The need for Social Security Reform in the next years is hardly a matter of debate. Therefore, the widespread believe among Americans that Social Security will not be able to pay benefits in the long run at the level that was anticipated, does not come as a surprise. The government acknowledges the situation, and predicts that substantial benefits cuts will be necessary, yet no legislation has been passed to tackle the problem. Researchers, however, have rarely modeled the uncertainty over Social Security reform and benefit levels, and how they affect claiming behavior and retirement. The purpose of this paper is to assess the extent to which these perceptions of future cuts might explain the puzzle of earlier take-up despite bigger penalties to doing so in the presence of increasing longevity. By introducing a small amount of uncertainty (based on self-reported responses to questions regarding expectations over future cuts) of a relatively small cut (compared with what the government reports as necessary to solve the crisis) in a dynamic life-cycle model of retirement, we are able to match the claiming behavior observed in the data, without relying on heterogeneous preferences. Our results support the hypothesis that expectations over future benefits are affecting current behavior.  We find that a mis-specified dynamic retirement model would erroneously predict that an increase in the NRA would delay claiming behavior and increase labor supply at older ages. Once the appropriate earnings test incentives are modeled, and we account for the probability of reforms to the system, an increase in the NRA has little effect on claiming behavior, and it can even increase the proportion of individuals claiming before the NRA.