What's the difference between a border adjustment and a tariff? And who pays?

Articles like this talk about options for how the administration would pay for the wall.


What’s not clear is who is paying these various taxes. If you tax imports, you tax imports. Why does it matter whether you call it a border adjustment or a tariff?

And in either case, who is paying? The foreign exporter, or the American importer?

The second article quote mainly answers the question ‘what’s the difference?’.

‘Who pays?’ is a much more difficult question. But keep in mind that’s just as true with existing taxes. Right now US domiciled corporations pay a tax on their net income, the difference between revenues and expenses (the income is the revenue, the deductions are all the expenses to make that revenue). It doesn’t matter where in the world the economic activities occur, just that the company’s official domicile is the US.

Who ‘really’ pays that tax? It’s collected from the corporations, but it’s unclear and debatable what proportion of it is reflected in higher prices for goods, lower wages for employees of the company, or lower return on capital for the owners (shareholders). Simplistic political discussion implicitly assumes the last of those (or even more unrealistically speaks of a ‘corporation’ as some imaginary thing unto itself separate from its owners and employees), but it’s not necessarily true.

The border adjusted cash flow tax works in contrast basically as follows: it’s nominally on the same thing, net income of the company, except the company is not allowed to deduct the cost of imports as an expense, only the cost of domestic inputs. And revenue from exports is excluded from income. Likewise the revenue and expense from wholly foreign transactions (inputs the company buys overseas to make products also sold overseas) is excluded.

The main arguments in favor of the border adjusted cashflow tax are removing the incentive to change company domiciles to avoid tax (no longer matter, ‘tax inversion’ would have no purpose), and the incentive to site operations in a particular place just for corporate tax reasons. And it would being generally simpler and more uniform at least initially: lower rate, broader base than the current US corporate income tax.

In theory the border adjusted tax would not be a trade policy, since the value of the currency of a country adopting it would increase to offset the different treatment of imports and exports. However politically it also has the advantage of appearing to fulfill a promise to ‘reduce trade deficits’, though that’s not a likely outcome actually.

The potentially serious disadvantage is secondary effects from the currency value increase in case of the US$ as world reserve currency. Stuff such as the effect on foreign entities with US$ debt but local currency revenue.

That’s only half of the equation. Who gets the money?

That question is referred to as tax incidence in economics, and it’S a complex matter. Long story short: The person who, economically, foots the bill of a tax (or tariff or border adjustment or whatever you call it - the term doesn’t matter) at the end of teh day can be very different from the person from whom the tax is, legally, collected.

Surely, in every case, it’s the end user who pays. A business has to make a profit, so must sell the widgets for more than the total cost of sourcing them. If part of that cost is transport, tax, duty, bribes, or the CEO’s new car, it is just another cost that has to be taken into account.

Take a look at the concept of tax incidence in the post above you. While there are edge cases in markets for certain goods where it’s only one party that pays, it’s usually spread between everyone involved.

It’s sure some person pays. Corporations are a legal fiction and saying ‘corporations pay’ is always nonsensical, I agree that far. However it’s not sure which ‘end user’ pays between customers for the company’s products, the owners of the company, and the employees. In an idealized case of a closed economy you can argue the incidence of a conventional corporate income tax is wholly on the owners, which is the assumption underlying its existence: that it mainly taxes owners, who tend to be rich, and is thus progressive. In the real world the incidence is more doubtful.

Another way to look at the border adjusted cash flow tax however, which is somewhat different than a corporate income tax albeit now presented as a reform of the corporate income tax, is to a VAT. A for example 15% VAT* combined with the elimination of a 15% payroll tax would be economically identical to a 15% border adjusted cash flow tax on companies retaining a 15% payroll tax. Or IOW a border adjusted cash flow tax is a VAT with an additional deduction allowed for domestic payroll expenses.

*border adjusted, charged on imports not on exports, as is virtually always the case.

To put a little more flesh to the bones of my last post, the bottom line of the concept of tax incidence is: The rate at which the incidence of the tariff is split between buyer and seller depends on the relative price elasticities of supply and demand. This can be translated into how easily, relatively speaking, buyers and sellers can switch to alternatives when the tariff is introduced. Which is intuitively plausible: The mor easily you can find a substitute for the product you’re buying or selling, the more you can pass off the burden of the newly introduced tariff to the other side who can’t switch as easily.

Another question is whether or not there are different treaty consequences for the two (for example NAFTA and World Trade Organization agreements).

Who pays is simple. Who wants the goods across the border? The Customs won’t allow anything across without proof of payment. So either the buyer or the seller pays the US government. They work it out between themselves. A trucking company that tries to bring stuff across the border without proof of payment will get fined. (Similarly, about a decade ago that was a problem on the US-Canada border - shipments without customs paperwork faxed ahead by at least 2 hours would face hefty fines due to Patriot Act paranoia.)

So the obvious question is, who owns the goods while they are sitting on the truck in transit? Most likely that owner has to pay the duty.

And it’s a no-brainer that the costs get passed on, so the cost is ultimately paid by the consumer. It encourages a company to make the same product entirely in the USA; labour may cost more in the USA, but the competition from Mexico becomes more expensive.

Could somebody give an example with numbers?

Here’s my attempt:

Assume Walmart imports TV sets from Mexico. The cost paid to the manufacturer is $100. The cost to Walmart (transport, sales) is $10. Walmart sells the set for $120, getting a profit before tax of $10, or $6.50 after tax (35%).

Now if the border adjustment tax is like (20%)VAT, then Walmart will have to add $20 to the retail price. The set will cost $140, with Walmart paying that extra tax to the government.

So far so good.

According to this, “the company cannot deduct the cost of the imported [goods] as a business expense.”


If I take that literally, then Walmart will have to sell the set for $200! $100 goes to the maker. $20 goes to the government. $10 goes to costs. But their profit before tax would be calculated as $170! Since the $100 is not a business expense? The corporate tax would be 35% of $170 = ~$60 with only $10 after tax profit remaining.

What am I misunderstanding?

Also how does this subsidize exports? Suppose Boeing exports an airplane to Mexico for $100 million. Do they get $20 million back from the Gov’t?

Then the value of the US dollar goes up, so everything imported is cheaper again, so the cost to the consumer is zero.

Except that the American value of the Mexican factories has gone down, so the rich American owners have lost money, and the owners of American capital have gained money.

Except now there is a whole border compensation tax going on, so the cost and trouble of collecting and paying the tax gets subtracted from everything and everybody is poorer.


It’s really hard to say, other than that the idea “it’s a no-brainer” isn’t likely to be true.

I would also imagine that smugglers will put their skills to work in what will become a very lucrative, but low risk compared to narcotics, enterprise.

I read recently where a business was smuggling garlic from Norway to Sweden. Norway has no import tax on garlic (no home production to protect) but the EU (Sweden) does at 20% (Spanish garlic is protected). Norway and Sweden generally have a free trade culture apart from specific products (like garlic) These guys were shipping it in by the container load. Of course, this means that we in the UK pay 20% more for garlic than we would if we imported it from China.

Lets say Señor Smuggler buys a load of mangos in Mexico and ships them to Puerto Rico. Once there they can magically become Puerto Rican mangos and can be shipped to the USA tax free.

Who is the tariff paid to?

It’s not really. It’s an area where common knowledge is wrong because most people don’t understand tax incidence. If you are looking like a broad tariff across many markets with differing elasticities of supply and demand the answer is likely different than that. Generally some of the tariff is likely to be borne by the consumer in higher prices and some will be borne by the seller in the form of reduced profits. In specific markets where very specific conditions will have all the tax borne by the consumer. There are also conditions where the seller is unable to pass any of those extra costs onto the consumer.

  1. The first comparison is correct. Under current tax law assuming those revenues and costs, the tax is $3.50, 35% of $10 net profit. Under the border adjusted cash flow tax, and incorrectly assuming for purposes of explanation that the cost of the imported set itself remains the same in USD, the taxable amount becomes $110 rather than $10, because the $100 cost of the import is no longer deductible, and the tax $22 if the rate is 20%. Thus Walmart would have to raise the sale price from $120 to $143.125 to make $6.5 profit, ie raise the cash flow to 143.125-110=33.125, pay a tax of (143.125-10)*.2=26.625, and end up with $6.5 after tax as before.

However, there is no reason to believe the USD cost of the TV would remain at $100 rather than sinking, as the USD strengthened. Also the example is extreme in that domestic rent or capital cost of store property ownership, staff, overhead etc. is typically a much larger % of sales price for retailers of imported goods than in the example.

  1. For simplicity if Boeing’s inputs are all domestic (many are not) and it costs $100mil to produce a plane it sells for $120mill, now it gets $20mil*.65=$13mil after tax. Under the border adjusted cash flow tax there is no tax on this transaction and the plane price can be lowered to $113mil for the same after tax profit.

But again it’s likely to have to lower the USD cost of the plane because the USD will strengthen because the tax otherwise reduces the supply of USD’s from exporters to the US who want to get rid of them, and increases demand for USD’s from buyers of now cheaper (static basis) US exports.

This paper has loads of examples

Melbourne’s post hints at what about such a tax is more likely to upset the apple cart and it’s the effect on capital from the repricing of the USD, not so much impact on consumers. In particular it could make insolvent some foreign entities which have borrowed in USD but have revenue in their own currency, ie specifically away from goods trade with the US (where the revenue would be in USD, albeit perhaps affected by the tax).
I don’t agree however with the implication this tax is more complicated than the current US corporate income tax: it’s much less so given all the special features of the current code which would be cleared out, at least until special interests gradually inserted new ones. Clearing out that underbrush periodically is an argument in favor of changing the system periodically, even if the new system is only equally good in theory.

eta: Assuming Boeing exports that plane. For domestic sales it just becomes 20% tax on the profit rather than 35%, assuming the company was really paying 35% effective, which many companies don’t due to the complexity of the existing code. The new tax would be simpler, again at least until it gradually got mucked up over time, which is pretty much a feature of the political system.

Thanks, Corry El! Have to study your numbers and that link!