Evaluating the Car Ownership Control Policy in Shanghai: A Structural Vector Auto-Regression Approach

Abstract

Beginning in 1986, Shanghai has implemented an interesting car ownership policy, namely an auction of the right to register private cars. The auction was modified several times over the years. Most recently since April 2013, the authority added price caps to mitigate rapid price run-ups. In this paper, we propose a structural vector auto-regression (SVAR) approach to evaluate the performance of the Shanghai auction since 2002. The SVAR is flexible enough to incorporate alterations to the auction system into the modelling structure. We find three key results: (1) at least before 2008, short-term price fluctuations can be managed by varying the quota; (2) the number of bidders, or demand, ultimately drives the price in later years; (3) the impact of one standard deviation increase in demand takes 6 months to one year to dissipate, casting doubt on the viability of the new auction procedure, which has accumulated more than 150,000 bidders in waiting since April 2013. We also analyze how socio-economic factors affect quota, price, and the number of bidders using a simultaneous equations model. Due to data limitations, such an analysis is less conclusive. We argue that the SVAR framework can be improved and applied to other cities in evaluating their car ownership practices.