The Americans with Disabilities Act (ADA) will have a significant effect on paratransit demand and system operating costs in the United States. One of the main components of the regulations is the restriction of any practice or pattern of trip denials. This paper presents an econometric model designed to account for the simultaneous interaction between paratransit trip demand and the supply determined denial rate. The model is estimated using two‐stage least squares applied to paratransit operating data for New York City and Seattle. The results show that lower denial rates (i.e., improved system reliability) lead to higher demand levels and subsequently higher operating costs. Using demand elasticities from the model, the principle of compensating variation is used to develop a cost‐benefit analysis of moving to lower denial rates. The analysis indicates that the optimal denial rate (that which maximizes net benefits) lies above zero percent.