This article appears in the Summer 2018 issue of The American Prospect magazine. Subscribe here.
The 2017 Republican Tax Act, as passed by Congress and signed into law on December 22, 2017, represents the most far-reaching reform of the U.S. international tax rules since 1962. Most importantly, for the first time since the income tax was enacted in 1913, it changes the rule that U.S. resident taxpayers have to pay tax on all income “from whatever source derived.” Under the Tax Act, dividends paid to U.S. corporate shareholders from their foreign subsidiaries are exempt from U.S. tax. That remains true even if the dividend was paid out of earnings that were not subject to foreign tax in the country where the subsidiary is incorporated.
Despite claims that tax reform would simplify the tax code or produce more domestic investment, this provision is an open invitation to U.S. multinationals to shift even more of their income to their foreign subsidiaries, because before the Tax Act, the main disincentive to such shifting was the difficulty of repatriating the income without incurring tax on the dividends. However, supporters of the law claim it will not create an incentive to shift because of certain anti-abuse provisions. In addition, supporters say that it will eliminate the incentive for U.S. multinationals to migrate offshore through “inversion” transactions, and that the repatriation of previously accumulated income will create U.S. jobs.
Unfortunately, none of these claims is true. The Tax Act creates additional incentives to shift income offshore for purposes of tax avoidance, and what is worse, it creates incentives to shift actual jobs. In addition, the incentive to engage in inversion transactions remains. Finally, the post–Tax Act evidence shows that repatriations under the law are used to reward shareholders, not create jobs.
Will the Tax Act Stop Income Shifting?
The basic structure of U.S. international tax rules encourages income-shifting out of the United States. That is because since the origins of the income tax in 1913, U.S. residents were subject to tax on worldwide income, but foreign residents were only subject to tax on U.S.-source income. Moreover, a foreign corporation was defined as a non-U.S. resident for tax purposes even if it was a pure shell that was entirely managed and controlled from the United States. Thus, it was easy for a domestic corporation to create a foreign subsidiary and arrange to book profits offshore. This rule differs from the definition of residency adopted by the United Kingdom and most other industrialized countries, which define any corporation that is managed and controlled domestically as a resident for tax purposes.
Until the 1990s, the resulting incentive to shift income out of the United States was mitigated by a set of anti-abuse rules known as Subpart F, which was enacted in 1962 (over intense opposition by U.S. multinationals). Under Subpart F, U.S. parents of foreign subsidiaries were taxed currently on investment income and other types of mobile income that was likely subject to little or no tax offshore. As recently as 1997, Subpart F ensured that the benefit of shifting income offshore was sufficiently limited to the point that taxing all foreign subsidiaries of U.S. multinationals would only have raised less than $10 billion.
However, in 1996, the Clinton administration either blundered or deliberately did a huge favor for corporations, and enacted a regulation called “check the box,” which enabled U.S. multinationals to shift mobile income among their foreign subsidiaries without triggering Subpart F. The result was an astonishing growth in offshore accumulation, which exceeded $3 trillion by the time the Tax Act was enacted and mostly reflects income from intangibles developed in the United States. In tax jargon, an intangible refers to software, patents, and other types of intellectual property that can be treated as being located offshore even if they were developed in the United States and are protected by U.S. patent and copyright laws.
Apple is a prime example of how U.S. tax law before the Republican bill incentivized profit-shifting. The company’s profits are based on software developed in the United States. The software was shifted to Bermuda using a technique called cost-sharing, which is approved by the IRS. Basically, Apple created a paper subsidiary based in Bermuda, which participated in the cost of developing the software for a new Apple product. Apple could contribute this money to the subsidiary. Once a product had been developed, Apple could share the profits with the Bermuda subsidiary at the same percentage rate as the cost. Apple then set up two Irish companies, one of which employed actual people in Ireland, while the other was treated as a Bermuda resident for Irish tax purposes. In between the two Irish companies, Apple set up a Dutch company. The profits from the sales of Apple products outside the United States were then attributed to the operating Irish company, but then shifted via the Dutch company to Bermuda, which does not tax those profits. The result of this “double Irish Dutch sandwich” structure was that Apple accumulated $128 billion in Bermuda with an effective overall tax rate below 5 percent. Most of these profits stemmed from software developed in the United States.
The Tax Act on its face will do nothing to reverse this trend, because it eliminated the tax on dividend repatriations, the last barrier to profit-shifting. However, Tax Act supporters point out two provisions designed to counter the incentive to shift: the Global Intangible Low-Taxed Income (GILTI) and the Foreign Derived Intangible Income (FDII) provisions.
Under GILTI, U.S. parents of foreign subsidiary corporations are taxed on GILTI, which is defined as their profits in excess of a 10 percent return on tangible investments offshore. This provision is designed to limit profit-shifting incentives, especially for multinationals like Apple, Google, and Amazon that derive their profits from U.S.-developed intangibles and have low levels of tangible investments offshore.
However, GILTI suffers from two major problems. The first is the tax rate, which at 10.5 percent is half of the new U.S. corporate tax rate of 21 percent. This lower rate creates a continued incentive to shift profits offshore, even if they are subject to GILTItax. The lack of tax on future repatriation dividends strengthens this incentive.
The second problem is the definition of GILTI as income in excess of a fixed return on tangible investments. This provision creates an incentive to shift actual jobs offshore, because the higher the investment in factories overseas, the lower the GILTItax. Thus, if Apple wants to avoid the new GILTItax on its Irish profits, it should expand its actual operations in Ireland, where the top corporate tax rate is 12.5 percent—significantly lower than 21 percent. Or it could choose to migrate its operations to other locations that have an even lower effective corporate tax rate. There is a proliferation of special tax zones and patent boxes in the European Union and elsewhere that offer companies like Apple an effective tax rate well below 5 percent.
Proponents of the Republican Tax Act realize this, and in response enacted FDII. FDII is the mirror image of GILTI, and provides for a lower rate on domestic income from exports that exceeds the same 10 percent return. The idea is to encourage exporting from the United States, rather than from overseas.
But FDII has its own problems. First, the tax rate under FDII is 13.125 percent, which is higher than the GILTI rate, so the incentive to shift remains. Second, under GILTI, income below the 10 percent “hurdle rate” of return on tangible investment is simply not subject to U.S. tax currently or upon repatriation. Under FDII, on the other hand, income below the hurdle rate is subject to the full 21 percent U.S. corporate tax.
Finally, FDIIis a blatant violation of the World Trade Organization’s export subsidy rules, which prohibit lower tax rates for income from exports. It is highly likely that at least one of our major trading partners will challenge the FDII in the WTO, and the almost certain result would be a ruling against the United States that would enable hefty sanctions. We have been through this in the past, as the WTO has struck down U.S. export subsidies repeatedly, most recently in 2004. If this happens, the United States will have to either accept the sanctions, which can be imposed on any U.S. goods (oranges from Florida were the target in 2004, a presidential election year), or the United States can abolish the FDII, like it did with the export subsidies in 2004. In that case, GILTI will create an even more powerful incentive to shift income overseas.
The Congressional Budget Office has confirmed that the incentive to shift profits out of the United States remains after the Tax Act:
The GILTI and FDII provisions affect corporations’ decisions about where to locate tangible assets. By locating more tangible assets abroad, a corporation is able to reduce the amount of foreign income that is categorized as GILTI. Similarly, by locating fewer tangible assets in the United States, a corporation can increase the amount of U.S. income that can be deducted as FDII. Together, the provisions may increase corporations’ incentive to locate tangible assets abroad.
Before your eyes glaze over at the sheer complexity, here is the bottom line: In essence, the Republican sponsors and corporate lobbyists were torn between two rival impulses. First, they wanted to cut corporate taxes as much as they could get away with. But second, they wanted to create the appearance of reform and limit the more flagrant forms of offshore tax avoidance, which was also necessary to keep the Tax Act from further ballooning the deficit. Basically, the first impulse won. But in their effort to limit the worst abuses, the bill’s drafters added several layers of complication that will benefit mainly accountants. At the end of the day, offshore corporate schemes to avoid U.S. taxation enjoy an even larger menu of maneuvers to choose from, and ordinary taxpayers who cannot resort to offshore dodges are fleeced more than ever.
Will the Tax Act Stop Inversions?
Before the new law, there were two waves of “inversions,” which are transactions in which the U.S. parent of a multinational becomes a subsidiary of a new foreign parent—essentially a sham transaction to avoid taxes. In the first wave, from 1994 to 2004, the foreign parent was typically a shell incorporated in Bermuda. This type of inversion was stopped in 2004, but the result was a second wave in which the new parent was incorporated in low-tax locations like Ireland and owned both the old U.S. parent and a foreign entity that was at least 40 percent as big as the U.S. group. The Obama administration waged a long fight against these types of inversions and was able to stop some of them, but others continued.
The Tax Act was supposed to stop inversions, because one of the reasons to invert was to avoid the tax on dividend repatriations, which the act abolished. However, many inversions were motivated by other factors, such as the desire to load up the old U.S. parent with debt and reduce its effective tax rate through interest deductions. The Tax Act puts some limits on these types of deductions, but not enough to eliminate the appeal of inversions.
More importantly, the first wave of inversions was motivated to a large extent by the desire to avoid Subpart F. This rationale was eliminated by the adoption of “check the box” in 1997, but under the Republican Tax Act, it comes back in the form of GILTI. Any multinational affected by the GILTItax can avoid it by inverting. Thus, one would expect inversions to continue. For example, Ohio-based Dana Incorporated announced on March 9, 2018, that it was planning on inverting to the United Kingdom. In The Wall Street Journal, the company’s CFOsaid that “even with the new tax legislation, there is a benefit for us.” The company expects that this move will reduce its tax liability by around $600 million over several years.
In addition, the Obama administration faced major problems with combating second-wave inversions because (with the GOPcontrolling both houses of Congress) it could not change the statute. Specifically, it wanted to define the acceptable threshold for the foreign partner in a true cross-border merger at more than 50 percent rather than 40 percent, and also define U.S. tax residency of the resulting group depending on from where the parent was managed and controlled. However, Congress refused to enact these provisions, which would have effectively killed inversions, since no inversion ever results in moving the actual corporate headquarters. The Tax Act provided a unique opportunity to enact these proposals in the context of a thorough rewrite of the tax law, but that opportunity was missed—the law contains almost no new anti-inversion provisions, and those that it does contain do not apply to second wave–type inversions.
Will the Tax Act Create Jobs in the United States?
As we have seen, the Tax Act incentivizes multinationals to move jobs out of the United States because of the structure of the GILTI rule. But proponents of the Tax Act argue that it will create U.S. jobs because it eliminates the tax on repatriations, so that most of the previously accumulated $3 trillion will flow back to the United States.
While it is too early to tell, the evidence so far is that the Tax Act has not resulted in significant job creation: The current low unemployment rate environment preceded the bill. And as of a Senate Finance Committee hearing on May 8, 2018, the Tax Act has resulted in more than $480 billion in stock buybacks, but only $7 billion in bonuses to employees.
We do have significant evidence from a previous attempt to create jobs through reduced repatriation taxes. In 2004, Congress enacted a one-year amnesty for repatriations with a rate of 5.25 percent (instead of 35 percent). About $300 billion was repatriated, but no jobs were created—the funds were used for share buybacks, just as they are being used now. Indeed, most of the companies that took advantage of the 2004 amnesty laid off workers at the same time. There is no reason to assume the experience under the Tax Act will be different.
What Can the Democrats Do?
The international tax provisions of the Tax Act fail to achieve their most important goals. They do not eliminate the incentive to shift profits out of the United States, and they create an additional incentive to shift jobs. They do not stop the incentive to invert, and they are unlikely to create any jobs.
The Tax Act is unlikely to be amended as long as the GOP controls Congress. But let’s assume the Democrats take over in November. What changes should they make?
One option is of course to try to repeal the entire law. But while I would support repeal of some of its provisions, like the terrible new regime for taxing pass-throughs, I think the approach to the international tax provisions should be different.
While the Tax Act is deeply flawed, on the international tax front it does represent an improvement over prior law. In particular, GILTI ensures that U.S. multinationals will not be able to accumulate trillions offshore without paying any U.S. or foreign tax. Other provisions ensure that both U.S. and foreign multinationals will find it harder to strip taxable income out of the United States.
Thus, I would propose amending GILTI by a) eliminating the exemption for income below the hurdle rate, which incentivizes job shifting, and b) setting the GILTI rate at 21 percent, like the normal U.S. corporate tax rate. At the same time, FDIIcan be repealed, because it will lose its anti-shifting rationale (and is likely to be repealed in any case if the United States loses a WTO challenge).
These changes will create a regime in which U.S. multinationals will be subject to a 21 percent tax on their entire worldwide income. They eliminate any incentive to shift income or jobs overseas. At 21 percent, the tax rate is low enough that it does not create competitiveness concerns, because that is the approximate average effective tax rate of our main competitors. These foreign effective rates are likely to go up when the EU Anti-Tax Avoidance Directive goes into full effect in January 2019. While there has been an overall decline in corporate tax rates in the European Union, that has been balanced by expansion of the corporate tax base, and the EU (minus the United Kingdom) is unlikely to engage in a “race to the bottom” following the Tax Act.
Once the income is currently taxed at 21 percent, there is no reason to tax it again when it is repatriated, so the Tax Act’s repatriation provisions can be retained.
There will be a stronger incentive to invert under these proposals to avoid the 21 percent tax on GILTI, but that can be addressed by enacting the Obama anti-inversion proposals. No U.S. inversion ever resulted in an actual shift of corporate headquarters, and there are no inversions in countries that define corporate residency based on the location of headquarters.
In 1961, the Kennedy administration proposed eliminating the shifting incentives by currently taxing U.S. multinationals on their foreign income. This proposal was rejected out of competitiveness concerns, even though at the time U.S. multinationals had no serious competition. In 2018, there is more competition, but with a corporate tax rate of 21 percent and EU moves to raise the effective tax rate, the Kennedy proposal can be enacted with no adverse effects on U.S. multinationals. That is the only way to prevent continuing incentives to shift income and jobs overseas. And if the European Union is serious about cutting tax competition and raises rates, the United States can also increase the rate on corporate income, which is now at a postwar low.