Introduction

Financial liberalisation has brought about a striking shift in banking policy in India and has led to changes in the regional and sectoral pattern of banking in the country. There has been (a) a large-scale closure of commercial bank branches in rural areas; (b) a widening of inter-State inequality in credit provision, accompanied by a fall in the proportion of bank credit directed towards rural areas as well as regions where banking has historically been underdeveloped; (c) a sharp fall in the growth of credit flow to agriculture followed by a revival in the 2000s, but with a changed pattern of distribution of agricultural credit in favour of urban-based borrowers and corporate and institutional groups; (d) the relative exclusion of the disadvantaged and dispossessed sections of the rural population from the formal financial system; and (e) the strengthening of moneylending in the countryside [see Ramachandran and Swaminathan (2005), Shetty (2004), Chavan (2005), Ramakumar and Chavan (2007), and Chavan (2010)].

These changes are associated with a reversal of the policy of social and development banking, a policy introduced after bank nationalisation in 1969. The policy of social and development banking aimed at expanding the geographical and sectoral reach of the banking system. Further, it also aimed at bringing the deprived sections of the population within the ambit of the banking system. After nationalisation, targets were fixed for opening bank branches in un-banked or under-banked regions. Further, targets were also fixed for lending at regulated rates of interest to priority sectors including agriculture and the “weaker sections,” comprising socio-economically backward sections of the population.

By contrast, the approach in the period of financial liberalisation has been to allow for a market-determined and market-oriented mode of operation of the banking system. The focus of policy has been on increasing the efficiency, narrowly defined, and profitability of the domestic banking sector in order to enable it to compete with the rapidly expanding foreign banking network in the country. As a result, interest rates on almost all categories of loans have been deregulated.¹ While the norms applicable to lending to priority sectors have been left untouched, the definitions of various priority sectors, particularly agriculture, have been widened considerably, leading to a dilution of the very notion of the “priority” attached to these sectors.²

It is in the period of financial liberalisation that the new term “financial inclusion” has been coined. Financial inclusion may appear similar in name to the policy of social and development banking, but is inherently different with respect to policy essentials. “Financial inclusion” has been defined as the provision of affordable financial services to those who have been left unattended or under-attended by the formal agencies of the financial system without compromising on commercial and profitability considerations in order to ensure the “long-term sustainability” of such services (RBI 2008, p. 204).

The recent approach to financial inclusion is more individual-specific than previous policy. It incorporates two main instruments: first, a no-frills basic saving bank deposit account facility (including a small overdraft facility); and secondly, a collateral-free small borrowal facility under microfinance at market-determined rates of interest (RBI 2006b). Both these instruments have been used rather rigorously by banks in the 2000s, either directly or through intermediaries. Banks’ intermediaries mainly include Business Correspondents (BCs) and Micro-Finance Institutions (including non-banking financial companies, trusts, and cooperative societies). Further, there has been a growing emphasis on the introduction of advanced technological solutions by banks and their intermediaries. These solutions include hand-held devices and mobile phones, measures intended to bring down the costs of administering the large numbers of small-volume transactions required under financial inclusion.³ The RBI also places emphasis on designing market-based regulatory incentives for banks for financial inclusion. The RBI has set targets of providing, by 2012, banking services through a banking outlet to every village with a population of over 2,000.⁴

Data Base

Dalits constitute an important segment of the rural poor in India. The policy of financial inclusion is expected to improve the access that rural Dalit households have to banking services in the period of financial liberalisation. In order to evaluate this claim, I analyse the trends in (a) the debt profile of rural Dalit households and (b) the flow of bank credit to Dalits since the early 1990s.

I use secondary data from the All-India Debt and Investment Surveys (AIDIS) in order to analyse the debt profile of rural Dalit households. The AIDIS provide data on the debt profile of rural households classified into various socio-economic groups, including Dalits. These data are available from the two most recent rounds of the survey, conducted in 1991-92 and 2002-03.⁵ Further, I use data from six sample surveys of Small Borrowal Accounts (SBAs) conducted by the RBI from 1993 onwards to examine changes in the supply of commercial bank credit to Dalits. These are sample surveys, and were conducted in 1993, 1997, 2001, 2004, 2006, and 2008.⁶

Debt Profile of Rural Dalit Households

According to the AIDIS, in 2002, more than half of the total debt outstanding of Dalit households in rural India was from informal sources (Table 1). The share of formal sources in the total debt of Dalit households was only 44.8 per cent, much lower than the corresponding share (59 per cent) for non-Dalit households. Among formal sources, the largest share of debt of Dalit households was owed to commercial banks, followed by cooperatives. Among informal sources, professional moneylenders were the single most important source of debt for these households.

An inter-round comparison of the AIDIS data from 1962 onwards shows that, with regard to the share of formal sources in the total debt of all rural households, there was a distinct break in the overall trend after 1992. The share of formal sources, commercial banks in particular, rose steadily between 1962 and 1992, and then fell between 1992 and 2002 (Table 2). The rise in the share of formal sources was particularly striking between 1972 and 1982, the period following the establishment of the policy of social and development banking.⁷

Separate data on Dalit households are not available from the AIDIS rounds before 1992. Nevertheless, the data do indicate that Dalit households in rural areas gained new and often unprecedented access to formal sector credit. The Integrated Rural Development Programme, which “channelled funds on a hitherto unprecedented scale for creating supplementary incomes amongst the relatively poor in rural areas all over India” (Guhan 1986) – was the most important means of such access.⁸

Between 1992 and 2002, as was the case with all rural households, the share of formal sources in the total debt of rural Dalit households declined. However, the decline for Dalit households was greater than for non-Dalit households (Table 1). Debt from formal sources as a percentage of the total debt came down by about 16 percentage points between 1992 and 2002 for Dalit households as compared to five percentage points for non-Dalit households.⁹ Thus, in the 1990s, Dalits suffered more than others from the cutbacks in formal credit to rural areas.

AIDIS data also show that, between 1992 and 2002, the number of Dalit households reporting at least one outstanding loan from formal sources fell by about five percentage points. The corresponding fall for non-Dalit households was only about one percentage point (Table 3).

As outstanding debt is a stock variable, it misses out on loans that are taken during the survey year and settled prior to the date of survey. Hence, it is also important to look at the incidence of fresh loans taken during the survey year. The percentage of Dalit households that reported at least one fresh loan from the formal sector during the survey year was 4.7 per cent in 2002-03 as compared to 7.6 per cent in 1991-92 (Table 4). At the same time, there was a rise between 1991-92 and 2002-03, by about four percentage points, in the number of Dalit households that reported at least one new borrowing from informal sources.

The fall in the share of debt from formal sources among Dalit households between 1992 and 2002 was attributable to a sharp fall in debt outstanding to commercial banks rather than from other formal sources. Of the total reduction of about 16 percentage points in the share of debt from formal sources between 1992 and 2002, about 13 percentage points were on account of commercial banks (Table 1). In 1992, commercial banks were the largest source of debt for Dalit households; in 2002, they were replaced by professional moneylenders. Between 1992 and 2002, the share of rural Dalit households that owed at least one loan to commercial banks fell by about three percentage points (Table 3). During the corresponding period, there was a rise of a little over four percentage points, from 3.2 per cent to 7.7 per cent, in the share of Dalit households that reported at least one loan outstanding from professional moneylenders.

When households were classified by the rates of interest at which they borrowed, it was clear again that informal sources predominated in Dalit debt portfolios (Table 5). Between 1992 and 2002, there was a decline in the share of debt taken at annual interest rates ranging between 6 and 15 per cent and a rise in the share of debt taken at interest rates above 20 per cent. Another disturbing feature was a rise between 1992 and 2002 in the share of total debt taken at compound rates of interest by Dalit households (Table 6). In the same period, the proportion of interest-free debt and debt taken at concessional rates of interest declined. These findings reflect the rise in the percentage of debt taken from informal sources, especially professional moneylenders.

The AIDIS data show that the fall in the share of formal sources in the total debt of rural Dalit households between 1992 and 2002 occurred in every State of India other than Maharashtra and Himachal Pradesh (Table 7). In Maharashtra, the rise in the share of formal sources of debt was mainly on account of cooperatives and not commercial banks.

The AIDIS data indicate that cooperatives were less important as a source of credit than commercial banks for rural Dalits. In 2002, credit cooperatives had a share of 18.3 per cent in the total debt of rural households, and commercial banks a share of 21.6 per cent. Nevertheless, there was some increase in the share of cooperatives, from 15 per cent to 18.3 per cent, between 1991-92 and 2002-03, a period in which the share of commercial banks declined significantly, from 34.6 per cent to 21.6 per cent (Table 1).

Supply of Commercial Bank Credit to Dalits

Credit given by commercial banks to Dalits is included in banks’ advances to “weaker sections.” Domestic commercial banks (public and private sector banks, but not foreign banks) have been required to direct at least 10 per cent of their (adjusted net) bank credit to “weaker sections” as part of priority sector lending. This sub-target for “weaker sections” was introduced by the RBI in 1980 as part of its social and development banking policy.¹⁰ Weaker sections were then defined as small and marginal farmers (having land holding of less than 5 acres), landless agricultural labourers, and tenant farmers, in addition to Dalits and Adivasis.

Over the years, new groups have been included as part of the definition of “weaker sections,” including members of minority communities engaged in specified occupations. More recently, in the 2000s, loans to Self-Help Groups have also been included as part of credit to “weaker sections.” All these new sub-categories have been clubbed together, thus effectively reducing the importance that each of these sub-categories individually deserves as a weaker section.

Despite the fact that the category of weaker sections has been broadened in the recent years, banks have consistently failed to meet the prescribed target of lending (Table 8). At the aggregate level, there has been an upward movement in the share of credit given to weaker sections (that is, of adjusted net bank credit) since 2007 onwards for both public and private sector banks. However, the picture at the aggregate level concealed the disturbing trends at the disaggregated level. First, at the bank group-level, the performance of private sector banks has been poor when compared with public sector banks. The credit allocation to weaker sections by private sector banks has been far lower than the prescribed target. Secondly, at the bank level, the performance has been even poorer. In every year between 2006 and 2009, about 40-70 per cent of public sector banks failed to meet the target. Further, over 80 per cent of the private sector banks failed to meet the target in each of these years. In fact, in 2006-07 and 2007-08, no private sector bank met the target. Even in 2010, when the percentage of credit given to weaker sections at the aggregate level was at its highest level since 1993, 15 out of 22 private sector banks (68 per cent) and 8 out of 27 public sector banks (30 per cent) failed individually to meet the target for lending to weaker sections (RBI 2010a).¹¹

An analysis of the Small Borrowal Accounts (SBAs) shows even more disturbing trends as regards the access of Dalits to bank credit in recent years. SBAs are accounts with an individual credit limit of up to Rs 2 lakh (the limit was Rs 25,000 till 1998). In 2008, about 56 per cent of the total SBAs were held with rural and semi-urban branches of commercial banks and about 37 per cent of these accounts were for direct finance under agriculture and allied activities (Table 9). Thus, credit flow through SBAs forms an important part of credit flow to rural areas.

The survey of SBAs has shown a sharp fall in the percentage share held by Dalits in the total number of small borrowal accounts as well as in the total amount of bank credit outstanding between 1993 and 2008. In 2008, both these shares were only about one-sixth of their corresponding shares in 1993 (Table 10).

On an average, there were only 16 small borrowal accounts per 1,000 Dalits in the population in 2008 as compared to 77 in 1993 (Table 11).¹² Further, credit per capita (adjusted for price change) for Dalits fluctuated around a declining trend across the survey years, recording a particularly sharp fall between 2006 and 2008.¹³ The amount of credit per capita for non-Dalits was nearly seven times the credit received by Dalits through SBAs in 2008. The gap between credit per capita for Dalits and non-Dalits widened substantially between 1993 and 2008.

Caste differences were aggravated by gender differences. In 2008, Dalit women, on an average, received only about Rs. 4 of bank credit per SBA for every Rs. 100 received by non-Dalit and non-Adivasi women (Table 12). In 2008, Dalit women obtained less than one rupee of credit per SBA for every Rs. 100 received by non-Dalit and non-Adivasi men. Further, the average amount of credit per account going to Dalit women vis-à-vis women and men from non-Dalit/Adivasi categories was on a rapid decline between 1997 and 2008.

Concluding Observations

According to the All-India Debt and Investment Survey, for Dalit households, commercial banks were the most important source of credit in 1992. There was, however, a sharp fall in the share of debt from commercial banks between 1992 and 2002. The vacuum thus created was filled in by professional moneylenders. While professional moneylenders did emerge in 2002 as an important source of credit for other rural households as well, their hold over Dalit households was much stronger. The expansion of informal debt meant an increased and onerous interest burden on rural Dalit households.

Data from the supply side indicated a continued failure by domestic banks to meet the targets set for credit to “weaker sections” since the early 1990s. In the 1990s and 2000s, the proportion of commercial bank credit to Dalits through Small Borrowal Accounts also recorded a sharp fall.

To conclude, available secondary data show an increasing exclusion of rural Dalits from the early 1990s with respect to access to affordable formal sector credit. This also explains the growing grip of informal sources on their debt portfolio in recent years. Hence, similar to the imbalance in credit flow emerging at the sectoral and regional levels, credit allocation across socio-economic groups too has become increasingly unequal in the period of financial liberalisation notwithstanding the emphasis on financial inclusion.

The exclusion of the disadvantaged and dispossessed sections is intrinsic to the functioning of markets.¹⁴ There is, therefore, an inherent conflict between allowing the banking system to be driven by market forces and expecting greater inclusion from the system. The growing exclusion of rural Dalits from formal credit is a sign of this conflict. It underlines the need to revive the policy of social and development banking, with a stronger commitment than before to the objective of social redistribution.

Acknowledgement: The views expressed in this paper are personal views of the author and do not represent the views of the organisation to which she belongs. The author thanks V. K. Ramachandran, Madhura Swaminathan, S. L. Shetty, Balwant Singh, and R. Ramakumar for useful comments and suggestions on an earlier draft of this paper.