Experimental evidence on the relationship between tax evasion opportunities and labor supply

Motivated by the observation that access to evasion opportunities is distributed heterogeneously across the labor market, this paper examines the extent to which labor supply elasticities with respect to tax rates depend on such evasion opportunities. We first set up a theoretical model to formally show that labor supply responses depend on access to evasion. The model is then tested in a lab experiment in which all participants undertake a real-effort task over several rounds. Subjects face a tax rate, which varies across rounds and are required to pay taxes on earned income. The treatment group is given the opportunity to underreport income while the control group is not. We find zero labor effort responses to tax rates in the control group and positive statistically significant adjustments in the treatment group; suggesting that both groups indeed react differently to taxes.