INTERNATIONAL FINANCIAL MANAGEMENT

INTERNATIONAL FINANCIAL MANAGEMENT

HARVARD BUSINESS SCHOOL
EC COURSE PAPER

DECEMBER 2005


“INTERNATIONAL CAPITAL BUDGETING IN PRACTICE”


MARC LIEN
70605147

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

INTRODUCTION

I wrote this International Financial Management Paper with four objectives in mind:

• To develop a survey instrument aimed at providing a rich description of the
practices of international corporate finance, allowing us to infer whether
corporate actions are consistent with academic theories.
• To construct a practical step-by-step approach to implementing such a survey
with the objectives of maximizing response rate and minimizing biases. This
approach would represent a plan of action to go from idea conception through
to publication.
• To find evidence from prior surveys supporting the theory that multi-national
corporations consistently over state their cost-of-capital.
• To improve my understanding of international corporate finance concepts and
to provide an opportunity to research and explore the challenges and tradeoffs
that financial executives face.

I have attempted to provide a real practical aid to the future undertaking of a survey
on international capital budgeting practices. To best achieve this aim, I structured
this report in the form of a chronological project plan. The major steps in the eight
point plan are:
1. Understand previous academic research
2. Define research objectives
3. Develop draft research instruments
4. Refine survey instrument
5. Select target companies
6. Construct delivery mechanism
7. Design marketing approach
8. Undertake research

Marc Lien

International Financial Management
Page 2
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

1. UNDERSTAND PREVIOUS ACADEMIC RESEARCH

Prior to engaging in a new piece of academic research, I felt it important to review
prior literature. Firstly, through a review of contemporary publications on a broad
range of International Financial Management topics, I felt better capable of creating
an informed survey instrument. Secondly, through a detailed critique of 20 previous
surveys on Capital Budgeting since 1970 (see Appendix A), I was able to both identify
best practice in survey construction and execution and, finally, this comprehensive
review of prior survey literature enabled me to identify where any gaps exist in terms
of academic understanding that could guide this piece of work.

My survey review highlights three main routes that prior research has followed:

1. Determine whether a gap exists between capital budgeting practices in the
field and those espoused by academics

A broad range of surveys going back to Mao (1970) [1]1 have looked to
understand and explain differences between the behavior of finance
practitioners and the normative view held by the academic world. This body
of research has captured three aspects of finance practice: how firms use
capital budgeting techniques, how they incorporate and manage risk and how
they determine their cost of capital.

Klammer [2], Gitman [4], Schall [5] all identify that managers in firms have a
preference to use simple payback techniques over more sophisticated DCF
ones although this trend is changing over time (Block [7]).

Later studies suggest the level of financial sophistication may be increasing in
some areas but not in others. Blazouske [10], for example, notes that although
firms use advanced capital budgeting techniques, few use any method of risk
adjustment or management science techniques.

Ryan (2002) [19] reviews past U.S surveys to see whether capital budgeting
practices have changed in Fortune 1,000 firms over time. She concludes: “it
appears the views of academics and senior managers of Fortune 1,000
companies on basic capital budgeting techniques are in stronger agreement
than ever before.”2

1 Numbers in square brackets ‘[ ]’ reference surveys in Appendix A
2 Ryan, P., “Capital Budgeting Practices of the Fortune 1000: How have things changed?”, Journal of
Business and Management, Vol. 8, No. 4, Winter 2002
Marc Lien

International Financial Management
Page 3
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE


The most thorough survey to date is by Campbell (2001) [16] which is unique
given its large sample size, broad scope and the availability of firm and
manager-specific characteristics.

2. Review whether capital budgeting procedures differ across industries,
countries or firms of different size

Given that the line of inquiry around managers trending towards normative
approaches had been exhausted, at least in a domestic context, researchers
turned their attention to seeing whether these trends were observable across
multiple dimensions.

Many started investigating differing practices across countries: Kester [9]
(Singapore), Blazouske [10] (Canada), Summers [11] (Europe and Asia),
Sandahl [12] (Sweden), Lazaridis [13] (Cyprus), Brounen [18] (UK,
Netherlands, France and Germany). One of the most interesting results from
these studies is the differing shareholder orientations of firm between
countries and the impact this has on capital management practices.

A second dimension was to look at whether sophisticated techniques were
equally prevalent across industries. Block (2005) [15] found that there were
significant differences in results relating to goal setting, determining the
required rate of return and accounting for the portfolio effect.

The impact of firms of different sizes was also explored. Brounen (2004) [18],
for example, found that smaller firms rely mostly on payback whilst larger
ones on NPV.

These differences in practices between countries, industries and firms of
varying size present opportunities for further research but also highlight areas
of potential bias in any future survey.

3. Investigate the challenges of capital budgeting for Multina ional Corp
t
orations

In addition to comparing capital budgeting practices in a domestic setting
across different countries, some researchers have begun to explore capital
budgeting practices in multinational corporations (MNCs).

Oblak (1980) [6] identified that many MNCs use DCF methods and that they
adjust for risk in their evaluation of foreign investment opportunities. He
Marc Lien

International Financial Management
Page 4
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

notes that MNCs have not significantly changed their way in which returns
from foreign projects are measured or the determination of the appropriate
discount rate. Block (1984) [7] followed up with a review of capital budgeting
methods in MNCs but did not explore the differences between domestic and
foreign project investment.

Goddard (1990) [14] dove deep into the political risk assessment process in
MNCs and found that the analysis was predominantly subjective and that
there was little integration of risk analysis into the formal capital budgeting
processes.

I would suggest that the paper that throws most light onto the international
capital budgeting processes is Block (2000) [17]. He explicitly examines the
extensions in theory and practice to domestic techniques in 146 multinational
corporations. He identifies a number of misapplications such as “applying
corporate wide weighted average cost of capital to foreign affiliate cash flows
rather than to cash flows actually remitted to the corporation.” Also that “risk
is frequently measured on a local project basis (in a foreign country) rather
than considering the portfolio effect on the total corporation.” He
summarized by saying: “Ultimately, it is shown that the survey respondents
hedge against the uncertainty of the procedures by adding a premium to the
weighted average cost of capital.”

Block’s paper is excellent in capturing the very latest practices in international
capital budgeting. He explores corporate policy, foreign investment and risk,
capital structure, cost of capital and operations considerations.


Marc Lien

International Financial Management
Page 5
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

2. DEFINE RESEARCH OBJECTIVES

It is important to determine the particular objectives of this new piece of work before
progressing as the structure, content and approach are dependent on the objective. I
can see four potential directions that it can take:

1. Extend existing survey research. This could, for instance, take previous work
on multinational capital budgeting carried out with the help of US firms and
repeat and extend the exercise with South American, European and Asian
firms.
2. Perform targeted research through a series of in-depth interview alone, to
explore a particular hypothesis. For example, interview with the CFOs of 5
private multinational conglomerates and CFOs of 5 publicly traded
multinationals to see explore whether potential differences in shareholder
impact capital budgeting approaches.
3. Survey firms to cover a new aspect of international finance that has yet to be
investigated in the field. Given the broad nature of previous multinational
surveys, there is plenty of scope for a deeper dive into one or two aspects. For
instance, how to multinational firms go about assessing risks prior to agreeing
to an international investment project.
4. Survey firms to test a hypothesis. For example: firms overestimate the risks
involved in investing in international projects and this overestimation is
reflected in their capital budgeting analysis.
Marc Lien

International Financial Management
Page 6
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

3. DEVELOP DRAFT RESEARCH INSTRUMENTS

Types of instrument
Researchers frequently use large sample analysis and clinical studies. Large sample
studies can increase confidence through their statistical power but are unable to gauge
answers to qualitative or probing questions, the research has to infer the answers.
Clinical studies offer excellent details and allow many of the biases discussed above to
be overcome but the small sample set only makes it impossible to generalize with any
confidence beyond the sample population.

The survey approach falls somewhere in between these two extremes in that a
moderately large sample of firms can often be attained (up to 350 in the case of
Campbell, 2001, [16]) at the same time a being able to ask for quantitative answers to
specific questions that externally available financial data does not provide.

Key issues with selecting survey method
Population issues
If the survey population is global multinational firms, it is important to compile a list
of these firms. Reviewing the various databases available at the Baker Library, I could
find no such ready made list. Step 5 details one approach that can be taken to compile
a list.
It is also unclear what is meant a multinational firm for the context of the survey.
One would need to decide the following in defining the population:
• Ownership – public or private
• Subsidiary definition – wholly owned, >50% or 0-50%
• Number of foreign subsidiaries – >0, arbitrary threshold, active international
investor - one or more investments in last 12 months or >x% of sales that are
foreign
• Industry – exclude or not investment and banking sectors
• Size – only top 500 firms on each stock exchange, range of small to large firms
given that capital budgeting practice varies along this dimension
• CAPEX spend – limit or not according survey to only firms that have had
growing CAPEX over last five years
• Growth – limit or not survey to firms that have been increasing shareholder
value
In addition to defining the survey, we need to determine who the recipient should be.
The research is focused on firm behavior yet it is unclear how closely related firm
behavior is to the expressed behavior of any one individual. Some people within the
firm may not have the knowledge required to answer the questions so the intended
recipient should be clearly defined. I would recommend targeting the CFO as that is
Marc Lien

International Financial Management
Page 7
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

the one person in the firm who would have an overview of all the financial aspects be
it capital budgeting, capital structure, performance measurement etc. It is not quite as
obvious how to avoid the completion of the survey being deferred to someone
without the appropriate oversight. In any case, it will be a challenge to identify
name, address and recipient of the CFO of our MNC population.

Question issues
Some of the decisions around the specific questions on the survey include:
• Should questions be open-ended or structured? Structured questions may be in
the form of multiple choice; for example, “How do you determine the target
leverage for the project?” a. same as firm’s target leverage, b. debt capacity of
project, c. level of cash-on-hand for investment. Answers to multiple choice
questions should be mutually exclusive and collectively exhaustive so some
answers sets might require an ‘other’ category.
Multiple response questions could also provide an insight into how many
different tools or approaches a firm takes. For example, “Why does your
company invest abroad?” a. access to new markets b. access to raw materials c.
improved production efficiency d. development of new knowledge e. political
safety f. fear of losing a market g. “bandwagon” effect h. strong competition at
home i. diversification benefits.
Score assignments might also be useful to gain an idea of perceived relative
importance of factors. For example, “When considering an investment, how
concerned are you about each of the following political risks [0-6]?” a.
expropriation b. inconvertibility c. imposition of a new tax d. removal of
agreed subsidy e. implementation of new tariffs f. creating barriers to sourcing
g. unilateral changes to key contract provision h. ethnic strife. Forced ranking
could also be used.
• Complexity of questions is also important. A concern I have is that the survey
could ask a series of specific questions about risk management, cash flow
analysis etc. but some firms may not manage any risks or analyze cash flows
beyond payback period. This would reduce the useable data for some of the
questions.
• It is worth considering the use of screening questions to see whether the CFO
actually did respond to the survey or whether the task was delegated. For
instance, the anonymous survey could capture the age of the respondent. This
age could then be cross-checked with the age of the CFO as noted in CapitalIQ
or the like. Alternatively, we could ask the respondent for how many years
they have been with the firm. We could then use this as an extra check when
looking to explain outlier data.
• The likelihood of completion may be partly dependent on the type of question
asked. Is it, for example, worth excluding survey questions that require the
Marc Lien

International Financial Management
Page 8
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

respondent to look information up. e.g. “What percentage of sales comes from
foreign subsidiaries?” I recommend circumventing this problem by converting
the answers to multiple choice: “0%, 1-25%, 26-50%, 51-75%, >75%”

Bias issues
Bias may creep into the survey items in a variety of ways.
• Ambiguity - Some of the financial terms and concepts may not be commonly
understood. For instance, I can see people interpreting the phrases, ‘sensitivity
analysis’ and ‘scenario analysis’ incorrectly. It would help to add a description
of these phrases: “Sensitivity analysis allows for the change in one input
variable at a time, such as sales or cost of capital, to see the change in NPV”
and “Scenario analysis allows for the change in more than one variable at a
time, including probabilities of such changes, to see the change in NPV”.
• Social desirability – Respondents are likely to want to look good in the eyes of
others if the survey is not anonymous. Making the survey anonymous,
however, will require a separate set of questions that capture key
characteristics of the company and the manager so that analysis can still be
performed. Even if using an anonymous instrument, it might be the case that
a respondent answers based on what they believe should be done or what they
think is done; both of which might be different from what is actually done.
• Non-response bias

Aggarwal (1980) discusses the limits of the usefulness of management science models
in more detail.

Developing Survey Questions
My original objective was to develop a completed survey which covered international
capital budgeting practices for multinational corporations and I worked on iterating a
series of potential questions. Subsequently, however, and near the completion of this
paper, I came across additional surveys by Block and Goddard (as described above)
which survey the exact elements that I had originally set out to do. My guiding
principal for this project was to provide useful data for implementation in the future
so I decided to not repeat the survey design based on my guess at what the research
might try and achieve.

Additionally, it became apparent that I am not the best judge of the final set of
questions to include in a survey – the effectiveness of the questions in eliciting a
usable response is. Theory on survey pre-testing (see step 4) recommends starting off
with a much larger set of questions than will end up in the final instrument.

Marc Lien

International Financial Management
Page 9
70605147
EC Course Paper

---

INTERNATIONAL CAPITAL BUDGETING IN PRACTICE

I therefore decided that the most effective use of this element of the report was to
collate and structure all the questions from all available surveys so as to provide a
useful reference guide when creating a new survey for future research upon clarifying
the objective as describe in step 2.

Appendix B provides a categorized list of all the research questions asked in 10
available surveys, and their multiple-choice answer options cross-referenced to their
source.

The categories I used when reviewing the survey questions are:

• Firm characteristics (industry, financials, international)
• Management characteristics
• Corporate policy (capital budget, capital budgeting process, capital rationing)
• Management perceptions (purpose of firm, time horizon)
• Capital budgeting (required rate of return, project selection, techniques
• Determining the cost of capital (methods, risk adjustment, management
science techniques)
• International investments (risk analysis, capital structure, cost of capital,
operations, income measurement)

In addition to this survey question guide, I posit some questions of my own which I
believe have not been asked in any survey to date. My questions and recommended
multiple-response options (Appendix C) fall into two categories: Corporate strategy
and foreign investments, and international capital budgeting.


Marc Lien

International Financial Management
Page 10
70605147
EC Course Paper

---