GILTI (Global Intangible Low Tax Income) is the gift that keeps on giving – claiming US tax jurisdiction over the income of corporations owned by US “persons” on an ongoing basis. While the transition tax was painful, it was a one-off. For calendar year taxpayers, GILTI will apply starting with the 2018 US tax return – so it’s actually been in place for almost 11 months now. But the IRS has only just issued some of the relevant regulations and there are many questions that remain unanswered. Comments on the first set of proposed regulations are due on 26 November, so I’m going to start by considering a comment submitted by Arnold&Porter on behalf of the Israeli Ministry of Finance. In subsequent posts I’ll go back and discuss the purpose of GILTI and whether the actual legislation does what it says.
First a quick summary of how GILTI works. GILTI applies to all US Shareholders (a defined term in the tax code) of Controlled Foreign Corporations (CFCs – another defined term). If you’re a US taxpayer and you own a small business that is (a) organised outside the US, and (b) taxed on your US tax return as a CFC (that is, if you’re filing form 5471), then you need to consider GILTI.
The rough idea behind GILTI was to ensure that US-owned corporations weren’t moving easily portable income into low-tax jurisdictions to avoid US tax. The basic idea is that any “intangible” income inside a CFC will now be taxed as if it had been distributed (a deemed distribution). As always, the devil is in the details – and there are plenty of details, many of them designed to protect the US tax base from aggressive tax planning by multinationals with highly paid international tax lawyers.
Last month the IRS issued the first set of proposed regulations on GILTI. These proposed regulations cover the general rules on how to measure GILTI and how to allocate expenses and interest when computing GILTI. They do not cover the rules for computing foreign tax credits to offset GILTI, which will be crucial in determining the final tax liability under the new rules. Comments on this first tranche of GILTI regulations are open until 26 November 2018.
Among the comments already submitted is a comment by Arnold&Porter on behalf of the Israeli Ministry of Finance arguing for changes in the regulations with regard to the exceptions to GILTI income. It is significant that the Israeli government is (through a proxy) commenting on the internal regulations of another country. In this case, tax on GILTI is clearly an attack on the tax base of other countries and should be called out as such. I think other countries (and the EU) should be encouraged to comment as well.
Arnold&Porter provide an excellent analysis of the history of the CFC rules. Under prior law, undistributed income of a CFC was only taxable in the US if it was classified as “Subpart F” income – this was essentially passive income (e.g. interest and dividends) on which insufficient foreign tax had been paid plus some related party income. Active business income could be retained inside the CFC and US tax would be paid when the income was distributed.
With GILTI that changes. Now some of the active business income of a CFC can be classified as GILTI and included in the taxable income of US Shareholders whether distributed or not. This is a significant shift in US tax policy.
The key question is how GILTI is measured. In a nutshell, the idea is to measure “intangible income” as all “tested income” less some approximation of tangible income. If income is excluded from “tested income” it will not end up in GILTI.
The law defines “tested income” as any CFC income that isn’t subpart F income or hasn’t been excluded from subpart F income solely because “sufficient” foreign tax has already been paid (there are other exclusions, but they are not relevant in this context). Parsing the high tax exception is the main issue raised. So, what does the legislation actually say? §951A(c)(2)(a)(i)(III) excludes from “tested income”
Confusing, right? To make sense of this you need to know that §954(b)(4) excludes passive income from subpart F income if it has already been taxed at a rate greater than 90% of the US corporate tax rate (90% of 21% is 18.9%). The big question is whether this clause means that all income taxed at an effective rate greater than 18.9% is excluded from GILTI “tested income”, or whether this exception applies only to passive income that would otherwise have been subpart F income.
The proposed regulations essentially say that this high-tax exception applies only to passive income.
Of course that interpretation just doesn’t make sense. Why would active business income that’s already been taxed at rates higher than 18.9% be treated worse than passive income? Arnold&Porter argue that this is not what Congress intended and that, therefore, the regulations should be re-written to apply the high tax exception to all income, active and passive.
Of course, this fix would help US taxpayers in most high tax countries – including Australia where the tax rate on small corporations is now 25%. However, it doesn’t address the real problem – that the US is front-running the tax base of other countries.
This problem becomes clear when you read through the numerical example provided in the Arnold&Porter comment. The example considers a dual citizen resident in Israel with a software development company. Remember that GILTI is income that has not yet been distributed from the CFC that is taxed to the shareholder. But GILTI does not create a taxable transaction in Israel (or any other foreign jurisdiction where a CFC and its US Shareholder might be resident). The taxable transaction in Israel occurs when a dividend or other distribution is paid. So, when GILTI is taxed, there’s no Israeli tax to use as FTC (Foreign Tax Credit) to offset US tax. And, later, when the dividend is taxed by Israel, it’s Israeli source income, so the previously paid GILTI cannot be used as FTC.
Wait a minute, what about FTC carryovers and carrybacks? Can’t we carry the Israeli tax back and amend the original US return to reduce US tax on GILTI? Unfortunately, that won’t work for two reasons: (1) You can only carry excess FTC back one year on a US return, so if the dividend is paid years later, a carryback is not possible; and (2) GILTI is in a separate basket for computing FTC and FTC for GILTI cannot be carried back or forward at all. For an individual shareholder who does not make a §962 election, there is really no way to offset tax on GILTI with the tax paid at home on the same income.
GILTI is essentially a raid on the tax base of every country with US CFCs – especially where those CFCs are owned by that country’s own tax residents and citizens. The US is taxing the income first by creating a deemed distribution, then disallowing the use of foreign tax credits generated in any other year to offset the resulting tax liability, a case of clear double taxation. This is what other governments should be pushing back on – especially with regard to the application of this tax on individual US shareholders who are tax residents of other countries.
7 thoughts on “Explaining GILTI”
Thank you for this explanation of GILTI and the issue identified in the Arnold Porter submission.
I know that these posts take a lot of work. Toward the end of your post you write:
“For an individual shareholder who does not make a §962 election, there is really no way to offset tax on GILTI with the tax paid at home on the same income.”
But, I could ask that you do a “follow up” post which would:
1. Explain how the S. 962 election might/could work to mitigate the impact of GILTI; and
2. Apply your explanation to the example in the Arnold Porter submission.
Thanks again for a great post!
Thanks, John. I have a series of posts on GILTI planned. The operation of §962 will certainly be included.
The 5471 is the same form that USCs are supposed to file if they are “deemed“ to be shareholders in their NRA-spouse’s corporation, correct? Even if the USC has no involvement in the spouse’s company, the 5471 is technically required, I believe. So far I’ve only heard stories from actual USC business-owners about the consequences of the transition tax and GILTI, but what about the ramifications for those “deemed US shareholders“ who in reality never touch or have anything to do with the spouse’s corporation? If my assertions here are true, then this “first strike“ against other nation’s tax bases is even more scandalous and perverse than you make it sound in your article, Karen. I’m horrified. In effect, the US has found a way to tax non-citizen, non-resident persons. Please correct me if I am wrong.
Petlover, I don’t think it’s quite that bad. When determining US Shareholder status, the attribution rules are modified by §958(b) which states that shares owned by a nonresident alien individual cannot be attributed to a related US citizen.