International Financial Integration and Entrepreneurship

Laura Alfaro Andrew Charlton Harvard Business School London School of Economics

June 5, 2006 (PRELIMINARY)

Abstract

We explore the relation between international financial integration and the level of entrepreneurial activity in a country. Researchers have stressed the role of new firm activity and economic dynamism on growth. Yet, the empirical effects of international capital mobility on entrepreneurial activity have received little attention in the literature albeit the ambiguous theoretical predictions and the intense policy debate. Using a unique comprehensive data set of nearly 24 million firms in a close to 100 developed and developing countries in 2004, we find higher entrepreneurial activity in countries with fewer restrictions to international capital flows. Our results are robust to using other proxies for entrepreneurial activity such as entry, skweness of the firm-size and firm-age distribution and controlling for other variables that might affect the business environment such as the size of the market, regulation, enforcement of property rights and other institutional variables. We also use different empirical specifications in our analysis. We further explore various channels through which international financial integration can affect entrepreneurship and provide consistent evidence to support our results. In particular, we find that the presence of downstream foreign firms has a positive effect on firm activity (a foreign direct investment channel) and we find higher entrepreneurial activity in sectors more reliant on external finance (a capital/credit availability channel). Overall our study provides suggestive evidence that international financial integration has been associated with higher levels of entrepreneurial activity. JEL Classification: F21, F23, F35, G15, G18, O19. Key Words: international financial integration, capital mobility, entrepreneurship, firm entry, capital controls, foreign direct investment. * Laura Alfaro, Harvard Business School, Morgan 263, Boston MA, 02163, U.S. (e-mail: lalfaro@hbs.edu). Andrew Charlton, London School of Economics, Houghton Street London, WC2A 2AE, U.K (e-mail: a.charlton(@)lse.ac.uk). We thank Dennis Jacques for helping us with the D&B data set and HBS and LSE for financial support. We further thank Pamela Arellano for excellent research assistance.

1

1. Introduction

In this paper, we explore the relation between international financial integration --a country’s

linkages to international capital markets-- and the level of entrepreneurial activity in a country. Researchers

have stressed the role of entrepreneurship, new firm activity and economic dynamism on economic growth.1

The empirical effects of international capital mobility on firm dynamism and entrepreneurial activity,

however, have received little attention in the literature albeit the intense academic and policy debates. Using

different measures commonly employed in the literature in a new data set of more than 24 million firms in

close to 100 developed and developing countries in 1999 and 2004, we find higher entrepreneurial activity

in more financially integrated countries and countries with fewer restrictions to international capital flows.

The theoretical effects of international financial integration on entrepreneurship are ambiguous.

The rapid rate of global financial integration, seen most directly through the explosive growth of foreign

direct investment (FDI), has raised concerns in both the public and academic communities about the

potential negative effects of international capital on the development of domestic entrepreneurs with

negative consequences to the economy as a whole. It has been argued that foreign enterprises crowd out

local efforts, and thus impart few, if any, benefits on the local economy. Grossman (1984) shows, for

example, that international capital flows, and in particular FDI, can lead to a crowding out of the domestic

entrepreneurial class.2 Hausmann and Rodrik (2002) argue that laissez-faire and in particular openness can

lead to too little investment and entrepreneurship ex-ante.3 Similar concerns were raised by the development

literature. Hirschman (1958), for example, warned that in the absence of linkages, foreign investments can

have negative effects on an economy (the so called 'enclave economies’). More generally, researchers have

argued that in the presence of pre-existing distortions and weak institutional settings, international capital

1 Entrepreneurship and firm creation are often described as the keys to economic growth (Schumpeter 1942). Lewis (1955, p. 182) for example notes that “economic growth is bound to be slow unless there is an adequate supply of entrepreneurs looking out for new ideas and willing to take the risk of introducing them. Thus private enterprise economy will be retarded if it has not enough businessmen, or if its businessmen are reluctant to take risks, whether because they cannot raise the capital or because they are timid by nature or because the differentials for risks taking are inadequate.” See Aghion and Howitt (1998) for an exhaustive survey of Schumpeterian growth models. 2 In addition, if foreign firms borrow heavily from local banks, instead of bringing scare capital from abroad, they may exacerbate domestic firms’ financing constraints by crowding them out of domestic capital markets. See Feldstein (2000), Harrison, Love and McMillian (2004) and Harrison and McMillian (2003). 3 Emran and Stiglitz (2005) argue that competition and financial liberalization can have a negative impact on the level of entrepreneurial activity.

2

mobility can increase the likelihood of financial crises; higher volatility and risk which can reduce

entrepreneurship and innovative efforts in a country.4 As Eichengreen (2001) notes “capital account

liberalization, it is fair to say, remains one of the most controversial and least understood policies of our

day.”

On the other hand, access to foreign resources can enable developing countries with little domestic

capital to borrow to invest, and resource constrained entrepreneurs to start new firms. Availability of funds

has been found to be an important determinant of entrepreneurship.5 International financial integration can

also lead to the improvement of domestic financial sectors through increased competition with further

potential benefits to entrepreneurship.6 Financial integration should also facilitate international risk sharing,

thus lowering the cost of capital for many developing countries.7 Moreover, researchers have stressed the

potential positive role of knowledge spillovers and linkages from foreign firms on domestic firm activity

and innovation.8

A great example is given by the on-going debate in the Irish economy about the impact of foreign

capital flows and in particular FDI on local entrepreneurial efforts. Given the limited size of the indigenous

sector, one concern has been the potential crowding out of domestic entrepreneurship due to competition in

the labor and product markets. Others contend that local entrepreneurs have benefited from foreign capital

and in particular from interacting with foreign firms as suppliers or costumers or from previous experience

working on multinational firms.9

4 Some scholars assert that open capital markets may be detrimental to economic development. See Bhagwati (1998); Rodrik (1998), Stigliz (2002). 5 Evans and Jovanovic (1989) show theoretically that wealth constraints negatively affect entrepreneurship. Evans and Leigthon (1989) find evidence that credit constraints are a critical factor in the founding and survival of new firms. 6 See Levine (2005). 7 The increased risk sharing opportunities between foreign and domestic investors might help to diversify risk which in turn may encourage entrepreneurs take on more total investments or attempt new ventures, see Obstfeld (1994), Acemoglu and Zilibotti (1997). 8 See Markusen and Venables (1999) and Rodriguez-Clare (1996) for a formalization of the linkage effects from multinational corporations to domestic firms. Markusen and Venables (1999) propose a model which suggests that FDI will be associated with firm turnover. In their model the entry of foreign firms initially increases competition and forces the exit of domestic firms. However in the longer run multinationals may stimulate local activity through linkages with the rest of the economy. 9 A recent study of the Ireland’s indigenous software sector, for example, finds that one-third of entrepreneurs had worked in foreign firms immediately before the start-up of the new firm, and two-thirds had worked in foreign firms at some point their career, see O’Gorman et al. (1997) and Alfaro et al. (2005).

3

Whether international capital mobility is fostering or destroying entrepreneurship is a critical

question in academic and policy circles. Yet, empirical analysis of the effects of international capital

mobility on entrepreneurial activity and firm dynamism are all but absent in the literature. This is largely a

function of the difficulty of obtaining an international data set sufficiently comprehensive to support studies

of firm dynamism in both developed and developing countries. We overcome this problem by using a new

data set of private firms in 98 developing and developed countries in 1999 and 2004. Our data set contains

both listed and unlisted firms and contains more than 24 million firm observations across a broad range of

developed and developing countries at different stages of international financial integration enabling the

study of the differential effects of restrictions to capital mobility on entrepreneurial activity. Over the last

decades, barriers to international capital mobility have fallen in developed countries and diminished

considerably in many developing countries. Despite the recent trends, restrictions to international financial

transaction are still quantitatively important for many countries and de facto flows remain low relative to

those predicted by standard models in particular for developing countries.10

Identifying the effects of international financial integration on entrepreneurial activity, however, is

not an easy task. There is not one definition of what entrepreneurship is or entitles and hence not one

variable to measure it.11 Hence, we analyze a variety of measures commonly identifying and used in the

literature as proxies for various aspects related to entrepreneurial activity. We focus on firm entry, average

firm size and skewness of the firm-size as these measures better capture firm activity. We also study other

measures used in the literature such as age, the skewness of the firm-age distribution and vintage (a size

weighted measure of the average age of the firm).12 The literature has distinguished between de jure

indicators of financial integration which are associated with capital account liberalization policies and de

facto indicators which are associated with actual capital flows.13 We use both measures as they capture

different aspects of international capital mobility and financial integration. We also control for other

10 See Table 3 for stylized facts. See Alfaro, Kalemli-Ozcan and Volosovych (2006) for an analysis of the main trends related to international capital in the last thirty years. 11 There are different views the literature about the role of entrepreneurs emphasizing a broad range of activities including the introduction of innovation (Schumpeter, 1942), the bearing of risk (Knight 1921, Drucker 1970), the bringing together of factors of production (Say, 1803), and the creation of creation of organizations (Gartner, 1988). 12 See Desai, Gompers, and Lerner (2003) and Klapper, Laeven, and Rajan (2005) and Black and Strahan (2002). 13 See Prasad et al (2003) Edison el al. (2000) for a discussion of the different indices and measures used in the literature.

4

determinants found in the literature to affect the level of entrepreneurship such as the local development

level and institutional constraints. We use industry fixed effects to control for technological determinants of

entry into an industry.

We first study the cross-section properties of our sample in 2004. We find positive correlations

between the different measures of international financial integration and the different measures of

entrepreneurial activity in a country. More firm creation is observed in more integrated countries and

countries with fewer levels of restrictions to capital mobility. In particular, more capital controls seem to

reduce firm entry. Firms also tend to be younger in more financially integrated countries. Our results are

both statistically and economically significant. In our analysis, we pay particular attention to role of FDI. An

advantage of our data set is that it allows us to distinguish between foreign and local firms. We obtain that

higher foreign firm activity is correlated with higher firm dynamism in the host country.

As mentioned, our data set allows studying the determinants of the business environment in a broad

sample of developed and developing countries. In line with the literature, we find variables related to the

regulation of entry such as days to start a business to negatively affect entrepreneurial activity; non-

corruption on the other hand, has a positive and significant effect on the dynamism of the economy.

Nevertheless, in terms of our research question, when we control for these other determinants of

entrepreneurship, the measures of international financial regulation remains positive and significant.

Our results are robust to different measures and specifications. We compare our results for 2004 and

1999 using a difference in difference approach obtaining similar results. In addition, we follow the

methodology of Rajan and Zingales (1998) and Klapper, Laeven and Rajan (2005) and focus on cross-

industry, cross-country interaction effects (is entry more likely in an industry with a particular need when

the host country scores strongly on a characteristic that facilitates that need?). Following these authors, we

use the Unites States as a proxy for the “natural” rate of entry in an industry. We test then whether entry is

relatively higher or lower in a naturally-high-entry industry when the country has relatively high

international capital mobility. The results confirm our main findings.

The nature of our data allows us to explore some of the channels through which these benefits might

materialize. First, international financial integration might increase the total amount of capital in the

5

economy and improve the intermediation of capital (a capital/credit availability channel). Although small

firms might not be able to borrow directly in international markets, improved financial intermediation and

other firms’ international borrowing (and the government’s) might ease financing constraints until some of

the additional capital finds its way to new firms.14 Second, local firms might benefit from spillovers and

linkages from foreign firms (FDI channel).15 We test for the first channel by exploring whether there is

higher entry in firms more dependent on external finance as defined by Rajan and Zingales (1998). The

evidence suggests that, indeedm we observe greater entrepreneurial activity in more dependent on external

finance sectors. In terms of the FDI channel, we regress our entrepreneurship measures on the share of

foreign ownership in the same industry. We also test whether our measures of entrepreneurship are

correlated with the presence of multinational firms in downstream and upstream sectors. Given the difficulty

of finding input and output matrices for all the countries in our data, we follow Acemoglu, Johnson and

Mitton (2005) in using U.S. input and output matrices. Our results are consistent with our previous findings,

namely that foreign capital, in particular, FDI, leads to increased levels of entrepreneurship.

Important concerns in our analysis are related to policy endogeneity and omitted variables biases

issues in terms of establishing the causality between international financial integration and proxy variables

of entrepreneurial activity. Capital account liberalization and entrepreneurial activity may be positively

correlated with an omitted third factor. With respect to endogeneity of policy if, for example, a policy-maker

liberalizes the capital account in a country in anticipation of the improvements in external conditions, we

would then observe capital liberalization and firm creation.16 We take different steps to mitigate these

concerns. We control for other variables that might affect entrepreneurial activity. We believe the extensive

robustness analyses we perform eases concerns about potential omitted variables. We also look at different

proxies for entrepreneurial activity. We analyze firm characteristics as opposed to country characteristics

and test effects controlling for the different sectors. Even if average firm growth is correlated because of an

14 International financial integration can also lead to the improvement of domestic financial sector and risk sharing. Beakert and Harvey (2000) and Henry (2000) find evidence that stock market liberalizations lead to a reduction in the cost of capital. 15 Gorg and Strobl (2002) find that foreign presence reduces encourages entry by domestic-owned firms in the high-tech sector in Ireland. Javorkic (2004) finds that FDI fosters spillovers through backward linkages in Lithuania although her work does analyze firm entry patterns. 16 See Rodrik (2005) for a recent critical analysis on regressing economic outcomes on policy variables.

6

omitted common factor, it is hard to argue that this omitted common factor affects the relation between

capital flows and entrepreneurial activity in a systematic away for firms with different characteristics. We

control for exogenous growth using growth forecast from the Economist Intelligence Unit (EIU). This

measure captures imperfectly country’s growth opportunities that are plausibly exogenous, thus providing

an additional robustness check for possible endogeneity related to omitted variables. As an imperfect

attempt to account for possible endogeneity biases, we also use institution-based instruments for financial

integration from La Porta et al (1998), which have been used in the literature for international financial

liberalization and domestic financial development.17 This instrumentation strategy yields similar results and

confirms that our results are quite robust. Finally, we feel more comfortable in interpreting our correlation

as causation in as much as mechanisms consistent with such an interpretation are supported by the empirical

evidence. However, even after all of these tests, our estimates should be interpreted with caution.

We noted earlier the scarcity empirical work on the effects of international capital mobility the

entrepreneurial activity in a country. Several papers, however, have studied how different aspects of capital

account liberalization affect a firm’s financing constraints and the cost of capital. Chari and Herny (2004)

examine the effect of stock market liberalization in different type of firms in 11 emerging markets. The

authors find that when publicly-listed firms become eligible to foreign ownership they experience a

significant average stock price revaluation and a significant fall in the average cost of capital. Harrison,

Love and McMillian (2004) find FDI inflows to be associated with a reduction in firms’ financing

constraints while restrictions on capital account transactions negatively affect their financial constraints.18

Our results are broadly consistent with these findings.

Our paper also relates to the research on the effects of the external environment on

entrepreneurship. Desai, Gompers, and Lerner (2003) and Klapper, Laeven, and Rajan (2005) have studied

different aspects of the external environment on firm creation and entrepreneurship on a cross-section of

European countries. Work on other aspects related to entrepreneurship include Kumar, Rajan, and Zingales

17 See Imbs (2004), Kalemli-Ozcan, Sorensen, and Yosha (2003). 18 The authors use firm level data for 38 countries and 7079 firms from the Worldscope data base, which contains data on large publicly traded firms in which there is an investor interest. Foreign ownership data is obtained from Amadeus-Bureau Van Dijk database and Dun and Bradstreet’s. In contrast, Harrison and McMillian (2003), find that in the Ivory Coast for the period 1974-1987 borrowing by foreign firms aggravated domestic firms’ credit constraints.

7

(1999) finding that the average size of firms in industries dependent on external finance is larger in countries

with better financial markets, suggesting that financial constraints limit the average firm size; Beck,

Demirguc-Kunt, Laeven and Levine (2006) finding that financial development exerts a disproportionately

positive effect on small firms; Johnson et al. (2002) finding that investment by entrepreneurs is lower in

countries with weak property rights; Black and Strahan (2002) and Guiso, Sapienza and Zingales (2004)

finding that financial development fosters firm entry; Giannetti and Ongena (2005) study of the effects of

foreign bank lending on Eastern European firms’ firm growth; and Fisman and Sarria-Allende (2005) study

of the effects of regulation on entry on the quantity and average size of firms.19 Our work is also related to

cross country comparative studies in finance focusing on concentration, firm size and entry, Kumar, Rajan,

and Zingales (1999), Acemoglu, Jonhson and Mitton (2005) and Beck, Demirguc-Kunt, Laeven and Levine

(2006). Most of these papers, with the exception of the latter two and the Fisman and Saria-Allende (2005)

paper, use data from the Amadeus dataset for Western and Eastern Europe, which include countries’

information only for relatively large public trade firms. Our paper contributes to this literature by exploring

the determinants of entrepreneurship and firm dynamism in a broader sample of developed and developing

countries using data for both private and public firms.20 Our data sheds light on an important number of

countries for which consistent data is not available.

Finally, by focusing on the micro effects, our results contribute to the broader debate on the effects

of international financial integration. As argued by Schumpeter, firm entry is a critical part of the dynamism

of an economy. Previous work has documented the important effects of new firm entry and economic

dynamisms on economic growth. Obstacles to this process can have severe short and long run

macroeconomic consequences. Our results suggest that contrary to the fears of many, capital mobility has

not hindered entrepreneurship. On the contrary, our results suggest that international financial integration

has been associated with more firm activity and entrepreneurship.

The rest of the paper is organized as follows. Section 2 describes the data. Section 3 presents the

main empirical results. Section 4 discusses potential channels. The last section concludes. 19 Fisman and Sarria-Allende (2005) use industry data from UNIDO and complement with World Scope data for a sample of 34 countries. 20 Bartelsman, Haltiwanger and Scarpetta (2004) provide evidence on the process of creative destruction across 24 countries and 2 digit industries.

8

2. Data and Descriptive Statistics

2.1. Firm Level Data

The data used in this paper are from the Worldbase data. WorldBase is a database of both public and

private companies in more than 213 countries and independent territories.21 For each firm, WorldBase

reports the 4-digit SIC-1987 code of the primary industry in which each firm operates and for a few

countries the SIC codes of up to 5 secondary industries, listed in descending order of importance. The data

in Worldbase database is complied by Dun and Bradstreet from a number of sources with a view to

providing its clients with contact details and basic operating information about potential customers,

competitors and suppliers. Sources include partner firms in dozens of countries, from telephone directory

records, websites, and firms that self-register.22 All information is verified centrally via a variety of manual

and automated checks. Information from local insolvency authorities and merger and acquisition records are

used to track changes in ownership and operations.

Another advantage of the Worldbase is that its unit of record is the ‘establishment’ rather than the

firm. Establishments like firms have their own addresses, business names, and managers, but might be

partly or wholly owned by another firm. Thus our data is able to capture new entrepreneurial ventures

owned and capitalized by existing firms as well as by private entrepreneurs. We are thus able to construct a

more complete picture of how firms respond to capital market liberalization across both their intensive

margin (expansion of existing activities) and extensive margin (establishment of new activities).

We use data for 2004 and exclude information for which primary industry and year started

information were missing.23 We also excluded territories with less than 80 observations and those for which

the World Bank provides no data and for which capital market integration information was not available.

We do not consider government related firms (SIC 9X) in the analysis. With these restrictions, our final data

21 Even for the United States, publicly listed firms account for only 25% of jobs in the United States. Although it is hard to quantify this number for our broad sample of countries, presumably the importance of publicly traded firms is much more in US than most other countries. See Davis, Faberman, and Haltiwanger (2006) for differences in patterns in the US between publicly traded and private firms. 22 Firms self-register to receive a widely recognized DUNS business identification number. 23 We use the 2005 edition. Due to lags in data collection, the coverage for 2005 is incomplete. We also use information for 1999 (close to 6 million observations) in the difference-in-differences section and in the robustness. We performed a similar analysis with these data obtaining similar results, not reported due to space considerations.

9

set includes more than 24 million observations in 98 countries.24 In the data appendix we explain in detail

the criteria used to clean the sample. Table 1 presents the list of the countries in the data set.

2.1.1. Sample Frame

The number of observations per country in the Worldbase ranges from more than 7 million firms in

the USA to less than 90 in Burkina Faso (see Table 1). This variation reflects differences in country size, but

also differences in the intensity with which Dun and Bradstreet sample firms in different countries. This

raises concerns that our measures of entrepreneurship are affected by cross-country differences in the

sample frame. For example, in countries where coverage is lower, older and larger firms may be

overrepresented in the sample. This may bias our results if the country characteristics which determine the

intensity of sampling are correlated with our explanatory variables.

We address this concern in several ways. We compare our results for 2004 and 1999 and study how

changes in our measures of entrepreneurship between the two time periods are related to changes in capital

restrictions and capital mobility.25 This gives us more confidence that our results are not driven by the

sample frame, although it is still possible that changes in sampling procedure are correlated with changes in

financial integration over the same period. Second we repeat our specifications for subsamples which

include only the rich countries which are the most intensively sampled by D&B. Third, we deal with the

possibility that our results are driven by outliers in country/industry pairs with a small number of

observations by running regressions on medians rather than mean characteristics of industries, and by

excluding outlier observations. Fourth we include a measure of country sampling intensity26 in our

regressions and find that our results are robust.

Finally it is worth noting that the variety of sources from which the data are collected avoids a

sample selection problem present in previous studies. Many international databases collect firm data from

24 Acemoglu, Johnson and Mitton (2005) use the 2002 Worldbase file. They use all firms in 93 countries which report secondary industries, with a maximum of 30,000 per country (a limit imposed due to cost constraints). For those countries with more than 30,000 firms, the 30,000 largest were selected, ranked by annual sales. Their base sample consists of 769199 firms in 93 countries. 25 One potential concern is that data set is for a good year in the international arena --2004--. As well as analyzing the differences between the two time periods, we rerun our cross-section specifications with our 1999 data set (post Asian and Russia crisis, during Brazil crisis before Turkey and Argentina) obtaining similar results. 26 The ratio of number of firms in the database to GDP.

10

national authorities. Sample coverage that draws on such sources will thus vary across countries depending

on the parameters of the national statistical agency’s reporting requirements. The sample of firms entering

the database from different countries is therefore not random, but is determined by the local institutional

environment. These reporting requirements may be correlated with other national characteristics in our

regressions, potentially biasing the results. By contrast the Dun and Bradstreet data are collected from a

wide variety of sources including self-registration which reduces the scope for the sample frame to be

determined by national institutional characteristics. In the Data Appendix we compare the D&B data to the

U.S. (for which there is more wide coverage). The comparison illustrates that our data set seem to be well

suited for our analysis.

2.2 Entrepreneurship Measurements

How to measure entrepreneurship?27 Given the different perspectives in the literature on the role of

entrepreneurs in an economy, definitions have emphasized a broad range of activities including the

introduction of innovation (Schumpeter, 1942), the bearing of risk (Cantillon circa 1730, Knight 1921), the

bringing together of factors of production (Say, 1803), the creation of creation of organizations (Gartner,

1988), the pursuit of opportunity without regard to resources currently controlled (Sahlman et al. 1999). In

general, entrepreneurs are risk-bearers, coordinators and organizers, gap-fillers, leaders, and innovators or

creative imitators. Overall, entrepreneurship is often viewed as the exploitation of opportunities which exist

within a market and the establishment of a new venture/firm while adopting some of the risk

If there is no one way to define entrepreneurship, there is certainly no one way to measure it. Hence, we

use a variety of indices commonly used in the literature to proxy for entrepreneurship. Following Black and

Strahan (2002), Desai, Gompers, and Lerner (2003) and Klapper, Laeven, and Rajan (2005), we calculate

for each industry within a country rate of entry, average firm size, age, the skewness of the firm size and age

27 The dictionary defines en·tre·pre·neur as a person who organizes, operates, and assumes the risk for a business venture. [French, from Old French, from entreprendre, to undertake.]

11

distributions, and vintage.28 In the regression analysis we focus on the latter measures as they represent the

full characterization of the firm distribution. We focus our analysis at the 2 industry digit level.

i. Firm Size Distribution: Small firms play an important role in the economy as they are often

portrayed as a source of innovation, regeneration, change and employment. The fullest characterization of

industrial activity is given by the firm size distribution. We examine whether skewness of firm-size

distribution is related of the presence of institutional determinants of capital constraints. If capital

constraints are operative in shaping the nature of industrial activity, overall firms-size distribution and firm

size distribution of the younger cohorts should to be skewed.29 We also calculate the average firm size

measured by the average number of employees.30 Our size variable is the log of this average for each

country/industry pair. Note that this measure does not distinguish between small old firms and small young

firms. While the prediction is not unambiguous, we might expect that lower levels of capital rationing

associated with international financial integration would result in greater numbers of small firms being able

to enter and survive in the market.

ii. Firm Entry: Markets that provide an opportunity for more startup firms are said to be more

dynamic and entrepreneurial. We measure firm entry relative to the number of firms that existed in the prior

year.31 Percentage of firm entry is defined then as the number of new firms in t divided by the total number

of firms that existed in year t-1 in the country/industry pair. Greater access to capital and improvement in

financial markets associated with international financial integration should ease capital constraints and

positively influence entry decision in a country.32

iii. Firm Age Distribution: In the robustness section we also use as proxy for entrepreneurship the

skewness of the age firm distribution. In particular, young firms are said to invigorate an economy. We

28 The Global Entrepreneurship Monitor (GEM) has published some indexes on entrepreneurial activity. These data does not seem to be empirically consistent with other measures used in the literature and hence are not used in this paper. 29 Cooley and Quadrini (2003) and Cabral and Mata (2003) argue that in the presence of capital constraints, the firm size distribution will be skewed. In Cabral and Mata (2003), for example, firm growth depends upon investment and access to capital. Capital constraints affect younger firms that are likely to be capital rationed. 30 In this paper, we use firm and establishment interchangeably. 31 Due to lags in reporting and collecting, we classify a firm as firm as new if it less than two years old. See Klapper, Laeven and Rajan (2005) for a similar treatment. 32 This might depend on whether a country is exporting or importing capital, but there might still be an improvement in intermediation.

12

expect greater financial integration to be associated with more dynamic business environments and lower

average age of firms. We also use the average age of the firms for each country/industry pair.

iv. Vintage. In the robustness section we use vintage, a a weighed average measure of age.

Following Desai, Gompers and Lerner (2003) vintage is constructed within each country/industry pair and it

is the weighted average age of the firms in each industry, the weights being employees. This measure shows

how important young firms are to the productive capacity of an industry. Low vintage indicates that young

firms dominate the productive capacity. The predictions with respect to vintage are not unambiguous,

although we expect small and younger firms to benefit from greater access to international funds. Vintage

provides a broader measure of where the mass of industrial activity is over average firm size. As the authors

explain, if for example, lower capital constraints make it easier to get external financing at earlier stages, the

young sector would be smaller on average. At the same time, if firms are able to graduate to being older,

larger firms faster (or do not exit or fail as much) the fraction of actual employment in the young sector

would be lower. In this case, for example, average firm size would be lower and vintage would be higher.33

2.3. Capital Mobility Data

How to measure international financial integration? Assessing a country’s integration with

international financial markets is a complicated task. The process of international financial integration – that

is, the change in the degree to which a country’s government restricts cross-border financial transactions – is

complex and involves multiple phases. Markets may be liberalized gradually, or reforms can be anticipated

so the effects are smoothed.34 The literature has differentiated between de jure financial integration, which

is associated with policies on capital account liberalization and de facto measures related to actual capital

flows. De jure liberalization processes might not reflect the de facto liberalization process. If one part of the

system is liberalized, for example, investors may use it to circumvent other controls. Other reforms may not

be credible, or countries may not have access to foreign capital despite being officially open. Hence, we use

both de facto and de jure measures of financial integration.

33 In the Data Appendix we present a series of examples taken from Desai, Gompers and Lerner (2003) which help clarify the distinction between average firm size and vintage, 34 Anticipation and gradualness should bias our results away from finding an effect.

13

2.3.1 De Jure Measures

Most empirical analyses requiring a measure of capital account restrictions use an index constructed

from data in the International Monetary Fund’s (IMF’s) Annual Report on Exchange Arrangements and

Exchange Restrictions (AREAER). This is a rule-based indicator, that is, it focuses on de jure restrictions

imposed by the legal authorities in each country as compiled by the IMF’s AREAER. The index is

constructed from data on different restrictions presented in the survey appendix. In 1997 the IMF changed

the way they report the capital controls data.35 The new format provides greater detail than present prior to

1997. We use this new classification. The index uses data from different restrictions: capital market

securities; money market instruments; collective investment securities; controls on derivatives and other

instruments: commercial credits; financial credits; guarantees, securities and financial backup facilities;

controls on direct investment; controls on real estate transactions: controls on personal capital transactions.

A corresponding dummy variable takes the value of 1 if each of the restrictions is present in each country

each year and a zero otherwise. For each year, we average of the restrictions for each country. We use this

measure as a broader capital account indicator.36

2.3.2. De Facto Measures

Capital flows are usually divided into flows of foreign direct investment (FDI), portfolio equity

investment, and debt inflows. FDI data include greenfield investments (construction of new factories),

equity capital, reinvested earnings, other capital and financial derivatives associated with various

intercompany transactions between affiliated enterprises. Portfolio equity investment includes shares, stock

participations, and similar documents that usually denote ownership of equity. When a foreign investor

purchases a local firm's securities without a controlling stake, the investment is regarded as a portfolio

investment. FDI is equity participation giving a controlling stake. The IMF classifies an investment as direct

35 The 1997 IMF AREAR issue corresponds to the classification for 1996. 36 Note that the way the IMF index is constructed results in a general indicator that distinguishes varying intensities of capital restrictions in a very limited way. Although the new classification is a vast improvement over the previous measure, issues regarding circumvention of controls remain. Quinn (1997) improves the IMF restriction measure by reading through the IMF’s narrative descriptions and assigning scores of the intensity of capital restrictions. This measure is only available for intermittent years for some countries and not for our sample years, and as a result we do not use this indicator. See Edwards (2005) for criticisms of the IMF, AREAER index.

14

if a foreign investor holds at least 10 percent of a local firm's equity while the remaining equity purchases

are classified under portfolio equity investment. Debt inflows include bonds, debentures, notes, and money

market or negotiable debt instruments.

In the analysis, we use three main measures of de facto capital mobility.

i. Capital Inflows. Capital inflows to GDP are the sum of the change in the liabilities of FDI, equity

portfolio and debt from the IMF, IFS International Financial Statistics. For FDI, we use direct investment in

reporting economy (line 78bed). For portfolio equity investment, we use equity security liabilities (line

78bmd). For debt flows, we use debt security liabilities (line 78bnd) as well as other investment liabilities

(line 78bid). This measure allows capturing the new foreign capital available to the economy. Data is

calculated as a percentage of GDP in U.S. dollars (taken from the World Bank Development Indicators,

WDI).

ii. Inflows of Foreign Direct Investment: By using inflows of FDI as percentage of GDP we focus

on the potential benefits derived from FDI associated with technological transfers, spillovers and linkages

that go beyond the capital foreign firm bring into the country provides.37 Our data comes from the World

Bank, WDI.38

iii. Stock of Foreign Liabilities: In our analysis, the stock of foreign liabilities proxies the thickness

of the banking and equity relationships (both FDI and portfolio investment) with other countries. Hence, this

variable captures the effects on current entrepreneurial activity derived from existing foreign capital

relations– and not that past flows of foreign capital should affect current entrepreneurial activity. We take

the data from Lane and Milesi Ferretti (2005). Lane and Milesi-Ferretti (1999, 2001) (LM) construct

estimates of foreign assets and liabilities and their subcomponents for different countries in the 1970s,

1980s, and 1990s recently updated up to 2005 paying particular attention to these valuation effects.39 Data

is calculated as a % of GDP.

37 A limitation of this definition is that it may overestimate the amount of “new capital” in the economy, since it might simply involve a multinational enterprise buying out a local manufacturer. However, it is not clear to us that outward foreign investment should generate technological spillovers within the source economy. 38 The WB FDI inflows data correspond to the IMF, IFS, direct investment in reporting economy (line 78bed). 39 Lane and Milesi-Ferretti (1999, 2001) estimate stocks of portfolio equity and foreign direct investment based on the IMF, IFS flow data. In order to estimate FDI stocks, the authors cumulate flows and adjust for the effects of exchange

15

iv. Gross Capital Flows: Gross private capital flows to GDP are the sum of the absolute values of

direct, portfolio, and other investment inflows and outflows recorded in the balance of payments financial

account, excluding changes in the assets and liabilities of monetary authorities and general government. The

indicator is calculated as a ratio to GDP in U.S. dollars. Data comes from the World Bank, WDI. The trade

literature tends to use the sum of exports and imports to GDP as measure of openness. Similarly, gross

capital flows to GDP capture the overall foreign capital activity in the country.

In the robustness section we use as well the following measures.

v. Equity inflows (sum of foreign direct investment and portfolio inflows) and bank and trade

related lending which refers to commercial bank lending and other private credits calculated as a percentage

of GDP from WB, WDI. We use these variables to assess whether there is a different relation between

entrepreneurial activity and the different forms of capital flows.

vi. Net Capital Flows: Net flows to GDP allow us to focus on the net capital available to the

economy.40 We use a measure of net flows the sum of flows of foreign claims on domestic capital (change

in liabilities) and the flows of domestic claims on foreign capital (change in assets) in a given year. For FDI,

we use direct investment abroad (line 78bdd) and direct investment in reporting economy (line 78bed). For

portfolio equity investment, we use equity security assets (line 78bkd) and equity security liabilities (line

78bmd). For debt flows, we use debt security assets (IFS line 78bld) and debt security liabilities (line

78bnd) as well as other investment assets (line 78bhd) and other investment liabilities (line 78bid). Data is

calculated as a percentage of GDP.

2.4 Other controls

The literature has found variables such as the institutional and business environment as well as

industry characteristics to affect levels of entrepreneurial activity. The level of economic development is

rate changes. For portfolio equity stocks, they adjust for changes in the end of year U.S. dollar value of the domestic stock market. 40 We also used the measure of net capital flows constructed by the World Bank. Private debt flows include commercial bank lending, bonds, and other private credits, as well as foreign direct investment and portfolio equity investment. Result not reported since the World Bank re-evaluated the series in 2006 and took it out from the WD indicators.

16

likely to affect the attractiveness of becoming an entrepreneur. In the main specification we use the

(logarithm of) GDP per capita from the World Bank, World Development Indicators (WB, WDI) to proxy

for development. We also use the (logarithm of) GDP to control for scale effects and market size that might

affect entrepreneurial activity and the rate of GDP growth.

In the main specification we also control for institutional quality. We use data from the

International Country Risk Guide (ICRG), a monthly publication of Political Risk Services. In particular, we

use the variables non-corruption, law and order and bureaucratic quality which we expect to be positively

related to entrepreneurial activity.41 We also use the number of days required to start a business from the

World Bank, WDI. We expect this variable to have a negative impact on entrepreneurial activity. In the

robustness section we used additional proxies for institutional quality and regulation. We use additional

controls for regulation such as a business disclosure index and legal rights of borrowers and lenders index

taken from the WB, WDI.42 To control for financial development, we use Domestic Credit on GDP and the

ratio of M3 to GDP, both measures taken from WB, WDI. To capture uncertainty in the macro-economy,

we use inflation. Finally, we use the Economist Intelligence Unit’s growth forecasts as an imperfect control

for exogenous growth opportunities in the country. Detailed descriptions of all the data are included in the

Data Appendix.

3. Summary Statistics and Preliminary Results

Table 1 presents summary statistics by country for our main variables. For the U.S., for example, we

have more then 7 million firms followed up by France with more than 4 million. On the other hand, we have

87 firms for Burkina Faso and 99 for Zimbabwe. There is clearly wide variation in entrepreneurial activity

across countries. Countries like Denmark, Netherlands and South Korea exhibit high firm creation in 2003,

41 ICRG presents information for the following variables: investment profile, government stability, internal conflict, external conflict, no-corruption, non-militarized politics, protection from religious tensions, law and order, protection from ethnic tensions, democratic accountability, and bureaucratic quality. We do not use the whole index since our view is that some of these variables might have ambiguous or opposite effects on entrepreneurial activity (non corruption versus external conflict for example). In addition, we do not have a priori view on how some of these variables might affect entrepreneurial activity (democratic accountability for example). 42 Desai, Gompers, and Lerner (2003) and Klapper, Laeven, and Rajan (2005) use various measures from Djankov, et al. (2003) and Botero et al. (2002, 2003). These variables are not available for our sample period. We use similar variables as we were able to locate in other sources (WB, WEF, HF).

17

while Papua New Guinea and Yemen relatively low. Countries like Bulgaria and Estonia had high age

skewness while Kuwait and Cameroon low. Table 2 presents summary statistics by industry at the 2-SIC

code level. Table 3 presents summary statistics on de jure and de facto capital mobility for the countries the

sample. Netherlands and Zambia have the lower levels of de facto restrictions according to the IMF index,

while Zimbabwe and Vietnam the highest levels of restrictions. There is also widespread variability in

terms of the de facto flows of capital. Table 4 reports summary statistics for our main control variables. In

countries like Australia and Canada it takes from 2 to 3 days to start a business; in Brazil and India more

than 150 days. There is also great variation in terms of corruption and bureaucratic quality. Table 5 presents

the correlation matrix of the main variables. De jure restrictions to capital mobility are negatively correlated

with entry and positively correlated with age, size of the firm. Our data seems to be internally consistent but

also to be consistent with other studies in the literature regarding firm dynamics.43 For example, we observe

a negative relation between entry and young firms. Similar results are obtained with other measures.

4. Empirical Analysis

4.1 Cross sectional analysis

The purpose of cross-sectional analysis is to investigate whether there are variations in

entrepreneurial activity across countries which are correlated with capital mobility (de jure or de facto). We

run the following specification,

iciccic XK εδβα +++=Ε (1)

where icΕ corresponds the entrepreneurial activity measure in industry i of country c in period t, Kc

corresponds to measure of capital account integration, Xc corresponds to country level controls, iδ is a full

set of industry dummies, and εci corresponds to the error term. The industry dummies control for cross-

industry differences in the technological level or other determinants of entrepreneurship.44 Hence, in

43 See Bartelsman, Hatliwanger, Sarpetta (2004), Cabral and Mata (2003), Evans (1987), Sutton (1997). 44 Keppler and Graddy’s (1990) and Klepper and Simons’ (2000) results point to the importance of controlling for industry in the examination of firm entry and exit. Dunne and Roberts (1991) describe a certain industry characteristics

18

equation (1), we look at whether firms in a country with more capital mobility exhibit more entrepreneurial

activity relative to firms in a country with less capital mobility in the same industry. In other words, cross

country comparisons are relative to the mean propensity to “generate entrepreneurial activity” in an

industry.

We run specification (1) on the different measures of entrepreneurship: entry, firm size, and

skewness of the firm-size distribution and on different measures of capital account integration. For many

industries the rate of firm entry is zero or negligible. To account for this large number of zeros and our

upper bound at 1 we use a Tobit estimation model for the firm entry regressions. This specification allows

us to observe a regression line that is not heavily weighted by the large number of industries with a wide

range of characteristics but which did not generate any observed new firms in our sample period. Our main

control variables are (log of) GDP, (log of) GDP per capita, GDP growth, days to start a business, and

indexes of bureaucracy, non-corruption and law and order.

Table 6a-6c presents the main results of using both facto and de jure measure of capital account

integration. Overall our results suggest a negative and significant relation between different measures of

entrepreneurial activity and restrictions to capital mobility.

Table 6a presents results for firm entry as the dependent variable. In Column (1), the marginal effect

of the IMF index variable conditional on the dependent variable (the rate of firm entry) being uncensored is

-0.78.Again let's consider a movement from the 25th percentile (Ghana, 0.77) of the index to the 75th

percentile (New Zealand, 0.15) in the distribution of the index of restrictions, based on the results shown in

column (2), we have 0.48 percent more entry in an industry on average. This represents an 11 percent

increase in entry over average entry45 in an industry with average rates of entry like textiles and apparel with

which is economically significant. Columns (2)-(5) present the main result of controlling for de facto

measures of capital account integration. A movement from the 25th percentile (Tanzania, 0.34) of the Capital

Inflows/GDP variable to the 75th percentile (Denmark, 15.6) is associated, based on the results in Column

(2) with an increase in entry of 0.72 which represents a 16 percent increase in entry over average entry.

that explain much of inter-industry variations in turnover rates. They find the correlation between those industry characteristics and industry turnover pattern to be relatively stable over time. 45 Average entry in uncensored industries is 5.6 percent.

19

Similarly, based on the results in Column (3), an interquartile range movement in the FDI/GDP variable is

associated with an increase in FDI/GDP of 0.54 which is a 12 percent increase over the industry average.

In terms of the other control variables, our results are in line with the literature. GDP and market

size are positively related with entrepreneurship. We find significant positive of non-corruption and law and

order and bureaucratic quality. Days to start a business have a negative and significant effect on

entrepreneurship. To give some sense of the relative size of the effect of our capital mobility variable

relative to our controls, if we move up from the 25 percentile (UK) to the 75 percentile (Philippines) in the

distribution of the days to start a business variable (a difference of 36 days), based on the results shown in

column (2), we have 0.18 percent less entry in an industry on average.46 This represents a 4.0 percent

decrease in entry over average entry in an industry with average rates of entry like textiles and apparel – less

than half the effect of a similar interquartile change in the IMF index.

In Table 6b, the dependent variable is the log of employees in the industry/country pair. As seen in

column (1), an interquartile reduction in the IMF index (less restrictive controls) is associated with an

decrease in average firm size by 21 percent. A similar increase in the Capital Inflows/GDP and FDI/GDP

variables is associated with a decrease in the average firm size by 21 percent and 14 percent respectively.

In table 6c, the dependent variable is skewness of the firm-size distribution. We believe this variable

presents the fullest characterization of the firm activity in the economy. Column (1) of the table shows the

effect of the IMF index to be negative and significant on the skewness of the firm size distribution in each

industry. To get a sense of the magnitude of the effect of a reduction in the IMF index on the level of

entrepreneurial activity, let's consider a movement from the 25th percentile (Ghana, 0.77) of the index to the

75th percentile (New Zealand, 0.15) in the distribution of the index of restrictions, based on the results

shown in column (1) we have a fall in the skewness of –6.2, which represents 53% of the average industry

skewness. In terms of the effect of de facto measures of integration on the firm size distribution, a

movement from the 25th percentile (Tanzania, 0.34) of the Capital Inflows/GDP variable to the 75th

percentile (Denmark, 15.6) is associated, based on the results in Column (2), with an increase in skewness

of 5.0 which represents a 43 percent increase in skewness over the industry average.

46 In Column 1, the marginal effect of the IMF index variable conditional on the dependent variable being uncensored is -0.0077.

20

We performed additional robustness checks some of which we report on Appendix Table 1 and 2.

Our results are robust to using additional proxies for entrepreneurship, such as age, the firm-age skewness

and firm vintage (a size-weighted measure of average age) as well other measures of de facto financial

integration including equity flows (sum of foreign direct investment and portfolio flows) and bank and trade

credit and net capital flows. As the table the shows, our main results are robust to controlling for other

measures of regulation, and measures of macroeconomic instability and business cycles as well as the level

of domestic financial development.47 Appendix Table 3 shows are results to be robust to using only the

manufacturing sector, only developing countries and adding regional dummies.

Appendix A presents the results of using the methodology of Rajan and Zingales (1998) and focus

on cross-industry, cross-country interaction effects. Following these authors, we use the Unites States as a

proxy for the “natural” rate of entry in an industry. We test then whether entry is relatively higher or lower

in a naturally-high-entry industry when the country allows for international capital mobility. We present

results for the skweness of the firm size distribution. As seen in the Appendix Table A our main results

remain robust to using this methodology.

4.2 Difference in Differences

We compare our results for 2003/2004 and 1999/2000 using a difference in difference approach. We

focus on the skewness of the firm-size distribution, since as mentioned this variable provides a full

characterization of firm activity. We use an event study based on the difference in differences (DiD) method

(Card and Krueger [1994]). We measure the difference between the level of entrepreneurship in the two

periods for the group of countries which experienced liberalization in the interim, and for the control group

of countries which did not. The difference in differences is the difference between these two measures. This

model differences out all the individual characteristics of each observation and thereby controls for more

heterogeneity than the cross-sectional estimation. The model is:

47 We also controlled for education levels. Our main results remain robust to controlling for education levels. We do not the results as we do not have strong priori on how education levels may affect entrepreneurship. We also run our analysis in the robustness section using averages of the three previous years for net private capital flows and inflows of FDI in order to smooth any short-term business cycle effect. We do not report the results due to space considerations.

21

icIctic dDXLD νβγ +++=Ε (2)

Here L is a dummy indicating whether the country experienced liberalization as measured by a fall

in its IMF index. γ is the parameter of interest and captures the difference between the change in

entrepreneurship in liberalized countries and the change in control countries. The DiD estimator is given

by 0/1/ˆ == −= LicLic DEDEγ . The key identifying assumption in this model is that in the absence of

liberalization, both the liberalized and control observations would have experienced the same change in

entrepreneurship over the period. Given that we only have two periods of data our ability to test this

assumption is limited. For this reason we interpret the results from this specification with some caution. We

do however include differenced control variables.

Table 7 column (1) presents summary statistics which describe the differences between our 1999

and 2004 data. In particular, 56 per cent of countries in our sample had a lower IMF index in 2004 than

1999. Differences between the statistics summarizing the measures of entrepreneurship in the two samples

are generally small. Table 8, Column (1) present the main DiD results for de jure restrictions to international

financial integration suggesting a positive and significant relation between different measures of

entrepreneurial activity and capital mobility. In terms of economic magnitude Column (1) indicates that

industries in countries which liberalized in the period had a firm size distribution which, on average, had a

higher skewness than countries which did not liberalize by 3.8 equivalent to 33 per cent of the mean

industry skewness. Table 8 Columns (2-5) presents the main results from a related specification which, like

the DID method uses two time periods, but captures international financial integration with de facto

measures. Results suggest a positive and significant relationship between different measures of

entrepreneurial activity and de facto measures of capital flows. These results across two cross-sections of

the same data give us some confidence that our results are not driven by correlations between the sampling

intensity of our data provider and capital generating apparent correlations between observed industry

characteristics and capital mobility.

22

4.3 Endogeneity

Important concerns related to the previous findings are whether our measures of entrepreneurship

and international financial might be determined by a potential omitted third factor or that reverse causality

might be driving our results. To mitigate these concerns, we take different steps.

First, we believe the extensive robustness analysis undertaken in this section shows our

entrepreneurship measures and the international financial integration measures not to be determined by an

omitted third factor. Second, in addition to de facto measures, we use de jure measures which are less likely

subject to reverse causality.48 Third, we use as an imperfect proxy of forward-looking growth opportunities

growth forecasts by the EIU. Reassuringly, as shown in Appendix Table 2, column (6), our results are

robust to including this measure.

We also run IV regressions using instruments that are not subject to reverse causality and can

account for the institutional variation. La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997, 1998)

emphasize the importance of the legal origins on the current institutions. They examine the laws governing

investor protection, the enforcement of these laws, and the extent of concentration of firm ownership across

countries. They find that countries with different legal histories offer different types of legal protection to

their investors. Most countries' legal rules, either through colonialism, conquest, or outright borrowing, can

be traced to one of four distinct European legal systems: English common law, French civil law, German

civil law, and Scandinavian civil law. They show that countries whose legal rules originate in the common

law tradition offer the greatest protection to investors. As far as law enforcement is concerned, German civil

law and Scandinavian civil law countries emerge superior. The French civil law countries offer both the

weakest legal protection and the worst enforcement. These legal origin variables have been increasingly

adopted as exogenous determinants of for international financial liberalization and domestic financial

development.49 The last column in Appendix Table 3, presents IV results using legal origin variables.50

48 In particular, it may be possible that policy makers liberalize at a time when the world economy is booming or after they observe good economic outcomes. However, this does not seem borne out by the facts. Henry (2000b), for example, shows that countries do not pursue stock market liberalizations in response to investment booms. Using a probit analysis, Bekaert, Harvey, and Lundblad (2005) find that past GDP growth cannot explain liberalization. 49 See Imbs (2004), Kalemli-Ozcan, Sorensen, and Yosha (2003).

23

Albeit criticism of these instruments, overall, the instrumental variable regression support the conclusions

from the OLS regressions.

Finally, as explained before, in Appendix A we also follow the methodology of Rajan and Zingales

(1998) and Klapper, Laeven, and Rajan (2005) and focus on cross-industry, cross-country interaction

effects. This methodology, as explained by Rajan and Zingales (1998) and Klapper, Laeven and Rajan

(2005) enables us to address some issues associated with country effects.

Notwithstanding the battery of robustness tests, we acknowledge the difficulties of establishing

causation. At our most cautious, we can conclude that the within correlation of capital mobility and

entrepreneurship is positive and significant.51

5. Channels

The D& B data enables us to investigate several possible channels through which international

financial integration might affect entrepreneurial activity, i.e. does capital mobility affect entrepreneurship

through a change in the activity of foreign firms (a FDI channel) and related effects to domestic firms, or

through the availability of resources (a capital /credit availability channel)?

5.1 FDI Channel

We first test for the effect of international financial integration on entrepreneurial activity through

the effect that foreign firms (FDI) have on the creation of new firms.52 Initially we investigate the effects of

foreign firms on new firms in the same industry. In particular we run

iciciic XFirmsForeignofSharentry εδβα +++=Ε (3)

where Entryi refers to the percentage of new firms relative to the number of firms that existed in the previous

year in sector i in country c. The Share of Foreign Firms in sector i is number of foreign firms calculated as

50 The first stage results indicate that the legal origin variables are individually and jointly significant determinants of the IMF index. 51 In section 5 we present illustrate mechanisms consistent with such an interpretation. These tests allow us to be more confident of interpreting our results as causal. 52 As well as encouraging entrepreneurs by improving finance – foreign funds particularly FDI may increase entry by killing off unproductive domestic firms and creating opportunities for the entry of new firms in a restructured industry.

24

total firms in the industry i. Xx represent country-level controls. In Column 1 and 4 of Table 9 we find that

the presence of foreign firms has a positive effect on entrepreneurship in the same industry. An increase in

the share of foreign firms equivalent to moving from an industry in the 25th percentile of the distribution of

foreign presence to an industry in the 75th percentile is associated with an increase in the percentage of new

firms in the industry by 0.82 points, or an 18 percent increase over an industry with mean levels of foreign

firms. There is a large literature examining horizontal spillovers from FDI. For instance Caves (1974),

Blomstrom and Persson (1983), Blomstrom and Wolff (1994), Haskel, Pereira and Slaughter (2002) who

find a positive correlation between foreign presence and sectoral productivity and Haddad and Harrison

(1993), Aitken and Harrison (1999), and Djankov and Hoekman (2000) who find little evidence of

horizontal spillovers to domestic firms. The positive effects of FDI are often attributed by the replacement

effect of productive multinationals forcing domestic firms to exit. Both the positive and negative effects of

FDI are consistent with industrial restructuring, and ultimately, firm turnover. We find evidence that the

existence of multinational firms increases the rate of firm creation. This may reflect changes in the industry

as a result of large new entrants increasing their market share at the expense of some firms, and creating new

opportunities for others.53

We also test whether our measures of entrepreneurship are correlated with the presence of

multinational firms in upstream and downstream sectors.54 Given the difficulty in finding input and output

matrices for all the countries in our data, we follow Acemoglu, Johnson and Mitton (2005) and use U.S.

input and output (IO) matrices. As the authors explain, IO tables from the U.S. should be informative abou

input flows across industries in our different sample of countries as long as they are determined by

technology. For example, in all countries, car makers use tires, steel and plastic from plants specialized in

the production of these intermediate inputs. We use an IO table from the US Bureau of Economic Analysis.

We calculate the presence of downstream firms in industry i in country c as the

∑ ×=j

jcUSjic WZesenceStreamDown _Pr (4)

53 Note that our measure of firm creating encompasses both foreign and domestic firms. 54 See Javorkic (2004) for evidence that FDI fosters spillovers through backward linkages. See Alfaro and Rodriguez-Clare (2004) for a review of the literature.

25

where Zjc is the ratio of the inputs industry j in the US sources from industry i in the US to the total output

of industry i in the US from the US input output table. And Wjc is the total number of foreign firms in

industry j in country c as a percentage of the total number of firms in industry j in country c. Thus the

presence of foreign firms downstream from industry i is weighted by the volume of goods they purchase

from industry i. We estimate the following relation,

iciciic XecencemDownsStreantry εδβα +++=Ε Pr (5)

We also estimated a similar regression for upstream presence.

In Columns 2-6 of Table 9 we investigate the effect of forward and backward linkages on the

creation of new firms and the skewness of the firm-size distribution. Columns (2) and (5) show that the

effect of foreign presence on upstream industries is limited while columns (3) and (6) indicates that the

presence of downstream firms has a positive effect firm activity and on the creation of new firms. Our

results are also consistent with evidence on vertical spillovers from FDI. 55 Firm entry (and exit) may be

increased if multinational firms increase the demand for intermediate goods, or if they have higher

requirements for product standards and on-time delivery which create opportunities for new firms with

better technology or better operations. Foreign firms may also import more of their intermediate goods from

abroad, increasing competitive pressures on domestic suppliers. This is consistent with case study evidence

from Hobday (1995) who found that foreign investments in East Asia encouraged hundreds of firms to

supply them with components or assembly services. Also, our results are overall consistent with the

evidence of Harrison et al. (2004). The authors find that incoming FDI has significant impact on investment

cash flow sensitivities for domestically own firms and firms with no foreign assets. The authors argue that

their results are consistent with the hypotheses that foreign investment is associated with a greater reduction

in the credit constraints of firms which are less likely to have access to international capital markets. They

argue that this is plausible because incoming foreign investment provides an additional source of capital,

freeing up scare domestic credit which can then be redirected towards domestic enterprises.

55 See Javorkic (2004) for evidence that FDI fosters spillovers through backward linkages.

26

Overall, although our data do not permit to correct for some of the concerns associated with cross-

section analysis, our results are consistent with our previous findings, namely that foreign capital, in

particular, FDI, leads to increased levels of entrepreneurship.

5.2. Capital/Credit Availability Channel

In addition to an FDI channel, we also investigate the possibility that capital mobility affects

entrepreneurship through the availability of resources (i.e. a capital /credit availability channel). We

investigate whether firm creation in industries which are more reliant on external finance are positively or

negatively affected by our measures of international financial integration. We divide our sample into those

industries with high dependence on external finance measured by Rajan and Zingales (1998).

We run the specification in equation (1) for both high financial dependence industries and low

financial dependence industries across both de jure (the IMF index) and three de facto measures of capital

mobility. Table 10 shows our main results. In most cases we find that entrepreneurship in industries which

are more reliant in external finance is more sensitive to restrictions in capital mobility (columns 1 and 2),

and is more strongly affected by increased flows of finance (columns 3-8). In terms of the effect of de facto

measures of integration on the firm size distribution, a movement from the 25th percentile (Tanzania, 0.34)

of the Capital Inflows/GDP variable to the 75th percentile (Denmark, 15.6) is associated, based on the results

in Column (11), with an increase in skewness for highly financially dependent industries of 2.17 which

represents a 19 percent increase in skewness over the industry average. Whereas for less financially

dependent industries, the same change in Capital Inflows/GDO is associated, based on the results in Column

(12) with an average increase of just 9 per cent.

6. Conclusions

Using a new data set of 24 million firms in close to 100 countries, we found a positive relation

between measures of capital account integration and entrepreneurial activity in a country. Our work

describes a number of plausible channels through which international financial integration can affect the

dynamism of firms. One may argue that our results are expected from a neoclassical perspective. Access to

27

foreign capital and improve risk sharing should allow for new start-ups and opportunities in a country.

However, from a theoretical point of view and in light of the empirical findings on capital account

liberalization and growth our results may be surprising. In this regard, our view is that more micro analysis

is required to understand the effects of capital account openness in a country. At the very minimum, the use

of micro firm level data should enhance the general understanding of the process by which the effects of

liberalization are transmitted to the real economy.

7. References

Acemoglu, Daron, Simon Johnson and Todd Mitton, 2005. “Determinants of Vertical Integration: Finance,

Contracts and Regulation,” NBER Working Paper 11424.

Acemoglu, Daron and Fabrizio Zilibotti, 1997. “Was Prometheus Unbound by Chance? Risk,

Diversification and Growth,” Journal of Political Economy 105, 709-751.

Aghion, Philippe, and Peter Howitt, 1999. Endogenous Growth Theory. MIT Press: Cambridge.

Aitken, Brian J. and Ann E. Harrison, 1999. "Do Domestic Firms Benefit from Direct Foreign Investment?

Evidence from Venezuela," American Economic Review 89, 605-618

Alfaro, Laura, Steven. McIntyre and Vinati Dev, 2005. “Foreign Direct Investment and Ireland's Tiger

Economy,'' Harvard Business School Case705-009.

Alfaro, Laura, Sebnem Kalemli-Ozcan, and Vadym Volosovych, 2006. “Capital Flows in a Globalized

World: The Role of Policies and Institutions,” in Sebastian Edwards, editor, Capital Controls and

Capital Flows in Emerging Economies: Policies, Practices and Consequences. National Bureau of

Economic Research, forthcoming.

Alfaro, Laura and Andres Rodriguez-Clare, 2004. “Multinationals and Linkages: Evidence from Latin

America,'' Economia 4, 113-170.

Bartelsman, Eric J., John Haltiwanger and Stefano Scarpetta, 2004. “Microeconomic Evidence of Creative

Destruction in Industrial and Developing Countries,” mimeo.

Beck, Thorsten, Asli Demirguc-Kunt, Luc Laeven, and Ross Levine, 2006. “Finance, Firm Size and

Growth,” mimeo.

28

Bekaert, Geert and C. Harvey, 2000. “Foreign Speculators and Emerging Equity Markets,” Journal of

Finance 55, 565-613.

Bhagwati, J. 1998. “The Capital Myth.” Foreign Affairs 77 (3, May-June): 7-12.

Black, Sandra E. and Philip E. Strahan, 2002. “Entrepreneurship and Bank Credit Availability,” Journal of

Finance, 57, 2807-2833.

Blomstrom, Magnus and Hakan Persson. 1983. "Foreign Investment and Spillover Efficiency in an

Underdeveloped Economy: Evidence from the Mexican Manufacturing Industry,"

WorldDevelopment 11, 493-501.

Blomstrom, Magnus and Edward W. Wolff. 1994. "Multinational Corporations and Productivity

Convergence in Mexico," in W. Baumol, R. Nelson and E. Wolff (eds.) Convergence of

Productivity: Cross-national Studies and Historical Evidence. Oxford: Oxford University Press.

Botero, Juan, Simeon Djankov, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2003.

“The Regulation of Labor,” mimeo, Harvard University, March 2003.

Cabral, Luis and Jose Mata, 2003. "On the Evolution of the Firm Size Distribution: Facts and Theory,"

American Economic Review 93, 1075-1090.

Card, D. and Krueger, A. B., 1994. "Minimum Wages and Employment: A Case Study of the Fast-Food

Industry in New Jersey and Pennsylvania", American Economic Review 84, 772-793.

Caves, Richard E, 1974. "Multinational Firms, Competition and Productivity in Host-Country Markets,"

Economica 41, 176-93.

Chari, Anusha and Peter B.Henry, 2004. “Risk Sharing and Asset Prices: Evidence From a Natural

Experiment,” Journal of Finance 59, 1295-1324.

Claessens, Stijn, and Luc Laeven, 2003. “Financial Development, Property Rights, and Growth“ Journal of

Finance 58, 2401-2436.

Cooley, Thomas F., and Vincenzo Quadrini, 2003. "Common Currencies vs. Monetary Independence,"

Review of Economic Studies 70, 785-806.

Davis, Steven J., R. Jason Faberman, and John C. Haltiwanger, 2006. “The Flow Approach to Labor

Markets: New Data Sources and Micro-Macro Links,” NBER Working Paper 12167.

29

Desai, Mihir, Paul Gompers, Josh Lerner, 2003. “Institutions, Capital Constraints and Entrepreneurial Firm

Dynamics: Evidence from Europe,” NBER Working Paper 10165.

Djankov, Simeon and Bernard Hoekman. 2000. "Foreign Investment and Productivity Growth in Czech

Enterprises," World Bank Economic Review 14, 49-64.

Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2002. “The

Regulation of Entry,” Quarterly Journal of Economics 117, 1-35.

Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2003. “Courts,”

Quarterly Journal of Economics 118, 453-517.

Dunne, Timothy, Mark J. Roberts, 1991. “Variation in Producer Turnover across U.S. Manufacturing

Indsutires. Entry and Constability: an International Comparison. Edited by P.A. Geroskiand and J.

Shwalbach, Blackwell, 197-2003.

Dunne, Timothy, Mark J. Roberts, and Larry Samuelson, 1988. "Patterns of Firm Entry and Exit in U.S.

Manufacturing Industries," Rand Journal of Economics 19, 495-515.

Dunne, Timothy, Mark J. Roberts, and Larry Samuelson, 1989. "The Growth and Failure of U.S.

Manufacturing Plants," Quarterly Journal of Economics 104, 671-698.

Edison, H., M. W. Klein, L. Ricci and T. Sloek, 2002. “Capital Account Liberalization and Economic

Performance: Survey and Synthesis,” NBER Working Paper 9100.

Edwards, Sebastian, 2005. “Capital Controls, Sudden Stops and Current Account Reversal.” NBER

Working Paper 11170.

Eichengreen, Barry. 2001. “Capital Account Liberalization: What do Cross-Country Studies Tell Us? The

World Bank Economic Review 15, 341-365.

Evans, David S., 1987. "Tests of Alternative Theories of Firm Growth," Journal of Political Economy 95,

57-674.

Evans, David S., and Boyan Jovanovic, 1989. “An Estimated Model of Entrepreneurial Choice Under

Liquidity Constraints,” Journal of Political Economy 97, 808-827.

Evans, David S., and Linda S. Leighton, 1989. “Some Empirical Aspects of Entrepreneurship,” American

Economic Review 79, 519-535.

30

Feldstein, Martin. 2000. “Aspects of Global Economic Integration: Outlook for the Future. NBER Working

Paper 7899.

Fisman, Raymond and Virginia Sarria-Allende, 2005. “Regulation of Entry and the Distortion of Industrial

Organization,” mimeo.

Giannetti, Mariassunta and Steven Ongena, 2005. “Financial Integration and Entrepreneurial Activity:

Evidence From Foreign Bank Entry in Emerging Markets,” mimeo.

Gorg, Horg and Eric Strobl, 2002. “Multinational Companies and Indigenous Development: An Empirical

Analysis,” European Economic Review 46, 1305-1322.

Grossman, Gene, 1984. “International Trade, Foreign Investment, and the Formation of the Entrepreneurial

Class,” American Economic Review 74, 605-614.

Guiso, Luigi, Paola Sapienza and Luigi Zingales, 2004. “Does Local Financial Development Matter?”

Quarterly Journal of Economics 119, 929-969.

Haddad, Mona and Ann Harrison. 1993. "Are there Positive Spillovers from Direct Foreign Investment?

Evidence from panel data for Morocco," Journal of Development Economics 42, 51-74.

Harrison, Ann, Inessa Love and Margaret S. McMillian. 2004. “Global Capital Flows and Financing

Constraints,” Journal of Development Economics 75, 269-301.

Harrison, Ann and Margaret S. McMillian. 2003. “Does Foreign Direct Investment Affect Domestic Firms

Credit Constraints, Journal of International Economics 61, 73-100.

Haskel, Jonathan E., Sonia C. Pereira and Matthew J. Slaughter. 2002. "Does Inward Foreign Direct

Investment Boost the Productivity of Domestic Firms?" NBER Working Paper 8724.

Hausman, Ricardo and Dani Rodrik, 2003. “Economic Development as Self-Discovery,” mimeo.

Henry, Peter B., 2000. “Stock Market Liberalization, Economic Reform, and Emerging Market Equity

Prices,” Journal of Finance 55, 529-564.

Hirschman, Albert, 1958. The Strategy of Economic Development, New Haven: Yale University Press.

Hobday, Michael, 1995. Innovation in East Asia: The Challenge to Japan. U.K.: Edward Elgar Publishing

Imbs, Jean, 2004. “Trade, Finance, Specialization and Synchronization”, Review of Economics and Statistics

86, 723-34.

31

International Monetary Fund. Annual Report on Exchange Arrangements and Exchange Restriction, various

issues. Washington, DC: International Monetary Fund.

Javorcik, Beata S., 2004. “Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In

Search of Spillovers Through Backward Linkages,” American Economic Review 94, 605-627.

Johnson, Simon, John McMillan and Christopher Woodruff, 2002. “Property Rights and Finance,”

American Economic Review 92, 1335-1356.

Kalemli-Ozcan, Sebnem, Bent Sorensen and Oved Yosh, 2003. “Risk Sharing and Industrial Specialization:

Regional and International Evidence,” American Economic Review 93, 903-918.

Klapper, Leora, Luc Laeven, and Raghuram Rajan, 2005. “Entry Regulation as a Barrier to

Entrepreneurship,” mimeo, University of Chicago.

Klepper, Steven, and Elizabeth Graddy, 1990. "The Evolution of New Industries and the Determinants of

Market Structure," Rand Journal of Economics, 21, 27-44.

Klepper, Steven, and Kenneth L. Simmons, 2000. "The Making of an Oligopoly: Firm Survival and

Technological Change in the Evolution of the U.S. Tire Industry," Journal of Political Economy,

108, 728- 760.

Knight, Frank H, 1921. Risk, Uncertainty and Profit. Boston: Houghton Mifflin.

Kumar, Krishna, Raghu Rajan, and Luigi Zingales, 2002. “What Determines Firm Size?,” mimeo,

University of Chicago. 1999. C.E.P.R. Discussion Papers 2211.

La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert W. Vishny, 1998. “Law and

Finance,” Journal of Political Economy 106, 1113-1155.

Lewis, W. Arthur, 1955. The Theory of Economic Growth, London: George Allen and Unwin.

Levine, Ross, 2005. “Finance and Growth: Theory and Evidence.” forthcoming in Philippe Aghion and

Steven Durlauf, eds. Handbook of Economic Growth. The Netherlands: Elsevier Science. 2005.

Markusen, James and Anthony .J. Venables, 1999. “Foreign Direct Investment as a Catalyst for Industrial

Development,” European Economic Review 43, 335-338.

Obstfeld Maurice, 1994. “Risk-Taking, Global Diversification and Growth” American Economic Review 84,

1310-1329.

32

O’ Gorman, Colm, Eoin O’Malley and John Mooeny, 1997. “The Irish Indigenous Software Industry: an

Application of Porter’s Cluster Analysis.” Research Paper 3. Dublin: National Economic and Social

Council.

The Political Risk Services Group, 2005. International Country Risk Guide. The PRS Group, New York.

Prasad Eswar, Kenneth Rogoff, Shang-Jin Wei, and M. Ayhan Kose, 2003. “The Effects of Financial

Globalization on Developing Countries: Some Empirical Evidence,” International Monetary Fund

Occasional Paper 220.

Quinn, Dennis P., 1997. “The Correlates of Change in International Financial Regulation,” American

Political Science Review 91, 531-551.

Rajan, Raghuram, and Luigi Zingales, 1998. “Financial Dependence and Growth,” American Economic

Review 88, 559-586.

Rodriguez-Clare, Andres, 1996. “Multinationals, Linkages and Economic Development,” American

Economic Review 86, 852-873.

Rodrik, Dani, 1998. “Who Needs Capital Account Convertibility,?” Princeton Essays in International

Finance 207: 55-65.

Rodrik, Dani., 2005. “Why We Learn Nothing from Regressing Economic Growth on Policies,” mimeo.

Sahalman, W. A., Stevenson, H.H., Roberts, M.J. and Bhide, A.V. (eds), 1999. The Entrepreneurial

Venture. Boston: Harvard Business School Press.

Schumpeter, Joseph A., 1942. Capitalism, Socialism and Democracy. London: Unwin University Books.

Sutton, John, 1997. “Gibrat's Legacy,” Journal of Economic Literature 35, 40-59.

Stiglitz, Joseph, 2002. Globalization and its Discontent. W.W. Norton.

World Economic Forum, Competitiveness Report. Geneva: WEF. Various Issues.

World Bank, 2005, World Development Indicators. Washington D.C.: World Bank.

33

Appendix A: Rajan and Zingales (1998) Methodology

We also follow the methodology of Rajan and Zingales (1998) and Klapper, Laeven, and Rajan

(2005) and focus on cross-industry, cross-country interaction effects (is entry more likely in an industry with

a particular need when the country scores strongly on a characteristic that facilitates that need?). We use, as

did these authors, the U.S. as a proxy for the “natural” rate of entry in an industry and test whether entry is

relatively higher or lower in a “naturally high entry” industry when the country allows for international

capital mobility. This methodology, as explained by Rajan and Zingales (1998) and Klapper, Laeven and

Rajan (2005) enables us to address issues associated with country effects.56

icciciic KZ εγδθ +++×=Ε )( (A.1)

where icΕ corresponds the entrepreneurial activity measure in industry i of country c, δi corresponds to

industry dummies, γct correspond to country level dummies. The industry indicators correct for industry-

specific effects. Country dummies correct for country-specific variables. The focus of analysis in on the

interaction term θ between a country characteristics (Ktc) and industry characteristic, Zit. For country

characteristics, we focus on the “capital mobility measures”. For industry characteristics, we use the Rajan

and Zingales methodology (1998) and use the natural propensity to enter proxied by the US as the “natural

propensity to enter,” reflecting technological barriers in that industry like economies of scale or incumbent

organizational efficiencies obtained from experience. “Of course there is a degree of heroism in assuming

that entry in the US does not suffer from artificial barriers…”.57 However, the methodology only requires

that the rank ordering of entry in the U.S. corresponds to the rank ordering of natural barriers across

industries and this rank ordering to correspond to that of other countries.

56 This is equivalent to de-meaning the variables using their industry and country averages thus removing some of the sample selection problems. The interpretation of a positive coefficient on the interaction term would be that in countries with above average capital mobility, industries with above average “country characteristics” have higher than average rates of firm entry. 57 Klapper, Laeven and Rajan (2005, p.17).This approach is based on the assumption that there are industries that have “naturally” high entry barriers such as as capital intensiveness of production or technological complexity. Note that while this may be a strong assumption, all that matters for their argument to hold is that the rank order holds – that there is some intrinsic rank ordering of demand for external finance in manufacturing industries (some component of these barriers that is industry specific and invariant across countries). For a detailed description of their methodology, see Rajan and Zingales (1998).

34

Our hypothesis is that if international financial integration improves entrepreneurial activity then the

coefficient on the interaction term θ will be negative. This reflects the fact that we may expect industries

with higher potential levels of entrepreneurship to be more negatively affected by constraints on their access

to foreign capital. Appendix Table A.1 reports the main results. We find that θ is negative and significant

for the different proxies of capital integration. Moreover the magnitude of the relationship is economically

significant. In the case of the IMF Index, for example, as seen in n Column (1) we find that a movement

from the 25th percentile (Ethiopia, 0.90) to the 75th percentile (Costa Rica, 3.3) is associated with an increase

in entry of 0.29 percentage points for an industry with average potential entry such as textiles and apparel.

This is a 6 percent increase over the average entry for all industries in the sample.

Data Appendix A: The Dun and Bradstreet Data Set

a. Final Sample

We use data for 2004 and exclude information for which primary industry and year started

information were missing. Our original data set included 118 countries. We excluded territories with less

than 80 58observations and those for which the World Bank provides no data and for which capital market

integration information was not available. The final dataset used described in the summary statistics consists

in 24,606,036 establishments of 98 countries, which cover all economic sectors (SIC) with the exception of

Public Administration (Division J, group 9). Also, we dropped all establishments with started year before

1900. Because of some few and evident outliers in the sales reports, when estimated mean, median and

skewness, we dropped 6 observations: a firm with sales of 648.7 thousand billions in Denmark, a firm with

sales of 219.3 thousand billions in Spain, a firm with sales of 219.3 billions in Spain, a firm with sales of

32.7 thousand billions in Germany, a firm with sales of 5,6 thousand billions in Lithuania, a firm with sales

of 4,9 thousand billions in United Arab Emirates, a firm with sales of 352 billions in Nigeria, a firm with

sales of 291 billions in Chad, a firm with sales of 291 billions in Angola, a firm with sales of 121 billions in

Congo, and a firm with sales of 99 billions in Haiti. Likewise, we exclude firms that did not report sales. We

maintained data with certain information (for example employment) but missing other (sales). This was the

58 Most of these countries were in Africa and had less than 20 firms.

35

case mostly in less developed countries (Africa in particular). Our objective was to maximize the number of

observations for these countries. In the estimations of mean, median and skewness using employees, we

excluded those establishments that reported 0 in this category. The creation rate shows the number of

establishments reporting starting year in 2003-2005 period over all establishments. We defined foreign firms

as one having and upper most parent of a corporate family located in a country different from which the

survey firm operates.

b. Comparisons Dun and Bradstreet Data and US Census Data

To give some sense of the coverage of the D&B sample used in this study, there were 24,846,832

establishments59 recorded in the United States 2001/2002 business census (US Census Bureau, Statistics of

US Businesses). In our data, we have 6,185,542 establishments (for which we exclude establishments in the

total sample without the year started) About three quarters of all U.S. establishments have no payroll. Most

are self-employed persons operating unincorporated businesses, and may or may not be the owner's

principal source of income. There are 7,200,770 employer establishments with a total sales of $22 trillion

recorded in the US census. Our data contain 4,293,886 establishments with more than one employee with

total sales of $17 trillion. In the US census there are 3.7 million small employer establishments (less than 10

employees). In our data we have 3.2 million firms in the US with more than one and less than 10 employees.

This indicates a high proportion of small firms in the database which could be evidence of over-sampling of

small firms or underreporting of firm size. In our data we find that 9.9 per cent of establishments are new.60

The US Census61 found that 12.4 per cent of establishments were new in 2001/2002.62

59 The unit of record in the Dun and Bradstreet data is the “establishment” (a single physical location where business is conducted or where services or industrial operations are performed) as opposed to a “firm” (one or more domestic establishments under common ownership or control). The US census collects information on establishments as well as firms. 60 We define new as having a start year less than 2 years old. 61 Establishment and Employment Changes from Births, Deaths, Expansions, and Contractions, http://www.census.gov/csd/susb/usst01_02.xls 62 For firms with 1-4 employees this was 15.9 per cent, for firms with more than 500 employees this was 11 per cent.

Firm Level Data From Worldbase - D&B. In the analysis we use data at 2 digit SIC-1987 code.

Skewness - Employment Skweness of the firm size distribution measured by employment.

Size - Employment (Log) of the average firm size as measured by number of employees for each country/industry pair.

Entry Percentage of firm entry defined as the number new firms in t divided by the total number of firms that existed in year t-1 in the country/industry pair. Due tolags in data reporting and collection, we classify a new firm as new if it is less than two years old.

Skweness - Age Skewness of age

Age Average age of the firms for each country/industry pair.

Vintage Vintage is constructed within each country/industry pair and it is the weighted average age of the firms in each industry, the weights being given by the totalnumber of employees.

International Financial Intregation

De Jure

IMF’s Capital Account Liberalization Index

From the IMF's Annual Report on Exchange Arrangements and Exchange Restriction (AREAER). The index was constructed using data on multipleexchange arrangements, payments restrictions on current transactions and on capital transactions, and repatriation requirements for export proceeds. Acorresponding dummy variable takes the value of one if each of the restrictions was present in each country each year. In order for our IMF indicator to fallbetween 0 and 1 and increase with openness, our IMF capital control measure is the negative of the average of the four dummy variables for each country.

De Facto Capital Inflows From the IMF, IFS International Financial Statistics. Capital Inflows are the sum of the changes in the liabilities of FDI, equity portfolio, debt and derivative

flows. For FDI, we use direct investment in reporting economy (line 78bed). For portfolio equity investment, we use equity security liabilities (line 78bmd). For derivative flows we used financial derivative liabilities (line 78bxd). For debt flows, we use debt security liabilities (line 78bnd) as well as other investment liabilities (line 78bid). Data is calculated as a percentage of GDP in U.S. dollars t(aken from the World Bank, World Development Indicators).

cont.

Data Appendix B: Variables

Dependent Variable

Independent Variables

In order to illustrate the difference between Age and Vintage, assume two cohorts of firms, old (o) and young (y). Then, Average Size = (NoS^o+ NyS^y )/( No+ Ny); Average Vintage= (NoS^oAgeo +NyS^y Agey)/( No S^o+ Ny S^y); where N is the number of old or young firms, S^ is the average size of firms in each cohort and Age is the respective age of the twocohorts. No S^o is the employment in the old sector and Ny S^y is the employment in the young sector. For the moment, assume that S^o > S^y. Note for example that if Noincreases and S^o decreases such that the product No S^o is constant (total employment in the old sector is the same), then vintage will not change while average size will be smaller.Other example is if S^o does down and No goes up by more than enough to ensure that the population shits to the old sector (S^’o < S^o, No’ > No), o that No’ S^’o > NoS^o thenvintage will increase and average size will decrease. If S^y does down and Ny does not up enough so that the population in he young sector shrinks S^’y < S^y and Ny is kept at levelsso that S^’y N’y < S^y Ny, then vintage will increase and average size will decrease.

Inflows of Foreign Direct Investment

From the World Bank, World Development Indicators. Foreign direct investment is net inflows of investment to acquire a lasting management interest (10percent or more of voting stock) in an enterprise operating in an economy other than that of the investor. It is the sum of equity capital, reinvestment ofearnings, other long-term capital, and short-term capital, as shown in the balance of payments. Data is reported as percentage of GDP.

Gross Capital Flows From the World Bank, World Development Indicators. Gross private capital flows to GDP are the sum of the absolute values of direct, portfolio, and otherinvestment inflows and outflows recorded in the balance of payments financial account, excluding changes in the assets and liabilities of monetary authoritiesand general government.

Stock of Foreign Liabilities

From Lane and Milesi Ferretti (199, 2001, 2005) estimate stocks of portfolio equity and foreign direct investment based on the IMF, IFS flow data. In orderto estimate FDI stocks, the authors cumulate flows and adjust for the effects of exchange rate changes. For portfolio equity stocks, they adjust for changes inthe end of year U.S. dollar value of the domestic stock market.

Equity Flows From the World Bank, World Development Indicators. Sum of Foreign Direct Investment inflows and portfolio investment inflows as a percentage of GDP.Portfolio investment flows are net and include non-debt-creating portfolio equity flows (the sum of country funds, depository receipts, and direct purchases ofshares by foreign investors) and portfolio debt flows (bond issues purchased by foreign investors).

Bank and Trade Related Lending.

From the World Bank, World Development Indicators. Bank and trade-related lending covers commercial bank lending and other private credits calculatedas a percentage of GDP.

Net Capital Flows From the IMF, IFS International Financial Statistics. Net capital flows are the sum of the changes in assets and liabilities on FDI, equity portfolio and debtfrom the IMF, IFS International Financial Statistics. For FDI, we use direct investment abroad (line 78bdd) and direct investment in reporting economy (line78bed). For portfolio equity investment, we use equity security assets (line 78bkd) and equity security liabilities (line 78bmd). For derivative flows we usefinancial derivative assets (line 78 bwd) and financial derivative liabilities (line 78bxd). For debt flows, we use debt security assets (IFS line 78bld) and debtsecurity liabilities (line 78bnd) as well as other investment assets (line 78bhd) and other investment liabilities (line 78bid). Data is presented as a percentage

Macro-economic Data: From the World Development Indicators (WDI): (Log) GDP, (Log) GDP per capita, inflation (percent growth in the CPI), domestic credit to GDP,M3/GDP, inflation (change CPI). From the Economist Intelligence Unit: growth forecasts.

Institutional Quality Fom the International Country Risk Guide (ICRG), a monthly publication of Political Risk Services: Non-corruption (assessment of corruption within thepolitical system; average yearly rating from 0 to 6, where a higher score means lower risk). Law and order (the law subcomponent is an assessment of thestrength and impartiality of the legal system; the order sub-component is an assessment of popular observance of the law; average yearly rating from 0 to 6,where a higher score means lower risk). Bureaucratic quality (institutional strength and quality of the bureaucracy is another shock absorber that tends tominimize revisions of policy when governments change; average yearly rating from 0 to 4, where a higher score means lower risk.).

Regulation Form the World Development Indicators (WDI): number of days required to start a business; business disclosure index (0=less disclosure to 7=moredisclosure); legal rights of borrowers and lenders index (0=less credit access to 10=more access).

Dependence on External Finance

From Rajan and Zingales (1998). An industry’s external financial dependence is obtained by calculating the external financing of U.S. companies. Using data from Compustat, a firm's dependence on external finance is defined as: (Capex-Cashflow)/Capex, where Capex is defined as capital expenditures and Cashflow is defined as cash flow from operations. Industries with negative external finance measures have cash flows that are higher than their capital expenditures.

Independent Variable (cont).

Country # Firms Sales Mean (U.S.$)

Sales Skew.

Empl. Mean

Empl. Skew.

Age Mean

Age Skew.

% of New Firms

% Foreign Firms

Algeria 1,182 83,735,453 25 575 12 14 1.8 3.1 0.6Angola 195 1,041,687,230 6 748 9 16 2.0 5.1 23.1Argentina 8,627 11,476,813 42 107 17 17 1.8 4.7 8.3Australia 653,466 22,758,736 58 28 191 18 2.2 0.2 0.5Austria 207,939 3,396,636 178 11 84 18 2.0 5.5 1.6Belgium 639,073 1,410,931 208 7 697 16 1.8 6.1 0.8Bolivia 563 14,634,377 5 80 4 16 2.1 4.5 7.6Bosnia-Herzegovina 170 9,219,200 .. 89 4 17 2.7 2.6 1.8Brazil 263,090 5,024,083 165 46 89 18 1.8 0.3 0.8Bulgaria 2,196 33,377,919 1 169 19 9 5.1 6.6 3.0Burkina-Faso 87 550,460,000 0 583 9 16 1.3 2.9 10.3Cameroon 125 .. .. 242 5 20 0.9 2.8 11.2Canada 597,993 3,614,747 460 11 163 19 1.7 2.8 1.2Chile 3,218 158,195,679 7 161 10 17 1.9 5.1 6.2China 78,237 27,171,803 85 408 114 13 2.4 4.6 5.5Colombia 2,898 365,345,236 47 147 8 21 1.3 3.3 7.5Costa Rica 1,332 111,251,372 11 354 35 19 1.2 4.8 5.3Croatia 979 11,956,036 6 106 11 17 2.7 1.8 1.3Czech Republic 1,697,489 7,128,283 24 67 13 11 0.4 2.0 0.2Denmark 404,637 16,047,894 599 4 335 12 2.8 13.0 0.7Dominican Republic 1,536 252,400,351 10 177 13 19 1.6 2.7 2.9Ecuador 1,024 47,406,958 11 147 8 19 1.2 4.4 6.9Egypt 2,198 11,207,809 16 552 13 20 1.8 2.3 3.6El Salvador 664 9,380,212 4 173 6 19 1.1 3.3 6.2Estonia 1,383 10,735,116 8 87 8 11 6.0 2.2 18.7Ethiopia 132 40,453,156 6 926 4 22 1.3 2.3 1.5Finland 267,694 1,280,595 149 3 91 15 2.3 7.2 0.8France 4,024,287 87,223,814 1,712 3 1,053 12 1.9 11.9 1.1Gabon 76 350,188,281 0 139 4 21 0.7 3.3 11.8Gambia 26 1,527,588,000 0 142 3 16 1.2 5.8 0.0Georgia 106 1,500,000 .. 205 7 9 5.4 5.7 1.9Germany 1,228,884 395,830,182 712 17 502 19 2.2 5.4 1.0Ghana 521 34,316,958 10 189 10 16 1.7 3.4 2.9Greece 27,883 9,137,363 40 33 60 16 1.9 3.1 0.3Guatemala 679 33,722,935 12 139 13 15 1.8 4.0 5.2Honduras 450 13,174,706 4 163 5 19 1.5 3.3 7.6Hungary 66,585 3,665,310 132 41 95 12 4.3 1.6 2.3India 9,682 44,286,222 36 637 88 20 1.8 1.8 2.1Indonesia 682 24,132,541 10 688 10 18 1.4 2.1 9.2Iran 1,226 556,339,627 24 476 10 19 1.4 4.0 0.2Ireland 17,429 33,025,518 29 60 36 23 1.9 1.7 5.8Israel 68,164 4,393,917 57 25 62 19 1.5 0.9 0.0Italy 1,181,012 1,617,476 242 6 374 18 1.6 1.6 0.2Jamaica 424 60,191,309 3 153 6 21 1.3 2.6 5.9Japan 1,356,841 11,358,140 274 20 650 26 0.7 5.9 0.2Jordan 734 17,677,066 15 119 9 15 1.9 5.4 0.3Kenya 1,111 316,834,884 15 266 13 21 1.7 1.5 4.3Korea South 156,168 22,311,697 52 14 144 5 4.3 19.9 0.1Kuwait 922 24,250,342 9 337 8 20 0.8 4.6 0.5Latvia 1,386 28,113,799 22 110 18 9 4.8 9.1 15.9cont.

Table 1: Country Entrepreneurship Data: Summary Statistics -- 2003

Country # Firms Sales Mean (U.S.$) Sales Skew. Empl.

MeanEmpl. Skew.

Age Mean

Age Skew.

% of New Firms

% Foreign Firms

Lebanon 921 6,204,946 5 72 7 14 1.9 7.4 0.8Lithuania 1,248 158,591,913 .. 155 16 12 4.1 3.6 5.7Madagascar 124 858,036,511 8 591 8 20 1.7 3.2 6.5Malaysia 23,118 19,767,666 48 102 31 17 1.5 1.7 3.9Mauritius 358 99,870,675 11 253 4 21 1.6 2.2 0.3Mexico 23,817 456,475,129 106 123 23 17 1.8 2.9 8.1Morocco 2,295 25,991,883 30 202 17 20 1.6 4.4 4.2Mozambique 159 245,690,509 8 616 9 17 2.2 3.1 18.2Netherlands 1,042,095 3,131,474 700 8 184 13 2.7 12.5 1.2New Zealand 50,541 5,248,999 73 20 58 18 2.2 1.9 2.7Nicaragua 213 8,777,266 3 104 3 19 1.5 2.6 7.5Nigeria 1,088 247,463,286 6 254 11 19 1.7 2.3 3.3Norway 168,981 4,037,120 256 10 114 14 2.6 10.8 3.3Oman 405 54,477,517 13 806 8 17 1.4 4.0 0.7Panama 1,250 85,655,147 10 125 16 18 1.5 5.1 7.9Papua New Guinea 102 85,319,635 3 386 4 26 0.9 0.5 19.6Paraguay 411 953,146,299 6 118 9 18 2.0 3.9 7.1Peru 7,746 10,537,072 22 77 10 14 2.3 6.8 2.4Philippines 1,718 231,250,039 13 303 6 17 1.8 6.1 6.1Poland 4,619 13,260,042 28 114 16 13 3.9 1.2 15.2Portugal 488,633 676,946 143 5 103 13 2.3 7.2 0.5Romania 3,877 5,292,860 2 244 25 10 4.0 9.5 15.3Saudi Arabia 1,850 177,880,064 26 935 12 20 1.3 3.4 0.8Senegal 237 .. .. 176 6 21 1.4 4.4 3.0Singapore 63,277 11,820,648 55 30 38 13 1.1 9.1 3.9Slovakia 4,466 22,275,032 7 164 17 12 4.2 4.4 19.4Slovenia 3,265 12,386,865 17 73 15 18 2.2 2.5 2.8Spain 320,577 4,973,514 379 7 96 11 2.9 10.5 0.1Sudan 135 58,981,678 8 1,275 11 20 2.1 4.8 2.2Sweden 825,988 1,009,952 206 4 247 13 2.5 9.3 1.0Switzerland 271,689 9,204,443 137 30 160 16 2.3 6.7 2.7Syria 441 27,393,899 16 456 13 21 1.0 2.8 0.2Tanzania 179 5,610,461 1 257 5 15 1.9 2.0 6.1Thailand 1,471 55,999,591 32 443 8 16 1.6 2.9 5.8Togo 59 156,690,000 .. 160 4 20 2.3 3.4 6.8Trinidad & Tobago 563 45,745,842 7 176 12 21 1.4 2.2 3.4Tunisia 2,289 15,695,646 9 225 33 15 2.0 4.3 1.5Turkey 10,467 70,953,220 10 761 10 11 2.3 11.6 4.0Uganda 154 64,546,399 0 480 6 19 2.4 2.3 7.8United Arab Emirates 5,407 357,197,682 42 674 13 12 1.6 9.2 6.9United Kingdom 893,589 7,357,010 796 19 424 19 2.1 3.3 1.7Uruguay 934 48,099,972 12 107 12 20 1.6 3.7 10.0USA 7,389,228 2,406,644 802 9 2,351 18 1.9 6.1 0.2Venezuela 2,134 126,236,487 16 130 7 22 1.0 2.4 7.8Vietnam 114 114,793,677 5 1,073 10 10 2.8 7.5 1.8Yemen 189 36,046,492 6 981 4 23 1.0 1.1 1.1Zambia 112 435,607,919 6 1,215 10 19 1.3 4.9 14.3Zimbabwe 98 96,118,890 2 375 4 26 1.0 4.6 4.1

Table 1: Country Entrepreneurship Data: Summary Statistics -- 2003 (cont.)

Notes: Summary statistics correspond to D&B Data Set of 24 million firms. Counts do not consider SIC 9 (public sector). See Data Appendix fordetailed data description.

SIC # Firms Sales Mean (U.S.$)

Sales Skew.

Empl. Mean

Empl. Skew.

Age Mean

Age Skew.

% of New Firms

% Foreign Firms

SIC # Firms Sales Mean (U.S.$)

Sales Skew.

Empl. Mean

Empl. Skew.

Age Mean

Age Skew.

% of New Firms

% Foreign Firms

01 713,580 371,031 726 3 435 20 2 3.26 234 45 18,339 41,512,170 27 94 50 15 2 5.27 62302 428,394 354,196 197 2 105 20 2 3.95 128 46 3,402 112,349,111 17 59 7 14 3 2.59 2607 295,844 526,947 305 5 120 15 2 6.90 183 47 213,269 4,307,284 90 14 92 13 2 7.11 481208 128,309 431,695 290 3 178 11 3 8.96 94 48 79,048 20,836,546 81 49 158 10 3 12.26 155109 27,370 1,070,338 140 8 53 16 2 5.41 82 49 94,020 26,437,706 65 52 59 17 2 6.38 161510 6,630 36,115,356 19 176 28 17 3 2.29 213 50 1,145,973 6,793,763 823 14 465 16 2 6.02 2896412 2,219 37,816,832 19 768 12 20 2 2.66 45 51 939,051 8,613,254 220 15 453 16 2 4.83 1289913 22,876 54,532,790 40 106 88 19 1 4.08 590 52 181,317 1,855,490 255 7 252 19 2 4.41 63014 19,904 7,361,145 89 34 121 22 2 4.09 579 53 59,883 25,936,266 179 34 112 15 2 8.78 25815 696,335 1,901,223 172 8 650 18 1 6.40 732 54 510,605 3,017,759 136 9 418 15 2 6.84 257316 206,270 4,971,769 271 23 267 24 1 4.32 912 55 393,621 4,342,818 131 10 366 19 1 4.98 248417 1,765,406 137,907,664 1,206 6 473 17 2 6.50 2440 56 394,170 1,116,349 242 5 126 16 2 7.75 225020 208,737 12,668,835 159 40 126 19 2 6.12 2599 57 406,919 1,608,978 444 6 271 18 2 6.13 214821 1,009 129,953,155 19 380 11 21 2 5.15 123 58 987,537 661,171 241 8 176 13 2 8.33 372722 70,833 4,910,092 70 68 37 20 2 3.65 809 59 1,284,587 431,245,780 728 4 830 15 2 7.55 588823 127,738 2,907,002 113 32 48 16 2 5.00 558 60 60,375 54,469,078 132 85 43 26 1 5.39 223524 165,074 3,159,681 187 16 162 17 2 3.54 523 61 82,477 12,645,087 142 18 176 12 3 9.52 138325 93,738 2,704,394 64 18 42 18 2 4.45 562 62 101,485 7,669,455 112 19 146 12 3 13.15 140426 30,691 21,072,167 96 74 45 22 2 3.66 1276 63 44,338 438,666,659 169 99 78 20 2 5.69 164927 240,919 3,598,782 126 18 453 18 2 5.85 1726 64 217,775 2,285,519 367 7 210 17 2 5.24 106428 72,077 29,874,511 36 89 60 19 2 5.48 4778 65 1,147,555 1,103,200 515 5 822 16 2 9.93 486129 5,344 188,566,761 29 320 66 20 2 4.02 297 67 556,167 13,814,174 258 35 205 14 2 8.91 721630 72,595 9,534,537 80 57 55 20 1 3.51 2621 70 237,768 1,444,980 116 15 148 17 2 5.12 113231 29,394 3,203,448 40 49 24 19 2 3.41 325 72 897,896 325,497 263 4 177 14 2 7.19 124032 91,907 6,933,111 105 39 91 20 2 4.41 2371 73 1,878,877 4,724,003 1,175 11 236 11 2 9.22 1755133 46,587 24,965,634 53 120 49 20 2 4.03 1538 75 503,720 778,447 289 5 219 17 1 4.78 215534 279,551 4,085,868 120 24 82 19 2 3.48 3082 76 317,479 578,566 348 5 314 16 2 6.14 103535 227,043 9,654,378 147 38 75 20 2 4.65 6454 78 112,650 1,455,396 177 6 123 12 2 9.10 47036 150,398 40,090,440 301 84 42 16 2 5.32 4993 79 576,247 1,212,539 332 10 667 14 3 9.34 45437 56,890 65,109,353 167 103 61 19 2 5.75 2194 80 866,390 1,617,395 314 16 82 16 2 5.42 57338 72,627 9,876,270 124 40 241 17 2 5.50 2244 81 244,511 597,825 298 6 162 18 2 4.09 10439 130,857 2,160,329 82 16 59 16 2 5.54 831 82 310,302 2,755,343 175 27 111 19 2 6.43 38040 1,894 96,294,822 14 417 19 19 2 5.86 80 83 318,884 911,584 84 16 161 17 2 5.71 18641 150,949 1,710,587 103 31 349 15 2 6.71 697 84 18,920 907,950 18 10 23 21 2 5.26 2242 372,994 2,402,695 297 12 287 16 2 6.34 2183 86 627,854 1,017,928 275 8 387 24 2 4.08 17943 18,085 108,731,169 46 73 69 11 3 17.04 14 87 1,571,535 1,530,838 419 8 284 12 2 8.43 955644 35,435 8,918,267 92 29 89 18 2 6.34 627 88 3,757 392,338 23 2 31 6 4 22.85 045 18,339 41,512,170 27 94 50 15 2 5.27 623 89 131,361 771,817 79 8 182 12 2 10.67 466

Notes: Summary statistics correspond to all countries in our D&B Data Set. See Data Appendix for detailed data description.

Table 2: Industry Entrepreneurship Data Summary Statistics

De Jure De Facto De Facto De Facto De Facto De Facto

IMF index Capital Inflows/GDP FDI Flows/GDP Foreign

Liabilities/GDPGross Capital Flows/GDP

Net Capital Flows/GDP

Algeria 0.917 0.932 0.362Angola 0.846 10.233 1.388Argentina 0.615 -14.086 0.787 1.358 21.569 -12.242Australia 0.769 20.657 1.346 22.608 6.966Austria 0.308 25.267 2.875 2.052 34.424 -0.680Belgium 0.154 40.933 3.943 85.841 1.658Bolivia 0.308 7.047 2.062 1.333 12.405 0.451Bosnia-Herzegovina 0.462 5.475 0.825 19.697Brazil 0.538 2.737 2.006 0.776 6.726 -0.001Bulgaria 0.462 11.854 7.119 1.105 16.177 13.251Burkina-Faso 1.000 0.000 0.267 0.407 Cameroon 0.923 0.000 1.722 0.563 Canada 0.154 5.468 0.732 1.116 13.095 -2.135Chile 0.462 11.483 4.118 1.182 23.907 2.131China 0.923 5.365 3.777 14.275 3.725Colombia 0.846 2.402 2.184 0.707 12.646 1.054Costa Rica 0.000 1.406 3.298 0.687 10.627 3.399Croatia 0.846 25.246 6.936 1.262 33.621 16.602Czech Republic 0.385 7.011 2.781 0.986 19.554 6.058Denmark 0.154 15.566 0.559 2.078 36.851 -2.415Dominican Republic 0.615 10.437 1.874 0.880 19.790 -0.096Ecuador 0.231 4.116 5.716 0.996 11.116 0.852Egypt 0.462 -1.268 0.288 0.777 8.649 -6.946El Salvador 0.231 0.596 0.972 13.434Estonia 0.308 9.808 1.941 42.945Ethiopia 0.846 0.902 1.094 3.053Finland 0.385 11.654 2.123 2.076 43.560 -8.593France 0.154 25.753 2.450 2.064 23.361 -0.007Gabon 1.000 -3.160 0.876 0.664 -9.313Gambia 0.154 16.362Georgia 0.167 8.455 0.935 11.081Germany 0.077 11.167 1.064 1.591 19.517 -3.314Ghana 0.769 3.663 1.793 1.316 3.733 4.555Greece 0.231 25.465 0.416 1.403 22.873 3.518Guatemala 0.231 5.312 0.468 0.394 16.461 6.132Honduras 0.615 2.884 1.152 6.431Hungary 0.154 13.384 1.390 8.597India 1.000 4.594 0.711 2.463Indonesia 0.846 0.548 -0.250 3.958 -0.398Iran 1.000 0.088Ireland 0.154 228.747 17.304 9.499 328.066 -1.923Israel 0.154 4.611 3.520 1.160 13.357 -2.388Italy 0.154 12.113 1.126 1.236 17.807 1.764Jamaica 0.417 35.426 9.591 1.467 48.320 4.170Japan 0.154 1.293 0.145 17.077 -0.496Jordan 0.231 3.782 1.058 13.184Kenya 0.462 3.584 0.514 2.826Korea South 0.846 0.530 8.072 Kuwait 0.538 -0.161 0.253 49.560Latvia 0.308 2.709 1.226 28.695cont.

Table 3: Summary Statistics for Capital Mobility -- 2003

Country

De Jure De Facto De Facto De Facto De Facto De Facto

IMF index Total Inflows/GDP FDI Inflows/GDP Liabilities/GDP Gross Capital

Flows/GDPCapital

Inflows/GDPLebanon 0.615 1.799 2.433Lithuania 0.385 0.976 0.720 15.560Madagascar 1.000 -1.668 0.233 1.198 1.114 -2.163Malaysia 0.923 3.923 2.384 22.641 -3.196Mauritius 0.308 2.718 1.195 0.345 7.480 1.712Mexico 0.769 2.513 1.687 0.634 5.442 2.993Morocco 0.846 -0.460 5.213 0.801 9.281 -2.495Mozambique 1.000 10.298 7.793 1.088 11.294 8.629Netherlands 0.077 19.361 3.068 4.083 64.452 -4.689New Zealand 0.154 7.633 3.062 9.557 3.828Nicaragua 0.231 4.868 1.329 5.643Nigeria 0.462 2.083 0.905Norway 0.385 18.337 0.931 1.408 24.680 -8.715Oman 0.333 0.634 0.324Panama 0.000 -2.214 6.154 2.078 28.491 -0.105Papua New Guinea 1.000 2.961Paraguay 0.077 -0.568 1.506 0.636 12.495 2.872Peru 0.154 2.274 0.836 7.387Philippines 0.923 -0.881 0.402 39.893 -2.082Poland 0.769 5.560 1.968 0.849 11.288 4.560Portugal 0.385 32.251 4.469 2.457 53.703 -0.339Romania 0.385 7.652 3.238 0.653 8.207 7.726Saudi Arabia 0.769 -0.644 -0.276 0.251 16.709 -12.438Senegal 1.000 0.192 1.207 0.756 0.004Singapore 0.385 21.598 12.375 99.405 -21.255Slovakia 0.231 0.000 1.747 0.970 15.541 0.000Slovenia 0.538 7.554 1.215 0.844 18.858 1.802Spain 0.154 21.746 3.042 1.745 38.937 1.155Sudan 0.818 7.583 1.131 10.508Sweden 1.000 -5.531Switzerland 0.154 8.263 5.482 4.213 41.512 -7.856Syria 1.000 0.699 1.420 6.437Tanzania 1.000 0.343 2.409 0.979 -0.244Thailand 0.846 -3.455 1.364 10.983 -5.335Togo 1.000 1.125 1.313Trinidad & Tobago 0.308 5.856 1.086 Tunisia 0.923 6.206 2.161 1.344 5.276 4.433Turkey 0.769 5.749 0.650 0.729 6.772 2.953Uganda 0.154 8.090 3.084 0.949 6.714 5.706United Arab Emirates 0.385 0.206 0.000United Kingdom 0.154 40.495 1.153 3.737 71.360 2.174Uruguay 0.154 17.766 2.454 1.547 49.331 0.038USA 0.308 12.723 0.364 1.067 11.363 4.788Venezuela 3.020 0.723 12.892 Vietnam 1.000 3.702Yemen 0.308 -0.810 0.499 3.195Zambia 0.077 2.307 1.343 Zimbabwe 1.000 0.000

Notes: The IMF index is the average of the capital market securities; money market instruments; collective investment securities; controls onderivatives and other instruments: commercial credits; financial credits; guarantees, securities and financial backup facilities; controls on directinvestment; controls on real estate transactions: controls on personal capital transactions from IMF, AREAER. Total Capital Inflows/GDP are the sumof inflows of foreign direct investment, portfolio, derivative and debt flows, from IMF, IFS. FDI Inflows/GDP are from WB, WDI. Gross CapitalFlows/GDP are the sum of the absolute values of direct, portfolio, and other investment inflows and outflows excluding changes in the assets andliabilities of monetary authorities and general government from WB, WDI. Net Capital Flows are the sum of the inflows and outflows of of foreigndirect investment, portfolio, derivative and debt flows, from IMF, IFS. Foreign Liabilities/GDP from Lane-Milesi Ferreti. See Data Appendix fordetailed data description.

Table 3: Summary Statistics for Capital Mobility -- 2003 (cont.)

Country

Country Bureau. Qual.

Non-Corrup.

Law and

Order

Days to Start

BusinessCountry Bureau.

Qual.Non-

Corrup.Law and

Order

Days to Start

Business

Algeria 2.00 1.50 2.00 26 Latvia 2.50 2.00 5.00 18Angola 1.00 2.00 3.00 146 Lebanon 2.00 1.00 4.00 46Argentina 3.00 2.50 1.50 32 Lithuania 2.50 2.50 4.00 26Australia 4.00 4.50 6.00 2 Madagascar 1.00 4.00 2.50 44Austria 4.00 5.00 6.00 29 Malaysia 3.00 2.50 3.00 30Belgium 4.00 4.00 5.00 34 Mauritius Bolivia 2.00 2.00 3.00 59 Mexico 3.00 2.00 2.00 58Bosnia-Herzeg. 54 Morocco 2.00 3.00 5.00 11Brazil 2.00 4.00 1.50 152 Mozambique 1.00 1.50 3.00 153Bulgaria 2.00 2.00 4.00 32 Netherlands 4.00 5.00 6.00 11Burkina-Faso 1.00 2.00 3.50 135 New Zealand 4.00 5.50 6.00 12Cameroon 1.00 2.00 2.00 37 Nicaragua 1.00 2.50 4.00 45Canada 4.00 5.00 6.00 3 Nigeria 1.00 1.00 1.50 44Chile 3.00 2.50 5.00 27 Norway 4.00 5.00 6.00 23China 2.00 2.00 4.50 41 Oman 2.00 2.50 5.00 34Colombia 2.00 3.00 1.00 43 Panama 2.00 2.00 3.00 19Costa Rica 2.00 2.50 4.00 77 Papua New Guinea 2.00 1.00 2.00 56Croatia 3.00 3.00 5.00 49 Paraguay 1.00 1.00 2.00 74Czech Republic 3.00 2.50 5.00 40 Peru 2.00 2.50 3.00 98Denmark 4.00 5.50 6.00 4 Philippines 3.00 2.00 2.00 50Dominican Rep. 1.00 2.00 2.00 78 Poland 3.00 2.00 4.00 31Ecuador 2.00 3.00 3.00 92 Portugal 3.00 3.50 5.00 78Egypt 2.00 1.50 4.00 43 Romania 1.00 2.50 4.00 28El Salvador 2.00 2.50 2.50 115 Saudi Arabia 2.00 2.00 5.00 64Estonia 2.50 3.00 4.00 72 Senegal 1.00 2.50 3.00 57Ethiopia 1.00 2.00 5.00 32 Singapore 4.00 4.50 5.00 8Finland 4.00 6.00 6.00 14 Slovakia 3.00 2.50 4.00 52France 3.00 3.00 5.00 8 Slovenia 3.00 3.00 4.50 61Gabon 2.00 1.00 3.00 Spain 4.00 3.50 4.50 108Gambia 2.00 3.00 4.00 Sudan 1.00 1.00 2.50Georgia 25 Sweden 2.00 2.00 3.00Germany 4.00 4.50 5.00 45 Switzerland 4.00 4.50 5.00 20Ghana 2.00 2.50 2.00 85 Syria 1.00 2.00 5.00 47Greece 3.00 2.50 3.00 38 Tanzania 1.00 2.00 5.00 35Guatemala 2.00 1.50 1.50 39 Thailand 2.00 1.50 2.50 33Honduras 2.00 2.50 1.50 62 Togo 0.00 1.50 3.00 53Hungary 4.00 3.00 4.00 Trinidad & Tobago 3.00 2.00 2.00India 3.00 1.50 4.00 89 Tunisia 2.00 2.00 5.00 14Indonesia 2.00 1.00 2.00 151 Turkey 2.00 2.50 4.50 9Iran 2.00 2.00 4.00 48 Uganda 2.00 2.00 4.00 36Ireland 4.00 3.50 6.00 24 United Arab Emirate 3.00 2.00 4.00 54Israel 4.00 4.00 5.00 34 United Kingdom 4.00 4.50 6.00 18Italy 2.50 2.50 3.00 13 Uruguay 2.00 3.00 2.50 45Jamaica 3.00 1.50 1.00 31 USA 4.00 4.00 5.00 5Japan 4.00 3.50 5.00 31 Venezuela 1.00 1.50 1.00Jordan 2.00 3.00 4.00 36 Vietnam 2.00 1.50 4.00 56Kenya 2.00 3.50 2.00 Yemen 1.00 2.00 2.00 63Korea South 0.00 1.00 5.00 22 Zambia 1.00 2.00 4.00 35Kuwait 2.00 2.00 5.00 35 Zimbabwe 2.00 0.00 0.50 96Notes: Days to start a business, from WB, WDI; bureaucratic quality, non-corruption index, law and order from ICRG; property rights and employment regulation from WEF. See Data Appendix for detailed data description.

Table 4: Summary Statistics -- Main Control Variables by Country -- 2003

Entry Skew. Age Empl. Skew.

Empl. IMF Index FDI Flows/GDP

Gross Capital

Flows/GDP

Liabilities/ GDP

Net Capital Flows/GDP

Capital Inflows/

GDPLog GDP Log

GDPpcGDP

Growth

Days to Start

Business

Bureau. Qual.

Non-Corrup.

Law and

Order

Entry 1.000Skew. Age 0.209 1.000Empl. -0.187 -0.311 1.000Skew. Empl. 0.089 0.246 -0.417 1.000IMF Index -0.081 -0.186 0.418 -0.259 1.000FDI Flows/GDP -0.053 -0.081 0.050 -0.142 -0.079 1.000Gross Capital Flows/GDP -0.072 0.118 -0.130 0.055 -0.290 0.727 1.000Liabilities/GDP -0.008 0.208 -0.259 0.166 -0.406 0.623 0.925 1.000Net Capital Flows/GDP -0.098 -0.133 0.175 -0.058 0.116 0.173 -0.106 -0.186 1.000Capital Inflows/GDP -0.087 0.099 -0.097 0.057 -0.253 0.735 0.967 0.859 0.043 1.000Log GDP 0.054 0.353 -0.374 0.477 -0.269 -0.326 0.094 0.228 -0.226 0.096 1.000Log GDPpc 0.107 0.396 -0.565 0.398 -0.531 -0.063 0.356 0.488 -0.384 0.302 0.724 1.000GDP Growth 0.037 -0.081 0.286 -0.182 0.212 0.059 0.000 -0.096 0.154 -0.025 -0.176 -0.216 1.000Days to Start Business -0.170 -0.133 0.185 -0.202 0.166 0.037 -0.157 -0.268 0.113 -0.122 -0.275 -0.443 -0.044 1.000Bureau. Qual. 0.055 0.389 -0.532 0.343 -0.446 0.015 0.363 0.512 -0.322 0.302 0.616 0.868 -0.241 -0.413 1.000Non-Corrup. 0.157 0.378 -0.550 0.333 -0.378 -0.112 0.219 0.399 -0.414 0.141 0.512 0.732 -0.228 -0.295 0.766 1.000Law and Order 0.197 0.313 -0.368 0.265 -0.339 0.059 0.336 0.463 -0.186 0.305 0.376 0.660 -0.163 -0.506 0.679 0.667 1.000Notes: See Data Appendix for detailed data description.

Table 5: Correlation Table for Main Variables

De Jure De Facto De Facto De Facto De Facto

IMF index Capital Inflows/GDP FDI Inflows/GDP Foreign

Liabilities/GDPGross Capital Flows/GDP

1 2 3 4 5

Capital Mobility -1.175 0.047 0.238 2.629 0.095[0.594]** [0.017]*** [0.093]** [0.422]*** [0.013]***

Log GDP 0.847 0.867 0.806 1.393 1.244[0.110]*** [0.112]*** [0.117]*** [0.163]*** [0.117]***

Log GDPPC 2.342 2.039 2.644 1.234 1.447[0.225]*** [0.210]*** [0.207]*** [0.332]*** [0.237]***

GDP growth 0.485 0.376 0.486 0.540 0.319[0.064]*** [0.068]*** [0.067]*** [0.077]*** [0.064]***

Days to Start Business -0.005 -0.014 -0.002 -0.002 -0.013[0.005] [0.005]*** [0.005] [0.006] [0.005]**

Bureaucracy -2.828 -2.736 -2.893 -1.936 -3.064[0.250]*** [0.241]*** [0.258]*** [0.413]*** [0.264]***

Non-Corruption 0.397 0.499 0.323 0.323 0.406[0.201]** [0.200]** [0.206] [0.250] [0.206]**

Law and order 0.877 0.886 0.971 1.086 1.110[0.143]*** [0.141]*** [0.147]*** [0.180]*** [0.145]***

Constant -38.929 -36.698 -41.444 -50.872 -42.258[2.626]*** [2.652]*** [2.873]*** [3.745]*** [2.714]***

# Observations 5454 4334 4363 5077 4450Notes: All regressions include industry dummies and are estimated by Tobit. Robust standard errors are in parentheses denoting *** 1%, **5% , and*10% significance. See Data Appendix for detailed description of the data.

Table 6a: Capital Mobility and Entrepreneurship: Entry -- Cross Section (Tobit)

De Jure De Facto De Facto De Facto De Facto

IMF index Capital Inflows/GDP FDI Flows/GDP Foreign

Liabilities/GDPGross Capital Flows/GDP

1 2 3 4 5

Capital Mobility 0.349 -0.014 -0.060 -0.272 -0.016[0.073]*** [0.002]*** [0.012]*** [0.027]*** [0.002]***

Log GDP 0.009 0.030 0.009 0.055 -0.021[0.014] [0.015]** [0.015] [0.017]*** [0.016]

Log GDPPC -0.386 -0.480 -0.445 -0.293 -0.429[0.028]*** [0.028]*** [0.026]*** [0.034]*** [0.031]***

GDP growth 0.109 0.116 0.115 0.110 0.108[0.008]*** [0.009]*** [0.008]*** [0.008]*** [0.008]***

Days to Start Business 0.000 -0.001 0.000 0.000 0.000[0.001] [0.001]* [0.001] [0.001] [0.001]

Bureaucracy 0.209 0.251 0.223 -0.030 0.280[0.032]*** [0.034]*** [0.033]*** [0.046] [0.036]***

Non-Corruption -0.382 -0.398 -0.383 -0.371 -0.411[0.025]*** [0.027]*** [0.026]*** [0.028]*** [0.027]***

Law and order -0.031 -0.006 -0.021 0.042 -0.010[0.018]* [0.019] [0.019] [0.020]** [0.019]

Constant 7.231 7.546 7.970 6.287 8.686[0.320]*** [0.348]*** [0.356]*** [0.384]*** [0.353]***

# Observations 5348 4300 4980 4372 4273R2 0.48 0.51 0.48 0.49 0.50

Table 6b: Capital Mobility and Entrepreneurship: Employment -- Cross Section (OLS)

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standarderrors are in parentheses denoting *** 1%, **5% , and *10% significance. See Data Appendix for detailed description of the data.

De Jure De Facto De Facto De Facto De Facto

IMF index Capital Inflows/GDP

FDI Inflows/GDP

Foreign Liabilities/GDP

Gross Capital Flows/GDP

1 2 3 4 5

Capital Mobility -10.056 0.328 0.338 4.228 0.127[3.141]*** [0.069]*** [0.190]* [1.489]*** [0.069]*

Log GDP 6.162 6.421 3.205 6.563 5.429[1.044]*** [1.119]*** [0.315]*** [1.413]*** [0.728]***

Log GDPPC -0.56 0.111 0.808 0.673 0.306[1.173] [1.139] [0.501] [1.713] [1.289]

GDP growth -1.084 -1.476 -0.63 -0.744 -1.194[0.298]*** [0.393]*** [0.158]*** [0.346]** [0.322]***

Days to Start Business -0.056 -0.069 -0.008 -0.039 -0.037[0.029]* [0.032]** [0.013] [0.028] [0.027]

Bureaucracy -1.51 -2.435 -0.702 -3.049 -1.58[1.105] [1.239]* [0.457] [2.144] [1.306]

Non-Corruption 3.147 3.759 2.74 2.077 3.165[0.955]*** [1.033]*** [0.493]*** [1.024]** [0.858]***

Law and order 0.541 -0.494 0.522 0.52 0.478[0.755] [0.983] [0.434] [0.958] [0.915]

Constant -152.036 -173.048 -92.883 -177.506 -148.257[20.908]*** [23.692]*** [6.785]*** [31.241]*** [16.315]***

# Observations 4383 3644 3977 3544 3908R2 0.37 0.38 0.56 0.39 0.45

Table 6c: Capital Mobility and Entrepreneurship: Skewness of Employment -- Cross Section (OLS)

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity and corrected at thecountry level (clustering). Robust standard errors are in parentheses denoting *** 1%, **5% , and *10% significance. See Data Appendix fordetailed description of the data.

1999/00 2003/04

Entry 7.559 4.526Skewness (size) 6.071 9.521Size (log) 3.299 4.212Age 23.286 18.129IMF index 0.519 0.479FDI Flows/GDP 4.022 2.807Liabilities/GDP 1.304 1.454Gross Capital Flows/GDP 28.992 28.526Capital Inflows/GDP 13.440 13.088Net Capital Flows/GDP 0.626 0.215

Table 7: Comparative Summary Statistics(2003/04 -- 1999/00)

Notes: See Data Appendix for detailed description of the data.

De Jure De Facto De Facto De Facto De Facto

IMF index Capital Inflows/GDP

FDI Inflows/GDP

Foreign Liabilities/GDP

Gross Capital Flows/GDP

1 2 3 4 5

D Capital Mobility 3.803 0.380 0.383 6.074 0.005[0.893]*** [0.092]*** [0.199]* [1.884]*** [0.068]

D Log GDP 3.600 5.197 2.768 1.996 2.792[0.285]*** [0.555]*** [0.318]*** [0.515]*** [0.324]***

D Log GDPpc 1.862 6.445 3.813 3.925 3.559[0.442]*** [0.800]*** [0.476]*** [0.744]*** [0.471]***

D GDP growth -0.815 -0.405 -0.694 -0.285 -0.759[0.126]*** [0.234]* [0.132]*** [0.160]* [0.140]***

D Bureaucracy -1.792 3.224 3.067 -1.154 2.085[1.390] [6.180] [3.403] [4.544] [3.395]

D Law and Order -2.006 -2.154 -0.007 0.791 -0.270[0.473]*** [0.953]** [0.578] [0.622] [0.601]

D Non-Corruption 1.308 2.958 3.152 5.964 3.236[0.484]*** [1.186]** [0.715]*** [1.002]*** [0.726]***

Constant -78.419 -132.389 -67.113 -76.289 -91.398[16.547]*** [25.637]*** [15.322]*** [13.889]*** [10.602]***

# Observations 1672 1046 1215 842 1170R2 0.34 0.34 0.42 0.51 0.42 Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standarderrors are in parentheses denoting *** 1%, **5% , and *10% significance. See Data Appendix for detailed description of the data. Data forthe variable Days to Start a Business was not available for 99/00.

Table 8: Capital Mobility and Entrepreneurship: Skweness of EmploymentDifferences in Differences (04-99)

Skewness Skewness Skewness Entry Entry Entry

Same Industry Upstream Industry

Downstream Industry Same Industry Upstream

IndustryDownstream

IndustryOLS OLS OLS Tobit Tobit Tobit

1 2 3 4 5 6

Foreign firms -0.527 2.131 -54.625 0.181 1.520 22.419[0.070]*** [6.499] [14.647]*** [0.019]*** [1.347] [3.589]***

Log GDP 5.755 6.289 6.278 0.781 0.611 0.685[0.253]*** [0.291]*** [0.313]*** [0.065]*** [0.070]*** [0.073]***

Log GDPPC 1.679 1.283 1.117 2.131 2.508 2.319[0.485]*** [0.565]** [0.595]* [0.128]*** [0.138]*** [0.142]***

Days to Start Business -0.044 -0.041 -0.042 -0.006 -0.003 -0.003[0.012]*** [0.016]*** [0.017]** [0.003]* [0.004] [0.004]

Bureaucracy -1.362 -2.247 -1.946 -2.235 -2.545 -2.583[0.606]** [0.723]*** [0.776]** [0.159]*** [0.177]*** [0.183]***

Non-Corruption 4.030 5.221 5.149 0.315 0.369 0.457[0.481]*** [0.608]*** [0.647]*** [0.123]** [0.144]** [0.149]***

Law and order -0.100 -0.784 -0.810 0.708 0.573 0.515[0.344] [0.496] [0.522] [0.089]*** [0.118]*** [0.121]***

Constant -175.235 -179.378 -177.609 -34.03 -31.499 -31.524[7.659]*** [8.624]*** [8.981]*** [1.606]*** [1.761]*** [1.807]***

# Observations 4112 3512 3039 4971 4182 3624R2 0.37 0.36 0.37

Table 9: Channels -- FDI Foreign firms in Same, Up Stream, and Downstream Industries

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are inparentheses denoting *** 1%, **5% , and *10% significance. Foreign firms are the share of foreign firms to total firms. See Data Appendix for detaileddescription of the data.

IMF Index IMF Index FDI Flows/GDP

FDI Flows/GDP

Liabilities/ GDP

Liabilities/ GDP

Gross Capital Flows/GDP

Gross Capital Flows/GDP

Capital Inflows/GDP

Capital Inflows/GDP

(High) (Low) (High) (Low) (High) (Low) (High) (Low) (High) (Low)1 2 3 4 5 6 7 8 11 12

IMF Index -10.604 -9.473 0.777 0.600 8.124 5.577 0.158 0.049 0.444 0.211[2.173]*** [1.666]*** [0.292]*** [0.230]*** [1.376]*** [1.116]*** [0.032]*** [0.026]* [0.055]*** [0.044]***

Log GDP 6.514 5.828 6.324 5.518 7.167 6.302 6.575 5.589 6.768 6.071[0.391]*** [0.303]*** [0.416]*** [0.333]*** [0.520]*** [0.429]*** [0.428]*** [0.338]*** [0.447]*** [0.359]***

Log GDPPC -0.656 -0.461 0.938 1.199 0.689 0.033 0.215 0.784 -0.088 0.324[0.825] [0.631] [0.745] [0.581]** [1.122] [0.883] [0.935] [0.723] [0.879] [0.690]

GDP growth -1.272 -0.910 -1.469 -1.017 -1.024 -0.701 -1.533 -1.038 -1.622 -1.331[0.237]*** [0.177]*** [0.241]*** [0.185]*** [0.253]*** [0.203]*** [0.263]*** [0.200]*** [0.301]*** [0.232]***

Days to Start Business -0.058 -0.053 -0.066 -0.051 -0.050 -0.038 -0.052 -0.046 -0.076 -0.061[0.018]*** [0.014]*** [0.018]*** [0.015]*** [0.019]*** [0.016]** [0.021]** [0.016]*** [0.022]*** [0.017]***

Bureaucracy -1.876 -1.158 -1.887 -1.031 -5.127 -1.952 -1.994 -0.932 -3.042 -1.832[0.894]** [0.698]* [0.904]** [0.729] [1.362]*** [1.093]* [1.065]* [0.841] [1.011]*** [0.814]**

Non-Corruption 3.384 2.923 3.596 3.093 2.139 2.094 2.963 2.859 4.054 3.487[0.736]*** [0.555]*** [0.740]*** [0.574]*** [0.822]*** [0.657]*** [0.836]*** [0.642]*** [0.851]*** [0.662]***

Law and order 0.709 0.388 0.445 0.182 0.575 0.316 0.526 0.264 -0.584 -0.405[0.515] [0.392] [0.515] [0.403] [0.581] [0.467] [0.581] [0.450] [0.599] [0.470]

Constant -167.145 -150.157 -181.914 -161.322 -194.669 -173.314 -181.837 -160.273 -180.139 -165.620[11.075]*** [7.701]*** [11.853]*** [8.516]*** [13.284]*** [10.589]*** [11.866]*** [8.441]*** [12.366]*** [8.956]***

# Observations 2158 2225 2041 2117 1531 1601 1939 1979 1813 1831

R2 0.35 0.40 0.36 0.40 0.33 0.36 0.36 0.39 0.38 0.40

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are in parentheses denoting *** 1%, **5% , and *10%significance. See Data Appendix for detailed description of the data.

Table 10: Channels -- Financial Dependence (Skewness of Employment)

Skewness Entry Size Age Vintage Skewness Entry SkewnessOLS Tobit OLS OLS OLS OLS OLS OLS

1 2 3 4 5 6 7 8

IMF Index -3.890 -1.819 0.447 -0.355 0.377 -9.775 -0.677 -6.176[1.223]*** [0.758]** [0.100]*** [0.590] [0.834] [1.465]*** [0.398]* [1.611]***

Log GDP 5.916 1.052 -0.056 0.033 0.202 6.495 0.353 4.691[0.265]*** [0.167]*** [0.022]** [0.135] [0.190] [0.278]*** [0.078]*** [0.300]***

Log GDPPC 0.267 2.383 -0.471 -1.210 -0.225 -1.148 0.945 -1.801[0.457] [0.285]*** [0.038]*** [0.223]*** [0.314] [0.591]* [0.159]*** [0.571]***

GDP growth -0.872 0.565 0.094 -0.425 -0.432 -1.091 0.219 -0.822[0.126]*** [0.078]*** [0.009]*** [0.062]*** [0.087]*** [0.164]*** [0.044]*** [0.165]***

Days to Start Business -0.022 -0.017 0.000 0.006 0.017 -0.064 -0.009 -0.006[0.010]** [0.007]*** [0.001] [0.005] [0.007]** [0.012]*** [0.003]*** [0.014]

Bureaucracy -1.191 -3.155 0.282 1.811 1.601 -1.491 -2.389 -0.891[0.500]** [0.315]*** [0.036]*** [0.255]*** [0.361]*** [0.610]** [0.169]*** [0.537]*

Non-Corruption 2.455 0.002 -0.341 0.823 1.465 3.419 0.177 3.145[0.422]*** [0.267] [0.033]*** [0.212]*** [0.299]*** [0.490]*** [0.135] [0.472]***

Law and order 0.953 0.899 -0.011 -0.860 -0.621 0.724 0.687 0.197[0.294]*** [0.186]*** [0.021] [0.147]*** [0.208]*** [0.349]** [0.096]*** [0.381]

Rights Borrow/Lenders 1.217 0.097 -0.087 -0.319 -0.696[0.177]*** [0.112] [0.013]*** [0.090]*** [0.127]***

Bus. Disclosure -0.800 0.198 -0.030 -0.015 0.343[0.231]*** [0.147] [0.017]* [0.116] [0.164]**

M3/GDP -0.025 -0.010 -0.001 0.017 0.025[0.008]*** [0.005]** [0.001]* [0.004]*** [0.006]***

Inflation 0.083 -0.001[0.059] [0.015]

GDP Forecasts -0.723[0.366]**

# Observations 3123 3967 3102 3967 3967 4099 5010 2963R2 0.38 0.52 0.17 0.15 0.38 0.12 0.45

Appendix Table 1: Robustness -- Capital Mobility and Entrepreneurship (Cross Section)

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are in parentheses denoting *** 1%,**5% , and *10% significance. See Data Appendix for detailed description of the data.

Entry Skewness Entry Entry Entry Skewness Empl. SkewnessTobit OLS OLS Tobit Tobit OLS OLS IV

1 2 3 4 5 6 7 8

IMF Index -2.454 -3.932 -0.886 0.557 -29.524[0.803]*** [1.104]*** [0.430]** [0.179]*** [5.133]***

Log GDP 1.153 3.522 0.219 0.573 1.447 6.312 0.028 9.073[0.154]*** [0.220]*** [0.091]** [0.103]*** [0.205]*** [1.020]*** [0.015]* [0.665]***

Log GDPPC 2.033 -0.484 1.211 2.305 2.637 1.37 -0.513 -3.167[0.300]*** [0.406] [0.162]*** [0.199]*** [0.383]*** [1.095] [0.028]*** [1.115]***

GDP growth 0.422 -0.460 0.254 0.345 0.581 -1.582 0.124 -0.910[0.082]*** [0.115]*** [0.046]*** [0.065]*** [0.114]*** [0.422]*** [0.009]*** [0.217]***

Days to Start Business -0.042 -0.005 -0.018 -0.011 -0.012 -0.06 -0.001 -0.116[0.007]*** [0.010] [0.004]*** [0.005]** [0.008] [0.033]* [0.001]** [0.021]***

Bureaucracy -2.895 0.121 -2.366 -2.624 -2.044 -1.769 0.233 -0.945[0.277]*** [0.440] [0.180]*** [0.229]*** [0.514]*** [1.327] [0.033]*** [0.693]

Non-Corruption 0.183 1.196 -0.408 0.418 0.389 4.553 -0.437 2.778[0.250] [0.391]*** [0.159]** [0.188]** [0.447] [1.261]*** [0.027]*** [0.670]***

Law and order 0.446 -0.029 0.504 0.764 0.913 -0.517 -0.003 -1.092[0.202]** [0.264] [0.107]*** [0.134]*** [0.244]*** [1.077] [0.019] [0.448]**

Rights Borrow/Lenders-182.425 7.885

Bus. Disclosure [23.086]*** [0.349]***

GDP Forecasts 0.243[0.192]

Domestic Credit/GDP 0.175 0.014[0.010]*** [0.004]***

Bank and Trade Flows/GDP 34.863[18.711]*

Equity Flows/GDP 0.018[0.007]***

Net Flows /GDP 0.557 -0.017[0.179]*** [0.004]***

# Observations 3346 3280 4222 4450 3409 3710 4373 2094R2 0.42 0.12 0.38 0.50 0.36

Appendix Table 2: Robustness -- Capital Mobility and Entrepreneurship (Cross Section)

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are in parentheses denoting *** 1%,**5% , and *10% significance. See Data Appendix for detailed description of the data.

Only Manufacturing Only Rich countries Non-US Regional Dummies

1 2 3 4

IMF Index -5.649 -15.350 -9.086 -7.000[0.912]*** [3.286]*** [1.132]*** [1.053]***

Log GDP 4.013 8.685 5.212 5.847[0.402]*** [0.704]*** [0.262]*** [0.305]***

Log GDPPC -0.759 -0.988 -0.183 -0.420[0.291]*** [1.336] [0.308] [0.357]

GDP growth -0.422 -0.714 -1.108 -1.124[0.074]*** [0.410]* [0.093]*** [0.104]***

Days to Start Business -0.033 -0.064 -0.040 -0.044[0.009]*** [0.029]** [0.009]*** [0.009]***

Bureaucracy -0.410 -3.518 -1.428 -1.828[0.262] [0.929]*** [0.361]*** [0.376]***

Non-Corruption 1.720 4.678 3.049 3.124[0.246]*** [0.936]*** [0.323]*** [0.350]***

Law and order 0.251 0.001 0.807 -0.271[0.190] [0.585] [0.190]*** [0.239]

Asia -1.862[0.898]**

Europe 3.780[0.820]***

Middle East North Africa -3.091[0.723]***

South America -3.919[0.759]***

North America 13.993[4.732]***

Constant -97.002 -217.09 -139.359 -149.902[8.446]*** [22.030]*** [5.526]*** [6.605]***

# Observations 1404 2223 4315.000 4382R2 0.41 0.38 0.38 0.39

Notes: All regressions include industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are inparentheses denoting *** 1%, **5% , and *10% significance. Foreign firms are the share of foreign firms to total firms. See Data Appendix for detaileddescription of the data.

Appendix Table 3 – Robustness: Manufacturing, Rich Countries, Regional Dummies --Skewness

Entry Skew. Entry Skew. Entry Skew. Entry Skew.Tobit OLS Tobit OLS Tobit OLS Tobit OLS

1 2 3 4 5 6 7 8

New Firms in US * Inflows / GDP 0.006[0.003]**

Skewness Firms in US * Inflows / GDP 0.001[0.000]***

New Firms in US * FDI Flows / GDP 0.019[0.011]*

Skewness Firms in US * FDI Flows / GDP 0.002[0.001]**

New Firms in US * Gross Flows / GDP 0.006[0.001]***

Skewness Firms in US * Gross Flows / GDP 0.003[0.000]***

New Firms in US * Foreign Liabilities / GDP 0.065[0.014]***

Skewness Firms in US * Foreign Liabilities / GDP 0.091[0.009]***

Constant -0.779 -9.031 -0.018 3.5 0.372 -6.819 0.813 -7.998[5.997] [15.342] [6.723] [14.040] [6.130] [14.635] [3.573] [10.372]

# Observations 4737 4029 5728 4564 4852 3911 4054 2723R2 0.26 0.38 0.19 0.4 0.24 0.4 0.26 0.47

Notes: All regressions include country and industry dummies and are estimated by OLS with White's correction of heteroskedasticity. Robust standard errors are in parentheses denoting *** 1%, **5% ,and *10% significance. See Data Appendix for detailed description of the data.

Appendix Table A: Entpreneurship and Financial IntegrationBenchmark - U.S. (Rajan and Zingales Methodology)