Capital Budgeting Behaviour in Small Firms
Rational or intuitive investing?
Renze Vonk
Capital budgeting behaviour in small firms
- Rational or intuitive investing?
Author: Renze Vonk
Student number: 1258567
University of Groningen
Faculty of Economics
Specialization (doctoraal): Management Accounting & Control
First reviewer: dr. S. Tillema
Second reviewer: prof.dr. H.J. Ter Bogt
November 2006
Table of contents
1 Introduction...4
2 Theoretical framework...6
2.1 Economic / Finance theory...6
2.2 Economic capital budgeting techniques...10
2.3 Alternative capital budgeting techniques ...13
2.4 Capital budgeting sophistication over time...14
2.5 Contingency Theory ...14
2.6 Behavioural Theory ...18
2.7 Decision rule ...21
3 Methodology ...23
3.1 Sample selection ...24
3.2 Survey procedure and design ...25
3.3 Response ...25
3.4 Interview procedure ...27
4 Survey results ...28
4.1 Firm and manager characteristics ...28
4.2 Capital budgeting techniques ...29
4.3 Game theory and real option usage ...31
4.4 Behavioural aspects of decision making...32
4.5 Motives for capital budgeting techniques ...32
4.6 Analyses of the relationships...33
4.7 Summary of the main findings ...39
5 Interview results ...41
5.1 Investment selection criteria...41
5.2 Intuition as a basis for investments...44
5.3 Concluding remarks...45
6 Conclusions and further research ...47
1 Introduction
Allocating scarce resources among possible investment opportunities is one of the most important decisions top management has to make with respect to implementing the firm’s strategy (Bowman & Hurry, 1993). Making the right decisions is of crucial importance for maximizing shareholder value and chances of firm survival. The process of selecting and evaluating investment opportunities is the field of capital budgeting. Capital budgeting techniques help managers select n out of N investment opportunities with different returns and risks (Verbeeten, 2006). But how are decisions actually made? Do economic capital budgeting techniques dominate decision making, or are there other, more subjective, reasons relevant for decision making?
The economic basis for capital budgeting is the seminal work of Dean (1951) and Lutz and Lutz (1951). Based on their work, possible investments should be sorted on their rates of return. The goal of capital budgeting is to identify the best investments which maximize firm value. Ideally a long run capital budgeting plan should be made to adjust to both future demand and supply of capital. Based on economic theory, discounted cash flow (DCF) techniques, such as the internal rate of return (IRR) and net present value (NPV), have been developed as the main techniques for selecting investments. In practice, payback period (payback) and accounting rate of return (ARR) are also often used for capital budgeting. The academic literature generally distinguishes among naïve and sophisticated capital budgeting techniques (Haka et al., 1985). Sophisticated techniques, such as IRR and NPV, are favoured because of their economic validity, whereas naïve techniques, such as payback and ARR, are mostly used because of their low cost and easy use. Theoretical work in the field of finance (Trigeorgis, 1993) indicates that DCF techniques have shortcomings when future cash flows are uncertain. The real option approach (ROA) and game theory (GT) are techniques that deal with this uncertainty.
During the last decades numerous studies have investigated the use of the various capital budgeting techniques (e.g. Klammer et al., 1991; Pike, 1996). Results from these studies show that, although declining, a large part of the firms still use naïve capital budgeting techniques. In particular, smaller companies appear to make more use of naïve capital budgeting techniques. Remarkably, almost all studies on capital budgeting focused on the largest firms in the US and the UK. Research on small and medium sized enterprises (SMEs) has been limited (Runyon, 1983). It is striking that so little is known about these firms, while for example 99%1 of all firms in the
Netherlands are SMEs. This thesis attempts to contribute to the academic literature by
investigating capital budgeting practices of SMEs in the Netherlands. The first research question this thesis tries to answer is the following:
(1) What capital budgeting techniques do Dutch SMEs use?
But what factors drive SMEs to choose these techniques? Size is suggested to be an important factor, but are there other factors which influence the choice for the capital budgeting technique? Contingency theory suggests that organizational and environmental factors should fit with the design of the capital budgeting process. To examine which factors are important for SMEs, the following research question is formulated:
(2) Why do Dutch SMEs choose specific capital budgeting techniques?
In addition, this thesis will deal with some of the behavioural aspects of capital budgeting. Do decision makers always follow the results from economic techniques, or are there other motives for decision making? It is known from the field of behavioural finance, that decision makers do not always make economically rational decisions. Behavioural issues, such as managerial hubris (Roll, 1986) and heuristics and biases, influence the decision making process. Behavioural aspects are particularly important for smaller companies, because their decision making is more ad hoc (Perren & Grant, 2000). In addition, in many small firms the manager is emotionally involved with the company. The decision maker for small firms is often the founder or owner of the firm. The decision maker might therefore be more likely to take decisions based on entrepreneurial judgement and emotional feelings. To examine the influence these behavioural aspects have on the decision making process, the following question is formulated:
(3) How are capital investment decisions made in practice in Dutch SMEs?
The usage of the different capital budgeting techniques is analyzed using a survey of 223 companies in the metalworking industry. Six main decision makers were then interviewed to examine the motivation for the choice of particular capital budgeting techniques, and the way in which decisions are made in practice.
The remainder of this thesis is organized as follows. Chapter 2 presents a theoretical basis for capital budgeting and describes the main capital budgeting techniques used in practice. Chapter 3 describes the research methodology and data collection of the research. Chapter 4 presents the results and analysis from the survey. In chapter 5 the results from the interviews will be discussed. Finally, chapter 6 contains the conclusion and discusses the results of the research.
2 Theoretical framework
As discussed in the introduction, capital budgeting is the process of selecting and evaluating investment opportunities (Dean, 1951). In this context an investment is defined as committing resources that could be used for current consumption, so that more can be consumed later (Lister, 1984). The goal of capital budgeting is to select the best investments which maximize firm value (Hirshleifer, 1958).
Capital budgeting may seem as a purely economic trade-off between current and future consumption, but it is the domain of industrial engineers, operations research analysts and finance specialists as well (Thompson, 1976). In addition, Tversky and Kahneman (1974) have shown that psychological aspects should be considered too when studying decision making under uncertainty. This chapter discusses the different theories, the different capital budgeting techniques and qualitative aspects relevant for decision making. Where economic theory suggests that the decision making process should be a rational choice between quantitative techniques, the behavioural and contingency approaches suggest a more qualitative way of decision making. Since this thesis focuses on decision making in small and medium sized enterprises (SMEs), it will be interesting to see the influence of these qualitative and quantitative approaches in practice.
The chapter starts with the foundations of modern finance theory. Section 2.2 discusses the economic capital budgeting techniques which resulted from these theories. Practice has shown that, although these techniques are superior in use, other capital budgeting techniques are used as well. The results from various surveys show that primarily smaller companies use naïve capital budgeting techniques. Section 2.3 discusses these naïve techniques. Section 2.4 gives an overview of the various studies of capital budgeting sophistication over time, which show an increasing trend for discounted cash flow (DCF) techniques. Next, the contingency theory is discussed and various organizational factors which are thought to influence the choice for the capital budgeting technique are presented. Section 2.6 discusses the influence of behavioural aspects of decision making. Finally, section 2.7 examines the decision rule for decision making. That is, this section discusses the decision rule for capital budgeting techniques and a qualitative approach for decision making.
2.1 Economic / Finance theory
The foundation of modern finance is the economic utility theory. In utility theory each possible outcome is associated with a certain amount of utility to the decision maker. The basic
concept of utility theory is that decision makers are expected to be fully rational and to maximize their expected utility (Von Neumann and Morgenstern, 1944). The theory of cardinal2 utility is
based upon five axioms of choice under uncertainty: comparability, transitivity, strong independence, measurability and ranking (Fama & Miller, 1972). Other powerful theories have been developed from these utility axioms, such as the mean-variance theory, asset pricing theories and the option pricing model (Weston, 1981). The decision maker is assumed to aim at maximizing his expected personal utility when faced with investment decisions. When the manager is the full owner of an unlevered firm, it is easy to see that accepting the best decision for the firm is in the interest of the manager too.
In practice, however, most of the managers are only to a limited extent owners of the firm and most firms are levered to some level. Agency theory has been developed to discuss information asymmetry among firm owners (shareholders), decision makers (managers) and bondholders. Agency theory explains how sub-optimal value development of the company can exist when the manager owns only a fraction of a levered firm’s common stock.
2.1.1 Mean variance theory
Mean variance theory suggests that the expected return and variance are related to each other. Mean variance theory influences three major issues of corporate finance: security valuation, asset expansion and capital structure (Rubinstein, 1973). Concerning the field of capital budgeting, security valuation and asset expansion are central themes. The basic security valuation theorem of a security j is under certain assumptions3 as follows:
) , ( ) (Rj = RF +
λ
Cov Rj RM Ε (1)Where RF is the rate of return on a risk-free security,
Rj is the rate of return on security j,
RM is the rate of return on the market portfolio of risky securities, and
λ is positive constant.
When considering asset expansion, a firm should only accept a project if:
)
,
(
)
(
R
j0>
R
F+
λ
Cov
R
j0R
MΕ
(2)Where Rj0 = Xj0 / COSTj0 is the rate of return of the project,
2 When cardinal utility is used, the magnitude of utility differences is treated as an ethically or
behaviourally significant quantity; source: http://en.wikipedia.org/wiki/Utility_theory.
3 Assumptions from Rubinstein (1973): (1) single period context, (2) no restrictions on short selling and (3)
Xj0 is the entire marginal dollar return of the project, and
COSTj0 is the entire marginal cost of the project.
Based on equation 2 a project should be accepted by utility maximizing decision makers if the project’s expected return exceeds the appropriate risk-adjusted discount rate (using the security market line). Based on this mean variance theory, the Capital Asset Pricing Model (CAPM) and the Arbitrage Pricing Theory (APT) have been developed as the main models for asset pricing.
2.1.2 Asset pricing models
Building on mean variance theory, Sharpe (1964) and Lintner (1965) have developed the CAPM to value asset prices by determining the theoretically required return on an asset. The basic idea of the CAPM is that the return of an asset is fully related to the non-diversifiable risk, measured as beta (β). β measures how risky the asset is related to the market risk. The decision rule suggests that only projects above the capital market line should be accepted, as discussed in equation 2 (Rubinstein, 1973). Figure 1 illustrates the relationship between return and risk in a graphical way.
The weighted average of the cost of capital (WACC) shows the rate of return currently required by the equity and debt holders of the firm, based on its current projects. However, the WACC can only be used as a good indication of the minimum required return for a new project, when the project is of equal risk as the average risk of the company. When looking at figure 1,
Firm cost of capital WACCj
Security market line
Risk (β) Cov(Rj,Rm) Expected return E(Rj) Risk-free rate Rf 0 Firm risk Cov(Rj*,Rm) λ
•
A
•
B
only projects above the security market line should be accepted. Project B should therefore be accepted and project A should be rejected.
The arbitrage pricing theory (APT) is an alternative asset pricing theory that builds on the absence of arbitrage possibilities. For capital budgeting purposes, a project should be accepted if, at its costs, it appears to offer arbitrage opportunities (Ross, 1978). The APT is less restrictive in its assumptions than the CAPM and allows for an explanatory, as opposed to statistical, model of asset returns (Lister, 1984). For more information on the APT read Ross (1976) and Ross (1978).
2.1.3 Option Pricing Model
The option pricing model (OPM), developed by Black and Scholes (1973), can be used to value future uncertainty in payoffs of investments. Although Black and Scholes focus on options on financial assets, their model can also be applied to ‘real’ investment opportunities. As Miller and Modigliani (1961) have shown, the value of a firm consists of the value of its current assets and the value of its future growth opportunities. These future growth opportunities can be seen as ‘real options’. The value of these real options depends on the discretionary future investments of the firm (Myers, 1977). As Myers states:
“the distinction is … between (1) assets that can be regarded as call options, in the sense that their ultimate values depend, at least in part, on further discretionary investment by the firm and (2) assets whose ultimate value does not depend on further discretionary investment.”
Thus, the value of these assets with uncertain payoffs is affected by the firm’s future strategy. When the CAPM and the OPM are combined, a more complete model of security pricing can be created (Galai and Masulis, 1976). By combining the CAPM with the OPM Galai and Masulis attempt to integrate option value with systematic risk. They suggest that the choice of new investments will be affected by the firm’s leverage. Since stockholders and bondholders have conflicting interests regarding the risk of the cash flows, levered firms will choose different projects than unlevered firms. Debt holders bare the downside risk of a project, because the credit repayment becomes riskier as the risk of the project increases. The value of the stocks and bonds will be affected by unanticipated increases or decreases of risk.
Individual capital investment opportunities can also be described in terms of real options. When these investments have option-like and uncertain payoffs, management has flexibility in decision making based on the intermediate outcomes. This flexibility can be valued with the real option approach (ROA) or the decision tree analysis (DTA) and used for decision making (see McGrath et al. (2004) for an overview of various option-like investments). The valuation of
flexibility allows us to better understand the value of investments with uncertain payoffs. The way in which the OPM is used for capital budgeting is discussed in paragraph 2.2.3.
2.1.5 Agency Theory
If the manager is the only owner of the firm, it is easy to see that maximizing firm value is in his own interest. However, if the manager owns less than 100% of the shares in the firm, conflicts of interests between the manager and the (other) owner(s) of the firm may arise, because the manager can pursue personal as well as corporate goals. Jensen and Meckling (1976) define such an agency relationship as “a contract under which one or more persons (the principal(s)) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent.” When there is asymmetric information, agency problems can arise between the agent and the principal. Since the manager is only interested in maximizing his personal wealth, he will trade-off company goals and personal well-being (perking behaviour). To limit this behaviour, costs must be incurred. The agency costs consist of bonding expenditures, monitoring expenditures and a residual loss.
Because of asymmetric information, the shareholders cannot see if the agent is making the right decisions. The manager knows that the shareholders can only see the firm’s output, and is thus interested to show positive output. However, positive output (e.g. an increase in short term profits) does not necessarily mean that shareholder value is created. For example, a decrease in maintenance increases short term profits but may actually decrease shareholder value. A manager focusing on maximizing current profits and increasing sales can actually decline firm value, while boosting his financial and social (i.e. pride and appraisal) compensation.
As Jensen and Meckling (1976) have shown, the amount of ownership of the manager determines the trade-off between the maximization of corporate and personal goals.
2.2 Economic capital budgeting techniques
From the economic theories discussed above, two main capital budgeting techniques have been developed over time. These include the internal rate of return (IRR) and the net present value (NPV). These techniques show clearly when should be invested in a project opportunity.
2.2.1 Net present value
The net present value (NPV) is a discounted cash flow (DCF) technique to calculate the value return of a project. It measures the difference between the cost of a project and the value of a project (Breadley et al., 2001). Ideally NPV uses all cash flows of the project and discounts them at the appropriate opportunity cost of capital. The discount rate greatly affects the NPV outcome and is therefore of crucial importance. For this reason, it is important that, the discount rate reflects the appropriate risk-adjusted opportunity costs of capital. Once the appropriate discount rate is determined, a project should be selected when the NPV is positive. The project then creates more value than its costs. Using the NPV technique, a decision maker can also choose among competing projects and decide about the timing of capital investment or replacement decisions. By maximizing the NPV of the available funds, decision makers also maximize shareholder value. The NPV technique is considered to be the literature’s most favoured capital budgeting technique (Hirshleifer, 1958).
NPV offers a correct calculation under almost all circumstances, but there are exceptions. As Hirshleifer (1958) illustrates in a theoretical setting, NPV fails to give correct answers only when capital markets are assumed to be inefficient and investments are non-independent. In a more practical setting, one of the main problems when using NPV as a capital budgeting technique is the uncertainty of cash flows. NPV makes implicit assumptions about an expected scenario and fails to value managerial flexibility (Trigeorgis, 1993). When cash flows are highly uncertain, NPV becomes less usable. The use of the real option approach (ROA), decision tree analysis (DTA) and game theory (GT) can help in uncertain environments. Section 2.2.3 and 2.2.4 elaborate on these sophisticated techniques.
2.2.2 Internal rate of return
The internal rate of return (IRR) is a way to calculate the expected economic rate of return of a project. Similar to the NPV, it is a DCF technique and it discounts all cash flows. The discount rate is the internal rate of return of the project that yields a zero NPV. In general, a project should be accepted if the IRR of the project is higher than the project’s cost of capital or hurdle rate, which is the minimum required rate of return.
Although the IRR can correctly identify whether a project is profitable, it cannot be used to choose between mutually exclusive projects. The IRR fails to quantify the contribution to firm value as NPV does. A decision maker cannot maximize firm value when he has to choose among different projects with different IRRs. A project with a higher IRR not necessarily implies a larger
increase of value for the shareholders. IRR calculations can also be problematic when cash flows are different than the standard (-++++) cash flow scheme. If the project starts with a cash inflow followed by cash outflows, the standard IRR decision criterion (accept when IRR > hurdle rate) has to be reversed. When there are multiple changes in the sign of the cash flows, the IRR does not work at all (Breadley et al., 2001).
2.2.3 Valuing flexibility
When an investment decision is made, it is often uncertain how the project will develop. As new information becomes available about market conditions and future cash flows, the management may have the opportunity to alter its operating strategy (Trigeorgis, 1993). The management may, for example, decide to expand or abandon the project. When management can alter the project based on intermediate outcomes, this flexibility can be valued. Two main techniques to do this are the real option approach (ROA) and decision tree analysis (DTA). The ROA has been based on the option pricing model (OPM). “ROA is a controlled means of systematically identifying the interplay between intermediate outcome states and alternative managerial actions and specifically valuing managerial flexibility” (Alessandri et al. 2004). A real option gives the right, but not the obligation to take an action.
A main advantage of the ROA is that it allows to profit from the upside potential and limits the downside risk of the project. Where traditional DCF analysis has problems valuing uncertain outcomes (i.e. it assumes an expected scenario), the ROA systematically values all possible asymmetrical outcomes. According to Kemna (1993) another main advantage of the use of ROA is that it provides management a framework for a structured view of the project’s potential. Real Options can exist in various forms: the option to defer investments, the option to expand, the option to abandon or switch use, and the option to default during construction (Trigeorgis, 1993).
An alternative, somewhat simpler way to value flexibility is to use DTA. DTA involves discounting contingent future cash flows of all possible outcomes at the project’s cost of capital (Koller et al., 2005). This technique works best when the underlying risk is diversifiable because only then the correct cost of capital can be calculated. However, when the underlying risk is not diversifiable, it is unclear how to estimate the cost of capital correctly.
2.2.4 Game theory
Where most capital budgeting techniques are static, the basic idea behind game theory is that capital budgeting decisions can be strongly influenced by existing as well as potential competitors
(Huisman et al., 2003). Game theory (GT) takes asymmetric information and strategic interaction into account in its analysis (Allen and Morris, 1998). With GT, firms try to anticipate competitors’ moves and make their decisions based on this information. Information about competitors is of crucial importance for decision making. Under uncertainty and imperfect information, GT can provide firms with an incentive to invest early in order to gain a first mover advantage and eliminate investment opportunities for competing firms (Murto and Keppo, 2002).
2.3 Alternative capital budgeting techniques
In practice, it appears that, besides the discussed sophisticated techniques, naïve techniques are often used as well. The most important naïve techniques are the accounting rate of return (ARR) and the payback period (payback).
2.3.1 Payback period
A project’s payback period is the length of time before an investment recovers its initial investment (Breadley et al., 2001). Generally, the sooner the investment pays off its initial investment costs, the more attractive the project is. However, payback only considers cash flows occurring within this payback period. All cash flows beyond this period are not considered for evaluation. Projects are thus not judged on the total amount of cash flows, but only on the period which is needed to pay back the initial investment. Moreover, the payback technique does not discount any cash flows.
An adjustment of the payback technique is the discounted payback period. This is the number of years before the present value of the prospected cash flows equals the initial investment. This adaptation solves the discounting problem, but still does not include any cash flows after the payback period.
2.3.2 Accounting rate of return
The accounting rate of return (ARR) calculates the accounting return of a project by dividing total accounting income by the book value of assets (Breadley et al., 2001). The ARR uses accounting profits instead of cash flows. The problem with accounting profits is that they are subject to accounting rules and that there are multiple ways to calculate them. Accounting profit does not correspond with economic profit, because it does not take account of the capitalization of certain activities such as research and development (Fisher and McGowan, 1983). Accounting
profits are of limited use for capital budgeting because they do not consider cash flows and thus do not show investment needs and available cash flow for investing. As Fisher and McGowan further state: “accounting rates of return are useful only as they yield information as to economic rate of return4”. Another problem with ARR is that it does not consider the time value of money.
2.4 Capital budgeting sophistication over time
Since the seminal work of Dean (1951) and Lutz and Lutz (1951), researchers have studied the use of sophisticated capital budgeting techniques. Comparisons between these surveys show an increase in the use of discounted cash flow (DCF) techniques over the years (Pike, 1996; Ryan and Ryan, 2002; Rosenblatt and Jucker, 1979; Klammer et al., 1991). The extensive overviews of capital budgeting surveys in these articles show that, although declining, a large part of the firms still use naïve capital budgeting techniques. However, a proper comparison between these surveys is difficult for two reasons (Pike 1996). First, a main difference between the surveys is the firm size within the sample group. While most surveys focused on the largest firms, some surveys focused on smaller firms (Runyon, 1983). Second, surveys might overstate the usage of sophisticated techniques, because firms with sophisticated capital budgeting techniques are more likely to respond than firms with less sophisticated techniques (Rappaport, 1979). Surveys with different response rates are therefore more difficult to compare.
Although it is difficult to compare these surveys, comparisons do show that smaller firms make more use of naïve capital budgeting techniques. But why do smaller firms appear to make more use of naïve capital budgeting techniques? Are there organizational characteristics which influence the design of the capital investment process? Or do behavioural factors influence decision making more for smaller firms than larger firms? The next two sections try to determine some important issues which may influence this choice for more naïve capital budgeting techniques.
2.5 Contingency Theory
When looking at the reasons why smaller firms adopt less sophisticated capital budgeting techniques, it is important to look at the characteristics of these firms. Firms might differ from each other because of size, capital intensity, risk, managerial style, financial status (Pike, 1986), technology, organizational structure, and environment (Otley, 1980). According to Pike (1986),
this can lead to different degrees of formalization, sophistication, complexity and specialization of capital budgeting behaviour. The way capital budgeting is used in practice might be related to some of these organizational characteristics. This would give rise to a contingency theory perspective. The contingency approach of management accounting is based on the premise that no universally appropriate accounting system applies equally to all organizations under all circumstances (Otley, 1980). The contingency theory suggests that specific aspects of a capital budgeting system are associated with certain organizational and environmental factors. Otley (1980) states that the contingency theory is based on the linear framework shown in figure 25.
The design of the capital budgeting system should fit with the organizational design (Pike, 1984), which in turn should be based on the different contingent variables. SMEs might thus choose more naïve capital budgeting techniques because it better fits with their environmental and organizational characteristics. This would suggest that small firms have a good reason not to adopt sophisticated capital budgeting techniques.
As just showed in figure 2, the design and operation of the capital budgeting system should fit with the corporate context (Pike, 1984). Pike reached the conclusion that the sophistication of a company’s capital budgeting system and its performance are not related to each other. For example, a highly advanced and rigidly administered capital budgeting system might not fit with a small innovative organization because it constrains innovative ideas and entrepreneurship. Pike (1986) focuses on three aspects of corporate context that are assumed to be associated with the design and operation of a firm’s capital budgeting system: organizational, environmental and behavioural characteristics. Since the surveyed companies are all in the metalworking industry, environmental characteristics are assumed to be quite the same.
5 Otley (1980) designed this framework to show the links between the contingency factors and the type of
accounting information system (AIS) for achieving effective performance. Figure 2 is modified from Otley in a way that it focuses on the type of capital budgeting technique instead of the type of AIS.
Contingent variables
(e.g. size, environment, capital intensity)
Organizational design
(e.g. decentralization, managerial style)
Type of capital budgeting system
(formalization, sophistication, complexity, specialization)
Organizational effectiveness
2.5.1 Organizational characteristics
Size
As suggested by various surveys (e.g. Filbeck and Lee, 2000; Graham and Harvey, 2001), firm size is of major importance for the use of capital budgeting techniques. It is known from these studies that smaller firms use less sophisticated capital budgeting techniques. There may be two reasons for a positive relationship between size and capital budgeting sophistication (Kim, 1982). First, larger firms are more likely to have skilled full-time staff members for capital budgeting, whereas small firms tend to have no or just a few skilled employees. Second, large firms generally make more large and complex capital expenditures, which require the use of more sophisticated techniques.
Capital intensity
Capital intensity is the amount of fixed or real capital present in relation to the amount of labour. It is often measured as net fixed assets per employee. Capital intensity is sometimes used as a measure for technological sophistication (Pike, 1986). Highly capital intensive firms are suggested to operate more administratively oriented capital budgeting systems (Klammer, 1972). Formalization
Formalization refers to the use of formal rules and controls (Simons, 1987). Higher formalization is associated with more guidelines, rules, procedures, controls, and more planning. Pike (1986) relates a number of characteristics of formal behaviour to the sophistication of capital budgeting. More formalization is thought to be related to a higher level of capital budgeting sophistication.
Decentralization
The degree to which decision making authority is diffused in the firm is usually referred to as decentralization (Haka, 1987). More decentralization gives more authority to lower managers in the organization. Haka finds that the effectiveness of sophisticated capital budgeting techniques is positively related to the level of decentralization. According to Haka, more decentralization motivates managers to spend more time and effort on the collection of data and on budget related activities.
According to Perren and Grant (2000) smaller firms generally have a more decentralized and less formal way of decision making, but they do not adopt more sophisticated capital budgeting
techniques (Runyon, 1983). This seems to contradict Haka’s (1987) findings that more decentralization should be associated with more sophisticated capital budgeting techniques. If small firms are more decentralized but they use less formal and sophisticated capital budgeting techniques, there may be other (e.g. behavioural) reasons for the adoption of capital budgeting techniques.
2.5.2 Behavioural characteristics
Managerial style
Managerial style refers to the attitude of managers towards risks and the way capital budgeting decisions are made. This can differ between organizations from a very informal approach to a highly bureaucratic and documented approach (Lorange and Morton, 1974). Recalling figure 2, the organizational design should fit with the contingent variables. The managerial style should thus fit with the organization characteristics. A small and innovative company is likely to perform better with an organic and decentralized managerial style than a highly bureaucratic managerial style. According to the contingency framework, the capital budgeting system should fit with the managerial style in order to be effective.
Professionalism and financial skills
Since the manager is of crucial importance for making projects work, one can suggest that his educational background is of major importance for the use of a sophisticated capital budgeting system. Silvola (2005) finds that the educational level of the CEO is one of the major driving factors of adaptation of new management accounting practices. Additionally, Graham and Harvey (2001) find that among small firms, CEOs without an MBA education are more likely to use the payback criterion rather than the more sophisticated NPV or IRR technique. A higher educational level in finance or economics would thus suggest that more sophisticated capital budgeting techniques will be used.
History of the organization
Pike (1986) suggests that the organization’s historical ex post performance influences the need and benefits of a sophisticated capital budgeting system. According to Pike, the main advantage of sophisticated capital budgeting techniques is distinguishing marginally profitable from marginally unprofitable investments. Highly profitable firms are less likely to gain the
benefits of sophisticated capital budgeting techniques. Changes in the primary capital budgeting technique might therefore be related to recent changes in the organization’s performance.
2.6 Behavioural Theory
Traditional finance theory, based on utility theory, assumes that all individuals make completely rational decisions based on the five axioms of cardinal utility (Fama and Miller, 1972). The sophisticated capital budgeting techniques based on traditional finance theory were designed to aid rational decision making. In theory, decision makers are presumed to define and order goals, foresee all possible actions, to anticipate all possible outcomes from these actions, and to make a rational choice of the possible actions to optimize the goals (Northcott, 1992). In reality, decision makers often do not have information about all possible outcomes and they do not always make rational decisions. A wide range of articles has been written on the fact that decision makers may not act according to the principles of rationality when faced with risky and uncertain decisions (see e.g. Kahneman & Tversky, 2000). The field of behavioural finance studies these imperfections and limitations of rational behaviour. As March (1994) states:
“The core notion of limited rationality is that individuals are intendedly rational. Although decision makers try to be rational, they are constrained by limited cognitive capabilities and incomplete information, and thus their actions may be less than completely rational in spite of their best intentions and efforts”
As a result of these limitations, decision makers use four different simplification concepts: editing, decomposition, heuristics (and biases) and framing (see March (1994) for an explanation of these concepts).
When faced with uncertainty, individual decision makers have to make a number of subjective expectations concerning, for example, expected cash flows and the required rate of return. Decision makers are subjected to heuristics and biases when making these estimations. The main heuristics and biases that are important for the field of capital budgeting are described in the next section. Individual decision makers can also have other interests than maximizing shareholder value. As reflected in the agency theory, decision makers can act in their own interest when they own less than all equity. Agency theory also suggests that management may adopt risky projects that transfer wealth from debt holders to equity holders. These decisions, which are in the interest of the individual manager, might not be in the interest of the firm as a whole. This is known as the goal incongruence problem. Another distortion of rational economic behaviour in
the decision making is called hubris or managerial overconfidence, which is discussed in section 2.6.2.
2.6.1 Heuristics and Biases
There are many heuristics and biases discussed in the behavioural finance literature. However, most of them focus on individual behaviour concerning personal preferences for investing. These include heuristics like representativeness, availability, anchoring (Tversky and Kahneman, 1974) and affect (Slovic et al., 2002). With respect to capital budgeting, it is interesting to look at heuristics and biases which influence corporate decision making. Two important biases are the hindsight bias and the bias to overstate cash flows.
The hindsight bias is the tendency to see past uncertain events as if they were predictable and reasonable to expect. As a result of this bias, people perceive their (ex ante uncertain) prediction as being right and logical afterwards. For example, a manager has the opportunity to invest in a project which success depends on the world oil price. Since the oil price is ex ante uncertain, the manager does not know whether the project will be profitable. If the oil price turns out to be positive for the project, the manager is likely to take the credit for his lucky gamble as if it was skill instead of luck.
In theory, performance evaluation should be based on ex ante decision information (Lipe, 1993). In practice, however, most performance evaluations are affected by decision outcomes when decision makers tend to credit their performance. A famous illustration is made by Brown and Solomon (1987): students evaluated a capital budgeting decision differently when outcome information was present (success or failure) than when only ex ante information was present. As a result of this ex post performance evaluation, rational decision makers can be expected to use hindsight to boost their performance (Lipe, 1993).
Another bias is the overstating of cash flow predictions. Individuals are primarily concerned about their personal goals, such as their future career, remuneration, and status and power. These can be incompatible with the corporate goals, and can lead to a goal incongruence problem. This problem can exist due to information asymmetry between the shareholders (principals) and the manager (agent) as discussed in the agency theory. When the manager overstates the projected cash flows, the project may look better and signal better managerial performance. Hirshleifer (1993) adds that managers have an incentive to use investment choices as a tool for building their personal reputations or reputations of their firms. These incentives come in three main forms: visibility bias, which is manipulation of short-term indicators of success in a positive way;
resolution preference, which focuses on the timing aspect of good and bad news; and mimicry and avoidance, which concerns the managerial behaviour to take only (perceived) good actions and avoid (perceived) bad actions. As a result of these incentives, managers are likely to overstate projected cash flows and select sub optimal investments, which may even lead to value destroying projects.
The above section focuses on managers who overstate cash flows, which can be a problem in firms where the manager is not the owner. Since many general managers of SMEs are owners as well, the effect is not likely to be an issue. However, subordinates might also propose investments which are overstated. Since these individuals have their own personal goals, they might overstate cash flows of possible investments in order to achieve these goals.
2.6.2 Managerial overconfidence and hubris
An interesting aspect of firm behaviour occurs when mergers and acquisitions (M&A) take place. Empirical evidence shows that most M&A deals do not pay for the bidding company (Bruner, 2003). If there is no value in M&A for bidding firms, then why do firms continue to pursue M&A strategies to expand their business? Roll (1986) comes with an interesting explanation for this fact: hubris. Hubris is the perceived overconfidence of managers’ own performance versus the performance of others. Roll suggests that M&A decisions reflect primarily individual decisions. It is known from Kahneman and Tversky (2000) that individuals do not always make rational decisions when faced with uncertainty. Hambrick and Canella (1993) concur with Roll (1986) and show that a sense of superiority of decision making managers is a major reason why M&A transactions occur. Hayward and Hambrick (1999) show that several factors influence CEO hubris: recent acquirer performance, media praise for the CEO, CEO’s relative pay, combined chairman and CEO position, and the percentage of inside directors in the board. Thus when recent performance is good, the appraisal for the CEO is strong, and the corporate control is poor, the CEO is likely to become overconfident for future investments.
Although the hubris hypothesis mainly applies to M&A behaviour, it can just as well be extended to capital budgeting. For large and prestigious investments it is reasonable to expect similar managerial behaviour. Managers who undertook good investments in the past may be more likely to actively adopt new investments than managers who did not perform this well.
2.7 Decision rule
Now that the main theories have been discussed, this section focuses on the decision making process in organizations. The way how decisions are made is very important for understanding the capital budgeting process. The decision rule focuses on the question: “How is a decision to be made among alternatives in terms of values and their consequences?” (March, 1994). Based on economic theory the basic decision rule is accepting projects which have a positive NPV or an IRR greater than the hurdle rate (Solomon, 1956). However, behavioural and agency theory have shown that not all decisions are economically rational from a corporate perspective. The contingency theory suggests that small firms may have a less formal capital budgeting system and thus a more behavioural way of decision making. Entrepreneurship or human judgment is often thought of as a major driver for decision making in SMEs (McMahon, 2005). The next two sections discuss the more qualitative approach to decision making, which is likely to be important for small firms.
2.7.1 Judgment and Entrepreneurship
A large number of cognitive aspects have been discussed in the literature (see McMahon (2005) for an extensive overview). Because of the limited scope of this thesis, the focus will be on a qualitative review of human judgment and entrepreneurship as a whole in the decision making process. Forbes (1999) indicates that the effects of managerial cognition on decision making are likely to be more direct and immediate in new venture settings than in the context of larger, more established organizations. According to Forbes, this managerial cognition is more important for entrepreneurs than non-entrepreneurs because of high uncertainty and ambiguity, high novelty, high emotional involvement, and a relatively flat organization. This is in line with Wright et al. (2000), who suggest that this cognitive approach to decision making is associated with faster learning and innovation.
2.7.2 Emotion
Related to the affect heuristic (Slovic et al., 2002) is the aspect of emotion. As Kida et al. (2001) state, managers consider both financial information and affective reactions when evaluating the utility of a decision alternative. Their experiments have shown that the managers tended to reject decisions that elicited negative emotional responses even though these had higher
expected values in a financial sense. In small firms, an entrepreneur often has a high level of emotional involvement in his company (Forbes, 1999). The decision making process in small companies is thus likely to be influenced by emotional aspects of the decision maker.
3 Methodology
Let us recall the research questions defined in the introduction: (1) What capital budgeting techniques do Dutch SMEs use?
(2) Why do Dutch SMEs choose specific capital budgeting techniques? (3) How are capital investment decisions made in practice in Dutch SMEs?
The different research questions ask for different approaches. The first question can be analyzed using quantitative data. The second and third question call for a more qualitative approach, since it is difficult to quantify the cognitive processes which influence decision making. A survey was sent to 2930 companies active in the metalworking industry to get an indication of the usage of the various capital budgeting techniques in Dutch SMEs. A main advantage of the survey research method is its broad scope (Verschuren en Doorewaard, 1999). A survey is a proper tool to gain a large amount of information about a limited number of variables. The use of a survey allows for a quantitative analysis of the usage of the different capital budgeting techniques of the sample. The usage of these techniques can be statistically related to a number of variables used in the survey to gain information on the influence of these organizational aspects. In chapter 4 the results of the survey will be discussed.
Two main disadvantages of the survey method are its lack of dept, and the limited flexibility (Verschuren en Doorewaard, 1999). Therefore, some main decision makers were interviewed to get a better understanding of the reasons for choosing a particular capital budgeting technique and the influence of behavioural factors on decision making. In-dept interviews are effective in that they can reveal the process of decision making for capital budgeting. The interview methodology has three main advantages over the quantitative survey (Meredith, 1998). First, the open interviews allow for a better understanding of the capital budgeting process, since the questions mainly consist of why rather than what and how questions. Second, interviews have the opportunity to clarify any vagueness. Third, interviews lend themselves for exploratory research. For capital budgeting in SMEs, the behavioural way of decision making is largely unknown and thus asks for an exploratory research method.
Section 3.1 motivates the selection of the sample firms for the survey. Section 3.2 discusses the procedure and design of the questionnaire. Section 3.3 discusses the response rate and the characteristics of respondents to the survey. Finally, section 3.4 discusses the interview procedure and the design of the interview questions.
3.1 Sample selection
When studying capital budgeting techniques and decision making in small firms, it is important to examine the differences between companies. To have a successful survey on the practice of capital budgeting in SMEs, the survey sample has to meet some criteria. First, to be of interest for research on capital budgeting, the firms should be regularly faced with major capital investments. Second, to reduce the influence of external factors such as uncertainty, risk and other environmental influences, only firms in one industry should be chosen. Third, since the average response in surveys is very low, a large number of small and medium size firms need to be active in this industry.
Based on these three criteria, the metalworking industry was selected for the survey. The metalworking industry is characterized by relatively large capital investments in machines and equipment, and consists of a large number of SMEs. Firms were selected based on their industry BIK6 codes ranging from 27 to 29. These BIK codes are defined as: (27) producing metals in
primary form, (28) producing metal products and (29) producing machines and appliances. For the selection of firms, the REACH database was used. This database comprises data on 1,7 million Dutch firms and allows researchers to select firms and collect data on a large number of variables. The data were based on the year 2005. Based on the data from REACH, firms that met a number of criteria were selected. First, the firms had to be active in one of the three industry BIK codes. Second, only firms with a total number of employees between 10 and 250 were selected. The lower boundary of 10 employees was used to eliminate firms with hardly any structure for financial decision making, and without sizable investments. The upper boundary of 250 employees was selected in line with the upper boundary of a medium size firm as defined by the Dutch small and medium size representative organization MKB7. Third, further selection was
made based on the availability of an email-address and the permission of the company to use the address information from the database. Finally, the companies were required to have an active economic status8 based on the definition of the Dutch chamber of commerce (KVK). After selecting companies based on the qualifications discussed above, a total of 3096 companies were presented by REACH. After eliminating double and incomplete entries, the survey was sent to a total of 2930 companies.
6 Bedrijfsindeling kamers van koophandel (BIK). 7 http://www.mkb.nl/Het_midden-_en_kleinbedrijf
8 A company is economically active if at least one person is active in the company for more than 15 hours a
3.2 Survey procedure and design
As discussed above, a survey is a proper research method to obtain a large amount of data about the usage of the different capital budgeting techniques. It allows us to quantitatively analyze the usage of the different techniques and the relation between certain organizational and managerial characteristics and the sophistication of capital budgeting techniques. The questionnaire consisted of five questions. The first question was designed to show the usage of the four most commonly used capital budgeting techniques, and the second question referred to the usage of real options and game theory. The respondents were able to choose on a 5-point Likert scale with what frequency they used the specified technique. The third question required the respondents to indicate to what extent the calculations from capital budgeting techniques were decisive in decision making. Additionally, the survey tried to identify the main motivations for choosing a certain capital budgeting technique. In the final part of the questionnaire information about some characteristics of the firm, the general manager and the financial manager was requested. The term general and financial manager is used instead of CEO and CFO, because it better reflects the formal position of managers in small firms. The questionnaire can be found in appendix 1.
The survey was pre-tested by a general manager of a large organization and was then sent by email to all companies selected from the database. The questionnaire was personally addressed to the first representative registered by the KVK. This person is generally the owner and the general manager of the firm. If the name of this person was not available, the survey was addressed to the general manager. The questionnaire was designed as a website which was easy and quick to complete.
3.3 Response
A total of 2930 firms were selected based on the criteria discussed in section 3.1. From these firms a total of 2487 firms received the email notification about the survey. The remaining companies could not be reached by email. A total of 233 responses were received from the survey, which is a response rate of 9,4%. Although this may seem low, it has to be said that this survey was on a voluntary basis and the response rate is comparable to other surveys by major studies (Graham and Harvey, 2001; Traham and Gitman, 1995). From the responses, a total of 223 were found to be usable.
Two tests were performed to investigate the non-response. First, the numbers of employees of the respondents were compared with the total population based on the data from REACH (Wallace and Mellor, 1988). Table 1 indicates that the respondents are generally larger than should be expected from the population. This indicates that the results from the survey might deviate from the population if size turns out to be a significant factor influencing the selection and use of capital budgeting techniques.
Table 1. Comparing respondents’ size with the population.
Population Respondents Number of
employees Frequency Percentage Frequency Percentage
10-19 927 31,64% 35 15,91% 20-29 679 23,17% 43 19,55% 30-49 635 21,67% 57 25,91% 50-99 407 13,89% 53 24,09% 100-250 282 9,62% 32 14,55% Total 2930 100,00% 220 100,00%
Second, the results of early respondents (186) were compared with those of late respondents (37) to get an indication about the non-respondents (Wallace and Mellor, 1988)9. Results from
appendix 2 show that late respondents are significantly larger than early respondents. Additionally, the late respondents do seem to make significantly more use of the payback technique, and are significantly more likely to base decisions fully on the calculations of the different techniques. These differences between early and late respondents indicate that the non-respondents might be different from the non-respondents. In combination with the low response level a possible non-response bias cannot be ruled out. The results, however, do not indicate that the use of sophisticated capital budgeting techniques is likely to be different between respondents and non-respondents.
Respondents were asked in the survey to identify their formal position in the organization. Table 2 shows that a large majority, 66,36% of the respondents, is indeed the general manager. 12,27% identified themselves as the controller or treasurer and 6,36% as the financial manager. These results indicate that the probability of a downward bias, by inexperienced staff responding to the questionnaire, is likely to be insignificant. Chapter 4 further discusses the characteristics of the responding firms.
Table 2. Formal function of the respondents. Frequency Percentage General manager 146 66,36% Other 33 15,00% Controller / Treasurer 27 12,27% Financial manager 14 6,36% Total 220 100,00%
9 Late respondents were respondents who returned the form 2 days or more after the mail was received.
3.4 Interview procedure
To answer the second and third research question, a more in-dept research method is required. Quantitative information is usable for a graphical and numerical visualization of the techniques used. However, it is difficult to quantitatively analyse the motivation for the use of these techniques and the behavioural influences on the decision making process. Since this research is exploratory, a more qualitative research method is required. In-dept interviews are suitable for this qualitative approach.
A total of six decision makers in different companies were interviewed to analyse these aspects. A selection was made based on the background of the general manager to investigate the different perspectives of these decision makers. Two general managers were interviewed with a technical background and two with a financial background. One interview was held with a general manager who had a technical as well as a financial background. A general manager with a technical background might have other motives for choosing a capital budgeting technique than a manager with a financial background. As discussed in the previous chapter, financial skills possibly influence the design and sophistication of the capital budgeting system.
For somewhat larger companies the financial techniques are used by financial staff to calculate the financial consequences of investment opportunities. The decisions to invest and the design of the capital budgeting system, however, are often made by the management. Therefore a distinction is made between general managers and financial staff. Based on this distinction, one interview was held with a division controller. Although more interviews with financial staff would be useful, only one person was willing to cooperate.
The interviewed companies differ in size and industry segments. The activities range from the production of machines, the modification of metal parts, the construction of steel frameworks, to the resale of robots. These different activities call for different investments and are therefore somewhat difficult to compare. The number of employees ranged from 14 to only 54 and are thus not representative for all SMEs. Larger firms were not found cooperative for this thesis.
The firms were contacted by telephone. The interviews were face-to-face and semi-structured as they followed a standard set of questions, but proceeded less structured. The questionnaire can be found in appendix 6.
4 Survey results
This chapter analyzes the results from the survey. Five main issues will be discussed: some firm and manager characteristics, the usage of capital budgeting techniques, the usage of real options and game theory, the impact of behavioural aspects on decision making, and the motivation for choosing a capital budgeting technique. The first five sections of this chapter will present and discuss descriptive statistics from the survey. The sixth section of this chapter analyzes whether the usage of the different techniques is related to the characteristics of the firm and the general and finance manager. Finally, section 4.7 summarizes the main findings from the survey.
4.1 Firm and manager characteristics
Respondents were asked to describe some characteristics of the firm and the general and financial managers. This gives some information regarding the respondents of the survey. The answers can be used to investigate which organizational and behavioural characteristics of managers, as discussed in chapter 2.5, have a significant influence on the capital budgeting techniques used. The respondents were asked to specify the age, education and background of the general manager and financial manager, and the percentage of ownership of the general manager. Figure 3 shows the summary statistics of the firms and managers in the sample.
>59 50-59 40-49 <40 50,0% 40,0% 30,0% 20,0% 10,0% 0,0%
A: CEO age (years)
>59 50-59 40-49 <40 50,0% 40,0% 30,0% 20,0% 10,0% 0,0%
B: CFO age (years)
Univer-sity MBA HBO MBO LBO or lower 60,0% 50,0% 40,0% 30,0% 20,0% 10,0% 0,0% C: CEO education Univer-sity MBA HBO MBO LBO or lower 60,0% 50,0% 40,0% 30,0% 20,0% 10,0% 0,0% D: CFO education
Other Economic / Financial Technical 80,0% 60,0% 40,0% 20,0% 0,0% E: CEO background Other Economic / Financial Technical 80,0% 60,0% 40,0% 20,0% 0,0% F: CFO background 100% 51-99% 50% 1-49% 0% 40,0% 30,0% 20,0% 10,0% 0,0%
G: CEO stock ownership
Figure 3. (Continued).
Figures 3A and 3B show that nearly half of all managers are between 40 and 49 years old. The general managers are on average somewhat older than the financial managers. Figures 3C and 3D show the educational level of the general and financial manager. The results show that a large majority of the managers has a HBO (higher professional training)10 educational level. Only
23% of the general managers and 19% of the financial managers have an MBA or university degree as their highest educational level. Figures 3E and 3F show that a large majority of the general managers have a technical background and that a large majority of the financial managers have an economical or financial background. Finally, figure 3G shows the stock ownership of the general manager. 35% of the general managers appear to be full owner of their firm and another 35% are to some extent owner of their firm.
4.2 Capital budgeting techniques
The respondents were asked to identify the frequency with which they used the specified capital budgeting techniques. The capital budgeting techniques were limited to the four most commonly used: payback, ARR, IRR and NPV. Additionally there was some room for other techniques. The respondents were allowed to choose from a 5-point Likert scale where the number 1 was associated with always and 5 with never. Table 3 shows the average scores on these techniques. A number closer to 1 shows that the technique is used frequently, a number closer to 5 shows that the technique is little used. Table 3 shows that the payback technique is the technique used most often, followed by IRR.
Table 3. Mean frequency score of capital budgeting techniques. Technique N Mean Payback 222 1,92 ARR 220 2,81 IRR 217 2,49 NPV 215 2,92
To examine which techniques are used always or often (a score of 1 or 2), figure 3 is constructed. Figure 3 shows that payback is used always or often by 76,53% of all respondents, followed by the IRR which is used always or often by 59,62%.
payback IRR ARR NPV 80% 60% 40% 20% 0%
Figure 4. Percentage of respondents who always or often uses the technique
43,66% 59,62%
46,95%
76,53%
These results indicate that Dutch SMEs use more often the payback and ARR technique, and less often the NPV and IRR technique than the respondents in recent international studies (e.g. Graham and Harvey 2001), but compare better with the respondents in studies from the eighties (see Sangster, 1993, pp. 315). It is not surprising that the results differ from recent results since chapter 2 already illustrated that smaller companies are expected to make more use of less sophisticated techniques than larger companies. The trend towards more sophisticated techniques may be lagging for small firms.
Among the other techniques that were mentioned by the respondents, emotion and intuition (8 respondents), and urgency (5 respondents) were named most. These are not capital budgeting techniques, but are rather considerations for the capital budgeting process.
When taking a closer look at the combination of techniques, table 4 illustrates that the combination of all four different capital budgeting techniques is used by 22,4% of the respondents. The combination of payback, ARR and IRR, and only the payback technique are popular as well. The results further show that a total of 151 (67,7%) respondents often or always use any sophisticated technique, and 178 (79,8%) respondents often or always use any naïve technique. These results suggest that although the naïve techniques are more popular, a majority of the respondents does use sophisticated techniques. A total of 22 (9,9%) respondents indicated that they used none of the capital budgeting techniques often or always.
Table 4. Combination of techniques used always or often.
Frequency Percentage
Payback + ARR + IRR + NPV 50 22,42% Payback + ARR + IRR 26 11,66% Payback + ARR + NPV 5 2,24% Payback + ARR 16 7,17% Payback + IRR + NPV 13 5,83% Payback + IRR 20 8,97% Payback + NPV 8 3,59% Payback 32 14,35% ARR + IRR + NPV 4 1,79% ARR + IRR 2 0,90% ARR + NPV 0 0,00% ARR 2 0,90% IRR + NPV 5 2,24% IRR 10 4,48% NPV 8 3,59% No technique 22 9,87% Total 223 100,00%
4.3 Game theory and real option usage
The respondents were asked whether they calculated the impact of competitors on the value of the project (game theory) and the impact of flexibility of managerial decision making during the investment (real option approach). Table 5 shows the average scores for these techniques, based on the same 5-point Likert scale as discussed in the previous section. Figure 5 shows the percentage of firms which use the technique always or often (a 1 or 2 score).
Table 5. Mean frequency score of game theory and real options.
N Mean Game theory 222 3,41 Real options 222 3,14 Real options Game theory 80% 60% 40% 20% 0%
Figure 5. Percentage of firms that always or often uses the technique
33,78% 28,38%
The results from figure 5 show that respectively 33,78% and 28,38% of the firms use real options and game theory always or often in their calculation. These results may seem rather high since these techniques are very advanced and require considerable financial skills and financial data of the project. It is therefore questionable whether all respondents fully understood the question. The results from this question should therefore be viewed with caution.
4.4 Behavioural aspects of decision making
To examine how often firms make decision based on their intuition or emotion, the respondents were required to indicate how often calculations of capital budgeting techniques were decisive in decision making. Economic theory suggests that decisions should be based fully on the outcome of calculations. The respondents were again allowed to choose from a 5-point Likert scale as discussed earlier. Table 6 shows the results for this question.
Table 6. The impact of calculations on decision making. Frequency Percentage 1 (always) 26 12,09% 2 79 36,74% 3 79 36,74% 4 21 9,77% 5 (never) 10 4,65% Total 215 100,00%
Table 5 shows that only 12 % of the respondents indicated that their calculations were always decisive. This suggests that for the other 88% of the respondents behavioural factors do influence their decision making. Behavioural influences thus appear to be important for the capital budgeting process. Additional research is required to get a clear view of the particular behavioural factors that influence decision making. In the interviews this behavioural decision making is one of the main subjects.
4.5 Motives for capital budgeting techniques
The respondents were asked to identify the main reason for choosing a particular capital budgeting technique. The alternatives were formulated in line with Tse and Griffioen (1993). The results are shown in table 7.
Table 7. The most important reason for choosing a particular capital budgeting technique.
Frequency Percentage
Agrees with the objective of the firm 120 54,79% Easy and fast to apply 65 29,68% Takes into account time value of money 18 8,22% Takes into account timing of cash flow 8 3,65% Easy to understand 8 3,65%
Total 219 100,00%
Table 7 shows that more than 54% of the respondents stated that the objective of the firm was the most important reason for choosing a capital budgeting technique. The main objective of the firm can be expected to be value maximization for the shareholders. These results indicate that
most managers of Dutch SMEs choose the technique of which they think that it will maximize their firm value. Nearly 30% of the respondents gave easy and fast to apply as an answer.
As a secondary reason, agrees with the objective of the firm appears to be the most popular again, followed by easy and fast to apply and takes into account timing of cash flow (see table 8). Easy to understand and takes into account time value of money appear to be less important considerations in choosing a capital budgeting technique.
Table 8. The second most important reason for choosing a particular capital budgeting technique.
Frequency Percentage
Agrees with the objective of the firm 62 29,52% Easy and fast to apply 52 24,76% Takes into account timing of cash flow 47 22,38% Takes into account time value of money 27 12,86% Easy to understand 22 10,48%
Total 223 100,00%
The popularity of the easy and fast to apply option may explain the popularity of the payback technique, but further analysis is required to identify whether the motive for choosing a particular capital budgeting technique is a determinant of the technique that is used. Section 4.6 analyzes several possible relations.
The results from this survey differ a lot from the results of Tse and Griffioen (1993), who show a major dominance for the time value of money as the most important reason. The results of Tse and Griffioen also differ from the results from this survey, in the sense that only 8% of the respondents from their study chose agrees with the stated objective as the most important reason for choosing a particular capital budgeting technique. Further analysis is required to identify some of the main differences between the results in this thesis and of from Tse and Griffioen (1993).
4.6 Analyses of the relationships
In this section the results from the survey will be analyzed in more detail. First, the impact of firm size on the capital budgeting techniques used is analyzed. Firms were separated into two groups: small firms with 10 till 49 employees and medium sized firms with 50-250 employees.
