«Florence Eid florence.x.eid Middle East/North Africa Economist Emerging Markets Research & Strategy JP Morgan 125 London Wall London ...»
Institutions, Knowledge Assets & Financial Contracting:
Private Equity in an Emerging Market∗
Middle East/North Africa Economist
Emerging Markets Research & Strategy
125 London Wall
London EC2Y 5AJ
We thank all those who contributed to this paper especially Josh Lerner (HBS) and Antoinette Schoar
(MIT Sloan) for sharing their version of the questionnaire used, and Jeff Nugent and Mine Cinar for comments on a previous version. Dr. Jaroudi (Islamic Finance House) and Dr. Bahaa El Din (Investment Authority of Egypt) for consultations on Islamic finance, Dr. Omran (Ministry of Investment, Egypt) for his efforts in helping access the Egyptian market. We also thank all the firms that provided support and information on contracts: Abraj Capital (UAE), IT-Ventures (Egypt), EFG-Hermes (Egypt), Injazat Capital (UAE), Audi Saradar Group (Lebanon), Jordan Islamic Bank (Jordan), Stratum W.L.L. (Bahrain), Foursan Capital (Jordan), and Capital Trust Middle East (Lebanon), SEDCO (Saudi Arabia), The Olayan Group (Saudi Arabia), Al-Malaz Group (Saudi Arabia). As this draft is being finalized, the database compiled continues to expand and further acknowledgements will be due. This work would not have been possible without the valuable trust of private equity actors in the MENA region.
Abstract We draw on recent work in law and finance to examine determinants of structure and size of private equity contracts in the Middle East/North Africa region. The roles of firmspecific knowledge and ownership factors, and country-specific legal regime and economic environment factors are examined. While some sections of contracts are “hard-wired” to the investment environment of local markets, others take the form of bilateral agreements. Exit rights and other protective covenants not “indigenous” to local legal regimes are in use, signaling rapid knowledge asset transfer through this industry, to a region where this stock of know-how is recent.
Research Question Most financing in the developing world, including the Middle-East-North-Africa (MENA) region, is of a private nature while most research remains biased toward public financial markets. This paper draws on recent work in law and finance1 to study how private equity contracting responds to legal regimes, and the impact of that on valuation and size of financings. Empirical work on this question in emerging markets has included Latin America and Eastern Europe, but not the MENA region. We investigate whether firm level factors of a private equity group (knowledge asset base2 and percentage ownership) affects the structure of contracts in MENA or whether the contracting choice is “hard-wired” to investment environment factors of the country where the investment is taking place. We use two sets of explanatory variables: (1) external variables that relate to the investment environment in the country where a contract is drawn3, and; (2) internal ones, which are under the control of the private equity group.
Research Motivation We believe that unique characteristics of the MENA region today provide an excellent testing ground to examine whether external or internal factors affect local contracting
choices in the private equity sector:
We draw mostly on the work of Lerner and Schoar (2004), that of Kaplan, Martel and Stromberg (2003), as well as LaPorta, et al (1997, 1998, 2000).
We measure the experience of a firm by taking the fraction of the investment team with work experience in a U.K. or U.S. private equity group. This is not to be confused with the learning effect, which is gained by repeated investments in the MENA markets.
Throughout this paper we refer to the level of economic development of a country (GNI per capita), rule of law index (compiled by the World Bank) and origin of the legal system as the “investment environment” of a country.
1. Unprecedented increases in levels of liquidity in the region since 2001 have contributed to making the private equity sector the fastest growing asset class in the region, with a growth rate hovering around 300% over the past three years, and similar projections for the next three years. A statistically significant sample of contracts is now available as a unit of analysis;
2. The relative size of the private equity sector in the MENA region, vis a vis the overall capital market, is approximately 1%, a proportion similar to the relative size of the sector in the United States, for example.
3. In addition to its financial role, the private equity sector has been shown all over the world, including MENA, to have a central economic role, particularly in employment generation. In MENA there is evidence that employment generation rates through the private equity sector can be up to five times the average labor market expansion rates (Eid 2005). MENA’s most pressing challenge today is employment generation.
The Data The private equity sector is relatively new in MENA, with the oldest contracts dating from 10 years ago. The process of collecting data from private equity actors in MENA was a lengthy first-time experience for this sector. Using a questionnaire adapted from the work of Lerner and Schoar (2004, henceforth LS) and adjusted to regional specificities, we compiled a database of 86 contracts from 9 different private equity groups in 6 countries of the region. Twenty-two of these countries are drawn from common law countries while the 64 remaining ones are drawn from civil law countries, reflecting the relative distribution of the contract over these regimes in the region. The sample of contracts is representative, and constitutes approximately 20% of the population of contracts in MENA4. Large differences in GNI per capita exist in the region, between $1,087 and $20,703 -- a range interesting for empirical work.
Previous Work and Findings Research in law and finance has reached the following conclusions on the size of capital and debt markets given how well property rights of investors are protected. The presence, on average, of larger capital markets in common-law countries than in civil-law countries has been fairly conclusively related to the fact that the former countries protect property rights better. When it comes to financial contracts however, two sets of results have been obtained to date, derived from slightly different empirical data. While some research (Kaplan, Martel and Stromberg 2003, henceforth “KMS”) finds that experienced firms can contract around legal differences and replicate first-best5 contracts, other research (LS) finds that the structure of the legal system affects private contracts in a way that cannot be undone through private (bilateral) solutions.
Our Work and Findings While we follow closely the work of LS and KMS in collecting data and in the choice of explanatory variables (GNI per capita, rule of law index and the origin of the country’s legal system), our work diverges when it comes to controlling for experience.
KMS control for experience through whether the lead VC has previously syndicated with The full size of the database compiled to date is 139 contracts, constituting 30% of the private equity sector in the MENA region. This larger database includes private equity contracts drawn under Islamic Sharia, and is the subject of a separate paper that takes a first look at the role of financial contracting under religious law.
U.S.- style contracts as per KMS.
a U.S. VC, whether the VC is located in the U.S. or whether the VC is larger and older.
LS only control for whether a firm is located in the U.S. or U.K.
In a fact-finding survey conducted as part of previous research (Eid 2005), we were able to conclude that much of the knowledge asset base in the MENA private equity sector comes from investment team members with experience in U.S or U.K. firms. We observe that when they have worked on PE contracts abroad, investment team members tend to replicate these practices in the region. We therefore take the fraction of the investment team with experience in a U.S. or U.K. private equity group. On one end of this continuum, where the entire team has U.S. or U.K. experience this would be as if the firm were located in the U.S. or U.K., in line with LS. We also control for the relative size of investment6. We observe that when unable to protect itself through rights and covenants drawn from the law of the land, a private equity group tends to increase its share of ownership in a firm to gain control rights through board representation.
Cross legal system comparisons show that the investments done in common law countries in MENA are not significantly larger with a mean of $6.55 million and standard deviation of 6.51 as compared to $3.02 and a standard deviation of 5.22 in civil law countries. A simple two tailed Z test for proportions shows that 9 out of the 15 rights and features we test for, other than size and type of securities used, are statistically equal.
We perform multivariate regression8 by assigning the values of 1 and 0 to whether a right or covenant is being used or not (same coding technique as LS and KMS). We then run a regression by considering all the explanatory variables and compare to a
regression using external variables only. We reach the following conclusions:
We distinguish between the anti-dilution provision and the right to increase the share of investment pro-rata. In the first we ask whether there is protection for investors from down rounds even if the investor decides not to buy into the new round, while in the second it is simply the right of an investor to buy a portion of the newly offered securities equal to their original percentage ownership.
We limit our independent variables to a range between 0 and 1 by taking the natural log of GDP per capita and dividing it by ten. This allows us to compare signs as well as sizes of coefficients obtained in multiple regression equations.
1. Evidence of “hard-wiring”. In the contracts drawn to date, the types of securities used have been largely hard-wired to the investment environments of MENA countries, and have had little to do with the level of experience of investment firms. Take for example the results we obtain when testing for common stock, straight debt and convertible preferred shares (we also test for ordinary preferred shares in the longer version of this paper). We find that common shares are used in countries with low GNI per capita and where the rule of law is less established.
The coefficient for legal origin, though positive, is not large enough compared with those for rule of law and GNI to imply that this variable is a strong determinant of why common shares are used. Moreover the level of experience of a private equity group plays no role in the choice of security when it comes to common stocks, which is further evidence of hardwiring.
When testing for convertible preferred shares however, we find that legal origin and the rule of law index are stronger explanatory variables than GNI per capita especially when we remove PEG -related factors. Debt is used in civil law countries with high GNI per capita and the rule of law index is high. While both types of securities tend to be used by more experienced groups, the issue of hard wiring to legal origin is clear because when working in common versus civil law countries PEGs tend to replace the use of convertible preferred shares with debt.
security used is hard-wired to the local investment environment. The only difference is that we do not find that the origin of law is the reason why the securities are used but rather the overall environment, particularly the level of