r.i.c.e.

Mahatma Gandhi National Rural Employment Guarantee Scheme (NREGA) is the largest anti-poverty program in India’s history and in the history of the developing world. Government of India allocated $8 billion (Rs. 400 billion) to NREGA in 2011-12--30% more than the World Bank’s total lending and assistance for all rural development programs across the world.

There has been a great emphasis on transparency, accountability and continuous monitoring in implementation of NREGA. The program has a dedicated website with regularly updated data down to the village level. I cannot think of any other government program in India with such rich, current, and readily available information. NREGA’s rich data would be a researcher’s dream—if only the data were any good.

I started exploring the NREGA website only about two-three weeks ago. I started with state-level summary reports on total employment demanded and created under NREGA. The reports tell us that most states of India have managed to provide employment to almost every household that asked for work in 2009-10 and 2010-11. This seemed too good to be true. But how do we verify it?

Fortunately, recent rounds of Employment and Unemployment Survey (EUS) of NSS ask some questions about participation in NREGA from all rural households in the sample. One of the questions is:

If any member of the household worked in NREGA in the last 365 days (1= yes; 2= sought work but did not get it; 3 = not interested in working in NREGA.

Responses to this question show that in 2009-10, only 45% of households that sought work under NREGA actually got any work. NREGA website shows this fraction to be more than 98%, with little inter-state variation. We see large inter-state variation in the extent of rationing of employment opportunity under NREGA in the EUS data. Rationing is high in Maharashtra (84.02%), Punjab (83.35%), Bihar (78.45%), and Haryana (73.79%) and low in Rajasthan (15.50%), TN (19%) and AP (25%).

The EUS data seems more reliable while the official NREGA data is complete fabrication. NREGA Act promises 100 days of work to every rural household that asks for it. If the government fails to provide work, it has to pay an unemployment wage to the person. While the federal government pays NREGA wage, paying unemployment wage is state government’s responsibility. State governments tamper data to save their money and the result is unreliable data and the accompanying loss of accountability.

Designers of NREGA would have hoped that making data easily accessible would prevent corruption. In reality, the data itself has been tampered—massively, blatantly, obviously.