Creditor Protection and Valuation of Banking Systems

Transcription

1 Creditor Protection and Valuation of Banking Systems The Author December 1999 Department of Economics Some University Abstract There have been few studies that analyze the interaction between law, procurement and its subsequent effects upon financial markets. In this paper we provide some introspection to this link by analyzing the effect of protective legal regimes upon stock market valuation of banks. In particular, we assess the impact of creditors rights and their effectiveness in countries where these rights are truly enforced upon banking system capitalization. First, we find that creditor protection does affect positive and significantly banking-wide valuations. Secondly, we find that the effect of protection to creditors on market value is sizeable. In particular, banking systems valuations increase more than twofold when, in the case of borrowers financial distress: 1) secured creditors are paid first; 2) borrower firms incompetent management is removed or replaced; and 3) there are no restrictions to dispose of borrowers assets to repay creditors. Third and last, creditor protection has no impact if there is no effective judiciary system in place able to enforce creditors rights.

2 1. Introduction Do legal country-specific characteristics such as the prevailing protection to creditors and the efficiency of the judiciary system in a given country affect the way investors assess the value of banks? For instance, why were investors reluctant to buy shares of Mexican banks before Congress passed a new bankruptcy law reform but were instead willing to hold claims in U.S. equivalents? 1 Why are banks shares more valuable in some countries than in others, controlling for size and profitability? What role does the judiciary system play in the assessment of banks assets that may turn sour? Commercial banks play a central role as core intermediaries of the financial system in any country. In the modern world, banks are the oil in the engine of the economy. They provide liquidity to firms and households, increase the depth and breath of capital markets, and promote efficiency in the allocation of resources. In particular, banks channel resources to finance projects by holding debt claims (i.e., loans) in their clients by recurring to their depositors money. When banking systems collapse or banks operate ineffectively, their ability to provide financing is largely curtailed. Most importantly, since banks generally hold portfolios across various industries, any malfunctioning of the banking system translates into limited financing to the private sector thus jeopardizing economic growth in developed and underdeveloped economies alike. Generally, commercial banks represent the largest pool of creditors in a single economy. Since banks financial viability is so crucial to the overall economic well-being, one might think that this fragility should be accounted for by investors, in addition to their performance. If we believe the Efficient-Markets Hypothesis, then all public information concerning the financial soundness of banks should be properly incorporated in their stock prices. More importantly, investors should be willing to pay more for assets such as equity and debt because, in better protective legal regimes, the chances of firms filing for bankruptcy or being expropriated are lower. Then if investors are aware of the institutional determinants of banking performance in a given country, banks stock prices should properly reflect the particularities of such frameworks. 1 Lopez-de-Silanes (2000) describes in detail the changes incorporated to the new bankruptcy and insolvency law reform passed in Mexico until early in

3 In this paper, we analyze the effect of two important institutional determinants of private sector financing on the valuation of commercial banks. First, we hypothesize that investor protection and in particular, creditor protection, affects investors valuations of banks and overall banking systems in general. Secondly, we explore the effect that the exogenous legal environment, namely the efficiency of the judiciary system, might have on these same valuations. This paper follows on previous research on the more general relationship between financial development and economic growth. In particular, this paper can be seen as an application to banks of La Porta et al. (2002), where the authors model and analyze the effect of shareholder protection upon stock valuation of non-financial firms. Previous research has focused on the effects of corporate governance upon financial markets. In particular, some authors have tried to assess the impact of the quality of investor protection on the ownership patterns and structure of firms, their capability for external financing and, more generally, the degree of financial development across countries (La Porta et al. 1997, 1998, 2001, 2002). This research shows empirically that countries with weak investor protection and inefficient judiciary regimes have underdeveloped financial markets. Other authors have explored the regulatory side of commercial banking performance. For example, Barth et al. (1999) argue that restrictions on banks to incur in the securities underwriting business affects the performance (in particular, net worth) of financial systems. Therefore, tougher restrictions on banks translate into increased inefficiency, higher transaction costs, lower profitability, and reduced value. In this paper, we neglect the regulatory aspect of banks valuations and focus on the more interesting institutional side. More directly related to this research is the study of the effect of corporate insolvency procedures on firm behavior in Asia (Hussain and Wihlborg, 1999). In this case, the authors argue that firms from countries with the poor insolvency procedures were hit harder during the 1997 Asian financial crisis and exhibited higher default rates. According to this hypothesis, poor bankruptcy procedures including protection of creditors induced a larger number of firms into financial distress after the severe macroeconomic shock following the Asian crisis. The authors find a small and negative relation between creditor protection and the share of non-performing to total bank loans. However, their study is limited in the sense that the set of countries analyzed have a more or less homogeneous degree of creditor protection. 3

4 Despite all these efforts, the relationship between creditor protection and banks performance has been subject of no study. In the case of commercial banks, credit may be seriously hampered by an unfavorable milieu towards creditors due to lousy or flawed bankruptcy procedures. Furthermore, inefficient and corrupt judiciary frameworks seriously reduces the chances of banks to recover their assets in the case of distress, whether on mass or individual basis. In this sense, the law as written on paper and its enforcement play a fundamental role in the valuation of banking systems worldwide. The purpose of this paper is to construct an empirical proof of this hypothesis. The paper is organized in the following manner: Section 2 presents the model estimated for our empirical proof. Section 3 describes our methodology and the data used. Section 4 draws on the results and Section 5 concludes. 2. Model The literature on financial intermediation and corporate governance provides little if any hints at all as of how creditor protection should be modeled within a valuation perspective. We follow the path-breaking work of La Porta et al. (2002) whose model gives us a testable hypothesis claiming that firms in more protective legal regimes should have higher valuations. 2 The way it works is that investors incorporate the higher legal protection into their assessment of financial assets such as equity and debt through higher valuations. Empirically, it is even more difficult to assess which should be the correct specification between creditor protection and banking system valuations. For this reason, we undertake a simple and orthodox approach. We use fixed-effects regression analysis to estimate coefficients of institutional measures on normalized market capitalization of banking systems across a number of countries and years. The specification is formalized as follows. Let y i, t be the normalized market capitalization of the banking system of country i at time t. Denote country and time fixed-effects as respectively. Define C i, t as the normalized cash flows and i t α i and γ t, D, as the normalized size measured by deposits of the banking system of country i at time t. Similarly, define S i as the prevailing 2 La Porta et al. (2002) use Tobin s q-ratio as a valuation proxy. 4

5 shareholder protection in country i (see La Porta et al., 1998). We will refer to C i, t, D i, t and i the control variables hereafter. To complete the notation of our model, let δ l be the coefficient corresponding to the l-th control variable, for l = Finally, let β k be the coefficient corresponding to the k-th measure of creditor protection, for k = The reduced equation reads as, where, 4 y i, t = α i + γ t + δ1ci, t + δ 2 Di, t + δ 3Si + βk creditori + ε i, t k= 1 S as creditor i denotes the country-specific measures of creditor protection. The above equation shows the model that we use to fit our data. 3. Data and Methodology Our study covers 40 countries, from Argentina to Zimbabwe. For this paper, we assembled a database from three different sources. First, we used data from two previous studies: La Porta et al. (1998) on law and finance and Barth et al. (1999) on banking systems around the globe. Secondly, we used Primark s Worldscope database of financial information on firms trading in the main stock markets around the world. Third and last, we consult the International Country Risk Guide database on institutional and political outcomes across different countries. From La Porta et al. (1998) we take as given our measures of shareholder and creditor protection. We assume these measures as exogenous and therefore its use is valid in our empirical construction. Measures on shareholder and creditor protection are basic rights embedded in the law and come from the codification of company, bankruptcy, and reorganization laws prevailing in the countries under study. More specifically, the measures of creditor protection that interest us are automatic stay on assets, secured creditors first, reorganization restrictions, and entrenched management (see Appendix for exact definitions of variables). On the other hand, shareholder protection is an index that encompasses several shareholder rights (see Appendix). From Barth et al. (1999) we check for robustness of the measures of shareholder and creditor protection by comparing these with their regulation-equivalent outcomes. 3 From ICRG we borrow one variable, rule of law, which captures the law and order tradition in a given country via recurrent surveys across 3 Previous research has shown that regulation restrictions on banking systems are positively correlated with investor protection (Barth et al., 1999). 5

6 time. We use this variable to partition our sample into law prevailing systems and non-law prevailing systems. Finally, from the Worldscope database we extract panel data (at the firm level) for the years This database comprises all balance-sheet items for publicly-traded traded banks in more than 60 countries. However, we restrict ourselves to 40 countries due to missing observations. For each commercial bank, we compute normalized market capitalization ratios using the cumulative market value of banks shares and their total assets. We then compute a single measure of market capitalization for an entire banking system per country per year using the 10-largest commercial banks operating in the countries. If fewer banks were available in the database for a given country or year, we use all of the remaining observations to compute the overall banking system market capitalization ratio. In the same way, we compute measures of the cumulative mean and median profitability and size of each banking system. We obtain the best results by using cumulative control variables rather than means and medians. We thus omit reporting means and medians results in the paper. Next, we fit a model of country and year fixed-effects to estimate the direction and magnitude of better-quality creditor protection effect on banking system valuations. We encounter the alternative specification of random-effects, especially because stock market valuations might be correlated across countries. 4 However, we reject and do not report the hypothesis of a random-effects specification by performing a Hausmann test. We make the standard econometric assumptions for fixed-effects models, namely that the disturbance term is distributed normal with zero mean and constant variance. Nevertheless, we do not take as given the typical no-serial correlation assumption and instead use robust standard errors in our results. Additionally, we report Bonferroni-adjusted p- values for all coefficients, including the control variables. 5 4 It has been shown that stock market prices are highly correlated in a world of relative capital mobility (Edwards, 2002). Consequently, one might question our fixed-effects specification since poor performance of a country s stock market might be translated into better performance for some other country. A firsthand view would propose instead a randomeffects model, thereby voiding our assumption of conditional heteroscedasticity of the disturbance term. In our empirical test, however, this is not the case. 5 Bonferroni-adjusted p-values are calculated by multiplying the original p-values of coefficients using the null hypothesis of these coefficients being equal to zero by the number of inferen ces performed. 6

7 4. Results Table 1 summarizes the main results of the paper and shows that creditor protection is indeed an important determinant of banking valuation in some countries. The table shows three different regressions using different samples. The first regression runs only countries with law-prevailing judiciary systems, defined as countries with a rule of law score greater or equal than the sample median. The second regression runs only countries with non-law prevailing judiciary systems, defined as countries with a rule of law score less than the sample median. The third is the pooled regression and runs the complete set of countries. In general, the control variables are usually significant and their effect sizeable upon valuation. Except for countries with non-law prevailing judiciary systems, the coefficient of normalized cash flows affects positively two to threefold the market valuation of banking systems. Conversely and except for countries with non-law prevailing judiciary systems, the coefficient of normalized deposits (or size) affects negatively and, at the most, twofold the market valuation of banking systems. These results hold trivially since cash flows are the typical determinant of stock prices in the standard theory of valuation and deposits (or size) represent debt claims upon banks, which further reduce their expected market value. In addition, shareholder protection seems to increase from 20 to nearly 50 percent depending on the sample overall banks worth in the stock market. 6 Nevertheless, this result does not hold again in countries with non-law prevailing judiciary systems. In particular, the pooled regression shows mixed effects. Coefficients for automatic stay on assets and reorganization restrictions are not statistically significant but the remaining creditor protection variables tell another story. Secured creditors first or, the ability of banks to recover collateral on their loans before other unsecured creditors when borrowers face distress raises, in average, 96 percent the market valuation of banks shares. Similarly, entrenched management or, having the old management guard in the borrower removed or replaced during bankruptcy cases increases the banking system s worth by 93 percent. These results point to an important distinction made by investors and analysts when assessing the role institutionalized creditor rights upon their valuations. 6 We include shareholder protection per se as a control variable because we want to differentiate between the institutional effect of shareholder and creditor rights. 7

8 Yet, the most interesting results emerge when we partition the sample. When we look only at countries with a high rule of law score, the degree of creditor protection gives us a remarkably different picture. This regression shows that all but the reorganization restrictions coefficient are positively and statistically significant at the 1 percent level. The right to recover collateral first (i.e., secured creditors first) raises 1.25 times the valuations of banking systems. Likewise, the right to dismiss incompetent management or the ability to have the borrowers administration intervened by a court (i.e., entrenched management) increases by 52 percent the market capitalization of banking systems. In addition, countries whose laws do not impose an automatic stay on the borrowers assets so that creditors can freely pull out their collateral have banks that are worth, in average, 60 percent more. In contrast, the third regression indicates that creditor protection is irrelevant in countries with a low rule of law score or, equivalently, with non-law prevailing judiciary systems. Looking at the coefficients of creditor protection measures we note that none of them is significant, except for automatic stay on assets and that, only at the 10 percent level. Not even shareholder protection produces the slightest effect on market capitalization ratio and, similarly, the control variables seem to cause no change upon valuations. These results point out to the important link between rights and their enforcement. Apparently, better-quality creditor protection positively affects banking system valuations if and only if the judiciary system in place allows for enforcement of these rights. Finally, to complete the analysis we perform inferences to test whether the coefficients of our creditor protection measures are equal to zero. Using the Bonferroni-adjusted p-values, we reject the null hypothesis that these coefficients are equal to zero and restate the relevance of creditor protection on the valuation of banks Conclusions There have been few studies that focus their attention to the interaction between law and its procurement. In this paper we provide some introspection to this link by analyzing the effect of creditors rights and its effectiveness in countries where they are truly enforced. First, we find that creditor protection does affect positive and significantly banking-wide valuations. Secondly, we find 7 In this case, we multiply all p-values by four, since there are four measures of creditor protection that interest us. Using Scheffe s S-method does not alter the significance of the results for the non-law systems regression. 8

9 that the effect of protection to creditors on market value is sizeable. In particular, valuations increase more than twofold when, in the case of borrowers financial distress: 1) secured creditors are paid first; 2) borrower firms incompetent management is removed or replaced; and 3) there are no restrictions to dispose of borrowers assets to repay creditors. Third, creditor protection is irrelevant if there is no effective judiciary system in place able to enforce creditors rights. These results are robust and hold even controlling in several ways for statistical significance in our data. In addition, there is a public policy recommendation that can be drawn from this paper. Today, emerging countries struggle to build up capital markets. Commercial banks are commonly the spine of such financial frameworks. Appropriate protection to all creditors in the economy, including commercial banks, is a necessary condition for financial markets to prosper. Bankruptcy and reorganization laws are still ineffective and archaic in many of these countries. Reforming these laws in ways such that would include appropriate creditor protection should be a priority in order to increase both the breath and depth of capital markets in such countries. References 1. Barth, J., G. Caprio, and R. Levine. Banking Systems around the Globe: Do Regulation and Ownership affect Performance and Stability? NBER mimeo, November Edwards, S. Controls on Capital Inflows: Do they Work? forthcoming in Journal of Development Economics, January, Hussain, Q. and C. Wihlborg. Corporate Insolvency Procedures and Bank Behavior: A Study of Selected Asian Economies. IMF Working Paper WP/99/135, October La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. Legal Determinants of External Finance, Journal of Finance, vol. 52, no. 3, July La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. Law and Finance, Journal of Political Economy, vol. 106, no. 6, December La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. Courts: The Lex Mundi Project, Mimeo Harvard University, July La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. Investor Protection and Corporate Valuation, forthcoming in Journal of Finance, Lopez-de-Silanes, F. Reforming and Deepening Mexico s Financial Markets, forthcoming Council of Foreign Relations Paper,

10 Table 1. Fixed-Effects Creditor Protection and Valuation Regressions This table presents fixed-effects panel data regressions of the valuation of banking systems on the quality of creditor protection in 40 countries around the world. Robust standard errors are reported in parentheses Bonferroni-adjusted p-values are reported in brackets. The specification estimated is the following: 4 y i, t = α i + γ t + δ1ci, t + δ 2 Di, t + δ 3Si + βk creditori + ε i, t k= 1 market capitalization of the banking system of country i at time t, fixed-effects, δ l are coefficients of banking system-specific control variables, and country-specific creditor protection variables., where y i, t denotes the stock α i are country fixed-effects, γ t are time β k are coefficients of Law prevailing Non-law prevailing Pooled sample systems systems Dependent Variable: Market Capitalization of Banking Systems Cash flows a a (0.7921) (1.0036) (0.6279) [0.0000] [0.2050] [0.0000] Size of banking system a c (0.5827) (0.6948) (0.4656) [0.0000] [0.4080] [0.0560] Shareholder protection a a (0.0740) (0.1217) (0.0577) [0.0000] [0.4290] [0.0000] Automatic stay on assets a c (0.1465) (0.2171) (0.1659) [0.0000] [0.0600] [0.4010] Secured creditors first a b (0.1515) (0.4419) (0.3414) [0.0000] [0.3360] [0.0200] Reorganization restrictions (0.0612) (0.2723) (0.1495) [0.9600] [0.4440] [0.3000] Entrenched management a c (0.1241) (0.4963) (0.3890) [0.0000] [0.7640] [0.0720] Observations R a = significant at 1 percent; b = significant at 5 percent; c = significant at 10 percent 10

11 Appendix. Definitions of Variables Variable Definition Sources Market capitalization The sum of market capitalization of the 10-largest commercial banks operating in a given country normalized by the sum of the cash flows for the same banks. Worldscope database Cash flows The sum of cash flows of the 10-largest commercial banks operating in a given Worldscope database country normalized by the sum of the deposits of the same banks. We use this variable to control in our regressions for profitability of banking systems across countries. Size of banking system The sum of deposits of the 10-largest commercial banks operating in a given country normalized by the sum of the total assets of the same banks. We use this variable to control in our regressions for size of banking systems across countries. Worldscope database Shareholder protection An index aggregating shareholder rights in a given country. The index is formed by adding 1 when (1) the country allows shareholders to mail their proxy vote to the firm; (2) shareholders are not required to deposit their shares prior to the general shareholders meeting,; (3) cumulative voting or proportional representation of minorities in the board of directors is allowed; (4) an oppressed minorities mechanism is in place; (5) the minimum percentage of share capital that entitles a shareholder to call for an extraordinary shareholders meeting is less than or equal to 10 percent; or (6) shareholder have the preemptive rights that can be waived only by a shareholder s vote. The index ranges from zero to six. La Porta et al. (1998) from Company Laws Automatic stay on assets Equals one if the reorganization procedure does not impose an automatic stay La Porta et al. (1998) on the assets of a non-financial firm on filing the reorganization petition. from Bankruptcy and Automatic stay prevents secured creditors, such as banks, from gaining Reorganization Laws possession of their security. It equals zero if such a restriction does not exist in the law. Secured creditors first Equals one if secured creditors are ranked first in the distribution of the proceeds that result from the disposition of the assets of a bankrupt firm. Equals zero if non-secured creditors, such as the government and workers, are given absolute priority La Porta et al. (1998) from Bankruptcy and Reorganization Laws Reorganization restrictions Entrenched management Equals one if the reorganization procedure imposes restrictions, such as creditors consent, to file for reorganization; equals zero if there are no such restrictions. Equals one when an official appointed by the court, or by the creditors, is responsible for the operation of the business during reorganization. Equivalently, this variable equals one if the debtor does not keep the administration of its property pending the resolution of the reorganization process. Equals zero otherwise. La Porta et al. (1998) from Bankruptcy and Reorganization Laws La Porta et al. (1998) from Bankruptcy and Reorganization Laws Rule of law Assessment of the law and order tradition in the country produced by the country risk-rating agency International Country Risk (ICR). Average of the months of April and October of the monthly index between 1982 and Scale from zero to 10, with lower scores for less tradition for law and order. We classify as law prevailing judiciary systems those countries that have a score greater or equal than the median. Non-law prevailing judiciary systems are those with a less than the median score. International Country Risk guide. 11