Reforming State Corporate Income Taxes Can Yield the States Billions
This is a great time for the states to make changes that they should have made anyway.
By Darien Shanske, Reuven Avi-Yonah and David Gamage
Introduction
The federal government should provide states and localities with a hundreds of billions of dollars in aid. However, it is unlikely the federal government will do enough, fast enough. States and local governments, which generally operate under balanced budget constraints, are, accordingly, already making sweeping cuts. Such cuts will deepen the recession and cut services when they are most needed.
Rather than make such cuts, it would be better to raise taxes on those who can afford to pay. In this post, we will focus on one such set of taxpayers — large multinational corporations that have long evaded both the state and federal corporate income taxes. Better yet, the reforms we propose represent good tax policy more generally: they are fair, efficient and administrable.
Before proceeding, a skeptic might note that state corporate income taxes represent only about 3% of state taxes or about $53bn per year, and one prominent estimate of the states’ need is $500 billion, other estimates reach $1 trillion. There is no question that the reforms we will propose will not raise $500 billion. However, all of the reforms below do have the potential to raise substantial amounts of revenue. For instance, one reform, moving to worldwide combination, has been reasonably estimated to have the potential to raise about $17 billion per year for the states. That is a substantial amount on its own. Several of the other reforms below are of similar magnitude. Our back of the envelope estimate for a deemed repatriation tax, outlined below, indicates a revenue potential for the states that could be as high as $25 billion or even significantly higher. To be sure, the current recession might suppress some of these projections, but maybe not by that much given that some firms are going to be very profitable during this pandemic. Furthermore, if states put these reforms in place now their revenues will not only bounce back faster, but they may be able to borrow against this anticipated future revenue right now.
That the numbers here can be fairly large should not be too surprising; the yield of the state corporate income tax has fallen sharply over the last few decades, even relative to the decline in federal corporate income taxes.
[Chart source here.]
It is not a good thing, of course, that the states have let their corporate income taxes wither in this way. However, in the current crisis this represents an opportunity. The states need only engage in some ordinary good tax housekeeping to revive their corporate income taxes. Though there may well be a need for stronger medicine at some point, there is no reason not to start with reforms that are easy and sensible.
The proposed reforms, a thumbnail sketch:
1. Tax the repatriation: Multinational corporations stashed $2 trillion in profits abroad. Many of those profits were earned in the US and should have been taxed by the states, but have not.
2. Shift to mandatory worldwide combination: Multinational corporations are still shifting domestic profits abroad. The states can counter this by (reasonably) including these foreign subsidiaries in their corporate tax base.
3. Conform to GILTI: The new tax law signed by President Trump in 2017 made a half-hearted attempt to combat income shifting. Though shifting to worldwide combination would be better, conforming to GILTI at the state level is the least a state should do.
4. Income tax surcharge: A handful of major corporations will be very profitable over the next few years. States should add a temporary income tax surcharge to tax some of these profits.
5. Suspend certain tax credits: States have been very generous in giving large corporations credits from the corporate income tax. Credits of this magnitude were probably never a good idea, but, given the current crisis, it is a particularly apt time to suspend these credits.
6. Reform the sales factor: States generally divide the income of multijurisdictional corporations based on their share of a corporation’s sales within the state. Some taxpayers have gotten very good at manipulating the location of sales, but there are ways to make the formula harder to game.
Under the prior international tax law regime that ended last in 2018, the United States sought to tax the profits of multinational corporations on the basis of their worldwide income. At the same time, the old regime permitted U.S.-based multinationals to defer payment of tax on profits earned overseas until that money was brought home. Until a foreign subsidiary repatriated its profits to its US parent, the profits were not subject to U.S. tax. Naturally, large multinationals corporations left a lot of money — over $2 trillion — stashed abroad.
Much of that money represented profits on sales to U.S. residents and/or profits from the sale of intellectual property developed in the United States. Under the prior tax regime, U.S.-based multinationals deployed a number of well-known techniques to strip the profits from sales to U.S. customers and thus avoid being taxed on those profits by U.S. tax agencies, either federal or state. Those profits were then secreted abroad to escape tax.
The 2017 tax law deemed all of the deferred income repatriated and subject to tax. Unfortunately, the law then applied a special low tax rates to these profits, effectively exempting most of the profits from tax.
But all is not lost. States can still tax this deemed repatriation. And they can use the money to keep the lights on during the current recession. Most state corporate tax laws did not and do not reach the repatriation, but these laws can be changed so as to reach this income. Alternatively, and perhaps even better, state governments should subject the repatriation to a special one-time tax surcharge.
There are many reasons why taxing the repatriation is a particularly good way for state governments to raise desperately needed revenue. Consider four.
First, states would be recouping a national loss. The repatriated profits of multinational firms reflect a form of national savings that is now being squandered. While these profits went untaxed for years and years, other taxpayers picked up the slack and critical national initiatives went unfunded. Now, as the money returns home, its value to the rest of us has been gutted by low tax rates and the predictable use of the repatriated revenues for corporate stock repurchases rather than job creation.
Second, much of the untaxed profits squirreled abroad also escaped state-level taxes. Thus, for states, taxing these repatriated profits reflects satisfaction of an overdue tax bill avoided for years.
Third, since the profits have been earned, taxing them does not affect the corporations’ behavior or their competitive position.
Fourth, though we cannot offer precise revenue estimates, subjecting the repatriation to tax can raise large sums of money, which makes sense because even a small slice of $2 trillion is a large number.
Here is a back of the envelope estimate. Though we think that the states can tax all of the repatriation, let’s suppose instead they tax the 84% of the repatriation that many taxpayers have not yet paid tax on at the federal level because the federal tax law gave taxpayers a backloaded deferred payment option. That leaves “only” $1.6 trillion. Now states need to estimate what percentage of that revenue was shifted out of the US and what percentage was really earned abroad. A defensible estimate, based on the relative size of US GDP and/or empirical work on how much income is shifted out of the US relative to other countries, might arrive at 35%. That leaves us with about $560 billion to be taxed. We think a rate as high as 20% could be justified given the taxes these taxpayers avoided over the years, but suppose instead the states went for 5% or closer to the median CIT rate. This would raise $25 billion.
II. Mandatory Worldwide Combination
The states have long confronted the problem of how to tax the income of a multistate business. In general, as a matter of constitutional law, each state can only tax income generated within the state, but how does one calculate that for an integrated multistate business like a railroad or Apple? If one asks the taxpayer to do the calculation, then, naturally, the taxpayer will argue that most of its income is generated in a low-tax state or a no-tax state.
The states arrived at a unique and uniquely effective solution to this problem. Instead of asking a multistate business to divide up its income, the states asked the business to report all of its income, including income nominally earned abroad, and then apportioned some of the income to each state by means of a formula. The most common modern formula uses the percentage of sales within a state because, among other reasons, the location of a firm’s customers is difficult to game. So how much of Apple’s total income is generated within a state under this system? The answer is the same percentage as the percentage of Apple’s sales within a state.
Note that this method — mandatory worldwide combination — eliminates the incentive to shift income to a low-tax jurisdiction. It does not matter where the income is nominally earned because it all goes in the same pot and is multiplied by the usual formula (percentage of sales) to apportion the income.
The US Supreme Court upheld mandatory worldwide combination twice. Unfortunately, in the 1980s, our trading partners pressured the federal government, which in turn pressured the states, not to use mandatory worldwide combination. Instead, states offered — and still offer — MNCs what is called a “water’s edge” election, which, for the most part, allows MNCs not to combine income earned abroad. Naturally, the availability of a water’s edge election gives MNCs incentive to shift income abroad so as to escape state taxation.
Eliminating this election should therefore allow the states to tax a substantial portion of the income that MNCs continue to shift to low-tax jurisdictions. As noted above, one respected estimate is that the states could raise about $17 billionthrough moving to worldwide combination.
III. Conforming to GILTI
“Global Intangible Low-Tax Income” or GILTI is a category of income that was added to the federal tax code by the 2017 tax law. GILTI income is income nominally earned by the foreign subsidiary of a US corporation that the federal tax law deems to really have been earned someplace else, such as the US. In other words, by means of a formula, the federal corporate income tax uses GILTI to combat the same income shifting problem that is the target of worldwide combination.
Though there is a little complexity, all a state with a corporate income tax has to do is subject GILTI to state corporate tax and then that state too is combatting income stripping. The state can use the same formula to divide up the income of a multinational corporation that it would ordinarily use, more or less.
A state cannot conform to GILTI and adopt worldwide combination; they are substitutes because both are attempts to ferret out shifted income. A state could offer taxpayers a choice between them. As between worldwide and GILTI, worldwide is better because it simply includes all the income of a multinational corporation without the intrusion of the complicated federal formula for picking out suspect income. That said, conforming to GILTI may well be easier politically, as it requires little more than adding a sentence to the state corporate income tax that the state is taxing GILTI. Note that such conformity would raise a lot of revenue. The Penn Wharton Budget Model provided an estimate of $382bn in total GILTI for 2020. Assuming the states were to follow the federal government and tax half of GILTI and do so at a 5% rate, then taxing GILTI would yield the states almost $10bn/year.
IV. Temporary Corporate Income Tax Surcharge
Many businesses will suffer losses during the current recession. A few will prosper. It is not imposing a moral judgment to suggest that businesses that do relatively well should pay more in taxes to offset the increased budgetary costs of the downturn. Put concretely, we may be grateful for Amazon for its delivery service, but it still seems appropriate to tax it on its profits so that there will still be a main street when this is all over. It might arguably be particularly appealing to tax a firm that is profiting during the recession on its “excess” profits, but designing and implementing a new tax would be a heavy administrative lift.
Fortunately, the regular corporate income tax is already something of an excess profits tax, to the extent that, by definition, it will only tax profitable firms and those firms are likely to be more rare during the deep recession we are entering. As to those few profitable firms, a temporarily higher corporate income tax rate would reflects the need that we are now in a time where putting a greater share of profits towards the common good is especially urgent. For instance, suppose that a state were to adopt a 3% CIT surcharge for the next three years. This could work to raise some extra revenues in the depths of the recession from those most able to pay.
V. Temporary Suspension of Certain Corporate Tax Credits and Deductions
States offer a multitude of tax credits. The general consensus is that these do little to encourage economic development, but this is not the time to fight that battle. What we want to emphasize here is that certain tax credits have grown so large that they are undermining the state CIT and could dramatically reduce the value of the reforms above. Our primary target is state R&D credits. These are very expensive, costing almost $2 billion/year in California alone. That is about 20% of all California collects in CIT. Even worse, R&D credits can be stockpiled. It is likely that many of the taxpayers that shifted profits have also been stockpiling credits. We hope that a time will come to re-consider state R&D credits more wholesale. But, as with the temporary surcharge we proposed above, states especially need to be able to tax actors with greater ability to pay now, in this economic downturn. R&D credits should thus be suspended temporarily, for (say) three years.
VI. Reform the Sales Factor
As noted above, states have now shifted to dividing up the income of a multinational corporation by using a formula based on the percentage of sales within a state. This approach has turned out to be expensive. In 2015–16, the last time it calculated this number, California estimated that it loses about $1 billion per year because of this shift, about 10% of its total CIT collection. How can that be? California is a big market state after all. The reason is that taxpayers have gotten very good at gaming the sales factor. Reforming the sales factor could therefore raise a lot of revenue, especially if some of the other reforms listed above are also adopted. We propose two reforms below, with more details available here (see end of paper).
First, states generally now permit certain sales to be located where a middleman, e.g., a wholesaler, takes title to the goods. The problem with this rule is that it encourages taxpayers to sell to intermediaries in low-tax jurisdictions. The law of the sales factor should thus be refined so as to apportion a sale to its ultimate destination. For example, a corporation could look to information it retains in the usual course of business because surely businesses generally know where their customers are. If the corporation does not have this information and cannot obtain it from its wholesalers, then it could use any reasonable method, including population as a way to fill in the gaps.
A second reform we propose to the sales factor would be for corporations to be required to submit an accounting of where they would locate all their sales using the methodology they have used for the state in question. The corporation should also report where they are reporting sales to other states with the single sales factor. It might not seem odd to a California auditor, for instance, if California’s sales factor for a particular corporation was only 8%, so a bit below the state’s share of US GDP, but it would rightfully raise alarm bells if the disclosure of the results of the methodology revealed a 10% sales factor for Nevada — or the Cayman Islands.
