Stop saying a value-added tax is an export subsidy

It's not. No matter how many times people (even economists) say they are.

In a recent Financial Times opinion piece, Jason Cummins (Columbia Econ PhD 😩) argued that the Trump administration should impose a 25% tariff on European goods to offset an “unfair advantage” from Europe’s Value-Added Tax (VAT) system. According to Cummins, European exporters like BMW enjoy an “implicit subsidy” when they receive VAT rebates on exports, while American companies like GM are disadvantaged when selling to Europe.

This argument sounds logical on its surface. After all, BMW gets a VAT refund when exporting to America, while GM faces the full VAT when selling to Europe. Doesn’t this create an uneven playing field?

No. No way. Not at all.

This claim fundamentally misunderstands how VAT works and ignores the basic principles of taxation. The Tax Foundation has a great takedown of this idea from last month when White House deputy chief of staff Stephen Miller (not known for his expertise in international trade) tried to make this argument.

Let me see if I can drive the point home further with a little Economic Forces pizzazz.

The Mechanics of a VAT

I know from our fancy Substack analytics that most of our readers are American, who may not know the basics of a VAT. If you do know what a VAT is, then skip to the next section on the actual economics.

A VAT is a type of consumption tax, charged as a percentage of the price at each stage of production on the “value added” to a good or service. It’s called destination-based because it taxes goods where they are consumed, not where they are produced. This principle is the norm virtually everywhere in the world. In fact, almost every country (175 out of 193 U.N. members as of 2023) uses a VAT or similar tax system, and the U.S. is the only OECD nation without one.

Under a typical VAT system, exports are zero-rated—meaning the exporter charges no VAT on the sale to a foreign customer, and can refund (or credit) any VAT already paid on inputs used to produce the export. Conversely, imports are taxed as if they were produced domestically—when an imported good is sold to a local consumer, the same VAT applies as for a locally made good​.

This is what we call a border adjustment: tax is levied at the border on incoming goods, and rebated at the border for outgoing goods. The logic is straightforward: we want to tax domestic consumption of goods and services, regardless of where they came from, and not tax consumption that occurs abroad.

The Economics of a VAT

The key to understanding VAT border adjustments is tax incidence. Lesson number one in Econ 101 on taxation is to not look at how directly paying any tax is. That’s irrelevant. When economists analyze tax systems, they look at how the tax changes economic behavior by altering price signals.

We need to focus on two key economic principles:

1. Relative prices determine resource allocation. When businesses decide what to produce, and consumers decide what to buy, they respond to the relative total prices of different goods and services, not the split-out tax. (If only there were a theory of prices to help us understand prices…) If all prices rise by the same percentage, the real economic decisions people make don’t change.

2. Destination-based taxes maintain relative price neutrality. In principle (we will deal with complications later), a VAT increases all consumer prices by the same percentage without changing their relative values. Producer prices remain unaffected, keeping production incentives constant.

Imagine a car that costs $30,000 to produce before tax. Now compare four scenarios:

1. BMW sells the car in Germany (domestic sale): Germany’s VAT (let’s say 20% for simplicity) is added on the final sale. The German consumer pays 20% VAT, i.e., an extra $6,000, for a total price of $36,000. BMW forwards that $6,000 to the German government as VAT.

2. BMW exports the car to the U.S.: Since the car is exported, BMW does not charge German VAT. Any VAT BMW paid on parts or inputs is refunded by the German tax authority. The U.S. buyer pays the $30,000 price, and since the U.S. has no federal VAT, there’s no equivalent federal tax on that sale. (A state sales tax might apply at the point of sale, but we’ll come back to that.) The key point: the German government collects no VAT on an item consumed in the U.S.. This makes complete sense because that car’s being enjoyed by an American buyer, not a German resident.

3. GM sells the car in the U.S. (domestic sale): The U.S. has no VAT, so the American consumer pays $30,000 (ignoring any state sales tax). No federal consumption tax is collected. (In states with a sales tax, the consumer might pay, say, 7% extra to the state government, but again, the federal treatment is no tax.)

4. GM exports the car to Germany: When the car arrives in Germany, it faces the same 20% VAT as any car sold in Germany. So a German customer buying the American-made car pays $30,000 + $6,000 VAT = $36,000. That $6,000 goes to the German government. From GM’s perspective, it doesn’t owe U.S. tax on that export sale (since the U.S. doesn’t tax exports of goods), but its product will bear German VAT when consumed in Germany.

What outcome do we have here? In Germany, both the BMW and the GM car cost the same $36,000 after tax, and the German government collects VAT on both. In the U.S., both cars cost $30,000 before any state sales taxes, and the U.S. government collects no federal consumption tax on either. Each country taxes consumption within its borders—no matter where the product came from—and does not tax consumption outside its borders. This is precisely the goal of destination-based taxation: neutrality. Consumers in each country face the same tax on a given product, whether it’s domestically produced or imported​. And neither country’s producers carry their home consumption tax as a “ball and chain” when they go compete in foreign markets.

What If We Removed Export Rebates?

Let’s tackle head-on the notion that VAT rebates confer an “unfair advantage.” Imagine for a moment the opposite policy: what if countries did NOT rebate VAT on exports? Suppose Germany kept that 20% VAT on BMWs even when they ship to the U.S. In that case, when BMW sells a car to an American customer, the price would include German VAT—meaning an American consumer is effectively paying German taxes, which go into German coffers, despite the car being consumed in the USA. Worse, if the U.S. also tried to levy its own sales tax on that car (which it does), the poor American consumer would be paying tax twice: once to Germany and once to the U.S. Clearly, that would be a recipe for double taxation and a huge impediment to trade.

Conversely, what if countries didn’t tax imports? In that case, an imported car in Germany would avoid the 20% VAT that German-made cars have to charge, making a tax-free import far cheaper than a taxed domestic product—unfair to German producers and also likely to erode Germany’s tax base.

So, to avoid these problems, virtually all countries follow the destination principle: don’t tax exports, do tax imports. This ensures that each country’s tax applies only to its own consumers, creating a level playing field within each market. Far from being a mercantilist trick, this practice is about tax neutrality—making sure taxes don’t distort trade by favoring domestic over foreign goods or vice versa. In fact, economists usually consider the destination-based VAT less distortive for trade than an origin-based tax system. The OECD notes that the destination principle “places all firms competing in a given jurisdiction on an even footing” and achieves neutrality in international trade​. It’s also explicitly endorsed by WTO rules, which say that not taxing exports (and refunding any prior tax paid) “shall not be deemed to be a subsidy.”

As Martin Feldstein and Paul Krugman (economists from opposite ends of the political spectrum) showed in a simple paper decades ago, the argument that rebating VAT on exports provides a competitive advantage is simply wrong. “A VAT is not, contrary to popular belief, anything like a tariff-cum-export subsidy” but simply a tax on domestic consumption much like a sales tax. VAT with border adjustments has no lasting effect on trade competitiveness or balances. The FT piece simply ignores this widely accepted economic understanding. If anything, failing to rebate a consumption tax on exports would be the real distortion, akin to an export penalty.

VAT vs. Income Tax: The Real U.S./E.U. Difference

While VAT border adjustments don’t create trade advantages, there is one way that VATs can legitimately affect international trade patterns: when they replace or reduce income taxes in a country’s overall tax mix. This effect has nothing to do with border adjustments or export rebates, but instead stems from how differently these tax systems treat saving and consumption.

We’ve been talking a lot about sales taxes but income taxes are the big thing in the US. Income taxes hit both consumption and savings. When you earn money, you pay income tax whether you spend that money immediately or save it for the future. Then, if you save it and earn returns on those savings, you pay tax again on those returns. This creates a bias against saving and in favor of current consumption.

VATs, in contrast, only tax consumption when it actually occurs. If you earn money and save it rather than spending it, you don’t pay VAT until you eventually spend those savings. This eliminates the double taxation of savings that occurs under an income tax. The result is that VATs are more neutral between present and future consumption than income taxes.

When a country shifts from income taxation toward VAT, it effectively reduces the tax burden on saving relative to consumption. This tax shift tends to increase the saving rate as consumers respond to the changed incentives. With more savings and less current consumption, there’s a temporary improvement in the trade balance—exports remain steady while imports fall along with domestic consumption.

However, this isn’t permanent. Over time, accumulated savings finance higher future consumption, including imports. What looks like a trade advantage in the short run balances over the longer term. This effect would happen even with an origin-based consumption tax without border adjustments. The key factor is shifting away from taxing saving toward taxing consumption.

Ironically, this genuine economic effect of VATs on trade is rarely mentioned in political debates about border adjustments and “unfair subsidies.” The focus instead remains on export rebates, which, as we’ve seen, are economically neutral. It’s a classic case of focusing on the visible (export rebates) while missing the invisible (intertemporal consumption changes)—exactly the kind of error good economic analysis is designed to prevent.

The Messy Reality of Tax Systems

Of course, the textbook analysis of VATs we’ve explored describes idealized systems that don’t fully match reality. No actual tax system—VAT, income tax, or otherwise—perfectly matches the theoretical models economists use.

Real-world VATs deviate from the ideal in numerous ways:

• They rarely cover all consumption equally (food and necessities often face reduced rates)

• Services, especially hard-to-tax ones like financial services, often receive exemptions

• Informal economic activity escapes taxation entirely

• Administrative complexities create uneven compliance

• Small businesses may be exempt from VAT regimes

Similarly, our U.S. system has its own deviations from theoretical ideals:

• State sales taxes often tax business inputs multiple times, creating “tax pyramiding”

• Different states have wildly different tax bases and rates

• Services are inconsistently taxed across jurisdictions

• Our corporate tax system creates its own set of incentives and distortions

If we insisted on creating retaliatory tariffs to offset every deviation from theoretical tax neutrality, we’d quickly find ourselves in an unworkable situation. Virtually every country’s tax system could be accused of creating some kind of “unfair” advantage or disadvantage in some sector.

The WTO has recognized this reality by establishing rules that distinguish between acceptable border adjustments (like VAT rebates) and unacceptable export subsidies. This framework acknowledges that consumption taxes can and should be border-adjusted without being considered trade distortions while preventing countries from disguising actual subsidies as tax adjustments.

If we started imposing tariffs in response to other countries’ VAT systems, we would be opening a Pandora’s box. Other countries could just as easily identify aspects of our tax system that provide implicit advantages to certain sectors and respond with their own retaliatory measures.

Tl;dr: VATs Don’t Subsidize Exports

I know I’ve been holding my cards close to my chest. So let me say it once and for all:

VAT border adjustments (including export rebates) don’t create trade advantages.

Adjustments simply ensure that consumption is taxed once and only once—where it occurs. Without these adjustments, VATs would distort trade by effectively taxing some consumption twice and other consumption not at all. The economics is clear, and it (once again) comes from thinking about relative prices. Everything is about prices.

So the next time someone claims that European VATs disadvantage American exporters from a political supervisor or an actual economist, remember—it’s their misunderstanding of how consumption taxes work that’s causing the confusion, not unfair practices by our trading partners. If anything, America’s reliance on our patchwork state sales tax system (which often taxes business inputs) creates more embedded taxes in our exports than other countries’ VAT systems do in theirs.