Abstract
Motorcycle ownership and its usage increase dramatically especially in urban area of Indonesia due to rapid growth of economic development and lack of public transport infrastructure. Government of Indonesia and local government of several cities enacted Inspection and Maintenance program for in-use motorcycle to reduce motorcycle exhaust emissions. To analyze this policy, we apply bivariate probit model for the likelihood of CO and HC emission violations given a set of motorcycle characteristics which obtained from on-road measurement at Jakarta, Bandung and Surabaya in 2004. We found that maintenance quality and running kilometer per year play significant role in determining the probability of emission test. Then, GDP per capita and urban density influences personal motorcycle mobility per capita per year at city level data. By comparing estimation results among those cities, the probability of observed data at Bandung differed significantly compare to Jakarta's samples. Empirical study confirms that high mobility at low population density is better than low mobility at high density of the city.
