This paper estimates the incentive effects of replacement rate (measured as the ratio of disability benefits to an estimate of potential earnings), non-employment and disability policies on the hazard rate of disability entitlements. The incentives are formulated in a dynamic programming model within a counting process framework and estimated using the one percent Continuous Work History Sample (CWHS) of the Administrative data on around 3 million individuals. The paper examines the argument that the widening income gap between the rich and the poor in the eighties created strong incentives for workers to get onto the rolls by raising the replacement rates of the poor. The paper finds that while the replacement rate of the poor grew rapidly in the eighties, its effect on enrollment was not strong. The paper also shows that the effect of non-employment on the likelihood of one’s getting onto the program depends on the health status of the worker: a healthier worker (as inferred from the proxy variable for health constructed in this paper using work history) has a much lower probability of getting onto the program than a worker with adverse health shocks. Among various disability policies, the addition of Medicare benefits contributed most to the growth of disability entitlements; the mental listings and broadening of coverage to musculoskeletal and mental disorders have significant positive effect and the D&A and Welfare reforms have significant negative effect in explaining the pattern of disability entitlements in the 80’s and 90’s.