Delhi has now had two encounters with the odd-even pilot experiment. The first, in January, succeeded in reducing pollution by about 10-13 per cent, as we showed in a previous article (‘Yes, Delhi, it worked’, The Indian Express, January 19). But this time around, we find no evidence that the same policy, running from April 15 to April 30, affected pollution levels. This difference illustrates an important point we have made before: That the odd-even policy is inherently unpredictable and ultimately is likely to have an insignificant impact on pollution. In the long run, other approaches are likely to be better suited to reduce pollution. The difference also underlines why context matters. Different conditions require different measures, and government action in one arena can undermine efforts in another.

Before going further, it’s important to understand the methods we used to arrive at these estimates. We deployed what is known as a difference-in-differences approach. This involves comparing changes in pollution trends in an area where a policy did occur to a similar area where it did not. The point of this approach is to control for external factors that influence pollution outside of the policy. For example, April has seen crop burning and forest fires in Uttarakhand. Certainly, these factors influence air pollution levels and trends in Delhi. But they are likely to equally affect the rest of the NCR too. The method we use therefore can account for such external factors and then examines policy impacts. We used data only from government monitors in Delhi, Faridabad and Gurgaon…