Research

Abstract: This paper evaluates the distributional and welfare impacts of New York City’s recently implemented congestion pricing program, the first in the United States. Using the 2022 Citywide Mobility Survey, I estimate a discrete choice model of travel mode with heterogeneous responses to cost and time, embedded in a congestion feedback framework. The analysis distinguishes between two channels: a price-side effect, capturing the direct burden of the toll on travelers, and a revenue-side effect, reflecting how toll revenues are recycled. The results show that the current toll ($9 peak / $2.25 off-peak) reduces private vehicle trips involving the Manhattan Core by about 19,000 per weekday, with most shifts to subways and walking. Accounting for congestion feedback raises driving speeds and substantially increases welfare relative to a fixed-traffic analysis. The toll alone lowers consumer surplus by $0.32 per person-day, with larger relative burdens on lower-income travelers, while a uniform rebate raises net welfare by $2.24 and makes the policy mildly progressive. Distributional analysis shows that equity outcomes hinge on these two effects: the price-side burden follows income patterns, while the revenue-side outcome depends on policy design. Transit fare subsidies generate smaller average gains while reinforcing mode shifts. The findings highlight that the equity and efficiency of congestion pricing depend jointly on traveler behavior and revenue design, with lessons for other U.S. cities planning similar policies.

Abstract: The impact of public transit on relieving traffic congestion remains a topic of debate. This paper uses a panel event study and a difference-in-differences (DID) design to estimate the impact of an intensive marginal change in the public transit system on traffic congestion, using data from three new subway stations in New York City (NYC) that opened in 2017. The results show that on average, traffic delays in nearby road sections decreased by 12.0% to 15.0% after the opening of the new stations, a magnitude similar to the effects reported in the literature for the entire transit system. The study also calculates an annual direct congestion-relief benefit of the transit system, estimated to be $332,601. These findings provide evidence for the potential of marginal changes in the public transit system to relieve traffic congestion and support the importance of subsidizing public transit.

Abstract: Road Construction, proposed as a solution to severe congestion, might increase commuting time and decrease traffic speed due to poor management or delay in completion. This project combined the INRIX traffic data with 13,000 construction record data to estimate the network effect of road construction on traffic speed in the Manhattan area of New York City, calculate the exact amount of congestion cost caused by construction and evaluate the efficiency of penalty on late completion and feasibility of bonus for early completion in the construction contrast. The main finding is that an extra road construction project will decrease traffic speed by 1.0%-1.6%, causing commuting time to increase by 0.06%-0.09%. Regarding the distance to the construction, traffic speed will decrease by 1.6% when the average distance to the construction sites increases by 1%. The estimated cost of a decrease in speed of an extra road project is about $33,000- 104,000 per day for Manhattan commuters, which is relatively higher than the late penalty in construction contrast. This project suggests that a bonus for early completion or a higher penalty for late completion should be considered in road construction contrast.

Abstract: Road Construction, which is proposed as a solution to severe congestion, might increase commuting time and decrease traffic speed due to poor management or delay in completion. This paper sets up a traffic dataset with 2 million real-time traffic speed data and 13,000 construction record data from 2017 - 2019 to estimate the marginal effect of road construction on traffic speed in New York City and evaluate the efficiency of construction contrast. The main finding is that traffic speed will decrease by 3.47% on average when vehicles are 1km closer to the road project. Speed decreases slower when vehicles are further from construction. Besides, long-duration and "unplanned" projects have a greater effect on vehicle speed. The estimated cost of a decrease in speed is about $326-451 per hour, which is relatively higher than the penalty in construction contrast. Based on this, this paper suggests that a bonus for early completion or a higher penalty for late completion should be considered in road construction contrast.