Once More On States and GILTI
States Should Conform to GILTI (and why I believe that they will (eventually))
[Note: I consider some of these issues in greater depth here and here.]
The state corporate income tax has declined even relative to the federal corporate tax base, even as corporations have generally become more profitable. Taxing the mobile capital represented by corporate profits is difficult. Quite reasonably, some leading commentators have suggested that it is not worth it.
Along comes the TCJA with its signature corporate tax cut and purported shift to a territorial tax system . . . and a very odd thing happens. The TCJA contains significant provisions meant to backstop the erosion of the corporate tax base, not just in the US, but globally. The key provisions, GILTI and BEAT, are deeply flawed — as one would expect given this was a perversely quick attempt to address very complicated problems — but the basic intuitions behind these provisions are sound. Corporations clearly do engage in profit shifting. They do this in part by means of inter-group transactions (the target of the BEAT). Further, extraordinary profitability of corporations within a jurisdiction is also a good sign that profits were not earned there (the target of GILTI).
As has been well noted by others, the peculiar structure of GILTI in particular encourages other countries to maintain their corporate taxes at least to the extent that those taxes yield credits against what a taxpayer would owe to the US on its GILTI income. I would argue that GILTI (and the BEAT, though no state is up to this yet that I know of) represents a similar “gift” to state corporate income taxes. Consider that thirty years ago states had an innovative approach to combat income shifting — mandatory worldwide combination — and the federal government (under pressure from our leading trading partners) pressured the states to drop this method. Since then, the states have had few tools to combat income stripping. With GILTI, states need only conform to the new federal tax law in order to broaden their corporate tax base. Even if GILTI were not potentially quite significant as a tool to counter base erosion (see appendix below), the relative administrative ease with which states can broaden their tax base should make conformity compelling. (The states should also consider a return to mandatory worldwide combination.)
Naturally, the representatives of major corporate taxpayers have argued forcefully — and generally successfully, so far — that states should not conform to GILTI. They make a variety of arguments that can really be boiled down to two. First, they argue that GILTI conformity is bad policy because it means that states will be taxing foreign income, which is not their place. Second, because conforming to GILTI involves taxing foreign income, conformity is either flatly unconstitutional or must be done in such a way that vitiates its effectiveness.
As a matter of policy, as I have intimated above and elaborated more elsewhere, as have others (here and here), it makes eminent sense for states to piggyback on a chance to broaden their corporate tax base and combat base erosion. Some portion of GILTI income is displaced domestic income and, to the extent that some might not be, then all federal constitutional law requires that a state must use some reasonable method to estimate which income should never have left the domestic tax base.
As for the law, it seems clear (to me) that a reasonable method to distinguish domestic from foreign income should suffice. The legal argument that this is not sufficient relies on an aggressive reading of Kraft v. Iowa. The state in that case, Iowa, was a separate reporting state, as opposed to combined reporting (which most states with CITs are at the moment). Following the federal definition of taxable income, Iowa taxed dividends received from foreign subsidiaries, but not dividends derived from domestic subsidiaries. Despite several arguments in favor of the Iowa structure, including administrative convenience and the fact that this structure does not necessarily have any discriminatory impact, the Court struck down the Iowa law as facially discriminatory against foreign commerce.
The private bar has drawn the following rule from Kraft: a state cannot treat foreign-source income less favorably than domestic-source income. Applied to our case, the argument is that GILTI is foreign-source income and, as such, must be apportioned just like domestic source income or this will result in a discrimination.
There are multiple flaws with this argument. I will provide two here (more to come!). First, the Iowa statute was a simple binary — tax or no tax. There was no discussion of whether having a different apportionment formula for foreign-source income would be an issue. The general constitutional rules governing fair apportionment suggest that the states have considerable leeway in designing their formulas. Kraft does not stand for the proposition — and I know of no such case — that foreign-source income must be subject to the same apportionment formula as domestic income even if there is a good reason for the difference in treatment. I would argue that following the federal government’s (well supported) surmise that there is something peculiar about GILTI is a very good reason.
Second, the Court in Kraft emphasized that it was treating the dividends at issue in Kraft as foreign-sourced: “[T]he only subsidiary dividend payments taxed by Iowa are those reflecting the foreign business activity of foreign subsidiaries.” Thus the Court did not reach the question of whether it would be constitutional for a state to treat some nominally foreign earnings as actually earned domestically.
If the policy and legal arguments in favor of GILTI conformity are so strong, why has there been so little state response? Certainly the states have a lot to deal with and these issues are complicated and novel. The advocates for not conforming are, of course, excellent. But in the end I think the main reason the states have not acted is that they have not needed to. Times are still good, more or less. But there will be a downturn and states, subject to balanced budget constraints, will need to decide whether to cut essential services or raise taxes. Raising taxes is very likely the better option as a matter of policy, though likely harder politically. But perhaps not that much harder politically if more taxes are to be raised by conforming to the federal provisions meant to combat base erosion.
Short Appendix on the state revenue potential of GILTI
The JCT expects GILTI alone to increase federal revenue by about $11 billion/year; Kimberly Clausing estimates about $8 billion/year. Let’s take Professor Clausing’s number; if we work backwards from the rate (10.5%) and the 50% deduction, then total GILTI is $150 billion. Suppose a state uses the methodology I (and David Gamage) proposed for taxing the repatriation (and by the way, states should still do so!), then a state would ascribe about one-third of the $150 billion to the US (or $50 billion) and then apportion the $50 billion to the state using the state’s share of GDP. In California’s case, that would be 12% or $1.3 billion, which, multiplied by California’s tax rate (8.84%), yields about $100 million. But note that this number might turn out to be much larger. First, some prominent commentators think that large MNCs are going to have a very hard time reducing their GILTI because GILTI is measured as a percentage of assets held abroad. Second, remember that the JCT and Professor Clausing are calculating how much the federal government will net from GILTI, and the federal government, but not the states, extends an 80% foreign tax credit. I have not seen an estimate as to how much eliminating the credit affects the calculation, but it is surely significant.
(Note that the private bar has argued that the absence of foreign tax credits indicates that somehow the states would be failing to reach only displaced domestic income if they conform to GILTI. This would be a stronger argument if GILTI did not permit taxpayers to blend tax credits from different jurisdictions. Thus the approach of the federal government is likely quite over-generous. Further, the states can quite reasonably argue that excess profitability is sufficient evidence on its own for displaced income and that they can tax it so long as they use a reasonable apportionment formula, which formula takes the place of the provision of foreign tax credits.)
