Ashesh Prasann

The Spillover Effects of Public Works on Labor Markets: Evidence from National Rural Employment Guarantee Scheme, India

 

Public-works programs guaranteeing work at above-market wages are intended to provide security to the seasonally unemployed and are an increasingly used labor market and anti-poverty policy in developing countries. Impact evaluations of India’s National Rural Employment Guarantee Scheme (NREGS), the largest such program in the world, have estimated a variety of labor market effects, but this literature is agnostic about general equilibrium spillovers generated by the program, which could potentially bias these estimates. This paper tests for spillovers from NREGS to neighboring labor markets by exploiting plausibly exogenous variation in wage differentials and exposure induced by the program’s staggered rollout across contiguous districts. The results show that non-program districts exposed to program neighbors experienced an 8.7% rise in casual wages relative to unexposed districts and these spillovers are higher among women. These findings demonstrate that the impact of NREGS on the casual labor market is higher than previously estimated. They also provide empirical support for the theory that increased short-term migration to contiguous program areas is the mechanism generating spatial spillovers in neighboring areas which did not receive NREGS over the study period.

 

Author: Ashesh Prasann

 

Agricultural Worker Productivity and Malaria: Alternative Productivity Measures using Physical Activity

 

Individual labour productivity is difficult to measure without individually attributed worker output, typically only available in piece rate settings. This paper investigates a possible alternative – direct measures of physical activity using an accelerometer, as a measure of individual worker productivity, as this information may be feasible to collect in numerous labor settings. We establish the validity of this measure in an agricultural labor force where workers are paid per quantity of sugar cane cut. First, we compare worker physical activity with daily output and labour supply and observe that it is strongly positively associated with both measures. Second, we investigate the effect of a health intervention on physical activity using a phased-in randomized control trial. The intervention provides testing for malaria followed by treatment when workers test positive. Treatment effects on the treated using measures of physical activity indicate that workers daily average time sedentary reduced by 66 minutes and time in ‘light’, ‘fairly active’ and ‘very active’ physical daily activity levels increased for the average treated worker by 21, 37 and 11 minutes, respectively. These effects are similar to Dillon et al. (2014) who find strong effects of the same treatment on individual labour supply and piece rate productivity measures in a previous agricultural season. The emergent technology of wearable activity trackers can be feasibly incorporated into prospective studies of productivity and labour.

 

Authors: Oladele Akogun (Common Heritage Foundation), Andrew Dillon (Michigan State University), Jed Friedman (World Bank), Ashesh Prasann, Pieter Serneels (University of East Anglia)