The most recent story on the Burger King press page remains "Burger King Restaurants Bring Back Chicken Fries," but events have overtaken the chicken fries. After many media reports, Burger King and Tim Hortons have put out a press release confirming that they've been holding merger talks, and imagine what the catering is like. This announcement has set off a frenzy of condemnation, since any merger would be what is called an "inversion," turning the combined Burger Tim into a Canadian company and thus freeing it from its harsh but patriotic duty to pay U.S. income taxes.
This condemnation is a bit confused, so I thought it might be helpful to explain very simply (too simply!) how the U.S. corporate income tax system works. There is nothing new here, lots of people know this, but weirdly lots of people don't, too. It works like this:
- U.S. companies pay the U.S. Internal Revenue Service a tax rate of 35 percent on income they earn in the U.S.
- U.S. companies pay the IRS a tax rate of 35 percent on income they earn abroad, but they can credit the amount of foreign tax they pay against that liability. So for instance if you earn $1,000 in Canada, you owe $350 of taxes in the U.S., but you can reduce that by the $150 that you paid to Canada (which has a 15 percent corporate tax rate), so you only end up paying the IRS $200. But still your effective rate is 35 percent -- 15 percent to Canada, 20 percent to the U.S. -- though you can defer those U.S. taxes for a long time if you just invest the money offshore.1
And because basically no countries have a higher corporate tax rate than the U.S. -- and plenty have lower or zero rates2
-- that means that U.S. companies owe the IRS taxes on pretty much all the income they earn anywhere in the world.
- Foreign companies pay a tax rate of 35 percent on the income they earn in the U.S. And, for the most part, that's it: Their home tax authorities rarely charge them tax on income that they earn in the U.S.3
- Foreign companies mostly just pay whatever the tax rate is in the country where they earn their income, and don't bother writing extra checks to the U.S. tax authorities.
So the purpose of an inversion has never been, and never could be, and never will be, "ooh, Canada has a 15 percent tax rate, and the U.S. has a 35 percent tax rate, so we can save 20 points of taxes on all our income by moving." Instead the main purpose is always: "If we're incorporated in the U.S., we'll pay 35 percent taxes on our income in the U.S. and Canada and Mexico and Ireland and Bermuda and the Cayman Islands, but if we're incorporated in Canada, we'll pay 35 percent on our income in the U.S. but 15 percent in Canada and 30 percent in Mexico and 12.5 percent in Ireland and zero percent in Bermuda and zero percent in the Cayman Islands."
Once you understand that, it's obvious that the appeal of this strategy will vary in direct proportion to how much business you do in the U.S. and how much you do in, say, Bermuda. Now, you might naively say: I doubt a lot of multinational companies do a lot of business in Bermuda! But I just told you that was naive. The key tricks are really to:
- put a lot of your income in Bermuda, and
- then never pay U.S. taxes on it.
Inversion solves the second problem but not the first. The first can be solved, though. For instance, if you are a drug company, it costs you like a dollar to make a pill, and you sell it in the U.S. for $10,000. You might say, "well OK then I have $9,999 of net income in the U.S.," but again you are being naive. The right answer is:
- Your U.S. subsidiary makes a pill for $1.
- Your U.S. subsidiary licenses the patent on that pill from your Bermuda subsidiary for $9,995.
- Your U.S. subsidiary sells the pill for $10,000.
- Your U.S. subsidiary has $4 of net income, which is taxable.
- Your Bermuda subsidiary has $9,995 of net income, which is not.
It's more complicated than that, but that's the general idea. If the parent company is a U.S. company, then eventually that Bermuda sub's net income will be taxable in the U.S. anyway. But if the parent company is Canadian or Dutch or Swiss or whatever, then the Bermuda sub's income will never be taxed.
Anyway, Burger King. Burger King does not sell a lot of burgers in Bermuda. (It does apparently sell some in the Cayman Islands.) Out of 13,667 Burger King restaurants as of the end of 2013, 7,436 (54 percent) were in the U.S. and Canada (though they're not broken out more specifically than that); those U.S. and Canada stores provided 58 percent of the company's income.
So if it moved to Canada, the roughly half of Burger King's tax bill attributable to its U.S. restaurants wouldn't change. Only the half of the bill attributable to its restaurants abroad would -- and that only when it repatriates the money.4 Its restaurants in Canada would go from paying 15 percent to Canada now, and 20 percent to the IRS eventually, to paying 15 percent to Canada and zero percent to the IRS. Its restaurants in the Cayman Islands would stop paying taxes to anyone, because that is how it goes in the Cayman Islands.
And could Burger King move all of its income to Bermuda to avoid paying taxes on its actual income in Canada and Mexico and the U.S. and wherever? Ehh, not really. There's a reason that companies that just wake up one day and decide to invert tend to be pharmaceutical companies. The tax arbitrage really works for high-margin, IP-driven businesses. It doesn't work for low-margin, labor-and-raw-materials-driven businesses. You can't really assemble a burger in Bermuda and then sell it in Canada. A lot of the price of a $1 burger goes to paying for, you know, the burger, and the assembling of the burger. Relatively little of it goes to pay for the idea of the burger. That said, 80 percent of Burger King's revenue does come from franchises -- who basically pay Burger King for the idea of the burger -- so there are perhaps some savings available.5
So on the evil inversion scale, this ranks pretty low. It's not the classic pharmaceutical-company inversion designed to avoid paying taxes anywhere. Burger King will still pay 35 percent taxes on the half of its business that's located in the United States.
Consider a few other things. First, Burger King's 7,436 stores in the U.S. (and Canada) are fewer than it had in 2010 (7,550). Our great national scourge of millennials is afflicting the fast food restaurants: McDonald's is losing its cachet among younger consumers, and I doubt Burger King is much cooler. (Is it? Snapchat me your answer!) Over the same period, Burger King has added 700+ restaurants in Europe, the Middle East and Africa, 400+ in Latin America, and 400ish in Asia and the Pacific. It's owned by a Brazilian private equity firm, 3G Capital. Its owners and its future are outside of the U.S. A Brazilian company opening a burger joint in China wouldn't have to pay U.S. taxes on that store's income. And that's not not what Burger King is.
Second: This merger, if it happens, is a real merger with real business and capital markets purposes. The merger is not mainly about taxes; in fact, Tim Hortons and Burger King's effective tax rates are basically the same.6 Tim Hortons, I am given to understand, sells a lot of coffee and donuts, most of them in Canada. (Out of 4,485 stores at the end of 2013, 3,588 were in Canada.) I don't know, you could probably sell the coffee in the burger stores, or the burgers in the coffee stores, or good lord you could put a burger on a donut, that will probably win you cool points with millennials; millennials love things that are part donut and part thing that is not a donut. So there are business reasons for the deal. But if Burger King acquired Tim Hortons, the tax rate on all those Tim Hortons stores would go up: Instead of the regular 15 percent Canadian rate that they're currently paying, they'd have to pay 35 percent combined to U.S. and Canadian authorities.7 From a Canadian company's perspective, that hardly seems fair. Thus the inversion.
One more thing: This inversion is not all that inverted. Tim Hortons is actually bigger than Burger King, on revenue and net income though not on stock market capitalization. This is not just an aesthetic point. In the uproar about inversions, several bills have been proposed in Congress to stop them. There is the cleverly named Stop Corporate Inversions Act of 2014, for instance, and the similarly imaginative No Federal Contracts for Corporate Deserters Act of 2014. Each tries to stop U.S. companies from relocating abroad by means of a merger with a foreign company in which the U.S. company's shareholders end up with more than 50 percent of the shares. (Versus the current legal standard of 80 percent: As long as Tim Hortons shareholders end up with at least 20 percent of the combined company's shares, it's a valid inversion under current law.)
But each bill has an exception for mergers with a foreign parent that is, like, a real foreign parent. The Stop Corporate Inversions Act provides (emphasis added):
A foreign corporation described in paragraph (2) shall not be treated as an inverted domestic corporation if after the acquisition the expanded affiliated group which includes the entity has substantial business activities in the foreign country in which or under the law of which the entity is created or organized when compared to the total business activities of such expanded affiliated group. For purposes of subsection (a)(2)(B)(iii) and the preceding sentence, the term ‘substantial business activities’ shall have the meaning given such term under regulations in effect on May 8, 2014, except that the Secretary may issue regulations increasing the threshold percent in any of the tests under such regulations for determining if business activities constitute substantial business activities for purposes of this paragraph.
The No Federal Contracts one has similar language.8 Those Treasury regulations "provide that an expanded affiliated group will have substantial business activities in the relevant foreign country only if at least 25 percent of the group employees, group assets, and group income are located or derived in the relevant foreign country."
Burger King and Tim Hortons have about the same number of employees: 2,420 for Burger King and 2,558 for Tim Hortons (obviously not counting franchise workers). Assuming that at least half of Tim Hortons' employees work in Canada, and none of Burger King's do -- both assumptions are probably too low -- the employee test will be satisfied.
Tim Hortons has about CAD 1.7 billion ($1.5 billion) of property and equipment, versus $802 million for Burger King. Again, assume (falsely) that none of Burger King's assets are in Canada. Then assume that 80 percent of Tim Hortons' assets are (80 percent of its stores, and 82 percent of its revenues, are in Canada). Then Burger Tim will satisfy the asset test, with 46 percent of combined assets in Canada.
Tim Hortons gets CAD 2.7 billion ($2.4 billion) of revenue in Canada, out of a total of CAD 3.3 billion ($3.0 billion). Burger King's total revenue is about $1.1 billion. Even if Burger King makes zero dollars in Canada, the combined Burger Tim will get about 59 percent of its revenues in Canada, more than enough to meet the gross income test.
So if you want to use this deal as an example of why the inversion rules should be changed, be careful. For one thing, this seems like a pretty reasonable inversion, as inversions go. For another, even if the rules were changed to ban inversions, this inversion would still be allowed.
Because those taxes are just on income earned by U.S. corporations. So if you put your foreign operations in a foreign subsidiary, you can keep their earnings offshore and untaxed in the U.S. But if you ever want to bring those earnings back -- to invest in the U.S., or just to pay out to shareholders -- then you need to pay U.S. tax on them.
2 Here's a table from KPMG. You have to read the footnotes on the right: There are some weird numbers like Belgium and India being 33.99 percent, and it perplexingly lists the U.S. as 40 percent to account for state taxes. But nobody's above 35 percent except the U.S., Japan (35.64), the U.A.E. (55 percent! But only on oil companies?).
3 Here is a table from the Manhattan Institute listing territorial systems (Canada, France, Germany, Japan, Switzerland, the U.K, etc.) versus the many fewer worldwide-taxation systems (the U.S., Chile, Greece, Ireland, Israel, South Korea, Mexico). Tangentially related, and I cannot vouch for it, but with regards to individual (not corporate) tax, here's a fun table from WIkipedia showing that only the U.S. and Eritrea tax the foreign income of nonresident citizens.
4 To Canada, I guess.
5 I wouldn't count on it though. If the intellectual property is not already in a non-taxable subsidiary it would have to be sold there for fair value.
6 Because of deferral, of course: Burger King's income earned abroad is not currently taxed in the U.S.; it's taxed when it's repatriated. If it's invested in growing the foreign businesses then you it can be deferred for a long time. Burger King does not have a lot of trapped cash abroad, presumably because it spends the foreign cash on the growing foreign business.
7 Again though deferral would make this not so much an instant increase in the tax bill as just a trapping of cash abroad.
8 Subsection (b)(2):
(A) In general. -- A foreign incorporated entity described in paragraph (1) shall not be treated as an inverted domestic corporation if after the acquisition the expanded affiliated group which includes the entity has substantial business activities in the foreign country in which or under the law of which the entity is created or organized when compared to the total business activities of such expanded affiliated group.
(B) Substantial business activities. -- The Secretary of the Treasury (or the Secretary's delegate) shall establish regulations for determining whether an affiliated group has substantial business activities for purposes of subparagraph (A), except that such regulations may not treat any group as having substantial business activities if such group would not be considered to have substantial business activities under the regulations prescribed under section 7874 of the Internal Revenue Code of 1986, as in effect on May 8, 2014.
To contact the writer of this article: Matt Levine at firstname.lastname@example.org.
To contact the editor responsible for this article: Tobin Harshaw at email@example.com.