The economic and investment implications of higher tariffs

Complexity and trade

The traditional view of global trade has evolved very little over the centuries. Many voters (and politicians) seem to regard trade in quite simplistic terms—goods are made at home from domestic materials, packed onto a ship, and sold to a foreigner.

Modern trade is almost nothing like this. Two related trends have emerged over the years. First, up to two-thirds of global trade is not taking place between separate entities, but is simply moving goods around inside companies with a global footprint. The rise of global corporations with production facilities scattered across several countries means that trade may represent moving parts up and down an internal supply chain.

In part because of this, most of the increase in global trade in goods over the past thirty years has been due to increasingly complex supply chains rather than necessarily an increase in domestic consumers’ desire to buy “foreign-made” goods. This can be easily seen with the increased importance of intermediate goods (components, essentially) as a share of overall trade.

Thirty years ago, a company might import raw materials from an external supplier abroad, manufacture at home, and export a finished product to a final consumer. Today, a company’s supply chain likely involves its product passing among different subsidiaries located in perhaps eight or ten different countries before a final sale to an end user at home. This trend created large and politically significant shifts in the location where tradable goods were produced, as well as in the associated levels of employment, even as global trade volumes rose.

These changes in global trade mean that the impact of tariffs on the global economy has evolved in recent years. The economic pain of tariffs may fall on domestic producers as much as overseas companies. It means that using past trade taxes to judge the economic impact of future potential tariffs is not very helpful.

For example, US President Nixon imposed a universal 10% “import surcharge” by presidential proclamation in August 1971, with the aim of forcing other countries to revalue their currencies against the US dollar. However, this general tariff increase (the first since the infamous Smoot-Hawley tariff of 1930) was accompanied by government price and wage controls which in the very short term prevented any domestic inflation impact.

Ultimately, the price controls were a disaster, triggering product shortages and helping to fuel a wage-price spiral when they imploded; it is unlikely price controls would be repeated today. Nixon lifted the tariff by the end of the year. That would leave current-day US prices and economic output more exposed to the effects of a universal tariff. US imports of goods were 3.4% of GDP in 1971, compared to 12.7% in 2023 (see Fig. 2). The economic impact of import taxes (i.e., tariffs) were therefore significantly less than they would be today.

President Trump’s imposition of tariffs on selected imports from China may be looked upon as a more modern example of trade taxation. The value of goods imported directly from China that are subject to tariffs has fallen by over 50% compared to 2017 levels. However, the value of these same imports from other countries has risen by more than the value that China’s imports have fallen (see Fig. 3).

The rest of the world has generally replaced China in supplying taxed products to the US. In addition, some of this shift in supply chains may represent the rerouting of China’s exports via third countries. An export from China that stops in Canada may have a maple leaf sticker applied to the side of the box and recorded as a Canadian sale to the US. Mexico, Canada, and the euro area picked up three-quarters of China’s lost market share in tariffed goods between 2018 and 2020. The complexity of global supply chains can make it difficult to identify and tax products that are small parts of a larger manufacturing process.

What do tariffs actually do?

A tariff is a sales tax that is applied after goods arrive in a country. The tax is paid by the domestic buyer—and tariffs are often applied with the intention of changing the behavior of domestic buyers. The buyer (importer) is generally a company, not a retail consumer. The buyer pays a higher price, either because they pay the tax that the tariff represents or because they purchase a higher priced (or conceivably lower quality) alternative. In theory, the exporter could cut prices in order to accommodate the tax burden. In practice, this rarely happens. Analysis of the 2018 US tariffs against China suggests almost the entirety of the tax was borne by US buyers.¹

The nation in which a seller is located has little incentive to devalue its currency. While devaluation could offset the tariff’s effects, the tariff will increase import costs for the entire domestic economy (and increase the cost of imported components for use in export production, reducing the trade benefit). We therefore do not think of a tariff as being applied to a foreign country, like China, but as being applied to domestic buyers of goods made in China.

Inflation: As a tax, a tariff’s direct impact is on price levels. A tariff will produce a one-off increase in a product’s price. As such, the direct impact of the tariff is to raise inflation rates for a single year as the price reflects an additional tax burden. A tariff will not add to prices subsequently, although it may cause behavioral changes (wage demands, profit-led inflation, reduced competition, etc.) that produce second-round inflationary effects.

There is, however, a big difference between the price that is subject to a tariff and the price that the final consumer pays. There are two reasons for this. In the case of the US, less than one-third of US imports are consumer goods (including passenger cars). Almost 70% of US imports are inputs into domestic production—inevitably, given the role of complex supply chains in trade. A 10% tariff on a semiconductor used in a washing machine does not justify a 10% increase in the price of a washing machine.

Even for consumer goods there is a big difference between the import price level and the consumer price level. A lot happens to a good between arriving at a port (when the tax is applied) and when the consumer buys it. Costs related to advertising, domestic transport, warehousing, and retailing, not to mention profit margins at each stage, all have to be paid for—and these add to the consumer price. As a result, the import price for a consumer good may well be less than half the price paid by the consumer. Even if all of a 10% tax on a product sitting on the dockside in the Port of Los Angeles is passed through to the US consumer, on average that should represent something less than a 5% increase in the price the consumer pays at the store. To give just one example, almost a quarter of what US consumers spend on goods goes to pay for the costs (and profits) of the shop or website, and these are unaffected by the tariff.

Pricing power is also important. Where imports are intermediate goods, there is the possibility that the tariff will be absorbed by slightly smaller profit margins along the supply chain. In 2018, companies along US supply chains appeared to be nervous that they could not pass on price increases, and part of the tariff increases meant squeezed margins. If the recent inflation episode has changed the psychology of companies, passing on price increases may be more commonplace than in earlier tariff episodes. A specific, more aggressive form of this is the risk of profit-led inflation, where prices rise by more than is justified by the tariff. Consumers may think that a 10% tariff means that a 10% increase in the price that they pay is “fair” (when it clearly is not).

Perhaps the simplest way to think about the impact of a tariff—a tax on trade—is to consider the relative importance of imports to a nation’s economy. In 2023, imports of goods into the US were the equivalent of 12.7% of GDP. Had a universal 10% additional trade tax been applied to those imports, and had it been passed along supply chains to the consumer, that tariff would have added 1.3% to price levels in the US economy over the subsequent quarters.

Growth: A tariff is a tax, and tax increases are negative for economic activity because they imply consumers have less money to spend. It is, of course, possible that other taxes are cut to offset the tariff, making the tax revenue impact neutral. However, this may still be a net negative. A tariff is a sales tax, and sales taxes will often hit lower-income groups harder because they spend a higher proportion of their income. While an income tax cut may offset the revenue gains from the tariff, the distribution of the income tax cut is unlikely to exactly match the distribution of the tax increase represented by the tariff.

Tariffs may reduce the competitiveness of domestic firms through the impact on intermediate goods prices. A tax on imported intermediate goods (components like microchips) will increase production costs for domestic producers. That puts those manufacturers at a competitive disadvantage relative to foreign competitors. For example, looking at the tariffs from 2018/19, US users of components from China experienced a one-off increase in costs that Canadian users of those same components did not face. Canadian producers were given an immediate competitive advantage. Export Development Canada offered specific advice to help Canadian companies importing components from China that were subject to US tariffs for use in producing finished goods that would be exported to the US. The tariff could not be avoided completely, but as long as there was a sufficiently large difference between the component and the finished product, Canadian firms were making the same thing as their US competitors with lower-cost inputs.

Transnational companies that use imported components may switch production to foreign locations. That reduces domestic production and potentially domestic jobs. For smaller companies that do not have the option of switching locations, this competitiveness challenge will either threaten to reduce sales or alternatively squeeze profit margins.

Firms that use imported components specifically to make exports have a more complicated situation. They may be able to claim Foreign-Trade Zone status. The company effectively pretends that its factory sits outside the taxing country. If components are used to make goods for export, those components are not subject to tariff. This is not costless: As might be imagined, there is a lot of paperwork involved in this sort of process. But this can help to reduce the competitiveness damage of a tariff in a complex supply chain.

For example, in May 2024 BMW submitted a Foreign-Trade Zone request for its factory in Spartanburg, South Carolina to allow it to import components without tariff, which could be used to manufacture cars for export from the US. The exemptions would apply to a range of components from glass adhesives to crankshaft sensors.

Economic activity will also potentially be affected by retaliation from trade partners. As other countries put in place retaliatory tariffs, their own consumption growth will fall, thereby slowing global growth.

Growth: A universal tariff is likely to drag growth lower via three mechanisms:

1. Reducing domestic consumption,
2. Reducing competitiveness of domestic producers,
3. Impacting exports through competitiveness and retaliatory measures.

Lower-income consumers will be harder hit by a trade tax than higher-income consumers, because sales taxes are inherently regressive. Because lower-income consumers spend a higher proportion of their income, consumer spending is reduced. Unless very carefully constructed, tax cuts implemented in conjunction with tariffs are more likely to favor higher-income consumers who have a higher propensity to save.

Firms that rely on imported components will experience an increase in manufacturing costs. Costs will increase with the tariff, and this will reduce the domestic price advantage that a tariff offers. For example, a 10% universal tariff would increase the cost of US washing machine components. It would also increase the price of imported washing machines. If (hypothetically) 30% of the manufacturing cost of a US washing machine is made with imported components, the 10% universal tariff would increase the cost of a US washing machine at the factory gate by 3%. That has to be compared to the cost of a foreign washing machine at the docks (the same point in the supply chain) increasing by 10%. While there is a relative price advantage for the US manufacturer (their costs rise by less than for the importer), there is still an absolute price increase. Fewer people will be able to afford domestic-made washing machines as the foreign component element will add to those prices.

Retaliation: A universal tariff is almost certain to lead to a universal retaliation—and an equal tax increase is likely to be imposed on foreign consumers of domestic manufactured goods. This is not what happened with the Nixon shock in 1971, but that was a universal tariff with a very specific aim (currency revaluations). Although retaliatory tariffs were not imposed while foreign exchange negotiations were underway, retaliatory tariffs were being planned in case these negotiations had not led to a lifting of the Nixon trade tax.

While very specific items may be excluded, it is highly likely that a universal and indiscriminate trade tax by one country will be met with an equal trade tax in all other countries. As with selective tariffs, this may be tweaked so as to maximize the political damage. The more retaliatory tariffs are tweaked, the greater the risk of an escalating tit-for-tat trade war.

Other implications: A universal trade tariff is a very aggressive move; Nixon tariff aside, this has not been seen in advanced industrialized economies in the post-war era. While the tariff would be directed at goods, it may have implications for global capital flows. Countries that initiate a trade tax process are generally current account deficit countries. Current account deficit countries export assets to be able to import goods and services. One consequence of this is that foreigners may own a lot of the tariff country’s assets. That gives foreign governments a potential source of leverage.

In 1995, US President Clinton’s administration was on the brink of a significant trade conflict with Japan. One of the concerns that began to emerge at that time was that the Japanese government might pressure Japanese investors to sell their holdings of US financial assets. The US dollar fell sharply as the dispute continued. Global capital flows are more important today than they were thirty years ago, and weaponizing capital flows in a trade war is a real threat. A decline in US asset values caused by capital flow shifts would have negative effects on US consumers via the wealth effect and possibly higher mortgage rates.

4. Extreme tariffs

In the US, an extreme form of trade taxation has been suggested—not perhaps entirely seriously—as worthy of consideration. This is the idea of scrapping income tax and replacing the lost revenue with a suitable tariff rate. While countries have depended on tariff revenues in the past, including the United States in the nineteenth century, depending on such revenues is normally associated with far smaller governments than any modern society has. The tariffs were also applied to a very different, less integrated trading system.

It is hard to know what level of tariff would have to be applied to generate enough revenue to offset US income tax receipts. This is because the volume of trade would react violently, reducing the revenue by reducing the tariff’s tax base. The economic hit to growth would be sizeable and presumably also affect other possible tax revenues (including state and local tax revenues). To put some context on the idea, in 2023 mercantile imports into the United States were USD 3.1tr. and individual income taxes raised USD 2.2tr. Replacing income tax would require a tariff tax increase in excess of 100% and probably closer to 130% to 150%.

The scenario of replacing income taxes with exceptionally high tariffs is unlikely, but a more dramatic universal tariff is possible if economic nationalism takes hold. Several countries have demonstrated that rampant populism can lead to counterproductive economic policies and self-inflicted harm: the UK voted to increase barriers to trade in a national referendum, for instance. In the event that a universal tariff provoked a tit-for-tat retaliation from other countries, the extreme tariffs would be the direction of travel.

Inflation: The price level increase in this scenario would be extreme, at least initially. However, reported inflation would rapidly cease to reflect inflation reality as consumers would shift consumption significantly (and some goods may not be available at all because of the ensuing supply chain disruption). The fixed weights of different goods in the inflation basket would not match up to this adjusted consumption pattern. While the tariff would affect price levels in a one-off move, it is quite possible that there would need to be a sequence of tariff increases in order to raise the required revenue if tariff revenue were to replace a tax on income (and with the volume of imports falling).

The extreme nature of this sort of price move would also likely produce second-order effects. Anyone who has wage bargaining power would be sure to exercise it. Small businesses in non-traded sectors would also likely raise prices, e.g., the self-employed plumber or the nail salon pushing up prices so that their owners can maintain their real living standard.

Growth: Although it is possible that the extreme tariff scenario is tax-revenue neutral (income tax cuts offsetting the increase in taxes on trade), the combination of a) almost certain aggressive retaliation and b) the redistribution of the tax burden will make this a negative for growth. The motive for extreme tariffs (i.e., revenue raising) gives the tax an aura of permanence. If companies do not regard tariffs as a short-term negotiating tactic, they are likely to start to restructure their businesses in a way that would reduce potential growth in the US. The exit of the UK from the EU (also regarded as permanent, and increasing trade barriers in a far less aggressive manner) hints at potential implications.

Retaliation: In such an extreme scenario the United States would face significant retaliation. This may not be a universal tariff, as countries would prioritize US-produced goods where there are few alternative sources. But there would be a relatively swift adjustment of supply chains, and there are also relatively few areas of modern trade where the US has the sort of market dominance that provides immunity from retaliation.

More seriously, from the US perspective, at some point the dollar would almost certainly lose reserve status. There are many motives for holding a specific currency in reserve—rule of law and liquidity being significant. But ultimately there has to be an assumption that the reserve holder will want to buy the output of the reserve currency country in the future. If there is dramatic trade war between the US and the rest of the world, this assumption becomes implausible. The US has, for many years, exported its assets (often its government bonds) to purchase the goods exports of other countries. If the US seeks to effectively halt the purchase of foreign goods, foreigners are likely to respond by halting the purchase of US assets.

Other implications: Any attempt to replace income tax (broadly progressive) with a sales tax (broadly regressive) would increase inequality in a country. The social implications of this would be quite negative, and it is possible that social mobility would be further reduced.

Figure 5

Election outcomes and tariff scenario probabilities, in %

Figure 6

Cumulative impact to GDP over three years in two tariff scenarios, in %