Firm Financing and Growth in the Arab Region

Transcription

1 Policy Research Working Paper 7756 WPS7756 Firm Financing and Growth in the Arab Region Juan Jose Cortina Lorente Soha Ismail Sergio L. Schmukler Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Development Research Group Macroeconomics and Growth Team July 2016

2 Policy Research Working Paper 7756 Abstract This paper documents how firms in Arab countries issue equity, corporate bonds, and syndicated loans in domestic and international markets to obtain financing and grow. Using a new data set on issuance activity and firm performance, the paper finds that capital raising through these markets has grown rapidly since the early 1990s and involved an increasing number of issuing firms. Whereas the amounts raised (relative to gross domestic product) in equity and loan markets stand well with respect to international standards, bond issuance activity lags behind. Yet, bond financing has gained importance over time. Equity issuances primarily take place domestically, while bonds and loans are mostly issued internationally, display long maturities, and entail low levels of credit risk. Issuing firms are larger, grow faster, and are more leveraged than non-issuers. While issuers tend to be larger ex ante than nonissuers, the size gap between them seems to widen over time. This paper is a product of the Macroeconomics and Growth Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at The authors may be contacted at jcortinalorente@worldbank.org, sismail2@worldbank.org, sschmukler@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team

3 Firm Financing and Growth in the Arab Region Juan Jose Cortina Lorente Soha Ismail Sergio L. Schmukler * JEL Classification Codes: F21, F65, G00, G10, G15, G23, G31, L25 Keywords: Arab countries, capital raising, corporate bonds, domestic and international debt markets, equity, firm financing, global financial crisis, issuance activity, Middle East and North Africa region, syndicated loans * We are grateful to Shanta Devarajan and Ibrahim Elbadawi for encouraging us to write the paper. For generous support to produce this paper, we are grateful to the World Bank s Development Economics Department, Knowledge for Change Program (KCP), Middle East and North Africa Region, Strategic Research Program (SRP), and especially the Economic Research Forum (ERF) in Egypt. addresses: jcortinalorente@worldbank.org, sismail2@worldbank.org, sschmukler@worldbank.org.

4 1. Introduction Since the early 1990s, many countries in the Arab world have embarked on significant financial and economic reforms, involving internal and external financial liberalization, as well as efforts to increase the depth, scope, and efficiency of their financial systems. At the same time, Arab financial systems have shown considerable improvements over the last two decades. Although policy makers in the region have realized the importance of expanding the breadth of their financial systems and operating active capital markets, Arab financial systems are often perceived to be underdeveloped. In particular, financial markets are still highly bank-based, thin, tightly regulated, and dominated by government ownership. Within the Arab world, financial systems in the Gulf Cooperation Council (GCC) tend to be more developed and globally integrated than those of other countries in the region. 1 This paper uses a unique data set to provide a first documentation of the extent to which firms in the Arab region use capital and syndicated loan markets to obtain financing and grow. We address three main questions as follows. (1) What is the total amount raised in equity, bond, and syndicated loan markets by firms in the Arab region and how does their issuance activity stand with respect to the rest of the world? (2) How many and which firms actually issue equity, bonds, and syndicated loans? (3) How do the assets, turnover, and number of employees evolve for issuing relative to nonissuing firms? To address these questions, we assemble a comprehensive transaction-level data set on equity, corporate bonds, and syndicated loans issued in domestic and international markets over the period. The data include 719,242 individual security issuances conducted by 138,091 (listed and non-listed) firms from 96 countries. For Arab countries, we then match these data with balance sheet information on publicly listed firms. Our matched data comprise 1,462 firms over the period The Gulf Cooperation Council states include Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates. 1

5 2011 from 12 Arab countries. This allows us to document new patterns about the comparative behavior of the size and growth of issuing and non-issuing firms. This paper presents three main interrelated findings. First, over the two decades leading to 2014, there has been considerable increase in the issuance activity of equity and debt in Arab countries. However, while the capital raising in equity and syndicated loan markets stands well with respect to international standards, corporate bond markets still lag behind in the Arab region. Syndicated loan markets are especially active, with the highest issuance activity (relative to GDP) in the developing world. Nevertheless, the relative importance of corporate bond financing over total debt has increased over time, especially since the onset of the global financial crisis of Within the region, the GCC countries capture around percent of the total amount raised in equity and debt markets. Whereas the bulk of equity is issued domestically, debt issuances mostly take place in international markets and have relatively long average maturities. Corporate bonds issued by non-financial Arab firms during had the longest average maturity in the world (11.5 years), while syndicated loans presented the second longest (8.9 years). Corporate bond issuances also entail low levels of credit risk compared to other developing regions, according to Standard and Poor s credit ratings. Second, the growth in equity and debt markets in the Arab region has not been limited to an increase in the intensive margin. Namely, the number of issuing firms has substantially risen, indicating an expansion in the extensive margin. In particular, the total amount raised (number of issuing firms) has increased 21-fold (15-fold) in equity markets, 22-fold (6-fold) in bond markets, and 5-fold (3-fold) in syndicated loan markets between and These findings stand in contrast with other regions in the world, where the expansion in capital market issuance activity has been mainly associated with a growth in the intensive margin (Didier and Schmukler, 2013; Didier, Levine, and Schmukler, 2015). The expansion in these markets has also been associated with a decrease in firm 2

6 concentration, as shown by the decreasing proportion of capital raising activity captured by the top-5 and top-20 issuing firms. Third, firms that issue either equity, bonds, or syndicated loans are larger than non-issuing firms, as they have more assets, turnover, and number of employees. For instance, during the median equity-issuing firm had assets of around $240 million, which was almost 3 times the value of assets of the median non-issuer. Debt, and especially bond, issuers are even larger than equity issuers. Despite being larger, issuers grow at a faster rate than non-issuers. For instance, the growth rates in assets (turnover) for the median equity issuer and non-issuer were around 14 (17) percent and 8 (13) percent per year, respectively. Issuing firms are also more leveraged and hold more long-term debt compared to non-issuers. A difference-in-differences analysis shows that issuing firms are larger ex ante than non-issuing firms in terms of assets, turnover, and the number of employees. While both issuers and non-issuers have grown in size over time, issuers have grown at a faster pace, widening the size gap. Similar results are evident along the firm size distribution (FSD). Although we do not evaluate the causal impact of a firm issuing equity, bonds, or loans on its performance, the findings in this paper indicate that firms grow faster when they issue. The analysis in this paper relates to the literature on financial development and growth. A large number of studies argue that financial development is positively associated with overall economic growth. 2 Better functioning financial systems can improve information dissemination and reduce transaction and monitoring costs, leading to a more efficient allocation of resources. Moreover, more accessible financial services can be beneficial for the development of small and medium enterprises (SMEs), which tend to be underserved in developing countries (Beck and Demirgüç-Kunt, 2006; de la Torre, Martínez-Pería, and Schmukler, 2010; Beck, Demirgüç-Kunt, and Martínez-Pería, 2011). Another strand of literature stresses that larger and more liquid capital markets are associated with 2 See, for example, Levine (2005) for a review of the finance and growth literature. 3

7 higher aggregate economic growth (e.g. Levine and Zervos, 1996, 1998; Demirgüç-Kunt and Maksimovic, 1998; Henry, 2000; Beck and Levine, 2004; Bekaert, Harvey, and Lundblad, 2005; Levine, 2005). Capital markets are considered an important source of long-term financing, and hence can be an engine of long-term investment and growth in developing countries. Nonetheless, most of this literature uses aggregate measures of size as proxies for financial system performance, whereas only very few have investigated how real market activity is related to firm performance. Studying the use of capital markets by firms in the Arab world is interesting not only because those countries have undergone a number of significant changes in their financial systems during the past three decades, but also because research on financial development in the region is generally rather limited. For example, Elsafti (2007) indicates that the total market capitalization for Arab stock markets increased 18-fold between 1994 and 2005, while the number of listed firms increased 1.5- fold. Similarly, looking at six non-gcc countries, Finger and Gressani (2014) argue that although stock markets have considerably expanded in size, the number of stocks available for portfolio investment is still small compared to the international average. A recent report by the Arab Monetary Fund (2015) shows that the total Arab stock market capitalization was around $1,255 billion in the second quarter of 2015, with 1,503 listed firms. The GCC countries account for almost 87 percent of stock market size in the region. Market capitalization in Saudi Arabia alone was reported at $537 billion for the second quarter of 2015; a figure that was more than 3 times the market size of all the non- GCC countries together. A limited number of studies investigate the relation between the size of the financial system and aggregate economic growth in the Middle East and North Africa (MENA) or Arab Region (Abu- Bader and Abu-Qarn, 2008a, 2008b; Al-Rjoub, and Abu-Mhareb, 2006; Al-Zubi, Kar, Şaban, and Ağır, 2011; Al-Malkawi, 2012). Moreover, literature that specifically looks at the economic effect of capital markets is limited to an even fewer number of studies (Bolbol, Fatheldin, and Omran, 2005; Naceur 4

8 and Ghazouani, 2007; Falahaty and Hook, 2013). For example, studying the rate of investment growth for 83 firms in five Arab countries during , Bolbol and Omran (2005) find no significant relation between equity issuance and investment growth in a given year. While they find debt growth to be associated with higher capital expenditure, there is no information as to the nature or type of this debt. No study has investigated how and which firms in the Arab world issue equity, bonds, or syndicated loans to raise finance, and how this is related to their growth performance. So far, the literature has been mostly limited to the use of aggregate data to analyze the level of banking system or stock market development in the Arab region. This paper contributes to the literature by using a large transaction-level data set to present a detailed analysis of how Arab firms have used not only stock markets, but also corporate bond and syndicated loan markets to raise funds over a long period of time ( ). Moreover, using a uniquely matched firm-level data set of real issuances and firm characteristics, we are able to examine how widespread the use of equity, bonds, and syndicated loans by firms in the Arab world is, and how firms that actually issue in these markets evolve compared to non-issuers. While some interesting findings have been documented for several countries (e.g. Didier and Schmukler, 2013; Didier, Levine, and Schmukler, 2015), none have specifically looked at firms in the Arab region. The remainder of the paper is organized as follows. Section 2 describes the data. Section 3 describes the evolution and main characteristics of equity, bond, and syndicated loan markets in the Arab region. Section 4 examines which firms use these markets, focusing on firm characteristics such as size, growth, debt structure, and profitability. Section 5 studies the dynamics of ex ante and ex post firm performance and the evolution of FSD for issuing and non-issuing firms. Section 6 concludes. 5

9 2. Data To assess the capital raising activity of Arab countries in equity, corporate bond, and syndicated loan markets, we assemble a comprehensive data set on firms security issuances from 1991 through To benchmark the Arab region against other regions in the world, we include firms issuance activity from all over the world. Our data on firms capital raising activity come from the Thomson Reuters Security Data Corporation (SDC) Platinum database, which provides transaction-level information on new issuances of common and preferred equity, publicly and privately placed bonds, and syndicated loans with an original maturity of one year or more. Given that the SDC Platinum database does not collect data on debt issuances with maturities shorter than one year, the data set does not cover commercial paper. Because our analysis focuses on corporate financing, we exclude all public sector issuances, comprising securities issued by national, local, and regional governments, government agencies, regional agencies, and multilateral organizations. We also exclude mortgage-backed securities and other asset-backed securities. The data set includes 138,091 firms and 719,242 security issuances: 199,742 equity issuances, 294,159 bond issuances, and 225,341 syndicated loan issuances. The analysis focuses on 12 Arab nations, for which the data set includes 1,630 firms and 4,372 security issuances: 1,398 equity issuances, 650 bond issuances, and 2,324 syndicated loan issuances. To classify equities and corporate bonds as domestic or international, we compare the market location in which the securities are issued to the issuing firm s nationality. For offerings that take place simultaneously in more than one market, we consider tranches in each market as separate issuances. The data set includes 382,535 issuances in domestic markets and 105,916 issuances in international markets. For syndicated loans, the nationality of the banks that participate in the deal is used to distinguish between domestic or cross-border lending. Domestic loans are those in which only domestic banks participate in the syndication, whereas international loans entail the participation of at 6

10 least one foreign bank. The data set includes 108,016 domestic syndicated loans and 117,325 international syndicated loans. Firms are classified into financial and non-financial corporations according to their Standard Industry Classification (SIC) code. Firms with SIC codes between are classified as financial corporations. The data set includes 40,591 equity, 184,236 bond, and 45,403 syndicated loan issuances by financial firms (or 20, 63, and 20 percent of the total issuances of equity, bonds, and syndicated loans, respectively). To examine the comparative characteristics and performance of issuing and non-issuing firms in the Arab countries, we match the data set on security issuances from SDC Platinum with firm-level balance sheet information from the Orbis (Bureau van Dijk) database. The latter covers publicly listed companies, providing a rather homogeneous sample of firms. By omitting unlisted firms from the analysis using the matched data, the sample excludes firms that are (1) relatively small and sometimes informal, (2) likely to have different accounting standards, and (3) less likely to issue in capital markets. The final matched data set covers 1,462 firms from 12 Arab countries. In particular, our sample includes firms from Algeria, the Arab Republic of Egypt, Bahrain, Jordan, Kuwait, Lebanon, Morocco, Oman, Qatar, Saudi Arabia, Tunisia, and the United Arab Emirates. Moreover, we classify firms as issuers or non-issuers based on whether they issued equity, bonds, or syndicated loans at any point during our sample period. Because firm-level balance sheet information is only available from 2003 to 2011, we classify a firm as an issuer if it had at least one issuance during that period. We further classify whether firms are equity, bond, or loan issuers depending on whether firms issued any equity, bonds, or syndicated loans, respectively. The sample of non-issuing firms is held fixed throughout the paper. Non-issuing firms are those that did not have 7

11 any issuance activity between 2003 and In the SDC-Orbis data, 384 firms are issuers and 1,078 are non-issuing firms. 3 Our analysis focuses on firm size and growth, measured by the level and growth rate of total assets, turnover, and the number of employees. Firm assets and turnover are measured in constant 2011 U.S. dollars, using the U.S. consumer price index (CPI) to discount nominal values. The analysis also examines firm profitability, and other financial indicators such as return on assets (ROA), leverage (including bank and other types of financing), and the maturity profile of liabilities. 3. Capital Raising Activity in the Arab World The available literature has documented considerable progress in the development of the financial sector in the Arab world. Nonetheless, it is essential to analyze the extent to which such developments have translated into an increased use of capital financing by the private sector. The data used here focus on capital raising activity at the transaction level of equity, bonds, and syndicated loans. The issuance data show that there has been a sizable expansion in equity and debt markets in the Arab region since the early 1990s. The total amounts raised in equity, bond, and syndicated loan markets, relative to the region s GDP, have grown by factors of 8, 6, and 2, respectively, between and (Figure 1). In comparison to international standards, while equity and syndicated loan markets seem to perform relatively well, bond market activity still lags behind. For example, the total value of equity issuances by firms in the Arab world stood at 0.8 percent of GDP in , compared to 0.7 percent in Eastern Europe and Central Asia (ECA), 1.0 percent in Latin America and the Caribbean (LAC), 1.6 percent in Asia, and percent in developed countries. On the other hand, at 1.2 percent of GDP, bond market issuances remained low compared to approximately 1.6 percent in ECA, 2.0 percent in LAC, 2.6 percent in Asia, and percent in 3 Appendix Table 3 reports the number of issuing and non-issuing firm per Arab country. 8

12 the developed world. Although the importance of syndicated loans, relative to GDP, has decreased since the onset of the global crisis, their markets remain very active. In fact, at 3.6 percent of GDP, Arab syndicated loan markets were the most active in the developing world during The global financial crisis of seems to have had an effect on the nature of debt financing by Arab firms. Whereas the total amount raised in syndicated loans to GDP dropped by almost 12 percent during , it increased by 54 percent in bond markets (Figure 1). As a result, loan debt as a share of the total debt issued per year has declined by 16 percentage points between 2008 and 2009 in the Arab region, following a worldwide pattern (Figure 2). This pattern is consistent with recent literature highlighting a second wave of global liquidity (from banks and toward bonds) in developing countries after the global financial crisis (Shin, 2013; Acharya et al., 2015; IMF, 2015; Cortina, Didier, and Schmukler, 2016). Looking at the absolute value of issuances, the wide expansion in Arab equity and debt markets is also evident. The gross amount raised by firms in equity, bond, and syndicated loan markets has increased almost 21-fold, 22-fold, and 5-fold, respectively, between and (Figure 3, Panel A). Comparing the different markets, the total amount raised in debt was larger than the total amount raised in equity during the last study period. In particular, the total amounts raised through bonds and loans were 1.4 and 4.3 times the amount raised in equity, respectively. However, while most of equity market issuances take place domestically, the bulk of debt issuances by Arab firms are concentrated in foreign markets. International issuances accounted for less than 4 percent of the total capital raised in equity markets, but for 89 percent and 92 percent of the total in bond and loan markets during Almost all the funds raised with foreign debt and a large share of domestic debt issuances (around 49 percent in the last period) are denominated in foreign currency. 4 Most of the international issuances by Arab firms are conducted in a few, mostly developed, regions. The bulk of international equity is issued in the United Kingdom (61 percent), Eurozone (11 percent), and United States (8 percent). International bond issuances take place mostly in the Eurozone (61 percent), the United States (16 percent), and the United 9

13 The weighted average maturity of debt at issuance during the period was 7.7 years in corporate bond and 7.8 years in syndicated loan markets (Table 1, Panel A). Moreover, within the non-financial sector Arab firms issue very long-term corporate bonds and loans compared to other regions in the world. 5 The average maturity of issuances by non-financial firms in Arab countries is 11.5 years for bonds, the longest in the world, and 8.9 years for syndicated loans, the second longest in the world. A leading factor that might be behind this pattern is the intensive use of debt to finance infrastructure projects, whose funding usually involves large amounts and long maturities. Indeed, there is a large share of issuances coming from the transportation and energy sector in corporate bond markets. Borrowing within this utility sector usually entails infrastructure project financing. This sector accounts for more than 70 percent of the total amount raised in bonds by non-financial firms, which is large compared to international standards (Figure 4, Panel B). Syndicated loans for the transportation and energy sector also capture a larger share of the non-financial loan debt in the Arab region than in other regions in the sample (Figure 4, Panel C). Accordingly, project finance loans, a category that consists primarily of infrastructure projects, are used intensively in the Arab region (only exceeded by Asia). Loans for project finance have an average maturity of 13.3 years and account for about 31.6 percent of all syndicated loans contracted by Arab countries (Figure 5). On the other hand, the average maturity of loans extended for other projects is much lower at 5.2 years. 6 Consistent with this pattern, the median deal size of debt issuances (proceeds raised per issuance) in the Arab region is also much larger than in other regions of the world. The typical debt issuance (bonds and loans) Kingdom (8 percent). Similarly, the largest volumes of syndicated lending are originated within a few regions, mainly the Eurozone countries (25 percent), United States (23 percent), and Saudi Arabia (12 percent). 5 Financial firms typically go shorter term than non financials, pushing the overall average maturity downwards. 6 Most of the project finance lending around the world finances infrastructure (Blanc-Brude and Ismail, 2013). Moreover, most financing for infrastructure projects comes from syndicated loans. Engel, Fischer, and Galetovic, (2014) provide evidence that in the United States and other developed countries the ratio of bonds to syndicated loans for infrastructure financing is 1:5 to 1:6, respectively. The ratio in Asia (excluding China) is 1:8 and in Latin America, 1:3. 10

14 stands at more than $200 million, which is very large compared to international standards. For instance, the median bond (loan) size is $19 ($125) million in LAC, $44 ($70) million in Asia, and $117 ($97) million in G7 countries (Table 2, Panel A). Moreover, corporate bonds issued by Arab firms tend to have relatively low risk ratings compared to other developing regions. Almost 44 percent of the total capital raised with corporate bonds by firms in Arab countries is rated between A- and AAA, according to Standard and Poor s credit ratings (Figure 6, Panel A). One possible explanation behind such high credit ratings might be the fact that it is the very large, low risk firms that are involved in bond issuances. 7 To what extent does the expansion in capital and syndicated loan markets imply that a wider set of firms in fact use them? The growth in capital raising activity in the Arab region has been associated with a growth in the extensive margin. In other words, with the growth in market activity, an increasing number of firms have been using equity, bond, and syndicated loan markets to obtain financing over the years. The number of issuing firms in Arab equity markets has increased by almost 15 times, from 44 issuers in to 658 issuers in (Figure 3, Panel B). Similarly, the number of issuers in Arab bond (syndicated loan) markets has increased almost 6-fold (3-fold), from 24 (141) to 146 (483) issuers. These findings stand in contrast with other regions in the world, where the expansion in capital market issuance activity has been mainly associated with a growth in the intensive margin, that is, a small number of firms materially increasing their use of capital markets (Didier and Schmukler, 2013; Didier, Levine, and Schmukler, 2015). Although the increasing number of issuing firms has translated into a lower degree of market concentration around the top issuers over time, most of the capital raising activity in the Arab region remains captured by a few firms. For instance, while the total amount raised in equity markets by the 7 These high credit ratings are not driven by the dominance of the financial sector in bond issuances. Although around 69 percent of the total amount raised through bonds in the Arab region is carried out by financial firms, similar credit rating results are found when restricting the analysis to non-financial firms. 11

15 top-5 (top-20) equity issuers stood at 51 (88) percent of the total during , their market share dropped to 19 (45) percent during (Figure 3, Panel C). Similar patterns are also observed in debt markets. Respectively, the top-5 (top-20) bond issuers accounted for 69 (99) percent of the market in , which declined to 19 (51) percent in For loan markets, while the top-5 (top-20) issuers captured close to 35 (67) percent of the market in , their shares dropped to 12 (32) percent during the last period. These concentration figures remain high, especially in equity markets where the number of issuing firms is relatively larger (there were 658 equity issuers, 146 bond issuers, and 483 loan issuers during ). Looking at the different countries in the region, there is considerable heterogeneity in market activity, with the bulk of trading concentrated within a limited number of countries. More specifically, the GCC states accounted for 82 percent, 91 percent, and 90 percent of the total value of issuances in equity, bond, and syndicated loan markets during (Table 2, Panel B). 8 Those are rather high shares considering that such countries accounted for around 66 percent of the total region s GDP during the same period. Saudi Arabia and UAE seem to be in the lead, with the latter accounting for almost 60 percent of bond issuances in the region. Among non-gcc nations, the Arab Republic of Egypt seems to have the most active markets, followed by Morocco in equity and loan markets and Tunisia in bond markets. 4. Which Firms Use Equity, Bond, and Syndicated loan Markets? To study which firms use capital and syndicated loan markets we merge the transaction-level data set, from SDC Platinum database, with balance sheet information for listed firms from Orbis. We compare 8 Almost 20 percent of bond issuances and 18 percent of loan issuances in the GCC countries are in the form of Islamic finance issuances, while such markets barely exist in the other Arab countries. In particular, Bahrain, Kuwait, Qatar, Saudi Arabia, and UAE have the most active Islamic finance markets in the Arab region, with around 13 (19) percent, 21 (21) percent, 18 (16) percent, and 57 (26) percent of their bonds (loans) in the form of Islamic issuances. It is worth mentioning, though, that Malaysia has the most active Islamic bond market in the world, accounting for 90 percent of all Islamic bond issuances in our sample (around 2,100 issuances. 12

16 the characteristics of non-issuing firms with firms that issue different types of instruments. In particular, we compare: (1) firm size, measured by assets and turnover in 2011 U.S. dollars and the number of employees; (2) firm growth, measured by the annual growth rate of assets, turnover, and employees; (3) the liability structure of the firm, measured by the firm s leverage and the ratio of longterm debt to total firm liabilities; (4) firm profitability, measured by the ratio of retained earnings to assets, the return on assets (ROA), and the return on equity (ROE). When comparing those characteristics across issuers and non-issuers, we use the median firm for the sample period, after taking the average over time for each firm. For the rest of the paper, we use the matched SDC-Orbis data set on capital raisings and balance sheet information for the period Issuing and non-issuing firms differ along several dimensions. Issuers are much larger than firms that do not issue equity, bonds, or syndicated loans (Table 3). Moreover, debt issuers, and especially bond issuers, tend to be much larger than equity issuers. During our study period, while the median non-issuing firm in the Arab region had around $81 million in assets, the median equity issuer had assets of around $240 million, the median bond issuer had assets of around $10.4 billion, and the median loan issuer had assets of around $5 billion. Similar qualitative results are found when looking at firm turnover and the number of employees; issuers are typically larger in size than non-issuers, debt issuers are larger than equity issuers, and bond issuers tend to be the largest. Issuing firms also tend to grow faster than non-issuing firms. While the median non-issuer grew at a rate of 7.5 percent per year (looking at assets) between 2003 and 2011, the median equity, bond, and syndicated loan issuer grew at a rate of 13.9 percent, 19.5 percent, and 18.2 percent, respectively. Differences between issuers and non-issuers are also sizable when looking at turnover growth rates. Turnover grew at a rate close to 17 percent per year for the median issuer, compared to 12.9 percent for the median non-issuer. For the growth in employees, although issuers of equity, bonds, and syndicated loans seem to have quantitatively higher rates than non-issuers, only the 13

17 difference for bond issuers is statistically significant. In particular, the growth in employees for the median firm that issued bonds was 7.7 percent per year, compared to only 0.06 percent for the median non-issuer. Issuing and non-issuing firms seem to significantly differ with respect to their liability structure. First of all, issuing firms tend to be more leveraged than non-issuing firms, especially those that issue either bonds or loans. While the leverage of the median non-issuer firm stood at 35.9 percent during our sample period, a firm that issued equity, bonds, or syndicated loans had a leverage of 44.2 percent, 60.4 percent, and 50.3 percent, respectively. Issuers also tend to have a longer-maturity liability structure. The share of long-term debt in total liabilities for the median non-issuer stood at 9.5 percent, whereas equity, bond, and syndicated loan issuers had 14.3 percent, 39.4 percent, and 38.7 percent of long-term debt over liabilities, respectively. Looking at profitability for issuing and non-issuing firms, the findings are more ambiguous. According to the firm s ROE results, only bond and loan issuers seem to have significantly higher ratios than non-issuers, while the ROE for equity issuers is lower but insignificant. The ratio of retained earnings to assets show qualitatively similar results, although the estimates are barely significant. Comparisons of ROA, on the other hand, show that all type of issuers have lower returns on their assets than non-issuers. 5. How Does Firm Size Evolve for Issuing and Non-issuing Firms? In this section we look at how firm dynamics are related to their capital raising activity. In particular, we look at ex ante and ex post differences in the assets, turnover, and the number of employees of issuing firms, compared to a control group of non-issuers. We examine whether firms that issue equity, bonds, or syndicated loans grow faster than non-issuing firms. Moreover, we investigate how the relative performance of both issuers and non-issuers evolve over the whole FSD. 14

18 We use information on firm characteristics for 2003 and 2010 for all firms in the sample, and estimate mean difference-in-differences regressions on a constant, a dummy variable for issuing firms, a dummy variable for 2010 observations, and the interaction term of those two variables. To assess the evolution of firm size for issuers and non-issuers, we estimate the regression using the logs of assets, turnover, or the number of employees as dependent variables. The estimated coefficients from the difference-in-differences analysis are reported in Table 4 and should be interpreted as follows. The constant term reflects the average size of non-issuing firms in The issuer dummy coefficient measures the size of issuers relative to non-issuers in The 2010 dummy variable gauges the total growth of non-issuing firms between 2003 and The interaction term measures the additional growth of issuers, between 2003 and 2010, relative to non-issuers. That is, whether issuers grow on average more compared to non-issuers. The results in Table 4 show that, on average, firms that issue equity, bonds, or syndicated loans are larger ex ante than non-issuing firms. The differences are also economically significant. The estimates suggest that non-issuers in the Arab region had, on average, about $73 million in total assets in 2003, while firms that issued equity had $150 million in assets about 105 percent more. Bond and loan issuers were even larger in 2003, holding $3.5 billion and $2.1 billion in assets, respectively. Similar qualitative results are found for turnover. In particular, while non-issuers had a turnover of $20 million in 2003, the values were almost $27 million for equity issuers and around $250 million for debt issuers. Looking at the number of employees, the estimates are only statistically significant for bond and syndicated loan regressions. Compared to non-issuers in 2003, bond and loan issuers had around 286 percent and 479 percent more employees, respectively. Issuing firms not only start larger than non-issuing firms, but they also grow much faster. The coefficients for the 2010 dummy are positive and statistically significant in all the specifications. The estimates imply that, on average, non-issuers have expanded their total assets, turnover, and number 15

19 of employees by around 81 percent, 74 percent, and 25 percent, respectively between 2003 and Importantly, the coefficients for the interaction term imply that issuing firms tend to grow even faster than non-issuers. That is, issuing firms are not only initially larger, but there is actually ex post divergence in firm size between issuers and non-issuers. Even after taking into account the initial size differences, the estimates imply a sizeable additional growth for issuing relative to non-issuing firms between 2003 and For example, equity issuing firms had an additional expansion of around 61 percent in their total assets, 63 percent in turnover, and 48 percent in the number of employees. The additional growth rates have been especially high for bond and syndicated loan issuers. For example, the additional expansions in the value of total assets were around 74 percent and 96 percent for firms that issued bonds and syndicated loans, respectively. To further assess the dynamics of firm size for issuing vis-à-vis non-issuing firms across the entire distribution of firm size, we estimate four probability density functions that capture the FSD. More specifically, we estimate two kernel density functions for 2003 (one for issuers of either equity, bonds, or syndicated loans and one for non-issuers) and two analogous ones for The distributions are estimated for the logs of assets, turnover, and the number of employees as proxies for firm size. The results are presented in Figure 7 for equity issuers, Figure 8 for corporate bond issuers, and Figure 9 for syndicated loan issuers. Consistent with the previous analysis, three main findings emerge from the FSD dynamics. First, the 2003 distribution of issuers falls to the right of non-issuers across every firm size decile, implying that issuing firms are typically larger ex ante than non-issuing firms. Second, the FSD for both issuers and non-issuers shifted to the right between 2003 and 2010, implying that firms grew over this time period. Third, the distribution of issuing firms shifted more to the right compared to that of non-issuing firms, indicating that the former grew relatively faster. This is especially true for the FSD of bond and syndicated loan issuers, for which the distributions stand even farther to the right. These patterns complement the previous results from the 16

20 difference-in-differences analysis and confirm that the results not only hold for the mean firm, but issuers tend to be larger ex ante and grow even faster than non-issuers along the whole FSD. 6. Conclusions During the period of fast expansion in Arab financial markets since the early 1990s, to what extent have firms used equity, bonds, and syndicated loans to obtain financing? How many and which firms issued securities in those markets? How did the assets, turnover, and number of employees of issuing firms evolve relative to non-issuing firms? Despite their importance, these questions have never been previously researched for firms in the Arab world. Three main patterns are documented in this paper. First, between 1991 and 2014 there has been a fast expansion in the total amount raised in equity, bond, and syndicated loan markets by firms in the Arab region. The overall evidence suggests that while the level of activity in equity markets stands well with respect to other regions in the world, bond market activity lags behind. Nevertheless, the relative importance of bond financing has increased over time, especially since Capital and syndicated loan market activity in the Arab region exhibit considerable heterogeneity across its member countries, with the GCC nations capturing percent of the total market activity. Moreover, Arab debt issuances seem to be mostly funded from abroad, display lengthy maturity structures, and entail low levels of credit risk. Second, the documented market expansion has not been limited to an increase in the intensive margin, as the number of issuing firms has substantially risen, indicating an expansion in the extensive margin as well. In particular, the total amount raised (number of issuers) has increased by a factor of 21 (15) in equity markets, 22 (6) in bond markets, and 5 (3) in syndicated loan markets. Moreover, firm concentration has also shown some decline, as indicated by the decreasing proportion of capital raising activity captured by the top-5 and top-20 issuing firms. 17

21 Third, issuers of either equity, bonds, or syndicated loans are larger and grow faster than nonissuers in terms of total assets, turnover, and the number of employees. Issuers also tend to be more leveraged and hold more long-term debt, compared to non-issuers. All these patterns are more evident for debt issuers, and especially bond issuers. Moreover, while issuers tend to be larger ex ante than non-issuers, the size gap seems to widen over time. The reported differences hold over the FSD. The findings in this paper imply that the development of bond and syndicated loan financing in the Arab region has mostly been funded by international investors and banks. The reliance of Arab countries firms on international markets to obtain debt financing might make their economies more prone to external shocks. Moreover, most of the funds raised with bonds and loans by firms in Arab countries are denominated in foreign currency. Debt denominated in foreign currency can be risky if not properly hedged, as capital flight and currency depreciations could severely affect the balance sheet of Arab firms and increase their credit repayment burdens. However, this risk might be mitigated by the fact that Arab firms issue very long-term corporate bonds and loans compared to other regions in the world. These results also suggest that relatively smaller firms in the Arab world might be very constrained from issuing securities in debt markets, given the high costs of issuing internationally and the illiquidity of their domestic markets. To meet the liquidity and size requirements of international buyers, the minimum deal size abroad is usually larger than in domestic markets. The international issuance of securities also includes high legal costs to meet international regulations and international rating fees (Zervos, 2004). Given the additional costs associated with international issuances, firms issuing in international markets are typically larger than domestic issuers. Although it is difficult to speculate, one could argue that Arab countries could benefit from further developed domestic bond markets. Well-developed bond markets could allow firms to access alternative sources of funds other than bank finance, promoting a more inclusive and broader use of 18

22 long-term finance to the extent that the entry cost is small. Moreover, they could increase the competitive pressure on banking systems. As a starter, well-developed government bond markets could be considered as a cornerstone for the development of domestic corporate bond markets, as they could act as benchmarks for bond pricing and to create the necessary infrastructure for trading. However, they also might have crowding out effects on the private sector. Furthermore, by boosting the development of domestic bond markets, less reliance could be placed on international markets, which could also reduce the vulnerability of Arab nations to external shocks. Finally, the findings in this paper suggest that while only a small number of firms issue equity, bonds, and syndicated loans in the Arab region, they do not just do so to shape their capital structure, but also to finance investment opportunities and grow. This is important as issuing firms account for a significant part of the total business investment. For example, Farrant et al. (2013) report that bond issuers account for around one-third of the total investment in the United Kingdom. Accordingly, these firms could have a big economic impact, with arguably extensive spillover effects over the rest of the economy. These results suggest that a wider availability of external finance might allow Arab economies to grow faster. 19

23 References Abu-Bader, S., Abu-Qarn, A.S., 2008a. Financial Development and Economic Growth: Empirical Evidence from Six MENA Countries. Review of Development Economics 12(4), Abu-Bader, S., Abu-Qarn, A.S., 2008b. Financial Development and Economic Growth: The Egyptian Experience. Journal of Policy Modeling 30(5), Acharya, V., Cecchetti, S., De Gregorio, J., Kalemli-Ozcan, Ş., Lane, P., Panizza, U., Corporate Debt in Emerging Economies: A Threat to Financial Stability? Committee for International Policy Reform, Brookings Institution, Washington, DC. Al-Malkawi, H., Financial Development and Economic Growth in the UAE: Empirical Assessment Using ARDL Approach to Co-integration. International Journal of Economics and Finance 4(5), Al-Zubi, K., Al-Rjoub, S., Abu-Mhareb, E., Financial Development and Economic Growth: A New Empirical Evidence from the MENA Countries, Applied Econometrics and International Development 6(3), Arab Monetary Fund, Arab Capital Markets Performance, Quarterly Bulletin for 2015 Q2. Retrieved from: Beck, T., Demirgüç-Kunt, A., Small and Medium-Size Enterprises: Access to Finance as a Growth Constraint. Journal of Banking and Finance 30(11): Beck, T., Demirgüç-Kunt, A., Martínez-Pería, M.S., Bank Financing for SMEs around the World: Evidence Across Countries and Bank Ownership Types. Journal of Financial Services Research 39(1): Beck, T., Levine, R., Stock Markets, Banks, and Growth: Panel Evidence. Journal of Banking and Finance 28(3), Bekaert, G., Harvey, C., Lundblad C., Does Financial Liberalization Spur Economic Growth? Journal of Financial Economics 77(1), Blanc-Brude, F., Ismail, O., Who Is Afraid of Construction Risk? EDHEC-Risk Institute, Paris. Bolbol, A., Fatheldin, A., Omran, M., Financial Development, Structure, and Economic Growth: the Case of Egypt, Research in International Business and Finance 19(1), Bolbol, A., Omran, M., Investment and the Stock Market: Evidence from Arab Firm-Level Panel Data. Emerging Markets Review 6(1), Cortina, J.J., Didier, T., Schmukler, S., How Long Do Corporates Borrow? Evidence from Global Bond and Loan Issuances. Working Paper, World Bank, Washington D.C. 20

24 de la Torre, A., Martínez-Pería, M.S., Schmukler, S., Bank Involvement with SMEs: Beyond Relationship Lending. Journal of Banking and Finance 34(9): Demirgüç-Kunt, A., Maksimovic, V., Law, Finance, and Firm Growth. Journal of Finance 53(6): Didier, T., Levine, R., Schmukler, S., Capital Market Financing, Firm Growth, and Firm Size Distribution. NBER Working Paper 20336, National Bureau of Economic Research, Cambridge, MA. Didier, T., Schmukler, S., The Financing and Growth of Firms in China and India: Evidence from Capital Markets. Journal of International Money and Finance 39, Elsafti, A., Financial Sector Reforms in the Arab Countries. Economic Studies, Arab Monetary Fund, Abu Dhabi. Engel, E., Fischer, R., Galetovic, A., Finance and Public-Private Partnerships: A Roadmap. In Financial Flows and Infrastructure Financing. Edited by A. Heath and M. Read. Reserve Bank of Australia, Sydney. Falahaty, M., Hook, L. S., The Effect of Financial Development on Economic Growth in the MENA Region. Journal of Economic Cooperation and Development 35(3), Farrant, K., Inkinen, M., Rutkowska, M., Theodoridis, K., What Can Company Data Tell us About Financing and Investment Decisions? Bank of England Quarterly Bulleting, 53 (4), Finger, H., Gressani, D., Bolstering Financial Stability and Development. In Toward New Horizons: Arab Economic Transformation amid Political Transition, International Monetary Fund, 40-60, Washington, DC. Henry, P.B., Stock Market Liberalization, Economic Reform, and Emerging Market Equity Prices. Journal of Finance 55(2), IMF (International Monetary Fund), Corporate Leverage in Emerging Markets, a Concern? Chapter 3 in Global Financial Stability Report, Washington, DC. Kar, M., Şaban, N., Ağır, H., Financial Development and Economic Growth Nexus in the MENA Countries: Bootstrap Panel Granger Causality Analysis. Economic Modelling 28(1-2), Levine, R., Finance and Growth: Theory and Evidence. In Handbook of Economic Growth, edited by P. Aghion and S. Durlauf, Edition 1, volume 1, chapter 12, Amsterdam: Elsevier. Levine, R., Zervos, S., Stock Market Development and Long-Run Growth. World Bank Economic Review 10(2): Levine, R., Zervos, S., Stock Markets, Banks, and Economic Growth. American Economic Review 88(3):