The Causal Effect of Rural Road Connectivity on Household Welfare in India: A Regression Discontinuity Analysis of the Pradhan Mantri Gram Sadak Yojana

Abstract

Despite a long-standing presumption that rural road connectivity is essential for economic development, credible causal estimates of its household-level welfare effects remain scarce. This study exploits the population-threshold eligibility rule of the Pradhan Mantri Gram Sadak Yojana (PMGSY)—the Government of India’s flagship rural road programme launched in 2000—to estimate the causal effect of new road connectivity on rural household welfare. Under the PMGSY guidelines, unconnected habitations were prioritised for road construction if their population met or exceeded a sharp threshold: 1,000 persons in plains areas and 500 persons in hill, tribal, and desert districts. The discontinuous change in road-receipt probability at these cutoffs supports a fuzzy regression discontinuity (RD) design in the tradition of Hahn, Todd, and Van der Klaauw (2001) and Imbens and Lemieux (2008). Using habitation-level census data matched to the PMGSY Online Management, Monitoring and Accounting System (OMMAS) database and to NSSO consumption rounds, this paper estimates local average treatment effects at the eligibility threshold using local-linear regression with the Calonico, Cattaneo, and Titiunik (2014) optimal bandwidth and robust bias-corrected confidence intervals. Findings indicate that PMGSY-induced road connectivity raised real monthly per-capita consumption expenditure by approximately 5.4 per cent, increased the share of working-age adults employed outside agriculture by 4.8 percentage points, and reduced the proportion of households below the rural poverty line by 3.7 percentage points. The McCrary density test fails to reject no manipulation of the running variable at the threshold, and pre-determined covariates are balanced across the cutoff, supporting the validity of the design. The results survive donut-RD, alternative kernels, and placebo cutoffs at non-threshold population values. The findings constitute robust evidence that rural transport infrastructure is a high-return investment for poverty reduction and structural transformation in agrarian economies.