A Tale of Two Min Taxes

The Democrats’ reconciliation package is back on track, and one of its key pillars is the 15% corporate minimum tax.

No, not that one–this is the “book” minimum tax, based on financial accounting information. Not the global minimum tax, even though both are 15%. And both are book minimum taxes. And both are global.

And no, it’s not that other minimum tax, the one that was in the Biden administration’s 2023 Green Book. That one (also book, but not global) is supposed to be paired with the global minimum tax, although it’s unlikely we’d do that, even if we passed the global min tax. Which it also looks like we’re not going to do.

Instead of passing the global min tax that happens to be book, we’re passing the book min tax that happens to be global.

OK, now pay attention, because this is where it gets confusing.

Actually, let’s back up a bit. The accounting world has traditionally been divided into two sides, book and tax. Book–referring to the financial books, where the records are kept–is how information is organized for internal use and for public filings to shareholders. Tax is for, well, taxes.

They’re two parallel systems which are supposed to be kept separate. But they do have one common goal–to measure the profit of a firm or venture. How much money is it making? The shareholders, and Uncle Sam, want to know. It sounds like they should arrive at the same number, but profit is a tricky, elusive notion–essentially a man-made concept, not appearing "in nature," contingent on many seemingly arbitrary determinations.

In recent years, the two supposedly separate systems have become more and more intertwined, and the Dems’ latest legislation, the Inflation Reduction Act, will erode the distinction even further. And even if this bill should stall, the trend towards intermixing the two will continue.

There are a lot of similarities between the IRA’s book min tax and the global min tax, the one drawn up at the Organization for Economic Cooperation and Development, not just including the rate. At a glance, they almost seem identical, and the rhetoric used around both seems the same–”close loopholes,” “make them pay their fair share.”

But they actually have different histories, and they use bookmaking for different purposes. Understanding the different goals is crucial to understanding why they often come to different determinations on the same issues, despite their common baseline. (For example, the IRA book tax exempts research & development credits but not stock options, while the OECD min tax exempts stock options but not R&D credits. More on that in a bit.)

The IRA’s corporate minimum tax is ultimately a variation on an alternative minimum tax, a concept which has been with the U.S. tax code for decades. Normally they use the tax code, plus or minus some items, to work as a patch on the existing system.

The IRA book tax instead uses book accounting, with the premise that this is somehow “truer,” or at least less susceptible to downward manipulation, than taxable income. It’s a common refrain in public rhetoric on taxes–that companies paid 0% in taxes on $X of profit, implying that 0 is the wrong figure but X isn’t. After all, X is what the company reported to its own shareholders.

In some cases it’s undoubtedly true that the discrepancy is due to manipulation. In other cases, it’s an apples to oranges comparison–like complaining that a driver is going 70 kilometers per hour in a 50 mile-per-hour zone. Book accounting often uses different timing to measure costs, creating apparently wide divergences in profitability which even out over time.

Nevertheless, the political pull to tie taxation to book accounting is undeniable. In fact, it’s been tried before, albeit briefly. Some lawmakers tried to add a book AMT to the 1986 tax reform bill, but met heavy resistance from accountants and businesses. The “compromise” was to keep it in, but add a three-year sunset. Supposedly Congress was then going to consider extending it, but it was left to expire with few mourning its loss. Even for 36 months, it seemed like too much of a mess.

Part of the problem with this method is that while 0% effective tax rates for large corporations are not popular, often the incentives and mechanisms which achieve that rate are. Already, lawmakers have added exceptions for benefits like R&D and green energy credits. Undoubtedly in the future, interest groups will want other tax benefits to be protected as well, and will push either Congress or the Financial Accounting Standards Board, which sets U.S. accounting rules. With each new exemption, the rationale for using the “pure” financial book rules becomes weaker.

The global minimum tax began in a very different way. Its roots trace back to U.S. tax reform drafts drawn up in 2011 and 2014 by Rep. Dave Camp (R-Mich.), then chairman of the House Ways & Means Committee. Both of those included anti-abuse rules meant to target income kept in offshore tax havens–arrangements, usually formed around valuable intellectual property, that allowed companies to record massive profits in jurisdictions with low tax rates and not much of a workforce or infrastructure. Using kind of pornography "I know it when I see it" logic, "Option C" in the Camp draft tried to approximate the conditions where profit-shifting to foreign havens most often occurred, where there is disproportionate income from intangibles. The legislation, which became the basis for the tax on global intangible low-taxed income in the 2017 Tax Cuts and Jobs Act, defined intangible income as unusually high returns on depreciable, tangible property. The more physical, tangible stuff is in a jurisdiction, relative to profit, the less likely that there's funny business going on. Or so the law assumes.

The GILTI tax targets foreign intangible income of U.S. companies at a 10.5% rate, and because of how foreign tax credits work, it only applies when that income is taxed below 13.125% by other governments. Hence, it’s like a minimum tax–but applied to a smaller, narrower base of income. (Camp’s Option C would have targeted individual jurisdictions, but GILTI applies on an aggregate basis, for all foreign income.)

Democrats claimed that GILTI was flawed, by encouraging companies to move facilities offshore and allowing the blending of foreign jurisdictions. But European countries found the idea attractive, and it soon became part of the OECD’s digital tax project–Pillar 2 of a Two-Pillar Solution, kind of a grand bargain between source and residence countries to better capture the untaxed income of giant tech conglomerates. The idea eventually evolved into the 15% global minimum tax which the Biden administration helped reach an agreement on and touted as a major accomplishment. Although the project has since been beset with obstacles.

The book aspect of this came in as almost an afterthought. Because this project involves nations from around the world, they needed to find a way for the common measurement of taxable income. Creating some new universal tax code would have been too cumbersome and difficult, so the OECD opted to use financial accounting principles “as the starting point…because it provides a uniform measure of income that can be applied in all jurisdictions,” according to the OECD’s Pillar 2 commentary. (Of course, there are different global accounting standards and combining them has turned out to be very complex as well. But that’s another tangent.)

So here, the book aspect isn’t an anti-abuse rule itself, and it’s not premised on the idea that it’s more accurate than the tax code. It’s trying its best to reach the same results as national tax codes, if only they could be aggregated somehow. This is why the OECD’s book tax system takes into account timing differences, as well as the expensing of stock options–which most national tax systems allow to be deducted even though they don’t create a cash expense.

But the OECD didn’t create direct exceptions for R&D, green energy, low-income housing, or other tax incentives that governments might deem worthy. With so many different tax codes, that would have been unwieldy. And as the project began to be described in grander terms, to end the “race to the bottom,” it would have been incompatible with the goals. Towards the bottom, you’re likely to find a lot of those incentives.

So this is how you have two 15% corporate minimum taxes, that look so similar from afar but so different up-close.

If, as now expected, the U.S. doesn’t enact any legislation to implement Pillar 2 itself, would the IRA’s book minimum tax be close enough that other nations at the OECD may give the U.S. a pass? The organization already floated the possibility to exempt GILTI into the system–after all, GILTI is what inspired Pillar 2 in the first place. But there are enough important differences–especially that Pillar 2 is calculated per-country, not aggregated–that it’s going to be a tough gap to fill. (What gives some in the U.S. hope is that some of the differences make GILTI harsher to taxpayers than Pillar 2. For instance, GILTI doesn’t allow for loss or excess foreign tax credit carryforwards.)

And the IRA book min tax may not be enough to bridge the difference, so long as it uses this different system of income measurement. It’s ultimately a different tool for a different purpose, and whether it can fit will depend on how much incongruity the OECD members are willing to accept.

Whatever happens, it’s clear that the book/tax divide will continue to crumble, no matter how much practitioners on both sides want to keep it intact. It’s not just in these projects–the Department of the Treasury is increasingly using book income in its regulations, and FASB has been getting more and more pressure to consider tax outcomes in its rules. Tax systems everywhere are getting more international, and financial accounting is one of the few tools available to help them fit together.

But in my time reporting on these issues, I’ve found that astonishingly few practitioners really understand both the book and tax systems. Those few are about to become the most sought-after experts in taxes.

The rest of us are going to be spending a lot of time catching up.


DISCLAIMER: These views are the author's own, and do not reflect those of his current employer or any of its clients. Alex Parker is not an attorney or accountant, and none of this should be construed as tax advice.


This is a special issue of Things of Caesar, following the surprise announcement of a tax deal among Democrats. Next week's post will continue on schedule.


If you found this piece interesting, please consider subscribing to this newsletter for future posts about these topics.