151graham
8
STERN STEWART JOURNAL OF APPLIED CORPORATE FINANCE
HOW DO CFOS MAKE
CAPITAL BUDGETING
AND CAPITAL STRUCTURE
DECISIONS?
y John Graham and
Campbell Harvey,
Duke University*
e recently conducted a comprehensive survey that analyzed the
cu
ent practice of corporate finance, with particular focus on the
areas of capital budgeting and capital structure. The survey results
enabled us to identify aspects of corporate practice that are
*This paper is a compressed version of our paper that was first published as “The Theory and Practice of Corporate Finance:
Evidence from the Field” in the Journal of Financial Economics, Vol XXXXXXXXXX), and which won the Jensen prize for the best
JFE paper in corporate finance in 2001. This research is partially sponsored by the Financial Executives International (FEI)
ut the opinions expressed herein do not necessarily represent the views of FEI. We thank the FEI executives who responded
to the survey. Graham acknowledges financial support from the Alfred P. Sloan Research Foundation.
1. In the original JFE version of this paper, we show that our sample of respondents is representative of the overall
population of 4,400 firms, is fairly representative of Compustat firms, and is not adversely affected by nonresponse bias. The
next largest survey that we know of studies 298 large firms and is presented in J. Moore and A. Reichert, “An Analysis of the
Financial Management Techniques Cu
ently Employed by Large U.S. Corporations,” Journal of Business Finance and
Accounting, Vol XXXXXXXXXX), pp XXXXXXXXXX.
consistent with finance theory, as well as aspects that are hard to reconcile with
what we teach in our business schools today. In presenting these results, we
hope that some practitioners will find it worthwhile to observe how othe
companies operate and perhaps modify their own practices. It may also be useful
for finance academics to consider differences between theory and practice as
a reason to revisit the theory.
We solicited responses from approximately 4,440 companies and received
392 completed surveys, representing a wide variety of firms and industries.1 The
survey contained nearly 100 questions and explored both capital budgeting and
capital structure decisions in depth. The responses to these questions enabled
us to explore whether and how these corporate policies are inte
elated. Fo
example, we investigated whether companies that made more aggressive use
of debt financing also tended to use more sophisticated capital budgeting
techniques, perhaps because of their greater need for discipline and precision
in the corporate investment process.
More generally, the design of our survey allowed for a richer under-
standing of corporate decision-making by analyzing the CFOs’ responses in
the context of various company characteristics, such as size, P/E ratio,
leverage, credit rating, dividend policy, and industry. We also looked fo
systematic relationships between corporate financial choices and manage-
ial factors, such as the extent of top management’s stock ownership, and
the age, tenure, and education of the CEO. By testing whether the responses
W
9
VOLUME 15 NUMBER XXXXXXXXXXSPRING 2002
varied systematically with these characteristics,
we were able to shed light on the implications of
various corporate finance theories that focus on
variables such as a company’s size, risk, invest-
ment opportunities, and managerial incentives.
The results of our survey were reassuring in
some respects and surprising in others. With respect
to capital budgeting, most companies follow aca-
demic theory and use discounted cash flow (DCF)
and net present value (NPV) techniques to evaluate
new projects. But when it comes to making capital
structure decisions, corporations appear to pay less
attention to finance theory and rely instead on
practical, informal rules of thumb. According to ou
survey, the main objective of CFOs in setting debt
policy was not to minimize the firm’s weighted
average cost of capital, but rather to preserve “finan-
cial flexibility”—a goal that tended to be associated
with maintaining a targeted credit rating. And con-
sistent with the emphasis on flexibility, most CFOs
also expressed considerable reluctance to issue
common equity unless their stock prices were at
“high” levels, mainly because of their concern about
dilution of EPS. (As we shall argue later, although
such reluctance to issue equity is likely to be
consistent with finance theory’s emphasis on the
costs associated with “information asymmetry,” the
extent of CFOs’ preoccupation with EPS effects
seems to contradict the theory.)
The survey also provided clear evidence that
firm size significantly affects the practice of corporate
finance. For example, large companies were much
more likely to use net present value techniques,
while small firms tended to rely on the payback
criterion. And, providing some encouragement to
proponents of academics’ trade-off model of capital
structure (discussed in more detail later), a majority
of large companies said they had “strict” or “some-
what strict” target debt ratios, whereas only a third
of small firms claimed to have such targets.
In the next section, we
iefly discuss the design
of the survey and our sampling techniques (with
more details provided in the Appendix). Then we
eview our findings, first on capital budgeting policy
and next on capital structure decisions.
SURVEY TECHNIQUES AND SAMPLE
CHARACTERISTICS
Perhaps the most important part of survey
esearch is designing a survey instrument that asks
clear and pertinent questions. We took several steps
to achieve this end. After spending months develop-
ing a draft survey, we circulated the draft to a group
of academics and practitioners and incorporated
their suggestions into a revised version. Then, afte
getting the advice of marketing research experts on
oth the survey’s design and execution, we made
changes to the format of the questions and to the
overall design in order to minimize biases induced
y the questionnaire and maximize the response
ate. The final survey was three pages long and took
approximately 15 minutes to complete.
We mailed the survey to the CFOs of all (1998)
Fortune 500 companies and also faxed surveys to
4,440 firms with officers who are members of the
Financial Executives Institute (313 of the Fortune 500
CFOs are also FEI members).2 The 392 returned
surveys represented a response rate of nearly 9%.
Given the length and scope of our survey, this
esponse rate compared favorably to the response
ate for other recent academic surveys.3 We received
esponses from CFOs representing a wide variety of
companies, ranging from very small (26% of the
sample firms had sales of less than $100 million) to
very large (42% had sales of at least $1 billion). Forty
percent of the firms were manufacturers, and the
emaining firms were evenly spread across othe
industries, including financial (15%), transportation
and energy (13%), retail and wholesale sales (11%), and
high-tech (9%). Sixty percent of the respondents had
price-earnings ratios of 15 or greater (a group we refe
to later as “growth firms” when we analyze the effect
of investment opportunities on corporate behavior).
The distribution of debt levels was fairly uni-
form. Approximately one-third of the sample com-
panies had debt-to-asset ratios (expressed in book
values) below 20%, another third had debt ratios
etween 20% and 40%, and the remaining firms had
debt ratios greater than 40. We refer to companies
with debt ratios greater than 30% as “highly levered.”
2. FEI has approximately 14,000 members that hold policy-making positions
as CFOs, treasurers, and controllers at 8,000 companies throughout the U.S. and
Canada. Every quarter, Duke University and FEI poll these financial officers with
a one-page survey on important topical issues. See http:
www.duke.edu
~jgraham under “FEI Survey.” The usual response rate for the quarterly survey is
8-10%.
3. See, for example, E. Trahan and L. Gitman, “Bridging the Theory-Practice
Gap in Corporate Finance: A Survey of Chief Financial Officers,” Quarterly Review
of Economics and Finance, Vol XXXXXXXXXX), pp. 73-87; the authors obtained a 12%
esponse rate in a survey mailed to 700 CFOs. The response rate also compared
favorably to the response rate for the quarterly FEI-Duke survey, which usually runs
around 8-10%.
10
JOURNAL OF APPLIED CORPORATE FINANCE
The creditworthiness of the sample also showed
oad variation. Twenty percent of the companies
had credit ratings of AA or AAA, 32% had an A rating,
and 27% were rated BBB. The remaining 21% had
speculative debt with ratings of BB or lower.
Though our survey respondents were CFOs, we
asked a number of questions about the characteristics
of the chief executive officers. We assumed that CEOs
are the ultimate decision-makers and that CFOs act as
agents for the CEOs. Nearly half of the CEOs for the
esponding firms were between 50 and 59 years old.
Another 23% were over age 59, and 28% were between
the ages of 40 and 49. The survey revealed that
executives change jobs frequently. Nearly 40% of the
CEOs had been in their jobs less than four years, and
another 26% had been in their jobs between four and
nine years. We defined the 34% who had been in thei
jobs more than nine years as having “long tenure.”
Forty-one percent of the CEOs had an undergraduate
degree as their highest level of education. Another 38%
had MBAs and 8% had non-MBA masters degrees; 13%
had gone beyond the masters level. Finally, the top
three executives owned at least 5% of the common
stock in 44% of the companies.
These CEO and firm characteristics allowed us
to examine whether managerial incentives or en-
trenchment affected the survey responses. We also
studied whether having an MBA affected the choices
made by corporate executives. All in all, the variation
in executive and company characteristics permitted
a rich description of the practice of corporate
finance, and allowed us to make a number of
inferences about the extent to which corporate
actions are consistent with academic theories. Ou
survey differed from previous work in several ways.
The most obvious difference is that previous work
has almost exclusively focused on the largest firms.
Second, because our sample is larger than previous
surveys, we were able to control for many different
firm characteristics. As with all survey research,
however, it’s important to keep in mind that survey
esults represent CFO beliefs or opinions. We have
no way of verifying that such beliefs account for (o
are even consistent with) their actions. What’s more,
in some cases, corporate executives might be influ-
enced by a theory without knowing it. In this sense,
as Keynes once wrote, “practical men...are usually
the slaves of some defunct economist.”
CAPITAL BUDGETING DECISIONS
It is a major tenet of modern finance theory that
the value of an asset (or an entire company) equals
the discounted present value of its expected future
cash flows. Hence, companies contemplating invest-
ments in capital projects should use the net present
value rule : that is, take the project if the NPV is
positive (or zero); reject if NPV is negative.
But if NPV has been the dominant method
taught in business schools, past surveys have sug-