Institute for International Business Working Paper Series

IIB Paper No. 9 May 2007

Financial reforms in China and India: A comparative analysis

Wendy Dobson Director, Institute for International Business Rotman School of Management University of Toronto

Financial reforms in China and India: A comparative analysis*

Wendy Dobson Rotman School of Management, University of Toronto

Abstract: This paper surveys financial reforms in the worlds two most populous and rapidlygrowing economies. The contribution of financial systems to long term growth through the efficient mobilization and allocation of scarce capital is well documented in the literature. Indias financial system is popularly perceived to be better developed than Chinas, yet they share two significant weaknesses: under-developed corporate bond markets and bank-dominated financial systems. High levels of state ownership of banks are associated with misdirected lending and high costs of intermediation. The paper examines the institutional frameworks that determine incentives in these sectors and marshals empirical evidence that historical decisions and insufficient market reform suggest performance problems persist. These problems will become more evident when growth slows; indeed a crisis may be necessary to force change since prevailing high economic growth rates in spite of the weaknesses undermine the case for deeper reforms.

* This paper also draws on Dobson and Kashyap, The contradiction in Chinas gradualist banking reforms, forthcoming in Brookings Papers on Economic Activity 2006:2. Both papers are available at http://www.rotman.utoronto.ca/Dobson.

JEL classification numbers: P 51; O16; G20. Key words: comparative analysis; financial systems; China and India. Acknowledgements: This paper has benefited from discussion and comments at a seminar in Pudong organized by IFPRI, CES and SUFE.

1. Introduction Chinas financial system is often called the Achilles heel of its long-term growth prospects while Indias has generated enthusiasm as a source of strength. Although China has more money circulating in its financial system (in 2004 Chinas financial assets were 200 percent of GDP while Indias were 160 percent1) the freer play of market forces and diversity of financial institutions make Indias system more efficient in allocating capital. Government ownership of banks is common to both economies: at between 90 and 100 percent, these levels are the worlds highest (OECD 2005:140). Banks also dominate the domestic financial systems: in 2004 they accounted for 72 percent of total financial assets in China and 43 percent in India.2 The purpose of this paper is to compare financial reforms in the two countries and evaluate their potential impacts on growth and development. The finance-growth literature emphasizes that domestic financial systems influence a countrys economic development and growth through the roles they play in both the accumulation and allocation of capital.3 International evidence shows that over time public sector banks perform poorly by the usual market metrics on efficiency and soundness: returns on equity are low, expenses are high and they are dogged by non-performing loans. The analysis begins with a comparative evolution of the two financial systems in the next section with particular attention to incentive structures including prudential oversight, legal systems, ownership and corporate governance, and market monitoring. The third section examines banking and capital market performance and the fourth discusses prospects for achieving further reforms to promote soundness and efficiency. As both economies become more complex and integrated into the world economy their financial systems will be tested on their efficiency and their ability to withstand economic shocks. The final section concludes that further reforms are required. Efficiency as measured by the allocation of capital is improving but weaknesses in capital markets and government involvement in both directed lending and asset allocation by banks continue to create distortions in both countries. 2. Financial systems in China and India When they took power in the late 1940s governments in both countries sought to mobilize domestic savings and channel them into industrial development. Government ownership and intervention was the means to that end. Chinas planners abolished private property and institutions after the 1949 revolution when the state took over the financial system. Since the late 1970s, recreating financial markets and market-based institutions has been a work in progress. Indias leaders chose a mixed economy after independence from Britain in 1947. Markets continued to function alongside public institutions, but with ever-increasing bureaucratic oversight and regulation. Banks were mostly nationalized by the late 1960s but capital markets were permitted. Today, Indias financial system is the most-developed in the emerging market economies.

2.1 Banks Both countries financial systems are bank-dominated. Chinas assiduous savers have no alternative to the formal banking system (OECD 2005). Many Indians, on the other hand, mistrust banks and prefer to accumulate gold and real assets instead. Both governments require banks to serve social objectives but the Indian government is more transparent about the social outcomes being worth the price. Banks, whether public or private, must meet targets for rural access to banking services and lending to priority sectors and must allocate a required share of their investments to public sector bonds. The Chinese governments political priorities are to ensure gradual controlled liberalization of publicly-owned producers and economic growth sufficient to absorb millions of labor force entrants, migrants and laid off workers each year. Bank lending is still enlisted to finance much of this growth even as banks are modernized to meet new foreign competition. Historically, governments have owned banks as vehicles to finance industrial development.4 The underlying rationale was that market failures exist in developing economies because private banks respond only to private returns and fail to finance socially desirable projects or borrowers (usually small borrowers and sometimes very large projects). State-owned banks dominate the banking sectors in a majority of developing countries despite international evidence that countries with a large state bank presence have slower financial and economic development than countries that do not (Hanson 2004). For example, cross-country empirical studies by La Porta et al (2002) have found that state ownership of banks in a country in 1970 was associated with less financial development and lower growth and productivity through time. These effects were more pronounced in lower-income countries. The positive growth effects of increased private ownership of banks were also found to be statistically significant and economically meaningful. Indias banking system, which dates to the eighteenth century, is a mixture of public, private and foreign ownership. The rationale for nationalization in the late 1960s was ostensibly to increase banks support for economic growth using government ownership to reduce their vulnerability to connected lending, to force rural branching to encourage small savers, and to contribute to planned growth and equitable distribution of credit, particularly to small scale industry and farmers. Priority lending required by government meant 40 percent of net bank credit had to be directed to small scale industry, with 18 percent of this total directed to agriculture and 10 percent to weaker sections.5 Targets for banks to serve un-banked locations led to a rapid expansion of the banking sector but with most deposits concentrated in the public sector banks (PSBs). By 1990-91 PSBs accounted for 90 percent of total deposits and advances. PSBs retained their corporate structures but substituted government for independent directors. Specialized state-owned intermediaries that included the National Bank for Rural Development (NABARD), the Small Industries Development Bank of India (SIDBI) and various state finance

corporations also promoted the social objectives of finance (Patel 2004). Indias 1991 balance of payments crisis was a catalyst for liberalizing bank ownership and regulation. RBI reforms aimed to make the banking system more resilient to market shocks and to introduce arms length regulation to replace its hands-on approach (something that still over-shadows the banking system). PSBs were also permitted to access capital markets for up to 49 percent of their equity. In 1994 six new private banks were launched by government-owned financial institutions and three foreign banks entered the market. Two of the private banks, HDFC and ICICI, have since grown quickly and are noted for sophisticated management and technology and strong customer focus.6 Limited foreign entry was also permitted.7 In March 2006, India had 218 scheduled commercial banks, also known as SCBs (Table 1) that included more than one hundred regional rural banks. In China, after the 1949 revolution Chinas central bank, the Peoples Bank of China (PBOC), became a mono-bank responsible both for conducting monetary policy and for collecting savings at branches throughout the country. In 1984, PBOC deposit and lending functions were turned over to four state-owned policy banks whose loans supplied the countrys thousands of state-owned enterprises (SOEs) with their working capital requirements. Initially SOE losses were financed with public bonds. As the losses mounted, the central government moved in the mid-1990s to reduce the resource drain by forcing the SOEs to finance their needs with bank loans. Bank reforms began in earnest in 1995 when institutions and regulations were changed to transform them into commercial banks. Prudential norms for lending were introduced, banking, securities and insurance regulators created and regulatory standards tightened. Three policy banks were created to carry on policy lending functions and PBOC created regional heads supposedly with sufficient seniority to force bank lending on creditworthiness criteria. Chinas domestic banking system now consists of a large number of institutions almost all of which are owned by various levels of government (Table 1). The Big Four state-owned commercial banks dominate the system, accounting for more than half of banking assets, thousands of branches and hundreds of thousands of employees located throughout the country. Smaller but more numerous banking institutions are also government owned but are geographically concentrated. 2.2 Capital markets Direct finance through equity and debt markets provides a wider range of debt maturities and lower-cost capital to businesses and more choice for savers than is available from banks. Capital markets encourage efficient capital allocation through institutions that continuously monitor the performance of issuers through movements in share prices and threats of takeovers. Non-bank debt markets, however, are underdeveloped in both countries.

Indias capital markets are better developed than Chinas and play a more significant role in the economy. India has two national stock exchanges (Bombay Stock Exchange (BSE) and the National Stock Exchange) with the former established in 1875. Both have electronic trading platforms and as much as 70 percent of BSE market capitalization is accounted for by private firms and joint ventures (Farrell and Lund 2005). India also has futures markets and a venture capital market. Bond market infrastructure includes the Clearing Corporation of India (CCIL), established in 1999 to handle clearing and settlement. The government bond market is large because of fiscal funding requirements but because banks are required to hold government bonds the yield curve fails to play its full role as the market benchmark. The corporate bond market is under-developed, accounting for only one percent of total financial assets (the comparable figure for China is 5 percent) because of stringent regulations on registration and disclosure and high transactions costs for issuers (Farrell and Key 2005). Chinas capital markets are among the smallest in the world. There are two stock exchanges, established in 1990 in Shenzhen and Shanghai.8 Most of the available listings are those of the public shares of Chinas largest SOEs. High levels of government ownership in these companies and segmentation of the market have limited market liquidity. Two kinds of shares are available, one restricted to domestic investors and the other available only to foreign investors. Recent innovations to resolve these problems include the Qualified Foreign Institutional Investor (QFII) program, administered by the China Securities Regulatory Commission (CSRC) and the State Administration of Foreign Exchange (SAFE), which allows foreign investors to trade in the domestic market subject to restrictions that include a ten percent equity stake in any domestic company and prohibitions on acquisitions of non-tradable state-owned shares. Chinas bond market is in even earlier stages of development, serving mainly as a channel for government finance due to a highly restrictive regime facing corporate issuers. Although both the central bank and the securities regulatory commission are the regulators, the National Development Research Council, secretariat to the policy making State Council, must approve quota allocations and the issuance of corporate bonds. Banks continue to supply most debt finance. By September 2006, corporate bonds accounted for 3 percent of outstanding bonds compared to the 68 percent combined shares of government and central bank bonds.9 Corporate issuers, especially those listed on the Hong Kong Stock Exchange, are tapping into overseas debt markets instead. 2.3 Financial oversight Supervisory structures in China are still quite new and therefore evolving, while Indias are well-established, even entrenched. Both have flaws including uncertain and overlapping jurisdictions that affect banks in particular. In India, the Reserve Bank of India (RBI) carries out a number of roles that reflect its history. Not only is it the central bank, it is also the bank regulator, manager of the public debt and majority owner of the State Bank of India, Indias largest commercial bank. Mor et al (2006) argue that Indias banking oversight focuses on bank processes

rather than on credit quality and risk and that the reforms of the past decade have failed to rectify this problem. A central bank with multiple roles is often found in developing countries, but as an economy becomes more complex and makes more demands on the financial system, conflicts of interest are bound to arise, for example between monetary policy or debt management objectives and bank soundness and efficiency. We return to this issue below. In addition, Indias numerous state, urban and district cooperative banks and nonbank finance companies are subject to multiple overseers including the State Registrar of Cooperative Societies, Ministry and Finance and RBI, which weakens oversight and creates opportunities for regulatory arbitrage. Other government agencies including the comptroller and auditor general, Central Vigilance Commission (CVC) and Central Bureau of Investigation are also involved in bank oversight, particularly audits. Students of Indias banking system highlight this pervasive government involvement beyond what might be justified to prevent systemic risk as creating disincentives for efficient capital allocation.10 Capital market oversight is more efficient. Indias stock exchanges are the responsibility of the Securities and Exchange Board of India (SEBI). The insurance sector has been regulated since 2000 by the Insurance Regulatory and Development Authority (IRDA) which protects premium holders and ensures orderly growth of the industry. Recent international surveys have given Indias disclosure, accounting and board room standards and practices consistently high marks. 11 Chinas financial institutions are regulated by three main agents, all created since 2000: the China Banking Regulatory Commission (CBRC), the China Securities Regulatory Commission (CSRC) and the China Insurance Regulatory Commission. The PBOC shares responsibility for oversight and the rationale for the division of labor responsibilities and authority among them remain somewhat unclear.12 The CBRC is responsible for oversight of retail and wholesale banks but not investment banks which, along with securities houses are the responsibility of the CSRC. The PBOC is responsible for the safety and soundness of the financial system. The decisions of each agency are subject to approval by the State Council which reflects the political concerns of the party leadership. 3. Comparative performance of banks and capital markets 3.1 Banks Since the 1980s both governments have reformed and modernized the banking sectors but public ownership and social objectives have retained their importance and contribute to continuing distortions. China has retained near-universal government ownership at the same time that the central government has tightened prudential standards and oversight and modernized the incentive structures for bank managers. India also has tightened prudential standards and oversight but permitted more diverse ownership.

Despite public ownership of the banks in both countries the relationship between the state banks and state enterprises differs. In China, governments initially owned both the banks and the capital-intensive industrial SOEs who were their major borrowers. As the SOEs were transformed, however, they continued to rely on bank financing in large part because of government priorities to maintain sufficiently rapid economic growth to absorb layoffs. Indias banks, in contrast, have not been used to fund the SOEs directly; the SOEs are financed by the public treasury (but banks are required by RBI to invest a proportion of their assets in public sector bonds). Enterprise ownership is becoming less transparent in China as corporate entities evolve into a mix of state and non-state forms.13 Bank managers take the heat for the bad loans to loss-making firms. In contrast, Indian banks bad loans are concentrated in the government-mandated priority sectors. The cost of SOE financing is thus more transparent to the Indian taxpayer as subsidies and financial support appear in the public accounts and it is implicitly assumed that government will not default on these bonds. The burden of this practice falls on individual and corporate borrowers; as interest rates fell in recent years, bond prices rose attracting banks to the potential profits from those investments and crowding out lending and private investment (Farrell and Lund 2005). More recently, buoyant credit growth has led to interest rate rises and the sale of bonds as a windfall source of funds available for, among other things, removing bad loans as they appear on bank balance sheets. Indian reforms and bank efficiency Indias banking reforms in the early 1990s re-introduced market forces and began to reduce government ownership. Bank efficiency, measured by declining NPA ratios has improved. The dominance of publicly owned banks, however, penalizes more productive borrowers but it has motivated innovation by non-PSBs in priority lending where efforts are underway to create profitable business models (Mor et al 2006). NPA ratios have declined steadily since 2000 (Table 2) and banks have not been recapitalized since the late 1990s.14 Legislation passed since 2002 creates a framework to speed up the liquidation of defaulted loans.15 Creditor rights have been strengthened. Lenders are allowed to settle with borrowers out of court and to sell blocks of bad loans to investors. NPAs tend to be concentrated in PSB loans to the priority sector which accounts for between 40 and 50 percent of their NPAs (RBI website). Requirements for commercial bank presence in the rural areas has led to 71 percent of their branches located there, producing 33 percent of their deposits and accounting for 21 percent of their total loans (Patel 2004). On the positive side, small and medium-sized private companies account for 45 percent of all corporate loans and generate 23 percent of bank revenues (Farrell and Lund 2005). The relationship between ownership and bank efficiency in India is the subject of considerable enquiry and debate. Several studies using aggregate data provide some evidence of greater efficiency of non-PSBs while the findings from micro studies of specific banks reinforce these findings. Using aggregate data, Sarkar, Sarkar and

Bhaumik (SSB 1998), found only weak differences in ownership effects between private over public sector banks with little evidence of a differential in operating efficiency. A later study of cost efficiency found private banks superior to PSBs, but the impact of deregulation in the early 1990s showed no significant differences between the two groups (Kumbhakar and Sarkar 2003). Further, the time trend in cost efficiency shows both groups making progress. But SSB (1998) found moral hazard in PSBs where mangers were less careful in managing risk than were private or foreign banks. Indepth study of a single PSB concludes that Indias financial system under-lends and denies access to credit for many potentially profitable firms. A major reason is that priority lending is perceived to be risky relative to investment in government bonds. Loan officers fear being exposed to charges of corruption if loans go bad; they also lack expertise to evaluate potential profitability of such customers (Banerjee, Cole and Duflo 2006). A further investigation into the role of bank ownership in under-lending, finds private banks to be no less responsive to government-mandated lending priorities than the public banks, with the exception of agricultural lending. A comparison of the intermediary roles of public and private banks confirmed that PSBs are less aggressive than private banks as lenders, attractors of deposits and in setting up new branches. (Banerjee, Cole and Duflo 2005). Two other significant costs of public sector ownership modernization of skills and technologies -- are not as well documented in the literature. PSBs are inflexible in human resource practices because of public sector rules and regulations on salaries and personnel that make personnel restructuring difficult despite a willingness to pay the costs. PSBs also face difficulties attracting new staff and modern skills that are essential to adopting the latest risk management and credit evaluation systems because they are not permitted to compete with the higher financial rewards offered in the private sector. Lacking the necessary skills and technology, PSBs have been slow to introduce nationwide automated banking services and new products; they are therefore losing market share to private and foreign banks.16 In summary, the impact of continued high levels of state ownership and conservative regulation make the PSBs that dominate Indias banking system operate like narrow banks: they take deposits but channel a significant proportion of their investments into low risk government bonds. In 2005 government securities still accounted for 31 percent of bank assets and corporate loans for around 50 percent (IMF 2006b). But the picture is changing. As Indias economic growth has accelerated demand for and returns to bank lending have also improved. The stocks of government securities held by banks dropped in 2006 to the mandatory 25 percent level as banks responded to robust credit demand in retail products like mortgages. In addition, experiments with models designed to turn priority sector lending into profitable businesses have begun to show success. Rural branches have largely failed to deliver finance to the poor for the obvious reason that such lending is high risk and high cost business when borrowers lack collateral and credit histories (Basu and Srivastava 2006).

Efforts by NGOs and micro-finance institutions have successfully restructured borrower profiles in some parts of the country. These groups help savers (mainly women) to pool their savings and make small loans to members using peer pressure to enhance loan repayment probabilities. Some commercial banks are now experimenting with linkages through which the groups savings are deposited and borrowing facilitated by group guarantees posted as collateral. ICICI Bank is pioneering the securitization of micro loan portfolios using local entities as the administrators (Mor et al 2006). These efforts are in their early stages, but show some promise in addressing the transactions cost and risk problems inherent in the business. Of course, the test of such approaches will be whether credit quality is maintained in the next economic downturn. Chinas banks and efficiency Chinas challenge differs in that the banks to be transformed into commercial banks were originally policy banks, lending according to government priorities. Since 1998 China has followed a three-step strategy to introduce market-based incentive structures into the four largest state owned commercial banks (SOCBs). The test of these reforms lies in banks ability to allocate credit efficiently and manage risks successfully. This section explains why, by these criteria, the reforms are still insufficient. The first step in the three-step strategy was to inject capital to strengthen the capital bases of the banks to meet BIS standards. Since 1999, the banks have both written off NPLs and transferred them to four state-owned asset management companies (AMCs) which issued government-guaranteed bonds in return. The second step was to attract strategic foreign investors willing to purchase equity stakes in the banks (restricted to 20 percent for any single investor and 25 percent total), to contribute directors to the boards and assign foreign managers to the banks. The third step was to offer small amounts of equity to institutional and other investors through IPOs on the Hong Kong and Shanghai Stock Exchanges. The purpose of this step was primarily to force bank managers to increase the transparency of their reporting and to expose them to market evaluations by bank analysts.17 At the end of 2006 this process was near completion in three of the four SOCBs. As they cleared their balance sheets of legacy bad loans, they generated optimism that they had put their problems behind them. NPL ratios for three of the largest banks had declined to single digit levels in 2005 (Table 3). Estimates of the cost of removing the bad loans vary, depending on assumptions, but the government found the necessary funds, from the treasury and foreign exchange reserves. Looking to the future, the central issue is whether modern bank incentive structures have made Chinas misdirected lending (mainly to SOEs) a thing of the past. There are several reasons to argue that more needs to be done. First, government influence is still pervasive because of political priorities to maintain economic growth, job creation and social stability. Chinas growth is being

driven primarily by rapid rates of investment which grew at unsustainable rates of 34 percent in 2004, 16 percent in 2005 and continued into 2006. The banks mainstay borrowers are firms, many of them government-owned or -controlled. By the large banks own published financial statements corporate customers still account for between 70 and 80 percent of their loans.18 Second, in 2005, more than 40 percent of the industrial SOEs were losing money and current data indicate the losses at government-controlled firms continue to mount.19 While the largest SOEs are reporting burgeoning profits and financing new investments themselves, revenues and profits are concentrated among the large: in 2005 the ten largest accounted for over 53 percent of total revenues and the 165 SOEs owned by the central government accounted for more than 70 percent of SOE profits.20 Yet China still has 120,000 SOEs. The implication is that banks exposures are likely to be greater to the tens of thousands government owned or controlled firms whose profits appear to be much less certain. Third, the governments approach to unsustainable growth is to try to slow it by administrative guidance to the banks on the sectoral allocation of loans; but such guidance is a very blunt instrument that distorts credit decisions emphasizing sectoral restrictions rather than decisions based on the risk and productivity indicators of borrowers and projects. For example, a productive and profitable borrower in a restricted sector will be denied credit while a less productive borrower in a permitted sector will have access just the opposite outcome from that predicted by market forces. Small entrepreneurial borrowers also have difficulties accessing bank credit because of regulations requiring the banks to demand high levels of collateral.21 Micro level and anecdotal evidence adds to concerns about bank inefficiency and inability to price their loans to reflect credit risk. One study of the determinants of banks loan growth rates during the 1997-2004 period found that corporate profitability of the banks commercial customers had no effect on loan growth and that the large state owned banks were losing market share to other financial institutions more quickly in the provinces with more profitable customers. Data on loan pricing patterns at the banks since 2004 show that interest rates charged borrowers are very compressed around the benchmark rate, suggesting little ability or preference to price for risk (Podpiera 2006). Finally, governance in Chinas majority government-owned banks is a work in progress that affects incentive frameworks. While steps have been taken to increase representation by independent directors on boards, the involvement of Communist party officials, while declining, continues to be pervasive. Bank heads are members of the Central Committee (Naughton 2003); the CEO is often also the party secretary; bank performance is discussed at party meetings.22 These ownership and governance weaknesses contribute to attitudes among investors, depositors and customers that Chinas government-owned banks are too big to fail. The Big Four are more subject to external monitoring than they were, but capital injections and continued government involvement in their governance undermine

efficiency. Depositors believe they have blanket protection of their deposits even if the rate of return is low. Deposit insurance is one way to remove blanket protection, but no date has been set for such an innovation. Moreover, even if the formal rules change it remains to be seen whether depositors would actually be forced to bear losses should a bank fail. 3.2 Capital market development Chinas corporate bond market is a work-in-progress for several reasons. Issuers face numerous regulatory restrictions; bankruptcy laws have only recently been adopted; default procedures are not yet based on market principles; investors lack the transparency afforded by a credit rating system, modern accounting standards and transparency by issuers; market discipline has not been established and investor education is insufficient (Zhou 2005). Nevertheless, the stock exchanges are quickly assuming their intended roles as liquidity and credibility problems ease following such innovations to increase demand from foreign institutions as the QFII program and supply side developments including a recent spate of new share offerings by SOEs and several large state-owned banks. Indias under-developed corporate bond market shows few signs of revival. Stringent government regulations continue to make it costly and difficult to issue domestic bonds; infrastructure is inadequate: legal recourse is complicated and bankruptcy laws are ineffectual (Jadhav 2005). Another factor is the willingness of banks to issue 5-year credits that reduce the incentives for corporations to dis-intermediate the banks. And foreign issues have helped to satisfy Indian companies appetite for long term finance. Companies are permitted to issue foreign currency convertible bonds (FCCBs) and private placements to international institutional investors, a channel that avoids the costs and obstacles facing issuers of domestic instruments. Further capital market development is clearly needed in both countries. China has the larger task of developing the institutions and instruments of direct finance by creating appropriate institutional and regulatory frameworks that facilitate market forces and promote market discipline. 4. Prospects for further reforms to promote efficiency and soundness Momentum for reform exists in both countries within the constraints of continued state ownership of banks. Bank regulators have tightened prudential standards. The Indian government has also strengthened creditor rights following the 2002 Sarfaesi Act legislation; ownership reforms dating back to the early 1990s are bearing fruit as private banks create more competition for PSBs. While PSB efficiency has increased, however, they are still not allowed to consolidate or be acquired by private banks (Economist 2006). Declining central government fiscal requirements23 and recent buoyant growth have also reduced banks investment in government bonds to the statutory minimum. In China, three of the Big Four state owned banks now have relatively clean balance sheets. But the economic and credit booms in 2002-04 have tested their ability to evaluate and monitor new loans; bank loans have soared (reducing NPL ratios) in response to robust

10

demand in expanding industries. Of course the most potent stimulus to greater efficiency in financial institutions is more competition, modern management technologies and incentives to manage risks and rewards. In this section we assess the prospects for more competition, new knowledge and skills and better allocation of capital, particularly in the banks. One source of new knowledge and skills is foreign entrants and here the policies differ. As theoretical and empirical analysis suggests, foreign entrants improve financial sector efficiency by stimulating competition and injecting new products, skills and technologies.24 Chinas bold rationale for foreign entry was embedded as a precommitment in WTO accession talks. Foreign entrants are restricted to 25 percent total equity stakes but are expected to help change banks incentive systems by bringing modern skills and institutional structures to their domestic partners, an approach used in the industrial sector. State-owned banks have also been permitted to list on the Hong Kong Stock Exchange, primarily to encourage transparency and responsiveness to modern market evaluations of their profitability and efficiency.25 In comparison, RBIs road map for foreign entry26 laid out in its 2005 Annual Report is cautious. Foreign investors are restricted to 24 percent of the equity in a small number of private banks that the RBI wishes to see restructured. Larger equity stakes are permitted up to 49 percent if a banks board and shareholders approve. Wholly owned subsidiaries of foreign banks are accorded national treatment. In 2009 permitted activities will extend to other private sector banks but stakes in such banks will be limited to 74 percent. There is no explicit plan to reduce the public share of the PSBs or to allow nonstate stakes in those entities. A second stimulus to competition comes from ownership changes. Performance comparisons by ownership are difficult, but current market indicators are available for some of the largest banks (including private banks in India) in both countries (Tables 4 and 5). Chinas banks are much larger than Indias as measured by market capitalization, assets and loans. The State Bank of India is the only bank that comes close in size. China Construction Bank, which listed on the Hong Kong Stock Exchange in 2005, is not Chinas largest bank (rather it is the Industrial and Commercial Bank of China which listed in Hong Kong and Shanghai in late 2006) but its deposit share is much larger than those of the next two largest banks for which public information is available (Table 4). What do we learn about the relative efficiency of the state-owned banks? First, Indias private bank (HDFC) is the leader on most performance indicators (Table 5); Indian bank averages for net profits, ROA and ROE are higher than the Chinese averages. Net interest margins are high in both countries, reflecting the continued administration of interest rates and the opportunity afforded banks to cover their costs by maintaining large spreads. This indicator is a sobering one because it also reflects banks favored positions, lack of competition and low productivity, the costs of which are borne by borrowers and savers. The other main indicator of future prospects for the banking systems is the

11

potential size of future loan losses and uncertainty about the necessity of yet more bailouts. Reasons for skepticism that China has put its problems behind it were laid out in section 3. Future loan losses are not perceived to be a problem in India, although some analysts have noted persistent regulatory forebearance and non-transparent subsidies of some troubled state-owned financial institutions (Patel 2004). But Indias buoyant growth since 2003 has fueled rapid credit growth which suggests little room for complacency. Two potential triggers of future problems in China are perceived to be stiffer competition from foreign entrants and slower growth. Foreign competition is the less likely source since foreign entrants perceive Chinese customers as underserved and interested in new products that domestic banks will later learn to produce. Some customers might migrate to foreign banks but as in most other countries, domestic customers tend to stay with the brands they know unless things go wrong. The more likely trigger for problems will be slower domestic growth or a negative external shock. High domestic growth rates have been fueled by rising investment funded partly by bank loans. At the same time, state companies have not been required to pay dividends, allowing them funds to fuel even more investment whether or not it has a positive real rate of return. Such misallocation of capital, as the Bank for International Settlements has noted will eventually manifest itself as falling profits and this will feed back on the banking system, the fiscal authorities and the prospects for growth more generally.27 Another bailout will probably be required, but by most calculations it will be affordable. One way to estimate future loan losses is to age the Special Mention loans publicly reported by three of the Big Four into loan losses in 2007 or 2008. Assuming the Special Mention loans become non-performing in 2007, an estimated RMB 1.5 trillion new bad loans would appear from lending during the current boom (around US$ 200 billion) and about 7.2 percent of 2007 GDP (Dobson and Kashyap 2006). Lardy (2004) uses an alternative method examining the impact on fiscal sustainability of the implied added burden of interest obligations of AMCs to the banks and the increase in NPLs resulting from loans made during the credit boom in the 200204 period. In the event of a growth slowdown, these liabilities would negatively affect fiscal sustainability. In two alternative scenarios, where 20 percent and 40 percent of the new loans become non-performing, he finds that the debt-to-GDP ratio rises at first, but then declines through the period to 2013. In other words, fiscal sustainability is preserved during this period. Indias buoyant growth has brought a credit boom which causes similar concerns about the consequences of a future growth slowdown. Since 2004, credit has grown by 30 percent (IMF 2006b), partly because of the wider use of bank credit with financial deepening and partly because of misdirected credit. The sectoral breakdown of credit growth in Table 6 suggests other reasons to be concerned. Overall credit growth was nearly 28 percent, far in excess of the governments 23 percent target. The priority sectors accounted for 40 percent of this growth (IMF 2006b:57).

12

Sudden rapid credit growth often foreshadows deteriorating loan quality that will challenge risk management at the banks and raise questions about weaknesses in prudential standards and supervision. RBI data show that the PSBs had a higher NPA ratio in 2005-06 (3.7 percent) than the private banks (2.4 percent) and foreign banks (2.1 percent). The same data also show that the priority sector share of PSB non-performing advances had risen to 54.1 percent of total PSB NPAs from 48.1 percent the previous year. The PSBs dominate the surge in lending to priority sectors. IMF (2006b) performed stress tests to assess the vulnerability of the banking system to loan quality deterioration and concluded that most banks are able to maintain their capital adequacy ratios close to 9 percent and are therefore unlikely to cause systemic risk. 28 This discussion underlines the costs of distortions associated with high levels of government ownership. In China, this supports the assumption that government will do whatever is necessary to maintain soundness. In India, the PSBs dominate the concerns about a new crop of bad loans, suggesting that they are responding to public sector economic development priorities rather than risk-based lending practices. Public resources may be adequate to fix problems, but further bailouts only continue to distort incentives and divert public funds, for example, from Chinas current 5-year Program priorities to enhance public services in rural areas and speed urbanization. 5. Conclusion The financial systems in both India and China are developing but somewhat slowly relative to the speed with which their economies are increasing in size and complexity. Their banking systems are increasingly sound; they are also more efficient but they are still agents of capital accumulation rather than of innovation and technical change. The Indian government directs banks to finance government deficits and social projects at the expense of non-priority borrowers. Chinas huge bank-dominated stateowned financial system persists in lending to entities associated with the state regardless of their productivity. Banks in both countries are under-lending to the agents of economic change and job creation: small, entrepreneurial entities that lack political connections, government ownership, or government contracts and guarantees. Formal finance, despite official efforts in India, reaches a small proportion of entrepreneurs but a larger share than in China. Instead those entities must rely on retained earnings and informal finance at much higher cost of capital which suggests that the high growth rates are financed less by banks than by retained earnings and informal financial institutions in China and by Indias capital market institutions. This in turn suggests that governments believe growth to be adequate to support the cost of using banks to pursue political objectives. The implication is that both countries could have grown even faster if they had more efficient financial sectors. But as these economies become more complex the distortions in evaluating credits and managing risk will exacerbate vulnerabilities to shocks and industrial setbacks as well as uncertainty about governments future fiscal liabilities.

13

The analysis in this paper implies that more reform would be desirable to improve capital allocation by increasing competition and spurring development of capital markets in China. As discussed in the previous section there are ways to do this through more domestic competition (and consolidation of small entities) and more foreign participation. China has made better use of this channel than India, using foreign entrants to introduce modern skills, products and new technologies, encourage greater transparency, and fan the winds of domestic competition. Privatization (or at least more widely-held local ownership) is another channel and here India has the better record. Chinas success with gaizhi in the industrial sector (transformation of industrial SOEs by grasping the large and letting go the small, maintaining state ownership of selected large entities while restructuring and privatizing legions of smaller ones), should also be considered in the banking sector.29 Competition in Indias banking system could be increased though consolidation of the smaller PSBs (and indeed allowing takeovers by the more efficient and fast growing private banks). Market forces could also be allowed to influence how banks meet priority lending targets by making obligations tradable as Basu and Srivastava (2006) suggest. Increased competition for the banks from a better developed corporate bond market would stimulate efficiency. More competition for capital will also raise interest rates and the cost of servicing the public debt. In turn, Indian governments and public enterprises would have to reduce their dependence on bond finance by cutting their fiscal deficits. Greater clarity is needed about the future role of Chinas largest state owned banks. If majority government ownership continues, serious consideration should be given to changing the banks structures to segregate risk on the lending side or to align lending mandates with the inadequate incentives and capacity of these banks to evaluate credit risk accurately. There are at least two ways to segregate risk. One is to restructure the banking sector more clearly into policy banks, leaving the commercial banks as narrow banks that take deposits and invest them in low risk assets (a lesson to be learned from Indias experience). Another alternative is to divide the banks into good and bad banks, with the bad assets and deadbeat customers moved into the bad banks and the good banks freed to operate strictly on market criteria (Dobson and Kashyap 2006). Finally, certain regulatory changes can be envisaged that would modernize banks incentive systems. Best practice deposit insurance would create incentives for monitoring by depositors. Indias deposit insurance system, however, may not be the model as it is considered to be overly liberal and contribute to, rather than reduce, moral hazard (Patel 2004). More independent financial regulators (eg, freeing bank regulators from the central banks) would allow them the freedom to focus more effectively on soundness and efficiency. At present Chinas CBRC must satisfy both PBOC and the State Council before it can implement regulations. Indias plethora of oversight agencies includes RBI, the National Housing Bureau, as well as overseers of cooperatives, insurance and securities. RBIs multiple roles can conflict with bank soundness and efficiency despite RBI officials statements to the contrary. For example, its ownership position in the State Bank of India means that its views dominate those of minority shareholders in

14

commercial decisions. RBI responsibilities for debt management will influence its views on commercial banks investment requirements and its concerns with bank soundness may conflict with monetary policy decisions. Balasubramanian et al (2005)puts forward a thoughtful position, recommending the creation of a single Financial Services Agencytype regulator, with RBI playing a coordinating role among the existing regulators during a phased transition to the single agency. In the end, RBI would focus only on monetary policy. In conclusion, this survey of the financial systems in China and India indicates that reforms to date have reduced the risks of systemic banking crises but capital is still being misallocated by banks, corporate bond markets and related infrastructure are still underdeveloped. Both banking systems, with high levels of government ownership, are vulnerable to future growth downturns. Both financial systems lag behind the increasing complexity of the domestic economies and are likely to slow integration into world capital markets that comes with full capital account convertibility. Ironically, without a crisis, India which has all the components of a modern financial system seems unlikely to remove the political constraints on its full development. Similarly, in the absence of slower economic growth and a further bank bailout, China is unlikely to resolve the tensions between the political goal of gradual controlled change and an efficient commercial banking sector.

15

Table 1. Structure of the Banking Industry, China (2005) and India (2003-04) A. The Chinese Banking Industry, 2005
(CNY billion) Number of institutions 34,045 4 Assets (CNY) 374.697 196.580 Market share,% 100 52.5 Liabilities (CNY) 358.070 187.729 Market share,% 100 52.4

All banks Big Four commercial banks Joint stock 12 58.125 banks City 112 20.367 commercial banks Other 33,917 99.625 Source: CBRC website, accessed June 2006.

15.5 5.4

56.044 19.540

15.7 5.5

26.6

94.757

26.5

Notes: 1) All banks include policy banks, state-owned commercial banks, joint stock commercial banks, city commercial banks, rural commercial banks, urban credit cooperatives, rural credit cooperatives, postal savings, foreign banks and non-bank financial institutions. 2) Big Four commercial banks include the Industrial and Commercial Bank of China (ICBC), Agricultural Bank of China (ABC), Bank of China (BOC), and the China Construction Bank (CCB). 3) Joint stock banks include Bank of Communications, CITIC Industrial Bank, Everbright Bank of China, Huaxia Bank, Guangdong Development Bank, Shenzhen Development Bank, China Merchants Bank, Shanghai Pudong Development Bank, Industrial Bank, China Minsheng banking Co. and Evergreen Bank. 4) Other consists of rural commercial banks and rural credit cooperatives, policy banks, the postal savings bureau, finance companies, trust and investment companies and financial leasing companies.

B. The Indian Banking Industry, 2006*

(Rupees billion) Number of institutions 28 28 28 218 Deposits (Rupees) 16,225 4,282 1,137 21,644 Market share, % 75 19.7 5.3 100 Loans*(Rupees) 11,347 3,168 989 15,504 Market share, % 73.2 20.2 6.4 100

Public Sector Banks (PSBs) Private banks Foreign banks Total

Source: RBI A Profile of Banks, 2005-06; Statistical Tables Relating to Banks in India, Tables 3.1, 7.1 and 7.2. Accessed at www.rbi.org in December 2006. Notes: * This date refers to 2005 fiscal year which ended on March 31, 2006. These aggregate statistics refer to Indias Scheduled Commercial Banks (SCBs). They do not include regional rural banks, a number of cooperatives and development finance institutions.

16

Table 2. NPAs in Indian banks, 2000-05 (Rupees billion) NPAs Total Advances NPAs / Total Advances,% 2000 2001 2002 2003 2004 2005 2006 608 640 710 703 649 575 519 4,758 5,587 6,809 7,765 9,020 11,712 15,504 12.8 11.4 10.4 9.1 7.2 4.9 3.3

Source: RBI. Statistical Tables Relating to Banks of India, Table 7.1 Bank Groupwise Classification of Loan Assets of SCBs, 2000-05. Accessed at www.rbi.org.in. Table 3. Reported NPLs in Chinas Big Four banks, 2000 and 2005 (RMB billions; NPL% = % total loans)
Loans (2000) 1484.3 NPL% na Loans (2005) 2829.3 NPL% 26.2

Agricultural Bank of China (ABC) China Construction Bank (CCB) Industrial and Commercial Bank of China (ICBC) Bank of China (BOC)* Total Loans/GDP

ABC

1386.4

20.3

CCB

2458.4

3.8

2413.6

34.4

ICBC

3289.6

4.7

1505.8 6,790.1

27.2 28.7** 76.0

BOC

1800.1 10377.4

5.5 10.5 55.9

* reported for domestic loans only; **loans and NPLs for only 3 reporting banks The ratio of NPLs is based on the BIS five-category loan classifications Sources: Bank annual reports; BOC 2006 IPO Memorandum; CEIC data

17

Table 4. Market indicators, Chinese and India banks, 2004-2005 (US$ million)
Market cap. China (2005) Bank of Communications China Construction Bank (CCB) China Merchants Bank India (2004) State Bank of India (SBI) HDFC Bank Punjab National Bank 10,770 5,271 3,259 100,302 11,217 27,534 44,138 5,576 13,176 80,054 7,929 22,501 5,250 986 1,712 21.5 2.1 6.0 25,988 86,921 9,786 170,130 554,679 89,519 95,204 305,036 57,281 148,955 482,578 78,345 9,611 35,926 3,180 4.0 13.1 2.1 Total assets Loans Deposits Shareholders equity Deposit market share,%

Source: Ramos et al (2006)

Table 5. Indicators of Bank Performance, China and India, 2005 Net interest margin, % 2.8 2.7 Price earning ratio 18.96 22.6 13.8 21.0 11.1 11.2 25.6 9.0 13.9 Net profit ROA,% (% average assets) 0.6 0.7 1.1 0.6 1.5 0.9 1.5 1.2 1.2 0.6 0.7 1.1 0.6 1.6 1.0 1.4 1.2 na ROE, %

China (2005) Bank of Communications CCB 2.9 China Merchants 3.0 Bank India (2005) 3.2 State Bank of 3.2 India HDFC Bank 4.4 Punjab National 3.6 Bank Hong Kong 2.4 Source: Ramos et al (2006)

15.3 13.3 19.0 17.2 17.1 16.1 18.1 17.7 13.7

18

Table 7. Credit growth by sector, 2003-04 and 2004-05 (yoy, percentage change) Sector 2003-04 Priority sectors 24.7 Agriculture 23.2 Small-scale industry 9.0 Others 38.3 Industry (medium and large) 5.1 Petroleum -16.8 Infrastructure 41.6 Autos -5.8 Cement -11.5 Housing 42.1 Nonbank financial 18.9 companies Wholesale trade 10.1 Export credit 17.2 Gross (nonfood) bank credit 17.5 Source: IMF 2006b, Table V3. 2004-05 31.0 35.2 15.6 37.0 17.4 19.2 52.3 20.9 7.4 44.6 10.8 36.0 14.3 27.9

19

References Allen, Franklin, Jun Qian and Meijun Qian. 2005. Chinas financial system: past present and future. Forthcoming in Chinas Great Economic Transformation, edited by Loren Brandt and Thomas Rawski. Cambridge University Press. Balasubramanian, Ramnath, Raj Kamai, Leo Puri, Joydeep Sengupta, Toshan Tamhane and Renny Thomas. 2005. India Banking 2010: Towards a High-performing Sector. Mumbai: McKinsey & Co. Bank for International Settlements. 2006. Annual Report 2006. Basel: BIS. Basu, Priya and Pradeep Srivastava. 2006. Scaling-up Microfinance for Indias Rural Poor. In Aziz, Jahangir, Steven Dunaway, Steven Vincent and Eswar Prasad, Eds. China and India Learning from Each Other: Reforms and Policies for Sustained Growth. Washington, DC: IMF. Banerjee, Abhijit, Shawn Cole and Esther Duflo. 2006. Bank Financing in India. In Wanda Tseng and David Cowen, Eds. Indias and Chinas Recent Experience with Reform and Growth. New York: Palgrave McMillan. -------------------. 2005. Banking Reform in India. In Bery, Suman, Barry Bosworth and Arvind Panagariya. India Policy Forum, Volume 1. Washington, DC and New Delhi: Brookings Institution and National Council of Applied Economic Research. Bhattacharya, Saugata and Urjit R. Patel. 2003. Reform Strategies in the Indian Financial Sector. Paper presented at Conference on Indias and Chinas Experience with Reform and Growth. November. Accessed at www.imf.org January 2004. Beck, Thorsten. 2006. Creating an Efficient Financial System: Challenges in the Global Economy. World Bank Policy Research Working Paper 3856. Washington: World Bank. Dobson, Wendy and Anil Kashyap. 2006. The contradictions in Chinas gradualist banking reform. Brookings Papers on Economic Activity 2006:02.Washington, DC: Brookings Institution (available at http://www.rotman.utoronto.ca/dobson). Dobson, Wendy and Pierre Jacquet. 1998. Financial Services Liberalization in the WTO. Washington, DC: Institute for International Economics. Economist, The. 2006. Naturally gifted in A survey of international banking. May 20. Embassy of the Peoples Republic of China to the United States of America. 2006. Profit of SOEs exceeds 900 bln yuan in 2005. February 24.

20

Farrell, Diana, Susan Lund and Leo Puri. 2006. Indias Financial System: More market, less government. The McKinsey Quarterly. August. Accessed at www.mckinseyquarterly.com, September 2006. Farrell, Diana and Susan Lund. 2005. Reforming Indias Financial System. The McKinsey Quarterly special edition. Accessed at www.mckinseyquarterly.com October 2005. Farrell, Diana and Aneta Marcheva Key. 2005. Indias Lagging Financial System. The MicKinsey Quarterly. No. 2. Accessed at www.mckinseyquarterly.com November 2006. Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge MA: Harvard University Press. Hanson, James A. 2004. The Transformation of State-Owned Banks. In Caprio, Gerard, Jonathan L. Fiechter, Robert E. Litan and Michael Pomerleano, Eds. The Future of State-Owned Financial Institutions. Washington, D.C.: Brookings Institution Press. International Monetary Fund. 2005. Peoples Republic of China: Staff Report for the 2005 Article IV Consultation. Washington, DC: IMF. July 8. --------------. 2006a. India: Staff Report for the Article IV Consultation. IMF. February. --------------. 2006b. India Credit Growth and Related Risks in India: Selected Issues. January 12. Jadhav, Narendra. 2005. In Jahangir et al, Eds Jahangir, Aziz, Steven Dunaway, Steven Vincent and Eswar Prasad, Eds. 2005. China and India Learning from Each Other: Reforms of Sustained Growth. Washington, DC: IMF. Kumbhakar, Subal C. and Subrata Sarkar. 2003. Deregulation, Ownership, and Efficiency Change in Indian Banking. Arthaniti. 2:1-26. LaPorta, R., Lopez-de-Silanes, F., and Shleifer, A. 2002. Government Ownership of Commercial Banks. Journal of Finance. 57:265-301. ----------------. 2003. What Works in Securities Law? NBER Working Paper No. 9882. August. Lardy, Nicholas. 2004. State-Owned Banks in China. In Caprio et al, Eds. Levine, Ross. 1997. Financial Development and Economic Growth: Views and

21

Agenda. Journal of Economic Literature. 35:688-726. Lewis, Sir Arthur. 1950. The Principles of Economic Planning. London: Allen & Unwin. Ma, Guonan, 2006, Sharing Chinas Bank Restructuring Bill, China & World Economy, Volume 14(3),19-37. Madgavkar, Anu, Leo Puri and Joydeep Sengupta. 2001. Revitalizing Indias Banks. The McKinsey Quarterly: special edition. September. Mor, Nachiket, R. Chandrasekar and Diviya Wahi. 2006. Banking Sector Reform in India. In Aziz, Jahangir et al, Eds. Naughton, Barry. 2003. The State Asset Commission: A Powerful New Body. China Leadership Monitor, No.8. Spring. Organization for Economic Cooperation and Development. 2005. Survey of China. Paris: OECD. September. Patel, Urjit R. 2004. Role of State-owned Financial Institutions in India: Should the Government Do or Lead? In Gerard Caprio et al, Eds. Podpiera, Richard. 2006. Progress in Chinas Banking Reform Sector: Has Bank Behavior Changed? IMF Working Paper WP/06/71. Washington, DC: IMF. Ramos, Roy, Ning Ma and Jennifer Meng. 2006. China Banks: where to for valuations? India banks read-across. Hong Kong: Goldman Sachs Investment Research. February 2. Reserve Bank of India. 2005. Reserve Bank of India Annual Report 2005. www.rbi.org.in. Sarkar, Jayati , Subrata Sarkar and Sumon K. Bhaumik (SSB). 1998. Does Ownership Always Matter? Evidence from the Indian Banking Industry. Journal of Comparative Economics. 26:262-281. Union Bank of Switzerland (UBS). 2006. China Bond Market Research. December 4. Zhou Xiaochuan. 2005. Chinas corporate bond market development: lessons learned. November 17-18. BIS Papers No. 26.Accessed at www.bis.org, September 28, 2006. Endnotes
1 2

Farrell et al 2006. Farrell et al 2006. Corporate debt accounted for 35 percent and equity for 34 percent of total financial assets in the United States that year.

22

3 4

See for example, Levine (1997); Allen and Gale (2001); Beck (2006). Gerschenkron (1962) was one of the first to make the case that in a weak institutional environment, private banks are unable to overcome deficiencies in information and contracting. Lewis (1950) advocated government ownership of banks to develop strategic industries. 5 Priority sector lending is monitored by the Ministry of Finance and reported regularly. See www.indiabudget.nic.in. 6 ICICI was founded as a state development bank in 1955; it formed a commercial banking subsidiary in 1994 which merged with the parent in 2002 as a single publicly-listed company. HDFC Bank was also created in 1994 by a state-owned mortgage company. 7 For example, ING bought 20 percent of Vysya Bank and Chase Capital acquired 15 percent of HDFC Bank (Madgavkar et al 2001). 8 In 1990 ten companies had listed on these exchanges; by the end of 2006, the number had grown to 1500. 9 UBS (2006). 10 See Battacharya and Patel (2003); Patel (2004); and Banerjee et al (2006). 11 See LaPorta et al (2003). 12 The single regulatory model typified by the Financial Supervisory Agency in the UK has been studied; a regulatory merger is a future possibility. 13 For example, reported corporate entities in bank annual reports include state, collective and shareholding forms of ownership as well as private, foreign and other. 14 By 2002 capital injections had cumulated to Rs. 225 billion (Patel 2004), or roughly 10 percent of nominal 2002-03 GDP, a number that does not include a number of indirect bailouts of troubled public sector financial institutions, such as temporary tax exemptions and government guarantees. 15 This legislation is known as the Securitization and Reconstruction of financial Assets and Enforcement of Security Interest (Sarfaesi) Act of 2002. 16 Authors interviews, Mumbai, November 2006. 17 Dobson and Kashyap (2006) make a conservative estimate at 10.8 percent of 2005 GDP for the big four banks, while Ma (2006) estimates 19.4 percent as the cost for the entire banking system. 18 The China Construction Bank, which breaks out its loans by the legal form of borrower, reports that loans to SOEs grew by nearly 9 percent in the first six months of 2006. Another large bank reports the 10 largest borrowers; half are SOEs (see Dobson and Kashyap, 2006: 9). 19 See Dobson and Kashyap (2006). 20 See Embassy of PRC (2006). 21 OECD (2005). 22 It is worth noting that the pervasive role of the party does not appear in the IPO memorandums of any of the three banks that went public in 2005-06. 23 IMF (2005) reports Chinas overall budget balance as a deficit of 2.1 percent of GDP while IMF (2006a) reports Indias central government balance in 2005-06 as a deficit of 4.3% of GDP and the general government balance as a deficit of 7.7 %. 24 See Dobson and Jacquet (1998) for a review of the literature and evidence on market entry by foreign banks. 25 Even so, this part of the modernization strategy has drawn criticism that banking assets should not be sold to foreigners and that the price on what has been sold has been too low. 26 RBI (2005:137) Box V.2 Road Map for Presence of Foreign Banks. 27 Bank for International Settlements (2006). 28 One exception is four old private banks accounting for 12 percent of bank assets whose capital bases would erode to dangerous levels. 29 Recent listings by Bank of China and Industrial and Commercial Bank of China on both the Hong Kong and Shanghai Stock Exchanges are moves in this direction.

23