Transportation investments have profound indirect effects on urban development. Improving transportation infrastructure reduces travel times, and thus, enhances accessibility of the population to employment and vice versa. In this procedure, the cities whose accessibilities mostly benefit from the transportation improvements would attract more population and capital. This paper attempts to quantify these impacts in the megaregions, and to prioritize the transportation investments according to their potential effects on the sustainability of the future development in these regions. For this put-pose, the problem is formulated as a bilevel programming model. At the upper level, subject to a budget and environmental constraints, the optimal subset of projects is chosen from the candidate set. The objective is to maximize the total production of the megaregion at the end of the planning horizon. The final product of each city is determined according to its resultant population and physical capital. Under the network investment decisions of the upper level problem, the lower level represents the route choice behavior of users. Because the focus of the study is interurban road networks, which have sparse connectivity, a simple all-or-nothing trip assignment technique is employed at this level. To illustrate the application of the proposed model, the Northeast Megaregion of the United States is used as the case study. The exact solution is obtained for this example and compared to the regional development without improving the transportation network. (C) 2014 American Society of Civil Engineers.