The Differences-in-Differences Approach with overlapping differences - Experimental Verification of Estimation Bias

Billede af publikationens forside
23-05-2014
DREAM

This paper demonstrates a hitherto overlooked problem with the differences-in-differences approach where it is applied to panel data and with overlapping differences.

Abstract

This paper demonstrates a hitherto overlooked problem with the differences-in-differences approach where it is applied to panel data and with overlapping differences. A differences-in-differences model with three-year overlapping differences has frequently been applied to measure the behavioural effects of changes to taxation. The problem is that an inherent technical flaw means the differences-in-differences model with overlapping differences seriously underestimate tax elasticities. It is demonstrated that the severity of the problem will depend on the actual situation – such as the nature of the tax changes – but under reasonable and substantiated circumstances the long term tax elasticity is underestimated by more than half. It is, however, not possible to assess the magnitude of the underestimation with any certainty and it is therefore recommended that the method is unreliable and should not be used. In contrast, the error correction model is an alternative dynamic specification that does not suffer from the flaw of the differences-in-differences model with overlapping differences. With explicit distinction between short and long run effects, the error correction specification is recommended as an alternative method for estimating behavioural effects of tax changes.