Research: Working Papers and Works-in-Progress

Financial Access and Gender Gap in Entrepreneurship and Financial Inclusion: Evidence from Rural India (with Sandhya Garg) IEG Working Paper No. 442 [Presented at Winter School, Delhi School of Economics, December 2020; Presented at Annual STEG Conference, January 2021]

Abstract: Can expansion of bank branch network reduce gender-gap in entrepreneurship and financial inclusion in Indian rural economy? To explore this issue, we construct a novel village-level panel data where we observe the distance of each un-banked village from its nearest village/town with bank branch; and village-level enterprise data of three economic census rounds of 1998, 2005 and 2013. To account for endogenous placement of bank branches, we use a difference-in-difference methodology. Presence of a bank branch within 5km of an un-banked villages between 2005 and 2013 (Treatment Group) mitigates the gender gap in access to formal finance, increases female entrepreneurship and shifts male entrepreneurship from agricultural to non-agricultural sector. The increase in number of female-owned enterprises is fully driven by the increase in female-ownership in non-agricultural sector. Our results are robust to unobservable village and year effects, and presence of alternative village-level infrastructure.

What determines Enterprise borrowing from Self-Help Groups? A Predictive Assessment using Supervised Learning Algorithms (with Madhura Dasgupta)

Abstract:  Despite several advantages of borrowing from Self Help Groups (SHGs), why does enterprise borrowing remain low from this source compared to informal channels and self-nancing, in India (MoSPI, 2020)? To answer this question, we develop a novel village-enterprise matched dataset and apply a wide array of supervised learning methods which predict an enterprise’s choice between SHG vis-a-vis informal credit as the major source of nance. With an AUC of 94%, Xtreme Gradient Boosting (XGBoost) model (Chen and Guestrin, 2016) performs the best. We conduct perturbation and counterfactual analyses to interpret the predictions of our model (Molnar, 2021). Distance from urban areas and socio-demographic factors such as literacy rates and sex ratio in a village are amongst the most important contributors in determining SHG loan pick-up. In contrast, village-level infrastructure such as roads and financial access points like commercial banks play a much smaller role in explaining demand for credit from SHG. We also extend the model to estimate the potential demand for SHG loans
among the universe of firms outside any financial net, that is, the self-financed firms. The potential for SHG inclusion among these firms remains low and skewed toward the southern districts. The paper contributes to the existing literature by highlighting a need for reorientation of financial inclusion strategies.


Financial Access of Unbanked Villages in India from 1951-2019: A Spatial Approach (with Sandhya Garg), IEG Working Paper No. 403 [Revise and Resubmit]

Abstract: We develop a novel measure of financial access at the village level by finding the Euclidean distance of unbanked villages to the nearest village or town with bank branch for each year from 1951-2019. We use this measure of financial access to evaluate different bank branch expansion policies over last seven decades. Particularly, we assess how proximity to banks changes in four different regimes of bank branch expansion—pre-Social Banking Phase (1950-1969), Social Banking Phase (1969-1990), Liberalization Period 1 (1990-2005) and Liberalization Period 2 (2005 onwards). We find that social banking policy led to a rapid decline in distance and, thereby, increased financial access. These gains became restricted from 1990 to 2005 as the policy of mandatory quotas on bank branch opening was withdrawn. However, financial access improved again from 2005 onwards when RBI introduced incentive-led policies for bank branch expansion. The results suggest that sound, predictable and incentive driven methods can provide both efficiency and equity of public service provision. The possibility to replicate our measure of financial access in other areas of policy is discussed in conclusion.


Career Concerns and Distortion in Credit Uptake: Evidence from India’s Lead Bank Scheme [SERI Working Paper No. 7; Presented at ISI, Delhi, December 2020; World Bank-Conference, February 2020; SERI, July 2020; Accepted at IOEA, 2020; Covered in Ideas for India]

Abstract: In the Lead Bank Scheme of India, a public sector bank, known as the Lead Bank of the district, reduces bottlenecks in financial service delivery. Another public sector bank, known as the Convenor Bank of the state, monitors the effort of Lead Banks across districts within a state. Given this organizational design, for some districts, defined as aligned districts, Lead and Convenor banks fall within the boundary of the same firm. I nd that in aligned districts credit uptake is higher by 21%, consistent with higher effort by Lead Bank personnel owing to lower monitoring costs (Williamson, 1981) and higher career concerns (Holmstrom and Roberts, 1998) within a rm. I conduct two tests to identify the impact of alignment within a district on financial inclusion outcomes. First, using a plausibly exogenous (to district characteristics) change in alignment status of a district, I nd that credit increases by nearly 28% when a district becomes aligned. Further, after an exogenous negative income shock of bad rainfall, saving withdrawals are lower in aligned districts, consistent with higher credit availability (Eswaran and Kotwal, 1990). The results show how organizational pressure within commercial banks in India may distort credit lending across districts.


Value of Managers in the Internal and External Labor Market: Evidence from an Insurance Firm in India [Under Review] (Presented at Society of Labor Economics Conference, Toronto, May 2018)

Abstract:  Do good managers improve the productivity of workers within the firm or are they better at spotting talent from the external labor market? In an Indian-insurance firm, I find that a new recruit chosen by a manager with high team productivity outperforms the new recruit chosen by a manager with low team productivity. This variation could stem from a good manager’s skill of recruiting talent or it could be her value addition to agent performance. To find manager value addition (MVA), I use the firm’s internal labor market policy under which agents move across managers. Though, there exists substantial variation in MVA, managers of high productivity teams do not have high value addition. This indicates that when an agent moves to a high productivity team, change in his performance is not attributable to the manager of the high productivity team. Thus, a manager increases team productivity by recruiting more productive
workers from the external labor market. To the best of my knowledge, this is the first paper to complement a manager’s role in selecting good workers from external labor market against her contribution to an individual worker in the internal labor market of the firm.


  1. Impact of Adaptation Strategies on Business Expectations: Evidence from India during the Pandemic (with Bornali Bhandari, Ajaya Sahu and KS Urs) [Under Review] [Available on SSRN]
  2. Model Selection Inference for Causal Impact of Clusters and Collaboration on MSMEs in India [Under Review]
  3. Business Sentiments in the times of COVID-19 (with Bornali Bhandari, Ajaya Sahu and KS Urs) [Under Review] [Available on SSRN]
  4. Public Goods, Government Procurement & Upstream Prices: Evidence from Indian Pharmaceutical Industry (with Chirantan Chatterjee, IIM Ahmedabad)
  5. Price and Tax Elasticity of Alcohol Consumption: Evidence from Seven Indian States (with Soumi Roy Chowdhury)