by Sambhavi Dhingra, Arpita Agarwal, and Snehal S. Herwadkar^ Priority sector lending (PSL) in India has been used as a policy intervention tool to direct credit to the needy sectors of the economy. Leveraging quarterly bank-level data from March 2006 to March 2023, the empirical analysis suggests that the introduction of priority sector lending certificates increased priority sector lending. The analysis further suggests that PSL is responsive to its asset quality and higher PSL growth appears to improve banks’ overall asset quality. Introduction Commercial banking forms a dominant part of India’s financial system, with an ever-widening scope and reach. The prevalence of informal credit institutions and unequal access to banking services necessitated initiating financial inclusion policies. In India, priority sector lending (PSL) falls under the overall structure and plan of furthering the inclusion objective. Primarily, sectors of the economy that impact large sections of the population and are employment-intensive fall under the priority sector umbrella (RBI, 2007). These mainly constitute loans for agriculture and allied activities, micro and small enterprises (MSE), housing, exports, education, and priority sector loans to weaker sections. PSL guidelines in India have been reviewed and revised periodically to align them with emerging national priorities and bring a sharper focus on inclusive development. The aim of this article is to examine the trends of the PSL programme in India, while evaluating its major drivers. It also gauges the impact of PSL on the asset quality of banks. The rest of the article is arranged as follows. Section II provides an overview of literature on the rationale for directed lending and the cross-country experience. Section III outlines the evolution of PSL in India and the performance so far. Section IV presents the research methodology along with the empirical analyses, while section V concludes the article. II. Review of Literature The prevalence of high transaction costs and information asymmetries can restrict credit flows to the relatively risky but economically productive activities and borrowers. In the absence of complete information about expected project returns, banks tend to base their lending decisions on observable risk characteristics and/or the availability of good collateral, keeping credit flowing to the traditionally viable borrowers. This risk of underserving such projects, which already have marginalised access to credit, can prompt governments to implement PSL-type programmes (Mundra, 2017). The problem of asymmetric information curtailing provision of credit becomes even more pronounced in the agriculture sector. Agriculture plays a critical role in economic development by generating employment, ensuring food security, and alleviating poverty, especially in emerging market economies (EMEs) like India. However, given the inherent uncertainty in agricultural yield and prices due to their heavy dependence on rainfall and recurrent climate shocks, farmer borrowers can often face restricted access to credit (Calomiris and Himmelberg, 1993). In the absence of formal credit, it becomes even more essential to bring them under the wing of directed lending programmes (Chakrabarty, 2012). Similarly, small-scale industries are often ‘informal’ in nature and the lack of adequate documentation is a major constraint in financing them. These industries, that usually have low capital-output ratio and are labour intensive, may not necessarily be amongst the preferred clientele of commercial banks and, therefore, could be credit-starved. Directing credit to such industries is especially beneficial in labour-surplus developing economies, as it generates employment and reduces income inequality (Kohli, 1997). There are divergent views on efficacy of directed lending, particularly regarding its commercial viability. For instance, a study conducted in Tamil Nadu for 1984-86 showed that nationalised banks, which lent heavily to agriculture, participated substantially in government loan schemes, and directed significant amounts to borrowers from weaker sections and scheduled castes/ scheduled tribes (SC/ST), had the highest recovery rates (Narayana, 1992). PSL does not adversely impact banks’ non-performing assets (NPA) ratios (Gaur and Mohapatra, 2020). On the other hand, according to some studies, labour and administrative costs associated with lending to the priority sector were higher than that of lending in the unreserved sector for four Indian public sector banks (PSBs), and such loans contributed more than proportionately to the NPAs of banks (Banerjee and Duflo, 2000; Banerjee and Duflo, 2014). The potential macroeconomic effects of priority lending have elicited a wide range of responses in existing literature. Under certain conditions, government intervention in the form of directed credit programmes would not only usher in financial development, but also provide important guidelines in ensuring sustainability of institutions (Chakrabarti et al., 2019). However, another study shows that small firms sacrificed their expansion to have access to priority sector lending (Bhue et al., 2019). By enhancing employment opportunities and promoting social equity, directed credit for priority sectors also bears a spillover impact for non-priority sectors, ensuring balanced economic growth (Gaur and Mohapatra, 2020). Such programmes play a significant role in addressing the welfare objectives of reducing income inequality and dependence on informal credit, as well as poverty alleviation (Federal Reserve Bank of San Francisco, 2014). As Muhammad Yunus (1987) commented, if the implementation of PSL is faulty and NPAs are high, “one should not be quick to blame the people of the recipient country for the failure; rather one should blame the designer of the credit institution that failed to do the job.” The mechanisms of directing credit take various forms, such as interest subsidies, interest rate caps, direct credit by government, government guarantees, lending quotas for banks, lending through development finance institutions (DFIs) or a combination of these methods. Some form of PSL exists in many countries, especially EMEs. Indonesia, for instance, directs priority lending to small and medium enterprises (SMEs), with quota of 20 per cent of its total portfolio (ILO, 2019). Malaysia and Vietnam implemented interest rate discounts for lending to priority sectors (Federal Reserve Bank of San Francisco, 2014). The policy to direct credit was an important factor contributing to strong economic performance in the East Asian countries (World Bank, 1993). The Japan Development Bank was instrumental in increasing incremental lending from private banks, providing improved access of credit to new firms and in generating new investment in post-war Japan (Horiuchi and Sui, 1993). In Korea, government-led direction of credit helped overcome pervasive market imperfections and channelled new borrowings into investments, which led to further economic growth (Cho and Hellman,1993; Werner, 2002). French loan guarantee program significantly impacted the development of newly created firms and enabled the targeted firms to systematically raise more external finance, pay lower interest expenses, and enjoy higher growth rates than other similar firms (Lelarge et al., 2010). In the US, while directed lending programmes have succeeded in increasing credit to the targeted group, they have not necessarily led to an increase in investment by that group (Schwarz, 1992). In some cases, inefficient implementation has also led to increased income inequality instead of promoting a more equitable distribution of resources, like in Costa Rica, where subsidised credit increased income accruing to the wealthiest 10 per cent of the population instead of the target group (Vogel, 1984). In many countries, these programmes proved to be particularly costly for the banking industry in terms of high NPAs, lower profitability, and higher variable costs involved in accomplishing targets, along with bearing moral hazard concerns. In Indonesia, the banking system’s SME loan portfolio exhibited inferior asset quality compared to the aggregate portfolio (Federal Reserve Bank of San Francisco, 2014). In South Korea and Japan, directed lending programmes are generally viewed to have been efficaciously implemented, with multiple factors being responsible for their success. Interventionist policies in these countries have been supplemented by appropriate institutional mechanisms. In Japan, once a firm got priority access to credit, they were restricted from further borrowing under the PSL and credit was extended to new borrowers. In Korea, if performance standards were not met by the beneficiaries, loans were either recalled or new credit was denied. Strict performance standards, accompanied with effective monitoring mechanisms, ensured the success of these programmes. To summarise, the available literature presents a range of views on the efficacy of directed lending programmes. The analysis undertaken in the article adds to the existing literature in the following ways. First, while most of the earlier papers in the Indian context were either theoretical or case studies, this article empirically evaluates the drivers of PSL. Second, the literature on the impact of PSL on asset quality of banks is scanty, especially for India; this article fills this gap. Third, it uses bank-level quarterly supervisory data, which adds more granularity to the analysis. III. Priority Sector Lending in India III.1 History and Evolution Priority sector lending in India has been a mainstay of credit control policies since the nationalisation of banks in 1969. In 1972, the description of priority sector and areas that qualified as such were formalised based on the recommendations of an informal study group of the Reserve Bank, and in 1974, a target of 33.33 per cent of total credit was fixed for the same, to be achieved by 1979. The target was further enhanced to 40 per cent in 1980, along with specific sub-targets for lending to agriculture and weaker sections, to be achieved by 1985. Since then, the guidelines have undergone further changes in terms of varying applicability to different types of banks, quantum of credit, targeted sectors and sub-targets, and treatment of shortfall by banks. Since April 2007, PSL requirement is being specified as a per cent of bank’s adjusted net bank credit (ANBC)1 or credit equivalent of off-balance sheet exposures (CEOBE), whichever is higher. In its present form, the PSL guidelines require domestic commercial banks [excluding regional rural banks (RRBs), and small finance banks (SFBs)] and foreign banks (FBs) to lend 40 per cent of their ANBC or CEOBE, whichever is higher, to the priority sector.2 Out of the total target, 18 per cent is prescribed for agriculture [10 per cent for small and marginal farmers (SMFs)3], 7.5 per cent for micro enterprises, and 12 per cent for weaker sections.4 The scope of PSL has gradually been widened to incorporate sectors that have assumed prominence in contemporary times. Accordingly, credit extended for promoting social infrastructure and renewable energy was included in PSL in 2015. COVID-19 led to a refocus of priorities, and the importance of investing in health and education infrastructure, along with increasing access to digital services, came to the fore. According to the latest PSL guidelines, loans up to ₹5 crore per borrower for setting up schools, drinking water and sanitation facilities, and loans up to ₹10 crore per borrower for building health care facilities (including under ‘Ayushman Bharat’) in Tier II to Tier VI centres are eligible for priority sector classification.5 III.2 Achievement of Targets and Sub-Targets Depending upon their risk profile and existing clientele, banks may resort to indirect routes, such as inter-bank participation certificates (IBPCs) and securitisation of priority sector loans, as well for achieving priority sector target and sub targets. In addition, PSL certificates (PSLCs) were introduced in April 2016, as recommended by the Raghuram Rajan Committee on Financial Sector Reforms (2009), to enable banks to achieve the PSL target and sub-targets in the event of a shortfall while incentivising surplus lending to the categories under priority sector. This also provides banks trading in PSLCs the advantage of specialising in their area of expertise and disbursing loans more efficiently. Trading of PSLCs is allowed for the four categories that have mandated targets under the Reserve Bank’s guidelines - agriculture, SMF, micro enterprises and general. Banks that still have shortfalls are required to contribute to the RIDF and other funds with NABARD/ NHB/ SIDBI/ MUDRA Ltd. Lending to the priority sector has generally remained above 40 per cent across time periods and bank groups, and the exact proportion is contingent upon, inter alia, the bank’s overall business strategy, reach, asset quality of such loans, and their expertise (Chart 1). In case of agriculture, PSBs, on most occasions, have fulfilled their target of 18 per cent; private sector banks (PVBs), which were earlier consistently short of the target, have in recent years aligned to the target. On the other hand, PVBs have fared better than PSBs in achieving the sub-target of lending 7.5 per cent of their ANBC or CEOBE to micro enterprises. Both PSBs and PVBs have met their targets for lending to weaker sections, with PSBs ahead of their private sector counterparts (Chart 2). IV. Empirical Analysis The empirical analysis is divided into two parts. First, despite the common regulatory mandate requiring banks to allocate 40 per cent of their ANBC/ CEOBE to the priority sector, significant variations exist among different banks. This section of the study investigates the potential factors that can influence the extent of banks’ PSL. Second, the study examines the impact of PSL on banks’ overall asset quality. IV.1 Drivers of Priority Sector Lending An exercise to determine the factors that affect PSL by banks is undertaken using fixed effects panel regression models. Quarterly data of PSBs and PVBs are used for the period March 2006 to March 2023. All the data are sourced from supervisory returns of the Reserve Bank. The following panel regression equation is estimated:  where PSL shareit is the share of priority sector loans in gross loans and advances6 for bank i at time t, PSL shareit–1 refers to the one period lagged value of PSL share, PSL GNPA ratioit–1 is the lagged asset quality indicator for priority sector loans, Branches to assets ratioit is a proxy for bank’s reach, and log(assets)it–1 is an indicator of the bank’s size. The PSLC dummy takes value 1 for all quarters starting June 2016 to capture the impact of the introduction of PSLCs. As the PSL shortfall calculations are based on data at the end of every financial year, a March dummyt is introduced, which takes value 1 for all quarters ending March7. This is to account for the possibility that banks may be accelerating their PSL in the last quarter to achieve their annual targets. αy are year fixed effects, ϑi are bank fixed effects8, and εit are standard errors clustered at bank level and adjusted for heteroscedasticity. Three alternate specifications are evaluated, with the dependent variables being the entire PSL share in total credit, PSL to agricutlure to total credit and PSL to MSEs9 to total credit. Results of the empirical estimation suggest that the asset quality of the priority sector portfolio plays a significant role in determining the PSL share of banks. Although PSL is mandated by regulatory requirements, banks take into consideration the usual risk-return trade-off when extending these loans. Wider bank reach, measured by bank branches-to-assets ratio, is positively associated with higher share of loans disbursed to the priority sector. Banks with greater brick-and-mortar presence are better placed to extend priority credit at grass-roots level. In the agriculture and MSE PSL specifications, rural branches-to-assets ratio and urban branches-to-assets ratio, respectively, have been used as explanatory variables. Results indicate that banks with bigger branch network in urban areas lend a greater share of their loans to priority MSEs. However, the results are not significant for priority sector agricultural loans for banks with a higher rural presence, possibly reflecting the presence of RRBs, SFBs and rural co-operative banks in these areas, which have greater expertise in agriculture financing. Bank size has a significant and negative impact on the PSL share. | Table 1: Determinants of PSL Share | | Variables | PSL share | Agriculture PSL share | MSE PSL share | | | (1) | (2) | (3) | | PSL share (L1) | 0.678***
(0.0435) | | | | Agriculture PSL share (L1) | | 0.810***
(0.0462) | | | MSE PSL share (L1) | | | 0.628***
(0.0486) | | PSL GNPA ratio (L1) | -0.225***
(0.0609) | | | | Agriculture PSL GNPA ratio (L1) | | -0.0971***
(0.0201) | | | MSE PSL GNPA ratio (L1) | | | -0.200***
(0.0386) | | Branches to assets | 0.436*
(0.251) | | | | Rural branches to assets | | -0.0572
(0.253) | | | Urban branches to assets | | | 1.340***
(0.458) | | Log (assets) (L1) | -1.453**
(0.533) | -0.636***
(0.220) | -0.580*
(0.292) | | PSLC dummy | 1.606**
(0.658) | -0.139
(0.390) | 1.603***
(0.425) | | March dummy | 1.465***
(0.346) | 0.471**
(0.184) | 1.149***
(0.164) | | Constant | 25.83***
(6.717) | 9.049***
(2.436) | 8.411**
(3.324) | | Observations | 2,162 | 2,162 | 2,161 | | Adjusted R-squared | 0.626 | 0.762 | 0.698 | | Number of banks | 33 | 33 | 33 | | Bank fixed effects | Yes | Yes | Yes | | Year fixed effects | Yes | Yes | Yes | Notes: 1. Figures in parentheses indicate robust standard errors clustered at bank level.
2. ***, ** and * represent 1 per cent, 5 per cent and 10 per cent levels of significance, respectively.
Source: Authors’ calculations. | The share of PSL in the overall credit by banks has gone up since the introduction of PSLCs, which has helped certain banks to develop a niche in specific PSL segments. Data suggest that such banks lend over and above the regulatory minimum to their specialised segments and convert these excess achievements into PSLCs, which they trade for a premium. The statistically significant and positive coefficient of the PSLC dummy indicates that its introduction has helped banks improve their overall and MSE PSL share. Peaking of PSL share in the March quarter is also captured in the model. IV.2 Asset Quality of PSL Historically, loans originating from priority sectors have had higher NPAs than their non-priority sector counterparts, of which the majority have been on PSBs’ books. However, the trend reversed in 2015, in part due to better recognition of NPAs after the asset quality review (John et al., 2016)10 [Chart 3]. Empirical evidence suggests that GNPAs of banks depend on bank specific and macroeconomic variables (Chavan and Gambacorta, 2016). In addtion to this, the focus of our estimation is to test whether banks’ PSL has an impact on their overall asset quality. For this, an exercise was undertaken using panel regression models on the same dataset used in the earlier subsection. where the dependent variable is the GNPA ratio of bank i at time t. PriorityLoanGrowthit–1 is the one period lagged year-on-year (y-o-y) growth in a bank’s PSL. AQRt is a dummy variable for asset quality review that takes value 1 for the quarters between September 2015 and March 2018, and 0 otherwise. Xit–1 are bank-level controls, inclusive of a dummy for bank group, lagged return on assets, and lagged log of bank size (sum of loans and deposits). αy are year fixed-effects, ϑi are bank fixed effects11, and εit are standard errors clustered at bank level and adjusted for heteroscedasticity.  | Table 2: Impact of PSL Growth on Bank’s Asset Quality | | Variables | Dependent Variable: GNPA Ratio | | (1) | (2) | (3) | (4) | | Priority advances growth (L1) | -0.0254*** | -0.0483*** | -0.0210*** | -0.0226*** | | | (0.00457) | (0.00841) | (0.00382) | (0.00455) | | AQR dummy | 0.757*** | 0.539*** | 0.742*** | 9.751*** | | | (0.185) | (0.153) | (0.180) | (1.667) | | Return on assets (L1) | -3.188*** | | -3.187*** | -3.238*** | | | (0.327) | | (0.308) | (0.319) | | Log (size) (L1) | -1.860*** | | -3.338*** | -3.401*** | | | (0.383) | | (0.684) | (0.711) | | Bank group dummy | 5.200*** | | | | | | (1.264) | | | | | Constant | 24.82*** | 4.205*** | 42.68*** | | | | (3.864) | (0.597) | (7.253) | | | Observations | 2,030 | 2,030 | 2,030 | | | Bank fixed effects | No | Yes | Yes | Yes | | Year fixed effects | Yes | Yes | Yes | No | | Time fixed effects | No | No | No | Yes | | Adjusted R-squared | 0.534# | 0.465 | 0.713 | 0.716 | | Number of banks | 33 | 33 | 33 | 33 | Notes: 1. Figures in parentheses indicate robust standard errors clustered at bank level.
2. ***, ** and * represent 1 per cent, 5 per cent and 10 per cent levels of significance, respectively.
3. #: Overall R-squared has been reported.
Source: Authors’ calculations. | As expected, primary results from the regression indicate that the GNPA ratios significantly increased during the AQR period. Results also indicate that higher growth in priority sector advances dampen the GNPA ratios of banks. In column (1) of Table 2, it is found that the coefficient of the Bank group dummy, which takes value 1 for PSBs and 0 for PVBs, is positive and significant as PSBs had higher GNPA ratios as compared to PVBs during the period under study. In column (2), bank fixed effects are added and the results remain robust to their inclusion. In column (3), other bank-level controls are added and it is found that asset quality improves with higher profitability and larger size of banks. Further, in column (4), even after adding time fixed effects to control for all exogenous time varying factors, the results remain robust. V. Conclusion Lending programmes like PSL have been put in place to enhance formal credit availability to the needy sectors. Cross-country literature on directed lending remains inconclusive on the impact it has on banks’ health. In one view, such lending can be successful when supplemented by appropriate institutional mechanisms, strict performance standards and policy framework. According to the opposing view, these loans, being mandated by regulatory requirements, may not align with the banks’ business interests and could potentially harm their overall asset quality, raising questions about their commercial viability. This article empirically evaluates these arguments. Using bank-level PSL data from March 2006 to March 2023, the study finds that the share of priority sector loans in banks’ total loan portfolio depends, inter alia, on the asset quality of such loans. Further, the introduction of PSLCs played a pivotal role in helping banks develop a niche in certain priority sectors, and consequently, increasing lending to these. The empirical analysis also suggests that high growth in PSL is not associated with a deterioration in banks’ asset quality. References Banerjee, A., and Duflo, E. (2000). Efficiency of Lending Operations and the Impact of Priority Sector Regulations. MIMEO, MIT. Banerjee, A., and Duflo, E. (2014). Do Firms Want to Borrow More? Testing Credit Constraints using a Directed Lending Program. The Review of Economic Studies, 81(2), 572–607. doi: 10.1093/restud/rdt046 Bhue, G., Prabhala, N., and Tantri, P. (2019). Can Small Business Lending Programs Disincentivize Growth? Evidence from India’s Priority Sector Lending Program. Indian School of Business. Calomiris, C. W., and Himmelberg, C. P. (1993). Directed Credit Programs for Agriculture and Industry: Arguments from Theory and Fact. The World Bank Economic Review, 7(1), 113-138. https://doi.org/10.1093/wber/7.suppl_1.113 Chakrabarty, K. C. (2012). Revised Guidelines on Priority Sector Lending: Rationale and Logic. Retrieved from https://www.bis.org/review/r120906b.pdf Chakrabarti, D., Sethi, P., and Bhattacharjee, S. (2019). Directed Credit, Financial Development and Financial Structure: Theory and Evidence. Applied Economics, 51(16), 1711-1729. https://doi.org/10.1080/00036846.2018.1528338 Chavan, P., and Gambacorta, L. (2016). Bank Lending and Loan Quality: The Case of India. BIS Working Papers, 595. https://www.bis.org/publ/work595.pdf Cho, Y. J., and Hellman, T. (1993). The Government’s Role in Japanese and Korean Credit Markets: A New Institutional Economics Perspective. The World Bank, Policy Research Working Papers Series, 1190. https://ideas.repec.org/p/wbk/wbrwps/1190.html Federal Reserve Bank of San Francisco. (2014). Priority Sector Lending in Asia. Retrieved from https://www.frbsf.org/banking/wp-content/uploads/sites/5/Asia-Focus-Priority-Sector-Lending-in-Asia-September-2014.pdf Gaur, D., and Mohapatra, D. R. (2020). The Nexus of Economic Growth, Priority Sector Lending and Non-Performing Assets: Case of Indian Banking Sector. South Asian Journal of Business Studies. https://www.emerald.com/insight/2398-628X.htm Horiuchi, A., and Sui, Q.-Y. (1993). The Influence of the Japan Development Bank Loans on Corporate Investment Behaviour. Journal of the Japanese and International Economies, 7(4), 441-465. doi:10.1006/ jjie.1993.1025 ILO. (2019). Financing Small Businesses in Indonesia: Challenges and Opportunities. International Labour Organization. John, J., Mitra, A.K., Raj, J., and Rath, D.P. (2016). Asset Quality and Monetary Transmission in India. Reserve Bank of India Occasional Papers, 37(1and2). Available at https://rbidocs.rbi.org.in/rdocs/Publications/PDFs/RBIOP37020520185C7D2C48AFA9427993B773BD194D 95A2.PDF Kohli, R. (1997). Directed Credit and Financial Reform. Economic and Political Weekly, 32(42), 2667-2676. http://www.jstor.com/stable/4405977 Lelarge, C., Sraer, D., and Thesmar, D. (2010). Entrepreneurship and Credit Constraints: Evidence from a French Loan Guarantee Program. International Differences in Entrepreneurship, 243-273. https://ideas.repec.org/s/nbr/nberch.html Mundra, S. S. (2017). Priority Sector Lending – Status, Issues and Future Agenda. Retrieved from https://www.bis.org/review/r170724f.pdf Narayana, D. (1992). Institutional Credit for Rural Development: Proper Risk-Management of Group Lending? Economic and Political Weekly, 27(39), A122-A127. https://www.jstor.org/stable/4398941 RBI. (2007). Master Circular- Lending to Priority Sector. Available at https://rbidocs.rbi.org.in/rdocs/notification/PDFs/78389.pdf Schwarz, A. M. (1992). How Effective are Directed Credit Policies in the US? The World Bank, Policy Research Working Paper Series, 1019. https://ideas.repec.org/p/wbk/wbrwps/1019.html Vogel, R. C. (1984). The Effect of Subsidised Agricultural Credit on Income Distribution in Costa Rica. In D. W. Adams, D. H. Graham, and J. D. Pishke (Eds.), Undermining Rural Development with Cheap Credit. New York: Routledge. Werner, R. A. (2002). A Reconsideration of the Rationale for Bank Centered Economic Systems and the Effectiveness of Directed Credit Policies in the Light of Japanese Evidence. Japanese Economy, 30(3), 3-45. https://doi.org/10.2753/JES1097-203X30033 World Bank. (1993). The East Asian Miracle. Available at https://documents1.worldbank.org/curated/en/975081468244550798/pdf/multi-page.pdf Yunus, M. (1987). The Poor as the Engine of Development. The Washington Quarterly, 10(4), 139–145. https://doi.org/10.1080/01636608709477624
|