University of Leicester

Centre for Management Studies

MBA (Finance) October 2008

Capital Budgeting Practices and Economic Development: A Comparative Study of Companies in Europe and West Africa
By

George Ekegey Ekeha

Email: ekegey24ge@yahoo.co.uk

March 2007

THIS DISSERTATION IS PRESENTED TO THE CENTRE FOR MANAGEMENT STUDIES, UNIVERSITY OF LEICESTER, UNITED KINGDOM. AND IT IS IN PART FULFILMENTS OF THE COMPLETION OF STUDIES TOWARDS THE AWARDS OF MASTERS OF BUSINESS ADMINISTRATION DEGREE (FINANCE OPTION). NO PART OF THIS THESIS IS TO BE USED FOR ANY PURPOSES, OTHER THAN ACADEMIC, WITHOUT THE OFFICIAL CONSULTATION WITH THE AUTHOR AND/OR THE UNIVERSITY AUTHORITIES.
GEORGE E EKEHA

MBA OCT. 2007

TABLE OF CONTENT LIST OF FIGURES AND TABLES................................................................................ 4 ABSTRACT....................................................................................................................... 5 PREFACE.......................................................................................................................... 7 1.0 INTRODUCTION....................................................................................................... 8 1.1 MOTIVATION OF THE STUDY ...................................................................................... 9 1.2 THE DEBT SERVICING CYCLE OF LESS DEVELOPED COUNTRIES .............................. 10 1.3 THE PROBLEMS AND RESEARCH HYPOTHESIS .......................................................... 11 1.4 ORGANISATION OF THE STUDY ................................................................................. 12 2.0 LITERATURE REVIEW ........................................................................................ 13 2.1 ECONOMIC DEVELOPMENT IN AFRICA ..................................................................... 13 2.2 THE CAPITAL BUDGETING DECISION ....................................................................... 15 2.3 STUDIES ON CAPITAL BUDGETING PRACTICES IN DEVELOPING COUNTRIES ............ 17 3.0 CAPITAL BUDGETING PROCESS AND PROJECT CLASSIFICATIONS... 18 3.1 CLASSIFICATION OF INVESTMENT PROJECTS ............................................................ 18 3.1.1Independent Projects ........................................................................................ 18 3.1.2 Mutually Exclusive Projects ............................................................................ 18 3.1.3 Contingent Projects ......................................................................................... 18 3.2 THE CAPITAL BUDGETING PROCESS .......................................................................... 19 3.2.1 Strategic planning............................................................................................ 20 3.2.2 Identification of investment opportunities ....................................................... 21 3.2.3 Preliminary screening of projects.................................................................... 21 3.2.4 Financial appraisal of projects........................................................................ 22 3.2.5 Qualitative factors in project evaluation ......................................................... 22 3.2.6 The accept/reject decision................................................................................ 23 3.2.7 Project implementation and monitoring .......................................................... 23 3.2.8 Post-implementation audit ............................................................................... 24

GEORGE E EKEHA

MBA OCT. 2007

4.0 DETERMINANTS OF CAPITAL BUDGETING PRACTICES......................... 24 5.0 SURVEY DESIGN AND METHODOLOGY ........................................................ 26 6.0 RESEARCH RESULTS AND ANALYSIS ............................................................ 27 6.1 COMPANY AND CFO CHARACTERISTICS .................................................................. 28 6.2 CAPITAL BUDGETING TECHNIQUES .......................................................................... 30 6.2.1 European CFOs ............................................................................................... 32 6.2.2 West African CFOs .......................................................................................... 33 6.2.3 European versus West African CFOs .............................................................. 34 6.3 COST OF CAPITAL ESTIMATION METHODS ............................................................... 35 6.3.1 European CFOs ............................................................................................... 37 6.3.2 West African CFOs .......................................................................................... 37 6.3.3 European versus West African CFOs .............................................................. 37 6.4 COST OF EQUITY ESTIMATION METHODS ................................................................. 38 6.4.1 European CFOs ............................................................................................... 40 6.4.2 West African CFOs .......................................................................................... 40 6.4.3 European versus West African CFOs .............................................................. 41 6.5 CAPITAL BUDGETING TECHNIQUES, COST OF CAPITAL AND COST OF EQUITY ESTIMATIONS: MULTIVARIATE ANALYSIS ..................................................................... 41 6.5.1 The Multivariate Analysis ................................................................................ 42 6.5.2 Capital Budgeting Techniques......................................................................... 44 6.5.3 Cost of Capital Estimation............................................................................... 46 6.5.4 Cost of Equity Estimation ................................................................................ 47 7.0 SUMMARY AND DISCUSSION ............................................................................ 49 LIST OF REFERENCES ............................................................................................... 51 RESEARCH QUESTIONNAIRES ............................................................................... 54

GEORGE E EKEHA

MBA OCT. 2007

LIST OF FIGURES AND TABLES FIGURE 1: THE CAPITAL BUDGETING PROCESS ............................................. 20 TABLE 1: COMPANY CHARACTERISTICS........................................................... 29 TABLE 2: CAPITAL BUDGETING METHODS USED BY CFOS......................... 31 TABLE 3: MOST FREQUENTLY USED METHODS TO MEASURE THE COST OF CAPITAL (% OF TOTAL)..................................................................................... 36 TABLE 4: MOST FREQUENTLY USED METHODS TO ESTIMATE THE COST OF EQUITY (% OF TOTAL) ....................................................................................... 39 TABLE 5: DETERMINANTS OF CAPITAL BUDGETING METHODS: MULTIVARIATE LOGIT ANALYSIS ....................................................................... 45 TABLE 6: DETERMINANTS OF THE MOST FREQUENTLY USED METHODS TO MEASURE THE COST OF CAPITAL: MULTIVARIATE LOGIT ANALYSIS ........................................................................................................................................... 47 TABLE 7: DETERMINANTS OF COST OF EQUITY ESTIMATION METHODS: MULTIVARIATE LOGIT ANALYSIS ....................................................................... 48

GEORGE E EKEHA

MBA OCT. 2007

ABSTRACT Over the years, efforts have been made to increase the developmental strides of African countries. Many projects move from donor countries like the United Kingdom and the United State into Africa to help improve the lives of the people. However, these efforts have not been able to redeem Africa from abject poverty and indebtedness to the West. Various projects that are targeted towards the reduction of poverty are normally completed with no changes in the lives of the people. These projects, in my opinion, have not been scrutinised to assess their capabilities of meeting some stated target.

Capital budgeting practices are some of the vital inputs in the decision-making process of embarking on investment projects. A very good analysis, scrutiny, implementation and monitoring of such projects could yield the expected results for the stakeholders (people of the country). According to Dayananda et al (2002), the capital budgeting practices are used to make investment decisions so as to increase shareholders value. Capital budgeting is primarily concerned with sizable investments in long-term assets, Brealey & Myers (2003). These assets may be tangible items such as property, plant or equipment or intangible ones such as new technology, patents or trademarks. Investments in processes such as research, design, development and testing through which new technology and new products are created may also be viewed as investments in intangible assets (ibid).

Dayananda et al (2002), argued that irrespective of whether the investments are in tangible or intangible assets, a capital investment project can be distinguished from recurrent expenditures by two features. One is that such projects are significantly large. The other is that they are generally long-lived projects with their benefits or cash flows spreading over many years. Sizable, long-term investments in tangible or intangible assets have long-term consequences (ibid). This implies that todays investment will determine the overall corporate strategic position over many years. These capital investments also have a considerable impact on the future cash flows of the organization and the risk associated with those cash flows. Capital budgeting decisions thus have a long-range impact on the strategic performance of the organization and are also critical to its success or failure.

GEORGE E EKEHA

MBA OCT. 2007

This paper compares the use of capital budgeting techniques by companies in Europe and West Africa, using data obtained from a survey between 225 European and 120 West African companies. The main aim is to analyse the use of capital budgeting techniques by companies in both economic blocs from a comparative perspective to see whether economic development matters in the choice of which technique to use.

The empirical analysis provides evidence that European CFOs on average use more sophisticated capital budgeting techniques than their counterparts in West African. At the same time, however, the results suggest that the differences between European and West African companies is smaller than might have been expected based upon the differences in the level of economic development between both economic blocs. At least, this is evident with respect to the use of methods of estimating the cost of capital and the use of CAPM as the method of estimating the cost of equity.

GEORGE E EKEHA

MBA OCT. 2007

PREFACE The work presented in this thesis was carried out over a period between late 2006 and early 2007 for the award of MBA (Finance) degree with University of Leicester. During this time and all my studies period I had help from several different people that I would like to thank.

First, I would like to thank my lecturer Jeremy French, administrators Hamza and Nikos at Citi Banking College and Peter Alfano at Centre for Management Studies, University of Leicester for their support and helpful advice during my period of studies. Secondly, I would like to thank all the responding CFOs for taking the time to participate in this study. I would also like to thank my family, my wife Mrs Beauty Ekeha, my mum Agnes Obri, who was looking after my kids during the period of my studies and all my kids, Norris Walter, Bright Mawusi, Urielle Jorgbenue and Suzzy Selase for their support during my studies. Finally, I would like to thank my Administrative Director, Mr. Samuel Boakye and all members of the Finance Department at Ghana Statistical Services, Ministry of Finance.

Throughout the period of my studies in the United Kingdom and work with this thesis, I have gained a lot of knowledge, experience, and insight of good business management in the area of capital budgeting techniques, and this would also help me contribute to the future research within this area and other managerial processes both in public and private sectors.

March 2007 GEORGE EKEGEY EKEHA

GEORGE E EKEHA

MBA OCT. 2007

1.0 INTRODUCTION This paper reports the results of a survey with respect to the current practices of capital budgeting techniques in two different economic blocs at two different levels of economic development: Europe and West Africa. The main aim of this paper is to analyse the use of capital budgeting techniques by companies in a comparative perspective to see whether economic development matters in the choice of techniques. Whereas several papers in the past have investigated the use of such techniques, this is one of the very few studies that use such a comparative perspective, comparing a more developed with a developing economy. This analysis was carried out using standard differences of mean tests and multivariate regression analysis to see whether there is a so-called country effect on the choice of capital budgeting technique. This means that the research tried to establish whether capital budgeting practices differ significantly between companies in the two economic blocs and whether these differences can be explained by differences in levels of economic development.

Again, only very few papers have addressed the determinants of capital budgeting practices using these types of analyses, let alone in a comparative economic perspective. Notable exceptions, among others, are Brounen, et al. (2004) and Payne, et al. (1999). Yet, both studies analyse the determinants of capital budgeting practices for a number of developed countries (The Netherlands, Germany, France, Canada, the U.S. and the U.K.). West Africa and Europe have been chosen for this comparison for the following reasons. The researcher was a Finance Manager in a government department of one West African country and considers West African countries as strongly emerging, yet still lessdeveloped economy in many respects, which has received a lot of attention in the economic and financial development literature during recent years. Moreover, the researcher also considers Europe as a typical example of a developed economic bloc and also most companies in this bloc have various investment interests in Africa. Finally, the researcher believes that most CFOs in African countries do not utilise the sophisticated capital budgeting techniques to scrutinise projects very well before selection. This resulted in various mismanagement and failure to achieve economic heights.

GEORGE E EKEHA

MBA OCT. 2007

1.1 Motivation of the Study Capital budgeting involves making investment decisions concerning the financing of capital projects by organisations. Making a good investment decision is important since funds are scarce and the investment is expected to add to the value of the organisation especially in Less Developed Countries (LDCs) and Third World poor nations. Capital investment decision is thus one of the requirements, if properly applied, that can help accelerate economic development. All countries of the sub-Saharan Africa expend an upward of 13.5 billion dollars per annum on foreign debt payment to rich foreign creditors, World Bank report (2005). Many countries in the third world borrowed huge sums of money in expectation that interest rates would remain stable. Many African countries accepted these loans for political and economic stabilization in the post independence era, however prominent problems such as corruption make these loans ineffective to save the recipients countries from their economic woes.

For example in Ghana, a governance and corruption survey was commissioned by the World
Bank, which was conducted by the Centre for Democratic Development (CDD GHANA). Evidence from the survey showed that public concern about corruption in the country is very high and that there is a widespread public perception that corruption has had a negative toll on productivity and efficiency of both the public and private sectors and consequent effects on popular welfare, CDD Ghana (2000). The Ghana Integrity Initiative (GII), a local chapter of Transparency International, has also on various occasions undertakes some educational programs on corruption and good governance through seminars and workshops for various interest groups in the country. One recent study on administrative costs faced by private investors in 32 developing countries most from Africa reported that it takes up to two or three years to establish a new business in many developing countries (Morisset and Lumenga Neso: 2002). Their study found that the most delays occurred in securing land access and obtaining building permits. The associated administrative costs were found to be positively correlated with estimates of the level of corruption and negatively correlated with the quality of corporate governance, degree of openness, and public wages, among others (Morisset and Lumenga Neso: 2002). The authors finally argued that the level of corruption or the lack of good governance is expected to influence administrative costs as bureaucrats and politicians are more likely to capture the extra rents (ibid). In fact, the corrupt practices of most executives in both public and the private sectors of these developing West African countries have led to increases in debts to their borrower countries with

GEORGE E EKEHA

MBA OCT. 2007

the intended targets of the loans not met. On the side of the creditors as well, many of these

loans were given in order to gain and or retain the loyalty of those corrupt regimes, which is the characteristic of African governments.

1.2 The Debt Servicing Cycle of Less Developed Countries These debt-trapped nations were under-developed and their debt crisis further plunge them into deeper economic crisis and abject poverty due to excessive borrowing. Most executives of these developing countries have the selfish tendency of mismanaging the various project assigned to them. Some managers of the projects are eager to satisfy their personal needs before thinking of the implementation of whatever projects has been assigned to them. This leads to poor budgeting, poor monitoring and hence poor implementation of the project. Governments of the nations have to then borrow more funds in order to complete and maintain the existing projects. Due to the fact that these loans were thoughtlessly accepted, and collected by most African governments, they had neither little implications for development nor benefit for the masses.

Finally the unreliable market prices in the worlds market for agricultural products and low-technologically manufactured goods, which make it particularly difficult for African countries to diversify and increase exports to hard currency markets. Thus making it difficult for them to earn their way out of the debt trap. In my opinion, the developed countries, like the USA and UK who have been prophesising their lengthy plans to alleviate Africa from its economic woes must endeavour to ensure some monitoring system such that the aids will go a long way to improve the investment capacity of the continent. International markets should also be opened to the African manufacturers in the said developed countries. Finally, loans must be channelled towards the transformation of the primary products into products worthy for the international market.

Notwithstanding, however, the researcher believes that these debt-ridden nations in the Sub-Saharan Africa are expected to make attempts at improving their economic status themselves through huge research and development leading into economic productivity. The concerns of the developed nations may be to no avail if these less developed nations

GEORGE E EKEHA

10

MBA OCT. 2007

do not take steps that will help relief their situation. The capital investment decision is thus one of the most critical and crucial decisions that any country or organisation can take to achieve economic development-thus by adding economic value. Since economic development depends on the multiplicity of viable corporate organisations and enterprises in the country, the approach adopted here is to demonstrate how capital budgeting, as an investment decision can help African countries promote corporate organisational growth by using acceptable techniques to identify viable projects. In other words, capital budgeting is an integral part of the corporate plan of an organisation, which reflects the basic objectives of an organization. The capital investment decision involves large sums of money and may introduce a drastic change in companies as well as the whole economy, when it is well scrutinised. For instance, acceptance of a project may significantly change a companys operation, profitability and create more jobs within the country. These changes might also affect investors evaluation of a company (Osaze, 1996:40-44).

1.3 The Problems and Research Hypothesis Most third world countries depend excessively on importation. They do not develop an enduring technological base that can support the growth of their economies. Their capital investment decisions are not usually well articulated. This may be due to the fact that their governments do embark on white elephant projects that gulp huge sums of money and are useless in terms of utility to the people. The projects often are abandoned halfway and in some cases, are only executed on papers. The current efforts of some African governments like those of Ghana and Nigeria, towards privatisation of hitherto government-owned firms and corporations is an indirect concession to the fact that the former investment decision pattern of the national government is not wise enough to alleviate their countries from poverty. In fact, most of the diversified companies have improved productivity and quality with enormous benefits to their countries. Considering the matter from the corporate perspective therefore, the researcher believes that capital budgeting decision is one of the decision-making areas of a financial manager that involves the commitment of large funds in long-term projects or activities. And these projects have a huge impact on the countys economic development.

GEORGE E EKEHA

11

MBA OCT. 2007

This study therefore seeks to examine the importance of capital investment decisions; the basic steps in making capital investment decisions and the techniques used in evaluating capital investment projects so that the overall countrys economy can grow from the corporate sector investments. It is also expected to show that the use of sophisticated techniques by both corporate and governmental CFOs will help in the development efforts of Africans and other poor nations. The researcher believes that most developed countries in Europe have achieved highs in todays competitive international market because they put money where it adds value. Investments are well scrutinised using various sophisticated techniques, both qualitative and quantitative, before final decision is arrived and such projects are well monitored until fully completed. It is my believe also that most African Countries remain in the low economic growth and poverty zone because CFOs dont make use of technical tools to analyse various investment projects, which have significant impact on the economic development. These differences might be due to the level of education, technology and economic development between the two economic blocs. Therefore, the researcher hypothesizes that CFOs of European companies will use net present value (NPV) and internal rate of returns (IRR) methods more often than their counterparts in West Africa, whereas the opposite will be true for the pay back (PB) and accounting rate of returns (ARR) methods.

An additional contribution of this paper to the existing empirical literature on capital budgeting practices is in terms of the countries for which the researcher had gathered data. Most previous studies focus on the United States and the United Kingdom and there are some few studies available for the Netherlands (Herst, Poirters and Spekreijse, 1997; Brounen, De Jong and Koedijk, 2004). The researcher is also aware of study on Capital Budgeting and Economic Development in the Third World Countries (Elumilade, et al 2006) but there were no comparisons with any developed economy.

1.4 Organisation of the study The paper is organised in seven different sections, with section one dealing with introduction, hypothesis and motivations of the study. Section two discusses literature

GEORGE E EKEHA

12

MBA OCT. 2007

review on capital budgeting practices and further discussed the capital budgeting process, classification of investment projects and alternative determinants of capital budgeting practices in sections three and four. This was followed by a discussion of the design of the survey in section five. Section six then provides the results of the survey and a discussion of the empirical analysis of determinants of capital budgeting practices. The paper ends with a summary and discussion of the results in the final section.

2.0 LITERATURE REVIEW In this section the researcher tries to outline previous studies relevant to this study. The section discussed various studies on economic development, capital budgeting among CFOs from various countries and if there is any comparative study between developed and developing countries.

2.1 Economic Development in Africa Economic development in Africa has not been steady. In fact, when compared to the situation in the Western countries like Europe, the conclusion is that countries of the third world are either qualified as undeveloped or mildly put underdeveloped. African scholars have tended to heap the blame on the Europeans; saying that colonialism or neocolonialism is the bane of Africas economic woes. This notion is referred to by Onigbinde (2003:21-25), as the Original Sin Fallacy. The present economic woe of underdeveloped countries (UDCs) according to this fallacy is that UDCs condition is original in relation to a so-called non-achievement, the present condition of the underdeveloped world is a historical product of capitalist expansion (ibid). The crisis of underdevelopment in Africa is also captured in the Africa at the Doorstep of Twenty-First Century by Adebayo Adedeji as sited by Onigbinde, 2003. According to him, African within the world is, poverty increased in both the rural and urban areas: real earning fell drastically; unemployment and underemployment rose sharply; hunger and famine became endemic; dependence on food aid and food imports intensified; disease, including the added scourge of AIDS, decimated population and became a real threat to the very process of growth development; and the attendant social evils-rime delinquency, there is a mess vengeance (Onigbinde, 2003: 78-79) .

GEORGE E EKEHA

13

MBA OCT. 2007

The United States Assistance for International Development (USAID), 1988-1992 (cited from Onigbinde, 2003:79-80), stated among other things that, approximately 180 million of sub-Saharan Africas 500 million people could be classified as poor, of whom 66.7 percent, or 120 million, are desperately poor. By every international measure, be it per capital income ($330), life expectancy (51 years), or the United Nations Index of Human Development (0.255 compared to 0.317 for South Asia, the next poorest region), Africa is the poorest region in the world, Onigbinde 2003. The solution to all these problems lies in the fact that firms are to embark on projects that would give rise to companys value which will by extension enhancing the desired economic development for the country. In the course of achieving these development efforts, the companys activities become more complex and corporate management assumes a sound financial position in the handling of problems and decisions therein.

In his study of The obstacles to investment in Africa, Professor Peter Montiel of the World Bank concluded among other things that One set of explanations is based on the view that investment projects with high economic rates of return are not as plentiful in Africa as the simple neoclassical growth paradigm would seem to imply. One argument is that for a variety of reasons, aggregate production functions may be characterized by lower levels of productivity in Africa than in creditor countries. An alternative or complementary story is based on generalizing the aggregate production function to include roles for human capital, public capital, and institutional capital Montiel (2006). He continued to say These effects raise questions about the abundance of investment opportunities yielding high economic rates of returns in Africa at the present time, (ibid). This conclusion suggests that, though not abundant, investment opportunities with high returns exist in African countries and when applied properly, it could bring economic growth to Africa. One of the best ways to scrutinise these opportunities is by using various techniques like the capital budgeting techniques, to access the profit potentials.

GEORGE E EKEHA

14

MBA OCT. 2007

2.2 The Capital Budgeting Decision Capital budgeting decisions are among the most important decisions the financial manager of a company has to deal with. Capital budgeting refers to the process of determining which investment projects result in maximisation of shareholder value, Dayananda et al, 2002. Generally speaking, there are four main capital budgeting techniques the manager may use when evaluating an investment project. In fact, there are other techniques that could have been considered, such as sensitivity analysis, real options, book rate of return, simulation analysis, etc. (Graham and Harvey, 2001, pp.196197). However, the researcher has chosen to focus on the most well known techniques to keep the study simple. The net present value (NPV) and internal rate of return (IRR) methods are considered to be discounted cash flow (DCF) methods. The payback period (PB) and average accounting rate of return (ARR) methods are so-called non-DCF methods, Brealey and Myers, 2003. From a pure theoretical point of view the NPV is considered to be the most accurate technique to evaluate projects. Yet, it is also the most sophisticated of the four, followed by the IRR method. Both non-DCF methods are considered to be less accurate, of which the PB method is the least sophisticated (ibid).

In the past, several studies of capital budgeting practices have been carried out. Most studies focus on companies in the U.S. Comparing survey results of capital budgeting practices in the U.S. over time generally seems to show that the analytical techniques used by executives have increased in terms of sophistication. For example, in one of the earliest studies reporting the results of questionnaires on capital budgeting practices, Klammer (1972) shows that in 1959, based on a sample of 184 large U.S. companies, 19 per cent indicated that they used DCF methods as their primary method to evaluate projects. The majority of companies used either PB (34 per cent of the total sample) or ARR methods (34 per cent) as their primary method of evaluation. In 1970, the picture had changed drastically: DCF methods were used by 57 per cent of the companies; 26 per cent used ARR and only 12 per cent used PB as their primary method of project evaluation (ibid). In a later study, Hendricks (1983) reports that in 1981 76 per cent of the companies in his sample studied used DCF methods as their primary tool. Only 11 per cent stated they used the PB method as their primary tool. Trahan and Gitman (1995)

GEORGE E EKEHA

15

MBA OCT. 2007

show that, based on a 1992 survey of 58 of the Fortune 500 large companies and 26 of the Forbes 200 best small companies, most companies used DCF methods as their primary evaluation tool, although these methods were more important for the large companies (88 per cent for NPV and 91 per cent for IRR) than for the small companies (65 and 54 per cent for NPV and IRR respectively).

A recent study by Graham and Harvey (2001), a comprehensive survey published on capital budgeting practices (using answers from a 1999 survey among 392 Chief Financial Officers (CFOs) of companies in the U.S. and Canada) showed that the NPV and IRR techniques are the most frequently used capital budgeting techniques. Their survey reported that 75 per cent of the CFOs always use NPV and 76 per cent always or almost always use the IRR method. Their survey results also show, however, that even though over time the use of the PB method has declined as a primary tool for project evaluation, it remains to be an important secondary instrument CFOs normally use. According to Hendricks (1983), in his 1981 survey 65 per cent of the companies in his sample used PB as a secondary measure. Trahan and Gitman (1995) show that in 1992, 72 per cent of the large and 54 per cent of the small companies used PB as one of the evaluation tools. In the 1999 survey of Graham and Harvey (2001) 57 per cent indicated they use the PB method as one of their evaluation tools.

The general picture that emerges from the previous short discussion also emerges from survey studies based on other U.S. as well as U.K., European and Australian companies (Gitman and Forrester (1977); Schall, et al. (1978); Kim and Farragher (1981); Shao and Shao (1996); Pike (1996) and Brounen, et al. (2004) ; Freeman and Hobbes (1991) and Truong, et al. (2005); Herst, et al. (1997) and Brounen, et al. (2004). A comparison of the results of these survey studies also showed an increasing sophistication with respect to the use of evaluation techniques over time. At the same time, however, it seems that companies in European countries report lower rates of the use of DCF techniques as compared to U.S. companies.

GEORGE E EKEHA

16

MBA OCT. 2007

Brounen et al (2004) replicate the Graham and Harvey (2001) survey in four European countries (U.K., France, Germany and the Netherlands; total sample was 313 companies) in 2002-2003 and find that for the U.K. companies in their sample 47 per cent states that NPV is (almost) always used as a tool of evaluating projects, whereas 69 per cent (almost) always use the PB. For the Netherlands these figures are comparable (70 and 65 per cent, respectively); for France and Germany the figures are even lower (42-50 per cent and 44-51 per cent, respectively).

2.3 Studies on Capital Budgeting Practices in Developing Countries A few studies have reported survey evidence on capital budgeting practices in the AsiaPacific region. These studies show a somewhat different picture. Wong, et al (1987) used information from a survey among a large number of companies in Malaysia, Hong Kong, and Singapore in 1985 and found that in these countries the PB method was the most popular primary measure for evaluating and ranking projects. For Malaysia this picture was confirmed in Han (1986). In a recent paper by Kester, et al. (1999), based on information from surveys of 226 companies in Australia, Hong Kong, Indonesia, Malaysia, The Philippines and Singapore in 1996- 1997, it was reported that the PB method was still an important method. Yet, DCF methods seem to have increased in importance as well. Excluding Australia from the sample of the countries studied, 95 per cent of the companies in the five Asian countries indicated that they use the PB method and 88 per cent of them said they use the NPV method when evaluating projects. In terms of importance (on a scale from 1 to 5, where 1 = unimportant and 5 = very important) both methods are rated almost equally important (3.5 versus 3.4) (ibid). When comparing these results to the results of studies for companies in Western economies, these figures seem to be very high. Comparing the results of the study by Wong, et al. (1987) with those of Kester, et al. (1999) does seem to suggest that the level of sophistication of capital budgeting techniques has increased quite rapidly during a period of just one decade within the developing countries in Asia.

GEORGE E EKEHA

17

MBA OCT. 2007

3.0 CAPITAL BUDGETING PROCESS AND PROJECT CLASSIFICATIONS 3.1 Classification of Investment Projects Investment projects can be classified into three categories on the basis of how they influence the investment decision process: independent projects, mutually exclusive projects and contingent projects, Dayananda et al (2002).

3.1.1Independent Projects An independent project is the one which the acceptance or rejection of does not directly eliminate other projects from consideration or affect the likelihood of their selection. For example, management may want to introduce a new product line and at the same time may want to replace a machine, which is currently producing a different product. These two projects can be considered independently of each other if there are sufficient resources to adopt both, provided they meet the firms investment criteria (ibid). This implies that the projects can be evaluated independently and a decision made to accept or reject them depending upon whether they add value to the firm.

3.1.2 Mutually Exclusive Projects According to Dayananda et al (2002), two or more projects that cannot be pursued simultaneously are called mutually exclusive projects the acceptance of one prevents the acceptance of the alternative proposal. Therefore, mutually exclusive projects involve either-or decisions alternative proposals cannot be pursued simultaneously. The early identification of mutually exclusive alternatives is crucial for a logical screening of investments. Otherwise, a lot of hard work and resources can be wasted if two divisions independently investigate, develop and initiate projects, which are later recognized to be mutually exclusive (ibid).

3.1.3 Contingent Projects Finally, a contingent project is the one which the acceptance or rejection is dependent on the decision to accept or reject one or more other projects. Contingent projects may be complementary or substitutes (ibid). For example, the decision to start an agricultural

GEORGE E EKEHA

18

MBA OCT. 2007

project in a West African village may be contingent upon a decision to build roads leading to the project sites. In this case the projects are complementary to each other. The cash flows of the farming project will be enhanced by the existence of good roads to transport inputs and outputs to and from the farm and conversely the cash flows necessary for the road maintenance will be enhanced by the existence of high road taxes paid by the trucks using the road. In contrast, substitute projects are ones where the degree of success (or even the success or failure) of one project is increased by the decision to reject the other project. For example, market research indicates demand sufficient to justify two restaurants in a shopping complex and the firm is considering one Chinese and one Thai restaurant. Customers visiting this shopping complex seem to treat Chinese and Thai food as close substitutes and have a slight preference for Thai food over Chinese (ibid). Consequently, if the firm establishes both restaurants, the Chinese restaurants cash flows are likely to be adversely affected. This may result in negative net present value for the Chinese restaurant. In this situation, the success of the Chinese restaurant project will depend on the decision to reject the Thai restaurant proposal. Since they are close substitutes, the rejection of one will definitely boost the cash flows of the other. Contingent projects should be analysed by taking into account the cash flow interactions of all the projects (ibid).

3.2 The capital budgeting process This section was adopted from Dayananda et al (2002), they stated that there are several sequential stages in the process. For typical investment proposals of a large corporation, the distinctive stages in the capital budgeting process are depicted, in the figure 1 below:

GEORGE E EKEHA

19

MBA OCT. 2007

Figure 1: The Capital Budgeting Process Corporate Goal Strategic Planning

Investment Opportunities Preliminary Screening Financial appraisal, quantitative analysis, project evaluation/analysis Qualitative factors, judgement & gut feeling Accept/reject decision on the projects Accept Implementation Facilitation, monitoring, control & review Continue, expand or abandon project Post-implementation audit
Source: Capital Budgeting: Financial Appraisal of Investment Projects (Dayananda et al 2002)

Reject

3.2.1 Strategic planning A strategic plan is the grand design of the firm and clearly identifies the business the firm is in and where it intends to position itself in the future. Strategic planning translates the firms corporate goal into specific policies and directions, sets priorities, specifies the structural, strategic and tactical areas of business development, and guides the planning process in the pursuit of solid objectives, Daft (2003). A firms vision and mission is encapsulated in its strategic planning framework. There are feedback loops at different stages, and the feedback to strategic planning at the project evaluation and decision

GEORGE E EKEHA

20

MBA OCT. 2007

stages indicated by upward arrows in Figure 1 is critically important. This feedback may suggest changes to the future direction of the firm, which may in effect, cause changes to the firms strategic plan Dayananda et al (2002).

3.2.2 Identification of investment opportunities According to Dayananda et al (2002), the identification of investment opportunities and generation of investment project proposals is an important step in the capital budgeting process. They proposed that the projects have to fit in with a firms corporate goals, its vision, mission and long-term strategic plan. Of course, if an excellent investment opportunity presents itself the corporate vision and strategy may be changed to accommodate it. Thus, there is a two-way traffic between strategic planning and investment opportunities. Deyananda et al (2002) went on to argue that this is very tactical level of the capital budgeting process because, some investments are mandatory for instance, those investments required to satisfy particular regulatory, health and safety requirements and they are essential for the firm to remain in business. Other investments are discretionary and generated by growth opportunities, competition, cost reduction opportunities and so on (ibid). Some firms have research and development (R&D) divisions constantly searching for and researching into new products, services and processes and identifying attractive investment opportunities. Sometimes, excellent investment suggestions come through informal processes such as employee chats in a staff room or corridor (ibid).

3.2.3 Preliminary screening of projects The next stage after identifying various investment opportunities is to do initial screening. It is obvious that all the identified opportunities cannot go through the rigorous project analysis process. Therefore, the identified investment opportunities have to be subjected to a preliminary screening process by management to isolate the marginal and unsound proposals, because it is not worth spending resources to thoroughly evaluate such proposals. Dayananda et al (2002) suggested that the preliminary screening may involve some preliminary quantitative analysis and judgements based on intuitive feelings and experience.

GEORGE E EKEHA

21

MBA OCT. 2007

3.2.4 Financial appraisal of projects The next stage after the initial screening of identified projects is to go through rigorous financial appraisal to ascertain if they would add value to the firm. According to Dayananda et al (2002), this stage is also called quantitative analysis, economic and financial appraisal, project evaluation, or simply project analysis. This project analysis may predict the expected future cash flows of the project, analyse the risk associated with those cash flows, develop alternative cash flow forecasts, examine the sensitivity of the results to possible changes in the predicted cash flows, subject the cash flows to simulation and prepare alternative estimates of the projects net present value. Thus, the project analysis can involve the application of forecasting techniques, project evaluation techniques, risk analysis and mathematical programming techniques such as linear programming. The financial appraisal stage will provide the estimated contribution that the project would make towards the increase of the firms value in terms of the projects net present values. If the projects identified within the current strategic framework of the firm repeatedly produce negative NPVs in the analysis stage, these results send a message to the management to review its strategic plan (ibid). It is noteworthy therefore that the feedback from project analysis to strategic planning plays an important role in the overall capital budgeting process. The results of the quantitative project analyses will therefore influence the project selection or investment decisions.

3.2.5 Qualitative factors in project evaluation Dayananda et al (2002), continued that when a project passes through the quantitative analysis test, it has to be further evaluated taking into consideration some qualitative factors. Qualitative factors are those which will have an impact on the project, but are virtually impossible to be evaluated accurately in monetary terms. They suggested the following factors for consideration: the societal impact of an increase or decrease in employee numbers the environmental impact of the project possible positive or negative governmental political attitudes towards the project the strategic consequences of consumption of scarce raw materials

GEORGE E EKEHA

22

MBA OCT. 2007

positive or negative relationships with labour unions about the project possible legal difficulties with respect to the use of patents, copyrights and trade or brand names impact on the firms image if the project is socially questionable.

They argued that some of the items in the above list affect the value of the firm, and some not. The firm can address these issues during project analysis, by means of discussion and consultation with the various parties, but these processes will be lengthy, and their outcomes often unpredictable. This stage will require considerable management experience and judgemental skill together with high level of think-tack to incorporate the outcomes of these processes into the project analysis. In some cases, however, those qualitative factors which affect the project benefits may have such a negative bearing on the project that an otherwise viable project will have to be abandoned.

3.2.6 The accept/reject decision Having done the critical quantitative and qualitative analysis, the NPV results from the quantitative analysis combined with those qualitative factors will form the basis of the decision support information. The analyst relays this information to management with appropriate recommendations. Management considers this information and other relevant prior knowledge using their routine information sources, experience, expertise, gut feeling and, of course, judgement to make a major decision to accept or reject the proposed investment project (ibid).

3.2.7 Project implementation and monitoring Once investment projects have passed through the decision stage they must be implemented by management without any further delay. During this implementation phase various divisions of the firm like sales and marketing, production and technical are likely to be involved. An integral part of project implementation is the constant monitoring of project progress. This would enable management to identifying potential bottlenecks thus allowing early intervention. Deviations from the estimated cash flows

GEORGE E EKEHA

23

MBA OCT. 2007

need to be monitored on a regular basis so that corrective actions will be taken when needed.

3.2.8 Post-implementation audit Dayananda et al suggest that, post-implementation audit does not relate to the current decision support process of the project; it deals with a post-mortem of the performance of already implemented projects. They said that, An evaluation of the performance of past decisions, however, can contribute greatly to the improvement of current investment decision-making by analysing the past rights and wrongs. The post-implementation audit can provide useful feedback to project appraisal or strategy formulation. For example, ex post assessment of the strengths (or accuracies) and weaknesses (or inaccuracies) of cash flow forecasting of past projects can indicate the level of confidence (or otherwise) that can be attached to cash flow forecasting of current investment projects (ibid). This might also be important because if projects are undertaken within the framework of the firms current strategic plan and they do not prove to be as lucrative as predicted, the audit information can prompt management to consider a thorough review of the firms current strategic plan.

4.0 DETERMINANTS OF CAPITAL BUDGETING PRACTICES As was shown in the previous section, over time, financial managers have applied various methods and procedures to determine which investments are beneficial to the firm. The choice of the evaluation method may therefore be determined by individual preferences of the manager and/or by the environment in which decisions have to be made.

While in the literature several factors have been mentioned as determinants of the choice of capital budgeting practices, in this paper the researcher wants to focus on the role that is played by the level of economic development in this respect. The review of studies of capital budgeting practices in the previous section showed that over time, the use of more sophisticated DCF methods has become more popular. This may be explained by various factors. First, financial markets have developed over time, making the use of DCF methods more applicable, convenient and necessary. Due to the development of financial

GEORGE E EKEHA

24

MBA OCT. 2007

markets (and especially stock markets) shareholders value maximization has gained high importance, which has pressured CFOs of companies to use DCF methods over other simpler and less accurate alternatives. Second, training of CFOs has improved over time, which may have enabled them to better understand and thus use more sophisticated techniques. Third, financial tools and programmes that help the CFO to determine which investments are beneficial to the firm have become increasingly sophisticated, which may also have stimulated the use of more sophisticated techniques. Finally, the increased use of computer technology and the related reduction in the cost of this technology may have stimulated the use of more sophisticated techniques.

This researcher believes that these factors are all related to increasing levels of development. More developed countries generally tend to have more sophisticated financial markets, Levine (1997) higher levels of human capital, Schultz (1988), Boozer et al (2003), and higher levels of technology, Evenson (1988). This would also mean that the level of economic development of a country and the sophistication of the capital budgeting techniques implemented by CFOs in that country are positively related. In general terms, therefore, it could be expected that CFOs of companies in more developed countries use DCF methods significantly more often than do their counterparts in less developed countries. The opposite may hold for the use of non-DCF methods. It is this hypothesis that will be investigated in this study, using information from Europe and West African CFOs with respect to their capital budgeting practices.

Although since the late 1980s some West African countries have seen some impressive economic growth, over the period, the researcher believes that at the beginning of the new millennium there was still a wide gap in levels of economic, human and technological development between West African countries and the developed countries such as Europe. Whiles the investment in high returns projects will facilitate the development efforts of these poor nations, the researcher also believes that the use of sophisticated capital budgeting techniques will help the West African CFOs to identify the most profitable projects and thereby helping their governments to achieve economic heights. The high level of economic and technological developments in the developed

GEORGE E EKEHA

25

MBA OCT. 2007

countries have facilitated the ability to make use of very sophisticated techniques, which are more likely to produce more reliable results. Therefore, the researcher hypothesizes that CFOs of European companies will use NPV and IRR methods more often than do West African CFOs, whereas the opposite will be true for the PB and ARR methods.

To test the hypothesis the researcher will also take into account other variables that according to the literature may also explain the use of capital budgeting practices, Brounen, et al. (2004) and Graham and Harvey (2001). These variables will be included in the multivariate analysis as control variables. In particular, the researcher included measures of the size of the firm, the industry to which the firm belongs, and the educational level and age of the CFO of the firm. Firm size is included because some papers have argued and indeed found evidence for the fact that larger companies are more inclined to use more sophisticated capital budgeting techniques (Payne, et al., 1999; Ryan and Ryan, 2002; Brounen, et al., 2004). One important reason for this may be that larger companies generally deal with larger projects, which makes the investment in the use of more sophisticated techniques less costly (Payne, et al., 1999). Based on this argument, the researcher expects to find a positive relationship between firm size and the use of DCF methods.

The measure of the educational level of the CFO is included, since it may be expected that CFOs with higher levels of education will have less problems in understanding and using more sophisticated capital budgeting techniques. Again, therefore, the researcher expects a positive relationship between the level of the educational background of the CFOs of the companies and the use of DCF methods. With respect to measures of the industry and age of the CFO, there are no specific and priori expectations about the nature of the relationship.

5.0 SURVEY DESIGN AND METHODOLOGY The data for the analysis have been obtained by using the results of structured questionnaires. The questionnaires were sent to 225 Europe and 120 West African listed and non-listed companies in the period between August 2006 and January 2007. The

GEORGE E EKEHA

26

MBA OCT. 2007

questionnaires consisted of a number of multiple choice questions related to capital budgeting practices of companies, questions specifying firm characteristics, such as size, foreign sales and industry, as well as questions asking for the age and educational background of the respondent.

With respect to the questions related to capital budgeting practices the researcher asked companies to indicate the frequency of the use of different project evaluation techniques (running from 0 to 4, where 0 = never and 4 = always), the cost of capital estimation method used most frequently, the use of methods to estimate the cost of equity. To increase the chances of getting responses from the companies, the researcher decided to keep the survey as short as possible. In total, I included only fifteen questions. The same set of questions was sent to European and West African companies. The questions were all structured in English and were sent by post. To increase the level of response, two reminders were sent to the companies: the first one was two weeks and the second three weeks after the original questionnaires were sent, all by email. The questionnaire was to be completed by the CFO of the company or any person in financial authority. The researcher received 36 responses, 28 from Europe and 8 from West African companies, resulting in a response rate of 12 per cent for the European and 6 per cent for the West African companies sampled. These response rates are somewhat on average to those found in other studies. For example, Graham and Harvey (2001) report a response rate of 9 per cent; Trahan and Gitman (1995) have a rate of 12 per cent and Brounen, et al. (2004) reports a rate of 5 per cent. Kester, et al. (1999) shows an average response rate for the five Asian countries of 15.5 per cent.

6.0 RESEARCH RESULTS AND ANALYSIS This section first describes and compares the characteristics of European and West African companies in the sample that was considered to be relevant as determinants of their capital budgeting practices. Next, it discusses the outcomes related to the answers to the questions on capital budgeting practices, focusing on the use of different capital budgeting techniques and methods used to estimate cost of capital and equity. Finally, the researcher present a univariate and multivariate analysis of the relationship between firm

GEORGE E EKEHA

27

MBA OCT. 2007

characteristics and capital budgeting practices for the European and West African companies in the sample.

6.1 Company and CFO Characteristics Table 1 shows the information on the characteristics of both the European and West African companies in the sample. With respect to total sales the table shows that the European companies on average report higher sales than the West African companies. While 36 per cent of the European companies have sales of more than 1 billion dollars, none of the West African companies reports sales in this category. About 55 per cent of the West African companies have sales of 100 million dollars or above, while the remainder have sales less than 100 millions dollars. If small companies are classified as having sales of less than 100 million dollars, medium-sized companies having sales of between 100-499 million dollars, and large companies having sales of 500 million dollars or more, then the figures indicate that majority of the European companies responding to this study fall within the large companies category, whereas the West African companies responding to the study mainly consist of medium-sized and small-sized companies.

GEORGE E EKEHA

28

MBA OCT. 2007

Table 1: Company Characteristics

Europe (%) Sales (US$ million): Less than 25 million 25 99 million 100 499 million 500 999 million More than 1 billion Foreign sales (% of total sales: No foreign sales 1 24% 25 49% 50 99% 100% Industry Manufacturing Technology Retail and Wholesale Transport and Energy Financial CFOs Level of Education: Professional (Undergraduate) MBA Non-MBA Master PhD Age of CFOs: Less than 40 years 40 49 years 50 59 years More than 60 Source: survey results 18 12 25 12 36 4 18 25 53 0 53 21 12 20 4 12 20 42 26 7 48 45 0 West African (%) 12 36 50 12 0 37 38 25 0 0 62 0 12 26 0 38 25 25 12 0 62 37 0

Additionally, with respect to the share of foreign sales both samples differ quite substantially. Whereas 75 per cent of the West African companies report that their foreign sales are zero or are less than 25 per cent of total sales, about 65 per cent of the European companies state they have 50 per cent or more foreign sales. Table 1 also shows that more than 50 per cent of the European and about two thirds of the West African companies are classified as manufacturing companies. It also provides information on CFO characteristics. In general, European CFOs seem to have a higher level of education. Whereas almost 70 per cent of the European CFOs have a non-MBA master or PhD, this is only 37 per cent for the West Africa CFOs. At the same time, 38 per cent of the West Africa CFOs have an undergraduate degree or professional qualification as their highest level of education; for the European CFOs this is only 12 per cent. With respect to the age structure CFOs in both countries are rather similar.
GEORGE E EKEHA

29

MBA OCT. 2007

6.2 Capital Budgeting Techniques The first question in the questionnaire relates to the capital budgeting practices of companies. Similar to Graham and Harvey (2001) and Brounen, et al. (2004), the researcher asked companies to rate different capital budgeting methods on a 4-point scale in terms of the frequency with which they are used (where 0 = never and 4 = always). This provides information with respect to the methods that are being used, and also with respect to the relative importance of the different methods. The following capital budgeting techniques are used: two DCF methods (NPV and IRR), two non-DCF methods (PB and ARR) and other techniques. As was already discussed above, from a pure theoretical point of view the NPV is the most accurate technique. Yet, it is also the most sophisticated of the four, followed by the IRR method. Both non-DCF methods are considered to be less accurate, of which the PB method is the least sophisticated method, Dayananda et al, 2002.

Table 2 shows the results of the responses from European and West African companies. First of all, the table shows the percentage of CFOs who indicate that they always or almost always use a certain capital budgeting method (scores 3 and 4). Next, the table shows the mean scores for the different methods in both continental blocs. Finally, the table shows the mean scores for different methods of different categories of companies, using the characteristics of companies and CFOs discussed in table 1 to categorize companies in sub-samples.

Before going into the analysis of the differences between the European and West African CFOs, let me shortly discuss this results and compare them to those of other studies relevant for this analysis. The row labelled % 3 and 4 scores presents the percentage of companies that indicate they use a certain capital budgeting method always (score = 4) or almost always (score = 3). The row labelled mean gives the mean score of all companies, using a 0 (never) to 4 (always) scale. The other rows show mean scores of different categories of companies, based on firm characteristics discussed in table 1. The figures in italics are t-test statistics based on standard differences of mean test, showing whether the averages for the different categories of companies are significantly different

GEORGE E EKEHA

30

MBA OCT. 2007

from each other. The t-test statistics shown for the mean scores of West Africa report whether they are significantly different from the mean scores reported for the Europe. (a), (b), (c) are significance levels of 10, 5 or 1 per cent respectively. Table 2: Capital Budgeting Methods Used by CFOs
NPV % responding 3 & 4 scores Mean Score Total Sales <$500 million $500 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: < 50% of sales 50% of sales 89 3.50 3.27 3.75 2.30(b) 3.55 3.38 0.70 3.52 3.47 0.22 3.52 3.48 0.22 3.55 3.45 0.43 50 2.51 4.43(c) 2.12 2.75 1.65(a) 2.95 2.16 2.15(b) 3.13 1.86 3.82(c) 2.43 2.67 0.58 2.97 1.50 4.21(c) IRR PB Europe (N = 42) 75 79 2.98 3.10 2.73 3.25 1.32 3.00 2.92 0.18 2.78 3.21 1.07 3.14 2.81 0.83 3.36 2.80 2.10(b) 3.10 3.08 0.09 3.17 3.00 0.61 3.00 3.19 0.68 ARR 7 0.24 0.09 0.40 1.33 0.10 0.54 1.76(a) 0.22 0.26 0,19 0.05 0.43 1.66(b) 0.20 0.27 0.31 12 1.00 3.92(c) 0.82 1.11 0.90 0.70 1.24 1.81(a) 1.00 1.00 0.00 1.03 0.93 0.31 1.03 0.93 0.31 Other 7 0.17 0.05 0.30 1.56 0.14 0.23 0.51 0.09 0.26 1.06 0.24 0.10 0.86 0.00 0.32 1.98(a) 0 0.02 1.73(a) 0.00 0.04 0.78 0.00 0.04 0.89 0.04 0.00 0.86 0.03 0.00 0.70 0.00 0.07 1.51

% responding 3 & 4 scores Mean Score Total Sales <$100 million $100 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: Yes No

3.15 3.40 2.82 2.82 0.82 2.17(b) West Africa (N = 8) 87 75 3.38 3.16 1.82(a) 0.35 3.29 3.43 0.64 3.45 3.32 0.63 3.48 3.27 1.01 3.43 3.27 0.77 3.42 3.29 0.60 3.00 3.25 1.22 3.00 3.28 1.40 3.00 3.32 1.61 3.10 3.27 0.78 3.03 3.43 1.88(a)

GEORGE E EKEHA

31

MBA OCT. 2007

6.2.1 European CFOs Table 2 shows that the NPV method is the most popular method among the European CFOs. 89 per cent of the respondents indicated they use this method (almost) always, and its mean score is 3.50, which is 0.40 above the second most popular method (the PB method). The IRR and PB method are quite comparable in terms of their mean scores and percentage of CFOs who say they use these methods (almost) always. The ARR method is clearly the least popular: its mean score is 0.24 and only 7 per cent of the CFOs in the sample stated that they use this method (almost) always.

If these results for the European CFOs are compared with those found in Brounen, et al. (2004), it seems that this finding has significantly higher percentages and mean scores for most of the methods reported in the survey. Brounen, et al. show that 70, 56 and 65 per cent of companies (almost) always use the NPV, IRR or PB method respectively, and they report mean scores of 2.76, 2.36 and 2.53, respectively. The differences between their findings and this study may be due to the fact that in their sample there are more smaller companies: almost 40 per cent of their companies have total sales between 25 and 102 million dollars, whereas in this study I have no respondent of the companies in this size category. In contrast, in this study 36 per cent of the companies have sales of more than 1 billion dollars, whereas in the sample of the study by Brounen, et al. only 20 per cent of the companies is in this size category.

Table 2 also shows the results of a standard difference of mean test of the mean scores of the NPV, IRR, PB and ARR method for the five different categories of companies listed in table 1 (size, educational level of the CFO, age of the CFO, industrial sector and percentage foreign sales of total sales). The results of these tests show that for larger companies the mean score for the NPV method is significantly higher (at the 5 per cent level) than for smaller companies. The opposite is true for the PB method. With respect to the PB method, the table also shows that companies with lower foreign sales have significantly higher mean score than companies with higher foreign sales. The other

GEORGE E EKEHA

32

MBA OCT. 2007

mean tests report either insignificant t-values or values that are only significant at the 10 per cent level. These results are comparable to similar tests results presented in Brounen, et al., who also find that larger companies have significantly higher mean scores for the NPV method. They, however, do not find significant differences for any of the other methods for any of the categories of companies they use in their study.

6.2.2 West African CFOs Table 2 shows that the IRR and the PB method are the most frequently used methods. 87 and 75 per cent of the CFOs in the sample stated that they use these methods (almost) always. The NPV method is used much less: only 50 per cent of the CFOs report they use this method (almost) always. Looking at the mean scores the IRR and PB method are comparable with 3.38 and 3.16 mean score respectively, whereas the NPV methods score falls behind with only 2.51. In the researchers opinion, the use of IRR scored very high in these countries because of the unstable nature of inflation and interest rates in those countries. As was true for the European companies sampled, the ARR method is the least popular. The mean score for this method is 1.00 and only 12 per cent of the West African CFOs in this sample say they use this method (almost) always. The researcher is unable to compare the results for West Africa with those found in other studies, simply because the researcher couldnt find any literature on capital budgeting practices in West Africa. Probably the best comparison to make here is by looking at the outcomes of a survey of capital budgeting practices for five Asian countries carried out by Kester, et al. (1999). This is based on the assumption that one can look at most of these countries as developing countries as compare to their fellow West African counterpart.

As already mentioned in section 2, Kester et al (1999) found that 95 per cent of the Asian companies in their sample indicate they use the PB method, whereas 88 per cent report they use the NPV method. The mean scores for both methods are 3.5 and 3.4 (on a scale of 5), respectively. Although it is difficult to make a simple comparison between the outcomes of the survey by Kester, et al and this study results, since the questions in their survey were slightly different from the ones used here, these figures nevertheless seem to suggest that CFOs in West Africa use the NPV method on a much less regular basis than

GEORGE E EKEHA

33

MBA OCT. 2007

their colleagues in other developing countries like those in Asia. The results of the standard differences of mean test of the mean scores of the NPV, IRR, PB and ARR method for the five different categories of companies and CFOs show that, for the higher educated and younger CFOs the mean score for the NPV method is significantly higher than for the lower educated and older CFOs. Also, companies with foreign sales have a significantly higher mean score for the NPV method than companies with no foreign sales. The mean score of the NPV method of the larger companies is higher than for the smaller companies, yet the t-value is only just significant at the 10 per cent level. The other mean tests report either insignificant t-values or values that are only significant at the 10 per cent level. Kester, et al. do not report mean scores of different categories of companies, so it is very impossible to make any comparisons.

6.2.3 European versus West African CFOs First, European CFOs seem to use the NPV method more often than their colleagues in West Africa. Whereas 89 per cent of the European CFOs indicate they (almost) always use this method, this is only true for 50 per cent of the West African CFOs. Instead, the IRR method is used more by West African CFOs than by European CFOs (87 versus 75 per cent). The differences with respect to the use of the NPV and IRR method in the Europe and West Africa are confirmed when looking at the mean scores. In Europe the mean score for the NPV method is 3.50 whereas in West Africa it is 2.51. For the IRR method, mean scores are 2.98 and 3.38, respectively. The differences of the mean scores with respect to the use of the NPV and IRR method between European and West African CFOs are statistically significant, as shown in table 2 (see the t-values in italics presented in the row below the mean scores for West Africa). Note, however, that for the IRR method the difference is only significant at the 10 per cent level. Although the PB method seems to be more popular among European CFOs responding to this study (79 per cent of the European CFOs indicate they (almost) always use this method, against 75 per cent for the West African CFOs), the difference between the mean scores of West Africa versus the Europe is not statistically significant. Finally, the ARR method is not used very much in both West Africa and the Europe; yet the analysis shows that the mean score for the West African CFOs is significantly higher than the score for their European counterparts

GEORGE E EKEHA

34

MBA OCT. 2007

(1.00 versus 0.24). The above discussion seems to suggest that the European CFOs are using the most sophisticated capital budgeting method (i.e. NPV) on a significantly more regular basis than their West African colleagues do. Instead, West African CFOs use the less sophisticated ARR method significantly more than the European CFOs. This result seems to partly confirm the research hypothesis that, based on the fact that there is quite some difference in the level of economic development of the two continental blocs, on average European CFOs will use more sophisticated capital budgeting methods than their West African colleagues.

6.3 Cost of Capital Estimation Methods The next important question in the questionnaire focuses on the methods that are used to estimate the cost of capital. Estimating the cost of capital is necessary when a firm applies discounting techniques like the NPV or IRR method. The researcher asked companies to indicate which method they use most frequently when estimating the cost of capital. In particular, CFOs are asked to make a choice out of the following set of possible methods, i.e. the project dependent (risk-adjusted) cost of capital (PDCC), the weighted average cost of capital (WACC), the cost of debt and other methods. Whereas the PDCC and WACC are the more sophisticated methods, the cost of debt is clearly the least sophisticated of the three methods. In fact, using the cost of debt for capital budgeting purposes is in most cases not appropriate. Yet, since in many cases projects are financed by newly issued debt, using the cost of debt is tempting, and also because of the ease with which it can be calculated.

Table 3 presents the results of the responses from European and West African companies. In particular, it presents the percentage of companies that indicates that a certain method of cost of capital estimation is the one they use most frequently. The researcher again used the characteristics of companies and CFOs discussed in table 1 to categorize companies in sub-samples. The percentages given in this table refer to the share of CFOs in each of the categories, using DCFs methods, who indicated that certain discount rate is the one they use most frequently. Total percentages for different categories of companies may add up to more or less than 100 per cent due to rounding errors.

GEORGE E EKEHA

35

MBA OCT. 2007

Table 3: Most Frequently Used Methods to Measure the Cost of Capital (% of total)
Project dependent Weighted Average (risk-adjusted) Cost of Capital (WACC) cost of capital (PDCC) Europe (N = 28) 10.7 67.9 Cost of debt (CD) Other Methods

% of total companies Total Sales <$500 million $500 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: < 50% of sales 50% of sales

14.3

7.1

14.3 5.0

64.3 70.0

21.4 5.0

0.0 20.0

13.8 0.0

62.1 76.9

10.3 23.1

13.8 0.0

13.0 5.3

56.5 78.9

21.7 5.3

8.7 10.5

14.3 4.8 10.0 9.1

71.4 61.9 65.0 68.2

14.3 14.3 20.0 9.1

0.0 19.0 5.0 13.6

% of total companies Total Sales <$100 million $100 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: Yes No

West Africa (N = 8) 12.5 50 5.9 21.4 47.1 57.1

25 47.1 17.9

12.5 0.0 3.6

15.0 16.0

70.0 40.0

15.0 40.0

0.00 4.0

21.7 9.1

69.6 36.4

4.3 54.5

4.3 0

13.3 20.0

43.3 73.3

40.0 6.7

3.3 0.0

19.4 7.1

54.8 50.0

25.8 35.7

0.0 7.1

GEORGE E EKEHA

36

MBA OCT. 2007

6.3.1 European CFOs The results in table 3 show that 67.9 per cent of the European companies state that they use the WACC for discounting purposes. Only 10.7 per cent of the companies use a project dependent (risk-adjusted) cost of capital. In addition, table 3 shows that a relatively large number of European companies (14.3 per cent) use the simple cost of debt as the discount rate. When looking at the results for different sub-samples of companies, a couple of points are noteworthy. Small companies use the cost of debt more often than large companies do: 21.4 versus 5.0 per cent. In addition, CFOs with higher levels of education make less use of the cost of debt than less educated CFOs. Less educated European CFOs do not use a project dependent (risk-adjusted) cost of capital at all, while 13.8 per cent of the higher educated CFOs indicate that they use this one the most frequently.

6.3.2 West African CFOs Of the West African companies, 50 per cent indicate that they use the WACC most frequently, 25 per cent mention the cost of debt, while 12.5 per cent state that they use the project dependent cost of capital most often. Compared to the European companies, West African companies appear to use the cost of debt more often. In addition, like European CFOs, West African CFOs with higher levels of education use the cost of debt less often than their less educated colleagues. Moreover, small West African companies use the cost of debt more often than larger companies. This result is consistent with those for the European companies in this sample. In contrast to their European counterparts, however, older West African CFOs are more inclined to use the cost of debt. The level of CFO education does not seem to influence the use of the project dependent (risk-adjusted) cost of capital.

6.3.3 European versus West African CFOs In all, when looking at the methods used to measure the cost of capital, the outcomes presented in table 3 suggest that the main differences between West Africa and Europe are with respect to the use of the cost of debt. In West Africa, the cost of debt is used more often than in Europe. Based upon the difference in the level of economic

GEORGE E EKEHA

37

MBA OCT. 2007

development between the two continental blocs, this may be expected, since the cost of debt is a relatively simple method of calculating the cost of capital. Moreover, most companies in West Africa are significantly small in size as compare to those in the European countries, hence the use of the simpler methods.

6.4 Cost of Equity Estimation Methods The researcher finally asked companies to indicate which methods they use to estimate the cost of equity. Estimating the cost of equity is necessary when a firm applies discounting techniques like the NPV or IRR method. The cost of equity is an input for calculating the project dependent (risk-adjusted) cost of capital and the WACC. The researcher asked companies to indicate which methods they use most frequently when estimating the cost of equity. In particular, companies were asked to indicate whether they make cost of equity estimations, and if they do, what type of method they use most.

Although there are several possible methods available, the survey results showed that companies basically use two (in the Europe) or three (in West Africa) different methods on a regular basis. The following methods were mentioned by the respondents: average historical returns on common stock, Capital Pricing Asset Model (CAPM), no estimation done, and other methods (e.g. dividend discount type of models). Of these methods, the CAPM can be seen as the most sophisticated model.

Table 4 presents the results of the responses from European and West African companies. In particular, it presents the percentage of companies that indicates that a certain method of cost of equity estimation is the one they use most frequently. The researcher again used the characteristics of companies and CFOs discussed in table 1 to categorize companies in sub-samples. The percentages between brackets in columns 3, 4 and 5 refer to the share of companies that indicate they do estimate the cost of equity. So, for instance if all the European companies in this sample stated they do estimate the cost of equity, 52 per cent uses the CAPM.

GEORGE E EKEHA

38

MBA OCT. 2007

Table 4: Most Frequently Used Methods to Estimate the Cost of Equity (% of total)
No Average historical Estimation returns on common done stock Europe (N = 28) 35.7 0.0 Capital Asset Pricing Model (CAPM) 33.3 (52.0) Other Methods

% of total companies Total Sales <$500 million $500 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: < 50% of sales 50% of sales

31.0 (48.0)

31.8 40.0

0.0 0.0

27.3(40.0) 40.0 (66.7)

40.9(60.0) 20.0 (33.3)

34.5 38.5

0.0 0.0

24.1 (36.8) 53.8 (87.5)

41.4 7.7 (12.5)

43.5 26.3

0.0 0.0

30.4 (53.8) 36.8 (50.0)

26.1 (46.2) 36.8 (50.0)

19.1 52.4 30.0 40.9

0.0 0.0 0.0 0.0

42.9 (53.0) 42.9 (50.0) 35.0 (50.0) 31.8 (53.8)

38.1 (47.0) 23.8 (50.0) 35.0 (50.0) 27.3 (46.2)

% of total companies Total Sales <$100 million $100 million CFOs Edu. (Master/PhD) Yes No Age of CFOs < 50 yrs 50 yrs or older Industry: Manufacturing Others Foreign Sales: Yes No

64.4 76.5 57.1

West Africa (N = 8) 4.4 (12.7) 0.0 7.1 (16.7)

24.4 (68.5) 17.6 (74.9) 28.6 (66.7)

6.7 (18.8) 5.9 (25.1) 7.1 (16.7)

55.0 72.0

5.0 (11.1) 4.0 (14.3)

35.0 (77.8) 16.0 (57.1)

5.0 (11.1) 8.0 (28.6)

43.5 86.4

4.4 (7.8) 4.6 (33.8)

39.1 (69.2) 9.1 (66.2)

13.0 (23.0) 0.0

66.7 60.0

3.3 (9.9) 6.7 (16.7)

20.0 (60.1) 33.3 (83.3)

10.0 (30.0) 0.0

51.6 92.9

6.5 (13.3) 0.0

32.3 (66.7) 7.1 (100.0)

9.7 (20.0) 0.0

GEORGE E EKEHA

39

MBA OCT. 2007

6.4.1 European CFOs The results in table 4 show that almost 36 per cent of the European companies in this study stated that in most cases they do not estimate the cost of equity. Of the companies that do regularly estimate the cost of equity, roughly half of them stated that they use the CAPM in most cases. When looking at the results for different sub-samples of companies, the table shows that 80 per cent of the manufacturing companies stated they do regularly estimate the cost of equity, a percentage that is much higher than that of the total sample of European companies. When splitting up the companies into two groups based on the age of the CFO, the table shows that companies with younger CFOs seem to be much less regularly making cost of equity estimations than companies with older CFOs (26 versus 44 per cent).

If we turn to the outcomes for those companies that do make frequent estimations of the cost of equity, the table shows that CFOs of smaller companies less frequently use CAPM as compared to their colleagues of larger companies. Moreover, CFOs with lower levels of education use the CAPM more often than highly educated CFOs do. Given the fact that CAPM is the most sophisticated method, the results with respect to the educational level of the CFO may be somewhat surprising and cannot easily be explained. For this, one might have to know more about what the respondents will include in the category other methods, since it turns out to be about 41 and 8 per cent for higher and lower level of educated CFOs respectively, but unfortunately this information is lacking in this questionnaire. One plausible explanation may be that other methods used by European companies consist of methods such as dividend discount models, which belong to the more sophisticated DCF-methods. For the other subsamples, there is no big difference in the use of methods between different types of companies or CFOs.

6.4.2 West African CFOs With respect to the West African companies in the sample, table 4 shows that the percentage of companies stating that they do not regularly make cost of equity estimations is much higher than for Europe: almost 65 per cent for West Africa versus 36

GEORGE E EKEHA

40

MBA OCT. 2007

per cent for Europe. The table also shows that there are quite some differences for several of the sub-samples on the issue of whether or not estimations of the cost of equity are made on a regular basis. In particular, smaller companies, companies with no foreign sales, and companies with CFOs who are older or have lower levels of education make cost of equity estimations (much) less frequently. Of the companies that do regularly estimate the cost of equity, almost 70 per cent stated that they use CAPM in most cases, whereas 13 per cent say that they use average historical returns on common stock as their estimation method. Looking at different sub-samples, the table suggests that higher educated CFOs use the CAPM much more frequently than CFOs with lower levels of education (78 percent versus 57 per cent). Moreover, older CFOs seem to use average historical returns quite often as their method of estimating cost of equity (34 per cent). These outcomes for sub-samples are more or less in line with what has been hypothesized above.

6.4.3 European versus West African CFOs To conclude the analysis in this sub-section, the outcomes presented in table 4 show that there seems to be quite some difference with respect to the use of techniques between Europe and West Africa. In particular, the results from the questionnaire seem to establish that the European CFOs are more inclined to use more sophisticated methods to estimate cost of equity. This outcome is in line with what may be expected based on the differences in the level of economic development between the two economies. It also seems to confirm what was already found before in table 2, that European CFOs use discounting techniques, and in particularly the NPV method, significantly more often than West African CFOs do. This probably explains the higher percentage of European CFOs reporting they make use of cost of equity estimations as compared to their West African colleagues.

6.5 Capital Budgeting Techniques, Cost of Capital and Cost of Equity Estimations: Multivariate Analysis The discussion in the previous sub-sections was based on comparing averages. Although the discussion provided some interesting results on the differences in the use of capital

GEORGE E EKEHA

41

MBA OCT. 2007

budgeting methods between Europe and West Africa, in this section, the researcher wants to go one step further by performing multivariate regression analysis. In particular, the researcher wants to investigate whether the use of different capital budgeting techniques and different methods of estimating the cost of capital is determined by a so-called country effect, i.e. to ask ourselves whether it matters if the company is European or West African when it decides on using a capital budgeting, cost of capital, and/or cost of equity estimation method. When investigating this country effect, the researcher made a consideration for other factors, such as market size, cultural and political environment, that may influence the choice of capital budgeting, cost of capital, and/or cost of equity estimation methods.

6.5.1 The Multivariate Analysis The multivariate analysis is set up as follows. The researcher estimated two different versions of three different models. The first version of the first model establishes to what extent the choice of a specific type of capital budgeting method is determined by the country effect. In the second version of this model, a number of control variables are added to see if the country effect still holds when adding other possible determinants of the choice of the capital budgeting method. The first version of the second model investigates whether the choice of a specific cost of capital estimation method is determined by the country effect, whereas in the second version of this model I again introduce a number of control variables to see if the country effect still holds even after controlling for other possible determinants of the choice of the cost of capital estimation method. The first version of the third model investigates whether the choice of a specific cost of equity estimation method is determined by the country effect, whereas in the second version of this third model I again introduce a number of control variables to see if the country effect still holds even after controlling for other possible determinants of the choice of the cost of equity estimation method.

With respect to the first model I tried to investigate the determinants of three different capital budgeting methods, i.e. the NPV, IRR and ARR method. The PB method has been left out, since the results in table 2 showed that for this method there was no significant

GEORGE E EKEHA

42

MBA OCT. 2007

difference in the mean scores between European and West African companies. The category of other methods has been left out due to the fact that only a very few number of companies in both Europe and West Africa indicated they used other capital budgeting methods. With respect to the second model I analyse the determinants of the decision to make estimations of the project dependent (risk-adjusted) cost of capital, the WACC, as well as of the cost of debt. The category other methods is left out, since the results in table 3 show that the percentage of both European and West African companies that using other methods is very low. With respect to the third model I analyse the determinants of the decision not to make estimations of the cost of equity, as well as of the CAPM and other methods. The control variables included are the same for all model specifications. Thus measures of size, level of education of the CFO, age of the CFO and type of Industry.

The dependent variables are binary dummy variables. In table 5, the dependent variables are created as follows: NPV = 1 if the score for a company for the NPV method is 3 or 4, it is 0 if the score of a company is less than 3; IRR = 1 if the score for a company for the IRR method is 3 or 4, it is 0 if the score of a company is less than 3; ARR = 1 if the score for a company for the ARR method is 1 or higher, it is 0 if the score of a company is 0. The dependent variables used in table 6 are defined as follows: PDCC = 1 if a company indicates that in most cases it uses a project dependent (risk-adjusted) cost of capital, it is 0 if this is not the case; WACC = 1 if a company indicates it uses the weighted average cost of capital on a regular basis to estimate the cost of capital, it is 0 if this is not the case; CD = 1 if a company indicates it uses the cost of debt as an estimate for the cost of capital on regular basis, it is 0 if this is not the case. Finally, the dependent variables used in table 7 are defined as follows: NOCC = 1 if a company indicates that in most cases it does not make estimates of the cost of equity, it is 0 if it does make estimations of the cost of equity on a regular basis; CAPM = 1 if a company indicates it uses the CAPM on regular basis to estimate the cost of equity, it is 0 if this is not the case; Other = 1 if a company indicates it uses another, not explicitly identified model to estimate the cost of equity, it is 0 if this is not the case. The researcher have excluded the share of foreign sales to total sales as one of the control variables, since with respect to this variable the

GEORGE E EKEHA

43

MBA OCT. 2007

West African and European companies cannot really be compared in the context of a multivariate analysis. For the West African companies in the sample 30 per cent does not have foreign sales and another 40 per cent has foreign sales less than 25 per cent. Instead, half of European companies in the sample have foreign sales of more than 50 per cent of their total sales.

The independent variables are also binary variables. I used the following variable specifications: West Africa = 1 if the company is a West African company, it is 0 if the company is European, this variable is used to measure the country effect; Size = 1 if an European company has total sales of less than 500 million dollars or if a West African company has total sales of less than 100 million dollars, it is 0 if a European (West African) company has total sales of 500 (100) million dollars or more; Education = 1 if the CFO of the company has a PhD or Master degree, it is 0 if (s)he has an undergraduate degree; AGE = 1 if the CFO of the company is 50 years or older, it is 0 if (s)he is younger; Industry = 1 if the company is manufacturing company, it is 0 if it is not. The figures in brackets are t-test statistics and (a), (b), (c) are significant levels of 10, 5 and 1 respectively. All estimations are carried out using the logit estimation method, Hosmer, D.W. and Lemeshow, S (1989) and Long S J (1997). The results of the multivariate logit analysis are presented in tables 5, 6 and 7.

6.5.2 Capital Budgeting Techniques Table 5 shows the results for the determinants of the use of the different capital budgeting techniques. The results provide the following picture. First, for the NPV method the country effect is negative and statistically significant (see column [1]). This result can be interpreted as supportive evidence for the fact that West African companies use the NPV method significantly less often than European companies do. This finding supports the hypothesis on the relationship between the level of development and the choice of the capital budgeting technique as discussed in section 1.3 of this paper. This result holds even if I include control variables for size, CFO education and age, and type of industry (column [2]). Moreover, the results show that the choice for the NPV method is also determined by the size of the company and the age of the CFO; both variables have a

GEORGE E EKEHA

44

MBA OCT. 2007

negative and statistically significant coefficient. This means that smaller companies and companies with older CFOs use the NPV method less often than larger companies and companies with younger CFOs do. Based on the discussion in section 4 of this paper, the outcomes with respect to the size variable are as expected. Table 5: Determinants of Capital Budgeting Methods: Multivariate Logit Analysis
NPV [1] Constant West Africa Size Education Age Industry Number of Observations 36 2.001(c) (4.20) -2.046(c) (-3.64) NPV [2] 2.957(c) (3.35) -2.324(c) (-3.61) -1.172(b) (-2.010) 0.525 (0.93) -1.212(b) (-2.15) 0.155 (0.27) 36 IRR [3] 1.036(c) (2.95) 1.043(a) (1.77) IRR [4] 1.585(b) (1.88) 0.807 (1.32) -0.593 (-1.00) -0.523 (-0.83) -0.293 (-0.50) 0.595 (1.02) 36 ARR [5] -2.001(c) (-4.20) 2.315(c) (4.10) ARR [6] 0.028 (0.04) 2.350(c) (3.68) -0.883 (-1.55) -1.626(c) (-2.61) -0674 (-1.15) -0.948 (1.53) 36

36

36

Secondly, table 5 shows that the country effect for the IRR method is positive and significant, which indicates that the IRR method is used more often by West African companies than by European companies (column [3]). Note, however, that the coefficient is only significant at the 10 per cent confidence level. If the control variables are introduced in the model, the country effect is still positive, yet it becomes insignificant (column [4]). This suggests that the choice for the IRR method may not really be different between European and West African companies. This finding is perhaps somewhat surprising in the light of the hypothesis on the relationship between the level of development and the choice of the capital budgeting technique as discussed in section 4, based on which it might have been expected that European companies are more regular users of DCF methods than West African companies. On the other hand, combined with the findings with respect to the use of the NPV method, these findings may make sense. It might be the case that in recent years European companies have been substituting the IRR

GEORGE E EKEHA

45

MBA OCT. 2007

method for the NPV method. Consequently the use of the IRR method by European companies has decreased.

Finally, table 5 shows that the country effect is positive for the ARR method, indicating that West African companies are using this method significantly more often than European companies do (column [5]), a result that still holds after introducing the control variables (column [6]). This result seems to be in line with the hypothesis that has been formulated on the relationship between the level of economic development and the use of capital budgeting techniques. Of the control variables, only the education variable is statistically significant and it has the expected negative sign, meaning that higher educated CFOs will use the ARR method significantly less, which is consistent with what is expected.

6.5.3 Cost of Capital Estimation Table 6 presents the results for the determinants of the use of the different methods of estimating the cost of capital. For the PDCC and WACC there were no statistically significant coefficients for the country effect variable, indicating that for these two methods of estimating cost of capital there is no difference in use between European and West African CFOs. Although the country effect is positive and statistically significant at the 10 per cent level in the bivariate model for the cost of debt estimation method (column [9]), this effect becomes statistically insignificant after the control variables were added (column [10]). In the extended, multivariate model, the size variable is positive and statistically significant at the 1 per cent level. Moreover the age variable is also positive and statistically significant at the 5 per cent level. The result for the size variable is in line with what may be expected based on the discussion in section 4.

GEORGE E EKEHA

46

MBA OCT. 2007

Table 6: Determinants of the Most Frequently Used Methods to Measure the Cost of Capital: Multivariate Logit Analysis
Project dependent (risk-adjusted) Cost of Capital (PDCC) [7] -2.251(c) (-4.28) 0.560 (0.84) Weighted Average Cost of Capita (WACC) [8] 0.693(a) (2.12) -0.560 (-1.26) Cost of Debt (CD) [9] -1.172(b) (-4.06) 0.891(a) (1.62) Cost of Debt (CD) [10] -3.766(c) (-3.61) 0.746 (1.13) 1.886(c) (2.75) -0.974 (-1.52) 1.298(b) (2.02) 1.213(a) (1.72) 36

Constant West Africa Size Education Age Industry Number of Observations

36

36

36

6.5.4 Cost of Equity Estimation Table 7 presents the results for the determinants of the use of the different methods of estimating the cost of equity. The results can be summarized as follows. First, the country effect is positive and statistically significant in the model explaining when companies do not regularly make cost of capital estimations (column [11]), which means that West African companies do make such estimations on a less regular basis than their European counterparts. This result holds even after the control variables were added (column [12]). This outcome seems to be in line with the results presented in table 5, showing that European companies do use the NPV method significantly more often than West African companies do.

GEORGE E EKEHA

47

MBA OCT. 2007

Table 7: Determinants of Cost of Equity Estimation Methods: Multivariate Logit Analysis

No Cost of Capital Estimation (NOCC) [11] Constant West Africa Size Education Age Industry Number of Observations 36 -0.588(a) (-1.83) 1.182(c) (2.64) No Cost of Capital Estimation (NOCC [12] -0.386 (-0.57) 1.283(c) (2.60) 0.308 (0.64) -0.464 (-0.95) 0.581 (1.23) -0.700 (-1.43) 36 Capital Asset Pricing Mode (CAPM) [13] -0.693(b) (-2.12) -0.435 (-0.91) Capital Asset Pricing Mode (CAPM) [14] 0.203 (0.28) -0.664 (-1.26) -0.765 (-1.45) -0.348 (-0.66) -0.798 (-1.55) 0.135 (0.267) 36 Others Others

[15] -0.802(b) (-2.40) -1837(c) (-2.68)

[16] -2.90(c) (-2.87) -1.662(b) (-2.30) 0.900 (1.40) 1.325(a) (1.80) 0.290 (0.46) 0.897 (1.36) 36

36

36

Secondly, the country effect is negative but not statistically significant in the models explaining the use of the CAPM (columns [13] and [14]). Thus, there seems to be no difference between European and West African companies with respect to the frequency with which they use the CAPM to estimate cost of capital. Third, table 7 shows that the country effect is negative and significant for the other methods, suggesting that West African companies use other methods less regularly than European companies do (column [15]). This result remains after adding the control variables (column [16]). If other methods can be interpreted as being dividend discount models which belong to the sophisticated DCF-methods then this finding supports the hypothesis on the relationship between the level of development and the choice of the cost of equity estimation methods presented in section 4. Since detailed information about the contents of the other methods category is lacking, this conclusion remains to be only tentative. The table also shows that the education variable is positive and statistically significant, confirming the idea that more developed countries make use of estimation methods that are positively related to the level of education of the CFO.

GEORGE E EKEHA

48

MBA OCT. 2007

7.0 SUMMARY AND DISCUSSION In this paper, it was argued that the use of capital budgeting practices might be related to the level of economic development. The researcher has given a number of arguments to support this argument. First, financial markets have developed over time, making the use of DCF methods more applicable, convenient and necessary. Due to the development of financial markets (and especially stock markets) shareholder maximization has gained its importance, which has pressured CFOs of companies to use DCF methods over other, more simple and less accurate alternatives. Secondly, training of CFOs has improved over time, which may have enabled them to better understand and therefore use more sophisticated techniques. Thirdly, tools and packages that help the CFO to determine which investments are beneficial to the company have become increasingly sophisticated, which may also have stimulated the use of more sophisticated techniques. Finally, the increased use of computer technology and the related reduction in the cost of technology may have stimulated the use of more sophisticated techniques.

This paper has investigated this hypothesis using information on the use of capital budgeting techniques by companies in the Europe and West Africa. This information was obtained from a survey among 28 European and 8 West African companies, who responded to the questionnaires sent to 345 companies. This minimum response was, in my opinion, due to the limited time allowed for the return of the questionnaires and some financial constrains. With this information, the researcher carried out the analysis using standard differences of mean tests and multivariate regression analysis to see whether the level of economic development matters for the use of capital budgeting practices. I focused on whether there was a so-called country effect, i.e. whether capital budgeting practices differed significantly between European and West African companies and whether these differences can be explained by differences in levels of economic development. The researcher was not aware of any other study in the literature that has looked at this issue.

The main findings of the analysis can be summarized as follows. First, European CFOs use the NPV method significantly more often than their West African colleagues do.

GEORGE E EKEHA

49

MBA OCT. 2007

Second, West African CFOs use the ARR method significantly more than European CFOs do. Third, CFOs of West African companies less often make cost of equity estimations as compared to European CFOs. These results may be explained by the fact that there is still a gap with respect to the level of economic, financial, human and technological development between the two continental blocs. At the same time, however, the study also found that the use of the IRR method does not seem to differ significantly between European and West African companies. The same is true for the estimation of the cost of capital and the use of CAPM as a method of estimating the cost of equity. The latter three results do not lend support to the central hypothesis of this paper.

Therefore, the researcher will restrain himself from drawing too strong conclusions with respect to the importance of the country effect as an explanation for differences in capital budgeting practices between the European and West African companies. However, the researcher still believes that there are some levels of economic factors among the determinants of the choice of capital budgeting practices.

It is therefore proposed that further research into this issue is required and that more and larger data sets should be created, in terms of the number of companies and individual company observations, as well as in terms of the selected countries included in the research.

GEORGE E EKEHA

50

MBA OCT. 2007

LIST OF REFERENCES 1. Boozer, M, Ranis, G, Stewart, F, and Suri, T, 2003, Paths to Success: The Relationship between Human Development and Economic Growth, Center Discussion Paper 874, Yale University: Economic Growth Center. 2. Brealey, R A and Myers, S C, 2003, Principles of Corporate Finance, International Edition, New York, McGraw-Hill Book Company. 3. Brounen, D, De Jong, A and Koedijk, K, 2004, Corporate Finance in Europe: Confronting Theory with Practice, Financial Management, 33, 4, pp.71-101. 4. Daft, R, 2003, Management, Sixth Edition, Thomson Southwestern, Ohio, 5. Dayananda, D, Irons, R, Harrison, S, Herbohn, J and Rowland, P, 2002, Capital Budgeting: Financial Appraisal of Investment Projects Cambridge University Press, Edinburgh. 6. Elumilade, D O, Asaulu, T O and Ologunde, A O, 2006, Capital Budgeting and Economic Development in the Third World: The Case of Nigeria, International Research Journal of Finance and Economics, Eurojournal Publishing, Issue No. 2 7. Evenson, R, 1988, Technology, Productivity Growth, and Economic Development, cited in: Gustav Ranis and T. Paul Schultz (eds.), The State of Development Economics: Progress and Perspectives, Oxford: Basil Blackwell, pp. 486-527. 8. Fang, C and Meiyan, W, 2002, How Fast and How Far Can Chinas GDP Grow? China and World Economy, Issue 5, pp. 9-15. 9. Freeman, M and Hobbes, G, 1991, Capital Budgeting: Theory versus Practice, Australian Accountant, Issue 61, Vol. 8, pp. 36-41. 10. Ghana Centre for Democratic Development, (CDD Ghana), August 2000, The Ghana Governance and Corruption Survey: Evidence from Household, Enterprises and Public Officials. Commissioned by the World Bank and conducted by CDD. 11. Gitman, L J, and Forrester Jr., J R, 1977, A Survey of Capital Budgeting Techniques Used by Major U.S. Companies, Financial Management, Issue 6, Vol. 3, pp. 66-71.

GEORGE E EKEHA

51

MBA OCT. 2007

12. Graham, J R. and Harvey, C R, 2001, The Theory and Practice of Corporate Finance: Evidence from the Field, Journal of Financial Economics, Issue 60, Vol. 2-3, pp. 187-243. 13. Han, C, 1986, The Sophistication of Capital Budgeting in Malaysian Companies, Omega, Issue 14, Vol. 2, pp. 175-181. 14. Hendricks, J A, 1983, Capital Budgeting Practices Including Inflation Adjustments: A Survey, Managerial Planning, Issue 31, Vol. 4, pp. 22-28. 15. Herst, A, Poirters, S and Spekreijse, H, 1997, Capital Budgeting Practices in the Netherlands, Western Decision Sciences Institute, available at Internet address: http://www.sbaer.uca.edu/research/wdsi/wdsi.html. 16. Hosmer, D W and Lemeshow, S, 1989, Applied Logistic Regression, New York: Wiley 17. Kester, G, Chang, R P, Echanis, E S, Haikal, S, Isa, M, Skully, M T, Tsui, K and Wang C, 1999, Capital Budgeting Practices in the Asia-Pacific Region: Australia, Hong Kong, Indonesia, Malaysia, Philippines, and Singapore, Financial Practice and Education, Issue 9, Vol. 1, pp. 25-33. 18. Kim, S H, and Farragher, E J, 1981, Current Capital Budgeting Practices, Management Accounting, Issue 62, Vol. 12, pp. 26-29. 19. Klammer, T P, 1972, Empirical Evidence on the Application of Sophisticated Capital Budgeting Techniques, Journal of Business, Issue 45, Vol. 3, pp. 387397. 20. Levine, R, 1997, Financial Development and Economic Growth: Views and Agenda, Journal of Economic Literature, Vol. 35, pp. 688-726. 21. Long, S J, 1997, Regression Models for Categorical and Limited Dependent Variable, Thousand Oaks: Sage. 22. Montiel, P J, 2006, Obstacles to Investment in Africa: Explaining the Lucas Paradox, Seminar Paper, IMF Seminar, Tunis, 28 February 12 March 2006 23. Morisset, J and Lumenga, N O, 2002, Administrative Barriers to Foreign Investment in Developing Countries. World Bank Policy Research Working Paper, Number 2848.

GEORGE E EKEHA

52

MBA OCT. 2007

24. Onigbinde, A, 2003, Development of Underdevelopment, Frontline Books, Ibadan, Nigeria 25. Osaze, B E, 1996, The Fallacy of R. O. I. As a Performance Measure, First Bank of Nigeria Plc, Bi-annual Review, Ibadan, Vol. 4, No. 9 26. Payne, J D, Heath, W C and Gale, L R, 1999, Comparative Financial Practice in the US and Canada: Capital Budgeting and Risk Assessment Techniques, Financial Practice and Education, Issue 9, Vol. 1, pp. 16-24. 27. Pike, R, 1996, A Longitudinal Survey on Capital Budgeting Practices, The Journal of Business Finance and Accounting, Issue 23, Vol. 1, pp. 79-92. 28. Ryan, P A. and Glen, P. R, 2002, Capital Budgeting Practices of the Fortune 1000: How Have Things Changed? Journal of Business and Management, Issue 8, Vol. 4. 29. Schall, L D, Sundem, G L and Geijsbeek Jr., W R, 1978, Survey and Analysis of Capital Budgeting Methods, Journal of Finance, Issue 33, Vol. 1, pp. 281- 287. 30. Schultz, T W, 1988, On Investing in Specialized Human Capital to Attain Increasing Returns, cited in: Gustav Ranis and T. Paul Schultz (eds.), The State of Development Economics: Progress and Perspectives, Oxford: Basil Blackwell, pp. 339-352. 31. Shao, L P. and Shao, A T, 1996, Risk Analysis and Capital Budgeting Techniques of U.S. Multinational Enterprises, Managerial Finance, Issue 22, Vol. 1, pp. 41-57. 32. Trahan, E A, and Gitman, L J, 1995, Bridging the Theory-Practice Gap in Corporate Finance: A Survey of Chief Financial Officers, The Quarterly Review of Economics and Finance, Issue 35, Vol. 1, pp. 73-87. 33. Truong, L G, Partington, G and Peat, M, 2005, Cost of Capital Estimation and Capital Budgeting Practice in Australia, Unpublished paper, Sydney: University of Sydney. 34. Wong, K, Farragher, E J and Leung, R K C, 1987, Capital Investment Practices: A Survey of Large Corporations in Malaysia, Singapore and Hong Kong, AsiaPacific Journal of Management, Issue 4, Vol. 2, pp. 112-123.

GEORGE E EKEHA

53

MBA OCT. 2007

35. World Bank, 2005, World Development Indicators 2005, Washington DC: The World Bank.

RESEARCH QUESTIONNAIRES These questionnaires are for the purpose of academic attainment in Masters of Business Administration (MBA) in Finance. The researcher intends to find out the use of capital budgeting methods among the developed and developing countries. It is also intended to establish if there be any relationship between level of economic, financial, human and technological development and the choice of capital budgeting methods. No part of any information provided in these questionnaires will be shared or used for any purpose other than as stated above. The researcher will therefore be very glad if the person answering these questionnaires is the Chief Finance Officer (CFO) or in authority of financial matters within the organisation. Thank you for your participation towards this study.

1. What is your position in the company? Financial Director Others

CFO

Financial Manager

(Please state) 40 49

2. Which of the following does your age fall? Less than 40 50 59 60 or above PhD MBA

3. What is your highest level of education? Non-MBA Masters

Undergraduate/Professional Financial

4. What industry does your company belong? Manufacturing Technology Transport & Energy

Retail & Wholesale

5. What is the total volume of annual turnover in dollars? Less than 25 million 25 99 million More than 1 billion 6. What percentage of the total sales is made outside the home country of your company? 100% 50 99% 25 49% 1 25% 100 499 million 500 999 million

No foreign sales 7. Do you use any capital budgeting method to assess projects when making investment decisions? Yes No

GEORGE E EKEHA

54

MBA OCT. 2007

8. If yes, which method(s) do you use? (Please select all that apply) Value (NPV) Internal Rate of Return (IRR) Others

Net Present

Pay Back (PB)

Accounting Rate of Return (ARR)

9. On the scale of 0 4 , with 4 being the method(s) used always and 0 being never, how would you rate the frequency of use for the selected methods above? NPV 0 1 3 4 10. Do you use any technical method to measure the cost of capital of projects when making investment decisions? Yes No Project 2 4 3 4 ; IRR 0 1 2 1 3 2 4 3 4 ; Others 0 ; PB 0 1 1 2 2 3

; ARR 0

11. If yes, which method(s) do you use? (Please select all that apply) dependent cost of capital (PDCC) (WACC) Cost of debt (CD)

Weighted average cost of capital Others

12. On the scale of 0 4, with 4 being the method(s) used always and 0 being never, how would you rate the frequency of use for the selected methods above? PDCC 0 2 1 3 2 4 3 4 ; WACC 0 1 2 1 3 2 4 3 4 ; CD 0 1

; Others 0

13. Do you use any technical method to estimate the cost of equity of projects when making investment decisions? Yes No Capital asset No

14. If yes, which method(s) do you use? (Please select all that apply) pricing methods (CAPM) estimate done

Average historical returns on common stock Others

15. On the scale of 0 4, with 4 being the method(s) used always and 0 being never, how would you rate the frequency of use for the selected methods above? CAPM 0 2 1 3 2 4 3 4 ; Average historical returns on common stock 0 1 2 3 4 1

; Others 0

Thank you for your time and invaluable contribution to this study.

GEORGE E EKEHA

55

MBA OCT. 2007