Sin taxes encourages fraudulent and cheating behaviors from targeted groups--- this uniquely increase market uncertainties which hurts growth.

Sin taxes (narrow taxes on specific goods or industries) are often justified as a key legislative tool to reduce unhealthy behaviors such as the use of tobacco and excessive alcohol consumption (Cawley & Ruhm, 2011). Research suggests these taxes are generally effective in decreasing the purchase and consumption of targeted products (Chaloupka & Warner, 2000; Xu & Chaloupka, 2011). In addition to directly impacting demand, however, sin taxes have other spillover effects. For example, Cook and Moore (1993) observe that beer-excise tax rates are negatively correlated with rape and robbery at the state level. Beatty et al. (2009) document that large tax differentials near borders induce cross-border shopping and smuggling of alcohol and tobacco. Rozema and Ziebarth (2017) find that smokers respond to increases in cigarette taxes by increasing their use of food stamps. Our study adds to the literature on the spillover effects of sin taxes by examining whether the taxpayers targeted by the tax are more likely to defraud other individuals in unrelated transactions. In general, this adds to our understanding of the true effects of sin taxes, and studying the forces that lead to fraud are particularly important because fraud can erode trust in markets.

There are two reasons why affected taxpayers might commit more fraud following a rise in a sin tax. First, tax increases can reduce the buying power of targeted taxpayers. Given the importance of incentives in spurring fraud (e.g., Charness et al., 2019; Dimmock & Gerken, 2012; Kajackaite & Gneezy, 2017), this reduction in purchasing power could increase an affected taxpayer's marginal value of income and, accordingly, their associated incentive to commit fraud. Second, being treated unfairly makes cheating behavior easier to rationalize, even when the potential victims of fraud are not the party behind an individual's unfair treatment (Houser et al., 2012). Surveys consistently report that affected taxpayers view sin tax hikes as unfair (Carlson, 2005; Dugan, 2014; Saad, 2002), which suggests that these affected taxpayers may find it easier to rationalize defrauding others in the aftermath of sin tax hikes.

We test the relation between sin tax increases and fraud using detailed taxi ride data, which allow us to identify drivers who cheat and drivers affected by a sin tax hike (an increase in cigarette taxes). We define cheating as overcharging on taxi rides, where drivers charge the “out of town rate” for rides taking place within New York City (following Paci, 2014 and Douthit, Millar, & White, 2021). We define treatment (smokers) and control (non-smoker) observations by identifying taxi drivers who smoke via tickets for violating the taxi regulator's prohibition on drivers smoking in cabs. We then examine how drivers' likelihood of committing fraud changes around a large (160%) increase in federal excise taxes on cigarettes. To control for permanent differences between smokers and non-smokers as well as time trends common to both groups we use a difference-in-differences methodology. We find that the treatment effect odds ratio on the test variable Smoker x Post Cigarette Tax Hike is consistently around 1.5, suggesting that in the month immediately after the tax hike smokers ramp up their cheating by about half (relative to baseline cheating rates, this represents an increase of about 1.1 fraudulent trips per month from a mean of 2.3 such trips per month per driver).