Tax evasion is basically non-existant

6/16/2014 11:51:00 AM
Tweetable
Do not collect $200 billion, that is.
...according to a new study.

The New York Times has been heavily marketing a column claiming that there's rampant tax evasion in the world, based on a study by a "Piketty protege" (their words). Yet, despite the fact that the Times is trying to make a big deal out of this, they've actually shown the opposite--the amount of lost tax revenues is trivial:
"If all of this illegally hidden money were properly recorded and taxed, global tax revenues would grow by more than $200 billion a year, he believes."
Even for just a single country like the US, $200 billion isn't all that much. It's not enough to fund Obamacare, or balance the federal budget, for example. In fact, global tax evasion is actually dwarfed by the amount of totally legal tax avoidance just within the US--corporate tax loopholes in the US alone amount to more than $300 billion in lost revenue each year. But we aren't talking about just the US--it's only $200 billion for the entire world, less than 0.3% of global income. So, it's chump change.

Tax evasion is an important issue in the study of optimal taxation. Typically, we think about optimal taxation and redistribution as balancing the tradeoff between the disincentive to work on the one hand versus the gains from a more equal distribution of income on the other hand. To optimize this tradeoff, we rely on empirical estimates of the elasticity of labor supply--in a world with no tax evasion, labor supply is directly proportional to labor income. Some microeconomic estimates--those that estimate the elasticity of labor with respect to changes in the tax code--are impervious to the question of tax evasion, because that would simply be baked into the estimates. But other studies that use non-tax related shocks for identification are vulnerable, because tax evasion drives a non-constant wedge between the labor supply decision and the reported taxable income. [side note: it's important to have both kinds of estimates, because as economists we should avoid assuming either that tax evasion doesn't happen, or that there's nothing that can be done legislatively to reduce it] This matters because the easier it is to hide income from the government, the less amount of redistribution we are able to accomplish, so that more tax evasion actually means we should aim for less redistribution.

Until this article, I worried that the existence of off-shore tax havens could actually be a big limitation on how much redistribution we could accomplish. But not anymore--this new study shows that the extent of tax evasion is trivial. The last argument against an expanded welfare state has been debunked.