To what extent have agricultural wages responded to the introduction of India's National Rural Employment Guarantee Scheme (NREGS)? Can delivery of NREGS benefits be improved by informing the target population of their entitlements under the scheme? What, if any, impact has NREGS had on income growth and poverty amongst the poor?

India's National Rural Employment Guarantee Scheme (NREGS) legally guarantees 100 days of paid work per year for any rural household that is willing to undertake unskilled or semi-skilled wage work. The high-profile scheme was introduced in 2005 and extended to all of India in 2008. Its twin aims are to provide a safety net for the poor in the form of locally available wage work, and to promote economic growth by channelling these labour resources towards the creation and restoration of local productive assets such as rainfall reservoirs, irrigation channels, wooded areas and roads. Low wage rates along with the manual nature of the work are intended to ensure that only the needy apply. However, as with many public schemes in developing countries NREGS faces serious implementation problems, and its real impact on growth and poverty is also being questioned.

One of the aims of the NREGS is to promote economic growth by creating and restoring rainfall capture systems, irrigation channels and other ‘productive assets' used by the poor. Local economic growth is difficult to measure, so instead we propose to look at local agricultural wages. The bulk of the rural labour force is employed in agriculture, and given that the poorest typically own little land of their own and rely on wage income, agricultural wages are a good proxy for the income and welfare of the poorest. Agricultural wages series therefore offer a way to measure and track ‘pro-poor growth'.

There are two ways in which the NREGS may be thought to have an influence on agricultural wages. One is by investing in capital which makes workers more productive and thereby increasing the return on labour input. The other is by changing local labour market conditions. If the scheme increases the demand for local labour under sufficiently attractive conditions, the market-clearing wage rate would be expected to increase. One might expect the effect to be particularly pronounced in the agricultural slack season, when labour demand is low. In addition, it is possible that agricultural labour will be more productive because the workers are generally better off and thus healthier.

Several long-running national data series collect agricultural wages at the local level. We will pair these data with information about the NREGS rollout schedule and intensity of operation in the same regions in order to assess the impact of the scheme on wages.

We will also investigate whether the delivery of NREGS services can be improved by informing the target population of their entitlements. We propose to undertake a campaign to inform intended beneficiaries of their entitlements under the NREGS. The campaign will be randomly allocated, and we will compare various measures of scheme efficiency before and after the campaign, across treatment and control locations. This will allow us to discuss the causal effect of providing entitlement information on the quality of service delivery in a large-scale and high-profile public programme.