How to Think About the Tariffs
This is bad policy, executed thoughtlessly. But it is worth thinking through exactly *why* it is bad.
The American people’s expected tax burden has gone up by ~2% of national income over the past two months.1
Like all sharp budget tightenings, this will reduce living standards—and these tax increases are targeted such that the poor will suffer disproporationately. The tax increases are also sufficiently complex and arbitrary that they will create substantial costs for businesses, reducing productivity and exacerbating the direct hit to incomes.
Perhaps most importantly, the actual announcement and implementation of these tax increases has made the incompetence and thoughtlessness of this administration even more obvious. The nonsensical “reciprocal” tariff rates published on April 2 were, as best as anyone can tell, generated by a chatbot. Officials repeatedly lied about how the rates were calculated, claiming that each economy’s “tariff and non tariff barriers” policies were quantified individually, when all they did was use a simple-minded formula based on bilateral trade balances in goods with the U.S., which imply nothing about anything.2 It is becoming increasingly difficult for the administration’s apologists to claim that there is some grand strategy here, or a secret plan, or a complex design that outside observers are simply failing to understand.
Traders have responded by placing a “moron risk premium”—to borrow a term from the U.K.’s minibudget misadventure—on U.S. assets, as well as marking down global growth forecasts. That helps explain why the U.S. dollar fell in response to the tariffs along with stocks, when standard theory (and prior experience) had suggested that an appreciating dollar would offset some of the impact.
The tariff hikes will harm Americans, they will harm people in the rest of the world, and they will likely fail to accomplish whatever they are supposed to do. Nevertheless, it is important to understand exactly why the tariffs are bad. Some commentators are already overcorrecting by reverting to old (wrong) views on trade and global imbalances, or by endorsing policy alternatives to tariffs that would be almost if not equally harmful.
I have three basic points:
Tariffs are bad mostly because (but not only because) they are a form of fiscal tightening
Trade and current account deficits are not inherently bad, but that does not mean they are always benign, either
These specific tariffs are the product of exceptionally weak analysis
Tariffs as Taxes
As I explained in Politico back in January:
Tariffs are taxes paid by importers on goods coming across the border. They can make U.S.-made goods look relatively cheaper — to Americans, anyway — compared to foreign-made goods. In theory, that would boost employment, wages and profits for American manufacturers compared to a world with no tariffs. But those gains would only come at the cost of forcing American consumers to spend more money to buy the same (or fewer) goods, which would mean there is less money available for everything else.
The net effect depends on how easily American workers and factories can ramp up production of goods that are currently imported. For goods where U.S. demand is relatively low but domestic capacity is rising rapidly thanks to government subsidies, such as battery electric vehicles, the benefits of tariffs could outweigh the costs. At the other extreme would be tariffs on imports of goods where demand is strong and domestic capacity is extremely constrained, such as coffee beans. There, tariffs would be closer to a sales tax that takes money from American consumers to reduce the federal budget deficit. In the middle are goods that Americans could make more of, but only by moving workers and machines away from other activities: more U.S.-made T-shirts, but fewer childcare workers.
In other words, tariffs “work” by reducing consumers’ purchasing power. To the extent that foreigners suffer, or “pay” the tariff, it is because they respond to this loss of their customers’ purchasing power with some combination of lower prices and lower volumes. It is possible, albeit unlikely, that import demand is so insensitive to price changes of this magnitude (especially given how much certain imports are marked up) that the entire cost of the tariffs is split between American consumers and American producers, wholesalers, and retailers, with foreigners no worse off than before.
These are not unique features of tariffs.
If Congress passed legislation to increase payroll taxes by 2% of GDP in an effort to reduce the budget deficit, the take home pay of many Americans would immediately drop, and this would almost certainly lead to lower spending on goods and services. To the extent that spending on imports falls in line with spending on everything else, foreign producers would effectively “pay” for the tax increase with some combination of lower prices and lower volumes. And to the extent that exports fall by less, because the rest of the world is not raising taxes on domestic consumers, the net result would be a contraction of the trade deficit that partly mitigates the drop in GDP.
Similarly, if Congress created a national sales tax worth 2% of GDP, or every state raised its sales tax such that the cumulative effect was worth 2% of GDP, or if Congress created a Value-Added Tax (VAT) worth 2% of GDP, all in an effort to reduce the budget deficit, then the effect would be even closer to a tariff. Consumers’ take home pay would not necessarily have changed, but the real value of that income would suddenly be lower. That would presumably lead to lower volumes of purchases, which in turn would eventually hit foreigners who wanted to sell to Americans, just as a tariff would.
Or, going the other way around, if Congress decided to cut federal government spending by 2% of GDP in an effort to shrink the budget deficit, that would also reduce Americans’ purchasing power. At least part of that cut would presumably lead to less spending on imports, and would therefore present producers in the rest of the world with the same set of choices they had faced when dealing with tariffs. Moreover, depending on the specific nature of the spending cuts, the hit to foreigners could be larger. Downsizing or closing military bases abroad without any offsetting spending increases or tax cuts, for example, would impose direct economic costs on the host societies.
A more complicated case is when there is no net fiscal tightening. But while the direct contractionary impact of higher taxes can be offset by higher spending, that does not mean that the net result of the higher taxes + higher spending is zero. It depends on the specific characteristics of the taxes and the nature of the spending. If the dollar value of the tariff increases imposed over the past two months were somehow balanced out by other measures that increased Americans’ total spending on goods and services, the immediate cost to American consumers and to foreign producers would be far smaller, although greater than zero.
If tariff revenues were used to pay subsidies to exporters, for example, then the net effect would be a transfer of purchasing power from local consumers to foreign consumers, while incomes would be redistributed from foreign producers to local producers. In that case, the cost would be that inefficient businesses might be kept alive at the expense of consumers stuck with inferior products. Besides lowering living standards and hitting productivity, this could encourage corruption as companies try to expand their tariff+subsidy regime at the expense of the consuming public.
But again, this is not unique to tariffs.
When a country’s currency depreciates, its importers lose purchasing power, while its exporters gain income through some combination of higher margins (if the prices charged to foreigners in their currency stay constant while costs in local fx go down) and higher volume (if prices charged to foreigners are allowed to fall). The flip side is that consumers in the rest of the world gain purchasing power as their currencies appreciate, while their exporters come under margin pressure and/or lose market share. It is possible to replicate the exact impact of a currency depreciation using tariffs and subsidies, but depreciations are much less controversial.
As it happens, the U.S. dollar has appreciated by 20% in real terms since January 2021, and by 35% relative to its trading partners since the summer of 2014.
There are other ways to replicate the tariff+subsidy combination. Writing in the teeth of the euro crisis, the economists Emmanuel Farhi, Gita Gopinath, and Oleg Itskhoki helpfully pointed out that countries with fixed exchange rates could do a “fiscal devaluation” by simultaneously lowering companies’ input costs (payroll taxes) while reducing consumers’ purchasing power by an equivalent amount (raising VAT). Unlike actual currency devaluations or tariffs+subsidies, these policies do not directly affect foreigners, but the consequence is the same.
From this perspective, it is strange to suggest that countries that want to “reduce unwelcome capital inflows” should embrace “a tighter fiscal policy that raises national saving” as if that were meaningfully different from tariffs, rather than roughly equivalent to tariffs.3 In both cases, the government is squeezing domestic consumers with the ultimate aim of reducing spending on imports.
Living Below Our Means vs. Positive-Sum Trade
It is worth rereading this passage from the end of The General Theory of Employment, Interest, and Money (1936):
If nations can learn to provide themselves with full employment by their domestic policy (and, we must add, if they can also attain equilibrium in the trend of their population), there need be no important economic forces calculated to set the interest of one country against that of its neighbours.
There would still be room for the international division of labour and for international lending in appropriate conditions. But there would no longer be a pressing motive why one country need force its wares on another or repulse the offerings of its neighbour, not because this was necessary to enable it to pay for what it wished to purchase, but with the express object of upsetting the equilibrium of payments so as to develop a balance of trade in its own favour.
International trade would cease to be what it is, namely, a desperate expedient to maintain employment at home by forcing sales on foreign markets and restricting purchases, which, if successful, will merely shift the problem of unemployment to the neighbour which is worsted in the struggle, but a willing and unimpeded exchange of goods and services in conditions of mutual advantage.
Unfortunately, many nations have not learned “how to provide themselves with full employment by their domestic policy”. Instead, for various reasons, they have come to rely on external demand, which in practice means mostly the U.S. (If you want the short version of why, you should read this, and for the long version you should read Trade Wars Are Class Wars.)
This is bad for them, because it means that their workers and consumers are living below their means, but it has also been bad for Americans. When people in the rest of the world spend less than they can afford on goods and services, they create a shortfall of income in the U.S. relative to what is needed to absorb U.S. production plus the rest of the world’s surplus production. In those circumstances, the only way the U.S. can maintain something resembling full employment is to consistently sell assets to cover that shortfall. The alternative would be to spend less, which would make both Americans and people in the rest of the world worse off.
Ideally, foreigners could be persuaded to stop leaving money on the table and raise their own living standards by enjoying more of the fruits of their labor. That actually seems to be happening in Europe right now, albeit only under extreme duress and at the cost of America’s international standing. But the rest of the major surplus societies do not seem to be moving in that direction.
This is a real challenge that must be managed, rather than ignored. The question is: how?
Focusing on the trade balance itself is a waste of time. The easiest way to compress the deficit is to crash the economy. That is, after all, effectively what those who recommend cutting government spending and raising taxes to boost “national saving” are recommending—with some of the domestic impact on employment potentially offset via currency depreciation that transfers some of the pain to foreigners.
Instead of that negative-sum approach, policymakers should concentrate on offsetting the specific symptoms of weak foreign demand. That means:
Setting the macro stance consistently hot enough to keep businesses operating at full tilt (without overdoing it)
Making sure that the associated financial inflows from abroad are sustainable, which in practice means making sure that federal government debt issuance is high enough to crowd out excessive private borrowing
Finding some way to ensure that domestic manufacturing is not inadvertantly displaced by foreign dumping/moribund export markets
Failing to do this means risking some unpleasant combination of underemployment, an unsustainable private borrowing boom, and deindustrialization. Getting it right is not easy, but the current administration is not even trying. Speaking of which…
The Inanity of the Recent Tariffs
There are many problems with this administration’s approach to trade, but I want to focus on one that I have not seen discussed as much, but that is nevertheless revealing. The tariff schedule published on April 2 is predicated on the assumption that the aggregate U.S. trade deficit should be understood as the sum of U.S. bilateral trade deficits with individual countries. That is why the chatbot formula they used to generate the “reciprocal” tariff schedule penalizes countries with larger bilateral goods trade surpluses with the U.S. regardless of their actual tariff rates.
But this assumption makes no sense, because trade occurs across borders for all sorts of different reasons. The only way to determine whether a country is contributing more on net to global demand or supply is by looking at its aggregate balance of payments with all countries, not individual bilateral relationships that are more often affected by product mix, geography, and corporate tax optimization.
One big clue that bilateral balances are meaningless is that the U.S. runs large trade surpluses with the Netherlands, the UAE, Hong Kong, and Singapore, even though all of those countries have massive trade surpluses with the world as a whole. The reason is simply that there are major ports in all of those places, and the U.S. data only show where exports land before they are shipped elsewhere.
Cambodians and Vietnamese were hit with some of the highest “reciprocal” tariff rates even though they almost certainly import some goods from the U.S. via Singapore, as do Indonesians, Malaysians, and Thais. Yet those are wrongly counted as “Singaporean” imports.
The U.S. also runs large trade surpluses with Australia, which also tells us nothing about the overall external position of either Australia or the U.S. Michael Pettis and I wrote about this instructive example in Trade Wars Are Class Wars:
Countries that spend more than they earn are not responsible for the current account deficits of their trade partners, regardless of what the bilateral data may indicate. America’s persistently large bilateral surplus with Australia, for example, does not explain Australia’s overall current account deficit, because Australians and Americans both spend more than they earn.
Australians happen to import more from the United States than they export to it, but this does not change the fact that both countries are in the same basic situation. Money earned from U.S. exports to Australia gets spent on gadgets or solar panels from China, which generates income to buy coal and iron ore from Australia.
As it happens, Australia’s trade deficit with the United States is more than offset by Australia’s trade surplus with China. That bilateral surplus is not enough to prevent Australia from having an overall current account deficit with the rest of the world, nor is it enough to prevent China from having a large surplus. The global relation is what matters.
The administration’s chatbot approach produced other strange outcomes. On March 31—Monday, believe it or not—the U.S. Trade Representative published its yearly National Trade Estimate Report on Foreign Trade Barriers. A reasonable person might have thought that this 397-page document would have informed the “reciprocal” tariff schedule, but of course it did not.
Amusingly, Brazil, which was dinged in the report for having “relatively high tariffs on imports across a wide range of sectors” and a “lack of predictability with regard to tariff rates” ended up getting the lowest possible tariff rate of 10% because the U.S. happens to have a trade surplus with Brazil. Interestingly the disconnect between the apparent trade barriers and the actual trade data does not seem to have influenced the administration’s thinking on the relationship between trade barriers, bilateral trade balances, and net trade balances.
The flip side is that some of the largest bilateral U.S. trade deficits—most obviously the one with Mexico—are also deeply misleading. As it happens, Mexico runs large trade and current account deficits with the rest of the world as a whole, and is therefore a net contributor to global demand for goods and services. As we wrote in the book, the implication is that penalizing Mexican exports for a consequence of geography would be stupid:
At best, Mexico’s shrinking bilateral surplus with the United States would be exactly offset by rising U.S. deficits with the rest of the world as those countries collectively lost income by selling fewer exports to Mexico…Mexico’s large bilateral trade surplus with the United States is mainly a consequence of its location next to the world’s largest consumer market. American, European, and Japanese manufacturers have spent decades establishing factories in Mexico to build components and assemble products to ship north.
Yet here we are.
The average effective tariff rate is up about 20-25 percentage points and goods imports are worth about 11% of GDP. The actual amount of money collected is likely going to be lower, and the amount depends on how much spending on imports falls relative to GDP. This is why others’ projections of additional revenues over the next 10 years are closer to 1% of GDP. I think it makes more sense to think about the increase in the expected tax burden by looking at the change in the tariff rate *before* GDP and imports fall. Also worth noting that some tariffs have yet to be imposed (pharma, semiconductors, copper, etc) but are supposedly in the pipeline.
One wag observed that this is effectively a “disparate impact” case.
Worth noting that the Irish and Spanish governments ran extraordinarily tight fiscal policies in the 2000s, and it did not help them.