties to raise
26

FIGURE B
2 (Constrained vs. Unconstrained) by 2 (Issue vs. No-issue) Framework

Figure B summarizes the logic and expectations of the additional test in 2 (equity-constrained vs. unconstrained) by 2 (issue vs. no-issue) matrix form. Regardless
of ex-post actual issuance, I predict positive association between ex-ante intentions to raise equity (i.e. degree of equity-constraints) and frequency of issuing
earnings guidance for both sub-groups. This prediction also implies that I do not expect to find any increase in frequency of issuing earnings guidance even if
the firm have actually issued equity when the firm is not equity-constrained (i.e. when they did not express their intentions to raise equity in advance).

27
TABLE 6
What is Driving All the Differences? Real Issue vs. Intention to Issue

Equity Issue No Equity Issue Debt Issue No Debt Issue Total Sample Total Sample
VARIABLES (1) Frequency (2) Frequency (3) Frequency (4) Frequency (5) Frequency (6) Frequency

Equity_Constrained 1.467** 1.613** 1.351**
(0.040) (0.022) (0.043)
Equity_Constrained*Equity_Issue 0.783
(0.198)
Equity_Issue 0.102*
(0.079)
Debt_Constrained -0.916 -0.974 -0.828
(0.360) (0.115) (0.130)
Debt_Constrained*Debt_Issue 0.214
(0.761)
Debt_Issue 0.028
(0.542)

Observations 9,868 20,642 10,419 20,091 30,510 30,510
R-squared 0.853 0.814 0.845 0.802 0.817 0.817
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 6 displays the sub-sample result of 2 by 2 framework shown in Figure B. The dependent variable, Frequency, is the number of management
earnings forecast issued in year t. Regressions include fixed effects for year and firm. All control variables are defined in the Appendix A.

28
equity in advance). Due to the fact that mitigating information asymmetry is less of a concern for
firms that are not equity-constrained, potential (marginal) cost of increasing voluntary disclosure
should outweigh potential benefit of increasing disclosure.
25

Figure B summarizes the logic and expectations of this additional test in a 2 (equity-
constrained vs. unconstrained) by 2 (issue vs. no-issue) matrix form. I find, in Table 6, that all the
signs and significances of the coefficients of interests are consistently same among the two groups
(i.e. actual issuance vs. no-issuance). The results from this analysis indicate that my results are
indeed driven by the ex-ante desire to issue the capital rather than the actual event of issuance
itself, further extending and refining the findings from prior studies (e.g. Lang and Lundholm
2000).
Increase in Frequency of Disclosure Reducing Information Asymmetry
So far, I have assumed that increased in disclosure reduces the information asymmetry for
equity-constrained firms.
26
To give more support to this assumption and also to strengthen the
causal link, I additionally test whether increase in disclosure actually leads to decrease in
information asymmetry using number of analysts following (#_Analyst) as a proxy. Intuitively,
increase in number of analyst following the equity-constrained firms would inevitably reduce the
information asymmetry between the firm and the investors. Similar to equation (3), I run the
following equation (4) to validate my predictions. I expect the coefficient ( β2) for the first
interaction term (Equity_Constrained •Frequency) to be positive and significant while I expect the
coefficient for the second interaction term (Debt_Constrained •Frequency) to be insignificant.

25
As mentioned in Chapter 2, examples of potential costs of frequent voluntary disclosure can be (1) greater
monitoring due to decreased information asymmetry and (2) risk of giving out proprietary financial information to the
rivals in the competitive industry.
26
Priors studies have already documented that the disclosure leads to less cost of obtaining external finance and higher
stock price (Ajinkya and Gift 1984, Botosan 1997). In addition to this first moment effect (i.e. higher stock price
considerations), I provide additional evidence that the disclosure from equity-constrained firms would have an impact
on second moment effect (i.e. reduce the variance of the stock price through decrease in information asymmetry).
29
TABLE 7
Increase in Frequency of Disclosure Reducing Information Asymmetry (Second Moment)
– Number of Analyst Following

(1) (2) (3)
VARIABLES #_Analyst #_Analyst #_Analyst

Equity_Constrained 0.046 0.038
(0.763) (0.806)
Equity_Constrained*Frequency 0.031** 0.032**
(0.043) (0.041)
Debt_Constrained -0.254 -0.257
(0.184) (0.180)
Debt_Constrained*Frequency 0.016 0.020
(0.773) (0.726)
Frequency 0.006 0.005 0.006
(0.221) (0.285) (0.227)

Observations 30,510 30,510 30,510
R-squared 0.796 0.795 0.796
Controls YES YES YES
Year FE YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 7 displays the result of equation (4). The dependent variable, Frequency, is the number of management earnings forecast issued in year t.
Regressions include fixed effects for year and firm. All control variables are defined in the Appendix A.

30
#_Analystit = α + β1Constrainedit + β2Constrainedit •Frequencyit + β3Frequencyit + ΣβjControlsit
+ ΓtYear_Dummy + ŋiFirm_Dummy + εit (4)

#Analyst = The number of analysts following the firm in the current quarter.

IBES Guidance
Table 7 presents the results of equation (4). Consistent with my expectation, I find positive
and significant values for Equity_Constrained •Frequency on columns (1) and (3) and insignificant
values for Debt_Constrained •Frequency on columns (2) and (3), indicating that the increase in
frequency of earnings guidance is positively associated with the number of analyst following that
proxies for decrease in information asymmetry.

3.6. Robustness Tests for H1
Lastly, I conduct a number of tests to check whether my primary findings are robust to other
characteristics of management earnings guidance and other types of voluntary disclosure. Due to
the fact that the management earnings guidance measures that I use as the main dependent
variables to proxy for voluntary disclosures are generally quantitative in nature, I also use
qualitative voluntary disclosure measures as my dependent variables to enhance the convergent
validity of the primary results.
Likelihood of Issuing Management Earnings Guidance
Firms that have already been providing management earnings guidance can choose to
increase the frequency of the guidance to reduce their information asymmetry problem. However,
for the group of firms that have never issued management earnings guidance before, they can
choose to begin issuing earnings guidance if they want to actively convey their information to
potential investors. To see if this is the case, I replace Frequency with Disclose to examine whether
the association predicted in H1 also holds when non-disclose firms convert to disclose firms by
estimating equation (5).
Discloseit = α + β1Constrainedit + ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (5)

31
Panel A in Table 8 reports the result of equation (5). The positive (1.256) and significant (p
< 0.05) coefficient on Equity_Constrained in column (1) and negative (-0.514) and insignificant
(p > 0.10) coefficient on Debt_Constrained in column (2) indicates that firms convert from non-
disclose firms to disclose firms when they are equity-constrained, whereas no change in voluntary
disclosure is observed when firms are debt-constrained. Result on jointly controlling for both types
of financing constraints shown in column (3) is also consistent. Overall, the results shown in Panel
B are consistent with H1.
Accuracy of Management Earnings Forecast
Along with the frequency of the management earnings guidance, the accuracy of the forecast
is also one of the important features of disclosure. Moreover, the (equity) constrained firms that
increase their frequency of earnings guidance would not benefit themselves by reducing
information asymmetry if their disclosure is inaccurate. Thus, by running the following equation
(6), I predict that the equity-constrained firms would issue more accurate earnings forecasts
conditioning on disclosing more. Accordingly, I expect negative and significant coefficient for
Equity_Constrained •Frequency and insignificant coefficient for Debt_Constrained •Frequency.
Panel B in Table 8 reports the result of equation (6). All in all, the results are consistent with my
expectation that the equity-constrained firms issue more accurate earnings forecast (conditioning
on the fact that they disclose more to reduce information asymmetry).
Forecast_Errorit = α + β1Constrainedit + β2Constrainedit •Frequencyit + β3Frequencyit
+ ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (6)

Qualitative Disclosures - Comments Spoken by Top Executives during the Conference Call
My first proxy for qualitative voluntary disclosure is the average percentage of comments
spoken by top management (e.g. CEO and CFO) during conference calls. Li, Minnis, Nagar, and
Rajan (2014) determine the amount of speech (both the number of times that a person spoke as
well as the number of characters spoken) for each individual on the conference call, enabling them
32

TABLE 8
Other Characteristics of Management Earnings Guidance

Panel A: Financing Constraint and Likelihood of Issuing Management Earnings Forecast

VARIABLES
(1)
Disclose
(2)
Disclose
(3)
Disclose

Equity_Constrained 1.256** 1.236**
(0.046) (0.042)
Debt_Constrained -0.514 -0.434
(0.566) (0.628)

Observations 30,510 30,510 30,510
Year FE YES YES YES
Control YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

The results replacing Frequency with Disclose are presented in Panel A of Table 8. Regressions include fixed effects for year and firm. The
dependent variable, Disclose, is an indicator variable that equals 1 if there is at least one management earnings forecast issued in year t, and 0
otherwise. All control variables are defined in the Appendix A.

(Continued on next page)

33
TABLE 8 (Continued)

Panel B: Accuracy of the Forecast Conditioning on Disclosing More

(1) (2) (3)
VARIABLES Forecast_Error Forecast_Error Forecast_Error

Equity_Constrained 0.750 0.815
(0.395) (0.350)
Equity_Constrained*Frequency -0.125** -0.136**
(0.021) (0.022)
Debt_Constrained 1.183 1.331**
(0.167) (0.050)
Debt_Constrained*Frequency -0.102 -0.136
(0.274) (0.133)
Frequency -0.023** -0.018** -0.022**
(0.015) (0.012) (0.023)

Observations 11,487 11,487 11,487
R-squared 0.522 0.521 0.522
Controls YES YES YES
Year FE YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel B of Table 8 shows the result of robustness test which predicts that the equity-constrained firms not only issues more frequent earnings
guidance, but also more accurate guidance compared to debt constrained firms. Regressions include fixed effects for year and firm. The dependent
variable, Error, is the absolute value of forecast error deflated by price (i.e., |actual earnings less management forecast|/EPS), multiplied by 100.
All control variables are defined in the Appendix A.

34
to compute the relative amount of text spoken by a specific individual such as the CEO or CFO.
Since most conference calls are typically quarterly events, they therefore convert the conference
call data to annual observations by averaging across all conference calls for a firm within a fiscal
year. Using the data provided from Li et al. (2014), I test whether my primary findings in the
previous section are robust to qualitative voluntary disclosures.
Panel A of Table 9 displays the results on average percentage of comments spoken by top
management during conference calls. Consistent with my prediction, while I find a positive (0.185)
and significant (p < 0.10) coefficient on Equity_Constrained in column (1), I find a negative (-
0.114) and insignificant (p = 0.323) coefficient on Debt_Constrained in column (2). The sharp
contrast among the two main coefficients in columns (1) and (2) indicates that the top management
in equity-constrained firms are more active in providing incremental explanations during
conference calls to reduce the information asymmetry than managers in debt-constrained firms.
Qualitative Disclosures – Firm-initiated Press Releases
My second proxy for qualitative voluntary disclosure is the total number of firm-initiated
press releases (Dai et al. 2015) of a given firm-year.
27
Since firm-initiated press releases not only
cover news related to earnings and financials but also other information that can additionally
inform investors, equity-constrained firms may also use the media channel to convey their firm
related information to the potential shareholders. Subsequently, I replace my main dependent
variable (Frequency) with Natural Log(# of firm-initiated press releases for the given firm-year)
in equation (1) and re-estimate my primary results once more. The results displayed in Panel B of
Table 9 indicates that the equity-constrained firms take advantage of firm-initiated press releases
much more than debt-constrained firms, as expected.

27
I retrieve firm-initiated press release information from the Raven Pack Database.
35
TABLE 9
Qualitative Disclosures

Panel A: Top Management in Action during Conference Calls

Dependent Variable
Average % of comments spoken by CEO and CFO during the conference call
(1) (2) (3)
Equity_Constrained 0.185* 0.186*
(0.067) (0.066)
Debt_Constrained -0.114 -0.116
(0.323) (0.313)

Observations 1,475 1,475 1,475
R-squared 0.780 0.779 0.780
Year FE YES YES YES
Controls YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel A of Table 9 displays the result on average percentage of comments spoken by top management during the
conference call. Regressions include fixed effects for year and firm. All control variables are defined in the Appendix
A.

Panel B: Firm-initiated Press Releases

Dependent Variable
Natural Log (# of firm-initiated press releases for the given firm-year)
(1) (2) (3)
Equity_Constrained 0.188** 0.186**
(0.046) (0.047)
Debt_Constrained -0.052 -0.043
(0.595) (0.661)

Observations 8,168 8,168 8,168
R-squared 0.865 0.865 0.865
Year FE YES YES YES
Controls YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel B of Table 9 displays the result on firm-initiated press releases. Regressions include fixed effects for year and
firm. All control variables are defined in the Appendix A.

(Continued on next page)

36
TABLE 9 (Continued)

Panel C: Readability of 10-K Filings

VARIABLES
(1)
Readability
(2)
Readability
(3)
Readability

Equity_Constrained 2.982* 2.809*
(0.056) (0.073)
Debt_Constrained -4.952** -4.788**
(0.012) (0.015)

Observations 5,855 5,855 5,855
R-squared 0.556 0.557 0.557
Year FE YES YES YES
Controls YES YES YES
Clustered by Firm Firm Firm
Firm FE YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel C of Table 9 displays the result on readability of 10-K filings. Regressions include fixed effects for year and
firm. All control variables are defined in the Appendix A.

Qualitative Disclosures - Readability of 10-K Filings
My final proxy for qualitative voluntary disclosure is the readability of 10-K filings utilized
by the Flesch Reading Ease Index. The Flesch Reading Ease Index rates text on a 100 point scale
and is calculated as 206.835 – (1.015•words per sentence) - (84.6•syllables per word) (Li, 2008).
28

The higher the Flesch Reading Ease index, the easier the text is to understand. Even though 10-K
filings are considered a mandatory disclosure, firms have a significant amount of discretion in the
readability of their 10-K filings. Thus, using the Flesch Reading Ease Index, I examine whether
equity firms are more willing to provide more readable information in their 10-K filings then debt
constrained firms to lessen the information asymmetry problem in the equity market. In Panel C
of Table 9, I find results supporting the prediction that 10-K filings provided by equity (debt)

28
I retrieve conference call information and Flesch Reading Ease Index data from Feng Li’s website.
(http://webuser.bus.umich.edu/feng/) I thank Feng Li for providing the database.
37
constrained firms are more (less) readable. Overall, empirical evidence on qualitative voluntary
disclosure is generally consistent with my main results on quantitative voluntary disclosure (e.g.
management earnings guidance).
Evidence from the Practice
To get some insights from the field, I also conduct a brief interview with a former CEO of
equity-constrained firm. The young equity-constrained firm which I interviewed was in R&D
intensive technology industry where they faced considerable amount of competition and
proprietary information risk. All the firms in that industry raised their fund through equity issuance.
The firm went public at 2007 but was seeking to raise more equity due to their lack of internal
capital to fund their promising M&A and R&D. They provided earnings guidance 4 times a year
since they wanted to remain visible to the potential investors. Most of the firms in this industry
went through SEOs frequently and all of them preferred issuing management earnings guidance at
least 4 times year rather than once a year due to the visibility issue. Moreover, equity compensation
consist huge portion of CEOs’ (executives’) compensation packages in this industry. Although I
wasn’t able to make a causal argument through an interview, I was able to verify that what has
been examined in this paper is consistent with evidence from the practice.

38
Chapter 4. Moderating Effect of Retention Incentives on the Relationship between
Financing Constraints and Voluntary Disclosure: Hypothesis Development

4.1. The Role of Managers’ Retention Incentives and Horizon on Frequency of Disclosure
The relation predicted in my H1 can be moderated by the managers’ horizons which, in turn
are affected by their incentives. Nagar, Nanda, and Wysocki (2003) find that firms’ disclosures,
measured both by management earnings forecast frequency and analysts’ subjective ratings of
disclosure practice, are positively related to the proportion of CEO compensation affected by stock
price and the value of shares held by the CEO. Using management earnings forecasts as a proxy
for voluntary disclosure and the CEO’s portfolio vega to measure the convexity of equity
incentives, Cho, Tsui, and Yang (2015) find a significantly positive association between managers’
vega and the issuance and frequency of forecasts.
29
Overall, these studies suggest that managers’
incentives shaped by compensation contracts increase managers’ tendencies to provide voluntary
disclosures and to help improve incentive alignment between managers and shareholders.
Going further, if managers’ retention incentives (i.e. horizon measured by the weighted
average of the vesting period of the compensation package) become longer (i.e. less short-term
oriented) due to their contract design, it is costly for them to leave the firm before the actual vesting
period as they will have to leave some remaining compensation (money) on the table. For this
reason, managers with long-term horizons in equity-constrained firms
30
have higher incentives to
actually raise (equity) capital and invest in their positive NPV projects
31
to maximize the firm

29
Due to the inevitable costs of voluntary disclosure such as litigation, penalty of not meeting the benchmark, or
giving away proprietary information, managers are normally reluctant to publicly disseminate the information unless
provided appropriate incentives. Nagar et al. (2003) and Cho et al. (2015) suggest that disincentives to disclose can
be outweighed by managers’ contractual terms such as increase in ownership of the firm or convex equity incentives
which secures the managers from the worst negative outcome.
30
Due to the fact that I do not make any predictions for debt-constrained firms in my H1, I continue to predict no
relationship between debt-constrained and voluntary disclosure even when managers’ retention incentives are
considered.
31
Generally, positive NPV investments in equity-constrained firms should be risky R&D projects.
39
value – i.e. managers’ retention incentives can be transformed into a strong commitment to raise
equity.
32
One key issue is that the investment (i.e. growth option) outcomes in equity-constrained
firms are only observed in more distant future. A manager with longer horizon is more incentivized
to make the (risky) investment and furthermore be willing to work hard (i.e. expend hidden effort)
to improve the informational environment by increasing the disclosure to reduce the cost of capital
when the equity is raised.
33

Since risky investments such as R&D projects can be extremely time sensitive, delaying the
project might be costly to the firms as their peers also race for R&D projects in the competitive
industry (Kim, Gopal, and Hoberg 2016). For this reason, R&D projects are more beneficial if
they are invested in an urgent manner rather than being delayed. However, prior studies document
that there are multiple short-term costs associated with R&D investments. R&D spending has a
negative impact on short-term accounting and stock performance (Dechow and Skinner 2000) due
to the fact that (1) R&D expenses are typically immediately expensed under U.S. GAAP and (2)
R&D projects often have greater uncertainty of the future benefits (Chan et al. 2001; Kothari et al.
2002, Cheng 2004).

32
Again, I note that equity-constrained firms in general are in high-technology R&D intensive industries where they
face high degree of competition. They are normally young-growth firms that need to fund their investments mainly
through equity and experience frequent corporate restructuring. These salient characteristics of equity-constrained
firms imply that the managerial labor market faced by equity-constrained firms, as well as capital market, is unstable
– i.e. it is unlikely that the manager (or the founder) of the equity-constrained firms will stay in the firm longer than
managers in other mature industries. Due to the fact that the value of firm depends on the long-term payoff from their
investments, equity-constrained firms normally pay out equity compensation to lengthen their managers’ horizon
(Erkens 2009).
33
Short-term horizon managers may still want to maximize the (short-term) value of the firm by enhancing the chance
of actual equity issuance through voluntary disclosure so that the stock price increases. Consequently, short-term
horizon managers can receive more money when they sell their shares upon (or even before) leaving the firm. In other
words, (1) if the current stock price reflects investor expectations of all future cash flows, and (2) assuming that there
is no agency problem between the manager and shareholders, short-term horizon managers in equity-constrained firms
can also actively issue earnings guidance. Thus, short-duration managers may also disclose (inaccurate or overly
optimistic) earnings guidance if she wants to boost up the stock price and leave the firm. However, the argument and
evidence stated afterwards make short-term horizon managers difficult to pursue this short-term strategy.
40

FIGURE C
BASIC TIMELINE

If managers’ retention incentives (i.e. horizon measured by the weighted average of vesting period of compensation package) become larger (i.e.
less short-term oriented) due to their contract design, it is costly for them to leave the firm before the actual vesting period as they will have to leave
their remaining compensation (money) on the table. For this reason, managers with long-term horizons in equity-constrained firms have higher
incentives to actually raise the (equity) capital and invest in their positive NPV projects to maximize the firm value – i.e. managers’ retention
incentives can be transformed into a strong commitment to raise equity. Key issue is that the investment (i.e. growth option) outcomes in equity-
constrained firms are only observed in more distant future. A manager with longer horizon is more incentivized to make the (risky) investment and
furthermore be willing to work hard (i.e. expend hidden effort) to improve the informational environment by increasing the disclosure to reduce
the cost of capital when the equity is raised.

41
As a result, current stock prices likely do not fully reflect the future benefits of R&D
spending (Lev and Sougiannis 1996) and short-term horizon managers in equity-constrained firms
who want to boost current accounting earnings have incentives to reduce R&D spending (Baber
et al. 1991; Dechow and Sloan 1991). Thus, if the short-term horizon managers are reluctant to
invest in their risky projects (i.e. R&D), they clearly have less incentive to disclose earnings
guidance. In contrast, long-term horizon managers in equity-constrained firms do not face any
problem regarding short-term stock price decline as their compensation package is vested after the
(positive) outcome of R&D project is realized (i.e. long-term value of the firm is maximized),
making them willing to issue earnings guidance more frequently.
Intense market competition can also have a strong impact on disclosure policy. Managers
are likely giving away sensitive information to the competitors which they can react on if the firm’s
future earnings guidance is revealed more frequently. (Foucalt and Fresard 2014). Whereas long-
duration managers can bear this type of short-term cost to maximize the long-term value of the
firm which their equity compensation is tied to, short-duration managers have less incentive to
bear this cost as frequent disclosure can lead to decrease in firm’s stock price while increasing peer
firms’ stock price through the learning effect.
Furthermore, given that rational investors in the (semi-strong efficient) public market already
know (1) whether firms are equity (debt) constrained and (2) the horizon of managers through
managers’ pay duration, real intention of short-duration managers – i.e. boost up the short-term
stock price – can easily be perceived. Thus, if short-term horizon (duration) managers also know
that the market will undervalue their disclosure in the first place, they should have less incentive
to issue guidance. However, this will not happen to long-duration managers since market also
knows the true intention – i.e. maximizing the long-term value of the firm - of long-duration
42
managers. Consequently, among the managers in equity-constrained firms, managers with longer
horizons are more motivated to actively issue earnings guidance than managers with shorter
horizons to reduce information asymmetry so as to avoid failure in raising funds.
34
Please also see
Appendix C for an analytical model that recapitulates the above arguments for following
hypothesis.
H2A: Managers with longer horizons in equity (debt) constrained firms will (do not need
to) to provide earnings guidance more frequently.
35

4.2. The Role of Managers’ Retention Incentives and Horizon on Accuracy of Disclosure
Accuracy of the earnings guidance is also another important factor to consider, in addition
to the frequency of the earnings guidance, when examining firms’ voluntary disclosure behavior.
Over 90 percent of managers surveyed by Graham et al. (2005) indicate that developing a
reputation for accurate and transparent reporting is a key factor motivating their voluntary
disclosures, including earnings forecasts (Healy and Palepu 2001; Skinner 1994; Stocken 2000).
Moreover, Hutton and Stocken (2007) report that the median stock price reaction to good news is
larger for firms with a reputation for providing accurate (and frequent) forecasts than for firms
without this reputation. Hui and Matsunaga (2015) also find changes in the annual bonus for both
the CEO and CFO are positively associated with changes in management forecast accuracy. Lee,
Matsunaga, and Park (2012) find that the probability of CEO turnover is positively related to the
magnitude of absolute forecast errors when firm performance is poor and that this positive relation
holds for both positive and negative forecast errors. In sum, evidence from these studies suggests
that the labor market rewards managers who make accurate earnings forecasts.

34
Conversely, managers in debt-constrained firms are not motivated to issue earnings guidance regardless of their
horizon since they can use their private communication channels with lenders.
35
In other words, the relationship documented in H1A and H1B is more (not) pronounced for firms in which managers
have longer (shorter) horizons.
43
Based on the findings from prior research, managers with longer horizons in financially
constrained firms will want to make not only frequent earnings forecasts, but also accurate earnings
forecasts to avoid any negative consequences from the capital market. Due to the fact that it is
beneficial for the long term horizon managers to stay in the firm longer, they have more incentives
to (1) put in effort to make accurate forecasts, (2) raise capital, and (3) invest in positive NPV
projects.
36
Since investments are normally long-term growth options, longer duration managers
have strong incentives to maximize the value of the growth option by focusing on capital raising
event through enhancing the frequency of voluntary disclosure.
On the other hand, managers with shorter horizon will have different incentives. As all (or
most of) their (equity) compensation should be vested in short period, they would have incentives
to maximize the firm value to increase their equity compensation when they can sell the shares
upon leaving the firm. However, they would have less incentive to put in their effort to issue
accurate earnings guidance if the cost of leaving the firm earlier than real earnings announcement
of forecasted earnings is low. Similar to the argument proposed for H1A, if rational investors fully
anticipate the incentives of short-term horizon managers in the equity constrained firms, the firm
value (i.e. stock price) will not be maximized even though the managers issue frequent earnings
guidance due to the less credibility in their disclosure. Thus, marginal benefit of issuing earnings
guidance for short-term horizon managers in the equity-constrained firms is less than that
compared to long-term horizon managers when the accuracy of earnings guidance is jointly
considered.
37

36
Events stated in (1), (2), and (3) are in a chronological order.
37
There are also reasons such as (1) long-term reputation building issues related to corporate disclosures and (2) career
concern issues that would lead to similar conclusion as well. This gives long-duration managers more incentives to
put more effort to be accurate (conditioning on disclosing more earnings guidance), while making short-duration
managers not disclose at all from the beginning.
44
Following the previous discussion, I posit that managers with longer horizons in equity-
constrained firms will issue not only frequent earnings guidance but also accurate earnings
guidance. I therefore make the following prediction for my latter half of the second hypotheses
38
:
H2B: Managers with longer horizon in equity (debt) constrained firms will (don’t need
to) to provide more accurate earnings guidance.

FIGURE D
Summary of the Hypotheses

Figure D shows the summary of the relationship between the two hypotheses.

38
See Figure D for a complete summary of the relationships between the hypotheses in the paper.
45
Chapter 5. Moderating Effect of Retention Incentives on the Relationship between
Financing Constraints and Voluntary Disclosure: Empirical Result

5.1. Variable Description for H2
I measure managers’ retention incentives and horizon using the duration of executive
compensation. Gopalan, Milbourn, Song, and Thakor (2014) develop a novel measure of executive
pay duration that reflects the vesting periods of different pay components, thereby quantifying the
extent to which compensation is short-term. Following Gopalan et al. (2014), I construct pay
duration as the weighted average of the vesting periods of the components of CEO compensation
- salary, bonus, restricted stock grants, and stock option grants - with the weight being the relative
size of each component using the Incentive Lab database which provides information related to
the vesting periods of equity compensation.
39
While the stock-based compensation measure
employed in prior work implicitly assumes that restricted stock grants and stock option grants have
equal vesting periods, the measure of pay duration explicitly incorporates the length of the vesting
schedules of different stock or option grants. Thus, the cost of the managers leaving the firm
increases in the duration of the managers’ pay, which further leads to an increase in managers’
retention incentives and horizon. Equation (7) shows how the pay duration variable is calculated
in detail.
𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑖𝑡
=
(Salary + Bonus) x 0 + ∑ Restricted stock
𝑖 x 𝑡 𝑖 𝑛 𝑠 𝑖 =1
+ ∑ Option
𝑗 x 𝑡 𝑗 𝑛 𝑜 𝑗 =1
Salary + Bonus + ∑ Restricted stock
𝑖
𝑛 𝑠 𝑖 =1
+ ∑ Option
𝑗
𝑛 𝑜 𝑗 =1
(7)

Where, ti equals the number of years over which the corresponding compensation component is to
be vested. ti decreases by 1 every year. See Appendix D for more information on the pay duration
measure.

39
The vesting periods for salary and cash bonuses are naturally zero.
46
5.2. Empirical Methodology for H2
My multivariate model for my second hypotheses tests whether the relationship documented
in equation (1) is further pronounced under managers with a longer duration – i.e. managers that
have higher retention incentives and longer horizons as stated in the H2A. Augmenting the model
shown in equation (1), I use the following model to examine the moderating effect of pay duration
on the relationship between financing constraints and forecast frequency.
Frequencyit = α + β1Constrainedit + β2Constrainedit •Long_Durationit + β3Long_Durationit
+ ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (8)

Where, Long_Duration indicates a dummy variable that equals to 1 if the managers’ duration is in
the top decile of the total sample, and 0 otherwise. H2A predicts that β2 is positive (not positive)
for Equity_Constrained (Debt_Constrained), indicating that the positive (no) relationship between
equity (debt) constrained firms and frequency of management earnings guidance is more (not)
pronounced under the firms with managers that have longer pay duration.
Lastly, I run equation (9) to test H2B which states that managers with a longer horizon in equity
(debt) constrained firms are likely (unlikely) to provide more accurate earnings guidance.
Forecast_Errorit = α + β1Constrainedit + β2Constrainedit •Long_Durationit + β3Long_Durationit
+ ΣβjControlsit + ΓtYear_Dummy + ŋiFirm_Dummy + εit (9)

H2B predicts that β2 is negative (not negative) for Equity_Constrained (Debt_Constrained),
indicating that longer horizon managers in equity-constrained firms not only provide more
frequent earnings forecasts but also more accurate earnings forecasts than those managers in the
debt-constrained firms.
5.3. Sample Selection and Descriptive Statistics for H2
From 8,796 firm-year observations which generated the primary result for H1, I lose 4,140
firm-year observations that are not included in Incentive Lab database for calculating managers’
pay duration. The final sample size for H2 is 4,656 firm-year observations. Table 10 provides the
47
descriptive statistics for Pay_Duration variable used in the empirical analysis. The mean value of
1.696 for Pay_Duration indicates that the weighted average vesting period of managers’
compensation package is less than 2 years, on average. However, managers’ in the highest decile
of the distribution have compensation packages that vest over more than 3 years. Moreover,
positive correlation between Equity_Pay and Frequency indicates that managers who receive
equity compensation are more likely to issue earnings guidance. Equity_Pay and Pay_Duration
are also positively correlated as Equity_Pay constitutes a major portion of Pay_Duration.
40

TABLE 10
Summary Statistics for H2

Panel A of Table 10 presents the sample selection process. The final sample size for testing H2 is 4,656 firm-year
observations.

Variable N Mean STD P10 Q1 Med Q3 P90
Pay_Duration 4,656 1.696 0.903 0.840 1.102 1.452 2.085 3.085

Correlations significant at the 0.01 level are shown in bold. All variables are defined in the Appendix A.

40
However, note that Equity_Pay captures the amount of managers’ equity compensation while Pay_Duration
captures the average vesting period of managers’ total compensation package.
Selection Process Number of firm-years
Final Sample for testing H1 (30,510)

Less: Firms not included in Execucomp, BoardEx, Incentive Lab
Database and without pay duration information

(25,854)

Final Sample for testing H2 4,656
Forecast_Frequency Forecast_Error Equity_Pay
Forecast_Frequency 1.0000
Forecast_Error -0.0957* 1.0000
Equity_Pay 0.0766* -0.0033 1.0000
Pay Duration 0.0360* -0.0240 0.2163*
48
5.4. Primary Results for H2
Testing H2A
Table 11 displays the result of H2A which predicts that the association documented in H1
is more pronounced in firms that have managers with longer pay duration. Consistent with my
prediction, while I find a positive (4.601) and significant (p < 0.05) coefficient on
Equity_Constrained •Long_Duration in column (1), I find a negative (-1.742) and insignificant (p
> 0.10) coefficient on Debt_Constrained •Long_Duration in column (2). The sharp contrast among
the two main coefficients in columns (1) and (2) indicates that managers in equity-constrained
firms with higher retention incentives and longer horizon – i.e. longer pay duration – are more
likely to increase the frequency of management earnings guidance to alleviate the information
asymmetry problem, while managers in debt-constrained firms do not consider increasing earnings
guidance regardless of their retention incentives and horizon.
Testing H2B
Table 12 shows the result of H2B which predicts that the managers in equity-constrained
firms that have high pay duration not only issue more frequent earnings guidance, but also more
accurate guidance compared to the same set of managers in debt constrained firms. As expected,
whereas I find a negative (-1.125) and significant (p < 0.05) coefficient on
Equity_Constrained •Long_Duration in column (1), I find a positive (0.562) and insignificant (p >
0.10) coefficient on Debt_Constrained •Long_Duration in column (2). Results in column (1)
indicate that the managers in equity constrained firms with higher retention incentives and longer
horizons – i.e. longer pay durations – have more incentives to issue more accurate earnings
guidance as they also increase the frequency. However, the result in column (2) indicates that
neither the accuracy of earnings guidance (nor the frequency) is a key concern for managers in
debt constrained firms. Overall, the results provided in Table 11 and Table 12 strongly support H2.
49
TABLE 11
Moderating Effect of Pay Duration on the Relationship between
Financing Constraint and Management Earnings Guidance Frequency

VARIABLES (1) Frequency1 (2) Frequency1 (3) Frequency2 (4) Frequency2 (5) Frequency3 (6) Frequency3
Equity_Constrained -0.728 -0.731 -0.344
(0.565) (0.592) (0.771)
E_Con*Long_Dur 4.601** 4.351* 3.522*
(0.029) (0.058) (0.071)
Debt_Constrained -1.500 -1.911 -1.344
(0.193) (0.131) (0.191)
D_Con*Long_Dur -1.742 -1.020 -1.580
(0.509) (0.726) (0.502)
Long_Dur 0.317* 0.168 0.246 0.100 0.312* 0.198
(0.083) (0.322) (0.211) (0.577) (0.066) (0.192)

Observations 4,656 4,656 4,656 4,656 4,656 4,656
R-squared 0.064 0.063 0.060 0.060 0.068 0.068
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 11 displays the result of H2A which predicts that the association documented in H1 is more pronounced in firms that have managers with
longer pay duration. Regressions include fixed effects for year and firm. The dependent variable, Frequency, is the number of management earnings
forecast issued in year t. All control variables are defined in the Appendix A.

50

TABLE 12
Moderating Effect of Pay Duration on the Relationship between
Financing Constraint and Management Earnings Guidance Accuracy

VARIABLES (1) Error1 (2) Error1 (3) Error2 (4) Error2 (5) Error3 (6) Error3
Equity_Constrained 0.347 0.601* 0.475
(0.197) (0.091) (0.127)
E_Con*Long_Dur -1.125** -1.294** -1.172**
(0.038) (0.029) (0.046)
Debt_Constrained -0.077 0.539 0.575
(0.826) (0.437) (0.411)
D_Con*Long_Dur 0.562 -0.112 -0.109
(0.298) (0.852) (0.858)
Long_Dur -0.055 -0.011 -0.086 -0.022 -0.086 -0.029
(0.138) (0.737) (0.129) (0.726) (0.117) (0.633)

Observations 2,208 2,208 2,208 2,208 2,208 2,208
R-squared 0.163 0.158 0.195 0.192 0.196 0.196
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 12 shows the result of H2B which predicts that the managers in equity-constrained firms that have high pay duration not only issues more
frequent earnings guidance, but also more accurate guidance compared to the same set of managers in debt constrained firms. Regressions include
fixed effects for year and firm. The dependent variable, Error, is the absolute value of forecast error deflated by price (i.e., |actual earnings less
management forecast|/EPS), multiplied by 100. All control variables are defined in the Appendix A.

51
5.5. Additional Analyses and Robustness Tests for H2
Ruling out Alternative Explanations - Retention Incentives vs. Amount of Incentives
I use a pay duration measure devised in Gopalan et al. (2014) to capture managers’ retention
incentive and horizon. Although this pay duration measure - weighted average of vesting periods
of the components of CEO compensation with the weight relative to the size of each component –
clearly captures the time (horizon) element of managerial incentives. The measure is also highly
correlated with the portion total compensation paid in equity. Thus, to rule out the alternative
explanation that my main results on pay duration (H2A, H2B) are driven by the amount of equity
incentives, I replace Long_Duration with High_Equity in equations (8) and (9), and examine
whether the results are consistent. High_Equity is an indicator variable that equals 1 if the firm is
ranked in the top decile for their manager’s ratio of equity pay to total compensation and 0
otherwise. Results shown in Table 13 indicate that neither of the relationships predicted in H2A
or H2B are observed using High_Equity as the key moderator. Thus, I conclude that it is not the
magnitude of the equity incentives, but the managers’ retention incentives and horizon that drive
the main results.
Earnings Guidance Frequency and the Actual Issuance
In Chapter 3, I find that the increase in frequency of earnings guidance for equity-
constrained firms leads to increase in equity financing for the following period. Table 14 presents
the results of equation (3) when I split the total sample into the highest and lowest pay duration
quintiles. Similar to my baseline results, I find that the relationship documented in Table 14 holds
only under equity constrained firms with high pay duration managers. Overall, the results in Table
14 strongly support my assumption that an increase in the frequency of voluntary disclosures
(jointly with high retention incentives) indeed helps equity constrained firms issue equity in the
near future.
52
TABLE 13
Placebo Test - Moderating Effect of Equity Pay on Financing Constraint and Management Earnings Forecast

Panel A: Management Earnings Forecast Frequency

VARIABLES (1) Frequency1 (2) Frequency1 (3) Frequency2 (4) Frequency2 (5) Frequency3 (6) Frequency3
Equity_Constrained 1.029* 1.346** 1.049*
(0.099) (0.037) (0.067)
E_Con*High_Equity 0.304 -0.080 0.177
(0.779) (0.944) (0.852)
Debt_Constrained -0.276 -0.519 -0.285
(0.655) (0.416) (0.609)
D_Con*High_Equity -1.963 -1.310 -1.678
(0.180) (0.397) (0.192)
High_Equity -0.214** -0.225** -0.229** -0.229** -0.166** -0.174**
(0.021) (0.013) (0.017) (0.016) (0.045) (0.035)

Observations 6,493 6,493 6,493 6,493 6,493 6,493
R-squared 0.053 0.053 0.054 0.053 0.062 0.062
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel B: Management Earnings Forecast Accuracy

VARIABLES (1) Error1 (2) Error1 (3) Error2 (4) Error2 (5) Error3 (6) Error3
Equity_Constrained 0.445 0.487 0.476
(0.430) (0.413) (0.402)
E_Con*High_Equity 0.178 0.002 -0.239
(0.809) (0.998) (0.743)
Debt_Constrained 0.739 1.247* 1.112*
(0.164) (0.051) (0.064)
D_Con*High_Equity -0.672 -0.967 -0.701
(0.112) (0.116) (0.179)
High_Equity 0.040 0.036 0.028 0.026 -0.001 0.003
(0.613) (0.569) (0.739) (0.700) (0.991) (0.957)

Observations 2,996 2,996 2,996 2,996 2,972 2,972
R-squared 0.047 0.048 0.075 0.077 0.072 0.074
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

53

TABLE 14
Joint Effect of Financing Constraint, Management Earnings Forecast Frequency, and Pay Duration on Post
Financing

Frequency1 Frequency2 Frequency3
Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D
VARIABLES
(1) ∆Equity (2) ∆Equity (3) ∆Debt (4) ∆Debt (5) ∆Equity (6) ∆Equity (7) ∆Debt (8) ∆Debt (9) ∆Equity (10) ∆Equity (11) ∆Debt (12) ∆Debt
Equity_Constrained -0.057 0.064 -0.054 0.084 -0.055 0.081
(0.622) (0.478) (0.639) (0.364) (0.633) (0.389)
Equity_Con*Frequency 0.038* 0.011 0.036* 0.003 0.040* 0.005
(0.070) (0.580) (0.075) (0.866) (0.089) (0.833)
Debt_Constrained 0.018 0.351 0.030 0.341 0.033 0.325
(0.884) (0.110) (0.822) (0.219) (0.807) (0.172)
Debt_Con*Frequency 0.001 -0.165*** -0.002 -0.189* -0.004 -0.185**
(0.970) (0.005) (0.931) (0.097) (0.883) (0.041)
Frequency 0.003** 0.002 -0.002 -0.009 0.004** 0.002 -0.004* -0.008 0.004** 0.002 -0.003 -0.008
(0.044) (0.317) (0.468) (0.219) (0.017) (0.324) (0.054) (0.222) (0.036) (0.481) (0.214) (0.264)

Observations 2,872 1,784 2,863 1,784 2,872 1,784 2,863 1,784 2,872 1,784 2,863 1,784
R-squared 0.386 0.265 0.238 0.337 0.387 0.265 0.242 0.326 0.386 0.264 0.240 0.316
Firm 268 292 267 292 268 292 267 292 268 292 267 292
Year FE YES YES YES YES YES YES YES YES YES YES YES YES
Controls YES YES YES YES YES YES YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

The results examining the joint effect of financing constraint, guidance frequency, and pay duration on actual issuance are presented in Table 14.
Regressions include fixed effects for year and firm. The dependent variable, ∆Equity ( ∆Debt), is net equity issuances scaled by the lagged total
assets. All control variables are defined in the Appendix A.

54
Qualitative Disclosures - Comments Spoken by Top Executives during the Conference Call
Similar to Chapter 3, I test whether my primary findings on H2 are robust to qualitative
voluntary disclosures. However, contents in qualitative disclosures are more likely to contain
proprietary information (which can potentially alter firms’ future earnings (Dye 1985)) than
quantitative earnings guidance. For this reason, I control for proprietary costs with a competition
measure based on the text based network industry classification measure (TNIC3) constructed in
Hoberg and Phillips (2010).
41
Table 15 presents the result of the moderating effect of pay-duration.
Although I do not find any significant results in the total sample, column (3) through column (8)
indicates that positive association between the average percentage of top management
explanations during conference calls and the degree of equity constraints is indeed more
pronounced under managers with longer pay durations, but only when the firm faces low product
market competition.
Private Debt vs. Public Debt
I further decompose debt-constrained firms into firms that intend to issue private debt (i.e.
bank loan) and firms that intend to issue public debt (i.e. bond) by dividing the total sample into
firms that have experienced borrowing a loan from the bank and firms that have never issued a
private loan. Using the DealScan database, I identify firms that have loan rate (All-in-drawn spread)
record up to year 2013 (i.e. end of my sample period) and consider these firms as private debt-
constrained firms as the debt-constraint score increases. If firms do not have any record in the
DealScan database, I consider these firms as public debt-constrained firms as the debt-constraint
score increases.
42
Since firms are not likely to be debt constrained if they can easily have access
to the public bond market, I have assumed that debt-constrained firms seek private loan in general.

41
I retrieve product market competition data from the Hoberg and Phillips data library. (http://hobergphillips.usc.edu/)
A huge thank you to Gerald Hoberg and Gordon Phillips for making the data public.
42
I acknowledge that this is a crude classification since even though firms are not captured in DealScan database,
firms still might be seeking to issue private loan for the first time.
55
TABLE 15
Moderating Effect of Pay Duration on Qualitative Voluntary Disclosures

Dependent Variable: Average % of comments spoken by CEO and CFO during the conference call
Degree of Competition (1)Total (2) Total (3) Low (4) Low (5) Medium (6) Medium (7) High (8) High
Equity_Constrained 0.449** 0.569 0.217 -0.134
(0.029) (0.481) (0.315) (0.716)
E_Con*Long_Dur -0.218 2.427*** 0.544 -1.790**
(0.598) (0.001) (0.295) (0.012)
Debt_Constrained -0.225 -0.333 0.161 -0.263
(0.237) (0.318) (0.619) (0.495)
D_Con*Long_Dur 0.577 2.235* 0.240 1.967
(0.153) (0.074) (0.678) (0.163)
Long_Dur -0.020 -0.016 -0.159 -0.127 0.079 0.053 0.056 -0.009
(0.460) (0.649) (0.117) (0.187) (0.179) (0.363) (0.233) (0.833)

Observations 478 478 158 158 159 159 161 161
R-squared 0.154 0.137 0.475 0.438 0.558 0.552 0.281 0.273
Firms 249 249 94 94 91 91 91 91
Year FE YES YES YES YES YES YES YES YES
Controls YES YES YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Table 15 presents the result on moderating effect of pay-duration. The dependent variable is the average percentage of comments spoken by top
management during the conference call. Regressions include fixed effects for year and firm. All control variables are defined in the Appendix A.

56

TABLE 16
Public vs. Private Debt

Public Debt Private Debt Public Debt Private Debt Public Debt Private Debt
VARIABLES (1) Frequency (2) Frequency (3) Frequency (4) Frequency (5) Error (6) Error
Debt_Constrained -0.665 -0.933* -2.075 -1.093 -0.385 -0.094
(0.503) (0.093) (0.292) (0.369) (0.253) (0.827)
D_Con*Long_Dur 7.900** -4.453* -0.696 0.804
(0.044) (0.083) (0.154) (0.247)
Long_Dur 0.095 0.264 -0.005 -0.008
(0.660) (0.195) (0.894) (0.862)

Observations 2,381 6,415 1,289 3,367 585 1,623
R-squared 0.069 0.060 0.110 0.082 0.291 0.212
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

The results examining different voluntary disclosure behavior among public debt-constrained firms and private debt-constrained firms are shown
in Table 16. As expected, the main coefficients of interest in all columns indicate that voluntary disclosure behavior of public debt-constrained
firms and private debt-constrained firms behave differently in marginal significance. The results in Table 8 provide suggestive evidence that long-
term horizon managers in public debt-constrained firms issue more frequent and accurate earnings guidance than managers in private debt-
constrained firms.

57
Consistent with my expectation, 6,415 firm-year observations (i.e. 73% of the total sample) had
experience issuing a private loan out of 8,796 firm-year observations.
Using this classification, I run my primary analyses for H1 and H2 again to see whether firms
seeking private loan and firms seeking public bond exhibit different voluntary disclosure behavior.
Specifically, I expect firms that public debt-constrained firms mimic more similar to equity-
constrained in terms of their voluntary disclosure behavior as public bond are more similar to
equity than debt in terms of their characteristics and features. The results are tabulated in Table 16.
As expected, the main coefficients of interest in all columns indicate that voluntary disclosure
behavior of public debt-constrained firms and private debt-constrained firms behave differently in
marginal significance. Although I refrain from making a strong conclusion, the results in Table 16
provide suggestive evidence that long-term horizon managers in public debt-constrained firms
issue more frequent and accurate earnings guidance than managers in private debt-constrained
firms.
Pay Duration of Top Management Team and the Pessimistic Nature
Several studies find that other executives in the top management team also have a strong
impact on firms’ voluntary disclosure policies (Bamber et al. 2010; Jiang et al. 2010; Hui and
Matsunaga, 2015). To verify that my main results are consistent with prior findings, I conduct the
managers’ pay duration analyses for H2 by extending the sample from firm-year-CEO
observations to firm-year-executive observations.
43
Untabulated results using the extended
observations are consistent with the main findings that top management teams with longer pay
duration – i.e. higher retention incentives and longer horizon – in equity constrained firms, issue

43
Using the compensation vesting period information of the top 5 executives, which is available in the Incentive Lab
database, I calculate top executives’ pay duration in the same way that I calculated CEO’s pay duration.
58
TABLE 17
Moderating Effect of Pay Duration on Financing Constraint and Optimistic vs. Pessimistic Management Earnings Forecast

Panel A: Interaction Test
VARIABLES (1) Pessmistic1 (2) Pessmistic1 (3) Pessmistic2 (4) Pessmistic2 (5) Pessmistic3 (6) Pessmistic3
Equity_Constrained -0.190 -0.257 -0.094
(0.531) (0.585) (0.828)
E_Con*Long_Dur 0.924
*
1.203* 0.945
(0.102) (0.060) (0.134)
Debt_Constrained 0.100 -0.586 -0.571
(0.791) (0.427) (0.443)
D_Con*Long_Dur
-0.548 -0.019 -0.067

(0.336) (0.979) (0.923)
Long_Dur 0.053 0.019 0.046 -0.010 0.035 -0.006
(0.172) (0.558) (0.470) (0.865) (0.564) (0.918)

Observations 1,123 1,123 1,123 1,123 1,114 1,114
R-squared 0.084 0.082 0.225 0.226 0.234 0.235
Year FE YES YES YES YES YES YES
Controls YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Panel B: Sub-Sample Test
Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D Long_D Short_D
VARIABLES (1) Pess1 (2) Pess1 (3) Pess1 (4) Pess1 (5) Pess2 (6) Pess2 (7) Pess2 (8) Pess2 (9) Pess3 (10) Pess3 (11) Pess3 (12) Pess3
Equity_Con 1.069* -0.349 1.545** -0.645* 1.540** -0.560*
(0.050) (0.191) (0.014) (0.060) (0.013) (0.083)
Debt_Con -0.372 -0.142 -0.715 -0.216 -0.733 -0.140
(0.428) (0.587) (0.234) (0.531) (0.223) (0.668)

Observations 1,436 925 1,436 925 1,436 925 1,436 925 1,420 921 1,420 921
R-squared 0.154 0.428 0.133 0.422 0.385 0.540 0.366 0.531 0.389 0.556 0.370 0.547
Year FE YES YES YES YES YES YES YES YES YES YES YES YES
Controls YES YES YES YES YES YES YES YES YES YES YES YES
Clustered by Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm Firm
Firm FE YES YES YES YES YES YES YES YES YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)
59
more frequent and accurate earnings guidance, compared to same set of top management teams in
debt constrained firms.
As a last analysis for this dissertation, I test how managers with high retention incentives
documented in H2B reduce the forecast errors. Table 17 shows that managers with high retention
incentives in the equity constrained firms issue earnings forecasts that are slightly lower than the
actual value in order to avoid any potential criticism of being overly optimistic from the investors
in the near future.

Chapter 6. Conclusion

This study examines the effect of financing constraints (i.e., firms’ desire to raise equity or
debt in the near future, as they face the risk of having to delay positive NPV investment projects
if they fail to raise capital) on voluntary disclosure behavior. While I expect that equity constrained
firms are more likely to actively increase the frequency of their management earnings guidance to
reduce their information asymmetry problems in the public market than debt constrained firms, I
also predict that this relationship is more pronounced when managers have higher retention
incentives and longer horizons. Consistent with my prediction, I find that (1) equity constrained
firms indeed issue more earnings guidance than debt constrained firms and (2) this result is driven
by managers in equity constrained firms that have longer pay durations. I also find that these
managers in the equity constrained firms not only issue more frequent earnings guidance but also
more accurate guidance.
Additional analysis shows that the improved information environment associated with an
increase in the frequency of management earnings guidance leads to an increase in actual (equity)
issuance in the following period. Moreover, I also verify that my main results are driven by firms’
60
ex-ante intentions to raise capital rather than the actual issuance itself since I find identical results
in a group of firms that actually end up raising capital and a group of firms that do not issue capital
ex-post. Further, given that management earnings guidance is a quantitative voluntary disclosure
in nature, I also show that my main results are robust to qualitative voluntary disclosure as well.
Overall, my findings suggest that equity constrained firms and debt constrained firms have
different voluntary disclosure policies as the former needs to raise capital through the public stock
market while the latter mainly finances debt capital through private relationships with lenders. Due
to the fact that it is more costly for managers with longer pay durations to leave the firm, managers
in equity constrained firms have more incentives to actively resolve the information asymmetry
problem by issuing more frequent and accurate guidance. This eventually leads to investment in
positive NPV projects after successfully raising actual capital.

61
Bibliography

Aboody, D., and Kasznik, R., 2000. CEO Stock Option Awards and the Timing of Corporate
Voluntary Disclosures. Journal of Accounting and Economics 29(1), 73-100.

Ajinkya, B. B., and M. J. Gift. 1984. Corporate Managers’ Earnings Forecasts and Symmetrical
Adjustments of Market Expectations. Journal of Accounting Research 22(2), 425–444.

Anilowski, C., Feng, M., and Skinner, D. J., 2007. Does Earnings Guidance Affect Market Returns?
The Nature and Information Content of Aggregate Earnings Guidance. Journal of Accounting
and Economics 44(1-2), 36-63.

Baber, W., P. Fairfield, and J. Haggard., 1991. The Effect of Concern about Reported Income on
Discretionary Spending Decisions: The Case of Research and Development. The Accounting
Review 66, 818–829.

Baik, B., Farber, D., Lee, S., 2011. CEO Ability and Management Earnings Forecasts.
Contemporary Accounting Research 28(5),1645-1668.

Balakrishnan, K., Billings, M. B., Kelly, B. and Ljungqvist, A., 2014. Shaping Liquidity: On the
Causal Effects of Voluntary Disclosure. The Journal of Finance 69(5), 2237–2278.

Balakrishnan, K., Core, J., and Verdi, R., 2014. The Relation between Reporting Quality and
Financing and Investment: Evidence from Changes in Financing Capacity. Journal of
Accounting Research 52(1), 1–36.

Bamber, L., Jiang, J., and Wang, I., 2010. What’s My Style? The Influence of Top Managers on
Voluntary Corporate Financial Disclosure. The Accounting Review 85(4), 1131-1162.

Bharath, S. T., Dahiya, S., Saunders, A., and Srinivasan, A., 2011. Lending Relationships and
Loan Contract Terms. Review of Financial Studies 24(4), 1141-1203.

Botosan, C. A., 1997. Disclosure Level and the Cost of Equity Capital. The Accounting Review
72(3), 323-49.

Chan, L., J. Lakonishok, and T. Sougiannis., 2001. The Stock Market Valuation of Research and
Development Expenditures. Journal of Finance 56, 2431–2456.

Cheng, Q., and K. Lo., 2006. Insider Trading and Voluntary Disclosures. Journal of Accounting
Research 44(5), 815–848.

Cheng, S., 2004. R&D Expenditures and CEO Compensation. The Accounting Review 79(2), 305-
328.

62
Cho, Y., Tsui, D., and Yang, H., 2015. The Role of Convex Equity Incentives in Managers’
Forecasting Decisions. Working paper, University of Southern California and Singapore
Management University.

Dechow, P., and D. Skinner., 2000. Earnings Management: Reconciling the Views of Accounting
Academics, Practitioners, and Regulators. Accounting Horizons 14, 235–250.

Dechow, P., and R. Sloan., 1991. Executive Incentives and Horizon Problem. Journal of
Accounting and Economics 14, 51–89.

Diamond, D. W., and Verrecchia, R. E., 1991. Disclosure, Liquidity, and the Cost of Capital. The
Journal of Finance 46(4), 1325–1359.

Dye, R., 1985. Disclosure of Nonproprietary Information. Journal of Accounting Research, 23(1),
123-145.

Erkens, D., K. Subramanyam, J. Zhang., 2014. Affiliated Bankers on Board and Conservative
Accounting. The Accounting Review 89 (5), 1703-1728.

Foucault, T., and L. Fresard., 2014 Learning from peers' stock prices and corporate investment,
Journal of Financial Economics 111(3), 554-577.

Farre-Mensa, J., and Ljungqvist, A., 2016. Do Measures of Financial Constraints Measure
Financial Constraints? Review of Financial Studies 29(2), 271-308.

Francis, J., Philbrick, D., and Schipper, K., 1994. Shareholder Litigation and Corporate
Disclosures. Journal of Accounting Research, 32(2), 137-164.

Gopalan, R., Milbourn, T., Song, F. and Thakor, A. V., 2014. Duration of Executive Compensation.
The Journal of Finance 69(6), 2777–2817.

Graham, J., Harvey, C. R., and Rajgopal, S., 2005. The Economic Implications of Corporate
Financial Reporting. Journal of Accounting and Economics 40(1-3), 3–73.

Healy, P., and Palepu, K., 2001. Information Asymmetry, Corporate Disclosure, and the Capital
Markets: A Review of the Empirical Disclosure Literature. Journal of Accounting and
Economics 31(1-3) 405–440.

Hirst, D. E., Koonce, L., and Venkataraman, S., 2008. Management Earnings Forecasts: A Review
and Framework. Accounting Horizons 22(3), 315-338.

Hoberg, G., and Maksimovic, V., 2015. Redefining Financial Constraints: A Text-Based Analysis.
Review of Financial Studies 28(5), 1312-1352.

63
Huang, Y., Jennings, R., and Yu, Y., 2016. Product Market Competition and Managerial
Disclosure of Earnings Forecasts: Evidence from Import Tariff Rate Reductions. The
Accounting Review In-Press.

Hui, K, W., and Matsunaga, S. R., 2015. Are CEOs and CFOs Rewarded for Disclosure Quality?
The Accounting Review 90(3), 1013-1047.

Hutton, A., and Stocken, P., 2007. Effect of Reputation on the Credibility of Management
Forecasts. Working paper, Boston College and Dartmouth College.

Jiang, J. X., Petroni, K., and Wang, Y. I., 2010. CFOs and CEOs: Who Have the Most Influence
on Earnings Management? Journal of Financial Economics 96(3), 513-526.

Kim, K., Gopal, A. and Hoberg, G., 2016. Product market competition and corporate venture
capital investment in the it industry: An empirical analysis. Information Systems Research,
27(2), pp.259-281.

Kothari, S. P., T. Laguerre, and A. Leone., 2002. Capitalization Versus Expensing: Evidence on
the Uncertainty of Future Earnings from Capital Expenditures Versus R&D Outlays. Review of
Accounting Studies 7, 355–382.

Krasker, W. S., 1986. Stock Price Movements in Response to Stock Issues under Asymmetric
Information. The Journal of Finance 41(1), 93–105.

Lang, M. H., and Lundholm., R. J., 2000. Voluntary Disclosure and Equity Offerings: Reducing
Information Asymmetry or Hyping the Stock? Contemporary Accounting Research 17(4), 623–
662.

Lee, S., Matsunaga, S., and Park, C., 2012. Management Forecast Accuracy and CEO Turnover.
The Accounting Review 87(6), 2095-2122.

Lennox, C., and Park, C, W., 2006. The Informativeness of Earnings and Management’s Issuance
of Earnings Forecasts. Journal of Accounting and Economics 42(3), 439–458.

Leuz, C., and R. Verrecchia. 2000. The Economic Consequences of Increased Disclosure. Journal
of Accounting Research 38(Supplement), 91–124.

Lev, B., and T. Sougiannis., 1996. The Capitalization, Amortization, and Value-Relevance of
R&D. Journal of Accounting and Economics 21, 107–138.

Li, O. Z., and Zhuang, Z., 2012. Management Guidance and the Underpricing of Seasoned Equity
Offerings. Contemporary Accounting Research 29(3), 710–737.

Lo, A. K., 2014. Do Declines in Bank Health Affect Borrowers’ Voluntary Disclosures? Evidence
from International Propagation of Banking Shocks. Journal of Accounting Research 52(2),
541–581.
64
Murphy. 1999. Executive compensation. In Handbook of Labor Economics (3B). Amsterdam, The
Netherlands: North-Holland.

Myers, S. C., 1977. Determinants of Corporate Borrowing. Journal of Financial Economics 5(2),
147-175.

Myers, S. C., and Majluf, N., 1984. Corporate Financing and Investment Decisions When Firms
Have Information that Investors Do Not Have. Journal of Financial Economics 13(2), 187-221.

Nagar, V., Nanda, D., and Wysocki, P., 2003. Discretionary Disclosure and Stock-Based
Incentives. Journal of Accounting and Economics 34(1-3), 283-309.

Rampini, A. A., and Viswanathan, S., 2010. Collateral, Risk Management, and the Distribution of
Debt Capacity. The Journal of Finance 65(6), 2293–2322.

Shivakumar, L., Urcan, O., Vasvari, F. P., Zhang, L., 2011. The Debt Market Relevance of
Management Earnings Forecasts: Evidence from Before and During the Credit Crisis. Review
of Accounting Studies 16(3), 464 -486.

Shleifer, A., and Vishny, R., 1989. Management Entrenchment: The Case of Manager-Specific
Investments. Journal of Financial Economics 25(1), 123-139.

Skinner, D. 1994. Why Firms Voluntarily Disclose Bad News. Journal of Accounting Research
32(1-2), 38–60.

Stocken, P., 2000. Credibility of Voluntary Disclosure. The Rand Journal of Economics 31(2),
359–374.

Verrecchia, R. 2001. Essays on disclosure. Journal of Accounting and Economics 32(1-3) 97–180.

65
Appendix A
Variable Definitions

Variable Definitions Data Source
Equity_Constrained = Hoberg-Maksimovic Equity-Delay Score (Hoberg
and Maksimovic 2015). Firms with higher values
are more similar to a set of firms that (1) are at risk
of delaying their investments due to liquidity issues
and (2) that indicate plans to issue equity
(presumably to address their liquidity challenges).

SEC Filings
Debt_Constrained = Hoberg-Maksimovic Debt-Delay Score (Hoberg
and Maksimovic 2015). Firms with higher values
are more similar to a set of firms that (1) are at risk
of delaying their investments due to liquidity issues
and (2) that indicate plans to issue debt (presumably
to address their liquidity challenges).

SEC Filings
Long_Duration = A dummy variable that equals to 1 if the managers’
pay duration is in the top decile of the total sample,
and 0 otherwise.

Incentive Lab
Frequency = Number of management earnings forecast issued in
year t. Frequency 1 (2) indicates the frequency for
current (future) period, and Frequency 3 is
combined variable of Frequency 1 and 2.
IBES Guidance
Forecast_Error = Absolute value of forecast error deflated by price
(i.e., |actual earnings less management
forecast|/EPS), multiplied by 100.
IBES Guidance
Disclose = Indicator variable that equals 1 if there is at least one
management earnings forecast issued in year t, and
0 otherwise.
IBES Guidance
∆Equity = Net equity issuances scaled by the lagged total
assets (Balakrishnan et al. 2014).

Compustat
∆Debt = Net debt issuances scaled by the lagged total assets
(Balakrishnan et al. 2014).

Compustat
Control Variables
Ln(Assets) = Natural log of market value of equity at the end of
fiscal year.
Compustat
Market-to-Book = Ratio of the market value of total assets to the book
value of total assets.
Compustat
Leverage = The sum of long-term debt and the current portion
of long-term debt divided by total assets.
Compustat

(Continued on next page)

66
APPENDIX A (Continued)

Variable Definitions Data Source

Control Variables
Asset Growth = Annual growth in total assets scaled by lagged total
assets.
Compustat
ROA = Net income less special items divided by total assets. Compustat
Increase = Indicator variable that equals 1 if the firm’s current
earnings increased compared to the previous
quarters’ earnings, and 0 otherwise.

Compustat
Loss = Indicator variable that equals 1 if the firm’s current
earnings is negative, and 0 otherwise.

Compustat
Earn_Volatility = The standard deviation of quarterly earnings scaled
by the total assets over the past 4 years.

Compustat
Ret_Volatility = The standard deviation of daily raw stock returns
over the last three years.

CRSP
Beta = Market model beta, estimated over the past year. CRSP
PP&E = Gross property, plant, and equipment divided by
total assets.
Compustat
SG&A = SG&A expenses deflated by total assets. Compustat
Inst_Own = The percentage of institutional ownership at the
beginning of the quarter t.

Thompson &
Reuters
Independent = The number of independent directors scaled by the
total number of directors in the board of the firm.

BoardEx
#Analyst = The number of analysts following the firm in the
current quarter.

IBES Guidance
FORTRANS = Indicator variable equals to 1 if the company reports
non-zero foreign currency translation, and 0
otherwise.
Compustat
Tenure = The number of years that the CEO has held the
position of chief executive officer.

BoardEx
Ln_Boardsize = The natural log of total number of board members. BoardEx
Ln_Auditfee = The natural log of audit fees paid to the auditor. Audit
Analytics
BigN = Indicator variable equals to 1 if the firm is audited
by one of the Big 4 auditors, and 0 otherwise.
Audit
Analytics
Segment = The natural log of the number of operating segment
plus the number of geographic segment.
Compustat
STD_OCF = Standard deviation of operating cash flow deflated
by total assets from period t-2 to t.
Compustat

(Continued on next page)

67
APPENDIX A (Continued)

Variable Definitions Data Source

Control Variables
Restatement = An indicator variable that equals 1 for firm-years in
which firm's reported earnings are restated for the
previous three years, and otherwise 0.

Audit
Analytics
MKT_Share = The percentage of the company’s sales to the total
sales of its industry, where industry is defined based
on three-digit SIC codes.
Compustat
Ln_Totalpay = Natural log of total compensation paid to top
executives. Total compensation includes total
current compensation, pension changes, restricted
stock granted, stock awards, and stock option
granted.

Execucomp
Dual = Indicator variable that equals 1 if CEO is also the
chairman of the board of directors, and otherwise 0.

Boardex

68
Appendix B
Validating Hoberg and Maksimovic (2015)’s Financial Constraint Measures

VARIABLES (1)∆Equity t+1 (2)∆Debt t+1 (3)∆Equity t+1_High (4)∆Debt t+1_High
Equity_Constrained 0.026* 1.113**
(0.085) (0.050)
Debt_Constrained 0.046** 1.614***
(0.019) (0.002)
Zscore -0.002*** 0.000 -0.038* -0.020**
(0.001) (0.573) (0.062) (0.023)
Cash_constrained 0.026*** -0.031*** 1.546*** 0.924***
(0.000) (0.000) (0.000) (0.000)
Size -0.011*** 0.004*** -0.534*** 0.246***
(0.000) (0.000) (0.000) (0.000)
MTB -0.011*** 0.002 -0.243*** 0.073*
(0.000) (0.259) (0.000) (0.096)
Asset_growth 0.045*** -0.007 1.263*** -0.316***
(0.000) (0.190) (0.000) (0.007)
RET 0.010*** 0.002 0.307*** -0.019
(0.000) (0.349) (0.000) (0.740)
STD_Sale -0.008 0.013 0.366 -0.029
(0.448) (0.229) (0.307) (0.911)
STD_CFO 0.110** -0.020 6.776*** -0.880
(0.037) (0.192) (0.000) (0.119)
Industry_leverage 0.002 -0.002 0.108 0.139
(0.593) (0.779) (0.399) (0.325)
CFO_Sale -0.067*** 0.031** -1.738*** -0.012
(0.000) (0.014) (0.001) (0.967)
CEO_Age 0.000 -0.000** -0.006 -0.005
(0.588) (0.016) (0.306) (0.225)
Operating_cycle 0.002 0.002 0.170* 0.119**
(0.467) (0.299) (0.091) (0.044)
Slack -0.000*** 0.000* -0.001 0.000
(0.000) (0.069) (0.274) (0.268)
Constant 0.044** -0.006 3.574*** -3.188***
(0.020) (0.757) (0.000) (0.000)

Observations 7,269 7,269 7,269 7,269
R-squared 0.257 0.064 0.225 0.111
Year FE YES YES YES YES
Clustered by Firm Firm Firm Firm
Industry FE YES YES YES YES
Robust p-value in parentheses (*** p<0.01, ** p<0.05, * p<0.1)

Results in Appendix B show that the Equity_Constrained (Debt_Constrained) variables used in this study is positively
associated with next-period equity (debt) financing, and this relationship is more pronounced when the firms raise
large amount of capital.

69
Appendix C
Model

Parameters
V1 = Value of the (equity-constrained) firm at the end of period 1 without voluntary
disclosure
V1’ = Value of the (equity-constrained) firm at the end of period 1 with voluntary
disclosure
V2 = Value of the (equity-constrained) firm at the end of period 2
p = Probability of having a successful R&D investment at the end of period 2 without
voluntary disclosure
p’ = Probability of having a successful R&D investment at the end of period 2 with
voluntary disclosure
α = Percentage of ownership which short-horizon (less retention incentives) CEO has
α' = Percentage of ownership which long-horizon (greater retention incentives) CEO
has

Assumption
A1: Due to the short-term costs related to frequent disclosure, firm value at the end of period 1
will be less for firms that disclose than firms that do not. (V1’ < V1)
A2: In the long-run (at the end of second period), the probability of R&D project being successful
is greater when the (voluntary) disclosure was made than no disclosure was made during the first
period. (p < p’)

Outcome for Short-term Managers
Situation 1) Leaving the firm after selling the shares at the end of period 1
Disclose = αV1’ < αV1 = No Disclose  Optimal Choice = No Disclose
Situation 2) Staying until the end
Disclose = αp’V2 > αpV2 = No Disclose
 However, if market does not fully believe the disclosure of short-term managers, then it is not
clear whether p would indeed increase to p’ upon disclosure in this case: (p  p rather than p 
p’).

Outcome for Long-term Managers
Disclose = α’p’V2 > α’pV2 = No Disclose  Optimal Choice = Disclose

70

APPENDIX D
Details on Pay Duration Measure

Operationalizing the Pay Duration Measure
Following Gopalan et al. (2014), I plan to measure pay duration as the weighted average of the vesting periods of the components
of CEO compensation: salary, bonus, restricted stock grants, stock option grants with the weight being the relative size of each
component.

𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑖𝑡
=
(Salary + Bonus) x 0 + ∑ Restricted stock
𝑖 x 𝑡 𝑖 𝑛 𝑠 𝑖 =1
+ ∑ Option
𝑗 x 𝑡 𝑗 𝑛 𝑜 𝑗 =1
Salary + Bonus + ∑ Restricted stock
𝑖
𝑛 𝑠 𝑖 =1
+ ∑ Option
𝑗
𝑛 𝑜 𝑗 =1
(1)

Where, ti equals the number of years which the corresponding compensation component is to be vested. ti decrease by 1 every year.
See Appendix B for more information on the pay duration measure. While the stock-based compensation measure employed in prior
work implicitly assumes that restricted stock grants and stock option grants have equal vesting periods, the pay duration measure
explicitly incorporates the length of the vesting schedules of different stock or option grants.
A principal objective of all the measures trying to capture manager’s horizon is to understand the mix of short-term and long-
term pay and hence the extent to which overall pay provides short-term incentives to executives. These existing measures include
the proportion of stock and option grants (noncash pay) in total pay, the delta and vega of the executive’s stock and option holdings,
and the correlation of executive pay with stock returns and accounting earnings. The newly composed pay duration measure is
superior in capturing the CEO’s (executives’) horizon compared to prior measures due to the following reasons.
(1) Just examining the amount of restricted stock or stock option as a proxy for mangers’ horizon can be problematic because
firms can self-select on paying equity compensation to their managers. (2) Pay duration measure explicitly accounts for the length
of the stock and option grants’ vesting schedules. Clearly, a large stock grant itself is unlikely to contribute to short-term managerial
incentives if it has a long vesting schedule. (3) While the delta and vega of an executive’s compensation portfolio capture its
sensitivities to movements in stock price and its volatility, respectively, they do not capture the mix of short-term and long-term
incentives in the pay contract, which pay duration measure does.

(Continued on next page)

71

APPENDIX D (Continued)

Numerical Example of Calculating Pay Duration
Let’s assume that a CEO receives same amount of annual salary ($100) and bonus ($250) every year, and restricted stock A
($700) that is scheduled to be vested in 4 years, restricted stock B ($300) to be vested in 1 year, option A ($500) to be vested in 3
years, and option B ($200) to be vested in 2years in year t. She did not receive additional equity compensation afterwards. The
following chart presents the pay duration of the CEO for each year, calculated based on equation (1).

Year t

Year t+1

Year t+2

Component Amount Period Duration

Component Amount Period Duration

Component Amount Period Duration
Salary 100 0
2.73

Salary 100 0
1.90

Salary 100 0
1.43
Bonus 250 0

Bonus 250 0

Bonus 250 0
Stock A 700 4

Stock A 700 3

Stock A 700 2
Stock B 300 1

Stock B 300 0

Stock B 0 0
Option A 500 3

Option A 500 2

Option A 500 1
Option B 200 5

Option B 200 4

Option B 200 3

As the year goes by, the vesting period of each equity compensation decreases by one and the numerical value of pay duration also
gradually decreases accordingly. If all components of equity compensation were to be vested immediately, the pay duration should
equal to 0. Moreover, if all components of equity compensation were to be vested in 1 year, the pay duration should equal to 0.83
which is the value identical to the ratio of equity pay to total pay.

Abstract (if available)

Abstract This dissertation examines the effect of financing constraints on voluntary disclosure behavior

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Voluntary disclosure responses to mandated disclosure: evidence from Australian corporate tax transparency

PDF

Insider editing on Wikipedia

PDF

Three essays on behavioral finance with rational foundations

PDF

Essays on the firm and stakeholders relationships: evidence from mergers & acquisitions and labor negotiations

PDF

Essays on the effect of cognitive constraints on financial decision-making

PDF

Who manages the firm matters: the incremental effect of individual managers on accounting quality

PDF

Executive compensation: the trend toward one size fits all

PDF

Dividend policy, earnings announcements, and analysts' earnings forecasts

PDF

Essays on information and financial economics

PDF

Studies on the creation of regulations in nascent drone industry

PDF

The effects of accounting performance and professional relationships on promotion, dismissal, and transfer decisions in a conglomerate

PDF

Leveling the playing field: unbiased tests of the relative information content of book income and taxable income

PDF

The interactive effects of incentive threshold and narcissism on managerial decision-making

PDF

Does mandatory adoption of international financial reporting standards decrease the voting premium for dual-class shares: theory and evidence

PDF

Evolution of returns to scale and investor flows during the life cycle of active asset management

PDF

“What you see is all there is”: The effects of media co‐coverage on investors’ peer selection

PDF

Levels of interest: the effects of teachers' unions on Congress, statehouses, and schools

PDF

Effects of flavorings in electronic cigarettes on the use and appeal of e-cigarettes among youth and adults

PDF

The effects of acceptance and commitment therapy-based exercises on eating behaviors in a laboratory setting

PDF

Faith, finances and the remaking of Southern California fundamentalism, 1910–1968

Asset Metadata

Creator Oh, Seung Hwan (Peter) (author)

Core Title The effect of managerial retention incentives on the relationship between financing constraints and voluntary disclosure

School Leventhal School of Accounting

Degree Doctor of Philosophy

Degree Program Business Administration

Publication Date 08/02/2018

Defense Date 04/10/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag disclosure quality,financing constraints,management guidance,OAI-PMH Harvest,pay duration

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Merchant, Kenneth A. (committee chair), Hoberg, Gerald (committee member), Li, Shelley Xin (committee member), Murphy, Kevin J. (committee member)

Creator Email peter.oh@mcgill.ca,seunghwo@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-48522

Unique identifier UC11672121

Identifier etd-OhSeungHwa-6613.pdf (filename),usctheses-c89-48522 (legacy record id)

Legacy Identifier etd-OhSeungHwa-6613.pdf

Dmrecord 48522

Document Type Dissertation

Format application/pdf (imt)

Rights Oh, Seung Hwan (Peter)

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 a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA