For this last guest post, I decided to eschew IP (well, to some degree, see below) and focus on a recent article of mine and Leandra Lederman—Enforcement as Substance in Tax Compliance—that (in my biased opinion) deserves more press. (Though, to be certain, Leandra would strongly disagree with my catchy title—see more on that below.)
In the article, we argue that the failure of the tax authorities to perfectly enforce the tax laws in effect can alter the substance of the tax laws, at least to the extent that taxpayers know the penalties for non-compliance and can reasonably predict how often the tax authorities conduct audits for certain classes of taxpayers.
This claim is quite different from the one that audit rates merely affect the propensity of taxpayers to “cheat.” Rather, the effective tax rate can be intentionally adjusted downward (and not just to zero) by the taxing authorities from the nominal tax rate, in effect, yielding an entirely new substantive law.
For instance, suppose there is a sales tax of 5% and that given current audit strategies, the effective tax rate is 4% in all industries, because 20% of the taxes go unpaid in each industry. Based upon the elasticity characteristics of the products in a given industry—in other words, the propensity of purchasers to continue buying as the price of a good increases—under certain circumstances, the tax authority can adjust its audit rates in one industry relative to another to change the compliance by taxpayers in each industry. This adjustment, in turn, can change the effective tax rates paid by a given industry. For instance, the tax authority might audit more heavily in an industry selling luxury cars than in one selling textbooks, yielding effective tax rates of 4.5% in the luxury car industry and 3.5% in the textbook industry.
The ability of the tax authority to generate different effective tax rates from the same nominal tax rate is an example of what we call “enforcement as substance,” namely the ability of differential enforcement strategies to effectively change the substantive law. A broader (and much less studied) question is whether “enforcement as substance” can be intentionally harnessed to improve social welfare.
Drawing upon work of mine in IP and others that shows that imperfect enforcement can decrease deadweight losses stemming from the issuance of patent rights, we show in our article that a “measured enforcement” policy (i.e., a conscious strategy to differentiate enforcement among different legal actors) can improve social welfare. In the tax context, measured enforcement can similarly be used to decrease deadweight losses caused by taxation, specifically by placing more of an onus on those industries with lower elasticities. Additionally, tax authorities can implement a quasi-Pigouvian tax on activities otherwise imposing negative externalities by upping enforcement of taxes on those activities.
Importantly, we show that under a variety of conditions, measured enforcement can yield welfare benefits while maintaining or even increasing the government’s total tax revenue. Some of these conditions are that the taxpayers be sophisticated and informed and respond rationally to incentives provided by the tax authority’s publicizing of its enforcement strategy. Generally, we believe large, sophisticated corporations—at least as an initial matter—would fit this bill.
Additionally, contrary to the title of this post—such a system should not be billed as one promoting “cheating.” Rather, like prosecutorial discretion in the criminal law context, here the tax authority would use its on-the-ground enforcement discretion to achieve optimal deterrence—namely, the use of its resources in those areas that most need it not solely to increase the public fisc, but to improve overall social welfare. Such allocation of resources of course is not tantamount to promoting cheating in the conventional sense (hence, the quotes around “cheat” in the title of this post).
One potential concern with such an approach is the possibility of abuse on the part of the taxing authority—adjusting enforcement not to benefit society, but as a means of political reward and punishment. Yet, our proposal yields no more room for abuse than under the current system. Indeed, under our framework, the taxing authority would publish its audit rates (otherwise, how could taxpayers adjust their behavior?), making its audit scheme much more transparent than it is now.
Of course, other potential concerns arise in such a scheme, such as separation of powers and horizontal equity (“treating like taxpayers alike”). We address these and other potential issues in detail in the article, concluding (of course) that none are fatal.
On a final note, thanks again to Gordon Smith to letting me guest blog the last few weeks—now, it’s back to the run-of-the-mill life as a law professor in the summer: writing law review articles (most of the time spent on the footnotes), lamenting the decline in law school enrollment, and enjoying the World Cup, Wimbledon, and the occasional barbeque.
TrackBack URL for this entry:
Links to weblogs that reference Why the Government Should (Sometimes) Help Companies “Cheat” on their Taxes: