Venturing into banks : the impact of bank credit on venture capital activity

Venturing into Banks:

The Impact of Bank Credit on Venture Capital

Activity

Abstract Bank finance and venture capital are two external funding sources for firms.

This research examines the impact of bank credit on venture capital for 28 countries

in the period 2005 to 2012. A regression is used to measure the impact of bank credit

on venture capital activity, measured by annual expenditure and completed deals per

country. The results show a negative but insignificant pattern between bank credit and

venture capital.

Rosa Hoogma

Student number 10430520

University of Amsterdam

Faculty of Economics and Business

Supervisor: Mark Dijkstra

Statement of Originality

This document is written by Rosa Hoogma, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

TABLE OF CONTENTS

1 INTRODUCTION"..."4"

2 LITERATURE REVIEW"..."5"

Bank finance"..."5"

Venture Capital"..."6"

Venture capital or bank finance"..."7"

Empirics"..."10"

3 METHODOLOGY & DATA"..."11"

Descriptive statistics"..."13"

4 RESULTS"..."16"

5 CONCLUSION"..."20"

6 REFERENCES"..."20"

APPENDIX A"..."23"

APPENDIX ... 24"

APPENDIX C"..."25"

"

"

"

1 INTRODUCTION

Start-ups are associated with uncertainty and information asymmetries (Sahlman, 1990). According to Jeng and Wells (2000) venture capital can be seen asasolution for the

difficulties in finding external finance sources by start-ups. Additionally, Bottazzi and Da Rin (2002) state that some start-ups are created because of venture capital. Prior research show that venture capital has exploded over the past decades; according to Ueda (2002) in 2001 venture capital investment was 100 times its value relatively to its value in 1977. However, the main external funding source is bank finance. Ueda (2004) suggest that venture capitalists provide capital to firms that have relatively higher growth potential, riskiness, return than firms who obtain external funding by banks. Furthermore, relatively low collateral increases financing through venture capitalists. De Bettignies and Brander (2007) explain that venture capital is chosen when it can add management contributions. Winton and Yerramilli (2008) show that the firm’s choice depends on the state of the economy. Research dedicated to the relation of other finance sources with venture capital is rarely. In particular, the relation between bank finance and venture capital. Therefore the research question is: do countries with less bank credit have more venture capital activity? Using annual panel data of the 28 member states of the Europe Union between 2005 and 2012, this thesis tests whether there is a relation between bank credit and venture capital activity measured by venture capital expenditure and venture capital completed deals on an annual basis. As it is suggested that in economic downturns bank credit is less attractive (Winton & Yerramilli, 2008) this thesis also tests the interaction of bank credit and GDP on venture capital activity. This thesis finds a negative but insignificant relation between bank credit and venture capital. This is in line with the theory provided by Carey, Prowse, Rea and Udell (1993), Ueda (2004), De Bettignies and Brandon (2007), and Winton and Yerramilli (2008) stating that venture capital is an substitute for bank credit. However, as the results are not significantly different from zero, the results should be interpreted with caution.

This thesis follows the following structure; the second part after the introduction contains theoretical background about how bank finance and venture capital work.

Furthermore the entrepreneur’s choice between venture capital and bank financing will be discussed concluding with empirical evidence. The third section describes the methodology, and data. The fourth section shows the results. The fifth section represents the conclusion.

2 LITERATURE REVIEW

Initial equity to start a business is for 31% of the firms funded by the owner himself or family and is therefore the main initialfunding source (Berger & Udell, 1998). When a firm is growing rapidly and the entrepreneur himself has insufficient resources to maintain growth, external sources must be sought (Berk & DeMarzo, 2011). Several studies have pointed out that venture capital contributed to innovation (Hellman & Prui, 2000; Kortum & Lerner, 2000). Venture capital accounts for 8% of innovation in 1992 in theUS, this increased to 14% in 1998. Kortum and Lerner (2000) investigated that revenue growth of venture capital backed companies are on average 36.8%, being 13% more than non venture capital backed firms. Even though venture capital is often associated with financing start-ups, commercial banks are the second main source of funding for new enterprises and therefore the biggest category for external funding1_{. Banks provide 19% of all entrepreneurial financing whereas} venture capital only covers 2% of total funding (Berger & Udell, 1998).

Bank finance

Banks provide capital by issuing loans to consumers and businesses (Miskin, Matthews &

Giuliodori, 2013). Irrespectively of the severity of the firm’s profits, the bank will always receivethe same profit; namely the interest rate. Therefore the bank focuses on avoiding or minimizing default or liquidation and does not pursue growth of the firm once the company is able to pay off his loan (Winton & Yerramilli, 2008). Collateral is a widely used solution for minimizing bad outcomes. Collateral is an asset, which has an intrinsic value and is offered by the borrower to secure the loan in case of default (Berk & DeMarzo, 2011). If the

company is not able to pay off the loan, the lender can seize the collateral to recoup its losses. According to studies (Jeng & Wells, 2000; Ueda, 2004; de Bettignies & Brander, 2006; Winton & Yeramilli, 2008; Ning, Wang & Yu, 2014), collateral is an obstacle for start-ups in lending from banks, because many new enterprises often have only a few tangible assets next to equity. Tangible assets include, for instance, machinery, inventory, and buildings. In most European countries (except, Germany, the Netherlands, and Switzerland) it is not common for banks to hold equity in commercial companies and for this reason equity cannot be used as collateral (Miskin, Matthews & Giuliodori, 2013). If the bank accepts equity as collateral and the company is not able to pay back the loan, the bank owns an equity share in the company.

""""""""""""""""""""""""""""""""""""""""""""""""""""""""

1 _{According to Berger and Udell (1998) other main sources are finance company loans (5%) and trade credit}

Not having reasonable collateral reduces banks’ willingness to take more risk associated with start-ups (Jeng & Wells, 2000). Therefore start-ups face difficulties obtaining bank loans (Winton & Yerramilli 2008). However, Bhaird and Lucy (2008) imply that firms can overcome this problem by using (if available) personal assets to avoid the problem of not having sufficient collateral.

Venture Capital

Venture capital is defined as equity or equity-linked investments made by institutions to young companies with high growth potential subjected to asymmetric information (Gompers & Lerner, 2001). Venture capitalists invest in areas of their expertise. In 1999 60% was invested in information technology (Gompers & Lerner, 2001). In the US in 2005 39% of venture capital was invested in software and biotechnology, these industries contributed to less than 5% of GDP. Retail and distribution, contributing together 31% of GDP, received significantly less venture capital (2%). Therefore De Bettignies and Brander (2007) conclude that venture capitalist invest in areas where their knowledge is most useful.

On average one third of the board of directors of the companies involved with venture capital are members of the venture capital firm. Venture capitalists try to add value to firms by using their knowledge of the market and the start-up process and using their network (Kortum & Lerner, 2000). Venture capitalists monitor their investment heavily, visiting their portfolio companies on average 18.7 times a year. By contrast, banks visit their portfolio companies only once or twice, depending on the probability of default (Winton & Yerramilli, 2008).

Although the companies invested in are characterised by high growth potential, they face high risk of default. Huntsman and Hoban (1980) observed three venture capital funds in between 1960 and 1975. 47 out of 110 portfolio companies generated a negative rate of return. Of these companies 19 generated a loss of 100%. According to Sahlman (2010) these results changed a little, 50% of the venture capital backed companies made losses in 2009.

In order to reduce risk, venture capitalists stage their investments (Sahlman, 2010). They invest in a certain stage of the firm, and if a project does not look profitable to them anymore they can easily stop investing when companies complete that stage. Jeng and Wells (2000) define three well-defined investment stages for venture capitalists, namely seed, start-up, and expansion investment. The first two stages are also referred to as early stage

investments. In all these stages, important developments occur. For example, during the seed investment the commercial potential of the product will be determined and during the start-up investment the product will be actually produced and sold on the market. According to Sahlman (1990) venture capitalists invest 15% of their total investments in early stage firms and 65 % in expansion stage firms. The remaining 20% is used for acquisitions.

When the contract between the venture capitalist and the firm expires, the venture capitalist will (partly) sell his shares (Jeng & Wells, 2000). Initial public offerings (IPOs) are stocks that are issued to the public for the first time and this is the most profitable exit mechanism for venture capitalists (Gompers & Lerner, 1998). In 1999 56% of the exit mechanisms were IPOs. An investment of 1.00 US Dollar in a venture capital backed company generated an average return of 195%. Comparing the same investment in a non-venture capital backed company this investment generated on average a return of 40% (Venture Economics, 1999 as in Jeng & Wells, 2000). Therefore the strength of the stock market based on high returns and IPO’s successes are likely an important determinant for venture capitalist to invest.

Venture capital or bank finance

One obvious difference between bank finance and venture capital is that bank finance is debt-based whereas venture capital is equity-linked. Another main difference is that venture capitalists add management contributions to the firm (de Bettignies & Brander, 2007). De Bettignies and Brander (2007) offer a theoretical model in which the entrepreneur has insufficient resources to fund the firm from internal sources and can choose between two outside investors, namely commercial banks and venture capitalists. The cost of the management contributions added by the venture capitalist for the entrepreneur is giving up ownership. The partial ownership causes entrepreneurs to weaken their incentive to provide effort. This is the case with venture capital and not with bank finance. According to De Bettignies and Brander (2007) managerial contributions are needed when entrepreneurial activity is low, the entrepreneur operates inefficient on a stand-alone basis and/or it can reduce the riskiness of strategies because of their knowledge of the market. Therefore,

venture capital is preferred to bank finance when venture capitalist can add managerial value. De Bettignies and Brander (2007) forget an important factor in order to establish the choice of the entrepreneur, namely the threat of expropriation of the project; the venture capitalist’s ability to duplicate the project or to continue the project on his own. The

probability of the firm’s success could be negatively influenced, as the venture capitalist can enter the market as well. The possible threat of expropriation is therefore a negative aspect of venture capital financing.

While De Bettignies and Brander (2007) explain the value added from venture capitalist after the investment, Ueda (2004) provides a model showing that venture capital add more value than banks before the investment is made. According to Ueda (2004) the

entrepreneur retains private information when negotiating with the bank, while the venture capitalist is assumed to be able to perfectly distinguish profitable from non-profitable

projects, because of their expertise. Additionally, the venture capitalist is able to expropriate the project, because of his expertise. Ueda (2004) concludes that the entrepreneur first discloses the project to the bank, as there is no threat of expropriation. In this way the entrepreneur can get at least the same pay off that he would get if he discloses with the venture capitalist. Ueda (2004) shows that banks focus on high collateral, because of the low screening abilities of banks and the resulting increased information asymmetry. High

collateral punishes the firm harder in case of default. Furthermore, the model predicts that firms that have high growth, high risk and high return are financed by venture capitalists as these aspects all raise the cost of asymmetric information.

Winton and Yerramilli (2008) design a theoretical model to determine the firm’s characteristics when financed by venture capitalist and by banks. Like Ueda (2004), their model predicts that venture capitalist finance firms with high risky strategies. The model is based on the assumption that venture capitalists monitor heavily and play an active role in making firms decisions. Banks are less skilled monitors (Winton & Yerramilli, 2008). When the firm’s strategic uncertainty is low, bank finance is optimal, as it has low screening and no costs arise in terms of giving up partial ownership. Alternatively, when uncertainty is high, firms choose for venture capital, as they are better suited to assess the firm’s situation and provide monitoring.

Additionally, according to Winton and Yerramilli (2008) the choice also depends on the state of the economy. Banks impose themselves to liquidity shocks, as they are subjected to ongoing need for funds. Random liquidity shocks are assumed to be small enough to be resolved by the bank itself. However, this makes them more vulnerable and less attractive compared to venture capitalists in economic downturns. During economic downturns banks face difficulties meeting the ongoing need for funds, causing banks to reduce their numbers of loans or even stop making loans. Venture capital fund has generally an investment term of 10 years and is therefore assumed to be very illiquid.

Ueda (2004), De Bettignies and Brander (2007), and Winton and Yerramilli (2008) suggest that venture capital is a substitute for bank finance assuming both are equally available. Carey, et al. (1993) agree about this. By analysing a conducted study by the staff member of the Board of Governors of the Federal Reserve System and the Federal Reserve Banks about the US debt market, Carey, et al. (1993) show that bank lending partly overlaps with venture capital (figure1)2_{. Both investments are made when the firm is relatively young.} However, in figure 1 can be seen that bank financing starts earlier than venture capital. Carey et al. (1993) argue that this is because small firms need relatively small investments. The amounts invested differ significantly regarding the two external finance sources. Investments

""""""""""""""""""""""""""""""""""""""""""""""""""""""""

made by venture capitalists in Europe are over 8.3 million euros (EVCA, 2014). European Investment Fund (2013) defined a bank loan as less than 0.25 million euros as proxy for small business lending3. Thus, banks provide in general smaller loans to businesses. A reason for that might be that banks require higher collateral on longer-term loans. As mentioned before, start-ups are often unable to provide sufficient collateral.

Furthermore, figure 1 shows that bank loans have a longer investment range. This is consistent with the assumption that venture capitalists prefer riskier firms compared to banks.

Figure 1: Overview bank finance and venture capital.

Carey et al. (1993) Ueda (2004), De Bettignies and Brander (2007), and Winton and Yerramilli (2008) show theoretical that venture capital and bank finance are substitutes. Ueda’s (2004) model predicts that firms that have low collateral, high growth, high risk and high return are financed by venture capitalists. Winton and Yerramilli (2008) suggest that banks finance relatively safer firms compared to venture capitalists. Furthermore, Winton and Yerramilli (2008) add that in economic downturns, bank capital is a less attractive finance source for startups compared to venture capital. De Bettignies and Brander (2007) conclude that venture capital finance is preferred when managerial value can be added. Therefore, substitution between venture capital and bank finance could be based on different needs for expertise.

""""""""""""""""""""""""""""""""""""""""""""""""""""""""

3 _{European Investment Fund defines small business as a company with less than 250 employees. This thesis}

Empirics

In a survey analysing 173 start-ups, which where no older than 10 years and located in Silicon Valley in the US, Puri and Hellman (2000) find that companies who prefer innovator to imitator strategy are likely to choose venture capital as funding source. Innovators are the firms that are first to introduce a product or service on the market. When the product is brought to the market, no close substitutes are offered yet. Imitators obtain leading a competitive position by following business strategies and production patterns from first movers on the market. As these strategies are already known, it can be assumed that imitators follow less risky strategies than innovators. Venture capital is considered by 21 out of 31 innovator companies, as a determinant of their existence: they would not have existed without. This was also the case by 11 out of 22 imitator firms (Puri & Hellman, 2000). Cosh, Cumming and Hughes (2009) support this suggestion by using a dataset of 2520

entrepreneurial firms in the United Kingdom which were formed between 1980 and early 1990s. Companies who recently developed new innovation are 35% more likely to obtain finance from venture capitalists.

Cosh, Cumming and Hughes (2009) analysed rejections rates in application for external funding sources, such as commercial banks and venture capitalists. 81% of the entrepreneurial firms approached banks for external source of funding, 66% of them got the full amount they asked for. 17% got partly the amount requested and the bank declined the remaining 17%. 9% of the entrepreneurial firms participating in the survey approached venture capitalists for external funding. 45% of them got the full amount, 9% got a part and 46% got rejected. In line with Ueda (2004), Cosh Cumming and Hughes (2009) argue that venture capitalists have better screening abilities and so decrease adverse selection. Because of high rejection rates compared to banks, Cosh, Cumming and Hughes conclude that less profitable firms approach venture capitalists rather than banks. One reason why less profitable firms approach venture capitalist is because they need management contributions (De

Bettignies & Brander, 2007). According to Cosh, Cumming and Hughes (2009) companies with relatively higher profits and/or higher assets value are financed by banks instead of venture capitalists. Furthermore, banks are 28% less likely to fund high tech firms. These results are in line with Ueda (2004) and Winton and Yerramilli (2008) saying that banks tend to lend to projects that are safer and have high collateral.

Ning et al. (2014) evaluated venture capital in the US from 1995-2011 using quarterly data of venture capital investments in total expenditures and in number of deals. By doing a regression analysis, they find that a strong stock market (measured by success of IPOs and high returns), positively affected venture capital activity. Additionally, they found

evidence during the financial crisis starting in 2008 when the stock market was weak and capital investment decreased overall. In 2007 4140 deals were made and valued at 31.85

billion dollar, in 2009 3080 deals were made and valued at 20.1 billion dollars. The financial crisis was characterized by bank’s balance sheet deterioration and caused worldwide bank performance to be at its lowest point since decades (Beltratti & Stulz, 2012). Balance sheet deterioration occurs when there is a decline in assets and/or rise in liabilities and causes net worth to decline (asset minus liabilities). Net worth can be used as a proxy for the value of banks (Berk & DeMarzo, 2011). Banks had to deleverage, sell their assets and decrease their loans to consumers and businesses. Drakos (2013) find that terms and conditions tightened on bank loans in the Eurozone during 2009-2011. Data was collected from a survey of the European Commission and Bank Survey. Firms participating in the survey could choose between tightening, being unchanged or easing terms and conditions of bank loans. Countries that were subjected to the sovereign debt crisis experienced higher tightening than other countries in the Eurozone (73% of the participated firms in Spain experienced tightened terms and conditions compared to 23% in Germany). This occurrence made it less attractive to borrow from banks. Even though overall venture capital investment decreased, Ning et al. (2014) found that venture capital investments in early stages remained relatively stable during the financial crisis and increased approximately 30% between 2009 and 2011. This is in line with the reasoning of Winton and Yerramilli (2008) that banks are more affected by liquidity shocks than venture capitalists. Another reason why venture capital activity in the early stages remained relatively stable and bank finance decreased could be that venture capitalists are better in assessing risk (De Bettignies & Brander, 2007).

"

"

"

"

"

"

3 METHODOLOGY & DATA

"

In order to examine the relation between bank credit and venture capital activity the following model is tested:

VCit = β0 + β1Bank Credit it + γ1Market Capitalizationit + γ2Unemploymentit + γ3R&Dit +

γ4GDPgrowthit + γ5 BankCredit*GDP + γ6Crisis + Country + Time + εit

i=1,….,n and t=1,….,T

i and t refer respectively to the country being observed and to the date at which it is observed. VCi,t is the dependent variable and represents venture capital activity in country i and period t.

The measure of venture capital activity is obtained in two ways, namely expenditure and number of completed deals (Kortum & Lerner, 2000). Both variables are expressed in annual changes. The independent variable Bank Credit represents annual changes of bank loans provided by commercial banks.

The control variable Market Capitalization is added to the model and stands for the change in market capitalization of domestic firms on the country’s stock exchange. Gompers and Lerner (1999), Jeng and Wells (2000) and Felix et al. (2007) argue that the strength of the public market is important. Therefore market capitalization from the country’s stock

exchange is added to the model.

Furthermore, GDP, Unemployment, and R&D are included as control variables and represent respectively to the change of GDP, unemployment and research and development expenditure per year. According to Schumpeter (1938), Gompers and Lerner (1999), Felix et al. (2007), and Ning et al. (2014) entrepreneurship and business creation is positively

correlated with the development of the economy. Economic growth translates into more opportunities for investors, and therefore might be associated with an increase venture capital activity. An expanding economy characterized by increased GDP and research and

development expenditure and a decrease unemployment rate is expected to benefit venture capital activity. However, Fèlix et al. (2007) imply that the relation is different between unemployment for the supply and demand side of venture capital. Unemployment could increase the incentive to start a new business and therefore the impact on venture capital activity will be ambiguous.

BankCredit*GDP

the change of GDP. There might be an effect on venture capital activity of bank credit

depending on GDP. Winton and Yerramilli (2008) argue that the choice between bank finance and venture capital depend on the state of the economy. Therefore it is expected that the variable BankCredit*GDP has a negative relation with venture capital.

A dummy variable is added to the regression to control the impact of the crisis. The dummy variable equals 1 if at which the time observed is between 2008-2012 and otherwise 0. In order to increase the accuracy of the relation between venture capital activity and bank credit, country and year fixed effects are added to the model.

For the empirical analysis, data on 28 European Union member states is collected: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Poland, Portugal, Romania, Slovak Republic, Slovenia, Spain, Sweden and United Kingdom. Data for venture capital is obtained from Thomson One database. Thomson One provides detailed and broad range financial statistics. The indicator loans provided by commercial banks is used to measure bank credit provided to businesses. Data is obtained from International Financial Statistics (IFS) database provided by International Monetary Fund (IMF). GDP growth rate, unemployment rate, research and development expenditure, and market capitalization is provided by the World Bank.

Descriptive statistics

Table 1 displays the descriptive statistics of the variables used in the regression. Appendix A contains descriptive statistics in which the standard deviation is decomposed in ‘between’ and ‘within’. Between and within respectively refer to the standard deviation between countries and within time periods. The ‘overall’ values are the same as displayed in table 1. The second column (Obs) shows the amount of observations. The timespan is from 2005 to 2012 for 28 countries, resulting in 224 observations. Data was for some countries not provided by the database used. One reason for this is that private firms do not have to disclose information about their activities. Resulting in 183 observations for venture capital expenditure and 201 observations for venture capital deals. The third column provides the mean. Venture capital expenditure has an average change of 6% whereas venture capital deals have an average change of 24%. The fourth column shows the standard deviation, and the fifth and sixth display the minimum value and maximum value. As seen in the table the minimum change per year is -100% for venture capital expenditure and deals. However, the minimum for change for bank credit is -20%. Additionally, the standard deviation from venture capital expenditure and deals is higher than bank credit (respectively, 67%, 112% and 10%). So it seems to be that bank credit is more stable during time.

Table 1: Descriptive statistics

Figure 2: Overview bank credit and venture capital activity in Europe

Figure 2 shows the total amount of bank credit and venture capital activity in European Union from 2005 to 2012. The total bank credit increased up to 2008 and decreased after the

financial crisis had started. Venture capital activity, both expenditure and deals, increased in 2005, but start declining even before the crisis has begun. Perhaps the venture capitalists were still recovering from the dotcom crisis in 2000 (Ning et al., 2014). After 2009 venture capital acitivity starts to rise slightly again. In 2012 venture capital expenditure are rising whereas venture capital deals are decreasing, suggesting an increase in average amount per deal. In Appendix B venture expenditure and bank credit per country between 2005 and 2012 are displayed. 0 1000 2000 3000 4000 5000 6000 7000 0 20000 40000 60000 80000 100000 120000 140000 160000 2005 2006 2007 2008 2009 2010 2011 2012 Venture capital activity € millions "" Bank credit € billions

Bank credit Venture capital Venture capital deals

Variable Obs Mean Standard

Deviation

Minimum Maximum

Venture Capital Expenditure 174 0.06 0.67 -1 2

Venture Capital Deals 220 0.24 1.12 -1 3.69

Bank Credit 223 0.05 0.10 -0.20 0.43

Market Capitalization 224 0.04 0.39 -0.77 1.23

Unemployment 224 0.05 0.41 -0.33 1.51

R&D Expenditure 196 -0.03 0.65 -0.32 0.34

Figure 3 and 4 are scatterplots of venture capital expenditure and deals and bank credit. No clear relation can be obtained between venture capital acitivity and bank finance since there is no visible uphill or downhill pattern. In appendix C the correlation between the variables is tabulated.

4 RESULTS

Table 3 Regression with venture capital expenditure as dependent variable

Variable 1 2 3 4 5 6 Bank Credit -0.11 (0.86) -0.66 (1.12) -1.00 (0.92) -0.50 (1.15) -0.10 (1.44) -1.42 (1.54) Market Capitalization 0.50** (0.23) 0.52** (0.23) 0.58** (0.29) 0.52 (0.39) Unemployment 0.09 (0.65) 0.16 (0.66) -0.05 (0.79) -0.05 (0.82) R&D expenditure 1.03 (1.26) 1.04 (1.26) 0.28 (1.48) 0.33 (1.46) GDP 0.02 (0.03) 0.04 (0.04) 0.05 (0.05) -0.06 (0.06) Bank Credit x GDP -0.14 (0.20) -0.14 (0.22) -0.21 (0.22) Crisis -0.16 (0.21) 0.21 (0.31) Constant 0.25*** (0.09) 0.20 (0.44) 0.20** (0.12) 0.18 (0.13) -0.09 (0.52) -0.09 (0.48) Country fixed effects

Year fixed effects

No No Yes No No No No No Yes No Yes Yes Observations 174 174 155 155 155 155 R2 0.00 0.11 0.04 0.04 0.15 0.26 Adjusted R2 0.00 -0.07 0.01 0.00 -0.09 0.01

***, **, * refer respectively to a significance level of 1%, 5%, and 10%. Standard errors are in parenthesis.

In table 3 results are tabulated from the regression with venture capital expenditures as dependent variable from 2005 to 2012. Country fixed effects and year fixed effects indicate whether the country specific effects are included in the regression. Even though the results (mostly) do not differ significantly from zero, the results seem to show a trend. In all six models bank credit has a negative sign, indicating a trend between bank credit and venture capital expenditure. Meaning that when bank credit increases, venture capital decreases and vice versa. The negative effect of bank credit is in line with Carey et al. (1993), Ueda (2004), De Bettignies and Brander (2007), and Winton and Yerramilli (2008) theories, stating that

venture capital and bank finance are substitutes. However, because the results are insignificant, they should be interpreted with caution.

Looking at the coefficients of the control variables, the results are not significant different from zero except market capitalization.Therefore the results should also be interpreted with caution. The positive sign of the coefficient of market capitalization (significant at a 5% level in model 3,4,5) shows that market capitalization has a positive relation with venture capital expenditure. Thus, capital expenditure benefits from a stronger stock market as stated by Gompers and Lerner (1999), Jeng and Wells (2000) and Felix et al. (2007) since IPOs are an attractive exit mechanism for venture capitalist. The impact of the unemployment in table 3 is difficult to determine, because the coefficient is positive in model 3 and 4 and negative in model 5 and 6. In addition, the numbers are insignificant. According to Fèlix et al. (2007) unemployment has an ambiguous impact since the relation is different for supply and demand side of venture capital. Unemployment could increase the incentive to start a new business. Research and development expenditure seems to have a positive relation with venture capital. The sign of the coefficient of GDP is negative, suggesting that GDP has a negative influence on venture capital expenditure. This implies that when the economy is contracting, venture capital expenditure increases. This is not in line with Schumpeter (1938), Gompers and Lerner (1999), Felix et al. (2007), and Ning et al. (2014) saying that economic expanding would lead to more venture capital as this would lead to more investment

opportunities.

Furthermore, the coefficient of the interaction term is negative. Implying that the higher GDP, the more negative effect of bank credit on venture capital expenditure. Similarly, the higher the bank credit, the more negative effect of GDP on venture capital expenditure. In other words, the higher the growth of the economy, the stronger the effect of bank credit on venture capital expenditure. This is in line with Winton and Yerramilli’s (2008) theory, stating that bank finance is more attractive to venture capital when the economy is expanding.

Controlling for time fixed effects, it does not affect the sign of the explanatory variable bank credit.Furthermore, the values of adjusted R2 _{are close to zero and sometimes} even below zero (can be interpret as zero). However, when the control variables are added to the model, the adjusted R2 _{shows a small increase (0.01). For this reason and because most} results are not significant different from zero all the above statements must be interpreted with caution.

Table 4 shows the results from the regression with venture capital deals as a dependent variable. This model shows approximately similar trends compared to the model shown in table 3, indicating a decrease in bank credit benefits venture capital deals

(significant at 5% level in model 6). However, in table 4 different results appear when looking at the variables market capitalization, unemployment, the interaction term bank

credit*GDP, and crisis. Again, the results are not significant. The coefficients of market capitalization show a negative pattern. Indicating a stronger stock market decreases venture capital deals. This is opposite of what Gompers and Lerner (1999), Jeng and Wells (2000) and Felix et al. (2007) expected. The coefficient of the unemployment is positive in all models, suggesting that unemployment increases venture capital expenditure, as supported by Felix et al. (2007).The coefficients of the interaction term have a positive sign. Implying that the higher GDP, the more negative effect of bank credit on venture capital expenditure. However, the higher bank credit, the less positive effect of GDP on venture capital deals. The impact of the crisis in table 3 is difficult to determine (insignificant and coefficients switch sign). However, the results of table 4 show a positive pattern. Indicating that venture capital deals increases during the crisis although these do not significantly differ from zero.

Table 3 Regression with venture capital deals as dependent variable

Variable 1 2 3 4 5 6 Bank Credit -0.29 (0.45) -0.06 (0.58) -0.56 (0.51) -0.98 (0.61) -0.65 (0.76) -1.21** (0.68) Market Capitalization -0.15 (0.12) -0.16 (0.12) -0.05 (0.15) -0.17 (0.13) Unemployment 0.40 (0.32) 0.44 (0.32) 0.39 (0.38) 0.42 (0.38) R&D expenditure 0.62 (0.61) 0.66 (0.61) 0.58 (0.69) 0.57 (0.69) GDP 0.03 (0.02) 0.02 (0.02) 0.03 (0.03) 0.01 (0.02) Crisis 0.17 (0.11) 0.28 (0.18)

Bank Credit x GDP growth 0.13

(0.11) 0.14 (0.12) 0.14 (0.12) Constant 0.07 (0.05) -0.10 (0.26) 0.06 (0.07) 0.07 (0.07) -0.25 (0.31) -0.02 (0.28) Country fixed effects

Year fixed effects

No No Yes No No No No No Yes No Yes Yes Observations 219 219 193 193 193 193 R2 0.00 0.06 0.03 0.03 0.10 0.08 Adjusted R2 _{0.00 -0.08 0.00 0.00 -0.10 -0.11}

***, **, * refer respectively to a significance level of 1%, 5%, and 10%. Standard errors are in parenthesis

Looking at table 3 and 4 together, the results show a negative pattern between bank credit and venture capital activity. Additionally, higher GDP lead to an even more negative

pattern between bank credit and venture capital activity. Higher bank credit increases the negative effect of GDP on venture capital deals, but to a less negative effect of GDP on the capital venture deals. However, most results are not significant different from zero all the above statements must be interpreted with caution.

5 CONCLUSION

This thesis tests if there is arelation between bank credit and venture capital activity in order to determine if countries with less bank credit have more venture capital activity. Although most results are not significantly different from zero and results should be interpreted with caution, a pattern can be obtained. The results show a negative pattern between bank credit and venture capital activity. In other words, countries with less bank credit have more venture capital activity. This is in line with main findings on the literature, saying that venture capital and bank finance are substitutes (Carey et al. 1993; Ueda, 2004; de Bettignies & Brander, 2007; Winton & Yerramilli, 2008). According to Ueda (2004) firms with high growth, high risk, and high return are financed by venture capitalists. Winton and Yerramilli (2008) suggest that banks finance relatively safer firms compared to venture capitalists. Furthermore, they state that in economic downturns bank finance is less attractive. According to De

Bettignies and Brander (2007) firms prefer venture capital to bank finance when managerial value is needed. So the substitution between bank finance and venture capital could therefore be based on different needs for expertise. Looking at the control variables, this thesis finds no significant results that these affect venture capital activity except market capitalization. Market capitalization has a positive and significant affect on venture capital expenditure. However, as most results are insignificant the statements above should be interpreted with caution.

8 REFERENCES

Beltratti, A., & Stulz, R. M. (2012). The credit crisis around the globe: Why did some banks perform better? Journal of Financial Economics, 105(1), 1-17.

Berger, A. N., & Udell, G. F. (1998). The economics of small business finance: The roles of private equity and debt markets in the financial growth cycle. Journal of Banking & Finance, 22(6), 613-673.

Berk, J. B., & DeMarzo, P. M. (2011). Corporate finance (2nd _{ed.). Essex: Pearson} Education.

Black, B. S., & Gilson, R. J. (1998). Venture capital and the structure of capital markets: banks versus stock markets. Journal of financial economics, 47(3), 243-277. Bottazzi, L., & Da Rin, M. (2002). Venture capital in Europe and the financing of

innovative companies. Economic Policy, 17(34), 229-270.

Bottazzi, L., & Da Rin, M. (2002). Venture capital in Europe and the financing of innovative companies. Economic Policy, 17(34), 229-270.

Carey, M., Prowse, S., Rea, J., & Udell, G. (1993). The economics of the private placement market (No. 166). Board of Governors of the Federal Reserve System (US).

Cosh, A., Cumming, D., & Hughes, A. (2009). Outside Enterpreneurial Capital*.The Economic Journal, 119(540), 1494-1533.

De Bettignies, J. E., & Brander, J. A. (2007). Financing entrepreneurship: Bank finance versus venture capital. Journal of Business Venturing, 22(6), 808-832. Drakos, K. (2013). Bank loan terms and conditions for Eurozone SMEs. Small

Business Economics, 41(3), 717-732.

Félix, E. G. S., Pires, C. P., & Gulamhussen, M. A. (2013). The determinants of venture capital in Europe—Evidence across countries. Journal of Financial Services Research, 44(3), 259-279.

Gompers, P. A., & Lerner, J. (1999). What drives venture capital fundraising? (No. w6906). National bureau of economic research.

Gompers, P., & Lerner, J. (2001). The venture capital revolution. Journal of economic perspectives, 145-168.

Hellman, T., & Puri, M. (2000). The interaction between product market and financing strategy: The role of venture capital. Review of Financial studies, 13(4), 959-984. Jeng, L. A., & Wells, P. C. (2000). The determinants of venture capital funding:

Huntsman, B., & Hoban Jr, J. P. (1980). Investment in new enterprise: Some

empirical observations on risk, return, and market structure. Financial Management, 44-51.

Kortum, S., & Lerner, J. (2000). Assessing the contribution of venture capital to innovation. RAND journal of Economics, 674-692.

Mac & Bhaird, C., & Lucey, B. (2010). Determinants of capital structure in Irish SMEs. Small business economics, 35(3), 357-375.

Miskin, F.S., Matthews, K., & Giuliodori, M. (2013). The economics of money, banking, and financial markets. Essex: Pearson education.

Ning, Y., Wang, W., & Yu, B. (2015). The driving forces of venture capital investments. Small Business Economics, 1-30.

Nofsinger, J. R., & Wang, W. (2011). Determinants of start-up firm external financing worldwide. Journal of Banking & Finance, 35(9), 2282-2294.

Sahlman, W. (2010). Risk and reward in venture capital. Harvard Business School Entrepreneurial Management case, (811-036).

Sahlman, W. A. (1990). The structure and governance of venture-capital organizations. Journal of financial economics, 27(2), 473-521.

Stock, J.H. and Watson, M.W. (2012). Introduction to Econometrics Global Edition. Pearson Education.

The European Private Equity and Venture Capital Association (2015). The EVCA Yearbook 2014

European Investment Fund (2014), Annual Report 2013

Timmons, J. A., & Bygrave, W. D. (1986). Venture capital's role in financing innovation for economic growth. Journal of Business venturing, 1(2), 161-176.

Ueda, M. (2002). Banks versus venture capital: CEPR Discussion paper 3411. Ueda, M. (2004). Banks versus venture capital: Project evaluation, screening, and

expropriation. The Journal of Finance, 59(2), 601-621.

Winton, A., & Yerramilli, V. (2008). Entrepreneurial finance: Banks versus venture capital. Journal of Financial Economics, 88(1), 51-79.

APPENDIX A

Descriptive statistics

Variable Mean Std. Dev. Minimum Maximum

Venture capital expenditure Overall Between Within 0.24 1.12 0.44 1.05 -1.00 -0.90 -1.58 3.69 1.01 3.41 Venture capital deals Overall

Between Within 0.06 0.67 0.14 0.65 -1.00 -0.25 -1.30 2.00 0.40 2.20

Bank credit Overall

Between Within 0.05 0.10 0.04 0.10 -0.20 -0.02 -0.21 0.43 0.12 0.38 Market capitalization Overall

Between Within 0.04 0.39 0.06 0.38 -0.77 -0.08 -0.79 1.22 0.17 1.21 Unemployment Overall Between Within 0.05 0.25 0.07 0.24 -0.33 -0.07 -0.39 1.51 0.19 1.46 R&D Overall Between Within -0.03 0.08 0.03 0.08 -0.32 -0.11 -0.27 0.34 0.02 0.32 GDP Overall Between Within 1.59 4.34 1.34 4.13 -17.95 -1.93 -18.94 12.23 4.35 11.25

Dummy Crisis Overall

Between Within 0.63 0.49 0 0.49 0 0.625 0 1 0.625 1

The table shows the number of observations (Obs), the mean (Mean), standard deviation (Std Dev) decomposed in before and within, minimum (minimum) and maximum (maximum).

APPENDIX B

Red line: Venture capital expenditure (red line) and bank credit (blue line) per country, 2005-2012. Y-axis shows amount expressed in million euros. X-axis displays the time.

APPENDIX C

Summary of correlation between variables

Variable Venture capital expenditure Venture capital deals Bank credit Market capitalizati on Unemploy ment R&D GDP Crisis Venture capital Expenditure 1.000 Venture capital deals 0.3884*** 1.000 Bank credit -0.0440 -0.0223 1.000 Market Capitalization -0.0771 0.1713*** 0.0240 1.000 Unemployment 0.0206 -0.0153 -0.1557** 0.1628* 1.000 R&D 0.0572 0.0473 0.0369 0.0133 0.0453 1.000 GDP 0.0364 0.0522 -0.2637*** -0.0785 -0.7784*** -0.0967 1.000 Crisis 0.1274* -0.0522 -0.4800*** 0.2787*** 0.3998*** 0.0112 -0.5482*** 1.000

R&D refers to research and development expenditure. ***, **, * refer respectively to significant at 1%, 5%, and 10%