A. Introduction

103. Financial reforms are a key ingredient of successful economic development. A large body of research has identified the various channels through which financial sector reforms can enhance growth performance and reduce poverty.⁴² In particular, more efficient financial institutions and markets should contribute to higher levels of savings collection and of private investment, through the diversification of savings product and the pooling of risks. Financial reforms should also ultimately lead to improved capital allocation and higher profitability through a better dissemination of information on economic opportunity and a rationalized assessment and monitoring of investment projects. Econometric studies show a positive impact of financial deepening on economic performance in the long run. For instance, according to a recent cross-country study, a 10 percentage point increase in financial deepening (measured by credit to the private sector in percent of GDP) is likely to increase the annual growth of total factor productivity by about 0.5 percent.⁴³ International experience suggests, however, that there is no mechanical link between credit expansion and growth. A precondition for a sustainable deepening of financial intermediation is the building of a strong and stable private-based financial sector that can resists shocks and strengthen confidence of depositors. Therefore, the positive impact of financial reforms may not be visible in the short-to-medium term.

104. Relying mostly on a recent in-depth study by the State Bank of Pakistan (SBP),⁴⁴ this chapter will focus on Pakistan’s banking sector. It will first briefly review the reforms in Pakistan during the 1990s, assess the current situation of the banking sector, analyze why the impact of the reforms has been limited so far and conclude with a discussion of present programs and policy options for further reforms. The last section is devoted to microfinance.

B. A Decade of Reforms

105. Pakistan’s authorities became aware in the late 1980s that the financial strategy pursued in the 1970s, including the full nationalization of banks, was leading the economy to a severe decline and should be reversed. Financial structural reforms were initiated in 1989 and implemented throughout the 1990s, with increasing speed after 1997. Returning to market-based monetary and exchange rate policies and shifting from direct state intervention to a modem regulatory approach were the two pillars of the strategy to reform the financial sector reform. While the direction of the reforms was clear, progress was uneven, slowed down by deteriorating macroeconomic conditions, strong vested interests, the increasing cost of the reforms, and political instability. These reforms were conducted in parallel with the liberalization of the foreign exchange payment system and the introduction of market-based government debt instruments.

106. Privatization. In 1990, all banks were still state-owned. Two out of the initial five nationalized banks, Muslim Commercial Bank (MCB) and Allied Bank (ABL), were privatized in steps between 1991 and 1993, while the banking sector was steadily opened to domestic, then foreign competition. As the balance sheet of the three remaining nationalized banks deteriorated strongly, further privatization efforts became dependent on the implementation of restructuring programs, including branch closures and staff downsizing through (costly) golden handshakes. Professional management of the three nationalized banks was established in 1996. Balance of payment crises, notably in 1996 and 1998, delayed the privatization efforts, which resumed significantly only in 2000. In 2000, the banking sector in Pakistan consisted of 39 commercial banks totaling PRs 1,641 billion of assets, equivalent to 49 percent of GDP, with 70 percent of the banks’ assets held by state-owned banks.⁴⁵ The state held a participation in eight banks, of which a majority stake in three large commercial banks (Habib Bank (HB), United Bank Limited (UBL) and National Bank of Pakistan (NBP)), and a minority participation in Muslim Commercial Bank (MCB) and Allied Bank. The rest of the banking sector was about equally split between 20 foreign banks, of which three large ones (ABN-AMRO, Citibank, and Standard Chartered Bank) and 12 private domestic banks (including Union Bank).⁴⁶ In September 2002, UBL was sold to a foreign group, while Habib Bank (HBL) has been brought close to the point of sale, with strategic shares to be sold to private investors by the end of 2002.⁴⁷

107. Financial liberalization. Credit ceilings imposed on individual banks were gradually liberalized and eventually abolished in 1995. They were replaced by the SBP’s indirect regulation of banks’ liquidity, through a system of cash reserve (CRR) and statutory liquidity requirements (SLR),⁴⁸ and through the conduct of open market operations. The automatic discount window was replaced by a three-day repurchase facility in 1992. Various subsidized credit schemes were eliminated or their rates were gradually linked to market rates, including the export finance scheme in 2000. Mandatory schemes were removed except for agricultural credit. Maximum lending rates were abolished in 1995.

108. Institutional strengthening. Autonomy was gradually granted to the SBP in several steps from 1994 to 2001. Prudential regulations were issued for banks in 1992, which were made gradually compliant with the Basel Accord Core principles. Banks’ minimum capital requirements were increased in 1997 and again in 2002. Starting in 1997, banking supervision was enhanced through a comprehensive set of soundness indicators and enhanced offsite and onsite inspection procedures. In 1995, nonbank financial institutions were requested to be rated by credit-rating agencies, followed by banks in 2000. Those ratings are now regularly published on the SBP’s Internet website.

109. Treatment of nonperforming loans (NPLs). Deteriorating macroeconomic conditions and political interference undermined banks’ portfolios during the last decade. The size of the NPLs became clearly apparent only at the end of the decade, reflecting a gradual tightening of the regulatory framework and improved bank supervision. In 1992, the SBP strengthened guidelines for loan classification and provisioning, and requested that all loans with payments late by more than 90 days received a special mention. Stricter disclosure requirements were put in place in 1997. In response to the emergence of a large outstanding stock of bad loans and to the requirements to clean the balance sheets of state-owned banks before their privatization, the authorities gradually strengthened legal and judicial procedures, starting in 1993. In 1997, new banking courts were established and a revised recovery law was promulgated. In October 1999, an accountability and recovery drive was initiated. The Corporate and Industrial Restructuring Corporation (CIRC) was established in September 2000 as an autonomous body under the Ministry of Finance, to “promote the revitalization of the economy by reviving sick industrial units.” Its main function is to recover minimum one-year overdue loans of PRs 30 million and more on behalf of the state-owned banks and development finance institutions. As of end-June 2002, some 10.2 percent of the total stock of NPLs had been transferred to CIRC at market prices, for about 19.2 percent of their face value. CIRC has at most three years to pay for the transfer (less if the bank is privatized), partly in cash (through the liquidation of the collateralized assets), partly in five-year bonds (for the residual).

110. Capital market reforms started in 1994 and accelerated from 1997 under a capital market development program supported by the AsDB. The three regional stock exchanges were modernized in 1997, based on transparent regulations. An independent regulatory authority, the Securities and Exchange Commission of Pakistan (SECP) became operational in 1999. In the near future, the SBP’s supervision will be limited to banks, with the SECP being the sole regulator of the nonbank financial institutions (NBFIs). The Insurance Ordinance promulgated in 2000 has started to improve the regulatory and supervisory framework in this industry, and new minimum capital requirements have been imposed, effective end-December 2002, However, several industries in the insurance and pension sectors remain dominated by inefficient state-owned near-monopolies, in particular the State Life Insurance Corporation (SLIC) and the Employees’ Old-Age Benefit Institution (EOBI).

C. A Banking Sector Assessment

111. During recent years, the SBP has made strides toward the implementation of the Basel Core Principles for Effective Banking Supervision, and considers the financial system to be in compliance or largely in compliance with most of the 25 principles. The few remaining shortfalls include (a) that banks do not seem to have in place a comprehensive risk management process; and (b) that bank supervision is not implemented on a consolidation basis, which may affect the results for those banks having specialized subsidiaries such as Modarabas and leasing companies.

Financial Soundness Indicators (FSI)

112. Since 1997, SBP has developed a framework for assessing financial institutions’ soundness based on international best practices. The framework (inspired by the Fund’s CAMELS system) comprises six major areas: capital adequacy, asset quality, management soundness, earnings and profitability, liquidity, and sensitivity to market risks. The SBP has not yet compiled on an aggregate basis a capital adequacy ratio based on tier I capital nor indicators of sensitivity to market risks, such as the maturity of assets and liabilities, and the net open position in foreign exchange to capital. The indicators reviewed below are extracted from this framework and are close, although not identical, to the Fund’s proposed core set of financial indicators.⁴⁹

113. Overall, the banking system in Pakistan still suffers from considerable weaknesses (Table IV-1). Although progress has been registered in many areas, FSIs reflect adverse factors that have affected the banking system since 1998 (capital outflows and strict monetary policy to counter the exchange rate crisis) as well as genuine efforts to restructure the banking sector. Consolidated results are heavily influenced by the still predominant state-owned banks:

• The capital adequacy ratio (measured on tier I and tier II capital) is above the 8 percent floor average. However, four banks were still below this limit in 2001—two being state-owned banks and two private domestic banks.

• The quality of assets as reported has not improved markedly (Figure IV-1): the ratio of NPL total gross is still close to 20 percent of bank credit. However, the SBP considers that the increasingly strict criteria used for assessing NPLs and more realistic reporting after 1997 may mask a substantial underlying improvement.⁵⁰ Indeed, flows of new NPLs are now limited and 55 percent of the outstanding stock has already been provisioned. As of end-June 2002, two-thirds of the NPL stock was held by the nationalized banks. A sizable reduction in this stock is expected in the coming months with the transfer of nonperforming assets to CIRC that precedes the privatization of UBL and HBL.

Table IV-I.

Pakistan: Selected Financial Soundness Indicators: Banking Sector, 1990–2001 ^1/

(In percentage points; unless otherwise stated)

Source: SBP

^1/Based on aggregate accounts.

^2/Difference between average income on earning assets and average cost on interest–bearing liabilities

^3/Cash and bank balances, investment in government securities and money at call.

Table IV-I.

Pakistan: Selected Financial Soundness Indicators: Banking Sector, 1990–2001 ^1/

(In percentage points; unless otherwise stated)

		1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000	2001
Capital adequacy
	Regulatory capital to risk-weighted assets average	n.a	n.a	n.a	n.a	n.a	n.a	n.a	6.0	12.5	12.2	11.4	11.4
	Number of banks below 8 percent	n.a	n.a	n.a	n.a	n.a	n.a	n.a	5	0	1	3	4
Asset quality
	NPL to total gross loans	17.6	18.9	17.4	20.0	21.7	19.3	19.2	20.1	19.5	22.0	19.5	19.6
Management Soundness
	Interest rate spread	4.0	4.4	4.4	4.8	4.6	4.9	3.8	5.2	5.3	5.6	6.4	5.9
Earnings and profitability
	Return on Assets	0.4	0.6	0.8	0.8	0.4	0.3	0.1	−1.2	0.5	−0.3	0.0	0.0
	Return on Equity	10.5	15.8	18.2	17.6	9.1	7.2	2.1	−38.8	9.1	−6.3	−0.3	−0.1
Liquidity
	Liquid assets to total assets	32.6	41.6	46.0	44.4	47.0	43.7	43.4	41.4	41.3	38.7	30.5	34.9

Source: SBP

^1/Based on aggregate accounts.

^2/Difference between average income on earning assets and average cost on interest–bearing liabilities

^3/Cash and bank balances, investment in government securities and money at call.

View Table

Table IV-I.

Pakistan: Selected Financial Soundness Indicators: Banking Sector, 1990–2001 ^1/

(In percentage points; unless otherwise stated)

		1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000	2001
Capital adequacy
	Regulatory capital to risk-weighted assets average	n.a	n.a	n.a	n.a	n.a	n.a	n.a	6.0	12.5	12.2	11.4	11.4
	Number of banks below 8 percent	n.a	n.a	n.a	n.a	n.a	n.a	n.a	5	0	1	3	4
Asset quality
	NPL to total gross loans	17.6	18.9	17.4	20.0	21.7	19.3	19.2	20.1	19.5	22.0	19.5	19.6
Management Soundness
	Interest rate spread	4.0	4.4	4.4	4.8	4.6	4.9	3.8	5.2	5.3	5.6	6.4	5.9
Earnings and profitability
	Return on Assets	0.4	0.6	0.8	0.8	0.4	0.3	0.1	−1.2	0.5	−0.3	0.0	0.0
	Return on Equity	10.5	15.8	18.2	17.6	9.1	7.2	2.1	−38.8	9.1	−6.3	−0.3	−0.1
Liquidity
	Liquid assets to total assets	32.6	41.6	46.0	44.4	47.0	43.7	43.4	41.4	41.3	38.7	30.5	34.9

Source: SBP

^1/Based on aggregate accounts.

^2/Difference between average income on earning assets and average cost on interest–bearing liabilities

^3/Cash and bank balances, investment in government securities and money at call.

View Full Size

Pakistan: Nonperforming and Default Loans as a Share of Total Gross Loans, 1990–2001

(In percent)

Citation: IMF Staff Country Reports 2002, 247; 10.5089/9781451830569.002.A004

Source: State Bank of Pakistan.

• Download Figure
• Download figure as PowerPoint slide

• Management soundness indicators do not yet reflect much impact from the restructuring and privatization process. The spread between the average interest rate on interest-bearing assets and liabilities was on an increasing trend till recently.

• The profitability of banks, as measured by the return on equity or the return on assets appears quite low in Pakistan, having declined since 1993. Large fluctuations in the ratios since 1996 seem to reflect mostly the entry of large provisions on NPLs, as well as the impact of the freeze on foreign currency deposits which deprived banks from a very profitable activity.

• Banks remain largely liquid, although on average less so than during the past decade. The gradual reduction in SLRs imposed on banks by the SBP, the development of banks’ refinancing through the SBP’s discount window and repo operations have significantly reduced banks’ liquidity needs.

• The SBP has not yet developed clear indicators of sensitivity to market risks. Exchange rate risks have probably been limited during the last decade because of various foreign exchange controls, some of which have been eliminated recently. The banks’ foreign exchange open position is limited to 10 percent of capital. Up to 1997, banks have been protected by the direct impact of large depreciations of the Pakistani rupee by forward cover provided by the SBP at a subsidized rate. Some banks may bear a significant interest rate risk on long-term government bonds.

D. Banks’ Limited Contribution to Financial Deepening in the 1990s

114. Reforms have not yet visibly improved the capacity of the banking sector to mobilize savings, finance the private sector, and provide services at low costs.

• Financial savings as a whole increased from 45 percent of GDP in 1992/93 to more than 60 percent in 2001/02 (Figure IV-2). The increase took place mostly in three discrete steps, in 1993/94, 1997/98, and very recently. Improved confidence in the Pakistani rupee over the last year and the easing of the drives toward tax compliance and accountability initiated in 1999 encouraged the repatriation of residents’ savings from abroad. However, most of the increase in financial savings was absorbed by the more attractive National Savings Schemes (NSS), which grew from 11.3 percent to 23.4 percent of GDP during the same period. Savings accumulated in banks over the same period remained mostly stable as a percentage of GDP. The dollarization of banks’ deposits fluctuated largely, however. FCDs amounted to almost 30 percent of residents deposits in 1998, on account of increasing pressures on the domestic currency and incentives provided by the SBP.

View Full Size

Pakistan: Stock of Financial Savings as a Share of GDP

Citation: IMF Staff Country Reports 2002, 247; 10.5089/9781451830569.002.A004

Source: State Bank of Pakistan.

• Download Figure
• Download figure as PowerPoint slide

• Banks’ contribution to the development of the private sector appears also limited (Figure IV-3). In proportion to GDP, credit to the private sector remains in the same 20–25 percent range, with a declining trend since 1998/99. The elimination of credit ceilings and constraints imposed on banks’ interest rates did not boost banks’ supply of credit, probably because the weight of NPLs had to be absorbed in the balance sheets, and better supervision forced banks to eliminate or reduce business with their most risky customers. The government and the public enterprises would not appear to have crowded out the private sector, since banks’ credit to the public sector has declined over the last ten years. The absence of credit expansion to the private sector seems rather due to demand factors such as a lower growth overall and the decline of the private investment ratio (by about 2 percent of GDP over the last decade). Higher real interest rates offered by banks had probably also a negative contribution.

View Full Size

Pakistan: Banks’ Credit by Main Sector

Citation: IMF Staff Country Reports 2002, 247; 10.5089/9781451830569.002.A004

Source: State Bank of Pakistan.

• Download Figure
• Download figure as PowerPoint slide

• Financial liberalization has considerably increased banks’ spread between average lending rates to the private sector and deposit rates, from about 2 percent in the early 1990s, to more than 8 percent (Figure IV-4). Since then, this spread has remained high, reflecting limited competition in the banking sector, and the increasing cost of provisions on account of NPLs.

View Full Size

Pakistan: Spread between Average Lending and Deposit Rates

Citation: IMF Staff Country Reports 2002, 247; 10.5089/9781451830569.002.A004

Source: State Bank of Pakistan.

• Download Figure
• Download figure as PowerPoint slide

E. Next Steps Toward Reform

115. A broad-based debate has been organized by the SBP to discuss the lessons from its Financial Sector Assessment and the future reforms. In particular, a conference in June 2002 in Karachi gathered private bankers, independent observers, government officials and donors’ representatives to discuss the SBP’s strategy for the financial sector. The SBP’s strategy and participants’ reactions are analyzed below. This section concludes with additional suggestions to reform the NSS.⁵¹

116. The SBP’s strategy identifies four directions for further reforms:

• The reshaping of the financial sector toward a three-tier structure: a first tier consisting of large universal banks; a second tier of specialized banks that would cater to specific sectoral needs or market imperfections; and a third tier of NBFIs. The implementation of this strategy would require a further restructuring and privatization of existing state-owned Development Finance Institutions (DFIs), as well as incentives for mergers and acquisitions of small and under-capitalized private banks and other financial institutions.

• An enhanced judicial, regulatory, and supervisory framework, for both banks (under the SBP) and nonbank financial institutions (under the SECP). Areas where improvements would be the most urgently needed include corporate governance in the financial sector; the quality of financial information disclosure, especially for NBFIs; improved prudential and risk management standards for NBFIs; and strengthened legal structure to facilitate foreclosure of collaterals and accelerating the recovery of bad loans. Banks and NBFIs’supervisors need also to improve their skills.

• Improved monetary policy and interest rate market mechanisms. This will imply among other steps, more active money markets and secondary markets for government securities, as well as a further rationalization of the NSS instruments that would reduce the distortion of savings allocation as well as the interest costs of domestic debt.

• Improved supply of financial instruments and services, including exportable ones, and further development of capital markets.

117. While most participants (and the staff) had little disagreement with the SBP’s directions for further reforms, the following suggestions emerged from the discussion with the financial sector stakeholders:⁵²

• The tax system and the fragmentation of financial markets. To promote a sound and large competition in the financial sector, income tax rates and withholding provisions need to be further harmonized. The tax system continues to discriminate against banks in the tax treatment of financial instruments. While income tax rates on banks have been reduced in 2002 from 58 percent to 50 percent, participants strongly supported the further steps that the authorities are considering to bring them in line with the standard corporate tax rate of 35 percent, which applies for other financial institutions such as leasing companies and investment banks. Banks’ specific provisions against NPLs are not tax deductible. Withholding taxes of 20 percent on income generated by short-term treasury bills maintain an artificial bias in favor of long-term instruments. The financial component of life insurance products are subject to the General Sales Tax (GST). The Provincial and Federal Governments levy stamp duties on securities at different rates. Voluntary contributions to provident funds and pension schemes should be exempted from income tax, while annuities should be taxed. There are also tax obstacles to the mergers between banks (and other entities) that should be removed. On a related issue, private banks should be allowed to compete for operations with public enterprises, currently they are still being forced to hold accounts with nationalized banks only.

• The role of market mechanisms and sound competition in reshaping the financial sector. While market imperfections may sometimes justify government involvement, careful examination is needed before embarking in new interventionist policies. There are examples of sound practices (see the following section on microfinance) but also less convincing approaches (the restructuring of the ADBP to continue to direct credit to the agricultural sector). Some participants questioned the necessity to force mergers of small banks that are appropriately capitalized (there seems to be a positive appreciation amongst the banking profession for some small banks, which, because of a thorough knowledge of their customers, provide valuable specialized services to some niche sectors of the economy).

• More efficient money and securities markets to reduce banks’ intermediation costs. Day-to-day interbank rates still fluctuate largely between SBP’s interventions, generally weekly. This problem and banks’ easy recourse to the SBP’s repo facility do not encourage improvements in liquidity management practices. To limit the day-to-day volatility of short-term rates, and in particular to set a floor on overnight rates on the interbank money market, a more transparent and active monetary policy is needed. More frequent open market operations or a standing facility could signal more effectively the SBP’s monetary stance. Greater predictability of short-term interest rates would permit banks to better manage their liquidity management position and would encourage them to reduce lending spreads. In this regard, the role of the SLR as a second line of defense for liquidity purposes does not seem to be consistent with the SLR-eligibility of long-term government instruments and the ineligibility of prime quality commercial paper. Specific measures are also needed to develop the secondary market for the new medium to long-term government bonds, the Pakistan Investment Bonds (PIBs) and for other fixed-income securities.

118. The authorities need to undertake a fundamental reform of the NSS and of the government department that issues the NSS, the Central Directorate of National Savings (CDNS). With time, the CDNS has been assigned, explicitly or implicitly, conflicting mandates: (a) design attractive savings schemes to finance the government deficit, at a time when inflation and balance of payments concerns imposed constraints on the domestic bank and foreign financing of the budget; (b) protect small savings holders from inflation and provide real income to vulnerable social groups; and (c) manage a substantial part of the government’s domestic debt and minimize its cost. The performance of public entities need to be assessed against clear mandates. The only mandate which is clearly relevant for the CDNS is the management of public debt. The design of attractive savings schemes for the public would be better run by private financial institutions. NSS instruments risk to divert savings from more productive private usage, using high interest rates to attract savers at the expense of present and future taxpayers. The NSS would also seem ill-suited to serve income distribution purposes, such as supporting vulnerable social groups.

119. Some progress has been achieved to rationalize the NSS. In 2001, the authorities started to auction the PIBs, and to use them as benchmarks for NSS rates, leading to a large reduction in the cost of these schemes over the recent years. Additional nonmarket distortions have been removed through the increase in premiums for premature cashing of long-term instruments. However, the premium of NSS instruments over corresponding PIBs of the same maturity is still high enough to prevent banks from collecting significant amounts of household savings on term-deposits. In addition, the government has little choice other than to borrow through this costly on-tap instrument.

120. In the short run, NSS rates should be further reduced to put them on an equal footing with similar government instruments. In addition, the administrative costs of maintaining a retail trade system of savings collection could be passed on to NSS holders. In the medium-term, the government would have a better control of its debt management by eliminating ail on-tap instruments and auctioning only a few instruments such as the treasury bills or the PIBs through primary dealers. The CDNS could progressively replace NSS instruments (starting with the most costly Defense Savings Certificates) with mutual fund instruments based on both government and private securities. Ultimately, the CDNS itself could be transformed into a private mutual fund, subject to strict prudential regulations. The authorities are considering some of these suggestions, in conjunction with a second set of financial market reforms that could be supported by the Asian Development Bank (AsDB).

121. There are no immediate plans to develop a deposit insurance scheme in Pakistan. Although mindful of the additional confidence that such a scheme would bring to the banking sector, the authorities are mostly concerned with the moral hazard issues attached to it.

F. Recent Steps Toward the Development of Financial Institutions for the Poor

122. In Pakistan, the poor, especially in the rural areas, are underserved by existing financial institutions or have to rely on expensive informal financing. Traditional credit to the agricultural sector, including through state-owned institutions like ADBP, is barely available to poor farmers, mainly for lack of usable collaterals. Banks do not collect the savings of the poor either, especially in the rural area because of high infrastructure costs: the majority of branch closures decided recently by the major nationalized commercial banks took place in the rural sector. Alternative sources of financing are costly; annualized rates offered by informal money lenders are generally in the range of 80–120 percent, but can reach 300 percent for short-term loans. Access to financial services is a key component of poverty according to poverty assessments conducted in Pakistan and elsewhere. The potential demand for micro credit is huge: according to the AsDB, the 6.3 million poor households potentially represent about US$2 billion of solvable credit demand.

123. The development of microfinance has been slow in Pakistan until recently. Traditional banks have offered limited service. The ADBP has special programs for the poor, with limited outreach (only 2 percent of landless farmers have ever received a loan from the ADBP). Most of these loans appear to be nonrecoverable. Some commercial banks (including the state-owned Bank of Khyber) have developed nonprofitable loan schemes in poor urban areas, mainly as a social obligation.

124. Until recently, the authorities’ efforts focused on NGOs to provide microfinance to the poor. About 100 NGOs provide microfinance services to 61,200 borrowers (for a total amount of PRs 1 billion) and 417,000 depositors. Most of the NGOs that supply microfinance are government-sponsored Rural Support Programs which run grant-supported schemes. In 1997, the authorities launched a new World Bank-supported program, the Pakistan Poverty Alleviation Fund (PPAF) with an endowment of PRs 500 million to extend micro credit to 200,000 households over 4 years through 20 NGOs. Results have been disappointing in the absence of real retailing capacity from NGOs.

125. In 2000, realizing that the current microfinance schemes would not reach the poor before many years, the authorities decided to establish a new self-sustained scheme, the Khushhali Bank. The objective is to offer financial services for 10 percent of the poor over six years, of which at least 40 percent would be women. This new entity is used as an experimental tool, to test a regulatory framework that could attract new actors in the field on a sound basis. The scheme will help build a social network of community organizations that would form the basis on which lending activity could develop, supported by peer-pressure mechanisms. More affordable compared with what informal channels offer, interest rates cover the rather substantial intermediation costs of such microfinance schemes (about 10 percent), while building reserves for two funds that protect against default risks and insure small depositors. The infrastructure is expected to be limited to a minimum and client-oriented: mobile units visit villages to collect deposits and loan applications. The new institution is owned by private shareholders, mostly banks, and has an independent supervisory board.

126. A new legal framework for microfinance institutions (MFIs) was promulgated in 2001, to provide a legal foundation for the Khushhali Bank. Recently, a general ordinance completed the framework with regulations for new entrants. This ordinance sets up different minimum capital requirements, according to the intended coverage: respectively PRs 500 million, PRs 250 million, and PRs 100 million for national, regional, and district coverage, against PRs 1 billion for a traditional bank. It also provides a ceiling for the maximum amount of loans that a MFI can grant (US$1,500).⁵³ A second MFI was established in 2002, the First Microfinance Bank, sponsored by the Agha Khan foundation and the IFC.

127. Microfinance had a late but promising start in Pakistan. After two years of operation, the Khushhali Bank has very encouraging results: As of July 31, 2002, about PRs 600 million have been lent to 54,000 clients, of which 43 percent were women. So far the default rate is no more than 0.4 percent. Although the authorities have used a top-down approach to set up the bank, the promising start of this institution seems to derive from a carefully designed concept and business plan and the incorporation of lessons drawn from experience accumulated in other countries. Overhead costs have been limited to the minimum; and staff has been hired from the private sector, so that conditions have been met for the development of a for-profit sustainable institution. The initial regulatory framework had been carefully designed to match the needs of the nascent institution. This framework has been recently adjusted and will continue to evolve over time to take into account the lessons from the ongoing pilot and adapt to the needs of the expected new MFIs.