World Development Vol. 32, No. 9, pp. 1487–1507, 2004 Ó 2004 Elsevier Ltd. All rights reserved Printed in Great Britain www.elsevier.com/locate/worlddev 0305-750X/$ - see front matter doi:10.1016/j.worlddev.2004.06.001 Imperfect Substitutes: The Local Political Economy of Informal Finance and Microfinance in Rural China and India

KELLEE S. TSAI * Johns Hopkins University, Baltimore, MD, USA Available online Summary. — Banking authorities in both China and India have attempted to limit most forms of informal finance by regulating them, banning them, and allowing certain types of microfinance institutions. The latter policy aims to increase the availability of credit to low-income entrepreneurs and eliminate their reliance on usurious financing. Nonetheless, the intended clients of microfinance continue to draw on informal finance in both rural China and India. This article argues that the persistence of informal finance may be traced to four complementary reasons––the limited supply of formal credit, limits in state capacity to implement its policies, the political and economic segmentation of local markets, and the institutional weaknesses of many microfinance programs. Ó 2004 Elsevier Ltd. All rights reserved.

Key words — Asia, China, India, informal finance, microfinance, rural finance

‘‘[O]fficial reports of the moneylender’s impending de- well-located rural households that have the mise are much exaggerated.’’–– Clive Bell on India option of keeping their savings in official (1990). financial institutions may lack access to formal ‘‘The fact that these private or underground credit sector credit and rely instead on a wide range of money houses exist and sometimes thrive in the coun- informal, curb market mechanisms. tryside even today has revealed that farmers need them.’’––People’s Daily on China (November 29, It is in this context that governments 2002). throughout the developing world have regarded informal finance as a negative reflection of deficiencies in the formal financial system. In both China and India, the traditional image of 1. INTRODUCTION the usurious moneylender adds an additional pejorative dimension to the official depiction of Developmental economists have long noted the complexity of providing effective rural credit delivery in large, agrarian countries such * Earlier versions of this paper were presented at the as India and China. 1 Establishing and main- Workshop on Local Governance in India and China: taining a network of rural financial institutions Rural Development and Social Change, Kolkata, Jan- is expensive, and managing their operations uary 6–8, 2003, and the Duke University Comparative is difficult in the absence of proper training, Politics Workshop, February 24, 2003. The paper bene- monitoring, and incentive structures. The fited greatly from the constructive input of the workshop operational challenges of rural financial inter- participants. I am also grateful for the insights of mediation are compounded by state develop- Richard Baum, Anirudh Krishna, Laura Locker, Eddie ment strategies that promote industrialization Malesky, Mark Selden, Suman Sureshbabu, Sarah and urbanization at the expense of agricultural Tsien, Fei-ling Wang, Steven Wilkinson, David Zweig, production. At the macrolevel, the notorious and four anonymous reviewers. They are, of course, scissors gap between agriculture and industry absolved from the article’s inadequacies. Financial sup- redistributes savings from rural to urban areas, port from the Ford Foundation Public Policy Grant thereby limiting the relative supply of rural Competition is gratefully acknowledged. Final revision credit. At the microlevel, this means that even accepted: 10 May 2004. 1487 1488 WORLD DEVELOPMENT informal finance: when the poor lack access to case studies from India and China, respectively, conventional sources of credit, they are to illustrate how the combination of credit exploited by loan sharks and other illegal curb supply, local political economic conditions, and market operators. Following this logic, the institutional characteristics of financial inter- prescription thus requires increasing state mediaries mediates the dynamics of rural efforts to eliminate informal finance, while finance. enhancing the availability of state-sanctioned financial intermediaries, especially microfinance programs devoted to poverty alleviation. Even 2. FINANCING RURAL DEVELOPMENT with these policy measures, however, small IN CHINA AND INDIA business owners and farmers continue to rely primarily on curb market finance in both China To understand the formal institutional con- and India. Moreover, in some cases, the scale of text against which curb market activities have informal finance actually increases in commu- flourished, this section highlights major chan- nities that have been targeted for a greater ges in the basic structure of rural finance. Both supply of official credit. This raises the question countries have established credit cooperatives, of why official attempts at limiting informal commercial banks, and poverty alleviation mi- finance and expanding the accessibility of for- crofinance programs in rural areas, but these mal finance may have such unintended conse- formal sector institutions have not displaced quences. One basic reason for the persistence of informal and semi-formal sources of credit. 2 informal finance is that the supply of formal finance is limited and insufficient to meet the (a) Formal financial sector demand for credit. A second explanation is that official state policies are not being implemented After India’s independence in 1947 and the properly. In addition to these economic and establishment of the People’s Republic of state-centric explanations, this article argues China in 1949, the 1950s represented a rela- that informal finance and formal finance are tively optimistic and ambitious phase for both imperfect substitutes for two additional, com- countries in establishing a national system for plementary reasons: First, because credit mar- agricultural finance. Both newly inaugurated kets are segmented by local political and social regimes shared the developmental goals of dynamics; and second, because government- promoting growth without exploitation, and sanctioned microfinance programs are often creating grassroots-level savings and credit structured in a manner that fails to serve its institutions to serve farmers. intended clientele. This suggests that informal Although India inherited a basic network of finance is not simply a manifestation of weak- credit cooperatives from the colonial era, the nesses in the formal financial system, but also, a Reserve Bank of India’s (RBI) first decennial product of local political, institutional, and All-India Debt and Investment Survey in 1951 market interactions. The analytical value in found that 93% of rural households relied on recognizing these local interactions lies in their informal finance (Bouman et al., 1989, pp. 12– ability to explain why developmental outcomes 14). This finding inspired a strong political deviate from state intentions. commitment to establishing formal sector The article proceeds as follows: Section 2 alternatives to the curb, which was popularly reviews the key expressions of formal and semi- viewed as being exploitative and even ‘‘evil’’ formal finance in China and India, and shows (RBI, 1954). Hence, throughout the 1950s and how the countries’ strategies in rural financial 1960s, the government actively promoted the intermediation compare with one another. expansion of cooperatives ‘‘to provide a posi- Both have relied on directed credit and tive institutional alternative to the moneylender encouraged the growth of microfinance pro- himself, something which will compete with grams, albeit to differing degrees. Section 3 him, remove him from the forefront, and put outlines the main expressions of informal him in his place (RBI, 1954, pp. 481–482)––or finance in China and India and discusses the more generally, to enhance the availability of extent to which they have been subject to state agricultural credit and alleviate rural poverty. regulation. Section 4 delineates four comple- In the mid-1970s, India’s rural financial system mentary explanations for why state efforts to went through another expansionary stage with substitute informal finance with microfinance the establishment of regional rural banks have not been successful, and presents two local (RRBs) at the district level, farmers’ service INFORMAL FINANCE AND MICROFINANCE 1489 societies at the village level, and further growth serving rural enterprises and households. Since of nonbanking finance companies. 3 Even then, central banking authorities have deliber- though the number of bank branches tripled ated over how to improve their performance during 1969–79, the government considered (Watson, 2003), and injected approximately rural access to be too low at 37,000 people per US$4 billion in recapitalization funds into the rural bank branch; therefore, in 1980 another RCC system because RCCs are technically seven commercial banks were nationalized to insolvent. As of mid-2003, RCCs accounted for extend their outreach in rural areas (AFC, 11.5% of total savings and 10.8% of loans 1988, in Nagarajan & Meyer, 2000, p. 172). In extended by formal financial institutions, and a quantitative terms, progress has been made on pilot reform scheme for decentralizing their this latter objective: according to the RBI, by management was underway in eight provinces 1998 India had a total of 64,547 RRB branches, (PD, November 30, 2003). which was equivalent to 17,000–21,000 rural citizens per bank branch. 4 But, the RRBs have (b) The rise of microfinance proven to be financially unsustainable and inefficient in loan delivery (Bhatt & Thorat, Given the inability of most formal sector 2001). banking institutions to reach rural populations Shortly after the founding of the People’s and the popularity of informal sector alterna- Republic of China, the Chinese communists tives, microfinance programs have emerged as a ordered the closure of all forms of private potential solution for bridging the gap between finance and banned popular forms of curb the supply and demand for rural finance. In market financing, including pawnbrokering both India and China, microfinance has taken and ‘‘loan sharking’’ (Hsiao, 1971). During the the form of subsidized loans in government- 1950s, China also set up a network of rural supported poverty alleviation (PA) programs, credit cooperatives (RCCs), but unlike the and various donor and nongovernmental cooperatives in India, China’s original RCCs organization (NGO)-lead endeavors. While the acted mainly as fiscal institutions that funneled actual expressions and overall scale of micro- credit between the state and the people’s com- finance differs in the two countries, the relative munes rather than serving as commercial effectiveness of these two main forms of mi- credit-granting institutions. It was not until the crofinance is similar. Specifically, subsidized commencement of market-oriented reforms in microloans in government-supported PA pro- the late 1970s that RCCs started to function grams tend to have low repayment rates and more as grassroots banking institutions that tend not to reach the intended clientele; and served rural households and collective enter- microfinance programs run by NGOs are more prises, and the Agricultural Bank of China effective in reaching poor clients when loans are (ABC) was re-established to handle larger scale structured in a financially sustainable manner commercial banking activities. 5 Meanwhile, in and use lending methodologies that are adapted the early 1980s, the Ministry of Agriculture to the particular economic needs of the inten- established a network of Rural Cooperative ded clients. Foundations (RCFs) to serve farmers, but the People’s Bank of China never considered them (i) Directed subsidized credit in public poverty formal ‘‘financial institutions’’ and succeeded in alleviation programs shutting them down at the end of the 1990s. Extending subsidized loans to low-income Indeed, throughout the reform era central areas and households has traditionally been the authorities have repeatedly waged national first, and perhaps least effective strategy that political campaigns to crackdown on the curb. governments use in their rural development In July 1998, China’s State Council even issued strategies, and India and China are no excep- formal ‘‘Provisions on the Cancellation of tions (Adams, Graham, & von Pischke, 1984; Illegal Financial Institutions and Activities,’’ cf. Morduch, 2000). The Indian Integrated which reiterated that illicit financial institutions Rural Development Programme (IRDP) was should be banned (Xinhua, July 22, 1998, cited established in 1978 with the mandate of in Tsai, 2002a, p. 1). extending microloans through the banking The elimination of RCFs left about 44,000 system to impoverished households and now RCCs at the township level (with about 280,000 regards itself as the ‘‘world’s largest program village branches) as the only formally approved for providing microloans to the poor (Sinha, nonbanking financial institution devoted to 2000, p. 66).’’ In its first two decades, the IRDP 1490 WORLD DEVELOPMENT extended Rs. 250 billion (US$12.3 billion) households would be directly targeted for sub- worth of subsidized loans to approximately 55 sidized poverty loans. In quite a change from million families who have an annual income of the previous mode of distributing subsidized less than Rs. 11,000 (US$305). 6 Given that 70 credit to local enterprises, in 1996 many of the million families live below the poverty line in counties adopted the Grameen Bank model of India, it is apparent that the IRDP has had group lending whereby groups of five borrow- significant outreach. In addition to the loans, ers would mutually guarantee the repayment of IRDP borrowers also receive a cash subsidy at their respective microloans in multiple instal- the time of loan disbursal equivalent to 25–50% ments (Bornstein, 1997; Holcombe, 1995; of the project cost (Nagarajan & Meyer, 2000, Khandker, Khalily, & Khan, 1995). These p. 170). The program has certainly disbursed a loans ranged from 1,000 to 2,000 RMB high volume of loans, but funds have been (US$120–240) and they continued to be subsi- misused via the subsidy component such that dized at the official PA lending rate of 2.88%. cash is diverted to local elites rather than the Once the decision was made to disburse PA intended borrowers; as a result, the program loans directly to households in officially desig- has had a repayment rate of only 25–33% nated impoverished counties, they were dis- (Sinha, 2000, p. 66). Meanwhile, the RRBs and bursed rapidly, almost quota style: primary agricultural credit societies have not performed any better. The RRBs have been By August 1998, official microcredit schemes were saddled with soft loans to priority sectors, while operating in more than 600 counties in 22 provinces, primary cooperatives have served mainly as with the largest programs (in Shaanxi and Yunnan) tools of political patronage. 7 Due to the non- reaching over 500,000 households...In 1999, with be- tween 30 and 40 million people still classified as poor, commercial orientation of these programs, the central government’s budget for the 8–7 Plan basically all of the formal sector institutions called for expenditures of Y24.8 billion ($3 billion), involved in microfinance have depended on of which Y15.3 billion ($1.84 billion, or 62%) was refinancing and recapitalization by apex insti- for loans funds (Conroy, 2000, p. 36). tutions on a regular basis (Nagarajan & Meyer, 2000, pp. 177–179). By 2000, the government had disbursed State-subsidized microfinance in China has US$775 million worth of subsidized microloans had a shorter history than in India, mainly (Tsien, 2001), and by 2002 nearly US$3.7 because China started poverty lending about billion (or half) of the central governments one decade later than India. To be sure, both poverty-relief funds were going toward pov- central and local governments in China have erty-relief loans (Xinhua, March 2, 2002). As in directed subsidized credit to particular sectors the earlier model of poverty lending, however, or industries, but that type of ‘‘policy lending’’ repayment rates in these government programs has not occurred in the name of microfinance have been low, i.e., less than 60%. Even though or poverty alleviation. 8 In 1986, a subsidized the Agricultural Bank of China (a state com- lending scheme for poverty relief was intro- mercial bank) took over the poverty lending duced, which targeted collective enterprises at program from the Agricultural Development the township and village level rather than Bank (a policy bank) in 1998, the People’s individual households (Rozelle, Park, Ren, & Bank of China (PBC) has not been involved in Bezinger, 1998). While official interest rates on monitoring the microcredit component of the loans ranged between 8% and 10%, the poverty Agricultural Bank of China’s operations, and alleviation loans charged only 2.88% annual the loans are treated more as social grants interest. As is the case with most subsidized rather than as commercial loans. In other credit schemes, the loans were distributed to words, the microcredit component of PA politically important enterprises and higher- lending has been treated as one-time fixes income households, and the repayment rates rather than exhibiting a commitment to sus- were about 50% (Park, 1999). tainable models of microfinance (Cheng, 2003). Providing subsidized loans directly to Meanwhile, the PBC has been encouraging households did not start until a few years into RCCs to extend microloans to rural house- China’s National 8–7 Poverty Alleviation Plan, holds. As of 2002, the PBC reported that RCCs introduced in 1993. As part of the strategy to had extended a total of 78.9 billion RMB raise 80 million people out of poverty in seven (US$9.54 billion) worth of microloans and that years (i.e., during 1994–2000), the central gov- 25% of all rural households in the country had ernment identified 592 poor counties where received such loans (CIIC, November 5, 2002). INFORMAL FINANCE AND MICROFINANCE 1491

Although RCCs report higher repayment rates ment of one million bank-linked SHGs by than the PA programs, as of year-end 2003, 2008. (Bansal, 2003, p. 24). their ratio of nonperforming loans was still Aside from participating in the SHG–bank quite high at nearly 30% (SIC, January 14, linkage model, over 500 NGOs serve as finan- 2004). cial intermediaries themselves by brokering funds between banks and low-income borrow- (ii) NGO and donor-managed microfinance ers. There are also a handful of cooperatives institutions such as SEWA Bank, the Indian Cooperative The involvement of NGOs in running mi- Network for Women, Tamil Nadu, and coop- crofinance institutions (MFIs) varies signifi- erative credit societies associated with the cantly in India versus China. This is due in part Cooperative Development Foundation that are to differences in the policy environment for involved in microfinance. Finally, about 10 both NGOs and nonbanking financial institu- organizations may be considered Grameen re- tions. While the government of India has pro- plicators. The largest ones are SHARE, Activ- moted the growth of self-governing NGOs and ists for Social Alternatives Trust, and Rural encouraged domestic development finance Development Organization, Manipur (Sinha, institutions to collaborate with them, China’s 2000, p. 70). NGOs are sponsored by a particular govern- Overall, MFIs in India have not been subject ment unit (making them government-organized to stringent regulations, especially those that NGOs rather than pure NGOs) or established are not registered as cooperatives or nonbank- by international donors. To date, India’s ing finance companies. Given the developmen- NGOs have had more extensive reach in mi- tal contribution of MFIs, the RBI has not crofinance than their counterparts in China, enforced Section 45S of the RBI Act, which but in both countries, few MFIs are financially prohibits savings mobilization from the public sustainable while the market for MFIs remains without RBI permission. Furthermore, finan- vast. cial liberalization since the 1990–91 economic In India, microfinance NGOs have generally crisis has loosened interest rate controls on taken one of the following three forms: self- microcredit, which offers MFIs in India the help group (SHG) programs that have linkages space to structure their loans in a financially with banks; cooperatives; or Grameen replica- self-sustainable manner. Whether this occurs, tors (EDA Rural Systems, 1996). Organized by however, depends in large part on changing NGOs, SHGs consist of 10–12 people with popular perceptions that low-income borrowers similar socioeconomic and demographic char- cannot afford commercially viable interest acteristics (e.g., low-income women in rural rates. areas). As of 2002, there were one million In contrast to the relative ease with which SHGs with 17 million members (Ashe, 2002, NGOs may register themselves and act as MFIs cited in Wilson, 2002, p. 221). The purpose of in India, China’s policy environment is much the SHGs is to help the members save small more restrictive. All NGOs in China must have amounts of money on a regular basis, to create an official government unit sponsor their an internal insurance fund for members to draw application to register as ‘‘social organizations’’ on in times of emergencies, to empower the with the Civil Affairs Bureau (Saich, 2000). As members through collective decision-making, such, China does not have purely nongovern- and to extend uncollateralized loans to group mental organizations engaged in microfinance members (Hannig & Katimbo-Mugwanya, even though they may be functionally equiva- 1999, p. 7). Since 1992, the National Bank for lent to NGOs. The introduction of the Gram- Agriculture and Rural Development (NA- een model of microfinance provides a good BARD) has experimented with creating link- example of the close relationship between ages between SHGs and banks, such that banks government entities and NGOs in China. lend through NGOs or directly to SHGs. As of The replication of the Grameen model in March 2003, over 444 banks were participating China first came about through the individual in microfinance linkages with 717,360 SHGs; in initiative of researchers at the Rural Develop- total, the SHG–bank linkage program had ment Institute of the Chinese Academy of served an estimated 7.8 million low-income Social Sciences (CASS) and international households (NABARD, 2002, 2003). 9 Ulti- donors; but to date, the most successful mately, NABARD hopes to reach one-third of Grameen replications are managed from an India’s rural population through the establish- office housed at CASS. With funding from 1492 WORLD DEVELOPMENT

Grameen Trust, the Ford Foundation, and the 3. THE INFORMAL AND SEMI-FORMAL Canada Fund, in 1994 a small group of CASS FINANCIAL SECTOR researchers led by Professor Du Xiaoshan established the Funding the Poor Cooperative As suggested already, despite the substantial (FPC) in Yixian, Hebei (Tsai, 2002a, pp. 200– expansion of rural financial institutions in both 202). To implement the project they collabo- countries over the last several decades, informal rated with the Yixian county-level Poverty finance still represents a major source of credit Assistance Bureau and the Civil Affairs Office. for farmers and petty traders. In China, a study As of March 2003, there were three FPCs in by IFAD estimates that farmers obtain four Yixian, Yucheng (Henan), and Nanzhao (He- times more credit from the curb market than nan) counties, respectively, and together, the from formal financial institutions (IFAD, 2001, three FPCs had served a total of 15,244 bor- p. C11), and another study of small business rowers (Du, 2003). With repayment rates owners found that the curb accounted for up to ranging from 95% to 99%, the FPCs are con- three-quarters of private sector financing dur- sidered the best examples of Grameen-style ing the first two decades of reform (Tsai, 2002a, microfinance in China. A central part of their pp. 36–37). In India, the 1992 AIDIS survey success has been structuring the loans in a revealed that nearly 40% of rural households manner that covers their operational costs, i.e., continue to rely on informal finance––or more at 16% effective interest per annum. 10 Scaling technically, ‘‘noninstitutional credit agencies,’’ up to extend their reach and experimenting which include agricultural moneylenders, pro- with nonGrameen lending methodologies is fessional moneylenders, traders, relatives and their next challenge. 11 friends, and others. 14 Table 1 outlines the Besides the FPC Grameen replications, primary forms of informal and semi-formal international donors have initiated over 200 finance in both countries and notes the extent microfinance programs throughout central and to which they are sanctioned or prohibited. In western China (Cheng, 2003, p. 123). The both countries, private transactions involving donors have all implemented their projects with high interest rates are in violation of banking different local governmental partners. For regulations, as are organizations that mobilize example, the AusAid project in Haidong, savings without registering with the appropri- Qinghai that started in 1996 collaborates with ate authorities. 15 Beyond those two restric- the Agricultural Bank of China and the Qing- tions, however, the legal marginalization of hai Commission of Foreign Trade and Eco- curb market activity has not been consistently nomic Cooperation; the Heifer Project defined or enforced. In practice, curb market International has been collaborating with the actors in both China and India have proven to Sichuan Animal Husbandry Bureau since 1985; be adaptable despite multiple rounds of disci- and since 1995, the International Crane Foun- plinary action by financial regulators. dation has implemented a Trickle-Up Program in Guizhou with the cooperation of the Guiz- (a) Grey areas in China’s curb market financing hou provincial Environmental Protection Bureau. 12 With few exceptions, the donor-ini- In China, the extremes of legal versus illegal tiated programs have been structured as pro- forms of financing are distinguished by whether jects with a limited lifespan rather than as MFIs or not they are sanctioned by the PBC, which aiming for sustainability (Cheng, 2003; IFAD, hinges on whether they mobilize savings from 2001, pp. 20–21; Park & Ren, 2001). Although the general public and offer/charge interest this may be attributed in part to the official rates above the repressed interest rate ceilings. interest rate ceilings on poverty loans, the FPCs Interpersonal lending and trade credit, for have shown that it is possible to build in a example, are among the most basic strategies higher, sustainable rate of interest in the that entrepreneurs use to deal with short-term Grameen model; and that rural borrowers are liquidity requirements. Small business owners willing and able to pay those rates. Indeed, a frequently borrow money from friends, rela- study of NGO MFI clients found that the tives, and neighboring shopkeepers. Wholesal- highest monthly interest rate that they would be ers may deliver goods to retailers on 10-day or willing to pay is 32.6%. 13 This is consistent even 30-day credit if they have an established with the popularity of informal financing relationship. Such practices are not illegal to mechanisms (discussed below) that charge even the extent that they do not entail interest above higher interest rates. the rates of state banks, 16 in contrast to those INFORMAL FINANCE AND MICROFINANCE 1493

Table 1. Legal condition of informal finance in China and India Type China India Interpersonal lending–– minjian jiedai––financial authorities do not Interpersonal lending––financial loans extended among interfere with casual, interest-free lending authorities do not interfere with friends, relatives, neigh- casual, interest-free lending bors, or colleagues Trade credit––merchandise hangye xinyong––neither sanctioned nor Trade credit, forward sales credit between wholesalers prohibited and retailers Moneylenders, loan gaolidai––all high interest lending is illegal Mahajan and Chettiar bankers–– sharks––loans from profes- Some are registered as finance com- sional and nonprofessional panies, trusts, banks, and partner- money brokers, typically at ship firms high interest rates Rotating savings and credit huzhuhui, hehui, biaohui, chenghui, juhui–– Chit funds––registered as companies, organizations (ROSCAs)–– permitted in localities where they have not partnerships, and sole proprietor- indigenously organized collapsed ships savings and credit groups Pawnshops––extend collat- diandang, dangpu––permitted when oper- Pawnshops––legal if licensed eralized loans with interest ated according to regulations Indigenous banks, money siren qianzhuang, private money houses–– Deal with short-term credit (hundis) houses, finance companies–– regarded as private banks, which are illegal; combined with trade for financing mobilize savings and extend most operate underground now trade––committees have made efforts collateralized loans to formalize them Rural cooperative founda- nongcun hezuo jijinhui––approved by MOA n.a. tions until closure by PBC in 1999 Social organizations, mutual huzhuhui, hezuo chu jijinhui (mutual Nidhi companies, mutual benefit benefit funds––registered assistance societies, cooperative savings societies, permanent funds (mainly in entities that are supposed foundations)––registered with MCA, but Tamil Nadu)––committees have rec- to serve lower-income not supposed to engage in for-profit ommended that they be regulated populations financial intermediation more stringently

charged by loan sharks or private money eliminated, but after the first one opened up houses. The latter are clearly illegal by PBC during the reform era in Chengdu in 1987, they standards because they reflect the higher mar- developed rapidly and by 1993, there were ket cost of capital in a financially repressed 3,013 documented pawnshops throughout the environment. Indeed, with the sole exception of country. Most were operated by various bran- Minsheng Bank, 17 private commercial banks ches of government agencies, including state are prohibited in China and the PBC has banks, policy departments, tax bureaus, cus- launched multiple ‘‘financial rectification cam- toms bureaus, and finance and insurance com- paigns’’ to shut down private money houses. panies (Li, 2000), though some simply Nonetheless, they have continued to operate registered as ordinary private businesses with underground, not only in the coastal south the Industrial and Commercial Management where private commerce is better developed, Bureau (ICMB). The official interpretation of but also in northern central provinces such as the ‘‘new’’ pawnshops was that they differed Henan (Tsai, 2002a, Chap. 5). fundamentally from the traditional exploitative Pawnshops straddle a finer line between ones. As explained in a Ministry of Finance being legal and not quite legal and provide a report, good example of Beijing’s regulatory ambiva- lence in dealing with unconventional financing It should be noted that today’s pawnshops in the mechanisms. Their re-emergence during the country are not entirely what they used to be. Pawn- reform era has been uneven and ambiguously shops in old China took in personal effects at very 18 regulated due to their usurious connotation. low prices when the owners were poverty-stricken. By 1956 private pawnshops were effectively However, such businesses today represent a medium 1494 WORLD DEVELOPMENT

for normal commodity circulation...The new-born share the attribute of being legal according to pawn brokering aims to serve the people and social certain governmental agencies, but not sanc- production (Zhongguo yinhang, 1993, pp. 240–243). tioned by the PBC. The establishment of rural cooperative foundations (RCFs) by the Despite this more favorable, revisionist Ministry of Agriculture in the mid-1980s evaluation of pawnshops, it became increas- exemplifies this phenomenon (Cheng, Findlay, ingly apparent that many were (illegally) & Watson, 1998; Du, 1998). As noted earlier, mobilizing savings deposits from the public and the PBC never recognized them as legitimate offering high rates of interest. 19 As a result, in ‘‘financial institutions’’ because another min- 1994 the PBC was granted administrative isterial bureaucracy created them. Nonethe- authority over pawnshops and two years later, less, by the early 1990s RCFs had been a PBC-lead crackdown on illicit financial established in approximately one-third of all institutions closed over half of the registered townships, and by 1998 there were over pawnshops, leaving only 1,304 shops with PBC 18,000 RCFs with over five million depositors licenses. 20 In a further attempt to circumscribe (Holz, 2001). Since RCFs were not permitted the financial malfeasance of pawnshops, they to mobilize deposits or extend loans like were reclassified in 2000 from being ‘‘financial formal financial institutions, they used institutions’’ under the PBC’s authority, to ‘‘a euphemistic terms for comparable transac- special kind of industrial and commercial tions; instead of paying interest on deposits, enterprise’’ regulated by the State Economic for example, they sold ‘‘shares’’ (rugu) and and Trade Commission (JJRB, 2000). In short, extended ‘‘capital use fees’’ (zijin zhan fei- over the course of the reform era, pawnshops yong). Like pawnshops and other forms of have been legally registered in some cases, informal finance, RCFs had a variety of registered with the incorrect local agency in governance structures and were more central others, and engaged in practices that are clearly to rural finance in some provinces than oth- illegal. ers (Park, Brandt, & Giles, 2003). Their While pawnshops are now technically sub- quasi-legal status proved to be short-lived, ject to central-level regulations, rotating sav- however. As part of broader national efforts ings and credit associations (hui) remain to rectify the financial system, in March unregulated in most localities. When hui 1999, the State Council announced the clo- involve relatively small groups of people (5–10 sure of poorly performing RCFs, and the members on average) who pool set monthly takeover of better performing RCFs by Rural contributions and rotate the disbursal of the Credit Cooperatives. These actions triggered collective pot of money to each member, local farmers’ protests in at least six provinces, governments usually consider them a produc- including Sichuan, Hubei, Hunan, Henan, tive form of mutual assistance among ordin- Guangxi, and Chongqing (AP, March 22, ary people, typically women. But if a member 1999; AFP, March 23, 1999). Apart from runs off with the collective pot early in the life RCFs, some de facto nongovernmental of an association, the members who have not financial institutions have managed to operate had their turn in collecting money are cheated above ground and serve private businesses by out of their contributions. In the coastal registering as social organizations, which are south, a handful of high-profile cases have administered by the Ministry of Civil Affairs. accumulated where various types of hui were These go by a variety of names, including exposed as fraudulent schemes organized by mutual assistance societies and cooperative con artists (Tsai, 2000). The large-scale cases savings foundations. The credit societies are were not traditional ROSCAs, however, but supposed to be nonprofit organizations that rather, ponzi schemes that are never sustain- serve impoverished populations. In practice, able because they generate extremely high however, they operate like RCFs or private returns by exponentially expanding the net- money houses in the sense that they mobilize work of investors. Hui collapses make head- savings, extend credit to private entrepreneurs lines, but they are actually relatively rare. As who may be well off, and use interest rates such, it is only in a small handful of localities that are higher than that set by the PBC. that hui have been banned by local govern- These types of social organizations should be ments. distinguished from those that are genuinely The ambiguous and shifting legal status of oriented toward poverty alleviation via mi- other curb market practices listed in Table 1 crofinance. INFORMAL FINANCE AND MICROFINANCE 1495

(b) Attempts at mainstreaming India’s example, moneylenders acts at the state level informal sector regulate nonborrowing lenders, while borrow- ing lenders (or intermediaries) are also subject Relative to China, India has a longer history to various types of regulation. 21 Furthermore, of state-directed credit for poverty alleviation, the RBI has tracked informal financial activi- yet its formal financial sector is more liberalized ties in official statistics as a means to measure and its informal financial sector, better docu- progress in expanding credit access into rural mented and more likely to take corporate forms areas. (Table 2 lists the official categories of than those of China. These apparent inconsis- informal finance as defined by RBI and Figure tencies may be attributed to the fact that 1 shows their relative share of the curb market India’s financial policy environment has also over time.) The extent of curb market regula- fluctuated considerably over the years. Post- tion and tracking in India stands in contrast to independence governments in India have been the situation in China where many types of concerned about the negative effects of infor- informal financing activities are simply banned. mal finance on rural welfare and made repeated After taking into account sampling and efforts to regulate and create institutional nonsampling errors in the decennial surveys, alternatives to the curb. Indeed, what most the main trend is that informal credit has cer- observers would regard as informal financial tainly declined as a percentage of total debt, intermediaries are registered under the Com- and both professional and agricultural money- panies Act, 1956 or regulated by the RBI. For lenders have reduced their share of the curb

Table 2. Breakdown of informal finance in rural India over time Type of noninstitutional sources 1951 1961 1971 1981 1991 Landlords 3.5% 1.1% 8.6% 4.0% n.a. Agricultural moneylenders 25.2 47.0 23.1 8.6 6.3 Professional moneylenders 46.4 13.8 13.8 8.3 9.4 Traders and commission agents 5.1 7.5 8.7 3.4 7.1 Relatives and friends 11.5 5.8 13.8 9.0 6.7 Others 1.1 7.5 2.8 5.5 4.9 Unspecified n.a. n.a. n.a. n.a. 3.8 Informal credit as share of total 92.8% 82.7% 70.8% 38.8% 39.6%a household debt Source: Reserve Bank of India, All-India Debt and Investment Survey, various years. a 1991 figures do not add up to 39.6% even though Table 5 of the 1991–92 AIDIS report clearly states that non- institutional agencies account for 39.6% of total rural household debt.

Landlords

60% Agricultural moneylenders 50% Professional 40% moneylenders 30% Traders and 20% commission agents 10% Relatives and 0% friends 1951 1961 1971 1981 1991 Others

Figure 1. Distribution of informal financing mechanisms, India 1951–91. 1496 WORLD DEVELOPMENT market over time. The decline of the money- completely informal, a number of chit funds in lender in official statistics reflects in part state India are registered as companies, partnerships, efforts to register and regulate professional and sole proprietorships under the All-India moneylenders during the 1950s. Some went Chit Funds Act 1982 or the state acts (Ruth- underground to avoid regulation and others erford & Arora, 1997). The state’s rationale for were probably re-classified as agricultural regulating them is to increase the security of the moneylenders or traders (Bell, 1990). In members’ contributions and to reduce the this regard, note that the first three official incidence of defaults. As such, organizers are categories of informal lenders––landlords, required to have licenses and make security agricultural moneylenders, and professional deposits with the Register of Chit Funds; the moneylenders––are not necessarily distinct cost of collecting the pot (i.e., the de facto from one another depending on the locality. interest rate) is capped at 30% of the size of the But generally speaking, landlord lenders extend pot; and chit funds are limited to a maximum credit to tenants; agricultural moneylenders of 60 months (Ghate et al., 1992, p. 197). These primarily deal with agricultural laborers and regulations have not had their intended effect, small farmers; and professional moneylenders however. Rather than increasing the stability of service a wider range of customers and may chit funds in general, many organizers have register themselves as companies, partnerships, gone underground and taken their members and trusts (Ghate et al., 1992, p. 45). (who seek higher returns) with them. Those in the fourth official category of In addition to chit funds, Nidhi companies or ‘‘traders and commission agents’’ are also mutual benefit societies are also an important known as indigenous bankers. In contrast to part of the nonbanking world of financial professional moneylenders who lend their own intermediation, especially in south India. money, indigenous bankers broker funds Incorporated under the Companies Act 1956, between banks and their clients, who tend to be Nidhis mobilize savings from their members traders rather than farmers (Schrader, 1994). and extend loans that are collateralized with The Shroffs of Western India, for example, jewelry and real estate (Nayar, 1992, pp. 197– provide a short-term credit instrument called 199). When nonmembers wish to make a darshani hundi to traders who need to travel deposit or borrow from a Nidhi, they take a great distances to purchase inventory and share of the Nihdi. Over the years, the state has transfer funds (Ghate et al., 1992, pp. 198–200). made repeated efforts to regulate these mutual In addition to serving as financial intermedi- benefit societies; and an Expert Group on aries, indigenous bankers are also business- Nihdis constituted by the Department of people themselves. 22 Besides trading, they may Company Affairs has recommended a host of operate commission agencies or hire-purchase additional regulations to professionalize their finance (HPF) companies, which are basically operations (PIB, 2002). leasing companies that finance automobiles and other goods over a fixed term for clients who lack sufficient cash to purchase capital goods 4. THE PERSISTENCE OF INFORMAL up front (Nayar, 1992, pp. 199–200). Even FINANCE though formal sector HPFs exist, one study found that informal HPFs finance a much State authorities in both China and India higher volume of vehicles than official auto have clearly recognized the importance of for- finance corporations––probably because lower- mal financial institutions in rural areas, income populations find the informal HPFs including the expansion of microfinance pro- more accessible (Das-Gupta & Nayar Associ- grams for poverty alleviation purposes. Finan- ates, 1989). cial regulators have also made repeated efforts Forms of informal finance in the other cate- to eliminate and/or regulate of curb market gory also include indigenous bankers who are activity. Why, then, has informal finance per- not registered as traders or commission agents; sisted and even expanded in both countries? unregistered finance corporations; nonprofes- Four complementary explanations may be sional moneylenders (other than those identi- derived from the perspective of supply-leading fied as friends and relatives); various types of economics, state–society relations, the local leasing, investment, and housing finance com- political economy of markets, and the institu- panies; ROSCAs (chit funds) and Nihdi socie- tional characteristics of lending programs. As ties. Unlike the ROSCAs in China, which are will be shown, the first two hypotheses capture INFORMAL FINANCE AND MICROFINANCE 1497 macrolevel dynamics, while the second two ristically test the supply-leading hypothesis by have explanatory leverage at lower levels of considering the impact of the government’s analysis. Specifically, the economic hypothesis large-scale poverty lending programs. Specifi- concerns the gap between the overall supply cally, if people were turning to informal finance and demand for formal credit, and the state– only because more institutionalized sources society hypothesis concerns limits in state were unavailable to them, then ceteris paribus capacity to reach the intended recipients of we would expect clients of microfinance pro- subsidized loans and microcredit. Both the grams to rely on subsidized poverty loans local political economy and institutional design rather than high-interest loans from the perspectives then explain why certain groups of curb. 23 Yet this turns out not to be the case. In people do or do not receive credit at the ground their surveys of MFI clients, Park and Ren level. The rationale for each explanation is (2001) discovered, discussed in more detail below. [O]ver 50% of households in program areas had out- standing loans from other sources, and that this per- (a) Neoclassical economics: a matter of supply centage was similar for both members and and demand nonmembers. The most common source by amount was Rural Credit Cooperatives (55% for members, Based on the logic of supply and demand in 46% for nonmembers), followed by informal sources. neo-classical economics, one reason that infor- Thus it does not appear that microfinance participants mal finance continues to play such an impor- lack access to other credit sources, whether formal or tant role in rural China and India may simply informal (p. 46). be because the amount of credit demanded by rural households exceeds that supplied by the Their study also found that the overall level of formal financial sector in rural areas. There- indebtedness is higher among microfinance cli- fore, to reduce the rural population’s reliance ents, and that only one-quarter of them would on the curb, official sources of credit should be have engaged in income-generating activities of increased. This is known as the supply-leading the same scale in the absence of these loans. approach to finance and development (cf. This suggests that although government-spon- Chandavarkar, 1992; Patrick, 1966). Table 2 sored microloans are not going only to rural shows that according to official statistics, the households that lack access to formal credit, relative dominance of the curb market in rural ‘‘microfinance lending relieves credit con- India has declined over time with the expansion straints at the margin’’ (Park & Ren, 2001, p. of formal credit institutions. Bell (1990) draws 46). on two independent World Bank surveys to Applying the supply-leading hypothesis to demonstrate, however, that the decennial RBI India and China explains in part the on-going surveys underestimate the true scale of informal popularity of informal finance among rural finance. This leads him to conclude: households, but it also raises a number of questions. Why is business getting better for India’s moneylenders amid the expansion in [A]lthough the moneylender did lose ground relative to [formal financial] institutions over the period from formal sector institutions and MFIs? If MFI 1951 to 1981, he remained a very important source of clients already have access to RCC loans, then finance to rural households, and the expansion of why are they receiving MFI loans in the first aggregate debt was almost surely so great as to imply place? Examining the issue from the perspective that his volume of business grew. of state–society relations helps to explain this disjuncture between the intended and actual In other words, despite significant increases recipients of targeted and microcredit. in the supply of bank loans and microcredit (over US$15 billion), rural households continue (b) State–society relations: a matter of state to draw on informal sources of credit. A more capacity recent study of credit rationing in rural India confirms that this is due to the combination of A second reason why increasing the official limited access to formal credit and continuing supply of credit has not translated into a high demand for such credit (Swain, 2002). matching decline in informal financial activity In rural China, the closure of RCFs elimi- is because official state policies are not being nated an important source of semi-formal implemented properly. This may occur in three financial intermediation, but we can still heu- main ways: First, state actors may not be 1498 WORLD DEVELOPMENT distributing targeted credit properly due to Aside from the top-down weaknesses in state insufficient knowledge of how to identify the capacity discussed above, nonstate actors may intended clients of subsidized credit and MFI also be responsible for distortions in policy programs. Second, state actors may intention- implementation (Migdal, Kohli, & Shue, 1994). ally divert credit from the intended recipients. In this case, nonstate actors would include Third, nonstate actors may interfere with the private economic actors such as financial proper disbursement of formal and MFI credit. entrepreneurs and politically important con- Taken together, all three types of implementa- stituents of society. First, the argument could tion failure could be interpreted as reflecting be made that the curb market thrives because weaknesses in state capacity (Evans, Reusche- informal financiers are determined to evade meyer, & Skocpol, 1985). banking regulations. In other words, no matter The first type of implementation failure is how much formal credit is available in rural rooted in the conditions under which formal areas and no matter how stringent the penalties credit is disbursed. In both India and China, are for violating state laws, a certain strata of conventional commercial banks do not have financial entrepreneurs will always endeavor to institutional experience in lending to rural cli- subvert state policies. After all, informal ents who lack an established credit history and finance persists even in advanced industrialized collateral or guarantor. Therefore, the typical countries with sophisticated financial systems. state response has been to require that The second main expression of noncompliance national banks allocate a certain portion of by societal actors may come from borrowers their lending portfolios to lower-income rural themselves. One could argue that lower-income households. Quota-style lending often does not farmers and rural traders boycott formal and achieve its substantive objectives, however, semi-formal financial institutions to undermine because the emphasis is placed on ensuring their legitimacy (Selden & Perry, 2000). Thus that a certain number of loans is disbursed, far, however, there is no evidence for this in rather than on the identity and needs of the India and China. Instead, it is more typically borrower. the privileged slice of the population that has When quota-style lending is accompanied by interfered with the implementation of PA subsidized interest rates––which has been the lending policies. In India, local politicians may case in both India and China’s PA loans––the extend subsidized credit to the upper tier of prospects of reaching the intended clientele are society, but after elections, loan recovery has further diminished. Instead of reaching lower- also proven to be difficult because ‘‘the credit income households, subsidized loans usually agencies’ bureaucracy is reluctant to touch the end up in the hands of local elites who do not influential rural elite who wield much formal feel obligated to repay the loans (Adams et al., and informal influence and considerable power’’ 1984). This common phenomenon relates to the (Yaron, Benjamin, & Piprek, 1997, p. 102). second implementation failure, whereby state The low repayment rates in China’s subsidized agents knowingly distribute credit to sectors of PA programs suggest that similar dynamics are the population that are not necessarily excluded in operation. 24 from the formal financial system. In India a Analyzing the persistence of informal finance number of government interventions in rural through the state–society lens takes us one step finance have been motivated by short-term closer to explaining why state financial policies political objectives that coincide with the elec- have had unintended outcomes, i.e., why state- toral cycle. While China’s political context subsidized credit and microfinance have not differs significantly from India’s, targeted credit gone to their intended recipients. But concep- and PA loans are similarly subject to political tualizing the curb market as an inverse function patronage at the local level. Compared with of state weakness and societal strength suffers participants in the FPCs and mixed NGO- similar problems as the supply-leading government programs, borrowers in the mi- hypothesis. Just as increasing the supply of crocredit projects run by local governments government-sanctioned credit does not auto- tend to be much wealthier and engaged in matically crowd out informal credit, strength- noncropping activities (Park & Ren, 2001). The ening state capacities in rural financial next section shows that the underlying reason intermediation does not necessarily come at the for this second type of implementation failure is expense of nonstate actors such as money- due to local market segmentation along politi- lenders and wealthier households if local agents cal and social lines. and institutions face competing political INFORMAL FINANCE AND MICROFINANCE 1499 incentives. In both China and India, intrastate credit cannot go to the highest economic bid- actors (such as local officials and bureaucrats der; instead, it is disbursed by credit officers. charged with loan disbursal) are just as likely as Moreover, as with any government-allocated nonstate actors to distort policy implementa- good or resource, the distribution of subsidized tion. credit and PA loans is political. Therefore, when PA loans do not reach the target popu- (c) Segmented markets: a matter of institutional lation, more often than not, examining local design and local political economy political and social hierarchies may reveal where the soft loans were distributed. Ultimately, rural credit markets are more Similarly, the cost of accessing informal finely differentiated than a dichotomous trade- credit also varies depending on the structure of off between state and society. Rather than local political and social networks. Interest-free assuming the perfect fungibility of credit lending only occurs among tight knit groups of (whether it be formal, semi-formal, or infor- people, typically close friends or relatives. mal), this explanation starts from the premise Members of ROSCAs usually know one that credit markets are segmented even at the another, or at a minimum, know the organizer grassroots level. This means that no single type or one other member. The higher rates of of credit can meet the needs of various potential interest charged by professional moneylenders borrowers, and no single type of credit is reflect in part the higher level of risk associated accessible to everyone (Hoff & Stiglitz, 1990). with lending to clients with unconventional The concept of segmented markets typically forms of collateral (if any). Even then, however, refers to the variation in preferences among accessing most forms of informal finance consumers in different economic strata, e.g., in requires some form of introduction. Local curb terms of loans for consumption versus pro- markets are also segmented, though not always ductive purposes, and the conditions of credit in expected ways. The following two cases from access such as collateral, third party guaran- India and China illustrate more concretely how tees, and savings requirements. In other words, local social, political, and economic dynamics various forms of credit are not functionally mediate the use of both formal and informal equivalent to the borrower. Both the institu- finance. 25 tional design and lending methodology of differ- ent forms of credit influence the relative (i) Tribal, caste, and occupational segmentation attractiveness of, for example, government- in a North Indian village subsidized loans versus unsubsidized microfi- In a diachronic study of a South Rajasthan nance loans from NGOs versus high-interest village that Jones (1994, Chap. 18) calls Chan- loans from the curb. All of them have different drapur, we can compare the nature of the local restrictions in terms of loan size, amount, credit market before and after a village bank repayment terms, preferred clientele, etc. was introduced. As of 1989, Chandrapur Vil- Hence, a borrower might take out a high- lage had a population of over 1,000 people in interest loan from a moneylender rather than a 200 households, within which were four main low-interest one from a government program social groups engaged in different economic because the former entails lower transaction activities: Hindu households engaged in caste- costs or because the latter requires that the loan based nonagricultural activities, Jain house- be used for productive purposes. holds prevailed in commercial and financial In addition to supply-side differences in the services, Jogis relied on income from working institutional types of rural credit, market seg- as migrants in Gujarat and Bombay, and the mentation also occurs along political and social Bhil population lived in the hinterland. Before lines, which further distorts the way local credit a village bank was introduced at the end of markets function in practice. Far from being a 1983, Chandrapur residents relied solely on pure market where prices (interest rates) reflect informal sources of credit. The records of a Jain the relative supply and demand of different shopkeeper (called B. Jain) who provided types of financing, formal and semi-formal pawnbrokering services revealed that even six sector credit for PA purposes often faces state- years after the village bank was established, B. mandated interest rate ceilings and is subsi- Jain’s lending volume had increased by over dized. That is to say, even during periods of 100% in nominal terms––from Rs. 53,351 credit scarcity, the cost of directed bank credit (US$5,455) in 1982–83 to Rs. 110,818 may be extremely low. By definition directed (US$6,756) in 1988–89––and the annual 1500 WORLD DEVELOPMENT number of loans had increased from 290 to 335, 52% of the bank’s loan portfolio was in arrears, but the interest rate had remained at 3% per and 30% was past due for over three years, i.e., month throughout the same period. 26 Mean- in default. By way of contrast, during the same while, the total number of pawnshops in the period 70% of the loans extended by B. Jain’s village increased from 15 in 1983 to 24 in 1989. pawnshop were repaid in full. Most remarkably, however, Jones found that Third, the local credit market was highly the volume of loans extended by pawnbrokers segmented on both the supply and demand vastly outstripped that of the village bank: sides. On the supply side, the lending method- ology varied considerably among different For Chandrapur, as a whole, a tentative estimate of sources of credit. The village bank did not offer pawnbroking loan volume is made by multiplying the types of services demanded by certain Rs. 110,818 (B. Jain’s loan volume) by the proportions groups in the community. In contrast, the of loan volume indicated by this shopkeeper for the popularity of pawnbrokering and mutual other 23 lenders in the village. Adding the figure to his own loan volume produces a total of Rs. finance groups may be attributed to their flex- 2,292,850 for all 24 pawnbrokers in the village: five ibility relative to restrictions associated with times the loan volume advanced by the bank during loans from the village bank. Villagers turned to 1988–89. the pawnshops to meet seasonal needs such as productive household consumption (e.g., A similar extrapolation from the 335 loans advanced housing construction, education, migration), by B. Jain, results in a total of 6,799 loans for all 24 pawnbroking businesses: 75 times the number of loans agricultural cultivation, and ceremonial expen- advanced by the bank in 1988–89, six years after it was ditures. At the other end of the income spec- established (Jones, 1994, pp. 18–24). trum, members of mutual finance groups used them to engage in money lending rather than In addition to the expansion in pawnbro- consumption or productive investment pur- kering, mutual finance groups emerged during poses. the same phase. By 1991, 50 out of the 200 At the same time, Chandrapur’s local credit households in the village were participating in market was also highly segmented along tribal these savings and credit groups, and by 1992, and occupational lines. For example, the Jogis the loan volume of mutual finance groups was who are on the Scheduled Caste lists were comparable to that of pawnshops and exceeded supposed to receive targeted credit from the that of the village bank (Jones, 1994, pp. 18–8, village bank, but due to their life cycle and 18–9). Why did a substantial expansion in curb consumption needs (e.g., weddings and funer- market activities follow the introduction of the als), they ended up relying on pawnbrokering village bank? loans from Jain shops. Meanwhile, only 23% of Each of the explanations outlined above the number of pawnbrokering loans extended offers insight into the question. First, from an by B. Jain went to local villagers, while 75% of economic perspective, one could infer that the the loans went to Bhil customers who focused overall demand for credit in Chandrapur sim- on agricultural cultivation in tribal settlements. ply increased dramatically over those years By 1989, Jain households themselves did not such that a single village bank could never have use the services of pawnshops because ‘‘to take fulfiled the demand. Indeed, during 1988–89 the such a loan would involve loss of prestige with village bank accounted for only 90 out of the fellow Jains (Jones, 1994, pp. 18–27).’’ Instead, total of 425 loans extended in the village (Jones, Jains not only enjoyed privileged access to 1994, pp. 18–23). 27 loans from the village bank, but also dominated Second, irrespective of credit supply, the vil- the ownership of pawnbrokering businesses lage bank itself was poorly managed and failed and accounted for one-third of the participants to carry out its intended mandate. Specifically, in mutual finance groups, which were geared the Chandrapur village bank was supposed to toward enhancing the volume of their informal service a total of 17 different villages, yet lending activities. Chandrapur village residents alone received In addition to intertribal and caste differen- over half (54%) of its loans. Furthermore, even tiation, informal credit markets are also seg- though the Bhil are Scheduled Tribe members mented along occupational and gender lines. and represent a specific target group of the This is reflected in the participation of savings bank, over half (52%) of these bank loans were and credit groups: of the 126 people partici- extended to Jain borrowers who are relatively pating in mutual finance, 125 are men; and the well off. It is also worth noting that in 1989, groups are organized by professional occupa- INFORMAL FINANCE AND MICROFINANCE 1501 tion such that the Jains form the Government nesses and accessing official sources of credit. Employees’ group, the Blacksmith Caste form Although members of the second branch are the School Staff group, and relatively few (8%) neither politically privileged nor persecuted, Bhil cultivators participate in mutual finance. they also have disadvantaged access to various These multiple dimensions of segmentation production inputs relative to the first branch. help to explain why the scale of informal It is important to point out, however, that finance actually increased after the introduction the political hierarchies in Lin Village have not of the village bank: not only did the village translated neatly into economic stratification. bank deviate from its mission, but ironically, During the Mao era, the third branch certainly the fact that most of the bank’s loans went to suffered more than most, but in the reform era, local curb market financiers (Jains) enabled the second and third branches have found ways them to expand the provision of informal to operate private businesses without going financial services to other groups in the village. through official channels. Given the paucity of arable land, 30 virtually every household in Lin (ii) Segmentation within a single surname village Village operates a small factory. Interestingly, a in South China member of the third Lin branch owns the While it may seem intuitive that a multitribal largest of these factories with over 30 employ- village would have a segmented economic ees––yet he has never borrowed from formal structure and therefore credit markets, the case sources of credit. Owner Lin explained, of a single-surname village in the southern coastal province of Zhejiang shows that strong It’s not worth it to me to apply for a loan from a state internal forms of differentiation are not bank or rural credit cooperative because the credit uncommon even in a village where everyone officers are dirty and rip me off given my family back- ground. If I applied for a 100,000 RMB (US$12,000) could be considered a relative of some sort. Lin loan, I would only receive 60,000 RMB (US$7,200) Village is comparable in size to Chandrapur because the credit officer would pocket the other Village, but unlike the latter it appears 40,000 RMB (US$4,800). Meanwhile, I would still homogenous: 95% of the households share the be expected to pay interest on 100,000 RMB. surname Lin and the village temple, which traces the Lin lineage back to the late Qing Owner Lin explained that households from dynasty. 28 Despite this shared ancestry, access the first Lin branch were more likely to borrow to various forms of credit is segmented along from state banks or RCCs because their rela- political, sectoral, and gender lines in Lin Vil- tives work there. Lacking such official connec- lage. tions, Owner Lin nonetheless managed to The political fault lines in the village are invest 700,000 RMB (US$84,000) in his based on the three branches of Lins that orig- motorcycle parts factory by using 100,000 inally settled in the village. The first branch was RMB (US$12,000) of his own savings, bor- very active during the Communist Revolution rowing 200,000 RMB (US$24,000) interest free and ended up with the most Communist Party from his four older siblings, and borrowing members. The third branch was the most 400,000 RMB ($48,000) at 24% annual interest prosperous one before the Revolution and was through moneylenders (yinbei). The latter loans thus subjected to considerable political perse- were guaranteed by his sisters who have good cution throughout the Mao era. For example, credit among moneylenders in the textile sector. during the Great Leap Forward, adult mem- As of 2001, Owner Lin still had the largest bers of the wealthiest household were sent to re- factory in town even though his family’s local education through labor (prison) camps and political status remained low. their spacious traditional courtyard home was Before Owner Lin’s motorcycle parts factory turned into the communal mess hall. 29 The was established in 1998, there were larger col- privileged position of the first branch has car- lectively-owned factories in the village and each ried over into the reform era. Even though of them raised their funds in different ways. For major decisions are supposed to be made by the example, at the outset of reform, there was an democratically elected Village Committee, iron factory, which relied mainly on RCC loans households from the first branch dominate because it was run by managers from the first village governance and the allocation of key branch of Lins. Later on, in the early 1980s, resources, including access to land and credit. about 25 households in the second branch set As such, members in the third branch have a up a plastics factory by pooling their savings difficult time contracting land for their busi- for four years and registered it as a collective 1502 WORLD DEVELOPMENT enterprise. This is called the ‘‘wearing a red how single-surname villages may be internally hat’’ strategy because the plastics factory was differentiated (Tsai, 2002b). Indeed, regardless really privately owned––registering it as a col- of the particular distribution of surnames at the lective gave it preferential land use and tax village-level, many other patterns of local seg- treatment. Meanwhile, clusters of smaller mentation may be identified in rural China household factories producing sugar, lime, depending on the structure of the economy, the paint, autoparts, and textiles tend to raise their nature of geographical constraints or resources, start-up and working capital in sectorally dis- the extent of external versus internal migration, tinctive ways, except in cases where extended and the often path dependent developmental families are involved in more than one sector. orientation of the local government (Unger, Given the rapid industrial transformation of 2002; Walder & Oi, 1999; Whiting, 2001; Wu, Lin Village in the first two decades of reform, it 1998). is unlikely that the local RCC and county-level Finally, despite the shared popularity of Agricultural Bank could have kept up with the informal finance, the financial landscape in Lin grassroots demand for investment and working Village differs significantly from that of Chan- capital. As such, the expansion in informal drapur. While the introduction of formal financing during the reform era is not surpris- finance to the latter had the unintended effect of ing. It is noteworthy, however, that the busi- expanding the volume of the curb, in Lin Vil- nesses that have received formal sector loans, lage the formal financial institutions have i.e., those run by the first branch, have not always been captured by local political elites performed as well as those financed by the curb. who are not especially adept at business (cf. This may be attributed at least in part to the Adams et al., 1984; Otero & Rhyne, 1994). The tendency of local political elites to view the vast majority of commercially successful oper- loans as grants rather than serious business ations in Lin Village have relied on informal obligations. financing mechanisms that do not involve the Gender represents the third major dimension first branch of Lins. This is typical of China’s along which credit markets are segmented in private sector as a whole. As of year-end 2003, Lin Village. In contrast to the male-dominated less than 1% of all loans extended by state savings and credit groups in Chandrapur, the banks were going to private entrepreneurs ROSCAs in Lin Village (called chenghui or hui) (PBC, 2004). Hence, even though there is are only managed by women. In China’s political and economic segmentation at the southern coastal provinces, women dominate local level, the expansion of informal finance in hui participation because they have better China is largely attributable to the limited developed social networks with one another, supply of formal credit to the nonstate sector. because they are more likely to remain in town year round (as opposed to men who may engage in seasonal migration), and because 5. CONCLUSION men are more likely to have other financing options (Tsai, 2000). In Lin Village, a handful The enduring popularity of informal finance of middle-aged women run ROSCAs full time, in rural China and India may be traced to a but most hui organizers have other income- number of complementary factors: First, for- generating activities as well. The organizer with mal financial institutions and microfinance the largest volume of hui in Lin Village, for programs are often unable to meet the demand example, is a doctor who operates the village for grassroots credit. But, merely increasing the clinic from the courtyard home of the third Lin availability of official credit may not reach the branch. At any given point in time, she runs up targeted population because it still needs to be to five hui in the range of 200,000 RMB disbursed in some manner. Even if the supply (US$24,000) each (i.e., 20 members contribut- of official credit were sufficient, credit officers ing 10,000 RMB/US$1,200 each meeting), and and poverty alleviation cadres charged with the the interest rates run up to 36% annually. Vil- task of extending loans to rural households lagers find Doctor Lin to be a trustworthy often face local pressures and incentives for organizer because as the village doctor, she credit distribution that deviate from the origi- knows everyone and is unlikely to flee town nal intentions of state authorities. This is with their money. especially the case when it comes to subsidized The manner in which credit markets in Lin microfinance programs because microloans are Village are segmented is only one example of readily treated as political patronage. Mean- INFORMAL FINANCE AND MICROFINANCE 1503 while, curb market operators at the grassroots intermediaries to local populations in the level generally have a comparative advantage in Scheduled Tribe and Scheduled Caste lists. serving rural households because they possess Meanwhile, the Lin Village case demonstrated better knowledge about local market actors and that from the perspective of borrowers with conditions. lower political status, formal sector credit is This is not to say, however, that informal actually more expensive to them than the curb. finance trumps formal finance in either eco- Informal and formal sources of finance are not nomic or normative terms, but rather, that top- necessarily in competition with one another down efforts at rural financial intermediation because they serve different segments of local are not likely to achieve their objectives if they society. are not structured in a sustainable manner and Analytically, if we accept that local-level implemented properly. It is no wonder that political and economic dynamics fundamen- subsidized PA programs have low repayment tally mediate developmental outcomes, then it rates when the loans are presented as develop- makes sense to transcend the conventional mental side-payments. MFIs that charge sus- state–society dichotomy by disaggregating both tainable interest rates, on the other hand, tend state and society. Just as local state agents may to have higher repayment rates; and while subvert central state objectives, different seg- reliable estimates of repayment rates in the curb ments of society may be at odds with one are not available, it is probably safe to say that another. Recognizing the complexity of grass- most informal financiers face hard budget roots segmentation ultimately has implications constraints. While the constant threat of for the local distribution of both governmental bankruptcy looms over the curb, the potential and nongovernmental sources of finance. promise of additional subsidies fuels targeted Commercial bank credit, subsidized loans, mi- microcredit. That is why informal finance and crofinance facilities, and curb market financing microfinance are imperfect substitutes. Rather all entail a mix of social, political, and eco- than crowding out informal finance, the infu- nomic incentives that are contingent on local sion of public and donor funds into microfi- context. Only when microfinance programs are nance adds another discrete source of credit in structured according to local needs and aimed local markets. As seen in the Chandahar case, at cost recovery will microfinance hold greater the establishment of a village bank enabled potential for displacing usurious forms of pawnbrokers to expand their role as financial informal finance.

NOTES

1. Useful compilation of the debates include Bouman but also officially sanctioned microfinance programs, and Hospes (1994) and Hoff, Braverman, and Stiglitz which include both subsidized and unsubsidized pro- (1993). grams, as well as state-sponsored and NGO-led pro- grams. Informal and semi-formal finance will generally 2. By definition, informal finance refers to financial be discussed together because both fall beyond the scope flows that occur beyond the scope of a particular of standard commercial and developmental/policy-ori- country’s formal financial system of banks, nonbanking ented financial institutions. As the article discusses, financial institutions, and officially sanctioned capital many forms of informal finance are subject to regulation markets. Most countries, however, also have a range of in India, while most forms of informal finance are simply financial intermediaries that are best described as semi- banned in China. formal because central banking authorities do not regard them as part of the formal financial system, but they may 3. RRBs represent a hybrid between cooperatives and be approved by some government agency or entity. In commercial banks; they were established specifically to India and China, the definitional boundaries among serve impoverished farmers, laborers, and microentre- informal, semi-formal, and formal finance have shifted preneurs in rural areas. due to changes in their political, macroeconomic, and regulatory environments. Furthermore, each of the 4. By 1999 India had a total of 140,000 branches of categories encompasses a wide range of different financ- various rural credit facilities, which is equivalent to one ing mechanisms. In this article, formal finance comprises formal financial institution per 5,600 rural citizens not only conventional banking and financial institutions, (Sinha, 2000, p. 66). 1504 WORLD DEVELOPMENT

5. In 1984 the responsibility for RCCs shifted from the 15. Note that ‘‘high’’ interest rates are defined as PBC to the ABC. Also, the Agricultural Development rates that exceed the interest rate ceilings in China Bank was established in December 1993 to handle the and the anti-usury laws in India. These limits are in policy lending functions of the ABC so that the latter the range of 10–12%/year for China and 24%/year for could devote itself to commercial banking activities. India.

6. During 1978–98, one US dollar ranged from Rs. 8.2 16. They are ‘‘legal’’ to the extent that they have not to 41.3. The conversion for Rs. 250 billion is based on an been banned explicitly. annual average rate of 20.3 over that period. The figure includes repeat assistance to the same families. 17. China Minsheng Banking Corporation was estab- lished by the All-China Federation of Industry and 7. Especially during the 1980s, nationalized banks had Commerce in February 1996. In November 2000, it went periodic loan melas, which entailed extending massive public by issuing 350 million A shares on the Shanghai quantities of subsidized loans to targeted sectors of Stock Exchange. society without regard for their creditworthiness. 18. Communist-era references to pawnshops in impe- 8. When it comes to subsidized loans for state-owned rial China condemned them as an expression of class- enterprises or collectives that employ large numbers of based exploitation. For example, Xin (1993). people, the argument could be made that propping them up has local employment and therefore, welfare impli- 19. For example, pawnshops in Xingtai, Hebei offered cations; but in the development field, microfinance refers annual interest of 40% to its depositors in 1991 (Xin, specifically to loans that are extended to individual, 1993). A more recent study found that some pawnshops small business owners rather than larger scale corpora- charge monthly interest rates between 5% and 8% (i.e., tions that have larger capital requirements. up to 72% interest annually), which is how they are able to offer depositors such high rates of return. China 9. Note that the participating ‘‘banks’’ include com- Online (September 9, 1999). mercial banks, RRBs, and cooperatives. For additional information about SHGs and what is known as the ‘‘new 20. The rectification effort was not entirely effective, microfinance,’’ see Bansal (2003), Satish (2001), and however. The PBC issued additional regulations Wilson (2002). throughout the late 1990s to standardize their operations and reiterate prohibitions against charging/offering 10. For a comparison of the performance of NGO, high interest rates––again, to limited avail in implemen- joint NGO-government, and purely government-run tation. microfinance programs in China, see Park and Ren (2001). 21. Many nonborrowing and borrowing lenders prob- ably do not comply with the various acts. Moreover, 11. Citibank has committed US$1.3 million to FPC via informal moneylenders fall beyond the scope of regula- Grameen Trust for expansion. Xinhua (November 19, tion. I thank one of the anonymous reviewers for 2002). pointing this out.

12. For a list of microfinance projects supported by 22. In pre-colonial and colonial India, Multanis, international donors, see China Development Brief Gujarati Shroffs, Marwaris, Nattukottai Chettiars, and (1999). Kallindaikurichy Brahmins represented the most prom- inent indigenous bankers. During the late colonial 13. NGO participants said they were willing to pay up period, many invested in industry and commerce (cf. to 32.6% in annual interest, while participants in Bagchi, 1972). government-run PA programs were willing to pay up to 21.4% annually and those in mixed NGO-government 23. Of course, formal and informal sector loans do not programs were willing to pay up to 20.2% (Park & Ren, have similar lending methodologies (in terms of size, 2001, p. 45). length, repayment schedule, collateral requirements, etc.). But this section of the article only focuses on the 14. The survey found 39.6% of rural households relied actual financial cost of formal versus informal credit to on ‘‘noninstitutional credit agencies.’’ RBI (2000), Table examine the issues of access to and demand for credit, 5. ceteris paribus. INFORMAL FINANCE AND MICROFINANCE 1505

24. This phenomenon is not specific to India or China 28. ‘‘Lin Village’’ is a pseudonym. This case is based (Adams et al., 1984; Otero & Rhyne, 1994). on fieldwork in Wenzhou, Zhejiang province, 2000– 01. 25. The local case study from India––‘‘Chandrapur Village’’ in South Rajasthan––was included in a larger project on microfinance in general. The local case study 29. During my visit in 2001, faded slogans from that from China––‘‘Lin Village’’ in Wenzhou, Zhejiang––was period could still be seen from the wooden beams. The part of a project on informal finance and rural indus- thought-reformed family was permitted to return to trialization in Wenzhou. their home in 1963, but as might be expected, during the Cultural Revolution much of the intricate artwork along the entryway, roof, windows, and walls was destroyed or 26. The US$ equivalent of the Indian Rupee went from damaged. Rs. 9.8 per 1 US$ in 1982–83 to Rs. 15.1 in 1988–89.

27. Note, however, that the village bank accounted for 30. Lin Village is ‘‘all mountains and water,’’ as the 80% of the total loan volume. locals put it.

REFERENCES

Adams, D. W., Graham, D., & von Pischke, J. D. China Development Brief (1999). 2, p. 2. (1984). Undermining rural development with cheap Conroy, J. D. (2000). People’s Republic of China. In credit. Boulder: Westview Press. ADB (Ed.), The role of central banks in Asia and the Agricultural Finance Corporation (AFC). (1988). Agri- Pacific. Manilla: ADB. cultural credit review: Role and effectiveness of lending Das-Gupta, A., Nayar, C. P. S., & Associates. (1989). institutions (Vol. V). Bombay: AFC. Urban informal credit markets in India. New Delhi: Ashe, J. (2002). Self-help groups and integral human National Institute of Public Finance and Policy. development. Waltham, MA: Brandeis University/ Du, X. S. (2003). Author’s correspondence, May 8. Catholic Relief Services. Du, Z. X. (1998). The dynamics and impact of the Bagchi, A. K. (1972). Private investment in India: 1900– development of rural cooperative funds (RCFs) in 1939. Cambridge: Cambridge University Press. China. Working Paper No. 98/2. Chinese Economies Bansal, H. (2003). SHG–bank linkage program in India: Research Center, The University of Adelaide, An overview. Journal of Microfinance, 5(1), 21–49. March. Bell, C. (1990). Interactions between institutional and EDA Rural Systems (1996). India: Micro-finance for the informal credit agencies in rural India. The World poor: An assessment of the status and efficacy of Bank Economic Review, 4(3), 297–327. microfinance institutions and programmes. Study Bhatt, N., & Thorat, Y. S. P. (2001). India’s regional prepared for the Asian and Pacific Development rural banks: The institutional dimension of reforms. Center, Kuala Lumpur. Journal of Microfinance, 3(1), 65–94. Evans, P., Reuschemeyer, D., & Skocpol, T. (Eds.). Bornstein, D. (1997). The price of a dream: The story of (1985). Bringing the state back in. New York: the Grameen bank. Chicago: University of Chicago. Cambridge University Press. Bouman, F. J. A.& Hospes, O. (Eds.). (1994). Financial Ghate, P., Das-Gupta, A., Lamberte, M., Poapongsak- landscapes reconstructed: The fine art of mapping orn, N., Prabowo, D., Rahman, A., & Srinivasan, development. Boulder: Westview Press. T. N. (1992). Informal finance: Some findings from Bouman, F. J. A., Bastiaansen, R., Van Den Bogaard, Asia. Manila: Asian Development Bank, Oxford Uni- H., Gerner, H., Hospes, O., & Kormelink, J. G. versity Press. (1989). Small, short, and unsecured: Informal rural Hannig, A., & Katimbo-Mugwanya, E. (Eds.). (1999). finance in India. New York: Oxford University Press. How to regulate and supervise microfinance? Key Chandavarkar, A. (1992). Of finance and development: issues in an international perspective. Proceedings of Neglected and unsettled questions. World Develop- the High-Level Policy Workshop, Kampala. ment, 20(1), 133–142. Hoff, K., & Stiglitz, J. (1990). Imperfect information and Cheng, E. (2003). Microfinance in rural China. In C. rural credit markets: Puzzles and policy perspec- Findlay, A. Watson, E. Cheng, & G. Zhu (Eds.), tives. The World Bank Economic Review, 4(3), 235– Rural financial markets in China (pp. 120–133). 250. Canberra: Asia Pacific Press at the Australia Hoff, K., Braverman, A., & Stiglitz, J. (Eds.). (1993). National University. The economics of rural organization: Theory, practice, Cheng, E., Findlay, C., & Watson, A. (1998). We’re not and policy. New York: Oxford University Press for financial organizations!: Financial innovation with- the World Bank. out regulation in China’s rural cooperative funds. Holcombe, S. (1995). Managing to empower: The Gram- MOCT–MOST: Economic Policy in Transition Econ- een bank experience of poverty alleviation. London: omies, 8(3), 41–55. Zed Books. 1506 WORLD DEVELOPMENT

Holz, C. A. (2001). China’s monetary reform: The Reserve Bank of India (RBI). (1954). All-India credit counterrevolution from the countryside. Journal of survey. Bombay: RBI. Contemporary China, 20(27), 189–217. RBI (2000). All-India debt and investment survey Hsiao, K. H. (1971). Money and monetary policy in com- (AIDIS). 1991–92––Incidence of indebtedness of munist China. New York: Columbia University Press. households, Part I. RBI Bulletin. Available: , February 8. (IFAD). (2001). People’s Republic of China: The- Rozelle, S., Park, A., Ren, C., & Bezinger, V. (1998). matic study on rural financial services in China. Rome, Targeted poverty investments and economic growth Italy. Available: . Rutherford, S., & Arora, S. S. (1997). City savers. New Jones, J. H. M. (1994). A changing financial landscape in Delhi: Department for International Development. India: Macro-level and micro-level perspectives. In Saich, T. (2000). Negotiating the state: The development F. J. A. Bouman & O. Hospes (Eds.), Financial of social organizations in China. The China Quar- landscapes reconstructed: The fine art of mapping terly, 161, 124–141. development. Boulder: Westview Press. Satish, P. (2001). Institutional alternatives for the Khandker, S. R., Khalily, B., & Khan, Z. (1995). promotion of microfinance: Self-help groups in Grameen bank: Performance and sustainability. India. Journal of Microfinance, 3(2), 49–74. World Bank Discussion Paper No. 306. Washington, Schrader, H. (1994). Moneylenders and merchant bank- DC: The World Bank. ers in India and Indonesia. In F. J. A. Bouman & O. Li, M. Y. (2000). Diandangye: ‘jinzi zhaopai’ xiexia Hospes (Eds.), Financial landscapes reconstructed: qianhou (Pawnshops: Future after removing the The fine art of mapping development (pp. 341–355). ‘‘gold store sign’’). Hexun caijing (Homeway Finan- Boulder: Westview Press. cial News), October. Selden, M.& Perry, E. (Eds.). (2000). Chinese society: Migdal, J. S., Kohli, A., & Shue, V. (1994). State power Conflict, change, and resistance. London: Routledge. and social forces: Domination and transformation in the Sinha, S. (2000). India. In The role of central banks in third world. New York: Cambridge University Press. microfinance in Asia and the Pacific: Country studies: Morduch, J. (2000). The microfinance schism. World Vol. 2. Asian Development Bank (pp. 61–89). Manila: Development, 28(4), 617–629. ADB. Nagarajan, G., & Meyer, R. L. (2000). Rural financial Swain, R. B. (2002). Credit rationing in rural India. markets in Asia: Paradigms, policies, and perfor- Journal of Economic Development, 27(2), 1–20. mance. Manila: ADB. Tsai, K. S. (2000). Banquet banking: Rotating savings National Bank for Agriculture and Rural Development and credit associations in South China. The China (NABARD). (2002). Ten years of SHG–bank linkage Quarterly, 161, 143–170. (1992–2002). Mumbai: NABARD. Tsai, K. S. (2002a). Back-alley banking: Private entre- NABARD (2003). Regional spread of SHGs as on preneurs in China. Ithaca: Cornell University Press. March 31, 2003. Available: . tions. The China Journal, 48, 1–27. Nayar, C. P. S. (1992). Strengths of informal financial Tsien, S. (2001). International projects left in the lurch as institutions: Examples from India. In D. W. Adams government weighs in. China Development Brief, & D. A. Fitchett (Eds.), Informal finance in low- 4(1). income countries (pp. 199–200). Boulder: Westview. Unger, J. (2002). The transformation of rural China. Otero, M.& Rhyne, E. (Eds.). (1994). The new world of Armonk: M.E. Sharpe. microenterprise finance. West Hartford, CT: Kumar- Walder, A.& Oi, J. (Eds.). (1999). Property rights and ian Press. economic reform in China. Stanford: Stanford Uni- Park, A. (1999). Banking for the poor. China Brief, II(2), versity Press. 9–15. Watson, A. (2003). Financing farmers: The reform of Park, A., Brandt, L., & Giles, J. (2003). Competition rural credit cooperatives and provision of financial under credit rationing: Theory and evidence from services to farmers. In C. Findlay, A. Watson, E. rural China. Journal of Development Economics, Cheng, & Z. Gang (Eds.), Rural financial markets in 71(2), 463–495. China (pp. 63–88). Canberra: Asia Pacific Press at Park, A., & Ren, C. (2001). Microfinance with Chinese The Australia National University. characteristics. World Development, 29(1), 39–62. Whiting, S. (2001). Power and wealth in rural China: The Patrick, H. T. (1966). Financial development and political economy of institutional change. New York: economic growth in developing countries. Economic Cambridge University Press. and Cultural Change, 14(2), 174–189. Wilson, K. (2002). The new microfinance: An essay on People’s Bank of China (PBC). (2004). Financial indus- the self-help group movement in India. Journal of try’s performance was stable in 2003. Available: Microfinance, 4(2), 217–245. , January 15. Wu, J. M. (1998). Local property rights regime in socialist Press Information Bureau (PIB). (2002). Government reform: A case study of China’s informal privatization. of India. Expert group on Nihdis recommends Unpublished doctoral dissertation. New York: continuation of regulatory measures. Available: Columbia University. , March 22. Shanghai: Shanghai wenyi chubanshe. INFORMAL FINANCE AND MICROFINANCE 1507

Yaron, J., Benjamin, M. P., Jr., & Piprek, G. L. (1997). China Online. Rural finance: Issues, design, and best practices. Jingji ribao (JJRB––Economic Daily). Environmentally and socially sustainable development People’s Daily (PD). studies & monographs series (Vol. 13). Washington, DC: The World Bank. Shanghai Information Center (SIC). Zhongguo yinhang Beijing guoji jinrong yanjiusuo Xinhua. (BOC Beijing Institute of International Finance). (1993). Zhongguo de jinrong jigou jiqi zhuyao jingying (China’s financial institutions and their primary management). Beijing: Zhongguo jihua chubanshe. FURTHER READING

Adams, D. W. & Fitchett, D. A. (Eds.). (1992). NEWS SOURCES Informal finance in low-income countries. Boulder: Westview Press. Agence France Presse (AFP). Asian Development Bank (ADB). (2000). The Associated Press (AP). role of central banks in Asia and the Pacific. China Daily. Manilla: ADB. China Internet Information Center (CIIC).