Using Movement of Exemption Cutoff to Estimate Tax Evasion : Theory and Evidence from Pakistan

The emerging evidence from developing countries suggests that the costs of evasion there are roughly a reverse-L-shaped function of earnings: they are extremely low up to a given threshold and then rise sharply with evasion. Embedding such evasion costs into the standard model generates a discontinuous earnings supply function. Taxpayers report true earnings at zero tax rate but evade the amount that can be evaded at a negligible cost as tax rate increases slightly above zero. I develop a methodology that exploits the discontinuity to provide a lower bound on tax evasion. Using a series of tax reforms and administrative data from Pakistan, I test the predictions of the model and estimate that at least 70% of self-employment and 1% of wage income is evaded in the country