In an attempt to reduce compliance costs and improve efficiency, recent environmental regulations have relied on market-based incentives. In most cases, the Environmental Protection Agency (EPA) selects an emissions cap and allows firms to trade pollution permits. The EPA typically selects the cap using a "bottom-up" approach to predict the costs of such regulations, forecasting how every affected firm will respond to the program. It is uncertain whether firms rely on the same predictions in making their compliance decisions; discrepancies between the actual and expected behavior of the affected firms could reduce the cost-effectiveness of the program. This paper uses stock prices to compare the predictions of the bottom-up studies with those of the affected firms.I focus on a recent tradable permit program, the Nitrogen Oxides Budget Trading Program (NBP). Started in 2004, the NBP requires electric generators in the Midwest and East to reduce their emissions or purchase permits from other firms. I compare utilities' stock prices with the prices that would have occurred in the absence of the new regulation. I make this comparison by exploiting variation in the location of generators owned by utilities; the control group consists of utilities without any generators in the NBP. I estimate that investors expected the program to reduce profits by about $2 billion per year (2000 dollars). Investors expected the NBP to primarily affect coal generators, which have larger baseline emission rates than other fossil fuel generators. These results agree with previous studies that used the bottom-up approach.