The Effect of Liberalization of Foreign Direct Investment (FDI) Limits on Domestic Equity Prices: Evidence from the Indian

Banking Sector1

B. V. Phani

Chinmoy Ghosh

John Harding

August 2004 Abstract

Foreign Direct investment (FDI) limits were liberalized in India to allow more than fifty-one percent ownership of private sector banks in February, 2002. Portfolios of private sector and government owned banks posted significant and large value gains surrounding the announcement, the gains by private sector banks being almost double that by government banks. The analyses show that the price increase is higher for smaller banks that have less debt, are less efficient, less productive, and burdened with non-performing assets. We conclude the evidence is consistent with the hypothesis that the valuation gains reflect the vulnerability to and premium of potential takeover of the inefficient banks following the liberalization.

Phani is at the Indian Institute of Management at Calcutta (IIMC), India;e-mail: phani@iimcal.ac.in. Ghosh & Harding are at the University of Connecticut, Storrs, CT 06268, USA; e-mail:Chinmoy.Ghosh@business.uconn.edu, John.Harding@business.uconn.edu. Authors wish to thank Florida International University's Center for International Business Education and Research (FIU CIBER) for their financial support. Authors are grateful to Professor Arkadev Chatterjea at IIMC for his help and encouragement during the early stages of the project. We also acknowledge assistance with data analysis from Rupendra Paliwal, a doctoral student at UConn. We are responsible for all errors.

The Effect of Liberalization of Foreign Direct Investment (FDI) Limits on Domestic Equity Prices: Evidence from the Indian Banking Sector 1. Introduction
The effect of foreign direct investment on the domestic economy has been widely debated in literature, but a consensus opinion has not emerged. Critics have attributed the Asian banking crisis to the growth of foreign direct investment following the liberalization of foreign investment restrictions. Generally, the argument runs that foreign investors create a destabilizing influence on stock prices. Stiglitz (1998) posits that unregulated capital flows render developing economies more vulnerable to fluctuations in supply of international capital. According to Dornbusch and Park (1995), foreign investors tend to follow positive feedback strategies which cause markets to overreact to fundamental changes in value. Radelet and Sachs (1998) attribute the Asian financial crisis to financial panic. Hamann (1999) concludes that currency crises lead to financial crises: collapse in exchange rates lead to the collapse of banks that underestimate exchange rate risk and accumulate vast currency reserves. Several other researchers including Delhaise (1998) blame the Asian crisis on overgenerous and indiscreet lending by banks, especially western banks, and then switching to overly strict lending policies when market turned sour. Kim and Singal (2000) characterize movement of hot money as a major concern with policy makers in developing nations. Hot money investment is highly sensitive to interest rate and future growth expectations, such that adverse changes in these factors result in large changes in international flow of capital which exacerbates the shock, destabilizing the economy. The authors further point out that when markets are integrated, excess volatility in the foreign market induces a similar effect in the domestic market which increases risk premium, and cost of capital, and reduces investment. In addition, opening the market increases the demand for and the value of domestic currency. Appreciation in exchange rate can adversely impact the countrys competitive position for goods and services in the world market. Finally, inflow of excess capital can bring in inflationary pressures. The potential benefits of opening domestic markets to foreign investors cannot be overlooked, however. A major benefit is the opportunity to attract foreign capital. Infusion of foreign capital enhances economic growth [Boyd and Smith(1996), Levine and Zervos (1996)]. More significantly, as Rajan and Zingales (1998), and Stulz (1999) demonstrate, integration with world market through relaxation of foreign investment restrictions reduces cost of capital. Stulz (1999) attributes the decrease in cost of capital to the improvement in managerial monitoring and governance following foreign ownership of domestic stock. The stringent disclosure

requirements foreigners impose improves transparency, enhances monitoring, and disciplines management through increased accountability. Kim and Singal (2000), Bekaert and Harvey (2000), Henry (2000) and Chari and Henry (2002) provide empirical evidence that stock market liberalizations induce growth in private investment and reduce systematic risk of securities that allow foreign equity participation. As Chari and Henry (2002) note, most of the empirical tests of the potential effects of liberalization of stock markets have used aggregate data. With aggregate data, however, only one aggregate stock price revaluation per country can be observed when stock market liberalization occurs, but it is difficult to incorporate firm level information. The authors assert that the gain in stock values that occurs at liberalization is the most direct signal that the policy change has reduced the cost of capital. Stulz (1999) observes that studies employing long time-series data to investigate the impact of liberalization suffer from the requirement that the cost of capital be held constant over the time period. However, if sudden relaxation of investment barriers induces unexpected decrease in cost of capital, stock prices will post unanticipated valuation gains and standard event study tests are more appropriate. Recently, event study tests of valuation effect of liberalization have been conducted by Kim and Singal (2000), Bekaert and Harvey (2000), Henry (2000), and Chari and Henry (2002). The first three studies use aggregate data and find evidence consistent with decreases in cost of capital following liberalization of investment restrictions. The first to use firm level analysis,2 Chari and Henry (2002) identify stock market liberalization dates for several countries and discriminate between firms that become eligible for foreign ownership when the market is liberalized and those that do not. They demonstrate that the valuation gain of stocks following liberalization is a function of the decrease in systematic risk and the cost of capital. Chari and Henry (2002) use monthly data for their analyses. No study to date has reported results on impact of liberalization on stock prices with daily data. Using a unique data set of daily returns of Indian banks, we provide evidence of significantly positive valuation effect associated with liberalization of foreign direct investment rules. On February 16, 2002, the Reserve Bank of India (RBI) announced that foreign entities will be allowed to make direct investment of 49 percent in private Indian banks. By far the most sweeping change in the slow, bureaucracy-laden process of liberalization Indian government has undertaken over the last
Stulz (1999) contends that for countries that have no access to international capital markets, initiation of ADR programs constitute liberalization events. The major impact of the ADR program can be expected to be on the cost of capital of the firm that lists the stock. Using daily returns, Forester and Karolyi (1999), and Miller (1998) find positive returns both around the announcement and listing dates of ADR programs. Miller attributes his finding to relaxation of barriers to international investment following ADR introduction.
2

decade, the RBI decision allowed foreign direct investors and foreign institutional investors a combined ownership of 98 percent of an Indian bank, giving them virtually complete control over operating decisions. An equally-weighted portfolio of private Indian banks registered excess returns of nearly 25 percent over the three days surrounding the announcement. Interestingly, a portfolio of government controlled banks for which the FDI limit was only 20 percent also posted significant gains at the announcement. We interpret the evidence as unambiguous support for Stulzs (1999) hypothesis that liberalization induces a reduction in domestic firms cost of capital eliciting a positive price reaction. Further, we provide evidence that the valuation gain is significantly related to the factors that make an individual bank a potential takeover target. This finding is consistent with the assertion that the reduction of the cost of capital is attributable to the monitoring benefits associated with control of domestic banks shares by foreign entities. The rest of the paper proceeds as follows. In the next section, we focus on the liberalization of the capital market in India with particular emphasis on the developments in the banking sector culminating with the RBI announcement on February 16, 2002. A chronology of the major events over the one-year period from May 2001 to March 2002 is provided. In section 3, we discuss the hypothesis on the impact of liberalization on domestic firms cost of capital and valuation. In section 4, the results of the standard event study analysis surrounding the important dates over the one-year period are reported. A test of the hypothesis that valuation effect of an individual bank is related to its acquisition probability is conducted in section 5. Specifically, the association between the abnormal return and selected firm-specific attributes is explored. Section 6 concludes the paper.

2.

Foreign Direct Investment (FDI) in Indian Banks

The traditional argument against foreign equity participation in domestic companies is

that these businesses often involve national and strategic interests and therefore, operational and strategic control must be retained to prevent a take-over or a buyout [Lam (1997)]. Until 1993, most Indian banks were 100 percent owned by the central government and private investment was allowed only in a handful of private banks formed around the 1940s. Further, foreign banks and financial institutions were allowed only 20 percent ownership stakes in Indian banks. In 1993-94, nine new banks were formed in the private sector and one co-operative bank was converted to a private bank. Banks were permitted to issue Certificates of Deposits (CDs) and offer foreign currency deposits to Non-resident Indians (NRIs) with exchange rate risk borne by the banks.

A major push towards liberalization occurred in 1995-96 when India committed to the World Trade Organization (WTO) recommendations and relaxed the requirement to continue shielding the priority sector from foreign equity participation. For the next five years, changes in the banking sector mainly aimed at allowing banks more flexibility in the design and marketing of products. On May 9, 2001, the Indian central ministry decided to increase Foreign Direct Investment (FDI) limits in private banks from the existing 20 per cent to 49 per cent along with increase in Non-resident Indian (NRI) investment from 40 per cent to 49 per cent. The reaction of the capital market was lukewarm at best. Analysts attributed the markets unenthusiastic response to two factors. First, the market was disappointed that, even under the revised rules, no foreign entity will be able to assume majority control of an Indian bank. It was recognized that significant differences existed between foreign and Indian banks with respect to labor and management policies, work ethics, and culture. These weaknesses can be corrected only if foreign banks are allowed majority control with boardroom implications on the entire bank, said one prominent foreign bank executive. Second, confusion ensued over the interpretation of the 49 percent rule -- investors were unsure if the 49 percent included investments by Foreign Institutional Investors (FII), and non-resident Indians (NRI). To clarify the confusion, the Ministry of Commerce and Industry issued a press note on May 21 specifying that FDI up to 49% from all sources is permitted in the banking sector on the automatic route3 subject to conformity with guidelines issued by RBI from time to time.4 On June 19, 2001, French financial giant BNP Paribus announced that it was exploring the possibility of acquiring an Indian Bank, but only after the government further liberalizes foreign investment norms. The disappointment intensified when Reserve Bank decided on The market reaction was uniformly negative as September 20, 2001 to put a limit on foreign institutional investment into a company at par with sectoral cap for foreign direct investment.
3

Under the Foreign Exchange Regulations Act (FERA) of 1973. most foreign investment was done in India with the prior approval of the Government of India. The New Industrial Policy of 1991 introduced an innovation by way of an Automatic Route not requiring approval from the government. It was limited to 35 industries and was subject to an investment ceiling of 51 per cent and still required approval of the Reserve Bank. Subsequently, automatic route has been expanded and at present, it is allowed in all industries except for certain select industries/sectors to issue shares to foreign investors up to 100 per cent of their paid up capital. Any company in these selected sectors required specific approval of the central government and in several of these sectors, foreign investors cannot invest beyond a certain percentage of the paid up capital. Banking was one of these protected sectors.
4 Banking was only one of several other industries covered in this memo. Other industry sectors that benefited from significant liberalizations at this time were manufacture of drugs and pharmaceutical, airports, defense, development of integrated townships, including housing, commercial premises, hotels, resorts, city and regional level urban infrastructure facilities such as roads and bridges, mass rapid transit systems; and manufacture of building materials, hotel and tourism, courier services, Mass Rapid Transport Systems in all metropolitan cities, including associated commercial development of real estate, and telecom services.

analysts expected the decision to adversely affect companies in sectors where the cap was at 49 percent or lower, ruling out any potential takeover attempts. Specifically, for banking where the cap was 49 per cent, foreign institutions would have to shed their investments to bring down the total foreign investment to the sectoral cap. On November 28, 2001, it was reported that CitiBank had evinced interest in Centurion bank which was looking for a buyer for the 26.2 per cent stake pledged with the bank by its promoter. CitiBank had earlier indicated that they would consider acquisitions in the Indian banking sector if regulations permitted. Talks did not proceed further, however, as they appeared to be interested in majority control which was not permitted under the current law. On February 1, 2002, Bank Brussels Lambert (BBL) informed the regulators of their intention to assume management control of Vyasa Bank in the private sector. BBL already held a 20 per cent stake in Vyasa Bank and wanted to buy out another 28.1 per cent owned by a promoter. If allowed, the total foreign holding in Vyasa Bank including the BBL stake and other institutional holding would exceed the RBI stipulated limit of 49 per cent. Bank stocks surged in anticipation that if the BBL proposal went through, it would open the door for other foreign banks contemplating acquisitions. On February 12, Mr. Kenneth Dunn, the US Deputy Treasury Secretary, who was visiting India at that time, urged the Governor of RBI, Mr. Bimal Jalan, to immediately take measures to promote foreign investment in India. On February 16, 2002, the Reserve bank, in a consolidated notification, laid to rest all doubts raised with regard to FDI in the banking sector by releasing the following decision: Foreign banks having branch presence in India are eligible for FDI in private sector banks, subject to the overall cap of 49 per cent (by way of FDI) subject to the approval of RBI. For the purpose of determining the ceiling in private sector banks under the automatic route following categories of shares will be included initial public offerings, private placements, fresh issuances of American depository receipts (ADRs)/global depository receipts (GDRs), and acquisition of shares from existing shareholders. However, FDI in the state-run banks, including the State bank of India will be permitted only upto 20 per cent. Automatic route does not allow transfer of existing shares in a banking company from residents to non-residents, however. Further, issue of fresh shares is not allowed to those foreign investors who have a financial or technical collaboration in the same or allied field. That would require approval from the Foreign Investment Promotion Board (FIPB) for FDI. While the decision appeared to facilitate formation of foreign majority stakes, RBI stipulated that no person holding shares, in respect of any share held by him, shall exercise voting rights in excess of 10 percent of the voting rights of all the shareholders. For public sector banks, the voting right was

capped at 1 percent of all shareholders. The limited voting rights does put serious restrictions on foreign banks ability to push their agenda without broad shareholder support, but as one foreign senior banker said, this is not so much an issue. Once you have management control and are making board decisions, the voting rights on equity shares hardly matter, (Times news Network, February 18, 2002). The Reserve bank announcement is silent on the issue of foreign institutional investors. But, market analysts interpreted the RBI circular as clearly excluding foreign institutional investment from the FDI limit. If so, it would mean that foreign holding in Indian private banks could go as high as 98 percent, 49 percent in FDIs plus the FII holding which was earlier raised from 40 percent to 49 percent in the 2001 budget. While the banking stocks posted hefty gains in response to the news, the market still had to wait for the confirmation from the commerce ministry or the finance ministry that FDI and FII limits were now independent and separate. The Finance ministers 2002- 03 budget unveiled on February 28 clarified that FII portfolio investments will not be subject to the sectoral limits for foreign direct investment except in specific sectors, but kept the issue in abeyance by adding that guidelines will be issued separately. This prompted a series of press reports critical of the central governments ambiguous stand on the issue. Finally, on March 19, the Minister of State of Finance told the Rajya Sabha that the portfolio investment by foreign institutional investors in the private sector banks would be outside the foreign direct investment limit of 49 per cent. We present a chronology of the developments in the Indian banking sector over the period May 2001 to April 2002 in Table 1. Two aspects of the Indian experience are particularly noteworthy. First, and this is possibly a characteristic that prevails in most developing countries, capital market liberalization policies are viewed with a lot of skepticism and there is general apathy towards them. Second, foreign banks are reluctant to undertake equity participation without sufficient control to be able to effect changes in management style and work culture. This implies that foreign banks consider monitoring of management a necessary condition for success in developing countries.

3. Theoretical Perspective
We follow Stulz (1999) to develop the hypotheses on the potential impact of foreign direct investment on the domestic stocks. In this model, liberalization has two significant effects on the domestic countrys capital market. First, discount rates fall when the domestic market allows international investment in its securities. Second, shareholders have more confidence in

management because globalization allows ownership by foreign entities, which improves governance and monitoring of management. Stulz argues that foreign direct investment reduces a countrys cost of capital by decreasing systematic risk, such that stock prices react positively to unanticipated announcements of liberalizations. To elaborate, in a market which is completely segmented from the rest of the world, domestic investors bear all the risk. As the capital market of the country opens up to foreign investors, they bear some of the risks associated with the countrys economic activities. If the entire world constitutes one capital market, we would have a global equity market where the beta of a risky security would be computed relative to the global market portfolio. Conceivably, in a global economy with a large number of countries, the volatility of the world portfolio could be reduced to zero such that risk premium of each country would fall to zero. Stulz demonstrates that as long as the correlation between the small countrys market portfolio and the world portfolio is not too high, or the volatility of the small countrys market portfolio is not too low, the small countrys risk premium, and hence its cost of capital will fall when it removes the barriers to international investment. Under this scenario, an individual firm with an increase in it cost of capital would be an exception rather than the rule. Lower cost of capital induces greater firm valuation only if managers can convince shareholders of higher expected cash flows in order to raise the necessary investment capital. Asymmetric information between managers and shareholders, however, renders managers claims suspect. Also, managers propensity and opportunity to overinvest in unprofitable projects, and increase firm size only to entrench and benefit themselves, make shareholders skeptical of their motives. Liberalization can improve information dissemination and managerial focus. The factors that benefit from liberalization to facilitate better governance of management include an independent board of directors, relationship with skilled investment bankers, the formation of large ownership stakes, and a pool of investors, who compete to gain control of poorly managed firms. In addition, globalization affords better protection of minority shareholders interest, and instills financial discipline by requiring compliance with strict disclosure laws. Greater disclosure encourages trading in the firms securities, improves liquidity, and reduces bid-ask spread, and cost of capital. Lower bid-ask spread also helps monitoring since shareholders are better able to liquidate their holdings if managers deviate from value maximizing policies [Bhide (1993)]. Stulz concludes that globalization enables firms to finance valuable projects, reduces the benefit of control to managers, and decreases deadweight costs associated with agency problems and asymmetric information.

This implies two testable hypotheses: Hypothesis 1: Liberalization of the capital market leads to decreases in the cost of capital, resulting in immediate valuation gains for domestic stocks that benefit from it. Hypothesis 2: The differential valuation gain for individual domestic stocks following liberalization is attributable to and associated with the improvements in managerial monitoring, greater disclosure of information, and the creation of an environment conducive to control contests. If integration with the world market enhances domestic security values, an event study focused on the time of liberalization can capture the impact of globalization on cost of capital. Using this framework, Chari and Henry (2002) demonstrate that change in valuation of individual firms induced by liberalization is positively associated with the change in systematic risk. They model the change in systematic risk after liberalization, but focus mainly on liquidity issues. Lam focuses on the average daily premium of foreign-owned shares over locally-owned shares. Our paper provides the first direct evidence on potential valuation gains at the individual firm level at the announcement of a sweeping change in foreign direct investment limits. Anecdotal evidence suggests that investors focused on the attractiveness of individual banks as potential acquisition targets for foreign banks under the new FDI regulations in India. As such, our analysis provides a direct test of the second hypotheses motivated by Stulzs model.

4. Performance of Indian Banks Around RBI announcements

The Reserve Bank of India currently lists 27 nationalized, government controlled banks (including The State Bank of India, and its 7 affiliates), 32 private banks, and 41 foreign banks. We reclassify three private banks -- IDBI Bank, UTI Bank, and Jammu and Kashmir Bank -- as government banks. government. IDBI Bank and UTI Bank are majority owned by IDBI, a financial institution, and UTI, a mutual fund, respectively. IDBI and UTI are fully owned by the central Jammu and Kashmir Bank is more than 50 percent owned by the central government, as well.5 Of the resulting new list of 30 nationalized banks and 29 private banks, 13 nationalized banks, and 11 private banks have no traded stock, and three private banks trade on

The RBI definition of Private Banks is governed by the RBI January 1993 guidelines revised on January 3 2001. Other than the minimum capital requirements and other related provisions the main criterion for classification as a private bank is that the bank should be a public limited company with shares listed on a stock exchange with minimum promoters capital of 40 percent and a lock up clause of five years. Companies, directly or indirectly connected with large industrial houses, could invest up to 10 per cent of the banks capital without having controlling interest in the new bank.

regional exchanges. We exclude these banks from our analysis as stocks on regional stock exchanges trade infrequently and accurate pricing data is difficult to obtain. One nationalized bank (Punjab National Bank) started trading in May 2002, and another (Indian Bank) has been under trading suspension since January 2001. One private bank (Bank of Madura) merged during our study period. Eliminating a total of 15 banks from each initial sample reduces the final samples to 15 government banks and 14 private banks. The 6 private banks formed during the 1993-94 period are referred to as the new private banks. The other 8 private banks, known as the old private banks, were formed more than 50 years ago before the independence of India from the British Raj. For each bank and the value-weighted market Index (SENSEX), we collected the daily closing price denominated in Indian rupees for the 303 trading days spanning January 1, 2001 to March 18, 2002. The data are obtained from the Prowess database of the Center for Monitoring of Indian Economy (CMIE), and the Bombay Stock Exchange (BSE). Daily abnormal return is calculated as the difference in the change in closing prices in Indian rupees between the bank portfolio and the market index. 6 Daily abnormal returns are cumulated over the entire period to generate the Cumulative Abnormal Returns (CAR). 4.1 Cumulative Abnormal Returns (CARs) In figure 1, we plot the cumulative abnormal daily returns (CAR) of the equally-weighted portfolios of government and private banks. For the overall period, the portfolio of government and private banks post cumulative abnormal returns of 40 percent and 15 percent, respectively. Both portfolios show significant abnormal gains of about 10 percent between January and February 2001. But, the portfolios give back all the gain over the next two months, the private banks under-performing the market by about 10 percent. The government banks more than recover the losses over the next few months and post gains in excess of 30 percent by September. At year-end, the government bank portfolio bests the market by a solid 20 percent. Private banks, however, sustain big losses and under-perform the market by over 20 percent by October. For a better perspective, it may be noted that during this period, the SENSEX lost about 30 percent of its value. In essence, private banks lost half of their value from March to October 2001. A mild recovery starts at this point and continues up to the middle of November. But both groups suffer significant losses over the next month. By year-end, the cumulative performance of the private
6

None of the sample banks paid any dividends over this period.

banks portfolio is more than 20 percent worse than the overall market. Both banking groups show very strong and robust gains from the new year. Between year-end 2001 to middle of February 2002, government banks gain nearly 20 percent, and gains of private banks exceed 35 percent, as compared to the SENSEX. As Table 1 shows, the significant dates in the chronology are May 9, September 20, and November 29, 2001 and February 2 and 16, 2002. The most significant movement in the value of either portfolio occurs around the beginning of February 2002. It is clear that although the decision to relax foreign investment barriers was taken in May 2001, private bank investors saw little promise in the proposed changes. In contrast, nationalized banks posted superior performance throughout 2001. Compared to the banks in the more developed nations, as well as the Indian private sector counterparts, the nationalized banking sector in India is notoriously poor in operating efficiency, work ethic, and managerial expertise. A potential explanation for the gains of nationalized banks is that their valuation was greatly depressed reflecting these inefficiencies. As such, they represented more attractive acquisition targets, with greater potential gains from a turnaround than the private banks.7 For the private banks, the confusion and the Indian governments persistent refusal to separate the FDI and FII limits hurt the investors. Investors were convinced that until foreign banks were allowed to acquire majority stakes, they would show little interest in Indian banks. Both portfolios show large and sustained gains from beginning of 2002 to the middle of February. The steady gains from the beginning of 2002 suggests that the market anticipated the February 16 RBI decision followed by the Finance Ministrys confirmation to separate the FDI and FII ownership limits, effectively allowing nearly 100 percent foreign ownership in Indian banks. 4.2 Daily Abnormal Returns (DARs) To capture the abnormal returns around specific dates, we conduct an event study analysis separately for May 9, 2001, February 1 and February 16, 2002, designating each of these dates in turn as day 0. One hundred days of daily returns is used for the estimation of the market model, and a period of 20 days preceding and following the event day represents the event period.
7

Mr. Shyam Bhat, general manager for investments at Tata TD Waterhouse Mutual fund at Mumbai said, Banks were always a value proposition in India. Their low price-earnings ratios reflected a fear of possible bad loans hiding in opaque balance sheets. That worry is still there, but now investors are paying closer attention to what was formerly a neglected industry. (Wall Street Journal, March 18, 2002, pp C12)

10

The Sensex is used as the market Index. One concern with the data is the incidence of missing observations. A firm is excluded if it has less than 60 days of data during the estimation period and less than 30 days of data during the event period. For the May 9, 2001 date, the final sample includes 27 banks, 14 private and 13 government banks. All banks except two had more than 90 days of data during the estimation period, and only one bank had less than 35 days of data during the event period. 19 banks had complete data during both periods. For the Feb 1, 2002 date, the final sample comprises 28 banks, 14 banks from each group. Four banks had less than 90 days of data during the estimation period, and two banks had less than 35 days of data during the event period. 19 banks had complete data during both periods. For the final February 16, 2002 date, the final sample includes all 29 banks. All banks except six had more than 90 days of data during the estimation period, and all 29 banks had more than 35 days of returns data during the event period. 19 banks had complete data during both periods. We use the following procedure to adjust for missing returns: when an estimation period contains a sequence of one or more missing returns, the first-succeeding non-missing return is not used. The reason is that the first nonmissing return is a multi-period return. Permitting multi-period returns could have unexpected consequences for parameter estimates. When a sequence of one or more returns is missing in the event period, the abnormal return computation period is adjusted to account for the multi-day character of the post-missing return. Results of the analysis are presented in Table 2. Cumulative abnormal returns (CARs) for the portfolios of government banks, private banks, and the combined samples are reported for the three dates for event windows of (-20, -2), (-5, +5), (-1, 0), (0, +1), (-1, +1), and (+2, +20), surrounding the event days. For each event window, we report the number of banks in each portfolio, the number of banks with positive abnormal returns, mean cumulative abnormal returns (CAR), t-statistic, and the generalized sign Z-statistic. The CARs for May 9, 2001 are in Panel A. CARs are not significant over any of the intervals, although 12 out of the 14 private banks experienced positive abnormal returns on days 0 and +1. This number is significant at the 5 percent level. The evidence indicates that the announcement of increasing FDI limits to 49 percent was a non-event. Without the opportunity to assume majority control, foreign banks were not expected to get involved in the Indian banking sector. As our earlier analysis of the performance of government banks over the next few months suggests, however, investors perceived the decision as a positive step towards complete liberalization. As Panel B shows, market reaction to the merger proposal on February 1, 2002, was significantly positive for all samples. The private banks gained 6.27 percent and government

11

banks gained 4.27 percent over days (0, +1) surrounding the announcement; 25 of the 28 banks in the total sample posted gains. Over the three-day period (-1, +1) surrounding the announcement, private banks gained over 7 percent, and government banks gained 3.5 percent; value increased for 24 of the 29 banks. The cumulative returns over the preceding 20days were not significant. This indicates that the merger announcement was unanticipated. The valuation effect is consistent with the notion that the market expected the merger talks to intensify pressure on the RBI and the Ministry of Finance to relax FDI and FII regulations, triggering a wave of mergers and consolidations in the banking sector. The valuation gains surrounding the February 16, 2002, RBI announcement reported in Panel C are strongly positive for both groups of banks. Over the 20 days preceding the announcement, a period which includes the merger proposal, the cumulative returns are nearly 20 percent for private banks, and above 23 percent for the government banks. These are both significant at the 5 percent level. Twenty-seven banks gained value over this period. Market was calling for the RBI to respond to the mounting pressure from the banking sector and allow foreign banks to acquire majority stakes in Indian banks. The gain over the two days (0, +1) is over 21 percent for private banks and over 3 percent for government banks, both significant at 5 percent level. Over the three-day period surrounding the announcement, private banks gain nearly 25 percent compared to the overall market and government banks gain over 6 percent. Shareholders of all 14 private banks in the sample enjoyed wealth increases over this interval. Over the next 20 days, neither group posts any significant changes in value. Overall, two patterns emerge. First, the RBI announcement on May 9, 2001, was largely ignored by the market, although it constituted -- particularly for the government bank investors -a signal of more relaxation to follow. The shareholders of private banks were disappointed by the ambiguity of the RBI release, as is evident from the persistently poor performance of this sector over the next several months, when it lost nearly 50 percent of its value. Second, valuations improved significantly during the merger talks on February 1, 2002, because the market felt that RBI would be forced to liberalize FII and FDI rules to allow foreign banks assume majority control of Indian banks. The reaction on February 16 was the culmination of a year-long wait. Consistent with the first hypothesis and Stulzs (1999) prediction, we find that removal of barriers to international investment induces immediate and significant increase in stock prices. Over the period May 2001 to February 2002, the financial press in India abounds in anecdotal evidence that investors attributed the value gains to the probability that foreign banks would now

12

be able to acquire majority control of Indian banks, remove the inefficiencies, demand transparency in reporting of financial information, and monitor management. These improvements would result in lower cost of capital, greater cash flow, and higher stock values.

5. Bank Attributes
To explore the factors that contribute to the valuation gain of an individual bank, we identify a set of variables that reflect the financial strength, and operational and managerial efficiency of a bank. In Table 3, we present the summary statistics for several selected attributes of the banks in our sample. Collected from Reserve bank of India, data items are as of December 31, 2001, except for the Balance Sheet items which are calculated as the average of the values at the beginning and end of 2001. Data for private banks (panel A), government banks (panel B), and average values for each variable for each class of bank are reported in the table. For each bank, we report whether it is a private or a government bank, and among private banks, we identify the ones formed after 1994 in italics. We measure bank size by the total number of branches nationwide, market value of common equity, total assets, and total deposits, in millions of Indian rupees. Return on assets and growth in assets, and the ratio of market value to book value of common equity, represent investment opportunities. Capital adequacy is measured in terms of debt to equity ratio, and capital adequacy ratio. Earnings quality is the ratio of net profits to total assets, and the ratio of interest income to total assets reflects operational efficiency. Productivity is proxied by net profit per employee, and the ratio of net operating assets to total assets represents the quality of bank assets. Liquidity is measured by cash balance as a fraction of total assets. To report ownership structure, we analyze the stock ownership by foreigners, the general public, and the management. Banks are arranged in the ascending order of market capitalization. On average,

government banks have higher marker capitalization, total assets and total deposits than private banks, and among private banks, the new banks are significantly larger. Government banks also have significantly wider network with a greater number of branches. The State Bank of India, which is by far the largest bank in India, has the maximum number of branches. The average Cumulative abnormal return (CumAr) over days (-1, +1) for private banks is 25 percent, and that for government banks is 11 percent. This shows a general pattern of an inverse relationship between CARs and size. The only two banks to sustain valuation losses around the announcement of increase of FDI limits are State Bank of India, and Corporation Bank, two of the largest banks in the sample. The evidence is consistent with the notion that there is less

13

information asymmetry about larger and older banks, such that they need less monitoring and therefore would benefit less from a takeover following the liberalization. The greater price reaction for smaller banks is also attributable to the possibility that smaller banks are more vulnerable to acquisitions by foreign banks. There is no significant difference in return on assets, debt-equity ratio, and capital adequacy ratio among private and government banks. It is worth noting, however, that the older private banks consistently have higher debt equity ratios, as do some government banks. The ratio of market value to book value, and growth of assets are higher for private banks, but the result appears to be driven by a few new banks. Private banks are more profitable, as indicated by the higher ratio of net profit to total assets; and, in terms of productivity, the new privates significantly outperform government banks. Private banks also have significantly higher nonperforming assets, but this result is dominated by the older banks. The level of cash balance is comparable among the two groups of banks. Foreigners, general public and insiders all hold significantly larger ownership stakes in private banks, particularly the new breed of private banks. In sharp contrast, foreign institutions tend to almost entirely avoid the government banks, with the exception of State Bank of India. In government banks, ownership of company shares by managers and directors is negligible. Managers and directors own less shares, on average, in old private banks than the new ones, an indication that the new private banks are possibly more efficiently managed and are therefore less vulnerable to takeover. The ownership by the public at large is more than double in new private banks than old private banks. The banks with the largest foreign holdings are Centurion Bank and ICICI Bank. Most of the merger talks preceding the RBI February 16 directive centered around these banks. In general, foreign banks that held ownership stakes in Indian banks appeared ready to assume controlling interest in these banks if RBI rules permitted. The pearson correlation coefficients between the variables are reported in table 4. Market capitalization, total assets, and total deposits are used in natural logarithms. The data show that among the significant relationships, larger banks earn higher return on assets, are operationally less efficient as indicated by lower interest income, have less non-performing assets, and maintain greater cash balances. As expected, profitability is higher in banks where employees are more productive, but both profitability and asset growth suffer as non-performing assets accumulate. Productivity (profit per employee) is positively correlated with the growth of assets, and negatively with non-performing assets. Interest income drops significantly as cash balance

14

increases as a ratio of total assets. As capital adequacy increases, bank performance, as measured by return on assets, improves significantly. 6. Determinants of Takeover Probability According to Stulz (1999), the differential valuation gain for individual domestic stocks following liberalization is attributable to the improvements in managerial monitoring, greater disclosure of information, and the creation of an environment conducive to control contests. Smaller banks, and banks run by management that is incapable of identifying investment opportunities under competitive pressure, are especially vulnerable to takeovers. Lack of investment opportunities manifest in low return on assets, low asset growth, and low ratio of market value to book value of equity. High debt level is a deterrent to takeovers, and adequate capital reserves enable target banks to thwart hostile attempts. Profitable banks, and banks with superior productivity are more expensive, and less attractive candidates for takeover. On the other hand, a bank laden with a high level of non-performing assets would be in jeopardy, and an easy target. Conversely, banks with high cash reserves are better equipped to withstand takeover threats, and negotiate a better price. Finally, ownership structure is an important determinant of takeover success. High insider ownership motivates management to greater effort and enhances bank value, and also gives management control and power to negotiate during a takeover attempt. Conversely, high ownership by foreign entities makes a bank more vulnerable after liberalization. Ordinary least square regression estimates with 3-day cumulative abnormal returns (CARs) over the interval (-1, +1) as the dependent variable are presented in table 5. Eleven different models are estimated. The independent variables are introduced sequentially to measure the impact of an individual banks size, management quality, capital adequacy, earnings quality, operational efficiency, productivity, asset quality, liquidity, and ownership structure on the probability of takeover and liberalization induced abnormal returns. The choice of variables in various models is dictated by the hypothesized relationships and potential multicollinearity problems. Model 1 shows that private banks posted significantly higher gains at the announcement. We include this dummy variable as a control in all subsequent models, and it retains its significantly positive sign regardless of the specification. However, a dummy variable identifying the new private banks has no explanatory power in any specification. Consistent with Stulzs hypothesis, the significantly negative sign for size measured as natural logarithm of

15

market value in model 2 demonstrates that smaller banks are more vulnerable.

The sign is

negative for total assets and total deposits as well, although we do not report these results in the interest of space. Return on assets has a significantly negative coefficient in model 3, which suggests that inefficient banks are perceived to be easy targets. assume heavy debt burdens. Model 4 and 5 show a significantly negative sign for debt-equity ratio, indicating that foreign banks are not expected to This variable remains significantly negative in all subsequent models. Capital adequacy ratio is insignificant. In model 6, growth of assets is insignificant, but the market to book ratio is significantly negative. This result is a clear indication that banks with superior investment opportunities are expected to be keep hostile suitors at bay. Earnings quality is the focus in model 7, and the significantly negative sign on net profit supports our contention that superior performance is a takeover deterrent. The significantly positive sign on interest income is intriguing. As hypothesized, model 8 demonstrates that banks with highly productive employees discourage takeovers. Conversely, as evidenced by the significantly positive sign of non-performing assets in model 9, accumulation of excess nonperforming assets weakens a bank against acquirers. Cash reserves have the opposite effect, as model 10 shows. Interestingly, insider ownership or any of the ownership variables have no discernible impact on takeover probability. In summary, the analysis reveals that private banks are more vulnerable to takeover. In general, banks that are smaller, less efficient, less productive, and have high accumulation of nonperforming assets are particularly vulnerable to hostile takeovers. We conclude that the regression models are generally consistent with the premise that abnormal stock price movements at the announcement of liberalization of foreign direct investment limits reflect the premium that acquirers must pay to takeover weak domestic companies that were protected by restrictive ownership limits. 7. Conclusion On February 16, 2002, the central government in India relaxed foreign ownership limits in the banking sector. Although the change made foreign control possible only in the private sector banks, a portfolio of Indian banks posted hefty gains at the announcement. Our objective in this paper is twofold. First, in contrast to the extant evidence which focuses on the aggregate stock price effect of FDI limits, we provide the first evidence of valuation changes at the

16

individual firm level. Second, we test the hypothesis that the valuation gain of an individual firm reflects a takeover premium, and is a function of the probability of takeover of the firm. The results demonstrate that valuation gains by private sector banks are significantly higher than government owned banks. Further, valuation gain is a function of an individual banks market value, investment opportunity and efficiency, labor productivity, earnings quality, and asset quality. Inefficient, and poorly managed banks with lower relative market valuation, and excess non-performing assets are likely to benefit most from a potential takeover, and post the largest gains. We conclude that our evidence is consistent with the notion that investors welcome the removal of protective barriers and the ultimate takeover of inefficient firms following the liberalization. As such, our study has important policy implications for third world countries where foreign ownership of domestic companies is still restricted to a level where takeover and control is too costly, and often, impossible.

17

Figure 1: Cumulative abnormal daily returns (CAR) of equally-weighted Portfolios of 15 Private banks and 14 Government banks in India between the period January 1, 2001, to March 18, 2002. Daily abnormal returns are calculated as the difference between the returns of the equally-weighted bank portfolio and the market index (SENSEX). Daily closing stock price in Indian rupees are collected from the Center for Monitoring of Indian Economy (CMIE) database Prowess and the Bombay Stock Exchange (BSE).

Cumulative Abnormal Returns (CAR): Jan 2001 - Mar 2002

50% Cum. Abn. Returns - Govt. banks 40% Cum. Abn. Retuns - Pvt. banks

30%

20%

10%

h
2/15/01 4/1/01 5/16/01 6/30/01 8/14/01 9/28/01 11/12/01 12/27/01 2/10/02

0% 1/1/01 -10%

-20%

-30%

18

Table 1 Chronology of events (May 9, 2001 Mar 19, 2002). Information is collected from various Indian press reports. DATE
May 9, 2001 May 10, 2001 May 12, 2001 May 21, 2001 Sep 20, 2001 Nov 29, 2001 Feb 1, 2002 Feb 8, 2002 Feb 13, 2002 Feb 16, 2002 Feb 27, 2002 Mar 19, 2002

EVENT DESCRIPTION
Central ministry decides increase FDI limits in banks from the current 20% to 49% along with increase in NRI investment from 40 to 49%. Earlier the FDI of 20% is sub cap of the overall cap of 40% from NRIs i.e if there is no FDI, NRIs can invest up to 40% and this can be done through automatic route. (The Economic Times, May 10, 2001) Clarification is issued by special secretary (banking) that the 49% includes NRI, FII and FDI investments. Any one of the above categories or some of them or all of them together can now take up to 49%. (Business Line, May 11, 2001) Limited reaction from the market with marginal increase in the stock prices of selected banks. (The Economic Times, Business Line, Financial Express, May 10 and 11, 2001 ) Formal notification of the increase in FDI issued vide Press Note no.4 (2001 series) by Government of India, Ministry of Commerce & Industry. RBI said in a statement that FII investment in a company will be governed by investment ceiling for FDI for specific sectors. For private sector banking, the investment limit would be 49 per cent. Citigroup said it will look at acquisitions in India if the regulatory framework permits. Since the bank has branch operations here, it can use its private equity arm to take an equity interest. CitiGroups interest involves acquiring the promoters stake in one of the new generation private sector banks. (The Economic times Nov 29, 2001). Bank Brussels Lambert (BBL) of the Dutch financial group ING announces its interest in hiking its stake in Vyasa bank from the current 20% which it holds. This is the first announcement of any Foreign Bank increasing its stake after the May 9 announcement permitting the same by Government of India. (The Economic Times Feb 1, 2002) RBI Governor, Dr Bimal Jalan, says that the central bank has sought a clarification from the Center on the total foreign direct investment (FDI) and foreign institutional investment (FII) limits in the domestic banking industry. (The Economic Times Feb 8, 2002) US deputy treasury secretary Mr. Kenneth Dunn joins foreign financial services firms in urging India to further open up investments in local banks during his visit to Bombay to discuss money laundering. (The Economic Times Feb 13, 2002) The Reserve Bank of India announces increase in FDI upto 49% in private banks, the limit for FDI in government banks is pegged at 20% vide Notification, DBOD.No.BP.BC. 68 /21.01.055/2001-02, dated February 16, 2002. The Reserve Bank of India (RBI) makes it clear to bankers that the 49 per cent foreign direct investment (FDI) ceiling includes only the following category of shares initial public offerings (IPOs), private placements, fresh issuances of American Depository Receipts (ADRs)/Global Depository Receipts (GDRs), and acquisition of shares from existing shareholders. This excludes portfolio investments by Foreign Institutional Investors (FIIs). But confirmation from central government must be awaited. (The Economic Times Feb 27, 2002) Minister of State for Finance, Mr. Balasaheb Vikhe Patil tells Rajya Sabha that the portfolio investment by foreign institutional investors (FIIs) in the private sector banks would be outside the foreign direct investment (FDI) limit of 49 per cent (Business Line, Mar 19, 2002)

19

Table 2
Cumulative abnormal returns (CAR) in percent for equally weighted portfolios of 15 nationalized banks, 14 private banks, and the combined sample of 29 banks surrounding May 9, 2001, February 1, and February 16, 2002. The Eventus (2002) software is used to generate the market model using the value-weighted SENSEX as the market Index. Closing stock price data are obtained from the Center for Monitoring of Indian Economy (CMIE) database Prowess and the Bombay Stock Exchange (BSE) website. Daily returns are calculated as the change in closing prices in Indian rupees. The Eventus software uses the Patell (1976) procedure to calculate the t-statistics. CARs are reported for intervals (-20, -2), (-5, +5), (-1, 0), (0, +1), (-1, +1), and (+2, +20) surrounding the event date. T-statistic and non-parametric generalized sign Z test statistic are in parentheses.
Private banks Event windows (-20, -2) (-5, +5) (-1, 0) (0, +1) (-1, +1) (+2, +20) Sample Size (# positive) Abn. Return (%), (t-stat, Gen. Sign Z) Nationalized banks Sample Size (# positive) Abn. Return (%), (t-stat, Gen. Sign Z) Combined sample Sample Size (# positive) Abn. Return (%), (t-stat, Gen. Sign Z)

Panel A: May 9, 2001 (RBI announces FDI Limits increased to 49 percent, but no clarification about the independence of FDI and FII limits)
14 (8) 14 (11) 14 (5) 14 (12) 14 (9) 14 (9) -1.52 (0.23, 0.84) 4.85 (0.96, 2.45*) -0.20 (-0.09, -0.76) 1.31 (0.61, 2.99*) 1.40 (0.53, 1.38) -0.12 (-0.02, 1.38) 13 (11) 13 (8) 13 (2) 13 (8) 13 (6) 13 (10) 6.85 (0.88, 2.72*) 1.48 (0.25, 1.05) -2.23 (-0.88, -2.28) 0.75 (0.30, 1.05) 0.25 (0.08, -0.06) 6.55 (0.84, 2.16) 27 (19) 27 (19) 27 (7) 27 (20) 27 (15) 27 (19) 2.52 (0.39, 2.49*) 3.23 (0.66, 2.49*) -1.18 (-0.56, -2.13*) 1.04 (0.50, 2.88*) 0.85 (0.33, 0.95) 3.09 (0.48, 2.49*)

Panel B: Feb 1, 2002: (Bank Brussels Lambert (BBL) of the Dutch group ING announces its interest in hiking its stake in Vyasa Bank from the current 20%. First announcement of any Foreign Bank increasing its stake in an Indian bank. )
(-20, -2) (-5, +5) (-1, 0) (0, +1) (-1, +1) (+2, +20) 14 (9) 14 (11) 14 (10) 14 (12) 14 (12) 14 (14) 0.60 (0.11, 1.21) 11.45 (2.72*, 2.28*) 1.18 (0.66, 1.74) 6.27 (3.49*, 2.81*) 7.19 (3.27*, 2.81*) 20.34 (3.68*, 3.88*) 14 (10) 14 (13) 14 (6) 14 (13) 14 (12) 14 (11) 5.59 (1.36, 1.81) 13.68 (4.38*, 3.42*) -0.35 (-0.26, -0.33) 4.27 (3.20*, 3.42*) 3.54 (2.17*, 2.88*) 10.23 (2.49*, 2.35*) 28 (19) 28 (24) 28 (16) 28 (25) 28 (24) 28 (25) 3.09 (0.72, 2.14*) 12.57 (3.84*, 4.03*) 0.42 (0.30, 1.00) 5.27 (3.78*, 4.41*) 5.36 (3.14*, 4.03*) 15.28 (3.55*, 4.41*)

Panel C: Feb 16, 2002 (The Reserve Bank of India announces increase in FDI upto 49% in private banks, the limit for FDI in government banks is pegged at 20%.)
(-20, -2) (-5, +5) (-1, 0) (0, +1) (-1, +1) (+2, +20) 14 (13) 14 (14) 14 (14) 14 (14) 14 (14) 14 (7) 19.64 (3.57*, 3.37*) 26.08 (6.23*, 3.91*) 9.59 (5.38*, 3.91*) 21.17 (11.87*, 3.91*) 24.75 (11.33*, 3.91*) 1.19 (0.22, 0.16) 15 (14) 15 (14) 15 (12) 15 (12) 15 (13) 15 (2) 19.90 (4.60*, 4.16*) 13.12 (3.87*, 2.86*) 9.65 (6.88*, 2.61*) 8.11 (5.78*, 2.61*) 11.06 (6.44*, 3.13*) -5.87 (-1.36, -2.57*) 29 (28) 29 (28) 29 (26) 29 (26) 29 (27) 29 (9) 19.78 (4.57*, 5.33*) 19.38 (5.88*, 5.33*) 9.62 (6.85*, 4.59*) 14.42 (10.26*, 4.59*) 17.67 (10.27*, 4.96*) -2.46 (-0.57, -1.73)

* significant at the 5 percent or lower levels (two-tailed test).

20

Table 4 Pearson Correlation Coefficients between selected attributes of 14 private banks and 15 government banks in India. Variables include natural logarithm of market capitalization (LnSIZE) in millions of Indian Rupees, natural logarithm of total assets (LnTA) in millions of Indian Rupees, natural logarithm of total deposits (LnDEP) in millions of Indian Rupees, ratio of net profit over total assets in percent (ROA), . ratio of total debt over total equity in percent (DEBTEQ), Capital Adequacy Ratio defined as the ratio of loan loss reserves to total assets (CAR), growth in assets in percent between 2000 and 2001, calculated on the value of assets as of year end (ASSETGr), ratio of total market value to total book value of equity (MVBV), ratio of total operating profit to total assets in percent (PRFTTA), ratio of net interest income to total assets in percent (INTTA), total profit divided by the total number of employees in percent (PRFTEMPL), ratio of non-performing assets to total assets in percent (NPATA), and stock ownership in percent by foreigners (FOR), public (PUB) , and insiders (INSIDE) as percent of total shares outstanding. Data for 14 private banks and 15 government banks are collected from Center for Monitoring of Indian Economy (CMIE) database Prowess. All data are as of December 31, 2002, except for the Balance Sheet items which are calculated as the average of the values at the beginning and end of 2001. Variables
LnSIZE LnTA LnDEP ROA DEBTEQ CAR ASSETGr MVBV PRFTTA INTTA PRFTEMPL NPATA CASHBAL FOR PUB INSIDE LnSIZE 1.00 0.82* 0.81* 0.37* 0.19 0.27 0.37* 0.62* 0.33 -0.44* 0.30 -0.57* 0.37* 0.27 -0.65* -0.22 LnTA 1.00 1.00 0.05 0.35 0.30 0.08 0.26 0.14 -0.42* -0.08 -0.37* 0.32 0.07 -0.68* -0.49* LnDEP ROA DEBTEQ CAR ASSETGr MVBV PRFTTA INTTA PRFTEMPL NPATA CASHBAL FOR PUB INSIDE

1.00 0.04 0.35 0.27 0.08 0.24 0.12 -0.41* -0.11 -0.35 0.30 0.04 -0.66* -0.51

1.00 -0.08 0.41* 0.34 0.44* 0.77* 0.09 0.36 -0.48* 0.02 0.01 0.01 -0.08

1.00 -0.01 -0.04 -0.14 -0.02 -0.17 -0.34 0.11 0.30 0.04 0.01 -0.22

1.00 0.06 0.04 0.76* -0.08 0.27 -0.71* 0.31 0.36 -0.22 -0.22

1.00 0.32 0.33 -0.29 0.68* -0.43* 0.14 0.32 -0.12 0.12

1.00 0.31 -0.26 0.44* -0.37* 0.21 0.25 -0.14 0.12

1.00 0.06 0.38* -0.73* 0.16 0.18 0.00 -0.13

1.00 -0.33 0.24 -0.69* -0.50* 0.39* -0.25

1.00 -0.54* 0.17 0.73* -0.20 0.47*

1.00 -0.25 -0.36 0.46* 0.09

1.00 0.47* -0.29 0.16

1.00 -0.21 0.56*

1.00 0.21

1.00

* significant at the 5 percent level.

21

Table 5
Ordinary Linear Regression (OLS) estimates of Cumulative Abnormal Returns (CumAR) over the interval (0, +1) against type of Bank (private or government controlled, PVT), natural logarithm of market capitalization (LnSIZE), debt-equity ration (DEBTEQ), capital adequacy ratio (CAR), return on assets (ROA), growth of assets (ASSETGr), ratio of market value to book value (MVBV), ratio of net profits to net assets (PRFTTA), interest income to total assets (INTTA), profit per employee (PRFTEmplratio of net performing assets to total assets (NPATA), ratio of cash balance to total assets (CASHBAL), and fraction of shares owned by insiders of the firm (INSIDE). CumAR represents the cumulative daily abnormal returns in percent over the interval (0, +1) from the market model estimated with 200 days of daily returns ending 30 days before the event date. The Bombay Stock Exchange Index (SENSEX) is used as the market proxy. Market capitalization is calculated in Indian Rupees (000) as the product of the number of shares outstanding and the share price. Capital Adequacy Ratio defined as the ratio of loan loss reserves to total assets. Data are collected from the Reserve bank of India, Center for Monitoring of Indian Economy (CMIE) database Prowess. All data are as of December 31, 2002, except for the Balance Sheet items which are calculated as the average of the values at the beginning and end of 2001. Tstatistics are in parentheses.
Variables PVT LnSIZE ROA DEBTEQ CAR ASSETGr MVBV PRFTTA INTTA PRFTEmpl NPATA CASHBAL INSIDE Intercept R-sqd adj. 11.05 (4.05)** 0.28 39.23 (4.91)** 0.51 19.88 (5.38)** 0.46 41.90 (4.52)** 0.53 26.82 (6.81)** 0.58 20.77 (5.71) 0.50 -19.51 (1.05) 0.61 20.12 (5.59)** 0.47 11.27 (3.37)** 0.51 20.84 (3.09)** 0.36 Model 1 13.69 (3.48)** Model 2 8.54 (2.41)** -4.59 (3.67)** Model 3 13.36 (3.91)** -13.04 (3.11)** -0.03 (1.79)* -0.19 (0.35) Model 4 8.63 (2.49)** -4.07 (3.19)** Model 5 13.31 (4.40)** -12.40 (3.15)** -0.04 (2.91)** -0.13 (1.14) Model 6 14.12 (4.26)** Model 7 14.63 (5.01)** Model 8 16.39 (4.53)** Model 9 10.78 (3.22)** Model 10 13.43 (3.60)** Model 11 19.56 (3.88)** -14.65 (4.04)** -0.04 (3.33)**

-0.04 (2.67)** -0.17 (1.41) -0.12 (2.02)**

-0.03 (2.12)**

-0.05 (2.99)**

-0.04 (2.76)**

-0.03 (1.76)*

-7.97 (3.73)** 4.74 (2.58)** -1.59 (2.45)** 2.33 (2.99)** -42.34 (0.87) -0.21 (1.62) 27.66 (7.07)** 0.60

**significant at 5 percent level.

22

References Bekaert, Geert and Campbell Harvey (2000). Foreign Speculators and Emerging equity Markets, Journal of Finance, Vol. 55, No. 2, pp. 565 613. Boyd, John and Bruce Smith (1996). The Coevolution of the Real and Financial Sectors in the Growth Process, The World Economic Review 10 (May), pp. 371 396. Chari, Anusha and Peter Blair Henry (2002), Risk Sharing and Asset Prices: Evidence From a Natural Experiment Research paper series, Graduate School of Business, Stanford University, Stanford, CA. Delhaise, Philippe F. (1998). Asia in Crisis : The Implosion of the Banking and Finance Systems, John Wiley & Sons (Asia) Pvt. Ltd., Singapore. Dornbush, Rudiger, and Y. C. Park (1995). Financial Integration in a Second-best World: Are we Still Sure About our Classic Prejudices, in Rudiger Dornbush and Y. C. Park, eds., Financial opening: Policy Lessons for Korea, Korea Institute of Finance, Seoul, Korea. Errunza, Vihang, and Darius P. Miller (1998). Market Segmentation and the Cost of Capital in International Equity markets, Working Paper, Texas A & M University, College Station, Texas. Foerster S., and Andrew Karolyi (1999). The Effects of Market Segmentation and Investor Recognition on Asset Prices: Evidence from Foreign Stocks listing in the U.S., Journal of Finance, Vol. 54, pp. 981-1014. Hamann, Javier (1999). Background to the Crisis in IMF-supported Programs in Indonesia, Korea, and Thailand: A preliminary Assessment, IMF, Occasional paper 178, Washington D.C. Henry, Peter Blair (2000a). Stock market Liberalization, Economic Reform, and Emerging market equity Prices, Journal of Finance, Vol. 55, No. 2, pp. 529 564. Henry, Peter Blair (2000b). Do Stock Market Liberalizations Cause Investment Booms? Journal of Financial Economics, Vol. 58 Nos. 1-2, pp. 301-334. Jensen, Michael, and William Meckling (1976). Theory of the Firm, Managerial Behavior, Agency Costs and Ownership Structure, Journal of Financial Economics, October 1976, pp. 305-360. Kim, Han E. and Vijay Singal (2000). Stock Market Openings: Experience of Emerging Economies, Journal of Business, Vol. 73, No. 1, pp. 25-66. Levine, Ross and Sara Zervos (1996). Stock Market Development and Long-Run Growth, The World Economic Review 10 (May), pp. 323 339. Miller, Darius (1999), The Market Reaction to International Cross Listing: Evidence from Depository Receipts, Journal of Financial Economics, Vol. 51, pp. 103-123.

23

Radelet, Steven, and Jeffrey Sachs (1998). The Onset of the Esat Asian Financial Crisis, Working Paper, Harvard University, Cambridge, MA. Rajan, Raghuram G, and Luigi Zingales (1998). Financial Dependence and Growth, American Economic Review 88 (June), pp. 559 586. Stiglitz, Joseph (1998). Boats, Planes, and Capital Flows, Financial Times, March 25. Stulz, Rene M. (1999a). Globalization, Corporate Finance, and the Cost of Capital, Journal of Applied Corporate Finance, Vol. 12, pp 8 25. Stulz, Rene M. (1999b). Globalization of Equity Markets and the Cost of Capital, Working Paper, Dice Center, Ohio State University. Stulz, Rene M. and Walter Wasserfallen (1995), Foreign Equity Investment Restrictions, Capital Flight, and Shareholder Wealth Maximization: Theory and Evidence, The Review of Financial Studies, Vol. 8, No. 4, pp. 1019-1057.

24