Working Paper: CEPR ID: DP17414
Authors: Sam Asher; Alison Campion; Douglas Gollin; Paul Novosad
Abstract: How do investments in agricultural productivity translate into development and structural transformation? We estimate the long-run impacts of India's irrigation canals, which span 300,000+ km and deliver water to 130,000+ villages. Drawing on high-resolution data on every household, firm, village, and town in India, we use three empirical strategies to characterize the direct and spillover effects of large increases in agricultural productivity. Our findings are consistent with a spatial equilibrium model in which labor is mobile, and urban areas have non-farm productivity advantages. In the long run, areas directly treated by canal irrigation have sharply higher agricultural productivity and population density, but similar non-farm employment shares to non-canal areas. Persistent consumption gains accrue only to landowners and structural transformation occurs almost exclusively through the concentrated growth of regional towns. In the long run, the substantial productivity effects of canals were equilibrated through the movement of labor across space rather than within locations across sectors.
Keywords: Agriculture; Agricultural Productivity; Irrigation; Structural Transformation
JEL Codes: O13; O18; Q15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Agricultural productivity (Q11) | Non-farm employment (J43) |
| Labor movement (J51) | Productivity effects of canals (O49) |
| Canal irrigation (Q15) | Structural transformation (L16) |
| Town growth (R11) | Structural change (L16) |
| Canal irrigation (Q15) | Agricultural productivity (Q11) |
| Canal irrigation (Q15) | Population density (J11) |