“Capital structures of surviving Fortune 500 companies: A retrospective analysis

for the past seven decades”

AUTHORS Wenjuan Xie

Wenjuan Xie (2024). Capital structures of surviving Fortune 500 companies: A

ARTICLE INFO retrospective analysis for the past seven decades. Investment Management and
Financial Innovations, 21(1), 98-115. doi:10.21511/imfi.21(1).2024.09

DOI http://dx.doi.org/10.21511/imfi.21(1).2024.09

RELEASED ON Tuesday, 30 January 2024

RECEIVED ON Friday, 01 December 2023

ACCEPTED ON Thursday, 18 January 2024

LICENSE This work is licensed under a Creative Commons Attribution 4.0 International
License

JOURNAL "Investment Management and Financial Innovations"

ISSN PRINT 1810-4967

ISSN ONLINE 1812-9358

PUBLISHER LLC “Consulting Publishing Company “Business Perspectives”

FOUNDER LLC “Consulting Publishing Company “Business Perspectives”

NUMBER OF REFERENCES NUMBER OF FIGURES NUMBER OF TABLES

34 5 7

© The author(s) 2024. This publication is an open access article.

businessperspectives.org
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Wenjuan Xie (USA)

Capital structures
BUSINESS PERSPECTIVES of surviving Fortune 500
LLC “СPС “Business Perspectives”
Hryhorii Skovoroda lane, 10,
Sumy, 40022, Ukraine
companies: A retrospective
www.businessperspectives.org
analysis for the past seven
decades
Abstract
Sixty companies on the inaugural “Fortune 500” list still remained on this list in 2020
and they have monotonically increased their leverage (median debt to total assets ra-
tio rose from 0.5% to 20.4%) over the past 70 years. This study applies factors from
classic capital structure theories to this sample and explains the dynamic choice of
debt usage. The methods employed include a Bayesian information criterion selection
process of explanatory variables and a set of pooled cross section and panel tests with
Received on: 1st of December, 2023 3,536 firm-year observations. The tests use an array of factors extracted from several
Accepted on: 18th of January, 2024 established theories on capital structure, including general economic growth, tax rate,
Published on: 30th of January, 2024 interest rate and many company-specific variables proxying profitability and growth
opportunities. The firm-level results first provide support to the free cash flow theory
© Wenjuan Xie, 2024
and confirm that company size and fixed assets proportion are the two factors associ-
ated with increased borrowing. Firms in the sample also actively respond to certain
debt market and macroeconomic conditions, and their leverage ratio is significantly
associated with credit spread and real interest rate. Further tests across subperiods
Wenjuan Xie, Ph.D., Associate and with risk measures illustrate the impact of expected inflation, investments activi-
Professor of Finance, Paul College of ties, and stock volatility, providing supporting evidence to the organizational theory.
Business and Economics, University of The main research conclusion is that large US companies adopt a balance sheet-based
New Hampshire, USA. approach to increase the use of debt, and they stay sensitive and versatile to market
conditions and risk landscape.

Keywords leverage, cost of capital, tax, interest rate, large-cap,

capital structure, growth, risk
JEL Classification G30, G32

INTRODUCTION
As capital structure is a central topic in corporate finance, there is a
vibrant strand of studies documenting and explaining the use of debt.
Chronologically, the academic explanations for U.S. publicly traded
corporations were taxes followed by bankruptcy costs and then agency
costs. Today, over 65 years have passed since the original Modigliani
and Miller (1958) piece set in motion what Myers (1993) called “the
search for an optimal capital structure.” During this time, the US
economy experienced many changes in the tax code, interest rates,
equity capital costs, inflation, government stimulus, corporate profit-
This is an Open Access article, ability, economic growth, and international competitions. A series of
distributed under the terms of the theories have also been developed to explain the dynamics of financial
Creative Commons Attribution 4.0
International license, which permits leverage in corporate America.
unrestricted re-use, distribution, and
reproduction in any medium, provided
the original work is properly cited. This study applies factors extracted from classic capital structure theo-
Conflict of interest statement: ries to the examination of large public companies’ capital structure,
Author(s) reported no conflict of interest and it aims to address the research question on what key factors from

98 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

the firms and the macroeconomy are significantly associated with increased use of debt. The research
design in this paper rigorously selects factors, comprehensively models a panel data sample, and pro-
vides support to the free cash flow theory and organizational theory.

1. LITERATURE REVIEW structure that maximizes (minimizes) the market

AND HYPOTHESES value (cost of capital) of a company. The literature
went on to identify more disadvantages associated
This study joins and contributes to the contem- with increases of leverage, such as the agency costs
porary theories of capital structure, which have created by conflicts of interest between creditors
their origin in Modigliani and Miller’s 1958 semi- and shareholders (Jensen & Meckling 1976). As a
nal work, “The cost of capital, corporation finance, firm takes on more and more debt, creditors be-
and the theory of investment.” This article is gener- come concerned about the “games” managers and
ally regarded as one of the watershed pieces that owners can play that will benefit the managers
led to the application of scientific principles to and owners at the expense of the creditors. Hence,
problems in finance even though it provides no creditors place more and more restrictions in the
explanation for the financing decisions observed loans and charge higher interest rates. These credi-
in practice (Smith, 1990). Since then, countless pa- tor reactions exacerbate the financial distress costs
pers have been written about the financial struc- and offset the tax advantages of borrowing. The
ture decision with the aim of explaining possible static trade off theory guides this study by identify-
connections between the financing and the mar- ing key factors that have been found to be associat-
ket value of a company. For example, Rajan and ed with leverage and summarized in many studies
Zingales (1995) study international data, Titman such as Hennessy and Whited (2005), Barry et al.
(2002) applies the theorem to integrated financial (2008), Campello et al. (2010), and He et al. (2021):
markets, and Welch (2004) and Warusawitharana tax rate, bankruptcy (financial distress) cost, mac-
and Whited (2016) further examine the dynamic roeconomic conditions, and agency cost. Free cash
of capital structure and equity capital markets. flow theory is then developed with a specific iden-
tification of the conflicts of interest between man-
For expository purposes, the theories can be cat- agers and shareholders, as elaborated in Jensen
egorized as the static tradeoff theory, the free cash and Meckling (1976) and further developed in
flow theory and the pecking order theory, which the last few decades by numerous studies such as
is more a description of observed behavior than Jensen (1986), Graham and Harvey (2001), Fee et
a theory per se. However, the theories are not mu- al. (2013), and Graham et al. (2015). Here, agency
tually exclusive. The following paragraphs indi- costs arise because managers expand the firm at
vidually summarize and contrast four theories of the expense of its owners by making negative net
financial structure, laying out the key factors to be present value (NPV) investments and using cash
considered in each theory, and connecting them to consume perks rather than distributing it to the
to the construction of empirical tests in this study. shareholders. To discourage these managerial ac-
tions, firms are advised to finance with debt rath-
The static tradeoff theory recognizes the fact that er than equity to ensure the investments are not
tax plays an important role in the determination made and the cash is not used for the benefit of
of capital structure. Modigliani and Miller show the managers but rather to finance the company’s
that an advantage for debt financing over equity growth. This theory provides further agency-cost
financing existed when interest was a tax-deduct- related variables as mentioned in Frank and Goyal
ible expense (Modigliani & Miller, 1963). However, (2009) and to be added to this study: firm size,
the use of debt is accompanied by disadvantages. growth opportunities, and asset structure.
Two of the first to be identified are bankruptcy and
financial distress costs. Bankruptcy and financial The pecking order theory further develops the de-
distress costs are assumed to increase with lever- bate by proposing the key assumption of the exis-
age ratios, and eventually these costs offset the tence of asymmetric information between manag-
tax benefits of debt leading to an optimal capital ers and public investors. As argued in Myers and

http://dx.doi.org/10.21511/imfi.21(1).2024.09 99
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Majluf (1984) and Gorton and Pennacchi (1990), and McKeon (2012), DeAngelo and Roll (2015), and
because managers have more or better informa- Huang and Ritter (2021). Additional factors, such
tion than outside investors, outside investors are as the investment sensitivity to internally available
skeptical whenever managers raise funds by sell- cash flows and the need/preparation for future ex-
ing equity capital. Investors take this as a sign ternal financing and investments (Poursoleyman
that the company is overvalued. This asymmetry et al. 2023), are also associated with maximizing
causes managers to prefer using inside equity (re- long-term organizational wealth, and are included
tention of earnings) as their first financing choice, in this study in the exploration of determinants of
followed by debt and lastly by issuing new shares the structure of debt.
of common stock. The more profitable a firm is,
the less debt or new equity it will need to issue to Guided by these theories and prior empirical re-
finance growth opportunities. Pecking order the- search and observing the stylized fact of almost
ory does not assume an optimal capital structure monotonically increased debt ratio of the large
or target debt ratio. However, it also does not ne- firms in the sample, this study serves the purpose
gate the advantages of tax shields or the costs of of jointly considering all theories and identify-
financial distress. What it predicts, other things ing how their representative factors are associated
being equal, is that firms needing more than inter- with leverage. Three testable hypotheses are thus
nally generated funds for undertaking investment developed as:
opportunities will exhibit increasing leverage ra-
tios. Many succeeding studies provide evidence of H1: The change in leverage ratio is positively as-
the validity of this theory, including Holmström sociated with firm size, information asym-
and Tirole (1998), Lins et al. (2010), and Begenau metry, and a firm’s need for capital.
(2020), among others. Recently using Slovakian
data, Horváthová et al. (2022) find an ROE thresh- H2: The change in leverage ratio is negatively
old based nonlinear leverage trajectory. The key associated with the cost of debt and firm
factors that further help explain leverage in this profitability.
study are thus extracted from previous empirical
studies, including measures of profitability and H3: The change in leverage ratio is positively as-
information asymmetry. sociated with tax rate and the cost of equity.

The final strand of literature focuses on a relatively

newer “organizational theory”. Myers (1993), draw- 2. DATA, METHODS,
ing on the work of Treynor (1981) and Donaldson AND RESEARCH ANALYSIS
(1984), extends the pecking order theory into an or-
ganizational theory of capital structure. These pa- This study employs quantitative multivariate
pers propose that firms maximize organizational methods and disentangle variables having posi-
wealth, defined as the market value of shareholders’ tive and negative explanatory powers as outlined
equity plus a more insider-driven employee surplus in the testable hypotheses. It constructs a sample
(future value of perks, overstaffing and above mar- that covers the 1950-2020 annual data of the 60
ket wages). Organizational theory makes a similar companies that were on the 1955 inaugural list of
prediction as that of the pecking order theory, es- Fortune 500 and remain on the list in 2020. Data
pecially with the abundance of cash. Debt capital sources are COMPUSTAT, CRSP, Bureau of Labor
will be used to fund growth if internally generated Statistics, National Bureau of Economic Research,
funds are not sufficient. However, when investment Federal Reserve Bank of St. Louis, Moody’s, Bureau
opportunities do not increase in proportion to in- of Economic Analysis, Federal Reserve Bank of
ternally generated funds, the “excess” cash flow will New York, and Nobel Laureate Professor Robert
mean that debt financing will not be needed to fi- Shiller’s website. All variable definitions follow
nance the positive NPV investments – over time, prior research and are required to be non-missing.
the debt ratio will fall as existing debt is paid off. In addition, all independent variables need to have
These points have been illustrated in extensive re- one-year lagged values. The final sample has 3,536
search such as Graham and Leary (2011), Denis pooled firm-year observations.

100 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

0.800 0.2

0.600 0.1
0
0.400
-0.1
0.200 -0.2
0.000 -0.3
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
Long Term Debt/ Total Assets Shareholder Equity/ Total Assets
Long Term Debt/ Sales Net Debt/Sales
Note: The left scale applies to LT debt/total assets, Shareholder equity/total assets, and LT debt/sales. The right scale applies
to net debt/sales.

Figure 1. Trends of median leverage ratios

To emphasize the stylized fact that leads to the ety of them in utilizing information from differ-
empirical examinations in this study, a few figures ent financial statements mitigates the concern to a
are first presented. What needs to be explained certain extent. What is readily apparent in Figure
for the surviving Fortune 500 companies is the 1 is the long-term trend increase in financial le-
almost monotonic increase in leverage from 1950 verage since 1950 for the surviving Fortune 500
through 2020. This long-term trend is apparent in companies. The median long-term debt to total as-
Figure 1 where four different measures of finan- sets ratio rose from 0.5% to 20.4% and the median
cial leverage are plotted: long-term debt to total net debt to total sales rose from a negative 0.05%
assets, common stockholders’ equity to total as- to 11.5%. The median long-term debt to sales rose
sets, long-term debt to sales, and net debt (long- from 0.4% to 27.5%. Some cyclicality exists around
term debt minus cash) to total assets. Although the long-term trend – especially in the 1970s and
all measures are book value based and possibly late 2000s – but it is the long-term trend that calls
subject to accounting choice distortions, the vari- for further exploration. So, how well, if at all, do
0.6 0.6
0.5
0.5
0.4
0.3
0.4
0.2
0.1 0.3
0
1950
1953
1956
1959
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
2010

0.2
-0.1
-0.2
0.1
-0.3
-0.4 0
Long Term Debt to Assets Long Term Debt to Sales
Effective Tax Rate Marginal Tax Rate
Note: The left scale applies to the effective tax rate, long-term debt/assets, and net debt/sales. The right scale applies
to the marginal tax rate.

Figure 2. Median leverage and tax rates

http://dx.doi.org/10.21511/imfi.21(1).2024.09 101
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

the factors identified in existing theories of capital Further attention is turned to growth opportuni-
structure explain this trend? ties (proxied by firm size), a key factor in the free
cash flow theory. Figure 4 is a time-series plot of
A few representative factors are plotted next for the yearly mean total assets value expressed in
trend analyses. Figure 2 focuses on the basics of 2005 dollars (converted using the GDP implicit
the static trade-off theory and contains a time- price deflator) as a multiple of its value in 1950. In
series graph of annual marginal and effective tax real terms, the median surviving firm in 2020 was
rates, the median long-term debt to total assets, 36.2 times bigger than it was in 1950. This long-
and net debt to sales ratios. The marginal corpo- term growth in the size of the firms, especially
rate tax rate dropped continuously from 42% to the accelerated growth in the recent two and a
21%. The effective corporate tax rate dropped from half decades, is positively correlated with the in-
40% to a low of 17.2%. During these years, leverage crease in leverage ratios, as illustrated by the mean
ratios rose on average, suggesting that they are not long-term debt to assets ratio in this figure. Not
explained by the trends in either effective or mar- included in this figure but included in later em-
ginal tax rates. pirical analyses are measures of operating leverage
and asset structure (the portion of long-term fixed
Moving on to firm-specific performance, which is assets). A mixed but generally increasing trend in
a key factor in the pecking order theory, Figure 3 fixed assets is consistent with the overall growth
presents a time-series annual plot of selected me- in total assets and with the expectation that in-
dian before tax profitability measures (EBIT to as- creased leverage should be associated with growth
sets and EBIT to sales) against the median long- and increased operating leverage.
term debt to total assets ratio. In general, the debt
to asset ratio has behaved as expected with respect The final quick check before formalizing the em-
to the before tax profit measures in the first two pirical methods framework is driven by the rela-
and a half decades: when before tax profitability tive cost of capital as outlined in Fama and French
exhibited a general trend of decline, leverage went (2002) and recently examined in Wang et al.
up. However, in recent decades the correlation is (2020), which is relevant to all capital structure
mixed: despite an increase in the operating return theories examined thus far. Presumably, decreases
on assets and no trend in operating profit margin, (increases) in the relative cost of equity should be
the debt ratio continues to increase. associated with reductions (increases) in leverage.

0.250 0.3

0.25
0.200

0.2
0.150
0.15
0.100
0.1

0.050
0.05

0.000 0
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010

Long Term Debt to Total Assets EBIT to Sales EBIT to Total Assets

Note: The left scale applies to LT debt/total assets and EBIT/assets. The right scale applies to EBIT/sales.

Figure 3. Median leverage and before-tax profitability

102 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

30 0.4000
0.3500
25
0.3000
20
0.2500
15 0.2000
0.1500
10
0.1000
5
0.0500
0 0.0000
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
Asset size constant dollars long term debt
Note: The left scale applies to asset size expressed in multiples of the 1950 value. The right scale applies to the ratio of long-
term debt/asset.
Figure 4. Mean asset size (2005 constant dollars) vs. mean long-term debt

The question becomes: How to measure the rela- with theories. For example, in the recent decades
tive cost of debt and equity? Figure 5 presents four when P/E ratio and equity risk premium declined,
measures: the 10-year average cyclically adjusted debt ratio did not decrease as expected, but ex-
P/E ratios (cost of equity), the ratio of 10X long- hibited an upward trend. An exception is the ob-
term interest rates to P/E ratios (relative cost of eq- servation that real interest rate is associated with
uity), the yield spread between Baa corporate rat- leverage as expected: in most years of rising inter-
ed debt and long-term government (USLT) bonds est rates, debt ratio declined, and vice versa. The
(10X percentage, risk premium), and the real inter- pattern suggests that leverage was sensitive to the
est rate (the long-term government bond rate mi- level of real interest rates, and it is consistent with
nus the GDP price deflator). Comparing the trends the finding of Barry et al. (2008). They “find strong
of these measures to that of the debt/asset ratio, it evidence that the amount of debt issued and the
is hard to draw clear cut conclusions consistent number of debt issues are related to the absolute

50 0.4000
45 0.3500
40 0.3000
35 0.2500
30 0.2000
25 0.1500
20 0.1000
15 0.0500
10 0.0000
5 -0.0500
0 -0.1000
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010

P/E Ratio i/pe Baa-USLT Debt/Assets Real interest rate

Note: The left scale applies to P/E ratio, 10X interest rate to PE ratio, and 10X Baa – USLT yield spread percentage. The right
scale applies to the ratio of debt/assets and real interest rate.

Figure 5. Cost of capital

http://dx.doi.org/10.21511/imfi.21(1).2024.09 103
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

level of interest rates and to their levels relative to Total debt, long-term debt, and net debt sepa-
historical rates” (Barry et al., 2008, p. 429). rately divided by total book assets. For robustness
checks, the ratio of total debt divided by market
With all these helpful but inconclusive single-pair value of assets is also used.
trend analyses, this study takes a holistic approach
and constructs an empirical framework linking The independent variables are the following one-
leverage to a battery of major variables driving year lagged factors, reflecting a collection of those
the contemporary financial structure theories and representing the various capital theories as well as
a group of macroeconomic variables as summa- those from considerations of the general econo-
rized in Barry et al. (2008). my and debt market conditions. Macroeconomic
variables Yt include: effective tax rate, top statutory
Employing a sample of balanced longitudinal da- tax rate, GDP growth, unemployment rate, reces-
ta and a procedure of multivariate regressions af- sion period dummy, expected inflation rate, credit
ter a Bayesian information criterion (BIC) based spread, real interest rate, and S&P 500 average P/E
core variables vector selection, this study tests ratio. Firm variables Xi,t include: firm size, fixed
the relationship between capital structure (prox- asset proportion, operating income, net income,
ied by leverage ratios) and a series of firm-year market-to-book ratio, and capital expenditure. All
variables (Xi, t) and a vector of macroeconomic firm-level variables are normalized by total assets.
variables (Yt) that are drawn from the theories,
which controlling for the macroeconomic con- Tables 1 presents the summary statistics of all
ditions. The baseline methods include a pooled firm-level variables expressed by decades, with the
cross-sectional regression and a panel data re- recent two decades divided by the 2008 financial
gression, depending on whether controlling for crisis. A first glance discovers many similar pat-
firm fixed effects: terns of the key variables as illustrated by the pre-
vious graphic presentations. For example, the ab-
(
Leverage ratioi ,t = α + β ⋅ X i ,t −1 )
(1)
solute value of debt and the leverage ratios all ex-
hibit monotonic increases, while company char-
+γ ⋅ (Yt −1 ) + ε i ,t . acteristics and macroeconomic conditions have
various turbulences.
All variable definitions can be found in Appendix
A. The dependent variable is mainly the leverage Table 2 presents the mean statistics of all macro-
ratio, and following prior literature, three alter- economic variables expressed by decades, with the
native definitions of book leverage ratio are used: recent two decades divided by the 2008 financial
Table 1. Descriptive statistics of firm variables
1950–1959 1960–1969 1970–1979 1980–1989 1990–1999 2000–2007 2008–2020
Obs. 531 529 529 529 529 423 466
Total Assets 686.51 1429.75 3722.24 11106.34 28269.40 54940.86 67799.72
Total Sales (Revenue) 820.47 1704.73 5019.57 12368.72 22026.18 39012.24 50506.58
Long-term debt 59.17 141.02 463.94 1681.02 5390.20 12941.66 16132.74
Cash 49.53 71.61 76.53 330.42 877.08 2985.53 5750.07
Net debt (Long-term debt minus cash) 9.64 69.41 387.41 1350.60 4513.12 9956.13 10382.67
EBIT 152.94 299.05 796.74 1820.42 3781.92 68888.86 8821.72
EBITDA 119.06 219.40 584.02 1225.59 2592.68 5016.29 6450.98
Long-term debt to total assets 0.09 0.10 0.12 0.15 0.19 0.24 0.24
Shareholders’ equity to total assets 0.00 0.71 0.52 0.40 0.24 0.25 0.30
Long-term debt to total sales 0.07 0.08 0.09 0.14 0.24 0.33 0.32
Net debt to total assets 0.01 0.05 0.10 0.12 0.16 0.18 0.15
EBIT to total assets 0.22 0.21 0.21 0.16 0.13 0.25 0.13
EBIT to total sales 0.19 0.18 0.16 0.15 0.17 0.77 0.17

Note: The total number of observations is 3,536, and decade-mean statistics are reported. All dollar amounts are in thou-
sands, and all ratios are in plain numbers. See Appendix A for detailed variable definitions.

104 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Table 2. Mean statistics of macroeconomic variables

1950-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2007 2008-2020
Obs. 10 10 10 10 10 8 12
Top statutory tax rate 0.51 0.51 0.48 0.43 0.35 0.35 0.32
Effective tax rate 0.44 0.40 0.33 0.31 0.28 0.23 0.17
Unemployment rate (%) 4.51 4.78 6.21 7.27 5.75 5.04 8.34
NBER defined recession years 4.00 3.00 5.00 3.00 2.00 1.00 2.00
Baa corporate bond annual yield (%) 3.91 5.64 9.30 12.81 8.54 7.09 6.31
Long-term government bond annual rate (%) 2.90 4.56 7.29 10.55 6.60 4.81 3.07
GDP price inflation annual rate (%) 2.46 2.41 6.66 4.69 2.22 2.56 1.70
Real interest rate (%) 0.43 2.15 0.63 5.86 4.37 2.25 1.37

Note: See Appendix A for detailed variable definitions.

crisis. The continuous reduction in statutory and debt, and net debt, all normalized by total assets. The
effective tax rates is observed, as well as helpful results illustrate that the two strongest firm factors
statistics in key indicators of economic growth are increases in the size of these firms and growth
and debt market conditions. The trends of these of the relative amount of fixed assets. Their positive
variables are not clearly related to changes in large significant association with leverage is robust across
companies’ leverage choices, neither positively nor three different definitions of leverage used as the de-
negatively. A multivariate empirical analysis fol- pendent variable. The economic magnitude is also
lows to find out the inherent associations of cer- significant. For example, a 1% increase in the fixed
tain variables, as well as to reflect on the applica- asset proportion is associated with 14 basis point in-
tion of capital structure theories in these contexts. crease in the net debt ratio. Market-to-book and capi-
tal expenditure are weakly associated with long-term
debt (column 2), but they lack explanatory power in
3. RESULTS the other two specifications.

Before the start of the baseline empirical test, it is While profitability does not seem to have explanato-
helpful to evaluate the relevant significance of as- ry power in this set of results, one can reconcile with
sociation between individual factors and the main the contemplation that these firms’ consistent posi-
dependent variable. A selection of core vector of tive performance mitigates the speculation of debt
variables is thus conducted using a process similar issuance timing game, making the leverage decision
to Frank and Goyal (2009, Table III). Specifically, relatively insensitive to whether it was a high profit or
the process begins with a regression that includes low profit year. Finally, the market-to-book and capi-
all factors (catering to different capital structure tal expenditure results appear to indicate an associa-
theories and somewhat correlated), and the study of tion of growth opportunity related debt market tim-
the R2 and the Bayesian information criterion (BIC). ing, which leads to the next set of empirical analysis.
Several worst performing variables are dropped These results provide evidence that H1 holds.
consequently. Largely reducing redundancy and
noise, this study arrives at a list of most relevant In Table 4 the model specification is expanded to in-
firm independent variables: size, profitability, fixed clude six macroeconomic variables previously select-
assets proportion, market-to-book ratio, and capital ed by a quick check of single R2 and BIC: top statuto-
expenditure. The macroeconomic variables selected ry tax rate, expected inflation, GDP growth, S&P 500
are top statutory tax rate, expected inflation, GDP average P/E ratio, credit spread, and real interest rate.
growth, S&P 500 average P/E ratio, credit spread, The three different definitions of the dependent vari-
and real interest rate. able continued to be used. With the addition of mac-
roeconomic variables, the model’s coefficient of de-
Table 3 presents the regression analysis of the base- termination improves, it provides some interesting
line model using only the firm variables. The three new results. First, the top statutory tax rate and GDP
columns present results using three different defini- growth do not explain the increase of debt, showing
tions of the dependent variable: total debt, long-term that the simple tax considerations and the economic

http://dx.doi.org/10.21511/imfi.21(1).2024.09 105
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Table 3. Firm factors associated with leverage

Total Debt / Book Assets LT Debt / Book Assets Net Debt / Book Assets
(1) (2) (3)
Total assets 0.0121*** 0.0099*** 0.0108***
(0.0037) (0.0029) (0.0031)
Operating income –0.1146 –0.1462 –0.1323*
(0.0820) (0.0989) (0.0693)
Net income –0.1451 –0.2012 –0.1971
(0.1792) (0.1996) (0.1889)
Fixed assets 0.1651*** 0.1226*** 0.1428***
(0.0448) (0.0405) (0.0466)
Mkt/Bk 0.0009 0.0011* 0.0010
(0.0006) (0.0006) (0.0007)
CapEx 0.0984 0.0931* 0.1002
(0.0622) (0.0489) (0.0711)
Adjusted R2 0.17 0.14 0.13
Obs. 3,536 3,536 3,536

Note: Variables are defined as in Appendix A. Standard errors are in parentheses. * indicates significance at the 0.10 level,
*** indicates significance at the 0.01 level.

cyclicality fail to influence the debt decisions of these Second, Table 4 shows that debt ratio is signifi-
large firms in the sample. This reinforces the obser- cantly positively related to credit spread, and
vations from the previous time-series graphs. negatively associated with real interest rate. The

Table 4. Firm factors and macroeconomic conditions: book value-based leverage

Total Debt / Book Assets LT Debt / Book Assets Net Debt / Book Assets
(1) (2) (3)
0.0109*** 0.0112*** 0.0098***
Total assets
(0.0032) (0.0033) (0.0029)
–0.1091 –0.1382 –0.1281
Operating income
(0.0795) (0.0862) (0.0737)
–0.1502 –0.1992 –0.1672
Net income
(0.1481) (0.1752) (0.1653)
0.1372*** 0.1183*** 0.1264***
Fixed assets
(0.0419) (0.0443) (0.0501)
0.0006 0.0009 0.0008
Mkt/Bk
(0.0005) (0.0006) (0.0006)
0.0638 0.0875 0.0986*
CapEx
(0.0498) (0.0562) (0.0523)
–0.1074 –0.1252 –0.1198
Top stat. tax rate
(0.0986) (0.1076) (0.1224)
0.3982 0.5561 0.7002*
Expected inflation
(0.2576) (0.3581) (0.3550)
0.2111 0.3071 0.2658
GDP growth
(0.1782) (0.2298) (0.1994)
0.0002 –0.0001 0.0002*
S&P 500 P/E
(0.0001) (0.0001) (0.0001)
0.1214*** 0.1731*** 0.1554***
Credit spread
(0.0319) (0.0423) (0.0377)
–0.0562*** –0.0621*** –0.0595***
Real interest rate
(0.0116) (0.0165) (0.0172)
Adjusted R2 0.31 0.26 0.28
Obs. 3,536 3,536 3,536

Note: Variables are defined as in Appendix A. Standard errors are in parentheses. * indicates significance at the 0.10 level,
*** indicates significance at the 0.01 level.

106 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

economic magnitude is also comparable to that in verage ratios. The literature cautions that the credit
prior literature: the net debt ratio increases by 16 spread result may not indicate a causality, as it can be
basis point per 1% of credit spread increase, and it argued that during periods of high debt supply rela-
drops by 6 basis point per 1% rise of real interest tive to demand, the debt market dynamics typically
rate. The explanatory power of these two variables lead to lower relative bond prices and thus to higher
is robust across three different definitions of lever- credit spreads. However, the real interest rate result
age used as the dependent variable. clearly indicates that firms indeed utilize opportuni-
ties when the cost of debt is low. This association is
Third, Table 4 shows that the expected inflation also confirmed in Barry et al. (2008).
is positively associated with increased debt but
does not exhibit statistical significance (except Also noteworthy in Table 4 is that, unlike the strong
for net debt ratio, column 3). In general, the real significant findings in Frank and Goyal (2009), the
value of tax deductions on debt is higher when expected inflation is positively associated with in-
inflation is expected to be high, but this depends creased debt but does not exhibit statistical signifi-
on whether the expected inflation rate is high- cance (except for net debt ratio, column 3). In gen-
er than current interest rates. Finally, there are eral, the real value of tax deductions on debt is higher
mixed and less significant results about P/E ratio, when inflation is expected to be high, but this de-
which leave the question on how the individual pends on whether the expected inflation rate is high-
firms’ debt issuance is related to the cost of equity er than current interest rates. Up to this point, only
capital still open for exploration. Taken together, one of the classic capital structure theories (free cash
these results make H2 accepted, but provide no flow theory) receives support from the findings of
support to the positive leverage-tax rate associa- this study, and it is sensible to further explore newer
tion as predicted in H3. theories. While no theory can be taken in isolation of
the consideration of general market conditions, the
organizational theory specifically addresses the time
4. DISCUSSION relevance between investment opportunities and the
need for debt. Further examination can thus help
This section discusses the main results, illustrates shed light on this relevance.
their support of selected capital structure theories,
and compares them with findings from prior re- Moreover, as debt usage is a dynamic choice (Denis
search. As a starting point, the main findings from & McKeon 2012; Huang & Ritter 2021), further dis-
Tables 3 and 4 confirm that with meaningful se- cussions are needed regarding whether firms in the
lection and control of leverage-relevant variables, a sample adjust the use of debt flexibly in response to
comprehensive set of firm-level factors representing their investment activities and risk exposure in the
growth opportunities, asset structure and informa- market. To examine the robustness of the results
tion asymmetry are jointly associated with changes above and facilitate extended discussion, more vari-
in capital structure choices. Specifically, results from ables and alternative variable definitions are includ-
the firm-variable baseline model in Table 3 are in ed, and additional tests are conducted. The newly
line with the findings of Frank and Goyal (2009). added firm-specific variables are loss carryforwards
This study argues that for the large and mostly stable / total assets, investment tax credit / total assets, and
companies in the sample, the “increasing leverage” market-adjusted individual annual stock volatility,
decision is primarily driven by the growth in total and their definitions are included in Appendix A.
assets and in the proportion of PP&E assets. That is,
the need for a larger size with a more fixed portion The first extension for further discussion repeats
makes firms raise more debt. This first set of results
the main estimation (as in Table 4) using alterna-
also partially supports the free cash flow theory. tive market value-based leverage ratios as dependent
variables. Table 5 illustrates that while these compa-
When expanding the model to include the explana- nies all have large market capitalizations, the fluctua-
tory power of macroeconomic variables, this paper tions in market value seem to have largely coincided
illustrates in Table 4 similar economic magnitude of with the changes of many independent variables,
the impact of credit spread and real interest on le- hence the debt ratios calculated against market val-

http://dx.doi.org/10.21511/imfi.21(1).2024.09 107
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

ues still exhibit similar statistical association with pendent variable, and the main model estimation
previously found explanatory factors: firm size, fixed results are reported across the seven decades using
assets proportion, credit spread, and real interest rate. dependent variables defined from both book value
H1 and H2 are supported using this alternative mar- and market value. As subsample sizes are smaller
ket value-based dependent variable. In addition, in and to ensure sufficient degree of freedom, two
two specifications (columns 2 and 3) the expected in- variables with the least explanatory power, net in-
flation becomes a significant factor. This new result come and S&P 500 P/E ratio, are dropped in this
provides inspiration for an attempt to reconcile with specification (see Table 6).
findings in Frank and Goyal (2009) and further ex-
amine how the expected inflation affects firms’ capi- In both Panel A (book asset-based leverage) and
tal structure decisions and more importantly, timing. Panel B (market value-based leverage) of Table 6, it
is evident that almost in all decades, the firm size,
To provide further contrast and discussion, the fixed assets proportion, credit spread, and real in-
next additional test studies subsamples by decades terest rate still remain significantly associated with
(note that the last two decades are actually 2000– leverage. Moreover, the explanatory power of the
2007 and 2008–2020 with the 2008 financial crisis expected inflation appears to be the strongest in
being a more meaningful divider). This treatment the 70’s, 90’s, and the first few years in the new
provides more insights on the macroeconomic fac- millennium before the 2008 crisis. This observa-
tors’ effects such as the expected inflation, on a tion is even more pronounced when market val-
more granular time scale. In Table 6, for concise- ue-based leverage ratios are used. Considering
ness only the net debt ratio (based on book assets major economic events and certain sector break-
and on market cap separately) is used as the de- throughs in these subperiods, this study conjec-
Table 5. Firm factors and macroeconomic conditions: market value-based leverage
Total Debt / Mkt Cap LT Debt / Mkt Cap Net Debt / Mkt Cap
(1) (2) (3)
0.0092*** 0.0103*** 0.0063***
Total assets
(0.0027) (0.0029) (0.0018)
–0.0726 –0.1028 –0.0982
Operating income
(0.0425) (0.0665) (0.0577)
–0.0738 –0.1384 –0.1173
Net income
(0.1003) (0.1173) (0.1228)
0.0816*** 0.0996*** 0.1001***
Fixed assets
(0.0246) (0.0302) (0.0412)
0.0004 0.0007 0.0006
Mkt/Bk
(0.0005) (0.0005) (0.0005)
0.0281 0.0693 0.0772*
CapEx
(0.0178) (0.0444) (0.0391)
–0.0915 –0.1093 –0.0892
Top stat. tax rate
(0.0718) (0.0829) (0.1018)
0.5717 0.6364** 0.7776*
Expected inflation
(0.3928) (0.3201) (0.3924)
0.1836 0.2847 0.2128
GDP growth
(0.1472) (0.1987) (0.1689)
0.0002 0.0001 0.0001
S&P 500 P/E
(0.0002) (0.0001) (0.0001)
0.0927*** 0.1193*** 0.1073***
Credit spread
(0.0228) (0.0362) (0.0311)
–0.0363*** –0.0483*** –0.0422***
Real int. rate
(0.0094) (0.0112) (0.0163)
Adjusted R2 0.29 0.22 0.24
Obs. 3,536 3,536 3,536

Note: Variables are defined as in Appendix A. Standard errors are in parentheses. * indicates significance at the 0.10 level,
** indicates significance at the 0.05 level, *** indicates significance at the 0.01 level.

108 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Table 6. Determinants of leverage across subperiods

1950–1959 1960–1969 1970–1979 1980–1989 1990–1999 2000–2007 2008–2020
Panel A. Dependent Variable: Net Debt / Book Assets
0.0121*** 0.0109*** 0.0094*** 0.0115*** 0.0102*** 0.0086*** 0.0091***
Total assets
(0.0037) (0.0031) (0.0021) (0.0033) (0.0028) (0.0021) (0.0030)
–0.1013 –0.1302 –0.1298 –0.1117 –0.1205 –0.1288 –0.1321
Op. income
(0.0699) (0.0826) (0.0674) (0.0635) (0.0711) (0.0756) (0.0784)
0.1314*** 0.1206*** 0.1277*** 0.1292*** 0.1189*** 0.1214*** 0.1225***
Fixed assets
(0.0528) (0.0479) (0.0531) (0.0498) (0.0426) (0.0492) (0.0487)
0.0010 0.0009 0.0008 0.0009 0.0010 0.0009 0.0008
Mkt/Bk
(0.0007) (0.0006) (0.0007) (0.0007) (0.0006) (0.0007) (0.0005)
0.0728 0.0876* 0.1016 0.0943 0.0922 0.1072* 0.0939
CapEx
(0.0491) (0.0443) (0.0572) (0.0510) (0.0465) (0.0545) (0.0478)
–0.1073 –0.1202 –0.1146 –0.1332 –0.1082 –0.1178 –0.1192
Stat. tax rate
(0.0925) (0.1183) (0.0897) (0.1218) (0.1094) (0.1003) (0.1174)
0.4973 0.3893 0.6014** 0.5105 0.6116** 0.5887* 0.6006
Exp. inflation
(0.4128) (0.3099) (0.2964) (0.4221) (0.3059) (0.2952) (0.4194)
0.2138 0.2384 0.2927 0.2615 0.3009 0.2582 0.2712
GDP growth
(0.1726) (0.1827) (0.1924) (0.1848) (0.1998) (0.1845) (0.1718)
0.1602*** 0.1582*** 0.1533*** 0.1493*** 0.1518*** 0.1722*** 0.1610***
Credit spread
(0.0410) (0.0392) (0.0327) (0.0321) (0.0310) (0.0492) (0.0413)
–0.0618*** –0.0561*** –0.0603*** –0.0583*** –0.0611*** –0.0571*** –0.0589***
Real int. rate
(0.0194) (0.0164) (0.0188) (0.0159) (0.0202) (0.0163) (0.0178)
Adjusted R2 0.28 0.26 0.31 0.29 0.32 0.29 0.23
Obs. 531 529 529 529 529 423 466
Panel B. Dependent Variable: Net Debt / Market Cap
0.0091*** 0.0084*** 0.0087*** 0.0055*** 0.0079*** 0.0051*** 0.0061***
Total assets
(0.0029) (0.0023) (0.0028) (0.0019) (0.0022) (0.0016) (0.0019)
–0.0926 –0.1182 –0.1073 –0.0817 –0.1031 –0.0921 –0.0827
Op. income
(0.0573) (0.0678) (0.0602) (0.0479) (0.0691) (0.0612) (0.0523)
0.1028*** 0.0971*** 0.1024*** 0.0877*** 0.0916*** 0.1071*** 0.0962***
Fixed assets
(0.0329) (0.0311) (0.0409) (0.0281) (0.0281) (0.0446) (0.0395)
0.0004 0.0006 0.0007 0.0007 0.0005 0.0007 0.0006
Mkt/Bk
(0.0004) (0.0005) (0.0006) (0.0008) (0.0005) (0.0005) (0.0006)
0.0711 0.0799* 0.0802 0.0479 0.0616 0.0591 0.0781
CapEx
(0.0482) (0.0403) (0.0415) (0.0329) (0.0396) (0.0385) (0.0407)
–0.1102 –0.1095 –0.0821 –0.0899 –0.0763 –0.0813 –0.0884
Stat. tax rate
(0.0828) (0.0700) (0.0912) (0.1015) (0.0718) (0.0927) (0.0812)
0.5813 0.5103 0.6998*** 0.4728 0.5882*** 0.7761* 0.5695
Exp. inflation
(0.4006) (0.4241) (0.2901) (0.3941) (0.2192) (0.3912) (0.3977)
0.1930 0.2185 0.1728 0.1996 0.2263 0.2008 0.2201
GDP growth
(0.1264) (0.1736) (0.1517) (0.1619) (0.1637) (0.1619) (0.1947)
0.0981*** 0.1106*** 0.1044*** 0.0992*** 0.1117*** 0.1022*** 0.1075***
Credit spread
(0.0278) (0.0318) (0.0291) (0.0303) (0.0325) (0.0281) (0.0309)
–0.0419*** –0.0398*** –0.0467*** –0.0439*** –0.0521*** –0.0402*** –0.0414***
Real int. rate
(0.0114) (0.0102) (0.0137) (0.0108) (0.0126) (0.0151) (0.0138)
Adjusted R2 0.27 0.21 0.33 0.3 0.27 0.26 0.23
Obs. 531 529 529 529 529 423 466

Note: Variables are defined as in Appendix A. Standard errors are in parentheses. * indicates significance at the 0.10 level,
** indicates significance at the 0.05 level, *** indicates significance at the 0.01 level.

tures that the general favorable economic growth ings as in Frank and Goyal (2009), DeAngelo
and market conditions contributed to the opti- and Roll (2015), and Poursoleyman et al. (2023).
mism and heavier use of debt. This provides par-
tial evidence to the time relevance argument of To further discuss the role of operational loss (as in
the organizational theory and echoes prior find- Horváthová et al. (2022)) and perceived risk (as in

http://dx.doi.org/10.21511/imfi.21(1).2024.09 109
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Table 7. Further exploration of firm-specific factors

Net Debt / Book Assets Net Debt / Mkt Cap
0.0122*** 0.0092***
Total assets
(0.0037) (0.0027)
–0.0927 –0.1009
Operating income
(0.0561) (0.0726)
–0.0078 –0.0051
Loss carryforward
(0.0119) (0.0044)
0.0046*** 0.0067**
Inv. tax credit
(0.0019) (0.0034)
0.0992*** 0.1173***
Fixed assets
(0.0276) (0.0382)
0.0005 0.0007
Mkt/Bk
(0.0006) (0.0005)
0.0186** 0.0222***
Stock volatility
(0.0095) (0.0087)
0.0461 0.0681
CapEx
(0.0328) (0.0372)
–0.1017 –0.0927
Top stat. tax rate
(0.0667) (0.0628)
0.5163 0.6772*
Expected inflation
(0.3725) (0.3417)
0.2209 0.3137
GDP growth
(0.1726) (0.2009)
0.1063*** 0.1127***
Credit spread
(0.0311) (0.0372)
–0.0727*** –0.0631***
Real int. rate
(0.0272) (0.0213)
Adjusted R2 0.36 0.29
Obs. 3,128 3,128

Note: Variables are defined as in Appendix A. Standard errors are in parentheses. * indicates significance at the 0.10 level, **
indicates significance at the 0.05 level, *** indicates significance at the 0.01 level.

Welch (2004) and Wang et al. (2020)) in the deter- Table 7 documents a slightly improved R 2 and con-
mination of capital structure, the final extended test firms the robust explanatory power of firm size,
turns direction to address more firm specific vari- fixed assets proportion, credit spread, and real
ables that have been recently found to be relevant interest rate. Among the three new variables, the
to leverage. Since the benefit of debt is directly re- investment tax credit and the stock volatility are
lated to tax deductibility/incentives, and prior liter- found to be positively related to debt ratios. Thes
ature finds the association between debt structure/ new results confirm the impact of corporate in-
volume and investment activities, this robustness vestment activities and cost of equity on the choice
check includes individual firm’s loss carryforwards of leverage: when the tax policy encourages invest-
and investment tax credit as two new independent ment activities and when equity is relatively more
variables. Moreover, as Wang et al. (2020) find the expensive due to perceived risk, firms increasingly
cost of equity being the key factor driving large turn to debt for external financing. The prediction
companies’ choices of equity financing, a market- about the positive association of leverage and cost
adjusted individual annual stock volatility is added of equity in H3 and existing findings in the litera-
as a firm-specific risk measure as well as a gauge of ture (Welch, 2004; Barry et al., 2008; Wang et al.,
the cost of equity. The main estimation is augment- 2020) are thus supported by this last set of results.
ed to include these three additional explanatory
variables as well as using debt ratios based on both As for future extension of this study, three more
book assets and market cap separately as the depen- sets of estimations can further enhance the ver-
dent variables. The sample size decreases slightly to satility of its empirical settings. First, though the
3,128 observations due to data availability. firms in the sample are all large, stable, and repu-

110 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

table with minimum default risk, it is worthwhile to recognize variations in individual debt terms and
characteristics. Including individual debt rating would be beneficial to address this point. In addition,
as uncertainty due to cash flow volatility is found to be directly related to companies’ dynamic adjust-
ments of leverage in earlier stages of firm life (Ren et al., 2023), including cash flow volatility and related
uncertainty factors may further shed light on U.S. large firms’ choice of capital structure. Finally, an
alternative model specification is to put these big firms into sector/industry context – this treatment can
further shed light on the trend of leverage increase. Adding the industry leverage as an independent
variable can potentially provide more insights but a caveat would be sample size constraints for sparsely
populated industries.

CONCLUSION
The purpose of this study is to apply classic capital structure theories and examine the main factors as-
sociated with an almost monotonic upward trend of leverage among the U.S. large firms. Focusing on
the 70-year debt financing of companies that were in the starting lineup of the Fortune 500 list and are
still going strong as of 2020, this study’s main results indicate that large “elite” firms not only actively
utilize debt market opportunities and are sensitive to the general interest environment, but also dy-
namically adjust debt level according to their growth opportunities, information asymmetry, and goal
of maximizing organizational wealth. These findings validate the free cash flow theory and provide sup-
port for the organizational theory of capital structure choices.

Further examinations discover that certain tax incentives and the perceived equity risk also contribute
to increased use of debt. In addition, more discussions are made in favor of critical macroeconomic fac-
tors, especially those that impact the cost and benefit of borrowing, being directly related to leverage
choices. These factors’ explanatory power provides favorable evidence for the organizational theory.

In conclusion, this study extends the ongoing scholarly discussion regarding corporate debt decisions
and makes a significant contribution to the corporate finance literature on the topics of the association
between capital structure and growth needs, in the context of macroeconomic environment shifts. After
comprehensively considering all capital structure theories, selecting representative factors, and con-
ducting rigorous empirical tests, this study presents findings that support the free cash flow theory and
organizational theory, clarify the corporate debt policy in general, and highlight large firms’ responses
in different time periods when the market conditions were quite different.

AUTHOR CONTRIBUTIONS
Conceptualization: Wenjuan Xie.
Data curation: Wenjuan Xie.
Formal analysis: Wenjuan Xie.
Investigation: Wenjuan Xie.
Methodology: Wenjuan Xie.
Project administration: Wenjuan Xie.
Resources: Wenjuan Xie.
Software: Wenjuan Xie.
Supervision: Wenjuan Xie.
Validation: Wenjuan Xie.
Visualization: Wenjuan Xie.
Writing – original draft: Wenjuan Xie.
Writing – review & editing: Wenjuan Xie.

http://dx.doi.org/10.21511/imfi.21(1).2024.09 111
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

ACKNOWLEDGEMENTS
The author declares no financial or personal relationships with other people or organizations that could
inappropriately influence my work. The author thanks the journal editor and anonymous referees, Ahmad
Etebari, Fred Kaen, discussants and participants at the Northeast Business and Economics Association
2022 annual meeting and the 2023 Global Finance Conference for their constructive critiques. All errors
remain my responsibility.

REFERENCES
1. Barry, C. B., Mann, S. C., Mihov, Financial Studies, 26, 567-601. 15. Hennessy, C., & Whited, T. (2005).
V. T., & Rodriguez, M. (2008). Retrieved from https://econpapers. Debt Dynamics. Journal of
Corporate Debt Issuance and the repec.org/article/ouprfinst/v_3a- Finance, 60, 1129-1165. Retrieved
Historical Level of Interest Rates. 26_3ay_3a2013_3ai_3a3_3ap_ from https://econpapers.repec.org/
Financial Management, 37(3), 3a567-601.htm article/blajfinan/v_3a60_3ay_3a
413-430. https://doi.org/10.1111/ 2005_3ai_3a3_3ap_3a1129-1165.
9. Frank, Murray Z., & Vidhan K.
j.1755-053X.2008.00019.x htm
Goyal, (2009). Capital Struc-
2. Begenau, J. (2020). Capital ture Decision: Which Factors 16. Holmström, B., & Tirole, J. (1998).
requirements, risk choice, and Are Reliably Important? Fi- Private and Public Supply of
liquidity provision in a business- nancial Management, 38, 1-37. Liquidity. Journal of Political
cycle model. Journal of Finan- https://doi.org/10.1111/j.1755- Economy, 106, 1-40. https://doi.
cial Economics, 136, 355-378. 053X.2009.01026.x org/10.1086/250001
https://doi.org/10.1016/j.jfine-
10. Gorton, G., & Pennacchi, G. 17. Horváthová, J., Mokrišová, M., &
co.2019.10.004
(1990). Financial Intermediaries Petruška, I. (2022). Indebtedness
3. Campello, M., Graham, J., & and Liquidity Creation. Journal and profitability – A threshold
Harvey, C. (2010). The Real of Finance, 45, 49-71. https://doi. model approach. Investment Man-
Effects of Financial Constraints. org/10.2307/2328809 agement and Financial Innova-
Journal of Financial Economics, 97, tions, 19(3), 13-27. http://dx.doi.
11. Graham, J., & Harvey, C. (2001).
470-487. https://doi.org/10.1016/j. org/10.21511/imfi.19(3).2022.02
The Theory and Practice of
jfineco.2010.02.009
Corporate Finance: Evidence 18. Huang, R., & Ritter, J. (2021). Cor-
4. DeAngelo, H., & Roll, R. (2015). from the Field. Journal of Fi- porate Cash Shortfalls and Financ-
How Stable Are Corporate Capital nancial Economics, 60, 187-243. ing Decisions. Review of Financial
Structures? Journal of Finance, 70, https://doi.org/10.1016/S0304- Studies, 34, 1789-1833. https://doi.
373-418. https://doi.orgS./10.1111/ 405X(01)00044-7 org/10.1093/rfs/hhaa099
jofi.12163
12. Graham, J., & Leary, M. (2011). 19. Jensen, M. (1986). Agency Costs
5. Denis, D., & McKeon. (2012). A Review of Empirical Capital of Free Cash Flow, Corporate
Debt Financing and Financial Structure Research and Directions Finance, and Takeovers. American
Flexibility: Evidence from Proac- for the Future. Annual Review of Economic Review, 76, 323-329.
tive Leverage Increases. Review of Financial Economics, 3, 309-345. Retrieved from https://papers.ssrn.
Financial Studies, 25, 1897-1929. Retrieved from https://papers.ssrn. com/sol3/papers.cfm?abstract_
https://doi.org/10.1093/rfs/hhs005 com/sol3/papers.cfm?abstract_ id=99580
id=1729388
6. Donaldson, G. (1984). Manag- 20. Jensen, M., & Meckling, W. (1976).
ing Corporate Wealth. New York: 13. Graham, J., Leary, M., & Roberts, Theory of the firm: managerial be-
Praeger. M. (2015). A Century of Capital havior, agency costs, and owner-
Structure: The Leveraging of ship structure. Journal of Financial
7. Fama, Eugene F., & Kenneth R.
Corporate America. Journal of Economics, 4, 305-360. https://doi.
French. (2002). Testing Trade-
Financial Economics, 118, 658-683. org/10.1016/0304-405X(76)90026-
Off and Pecking Order Predic-
https://doi.org/10.1016/j.jfine- X
tions About Dividends and Debt.
co.2014.08.005
Review of Financial Studies, 15(1), 21. Lins, K., Servaes, H., & Tufano, P.
1-33. https://doi.org/10.1093/ 14. He, W., Hu, M., Mi, L., & Yu, J. (2010). What Drives Corporate
rfs/15.1.1 (2021). How Stable Are Corporate Liquidity? An International Sur-
Capital Structures? International vey of Cash Holdings and Lines of
8. Fee, C., C. Hadlock, & Pierce,
Evidence. Journal of Bank- Credit. Journal of Financial Eco-
J. (2013). Managers With and
ing and Finance, 126, 106103. nomics, 98, 160-176. https://doi.
Without Style: Evidence Using
https://doi.org/10.1016/j.jbank- org/10.1016/j.jfineco.2010.04.006
Exogenous Variation. Review of
fin.2021.106103

112 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

22. Modigliani, F., & Miller, M. H. 26. Poursoleyman, E., Mansourfar, 30. Titman, S. (2002). The Modigliani–
(1958). The Cost of Capital, Cor- G., & Abidin, S. (2023). Debt Miller Theorem and the Integra-
poration Finance, and the Theory structure: a solution to the puzzle tion of Financial Markets. Finan-
of Investment. American Economic of capital structure, International cial Management, 31, 101-115.
Review, 48, 261-297. Retrieved Journal of Managerial Finance, https://doi.org/10.2307/3666323
from https://fdjpkc.fudan.edu. 19(1), 22-47. Retrieved from
31. Treynor, J. L. (1981). The Financial
cn/_upload/article/files/91/a3/ https://ideas.repec.org/a/eme/ijm-
Objective in the Widely Held
fb1aedf9496dbda48b8f6c1096de/ fpp/ijmf-03-2021-0155.html
Corporation. Financial Analysts
a36f7904-24d0-44be-8245-
27. Rajan, R. G., & Zingales, L. (1995). Journal, 37(2), 68-71. https://doi.
02dff89842d7.pdf
What Do We Know About Capital org/10.2469/faj.v37.n2.68
23. Modigliani, F., & Miller M. H. Structure? Some Evidence from
32. Wang, Z., Ettinger, M., Xie, Y., &
(1963). Corporate Income Taxes International Data. Journal of
Xu, L. (2020). The cost of capital:
and the Cost of Capital: A Cor- Finance, 50, 1421-1460. https://
U.S.-based multinational corpora-
rection. American Economic doi.org/10.1111/j.1540-6261.1995.
tions versus U.S. domestic corpo-
Review, 53, 433-443. Retrieved tb05184.x
rations. Global Finance Journal, 44,
from https://www.jstor.org/
28. Ren, D., Chen, Y., Maynard, A., & 100443. https://doi.org/10.1016/j.
stable/1809167
Pysarenko, S., (2023). Valuating gfj.2018.07.002
24. Myers, S. C. (1993). Still Search- the capital structure under in-
33. Warusawitharana, M., & Whited,
ing for Optimal Capital Structure. complete information. Investment
T. (2016). Equity Market Misvalu-
Journal of Applied Corporate Management and Financial Inno-
ation, Financing, and Investment.
Finance, 6(1), 4-14. https://doi. vations, 20(3), 48-67. http://dx.doi.
Review of Financial Studies, 29,
org/10.1111/j.1745-6622.1993. org/10.21511/imfi.20(3).2023.05
603-654. https://doi.org/10.1093/
tb00369.x
29. Smith, Clifford W. Jr. (1990). The rfs/hhv066
25. Myers, S. C., & Majluf, N. S. theory of corporate finance: A
34. Welch, I. (2004). Capital Structure
(1984). Corporate Financing historical overview. In Clifford
and Stock Returns. Journal of
and Investment Decisions When W. Smith, Jr. (Ed.), The Modern
Political Economy, 112, 106-131.
Firms Have Information That Theory of Corporate Finance (pp.
https://doi.org/10.1086/379933
Investors Do Not Have. Journal of 3-24). New York: McGraw-Hill.
Financial Economics, 13, 187-221. Retrieved from https://papers.ssrn.
https://doi.org/10.1016/0304- com/sol3/papers.cfm?abstract_
405X(84)90023-0 id=244161

http://dx.doi.org/10.21511/imfi.21(1).2024.09 113
Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

APPENDIX A
Definition of Variables
Total Assets: Book value from Balance Sheet. In thousands of US dollars in Table 1. Treated by natural
logarithm for all regressions.

Total market value of assets: Market value of equity (fiscal year-end closing price times share outstand-
ing) plus debt in current liabilities plus long-term debt minus deferred taxes. In thousands of US dollars
and used as denominator to standardize debt for independent variables.

Total Sales: Net revenue from Income Statement. In thousands of US dollars.

Total debt: Book value debt in current liabilities plus long-term debt from Balance Sheet. In thousands
of US dollars and standardized by total assets to be used as an independent variable in regressions.

Long-term debt: Book value from Balance Sheet. In thousands of US dollars. Standardized by total as-
sets or total market value of assets as independent variable in regressions.

Cash: Cash and cash equivalents from Balance Sheet. In thousands of US dollars.

Net debt: Long-term debt minus cash and cash equivalents from Balance Sheet. In thousands of US dol-
lars. Standardized by total assets or total market value of assets as independent variable in regressions.

EBIT: Earnings before interest and tax from Income Statement. In thousands of US dollars.

EBITDA: earnings before interest, tax, depreciation and amortization from Income Statement. In thou-
sands of US dollars.

Top statutory tax rate: Rate of the top bracket of corporate tax. In plain numbers.

Effective tax rate: Corporate tax paid divided by taxable income for the same year, from Income
Statement. In plain numbers.

Unemployment rate: Annual rate as published by the Bureau of Labor Statistics, in percentages.

Recession years: Defined by National Bureau of Economic Research, binary variable that takes the value
1 if the year is marked as in recession, and 0 otherwise.

Baa corporate bond annual yield: Moody’s average annual yield of Baa-rated 10-year corporate bonds,
in percentages.

Long-term government bond annual rate: Average 10-year US government bond annual yield, published
by Federal Reserve Bank of St. Louis, in percentages.

GDP price inflation rate: Annual rate of inflation as measured by changes in the prices of goods and
services produced in the United States, including those exported to other countries. Published by the
Bureau of Economic Analysis, in percentages.

Real interest rate: Nominal interest rate adjusted for inflation, from Rober Shiller’s website and Federal
Reserve Bank of St. Louis, in percentages.

114 http://dx.doi.org/10.21511/imfi.21(1).2024.09
 Investment Management and Financial Innovations, Volume 21, Issue 1, 2024

Operating Income: The operating income from Income Statement, standardized by total assets.

Net Income: The net income from Income Statement, standardized by total assets.

Fixed Assets: Net Property, Plant and Equipment from Balance Sheet, standardized by total assets.

Mkt/Bk: Market-to-book ratio, defined as total market value of equity (fiscal year-end closing price
times share outstanding) divided by total shareholder’s equity from Balance Sheet.

CapEx: Capital expenditure from Cash Flow Statement, standardized by total assets.

Expected inflation: Short-term (1 year) expected inflation rate based on GDP price, published by Federal
Reserve Bank of New York, in percentages.

GDP growth: Real GDP (in 2005 dollars) annual growth rate, published by Bureau of Economic Analysis,
in percentages.

S&P 500 P/E: 10-year average cyclically adjusted average price-to-earnings ratio of all S&P 500 firms for
a given year, provided by Robert Shiller’s website.

Credit spread: Moody’s average annual yield of Baa-rated 10-year corporate bonds minus 10-year trea-
sury bond yield of the same year, in percentages.

Loss carryforward: Net operating loss carryforward from Income Statement (as a result of previous
year’s deferred tax asset), standardized by total assets.

Inv. tax credit: investment tax credit from Balance Sheet, standardized by total assets.

Stock volatility: Annual standard deviation of individual stock’s daily excess return (individual stock
return minus that of the S&P 500 index), in plain numbers.

http://dx.doi.org/10.21511/imfi.21(1).2024.09 115