We compare the administrative data on the implementation of the Mahatma Gandhi National Rural Employment Guarantee Act (MG-NREGA) with the estimates of public works employment from a large-scale survey undertaken by India’s National Sample Survey Office. A widely held concern is that the administrative employment records are inflated by corrupt officials and therefore do not reflect the true employment provided. By comparing the administrative data with independent survey data, we provide an
estimate of the fraction of officially reported days that are actually worked.
The employment estimates from the survey data are between 42% and 56% of the employment reported in the administrative data. These numbers are similar in magnitude though generally lower than existing estimates. In particular, using a similar approach, Himanshu (2010) estimates that 70% of the reported days are independently confirmed by the NSSO data. Bhalla (2010) estimates the number to be 50% using an earlier time period.