DOES A FREE PRESS NURTURE ENTREPRENEURSHIP?

Transcription

1 DOES A FREE PRESS NURTURE ENTREPRENEURSHIP? Nabamita Dutta University of Wisconsin La Crosse Sanjukta Roy West Virginia University Russell S. Sobel West Virginia University Entrepreneurship is the main engine of economic growth and prosperity. Previous research has explored both the factors that make individuals more likely to be entrepreneurs and the economic policies that foster entrepreneurial activity. In this paper we explore, for the first time, the relationship between media freedom and entrepreneurial activity. A free press should increase entrepreneurial activity because it increases the flow of ideas and information, leading to both more new discoveries as well as an easier ability for entrepreneurs to market and sell new products and innovations. JEL Classification: L26; L82; O43 Key Words: Entrepreneur; Entrepreneurship; Media Freedom; Institutions Introduction The role of a free and vibrant media sector in communicating information to the citizens of a country is well established in the literature. The focus of previous research has been on how a free media helps to foster a more educated electorate, leading to a better ability of voters to hold democratically elected leaders accountable. The important role of a free media in promoting information flows within a democracy was even recognized by the founding fathers in the United States. In 1787, Thomas Jefferson wrote that he would prefer a situation of newspapers without government to one of government without newspapers. While the information flows provided by a free press have been shown to lead to a more efficient political sector (Stiglitz (2002), Coyne and Leeson (2004)), this paper asks whether a free press also contributes to a more efficient and vibrant private sector by increasing the rate of entrepreneurship. With a free press, entrepreneurial individuals will be exposed to a greater quantity and variety of knowledge and ideas, potentially leading to a greater rate of new discoveries based on the use of this information. In addition, a free press may make it easier for entrepreneurs to market or sell their new products and innovations, leading to a higher success rate for new ventures, and a more profitable climate for entrepreneurs. Previous literature on entrepreneurship has identified various demographic factors and economic policies that contribute to a higher rate of entrepreneurial activity. Southern Journal of Entrepreneurship 71

2 However, no prior research has examined the link between press freedom and entrepreneurial activity. Some studies have examined entrepreneurship within the media sector, i.e. media-entrepreneurship but this is not the focus of our analysis. Rather, we hypothesize that a free media leads to broader impacts on entrepreneurship more generally across all sectors in an economy through the more robust information flows present under a less regulated and government controlled media. The media sector in a country bears the brunt of channeling information throughout a society and is thus crucial in shaping the growth and development path of a nation. In this paper, we aim to explore how greater press freedom may actually lead to a higher rate of entrepreneurial activities. Greater press freedom can supply valuable information to potential entrepreneurs and thus augment the process of innovative activities. A free media brings forth national and international best practices as well as information on failed initiatives in businesses. When a country possesses a relatively free media, each individual has wider access to variety of information and viewpoints. In essence, a free media lowers the transactions costs of knowledge and information flows for entrepreneurs. While entrepreneurial alertness to new information, unique ideas, and local knowledge (i.e., asymmetric information) is critically important for the entrepreneurial discovery process (see Minniti 2004), a free media that allows for the dissemination of ideas and effective marketing is equally important in ensuring the growth and survival of new ventures. In contrast, in a country with a high degree of government control and regulation over media, we should expect systematic problems with information flows within society. Straightforward applications of basic public choice theory, such as the ideas of Buchanan and Tullock (1962) and Weingast, Shepsle and Johnsen (1981), would suggest that media policies set through collective-decision-making would result in a bias toward policies that generate clearly identifiable, immediate, and concentrated benefits to special interest groups, or the politicians themselves, at the expense of the long-run prosperity of the largely uninformed majority. Quite simply, it would be impractical to expect a highly government-controlled media sector to provide the same beneficial information flow to the citizenry of a country as is present under a free media. Because we focus on the role of media as an institution, the embedded cultural underpinnings that give rise to each country s different institutions play an important role. Freytag and Thurik (2006) explain that country specific cultural characteristics have a bearing on the preference for entrepreneurship, but not on actual initiatives. The media of a country is a representation of its culture, and thus the cultural background of a nation determines the content and outlook of the media. A free unbiased media embraces the positive traits around the globe and blends it with the existing indigenous culture. Culture can also impact the attitudes of citizens toward entrepreneurship. Beugelsdijk (2007) explores the impact of attitudes toward entrepreneurship and non-entrepreneurship on economic growth across European regions and finds that a greater entrepreneurial culture boosts economic growth. 72 Southern Journal of Entrepreneurship

3 In the light of the above, the context of our paper gains importance because we are examining the implications of a free media as the institutional and cultural information infrastructure within a country and gauging its repercussion on entrepreneurship. We hypothesize that a free media nurtures entrepreneurship. We test both the direct and indirect route of this influence. Directly a free press encourages enterprising behavior mainly by reducing transaction cost and increased marketing, publicity etc opportunities. Indirectly, a free press works effectively via the channel of economic freedom. A country endowed with a higher level of economic freedom provides a more productive and conducive environment for individuals to function in a market. It makes business endeavors easier when supplemented by a free press. We test our hypothesis using international data on entrepreneurial activity from the World Bank Entrepreneurship Survey for 95 countries from 2000 to 2005, matched with data on each country s media freedom from the Freedom House Index. Our results show that indeed, after controlling for other factors, media freedom has a large and significant positive impact on the rate of entrepreneurship across countries. Our results are robust to various alternative specifications and different measures of entrepreneurship. We begin by reviewing the existing research on both entrepreneurship and media freedom in Section II. We then describe our data and present our econometric analysis in Sections III through V. Section VI concludes by summarizing our results. Literature Review Within the economics literature, the modern understanding of the role of the entrepreneur is frequently associated with the work of Schumpeter ([1911] 1934). Schumpeter described an entrepreneur as an innovator who discovers new and more valuable combinations of resources. At any point in time there is virtually a limitless number of new possible ways to combine the inputs and resources available within a society. Even a deck of cards, with only 52 beginning inputs, can be arranged into a staggering 68-digit different number of possible combinations (80,658,175,170,943,878,571,660,636,856,403,766,975,289,505,440,883,277,824,000,00 0,000,000). Within the economy it is the entrepreneur who discovers new possible combinations of resources. A vibrant economy is one in which entrepreneurs can cycle through these many different combinations as quickly as possible, getting rapid feedback on whether the new combination is more valuable than existing ones. In a market economy, the profit and loss system serves this feedback role. 1 This process is both valuable and disruptive, as the creation of new products often leads to the obsolescence of others, a process Schumpeter termed creative destruction. 2 Within Schumpeter s 1 According to Sobel (2005), the key to prosperity is having a process in place that helps a state s resources discover which of these different goods or services have the highest value added in the marketplace. 2 In addition to being a disruptive force, entrepreneurship can also be equilibrating as was pointed out by Kirzner (1973). Kirzner viewed the entrepreneur as someone who discovers a previously unnoticed profit opportunity and acts on it. In this framework, the entry of new entrepreneurs into a profitable industry, and the resulting long-run equilibrium reached are also a result of entrepreneurship. Shane and Venkatamaran (2000) describe an entrepreneur as someone who bears the burden of discovering, evaluating and exploiting opportunities to create new goods and services. Southern Journal of Entrepreneurship 73

4 framework, a free press contributes to entrepreneurship basically by increasing the number of cards in the deck and the knowledge of the cards among individuals, allowing the discovery of more possible final combinations. Previous research has clearly demonstrated the importance of entrepreneurship for economic growth and prosperity. Reynolds, Hay, and Camp (1999) show that onethird of the differences in national economic growth rates can be attributed to differing levels of entrepreneurship. Supporting these findings, Zacharakis, Bygrave, and Sheperd (2000) study 16 developed economies and find that entrepreneurial activity explains approximately one-half of the differences in GDP growth. More recently, Henderson (2002) shows that entrepreneurs significantly impact economic activity at the local level by fostering localized job creation, increasing income, and connecting local economies to the larger global economy 1. Minniti and Lévesque (2008) show that even in the absence of significant R&D activities within a country, imitative entrepreneurship significantly contributes to economic growth. In summary, successful entrepreneurship expands the size of the economic pie by allowing individuals to generate more wealth from the limited resources available and a free press can contribute to this process by increasing the flow of information within an economy. A recent paper by Minniti and Bygrave (2001) treats entrepreneurs as individuals who learn from mistakes rather than treating them as rational agents in the perfect sense. Entrepreneurs are shown to realize from their mistakes, improve their decision choices and continuously update their stock of knowledge acquired from past experiences. In an environment with free flow of information, thus, entrepreneurs get a better chance to improve and update their knowledge base and performance. Minniti, Levesque and Shepherd (2009) also notes that information gathered through observation and participation play an important part in entry decisions of entrepreneurs and acquiring these two types of information involves a tradeoff, in terms of delay in entering the market. A free media which ensures free and ample flow of information would influence the amount of information acquired through observation and hence have an impact on the entry decision of entrepreneurs. Turning to the literature on media freedom, its role in the economic development of a nation has been stressed in a vast array of literature. The media industry, public or private, plays a significant role in any economy by forming the eyes and the ears of the populace, providing a voice for the people, spreading economic information, and emphasizing the interactions of the industry with the masses and the government. The three criteria for effective media are independence, quality, and reach to the people. These benchmarks ensure that information is reported without the fear of government and other interest groups, views are expressed from a wide variety of perspectives, and media has the capacity to generate political, social, and economic information to all segments of the society. 1 Lévesque and Minniti (2006) show that Becker s theory of time allocation explains the negative relationship between entrepreneurial initiatives and an individual s age that has been identified in the empirical literature. 74 Southern Journal of Entrepreneurship

5 Sen (1982, 1999) shows that an effective and free media sector, along with a multiparty democracy, is capable of preventing the occurrence of famines. According to Djankov, Mcliesh, Nenova and Shleifer (2003), economies with intense government ownership of the media suffer from poverty, high infant mortality rates, less access to sanitation, higher corruption, and less developed capital markets. Stiglitz (2002) points out the significance of the media in improving government accountability and transparency. The other various contexts for which the role of media has been studied are in solving principal (government)-agent (citizens) problems (Besley, Burgess and Prat, 2002; Besley and Burgess, 2001), in public policy (Spitzer, 1993) and in shaping corporate policy (Dyck and Zingales, 2002). Government control on the flow of media-provided information reaching the citizens has been shown to be detrimental for the development of an economy. A country with significant state control over the media provides additional temptation to politicians to abuse their power. According to Coyne and Leeson (2004), a free media can contribute to successful adoption of policies aimed at economic progress. According to them, successful economic development requires co-ordination of efforts by politicians along with the agreement of the populace on policies. They argue that media can play a successful role in shifting games of conflict to games of coordination between politicians and the populace. Further, Leeson (2008) has shown that economies with greater government control of the media have citizens who are politically ignorant and apathetic. In the context of Mexico, the privatization of media in 1989 brought into focus stories about government scandals to a large extent (Islam, 2002). Dutta and Roy (2009) have shown that greater foreign direct investment inflows enhance the level of media freedom. Further, Kassem (2002) points out that the Egyptian press is a prominent stronghold of the government and is prime target for corruption. Virtually all (90%) of the press is controlled or funded by one or more of Egypt s monarchies or military dictatorships, and its Office for Censorship of Foreign Publications monitors and censors 80 percent of Egypt s press. No prior research, however, has explored the impact of a free press on the rate of entrepreneurship. A free press, as discussed above, not only boosts economic performance but also brings in a liberated socio-economic and political environment. A free press is a source of unlimited and unbiased information from across the globe. Such information includes trends in consumer demand, business climates, market prospects, information on availability of venture capital and also national and international best practices. Such widespread availability of information surely reduces transaction cost of innovation as well as makes it easier for entrepreneurs to market their new products to consumers. We hypothesize that a free media should be a significant positive determinant of the rate of entrepreneurship across countries. This can be backed up from the perspective of either Kirzner or Hayek s views on entrepreneurship or even from Romer s perspective on technological innovation. While Kirzner viewed entrepreneurship as successful arbitrage opportunities, Hayek viewed entrepreneurship as innovation. Romer, in the context of growth theory, establishes that technological innovation is the key to Southern Journal of Entrepreneurship 75

6 economic growth. He views this innovation in the knowledge and skill set of the economy. A free media through information dissipation can take any of the above routes and encourage entrepreneurship. However, this is outside the scope of this paper. Here we choose to focus on the impact of a free press on entrepreneurship via any of the above routes. The paper, however, puts forth an extension of the above hypothesis. We claim that when coupled with high levels of economic freedom, a free press encourages entrepreneurial endeavors even more. This finds ample support in the literature which shows how important economic freedom is for entrepreneurship. Hall and Sobel (2008) use national and international evidence to show how more economic freedom is conducive for entrepreneurship. Kreft and Sobel (2005) emphasize that economic freedom leads to higher growth specifically because the former underlies productive entrepreneurial activities. Data Our data for entrepreneurship has been taken from the World Bank Entrepreneurship Survey, 2008 database. The database consists of cross-country annual data on the number of total and newly registered businesses for 95 countries during Appendix 1 provides a list of the countries in the data. For the purpose of analysis, we consider the time-span This is in consistency with the availability of data for other control variables. The main variables are business density which is the number of total registered corporations divided by total working age population and new density which is the number of newly registered corporations divided by the total working age population. The variable of prime importance is new density as it captures the new entrepreneurial initiatives within the country. Our explanatory variable of interest is media freedom. The data for this is taken from the Freedom House Index. Freedom House s survey data is the most comprehensive dataset available on global media freedom. The level of press freedom in each country is based on twenty three methodology questions divided into three categories: the legal environment, the economic environment and the political environment. The legal environment category judges laws that restrict media content, the ability of journalists to operate freely within the country, and the government s use of regulations to curb media operations. The degree of political control over the content of news media forms the basis for the political environment. The economic environment includes the structure of media ownership, transparency and concentration of ownership and the impact of corruption and economic institutions on media. The press freedom score ranges between 0 and 100 points. The Freedom House index assigns a higher score to lower levels of press freedom. Thus, for the convenience of our analysis, the scores have been rescaled so that a higher score denotes a more free media environment. The choice of other control variables is based on prior literature on the determinants of entrepreneurship. The control variables used are working age population as a percentage of GDP, male labor force participation rate, GDP per capita (in constant 76 Southern Journal of Entrepreneurship

7 2000 U.S. dollars), foreign direct investment (FDI) inflows, and financial sector development. While the first three are geographic and demographic factors, the latter two are important factors contributing to a successful and encouraging entrepreneurialfriendly environment. Finally, we control for economic freedom. We consider two different indices of economic freedom compiled by the Fraser Institute and Heritage Foundation respectively. All these data have been taken from World Development Indicators (WDI) 2007 database. For the alternative measure of financial development, we use the Beck, Demirgüç-Kunt and Levine (2000) database. Appendix 2 provides a description of all the variables used in the analysis. Methodology and Empirical Analysis Prior to performing our econometric analysis, it is worthwhile to examine the correlations in the raw data between the rate of entrepreneurship and press freedom. Figure 1, panels (a) and (b) show the scatter plots of our data for press freedom versus both business density and new business density respectively. Even in the raw data there appears to be a clear positive correlation between press freedom and the rate of entrepreneurship using either measure. Business Density Figure 1: Press Freedom and Entrepreneurship 1a) Press Freedom vs. Business Density Press Freedom Southern Journal of Entrepreneurship 77

8 25 1b) Press Freedom vs. New Business Density 20 New Density Press Freedom To explore our hypothesis controlling for other factors, we undertake pooled ordinary least square (OLS) analysis. Regressions based on pooled data combine time and space and, hence, allows greater variability of the data in comparison to a simple time-series or cross-section sample (Hicks 1994 and Schmidt 1997). We undertake a pooled OLS estimation for an unbalanced panel of 92 countries over the years 2000 to We run the following empirical specification: where Entrepreneurship it represents the specific measure of entrepreneurship used in the specification (either business density or new business density) for country i in time period t. Press Freedom it is our main variable of interest, and is the matrix of control variables. The control variables used for the benchmark regression are total working age population, GDP per capita constant, male labor force participation rate, level of financial development, foreign direct investment (FDI), and economic freedom. The proxy used for financial development is domestic credit provided by banking sector (we examine alternative proxies later). Based on our hypothesis, we expect to be positive implying that higher press freedom generates a higher rate of entrepreneurial activities. 1 Following previous literature, we have also checked the robustness of our findings by running both crosscountry and 5 year panel regressions. Our results remain unaffected with the alternate specifications. 78 Southern Journal of Entrepreneurship

9 Table 1: Pooled Ordinary Least Squares Estimates of the Impact of Press Freedom on Business Density 1 Independent Variable: (1) (2) (3) (4) (5) Press Freedom 0.246*** 0.249*** 0.243*** 0.202*** 0.199*** (0.082) (0.083) (0.081) (0.063) (0.056) Total Working Age Population 1.514*** 1.684*** 1.522*** 1.500*** 1.286*** (0.218) (0.217) (0.224) (0.221) (0.199) GDP per cap 2 (constant US $ 2000) Male Labor Force Participation Rate 0.6*** 0.7*** 0.5*** (0.1) (0.1) (0.1) (0.2) (0.2) * (0.248) (0.237) (0.252) (0.254) (0.232) Financial Development 0.063* * ** * (0.037) (0.0164) (0.0179) (0.0158) FDI (0.0153) * (0.0373) (0.0351) (0.0342) Economic Freedom 7.460*** 0.692*** (2.275) (0.159) Constant *** *** *** *** *** (32.78) (31.16) (33.56) (36.52) (32.23) Observations R-squared F statistic Notes: Robust standard errors in parentheses; economic freedom variable in column 4 is the Fraser Institute index and in column 5 is the Heritage Institute index; significance levels: *** = 1%, ** = 5%, * = 10%. Table 1 presents our benchmark results. To uncover any problems with correlations among the independent variables we run the regressions first considering the variables for financial development and FDI separately and then including them together in a single regression. We subsequently add the control for economic freedom. The results show that press freedom is positive and significant at the 1% level. Thus, the higher is the level of the press freedom within a country, the greater is the rate of 1 The regressions include time dummies that have not been reported in the table 2 GDP per capita is in constant 2000 US $ (billions). Southern Journal of Entrepreneurship 79

10 entrepreneurship. Regarding the controls, almost everything is significant except financial development. As expected, total working age population, GDP per capita, and male labor force participation rate have a positive impact on entrepreneurship rates. In column (2) we control for FDI inflows and do not include financial development. The results almost remain identical and the coefficient of press freedom remains significant at the 1% level. In column (3) we include both financial development and FDI inflows. The coefficient of press freedom is still positive and significant at the 1% level. In the subsequent pair of columns we also control for each country s level of economic freedom (first the Fraser Institute measure, then the Heritage Foundation measure). In both cases, economic freedom is positively associated with entrepreneurial activity, as expected, and press freedom remains significant and positive at the 1% level. The magnitude of the press freedom coefficients suggest that press freedom has a substantial impact on entrepreneurial initiatives. Column (4) results suggest that a percentage point rise in press freedom raises business density by 20 percentage points. A one standard deviation increase in press freedom increases business density by 5.3 percentage point. The impact is the same for column (5). 80 Southern Journal of Entrepreneurship

11 Table 2: Pooled Ordinary Least Squares Estimates of the Impact of Press Freedom on New Business Density 1 Independent Variable: (1) (2) (3) (4) (5) Press Freedom *** *** *** *** *** (0.0080) (0.0075) (0.0078) (0.0063) (0.0046) Total Working Age Population 0.131*** 0.136*** 0.133*** 0.174*** 0.106*** (0.020) (0.019) (0.020) (0.027) (0.018) GDP per cap 2 (constant US $ 2000) 0.09*** 0.09*** 0.09*** ** (0.02) (0.02) (0.02) (0.02) (0.02) Male Labor Force Participation Rate ** *** ** 0.105*** * (0.0236) (0.0222) (0.0234) (0.0337) (0.0228) Financial Development * *** * ( ) ( ) ( ) ( ) * FDI ( ) ( ) ( ) ( ) Economic Freedom 1.148*** 0.104*** (0.290) (0.0230) Constant *** *** *** *** *** (3.008) (2.825) (2.999) (5.405) (3.340) Observations R-squared F statistic Notes: Robust standard errors in parentheses; economic freedom variable in column 4 is the Fraser Institute index and in column 5 is the Heritage Institute index; significance levels: *** = 1%, ** = 5%, * = 10%. 1 The regressions include time dummies that have not been reported in the table 2 GDP per capita is in constant 2000 US $ (billions).

12 We run identical specifications in Table 2 but with an alternative measure of entrepreneurship, new business density. This measure gives an estimate of the amount of new business activities which are arising in the country in each time period. The results are almost identical to those found using total business density. The coefficient of press freedom is positive and significant at the 1% level in all specifications. Thus, the results imply that countries with more press freedom have higher levels of entrepreneurship. Economic freedom also remains positive and significant at the 1% level for new business density. Column (4) results suggest that a percentage point rise in press freedom raises new business density by 2 percentage points. A one standard deviation increase in press freedom increases new business density by 0.5 percentage point. The impact is the same for column (5). We also run our specifications by controlling for time dummies which help to control for year effects. We control for time dummies in alternate specifications with the two different proxies of entrepreneurship. The results remain robust to the inclusion of time dummies. The coefficient of press freedom remains significant at 1% level for both specifications. We also control for economic freedom indexes of Fraser Institute and Heritage and the results remain unaffected. In the above discussion we establish the direct impact a free media has on entrepreneurial endeavors. Now, we address the indirect route through which press freedom can affect entrepreneurship the channel of economic freedom. Press freedom will enhance economic freedom which, in turn, will augment entrepreneurial initiatives. In order to test this, we include an interaction term in our specifications. The interaction term is defined as press freedom of country in period multiplied by economic freedom of country in period. A positive coefficient of the interaction term would imply that for given levels of economic freedom, as press freedom rises, there are higher entrepreneurial initiatives. This positive association is enhanced for higher levels of press freedom. A negative interaction term would imply the opposite. We rerun the specifications of Table 1 and 2. The results are presented in Table 4. Columns (1) and (2) consider the economic freedom measure from Fraser while columns (3) and (4) consider the economic freedom measure from Heritage. The results show that interaction term is positive and significant implying that for given levels of economic freedom, as press freedom rises, entrepreneurial activities go up. To get a clearer picture, we consider the partial derivative of equation (2) with respect to press freedom. The equations is as follows, 82 Southern Journal of Entrepreneurship

13 Setting equation (2) equal to zero, we can find the critical level of economic freedom for which press freedom has no impact on entrepreneurial initiatives. Initially, we consider economic freedom measure from Freedom House. For business density, this critical score is found to be 6.03 and for new density the value is Thus, for low levels of economic freedom, press freedom, in fact, has a negative impact of entrepreneurial activities. Yet, for even a marginal increase in economic freedom, the association between press freedom and entrepreneurial activities is enhanced. Robustness We performed a battery of possible robustness checks on our data, and our results remained unchanged. In this section we discuss these alternative specifications and methodologies we used to ensure the robustness of our results. First, the presence of outliers in a sample can lead to biased estimates. Thus, we employed robust regression specifications to control for any possible outlier effects in our sample. Table 3 presents these results for both measures of entrepreneurial activity and both measures of economic freedom. For all specifications, press freedom remains positive and significant at the 1% level. Yet, robust regression specifications may not generate efficient estimators in the presence of outliers. When there are outliers present, Koenker and Bassett (1978) suggested quantile regression methodology which generates efficient estimators. The model is estimated based on the quantile s asymmetrically weighted absolute residuals of the median rather than the mean of the distribution. Quantile regressions are nonparametric and, thus, assume no underlying distribution. We rerun our specifications employing quantile regression methodology. The specifications of Tables 1 and 2 are rerun by estimating the model with weighted absolute residuals of the mean. The results remain robust to the alternate regression methodology. Similar to Table 1, we control for domestic credit provided by the banking sector and economic freedom measure from Fraser Institute. In alternate specification, we control for economic freedom measure from Heritage database. For business density measure of entrepreneurship, the coefficient of press freedom is significant at the 1% level for both measures of economic freedom. In case of new density, while the coefficient of press freedom is not significant for the economic freedom index from the Fraser Institute, it is strongly significant for the alternate measure of economic freedom. In all these specifications, times dummies are included. We also check our results by considering an alternate measure of financial development. The proxy considered is private credit given by banks. We run our 1 Economic Freedom varies from 1 to 8. Relative to that range, it might seem that press freedom can generate higher rate of entrepreneurial activities only when the level of economic freedom is sufficiently high. But for our sample, 333 out of 435 observations have economic freedom greater than or equal to 6.04 and 272 observations are greater than or equal to 6.4. The average for economic freedom is 6.7 for our sample. Thus, press freedom helps most of the countries for our sample. Southern Journal of Entrepreneurship 83

14 specifications by excluding and including time dummies and also running quantile regressions. The results remain unaltered for the all the alternate specifications. There can be potential problems of endogeneity in the specifications. Specifically, real GDP per capita can endogenous while real GDP per capita can be determinant of rate of entrepreneurship, entrepreneurial activities can also be determinant of real income. To overcome this bias, we consider initial values of real GDP per capita and rerun our specifications. Quantile regression methodology is considered to avoid outlier bias. The results remain unaltered although it is weakly significant for new density measure. We also check our results after controlling for private credit given by banks. In a similar way, levels of entrepreneurship can also be a determinant for the other control variables. To overcome the overall endogeneity bias, we consider cross sectional specifications. We consider initial 1 values for all control variables, including press freedom. For the dependent variable proxies, we consider the values for Private credit given by banks is used as a proxy of financial development to maximize the set of observations. We check our results for economic freedom data from both Fraser and Heritage. The results show that while the coefficients of press freedom are significant for all the alternate specifications, they are weakly significant for the ones with business density and controlling for economic freedom from Fraser Institute. In addition to running our main pooled OLS specifications, we also performed individual cross-sectional regressions for each of the five time periods separately. We also do further robustness checks by using an alternative measure of financial development (private credit given by banks and other financial institutions), and specifications including country unemployment rates, all of which resulted in identical results to the ones presented above. Thus, our results lead to a clear conclusion: the level of press freedom is a positive, significant, and robust determinant of the rate of entrepreneurship in a country. Conclusion This paper is an important contribution to the entrepreneurship literature. So far, research has shown that factors like liquidity constraints, government regulations, and economic freedom are important determinants of entrepreneurship rates across countries. In this paper we show that, accounting for all of these variables, one other factor a free media significantly impacts entrepreneurship rates as well. A free press generates information flows that lead to a higher rate of innovation and a greater ability of entrepreneurial ventures to succeed and penetrate the national marketplace. Our results are robust to alternative model specifications and the inclusion of alternative control variables. We thank the editor and the two anonymous referees for indispensable comments and suggestions. We gratefully acknowledge the financial assistance of the Kendrick Fund. 1 Since we have an unbalanced panel, the initial year is not the same year for all countries. 84 Southern Journal of Entrepreneurship

15 References Baumol, W.J., (1990), Entrepreneurship: productive, unproductive and destructive, Journal of Political Economy, 98 (5), Beck Thorsten, Asli Demirgüç-Kunt and Ross Levine, (2000), "A New Database on Financial Development and Structure," World Bank Economic Review, 14, Besley, Timothy and Robin Burgess (2002), The Political Economy of Government Responsiveness: Theory and Evidence from India, Quarterly Journal of Economics, 117, Besley, Timothy and Andrea Prat (2006), Handcuffs for the Grabbing Hand? Media Capture and Government Accountability, American Economic Review, forthcoming. Beugelsdijk Sjoerd (2007), Entrepreneurial Culture, Regional Innovativeness and Economic Growth, Journal of Evolutionary Economics,17, Buchanan, James M. and Gordon Tullock (1962), The Calculus of Consent: Logical Foundations of Constitutional Democracy, Ann Arbor: University of Michigan Press. Coyne, Christopher J. and Peter T. Leeson (2004), Read All About It! Understanding the Role of Media in Economic Development, Kyklos, 57, 21-4 Djankov, Simeon, Caralee McLiesh, Tatiana Nenova, and Andrei Shleifer (2003), Who Owns the Media?, Journal of Law and Economics, 46(2), Dutta, Nabamita and Sanjukta Roy (2009), The Impact of Foreign Direct Investment on Press Freedom, Kyklos, 62(2), Dyck, Alexander and Zingales, Luigi (2002), The Corporate Governance Role of the Media, CEPR Discussion Papers 3630, C.E.P.R. Discussion Papers. Freedom House (2006), Freedom of the Press 2006: A Global Survey of Media Independence. Karin Deutsch Karlekar (ed.), New York: Rowman and Littlefield Publishers, Inc. Freytag Andreas and Roy Thurik (2007), Entrepreneurship and its determinants in a cross-country setting, Journal of Evolutionary Economics,17, Gwartney, James and Robert Lawson ( 2009), Economic Freedom of the World : 2009 Annual Report. Vancouver, Fraser Institute. Southern Journal of Entrepreneurship 85

16 Hall, Joshua C., and Russell S. Sobel (2008), Freedom, Entrepreneurship, and Economic Growth, Chapter 6 in Alvaro Vargas Llosa (ed.), Lessons from the Poor: Triumph of the Entrepreneurial Spirit (Oakland, CA: Independent Institute). Henderson, J. (2002), Building the rural economy with high-growth entrepreneurs, Federal Reserve Bank of Kansas City Economic Review, 87, 45 70, Third Quarter. Islam Roumeen (2002), Into the Looking Glass: What the Media Tell and Why An Overview, in R. Islam ed. The Right To Tell: The Role Of The Media In Development. Washington D.C.: The World Bank. Kassem, Hisham (2002), How the Cairo Times Came to Be Published out of Cyprus, in: Alisa Clapp-Itnyre, Roumeen Islam and Caralee McLiesh (eds.), The Right to Tell: The Role of Mass Media in Economic Development. Washington D.C.: The World Bank. Kirzner, Israel M., (1997), Entrepreneurial discovery and the competitive market process: an Austrian approach, Journal of Economic Literature, 35, Kirzner, Israel M., (1973), Competition and Entrepreneurship, Chicago: University of Chicago Press. Kreft, S.F. and Russell S. Sobel (2005), Public policy, entrepreneurship, and economic growth, Cato Journal, 25 (3), Koenker, Roger W and Gilbert Jr. Bassett (1978), Regression Quantiles, Econometrica Econometric Society, 46(1), Leeson Peter (2008), Media Freedom, Political Knowledge, and Participation, Journal of Economic Perspectives, 22(2), Lévesque Moren and Maria Minniti (2006), The Effect of Aging on Entrepreneurial Behavior, Journal of Business Venturing, 21(2), Lévesque, Moren, Minniti, Maria and Dean Shepherd (2009), Entrepreneurs' Decisions on Timing of Entry: Learning from Participation and from the Experiences of Others, Entrepreneurship Theory and Practice, 33(2), Minniti Maria and Moren Lévesque (2008), Entrepreneurial Types and Economic Growth Journal of Business Venturing, Forthcoming Minniti Maria (2004), Entrepreneurial Alertness and Asymmetric Information in a Spin- Glass Model, Journal of Business Venturing, 19, Minniti Maria and William Bygrave (2004), A Dynamic Model of Entrepreneurship Learning, Entrepreneurship: Theory and Practice, Southern Journal of Entrepreneurship

17 Reynolds, P. D., Hay, M., and Camp, S. M. (1999), Global entrepreneurship monitor, Kansas City, Missouri: Kauffman Center for Entrepreneurial Leadership. Schumpeter, Joseph A. [1911] 1934, The Theory of Economic Development, Cambridge: Harvard University Press. Sen, Amartya (1982), Poverty and Famines: An Essay on Entitlements and Deprivation, Oxford, Clarendon Press, Sen, Amartya (1999), Development as Freedom, Oxford: Oxford University Press. Shane, Scott and Sankaran Venkataraman (2000), The Promise of Entrepreneurship as a Field of Research. Academy of Management Review, 25(1): Sobel, Russell S. (2008), Testing Baumol: Institutional quality and the Productivity of Entrepreneurship, Journal of Business Venturing, 23 (8), Sobel, Russell S. (2005), Entrepreneurship, in David R. Henderson (editor), The Concise Encyclopedia of Economics, Indianapolis: Liberty Fund, forthcoming. Spitzer, R. J., (1993), Introduction: Defining the Media-Policy Link, in R. J. Spitzer (ed.), Media and Public Policy. Westport, CT: Praeger. Stiglitz Joseph (2002), Transparency in Government, in R. Islam ed. The Right to Tell: The Role of the Media in Development. Washington, DC: The World Bank. The Founders' Constitution, Volume 5, Amendment I (Speech and Press), Document 8 The University of Chicago Press. The Papers of Thomas Jefferson. Edited by Julian P. Boyd et al. Princeton: Princeton University Press, Weingast, Barry R., Kenneth A. Shepsle, and Christopher Johnsen (1981), The Political Economy of Benefits and Costs: A Neoclassical Approach to Distributive Politics, Journal of Political Economy, 89(4): World Bank (2006), World Development Indicators, CD ROM. World Bank Entrepreneurship Survey, 2008 database Zacharakis, A. L., Bygrave, W. D., and Shepherd, D. A. (2000). Global entrepreneurship monitor: national entrepreneurship assessment, United States of America. Kansas City, Missouri: Kauffman Center for Entrepreneurial Leadership. Southern Journal of Entrepreneurship 87

18 Table 3: Robust Regression Estimates of the Impact of Press Freedom on Business Density and New Business Density 1 Dependent Variable: Business Density New Business Density Independent Variable: (1) (2) (3) (4) Press Freedom *** *** * (0.0266) (0.0292) ( ) ( ) Total Working Age Population 0.360** 0.418** *** *** (0.161) (0.178) ( ) (0.0142) GDP per cap 2 (constant US $ 2000) Male Labor Force Participation Rate 0.62*** 0.65*** 0.06*** 0.06*** (0.09) (0.09) (0.06) (0.08) *** *** *** (0.130) (0.142) ( ) (0.0126) Financial Development (0.0162) (0.0179) ( ) ( ) FDI * * *** *** (0.0164) (0.0177) ( ) ( ) Economic Freedom 0.428*** 3.563*** *** 0.270*** (0.0785) (1.138) ( ) (0.0941) Constant * (18.41) (20.58) (1.101) (1.821) Observations R-squared F statistic Notes: Standard errors in parentheses; economic freedom variable in columns 1 and 3 is the Heritage Institute index and in columns 2 and 4 is the Fraser Institute index; significance levels: *** = 1%, ** = 5%, * = 10%. 1 The regressions include time dummies that have not been reported in the table 2 GDP per capita is in constant 2000 US $ (billions). 88 Southern Journal of Entrepreneurship

19 Table 4: Pooled Ordinary Least Squares Estimates of the Interactive impact of Press Freedom and Economic Freedom on Entrepreneurship 1 VARIABLES Economic Freedom (Fraser) Economic Freedom (Heritage) (1) (2) (3) (4) Business New density ss density New density density Press Freedom *** *** *** *** (0.377) (0.0712) (0.250) (0.0442) Economic Freedom *** *** ** (2.748) (0.411) (0.234) (0.0304) Interaction 0.286*** *** *** *** (0.0570) (0.0109) ( ) ( ) GDP per cap (constant US $ 2000) ( ) (2.51e-05) ( ) (2.30e-05) Male Labor Force Participation ** * Rate (0.246) (0.0304) (0.228) (0.0211) FDI ** ** * (0.0174) ( ) (0.0147) ( ) Total Working Age Population 1.386*** 0.148*** 1.151*** 0.114*** (0.208) (0.0245) (0.194) (0.0169) Financial Development (0.0322) ( ) (0.0330) ( ) Constant (33.22) (4.360) (32.50) (2.452) Observations R-squared F statistic Notes: Standard errors in parentheses; significance levels: *** = 1%, ** = 5%, * = 10%. 1 The regressions include time dummies that have not been reported in the table Southern Journal of Entrepreneurship 89

20 Appendix 1: List of Countries in Sample Albania Greece Oman Algeria Guatemala Pakistan Argentina Haiti Peru Armenia Hong Kong Philippines Australia Hungary Poland Austria Iceland Portugal Azerbaijan India Romania Bangladesh Indonesia Russian Federation Belgium Ireland Senegal Bolivia Israel Singapore Bosnia and Herzegovina Italy Slovak Republic Botswana Jamaica Slovenia Brazil Japan South Africa Bulgaria Jordan Spain Burkina Faso Kazakhstan Sri Lanka Canada Kenya Sweden Chile Kyrgyz Republic Switzerland Colombia Latvia Syrian Arab Republic Republic of Congo Lebanon Tajikistan Croatia Lithuania Tanzania Cyprus Luxembourg Thailand Czech Republic Madagascar Tunisia Denmark Malaysia Turkey Dominican Republic Malawi Uganda Ecuador Maldives United Kingdom Egypt Malta Ukraine El Salvador Mexico United States Finland Morocco Uzbekistan France Netherlands Vietnam Georgia New Zealand Yemen, Rep. Germany Nicaragua Zambia Ghana Norway 90 Southern Journal of Entrepreneurship

21 Appendix 2: Data description and Sources Variable Name (source) Description Mean (Std.Dev.) Dependent Variables: Business Density (1) Number of total registered corporations divided (26.85) by total working age population New Density (1) Number of newly registered corporations divided by the total working age population 2.24 (3.16) Independent Variables: Press Freedom (2) Economic Freedom Index (3) Economic Freedom Index (4) Working Age Population (5) Male Labor Force Participation Rate (5) GDP per capita constant (5) FDI Inflows (5) Financial Development (5) Financial Development (6) An index representing the dependence of the press on government. The level of press freedom in each country is based on twenty three methodology questions divided into three categories: the legal environment, the economic environment and the political environment. It ranges from 0 to 100 which higher scores representing worse conditions. The scores have been re-scaled for this study. An index where higher scores represents better conditions. The index ranges from 0 to 10 with higher scores signifying better situation. The Index of Economic Freedom is built upon analysis of 10 specific components of economic freedom, some of which are themselves composites of additional quantifiable measures. The 10 component scores are equally weighted and averaged to get an overall economic freedom score for each country. The index ranges from 0 to 100. Population ages as a percentage of total population (%*100) Male labor force participation rate as a percentage of male population aged (%*100) GDP per capita (constant 2000 US $). Reported in billion US $. Net inflows of foreign direct investment as a percentage of GDP (%*100) Domestic credit provided by banking sector as a percentage of GDP (%*100) Private credit by deposit money banks and other financial institutions to GDP Sources: 1. World Bank Entrepreneurship Survey, 2008 Database 2. Freedom House 3. Fraser Institute 4. Heritage Foundation 5. World Development Indicators (2007) Database 6. Beck, Demirgüç-Kunt and Levine (2000) Database (26.27) 6.70 (0.88) (10.09) (5.92) (6.09) 8.82 (1.14) 6.99 (33.15) (57.16) 0.57 (0.47) Southern Journal of Entrepreneurship 91