Scott Sumner  

The shocking truth about border tax adjustments

PRINT
Rejoinder on UBI... Henderson on Squawk Box...

When a country does a currency reform, say exchanging 100 old pesos for 1 new one, it doesn't have major macroeconomic effects. Debt contracts and wage contracts get automatically adjusted in the same way, leaving both types of contracts the same in real terms.

The proposed border tax/subsidy plan is often assumed to be neutral, for similar reasons. In theory the dollar should appreciate by 25%, preventing the import tax from distorting trade. And there's lots of evidence from other countries that this would occur---similar provisions are part of most (all?) VAT systems. We know that this can work.

And yet I've always had a nagging feeling that this particular plan might cause problems. Matt Yglesias has a post that helps to explain my concern:

In the case of border taxes, it's true that if you look back at a textbook, it would say these taxes impact exchange rates rather than trade flows -- whether taking the exotic form of the House GOP's corporate tax reform, a European-style value added tax, or even an old-fashioned broad tariff. But like many other conclusions in economics, that's assuming "all else is equal," and the world's governments actually allow exchange rates to float freely.

Cahill's point is that it's very unlikely this would happen. Many emerging markets, he writes, "intervene in some way to limit volatility against the dollar." Over half of American trade is "against EM currencies, and China alone has a weight of more than 20 percent."

Some observers think you can wish this currency intervention away by assuming that foreign countries manipulate their currencies only to hold the price down and maintain competitiveness. But that's simply not true. China has been acting to prop up the value of its currency recently, and it's not the only country to do this.

"Since EMs intervene under both appreciation and depreciation pressure," Cahill writes, "it would seem more than competitiveness considerations are at work." . . .

The theoretical model's prediction that the tax impact will be fully neutralized by exchange rates has some empirical evidence to back it up. But that evidence, more or less necessarily, is drawn from countries that are a lot smaller than the United States and/or that were making more modest tax changes. For the world's largest economy to undertake a tax reform on the scale House Republicans are contemplating entails a 25 percent increase in the value of the US dollar.

That's simply much too big a change for world governments to watch from the sidelines.


A US border adjustment tax would be a massive contractionary shock for the global economy. The only question is what type of shock. Here are three possibilities:

A. The dollar appreciates by 25%.
B. The dollar is unaffected.
C. The dollar appreciates by 12.5%.

In case A, trade is unaffected, but foreigners with dollar denominated debts would immediately see their debts rise by 25% in local currency terms. This could create a global financial crisis.

In case B, dollar debtors would be unaffected, but you would have a massive trade shock. The real (after tax) exchange rate for foreign countries would immediately appreciate by 25% relative to the US dollar. Over time, equilibrium would be restored through a painful process of reducing wages and prices in foreign countries, until the real exchange rate moved back to equilibrium.

In case C, you have a fairly sizable debt shock and a fairly sizable trade shock at the same time. Emerging market exporters would lose exports at exactly the same time that their dollar denominated debts rose by 12.5% in local currency terms. Not good.

I've been trying to wrap my mind around the question of why this border tax would be such a big shock. I think it might have something to do with the fact that money inflows to the US caused by the export of Boeing jets would receive a 20% subsidy, but money inflows to the US in repayment of dollar debts would not receive any subsidy. Perhaps someone who knows more about border taxes/subsidies can clarify this issue.

It looks to me like a border tax would be a really big negative for emerging markets. It's one of those "pick your poison" situations. Do you want a debt crisis, or a jobs crisis?

Screen Shot 2017-02-09 at 12.22.06 PM.png

PS. I believe that Matt is wrong about "old-fashioned broad tariffs", which do in fact distort trade (unless accompanied by export subsidies.)

PPS. Full disclosure. I own overseas stock funds that could be hurt by the border tax. Also, in fairness, the border tax does have some positive aspects, making the corporate tax system simpler and less distortionary.




COMMENTS (19 to date)
BC writes:

"Also, in fairness, the border tax does have some positive aspects, making the corporate tax system simpler and less distortionary."

Scott, do you have an opinion on the claim that a destination-based cash flow tax (DBCFT) and VAT undermine tax competition? I think Daniel Mitchell of Cato likes to make this claim: [https://www.cato.org/blog/concerns-about-theborder-adjustable-tax-plan-house-gop-part-ii].

Also, I know that you favor turning our income tax into a consumption tax by eliminating taxes on capital gains and interest, say by removing the contribution limits on IRAs. But, if eliminating investment taxes turns an income tax into a consumption tax, and a VAT and DBCFT are also consumption taxes, then does that mean that all of these taxes are equivalent (in economic incidence)? Even in this case, I think a big difference would be that an income tax with no investment taxes would be the most transparent, where everyone can see how much tax they pay each year. When I hear that VATs and DBCFTs undermine tax competition because they are hard to avoid, and in addition they are hidden from consumers because they are embedded in prices, then that scares me from a public choice perspective.

HL writes:

A WTO non compliant tax that will likely test all peg / FX targeting systems around the world if domestic political backlash is to be averted at all cost...what could go wrong? No big country targets FX these days, right? ... oh wait.....

Rob writes:

I keep seeing the tax referred to as a border or import tax, but when I read the House GOP tax plan it doesn't seem that straightforward to me.

According to the Ways and Means blueprint - it's a complete repeal of the corporate income tax and replaces it with a 20% tax on revenue minus employee compensation costs, capital investments and revenues from exports. In other words - it's a 20% tax on stuff sold in the US regardless of whether it is imported or made domestically.

It seems to me some sort of border adjustment has to be done if (as this plan does) you move from a global to a territorial tax system.

If the above is accurate - I'm not sure why this plan would impact trade/currencies as much as the discussions about it seem to indicate. Is it that US exporters wouldn't be paying (US) corporate income tax? Or that imports would have less employee compensation costs and capital investments to minimize the "border" tax?

Scott Sumner writes:

BC, I don't favor a VAT as an add on to the current tax system, but rather as part of a replacement of income taxes. I like a system with VATs, payroll taxes (progressive) and carbon taxes. Maybe a land tax.

As long as the VAT replaces the income tax, I don't worry too much about the issues you raise. If it's an add on to income taxes then I do worry.

HL, Yes, that's what has me worried. And it also explains why I'd want to keep the system if it had already been in place for many years, but am very reluctant to suddenly spring it on the world.

Rob, No, the tax applies to imports but not domestic goods. Exports are subsidized.

Rob writes:

Thanks for replying, Scott.

I do not think that is correct. From the GOP 'blueprint':

"Taken together, a 20 percent corporate rate, a switch to a territorial system, and border adjustments will cause the recent wave of inversions to come to a halt. American businesses invert for two reasons: to avail themselves of a jurisdiction with a lower rate, and to access “trapped cash” overseas. Those problems are solved by the lower corporate rate and the territorial system, respectively. In addition, border adjustments mean that it does not matter where a company is incorporated; sales to U.S. customers are taxed and sales to foreign customers are exempt, regardless of whether the taxpayer is foreign or domestic."

I admit that most of the online discussion of this tax suggests that only imports are taxed but if you think about it - there's no way you could repeal the corp tax and come anywhere close to replacing that tax revenue by only taxing imports at 20%. The vast vast majority of coporate revenue would then be completely untaxed. (And that's before allowing immediate expensing of capital investments and deducting employee compensations costs.)

It would be helpful if there was actual legislative language to read - the blueprint is a sales document so it can be hard to parse out the details.

mariorossi writes:

I really don't think it's fair to compare the proposed corporation tax with a VAT.

A VAT is a consumption tax: it will shift demand for consumer to investments goods. While it's possible this will have an effect on the trade balance, there is no obvious reason for this: you can import investment goods as well as consumption goods.

The proposed tax creates an advantage for all domestic production factors: not only labour but land and raw materials as well. I don't think the currency adjustment would really eliminate such difference. If it is applied to commodities, you will need to see a price differential between comodities inside the US and comodities outside the US. Oil in the US will be X% more expensive than outside. Now that might not be that economically significant but it will certainly look wierd. The results would look very similar to a import duty on those commodities.

I also think that the standard text book model assumes that all currency flows are driven by real trade flows: there is little emphasis on the capital and financial system. Given that capital flows are so large nowadays, I find the full currency adjustment theory somewhat unrealistic.

Pithlord writes:

I am a Canadian. Canadians (who pay attention to economic news) are freaked out about this. Should we be?

Scott Sumner writes:

Rob, You are misunderstanding that statement. This is not a VAT, it's simply a tax on corporate profits. If it were a VAT it would raise vastly more revenue than the current corporate tax.

Mariorossi, Most textbooks I've seen do put a lot of emphasis on capital flows.

You said:

"Oil in the US will be X% more expensive than outside."

That's very misleading, as almost all developed countries have significant VATs. Right now, oil in America is much cheaper than elsewhere.

Scott Sumner writes:

Pithlord, Canadians with US dollar denominated debts should be concerned if it passes. (I doubt it will pass.)

I don't think it's a big concern to Canadian exporters.

Rob writes:

I've been doing a poor job of asking my question, but I appreciate your efforts to answer me regardless. I'll try one more time.

Despite being described often as a 20% tax on imports, House GOP tax plan is more complicated than that. Under the plan, the costs of imports would no longer be deductible from taxable income while revenue from exports would no longer included.

The 20% tax would apply to revenues from BOTH imported and domestically produced goods and services BUT more of the costs of domestic production* than imported production would be deductible.

As a result the effective tax rate on domestically produced and sold stuff would be less than on imported stuff.

My original question was basically is that effective rate really lower enough that the impact on imports would be as dramatic as many of the articles about it seem to imply? And with all of the export subsidies the current code has and shifting costs across borders how much difference would a 0% tax rate on exports really make?


*The House GOP plan claims to be greatly simplified with many fewer deductions. Some have even described it as a cash flow tax. If accurate that would imply the effective tax rate on domestic production would not be that much less than on imported production.

Mike W writes:

Continue to pursue this issue...you're my go-to source for attempting to understand it.

Scott Sumner writes:

Rob, You might check this out:

https://www.americanactionforum.org/research/14344/

I'm not sure how best to explain this. Recall that the cost of production of domestic goods (labor and capital) is tax deductible, but the cost of foreign goods is not deductible. That's what makes them so different. If the exchange rate rises by 25%, that offsets the tax advantage of domestically produced goods.

Thanks Mike.

HL writes:

There are fine distinctions, but I would basically see this as a cleverly disguised VAT. VAT adoption is a horrendous political proposition for the GOP that won due to "angry voters" last year. Imagine that you adopt a 20% sales tax and eliminate corporate income tax altogether. Can the GOP stomach the political fall-out? I mean a meager 3% hike in Japan caused all kinds of political mayhem for Abe! No way they can sell that kind of apparent tax burden shift (it really isn't, but the optics of it) to the electorate right now.

Once you transform the current corporate income tax structure into an odd variant of Japan's subtraction method VAT (entity-based VAT that relies on companies' books instead of transaction-based, credit invoice VAT that allows more straight forward audit trail), you basically have Ryan / Brady plan. Of course you also allow wage deduction, become neutral in the treatment of financing method (no interest deduction), and make all investments expenses. That's because this is a de facto VAT: you only worry about spending. Subtraction-based VAT is also administrative easier perhaps since it will rely on the same tax reporting structure as the current tax system.

In fact, the legality of Japan's subtraction method VAT is still an interesting topic. It has the same WTO compliance issue like the Ryan / Brady plan, but for some reason I can't find via web, there has been no challenge by other countries. Curiously, though, the low VAT rate might have been a mitigating factor (because it is only 8% right now). This is just my personal speculation.

mbka writes:

I agree with HL. This looks very much like a VAT to me. Taxation of revenue, not profits, is essentially taxation of final sales. Tax deduction of inputs to production is a hallmark of a VAT. Only difference here is "only if they're domestic". But if this is really done this way, then it's effectively a tariff and how can this not be illegal under current trade agreements?

HL writes:

It is basically illegal in the WTO framework because of wage deductions. But then you have the odd Japanese case that also violates the spirit of the agreement.

This is a fascinating debate. But as Scott has been highlighting, far, far more opaque and unclear (and probably hastily put together) than many people assume.

Miguel Madeira writes:

"In case A, trade is unaffected, but foreigners with dollar denominated debts would immediately see their debts rise by 25% in local currency terms."

But there is not also the opposite effect? Foreigners who own dollar denominated assets (including dollar banknotes under the bed) will see their assets rise by 25% in local currency terms. Ans americans who own assets denominated in foreign currency (for example, Portuguese public debt) will see their assets devalue in 20%.

The bigger problem that I see in the theory "dollar will raise 25% and everthing will remain the same" is:

1 - Exports will be 20% less in nominal value, but it will remain the same

2 - The same for imports

3 - Probably the same for foreign inflows of capital in the US (if a portuguese wants to invest 10.000 euro in US titles, will invest 20% dollars)

4 - But the outflows of capital from the US to the rest of the world? I don't see a clear mechanism making americans who want to buy foreign titles to reduce their expense in dollars in 20%

Rob writes:

Thanks, Scott. The Holtz-Eakin link was very helpful. A 25% adjustment intuitively seems too high to me but I shouldn't trust my intuition on this kind of math. Looking at the tables in that link there would be a wide spread in winners and losers were this to pass, which probably dooms it.

HL - not hastily put together. JEC economists have been working on this for a long time and it was unveiled last summer by Chairman Brady and Speaker Ryan.

Scott Sumner writes:

HL and mbka, This is definitely not a VAT (it would collect far less revenue for instance.) The key different is that under this plan wages are deductible, not so under a VAT.

And it is not a tariff, it does not distort trade.

HL writes:

A crucial error on my side (after insisting on the WTO compliance issue). Thank you for that clarification!

Comments for this entry have been closed
Return to top