COMBINING AN ENTITY-LEVEL CONSUMPTION TAX, IMPROVED SALES FACTOR APPORTIONMENT, AND A TAX ON A FEDERAL WINDFALL (THE QBI DEDUCTION).

• Author: Shanske, Darien

INTRODUCTION 450 I. STATE TAX POLICY OPTIONS 455 A. SALT Restoration 455 B. Pickup Federal Revenue 457 C. Expand State Fiscal Base with Entity-Level Consumption 460 Taxes: Four Good Reasons 1. Consumption Taxes Encourage Investment 460 2. Consumption Is Less Mobile than Income 460 3. A New Entity-Level Tax Can Tax Consumption Efficiently 462 4. The New Entity-Level Tax Can Tax Consumption Not 464 Currently Taxed (but Should Be) D. Into the Weeds with Entity-Level Consumption Taxes 465 E. Sidebar on State Corporate Income Taxes 469 F. Back to the TCJA 471 G. The Proposal, Finally 473 1. Consumption Tax 474 2. Tax on QBI 474 II. FORMULARY APPORTIONMENT: THE WEAK LINK? 476 A. Introducing Formulary Apportionment 480 B. UDITPA and the Sales Factor 483 C. The Constitution and the Single Sales Factor 485 D. Improving the Single Sales Factor 487 1. Background: Theories of Tax Compliance 487 2. Reform One: Craft a More Certain Route Amidst 489 Uncertainty 3. Reform Two: Focus on the Ends 497 4. Reform Three: More Information Please 499 CONCLUSION 499 INTRODUCTION

It is a truth universally acknowledged (1) that states do not have the fiscal capacity to take up the slack if the federal government dramatically cuts its support for healthcare or other safety net program. (2) Alas, it is the case that not only is the current federal government attempting to--and succeeding in (to some extent)--cutting safety net programs, the federal government has already taken a big step toward undermining the ability of states to continue financing the level of social services they currently finance. The big blow here was the partial repeal of the state and local tax (SALT) deduction; this deduction was, to a considerable extent, a subsidy for states that chose to fund more generous safety net programs using progressive taxation. (3)

But perhaps matters are not as grim as they seem for a state looking to do more for its citizens than what is set by a lowering federal baseline. We know that states have less fiscal capacity than the federal government, but they still might have quite a lot. I argue that in a situation--our situation--where the federal government does not tax very much or very well, and the state governments also do not tax well (their primary taxes date to the 1930s or earlier), then states have quite a lot of fiscal capacity they can tap. This Article is the first in a series to explore how states that want to increase their fiscal capacity can proceed.

The capping of the SALT deduction has motivated states to think more about their tax systems, and it turns out that states have several ways to restructure their revenue systems so as to at least mitigate the cost of the repeal of the SALT deduction. (4) Yet, at first blush, all these mechanisms can achieve is to return states to where they were in 2017. No one thinks state tax systems were in great shape in 2017. Thus, returning to 2017 would not help state citizens in the face of sweeping federal cuts nor would it raise revenue so as to finance more social services.

The basic idea for expanding state fiscal capacity is this: the federal government is leaving money on the table that the states can and should pick up. (5) Here is one relatively straightforward example created by the recent overhaul of the federal tax law: tax Qualified Business Income (QBI). As has been extensively discussed, the TCJA, for no particularly good reason, chose to tax one kind of business income at a lower rate than others--QBI. (6) The states can and should simply impose a surcharge on QBI so that the revenue lost to this ill-conceived federal tax benefit can still be used for public purposes. The choice to advantage QBI is, in fact, so misguided that it is arguable that a state that chooses to tax this benefit away is doing a favor to its economy by depriving its businesses of an incentive to engage in pointless restructurings so as to generate more QBI.

A tax on QBI would thus be a step towards picking up on a new unforced error committed by the federal tax system. But there is a still deeper flaw in the federal tax system. The flaw is that, alone among advanced economies, indeed among virtually all economies, the United States does not levy an advanced consumption tax, specifically a Value Added Tax (VAT). (7) As a result, in 2014, the last year for which data is available, consumption taxes in the United States only amounted to 3.8% of GDP, compared to the OECD average of 10.3%. (8)

This is bad news for us as a nation, but good news of a sort for the states because the states can tax the consumption that the federal government does not. And 45 states do tax consumption through retail sales taxes; (9) unfortunately, retail sales taxes are a far inferior way to tax consumption relative to a VAT. Worse news is that it is very unlikely that a subnational government can levy the most common and effective form of VAT--the credit-invoice method--without help from the national government. (10) And so it looks like we just demonstrated the melancholy, universally acknowledged truth with which we began. States are doing the best they can with the imperfect tools that they have.

There are, however, more esoteric forms of the VAT that could be levied at the state level. The credit-invoice method VAT, the VAT in common use around the world, like the typical U.S. retail sales tax, is ultimately collected on a transaction-by-transaction basis from consumers. The more esoteric forms of a VAT would be administered at the business entity level once a year, much like a corporate income tax, but the calculations would be different so that consumption, rather than income, would be taxed. The key insight as to why a VAT can be collected from business entities is straightforward. All personal consumption involves a buyer and seller; there is no reason you can't tax the seller at the end of the year on all the sales she has made.

"Wait," you say, "you are proposing to set up a whole new entity-level business tax that you acknowledge is suboptimal for taxing consumption?" Yes. First, as will be explained below, these new taxes, along with the old state retail sales taxes, could end up together taxing consumption pretty efficiently and better than using state retail sales taxes alone.

Second, federal tax reform has made the creation of business entity taxes more appealing than ever. This is because federal tax reform has capped the ability of individuals to deduct state and local taxes; it has not capped the ability of businesses to deduct state and local taxes. Thus, a shift to the use of business entity taxes is a shift to the use of federally deductible dollars to fund state and local services.

Third, the proposal here is not to tax consumption through a new entity-level tax at a rate comparable to European VATs or even state retail sales taxes. Because of the structural flaws necessitated by the fact that this is a subnational VAT operating without central government support, a high rate of tax could likely spur a lot of inefficient tax gaming. But a small rate of tax, 2% at the maximum, ought not motivate large-scale evasion. Furthermore, the second part of this Article explains how a key component of this new tax can be strengthened to further prevent

gaming.

Note that the inability to impose a high tax rate does not mean that such a tax would not raise substantial revenue because this would be a tax on a very broad base. When California considered a tax similar to the one proposed here in 2009, the proposed rate was 5% and the projected revenue was $39 billion. (11) If a 2% version of this tax raised even $10 billion/year for California, that sum would have significantly inoculated California from the cost of the almost passed proposals to abolish the Affordable Care Act. (12)

Indeed, as this article was going to press, Oregon passed an approximate version of the type of tax proposed here. The rate is low (0.57%) and the base is broad (gross receipts), though narrowed to prevent too much pyramiding because of business-to-business sales. (13) The tax is expected to raise about $1 billion per year for education, (14) which is quite significant given that the State of Oregon currently spends about $12 billion per year for education. (15)

Even if one is convinced on policy grounds, one might wonder how such taxes work. If we tax a big multistate business once a year on its sales, then how are we to know which sales happened in a particular state? The location of a sale is potentially a nebulous thing. Consider a scenario where a national accounting firm is providing accounting services to a national rental car firm. Can the taxpayer make a reasonable apportionment of its sales based on, say, where a car is rented?

Such a solution is called formulary apportionment, and the Supreme Court has blessed reasonable formulas since the nineteenth century and will, in all likelihood, continue to do so as any more searching approach amounts to a significant assault on a key component of state sovereignty in the name of the dormant Commerce Clause.

A further question is how well this formulaic approach will work in the face of determined tax evasion. Again, this concern is partially addressed through the use of a low rate, but given that taxpayers are already acting to game the same type of formulas in connection with the state corporate income tax, this is a real concern. Fortunately, there is good reason to believe that formulary apportionment can be improved so as to tamp down the gaming. As with the basic proposal--impose an entity-level consumption tax--this secondary reform, namely, improve formulary apportionment, should happen in any event because of its importance for the corporate income tax. I will discuss how this can be done with particular reference to the literature on tax compliance.

To return to the beginning, states have more fiscal capacity than they think, but accessing that capacity...