Assessing Driving Restrictions: Evidence from Delhi's Odd-Even Scheme

In this paper, we present the analysis of odd � even scheme implemented in Delhi. In a random fixed effects model, we use difference-in-differences (DD) and difference-in- difference-in-differences (DDD) research design that draw on temporal variation and spatial variation to identify the effects of the scheme. We use pollution levels as our outcome measure and define our treatment group to be the set of peak traffic hours in the restricted zone. We exploit temporal variation by using a DD strategy that compares the change in pollution during peak hours inside the restricted zone to the change in pollution during off- peak hours in the same zone. Spatial variation is exploited by using a DD strategy that compares the change in pollution during peak hours inside the restricted zone to the change in pollution during the same hours outside the restricted zone. Both types of variation are also exploited with a difference-in-difference-in-differences strategy. In the section 2, we present a review of the existing literature on driving restrictions. In section 3, we provide some more detail about Data sources, the odd-even program, along with background information about Delhi and its air quality. Our empirical strategy is described in section 4 and section 5 with Results.