If you build it (roads), will they (financing) come?

Investments in public infrastructure is a key component of economic growth strategy among emerging economies, and a particular focus of the Modi government. In general, policymakers and financial economist assume that financing will follow once the roads are built, and thus, facilitate the best use of new productive opportunities created by new road connectivity. However, many rural and agrarian economies suffer from chronic problems of financing, characterized by the absence of formal financial institutions and reliance on informal moneylenders who often are unreliable and charge usurious interest rates. Therefore, a key question remains: if you build it (roads), will they (financing) come?

Moreover, even if financing does follow infrastructure improvements, who will it benefit, the relatively rich villagers, or the poorer villagers who were excluded from formal finance to begin with? These questions are of immense relevance to policy makers but remain underexplored in extant research. One reason behind the lack of research in this area is that infrastructure is not placed randomly across regions. Any improvements in economic growth in a region that receives roads, compared to one that does not receive roads, cannot be exclusively attributed to investment in infrastructure. The improvements can also come from changes in other contemporaneous environmental or policy factors. Thus, research needs to find a way to compare regions that are similar in all aspects except for their infrastructure development. Fortunately, the Pradhan Mantri Gram Sadak Yojana (PMGSY) program in India provides a near perfect setting that solves this problem.

The PMGSY program prioritizes building roads in villages above explicit population thresholds at a point in time. For example, a village with a population right above a round figure, say 500, was to be prioritized to receive a road, relative to a village with a population right below 500. Thus, a comparison between two villages that are right above (say 510) and right below (say 490) the threshold, but are otherwise similar in geographic and economic factors, allows us to examine the effect of receiving a new road on financing, while holding other factors constant. In our recent paper, we take advantage of this setting to examine the impact of the PMGSY program on private financing to households (Roads and Loans by Sumit Agarwal, Abhiroop Mukherjee and S.Lakshmi Naaraayanan). Our analysis is made possible by our access to a proprietary loan-level dataset from one of India's largest private lenders (referred to as “the bank” hereafter) on loans made in the largely rural district of Ganjam in Odisha.

Our sample period starts in 2009, when the bank started lending and all villages in our sample were unconnected. By 2014, a previously unconnected village right above the population threshold is about 45% more likely to get a road than one right below the threshold, which validates that the villages above the population threshold are given priority to receive roads.

In terms of the number of households receiving loans from the bank, by 2014, when we measured differences in loans outcomes, 5.1% of households living in villages right below the threshold received loans. In comparison, this coverage jumped to about 8-9% in villages right above the threshold. Among the households that received loans, the loan amounts are 28-32% higher for those who lived in villages right above, versus below, the threshold. Overall, we find large financing responses to rural road connectivity.

Given the significant amount of financing, it begs the question what are these loans used for. To answer this question, we partition the loan amounts based on whether the financing was provided for productive uses, such as business expansion, asset acquisition, and working capital needs, or other non-productive uses, such as consumption needs, weddings, and festival expenses. We find that, villages above the population threshold (relative to below) received significantly higher loan amounts for productive uses of all types, and in particular, this difference is strongest economically for micro enterprise loans and crop loans. This is consistent with policy makers' views on benefits owing in the form of opening or expansion of small businesses like village grocery shops, or changing crop patterns from subsistence to more profitable market-based crops. Further, we also find lower loan amounts for non-productive uses.

Lastly, we examine the distributional consequences of rural road connectivity. This is a critical step in understanding the trickle-down effects of infrastructure development, as well as financial inclusion. We focus on individual-level demographic differences, and identify the segment of population that benefits most from road connectivity. We find that individuals with less assets benefit more in terms of receiving higher loan amounts. That is, people who did not have collaterlizable assets like a house or land can now borrow more. This is consistent with the view that new roads release collateral constraints, and improve financial inclusion for those who lack traditional collateralizable assets.

Interestingly, we also find that those with some school education and those with lower incomes seem to benefit more from connectivity. Thus, an important implication is that villagers who perceive the association between education and greater financing, upon receiving new roads, may invest more in their children's education.

Policymakers often talk about trickle down benefits of rural road connectivity for households, such as allowing farmers to shift to cash crops, move goods produced in villages to market places, employment generation within villages, etc. Our results have important implications for understanding such trickle down benefits of building large scale infrastructure and its distributional consequences. Our research is one of the first explorations of whether financing responds to productivity changes arising from large-scale road-building initiatives. In doing so, our research provides convincing estimates of the impact of road connectivity on access to, and productive utilization of, formal financial services.