US deficits: much more than manufacturing
Analyzing the exposure of the US to its own trade deficits, going well beyong manufacturing and jobs, into credit, demand and productivity.
A few months ago, after Liberation Day, I published a review of different theories around US trade deficits. Since then, I have been thinking about how the unraveling of those deficits could affect different industries inside and outside the US. More generally, I have been thinking about how deficits impact areas of the economy, such as prices of factors like labor, land, and capital, innovativeness, productivity, etc.
In today’s mainstream discourse, trade deficits are framed as a manufacturing or defense issue: a matter of lost factories or foreign dependence in strategic sectors. But I have come to believe that this framing misses the point. Trade deficits are not just a problem at the margin: they are a central force shaping the entire U.S. economy.
By financing a permanent excess of expenses above income, deficits sustain a credit cycle that inflates demand across all sectors. Not just goods and imports, but also real estate, finance, services, and technology. Without that excess demand that deficits finance, the productivity (and value) of most domestic service industries in the US, and of many of its champions, would be much lower.
In that sense, the real risk posed by deficits is not geopolitical or industrial, but systemic. They are not just a symptom of global imbalances, but are a core enabler of the U.S. credit cycle itself.
This article tries to explain the reasoning behind these opinions, offers some rough estimates where useful, and presents possible scenarios for what a reversal might look like.
Other articles on macro themes
Trade deficits are debt claims on the US economy
A trade deficit (or surplus) is a form of financing from one country (the surplus) to another (the deficit one).
In most forms of (successful) financing, lent or invested funds are usually returned. First, the capital provider gives capital to the capital recipient, which in exchange delivers a financial asset (debt or equity title). But then, the capital recipient returns that capital, in the form of principal, interest, or dividends, to the capital provider.
If a capital recipient (say, a company) were to receive funding indefinitely, eventually the value of the claims on its assets and future cash flows (its bonds or stocks) would decrease. Eventually, liquidity would dry up and the company would need to work a way around returning its financing.
The US has been an exception to this rule. When the US runs a trade deficit, the capital provider (foreigners) supplies goods or services, and the capital recipient (US economy) delivers financial assets (US dollars, Treasuries, mortgages, corporate bonds, munis, equities).
If the financing operation were to be reversed, the US economy would have to deliver goods or services and receive financial assets in exchange. That is, foreigners would buy goods and services from the US, and pay them with dollars or Treasuries.
However, this has (in the aggregate) never happened, and the US financial claims in the hands of foreigners have expanded continuously for at least 35 years. More importantly, this has not deterred foreigners from holding ever-expanding quantities of US financial assets.
This is the equivalent of a company that, without ever paying a dividend or doing buybacks, keeps issuing equity, and instead of paying its debts or interest, it just keeps issuing more debt. And yet, the debt of this company is considered the safest in the world.
With persistent deficits, there is no external constraint to the US credit cycle
The most important effect of the persistence of trade deficits (or the insistence of foreigners in acquiring US dollar assets, two sides of the same coin) is that the US financial system can create a lot of credit without the country running into a balance of payments problem, as it would in almost any other country.
In a regular country, the expansion of credit has to somewhat go in line with the expansion of wealth and productive capacities, or else eventually there will be problems. Credit creation cannot continue indefinitely because it eventually faces internal or external limitations.
Internally, if the production of goods and services is limited by some factor (say, availability of employees, or machinery, or land), then credit creation will lead to inflation. The Central Bank will probably act before this happens, tighten conditions (raise interest rates), and create a downward cycle in credit. The US has indeed faced these types of internal limitations, which have been managed promptly by the Federal Reserve, sometimes with very challenging results for the economy and equities as the credit cycle is reversed.
In other cases, however, the constraint is external. No economy is 100% self-sufficient, and therefore, as it grows, it demands more goods and services from abroad. Because demand for exports is not fueled by the internal credit boom, imports will grow faster than exports, and eventually (eventually can be a long time), the country will run a trade deficit. This can either lead to the Central Bank tightening credit conditions to cool demand, to a depreciation of the currency, or, in the worst cases, to a balance of payment crisis.
But in the US, this second type of constraint has never been enforced: foreigners are willing to accept the assets that the US financial system creates almost indefinitely (so far). Even more, because those assets remain in the system, they add fuel to the positive credit cycle.
Foreigners finance 15% of all goods and 25% of all debt
The simplest measure of the impact of the trade deficit on the US economy is the trade deficit as a percentage of GDP. This has been about 3% since the GFC, and had reached a high point of almost 6% just before the GFC (Fred). That doesn’t sound too serious. The US economy grows at 3% every year.
However, this measure underweights the importance of deficits in the economy for two reasons.
It fails to account for all the accumulated deficits, which remain in the US financial system, and help maintain the credit cycle going.
They dilute the weight of current deficits on total GDP, of which 80% is services. As we will see in the next section, service productivity (and consequently GDP) is inflated by the existence of deficits.
Starting with the accumulated debt, if we simply add up all of the annual trade deficits since 1992, we arrive at roughly $16 trillion in US financial assets held by foreigners. This is a reasonable low-end estimate of the foreign claims on the U.S. economy, based on how balance of payments accounting works. However, those claims have remained invested in the US over time, generating additional interest or capital gains. As a result, the current value of foreign-held assets is much higher. The BEA calculates that the Net Investment Position is actually negative by $26 trillion. To put this in perspective, that is the equivalent of 100% of all Treasury debt, or approximately 25% of all debt liabilities in the US. This does not mean that all of these foreign claims take the form of debt. Rather, they are spread across different types of assets, like cash, deposits, debt, traded equity, and direct investments. However, their size is significant.
On the goods side, estimating the weight of the trade deficit is more complex, since there are multiple ways to measure the value of goods used in the economy, depending on whether we look at final consumption, production cost, or value added.
First, we can compare the trade deficit with Personal Consumption Expenditures in Goods. In 2024, PCE was $6.2 trillion, while the trade deficit stood at about $1 trillion (Fred, already incorporating the positive effect from export services), which implies that the deficit represents 16% of all US goods consumption. This measure, however, has two limitations. First, consumption is not the only component of GDP expenditures, although it is the biggest (others are Investment and Government consumption). Second, and most importantly, PCE is a final expenditure measure of how much consumers spend, which includes not just the goods themselves but also embedded services and margins. For example, when we buy a Coca-Cola at the supermarket, the goods component is only a portion of the price we pay; the price includes not only the beverage but also labor, rent, logistics, and retailer margins.
To address this, we can turn to the producer-cost side using the BEA’s supply-use tables. According to this data, total product supply of manufactured goods at basic prices (i.e., before margins and distribution costs) was $9 trillion in 2023. That would make the $1 trillion trade deficit equal to about 11% of supply. If we exclude $1.6 trillion in direct exports, since these aren’t part of domestic consumption, the deficit-to-supply ratio climbs to 13.5%.
However, I should note that the calculations are not straightforward, and I’m not an expert on the database. I have the suspicion that the Total Product Supply above double-counts a lot of the supply as intermediates.
That said, the US Department of Commerce, using the same underlying BEA database, recently reported that gross purchases of manufactured goods (defined as consumption, investment, and government spending at producer prices, i.e., before margins) were $3.7 trillion. Of that total, 50% was imported.
If these numbers are more representative of reality, then the trade deficit effectively finances 25% of total manufacturing purchases and 50% of all manufacturing imports. These figures exclude other goods categories such as food, raw materials, mining, and oil & gas.
As another cross-check, we can look at the value added by private goods-producing industries, which stood at about $5 trillion (FRED). If we assume a total goods supply of $6.2 trillion (from a goods deficit of $1.2 trillion) and a trade deficit of $1 trillion, this would also imply that foreign producers account for 16% of the goods supply.
The effects are pervasive across all industries
From the discussion above, it is not hard to see that trade deficits affect far more than just manufacturing. While much attention is placed on the jobs and businesses lost in industry, the real source of fragility is the way deficits support the broader credit cycle, particularly by financing the excess consumption of goods. If that credit cycle were to reverse, the damage wouldn’t just fall on manufacturing or even goods retail industries that depend more directly on imports. It would ripple across the service economy, where no industry is immune.
Additionally, putting the credit cycle at the center of the analysis also changes the way I think about US manufacturing competitiveness. It’s not about low-wage foreign labor or industrial policy abroad. In my view, what really makes US manufacturing uncompetitive is the domestic competition for production factors, driven largely by non-tradable service sectors operating in a high-demand, credit-fueled environment. When services can consistently outbid manufacturing for labor, land, and other inputs, it becomes very hard for industry to survive, regardless of what’s happening overseas.
The deficits make all companies more productive
Any investor worth their salt knows that all companies look more productive and valuable during the positive part of the credit cycle. Yet, for some reason, this has been somewhat forgotten when thinking about the productivity of US companies and US assets.
This is not to deny that Americans and American companies are entrepreneurial, hard-working, innovative, and that the economy is open and competitive, therefore allowing only the most efficient companies to remain in the market. But it suggests that part of what we call “productivity” comes from being in this high-demand environment that the positive credit cycle finances.
Consider a retail company, like Home Depot. When credit expands, consumers spend more. Even with the same store, staff, and layout, Home Depot becomes more productive, selling more in nominal terms (dollars sold), but also in real terms (units sold).
That additional revenue allows the company to expand stores (benefiting construction), increase ad spending (benefiting Meta and Google), and pay for more services (benefiting lawyers, consultants, and accountants). The firms selling to Home Depot can spend more on the process to get the company as a client, and therefore can afford to pay for Salesforce or the latest marketing B2B SaaS product.
This is how one extra unit of demand circulates through the entire economy. But it doesn’t stop there. Being more productive, Home Depot can afford to pay more for factors. If the same employee can sell more products (without any efficiency gain from the employee himself, only because clients have more access to credit), then Home Depot can pay higher salaries. It can also pay more for the land where the stores will sit.
By starting to impact all forms of income (firms, employees, factors like land), the process reproduces across the economy, generating even more income downstream. This is a basic multiplicative effect.
Services end up expelling manufacturing via the cost of factors
In an economy with more credit, all businesses are more productive and are therefore able to pay more for factors like land and labor (or taxes for a bigger state, too).
Coming back to the previous example, in this credit boom economy, Home Depot can afford to pay higher wages and higher rents. The company competes for those inputs locally, within the US, often within a specific region or city. That’s the case for most service sectors, which are not tradable and therefore not exposed to international competition.
Manufacturing, however, doesn’t operate under the same conditions. It competes globally for factors. If it pays up for wages or land, then someone else in China or India, or even Germany, or Japan will outcompete the company. Further, because the US is a trade-open country, domestic buyers will naturally turn to the cheaper global supplier.
In a regular country, this dynamic would definitely result in pressure on the balance of payments and force an adjustment: either a cut of the credit cycle by the Central Bank, or a depreciation that equates the costs of the factors across the economies, adjusted for differences in their real productivity. Then, the US wages or cost of land would be more competitive versus Germany or China (again, adjusting for productivity factors like infrastructure, machinery, workers’ education, legal system, entrepreneurship, etc.).
However, in the US, that correction mechanism doesn’t operate, for all the reasons explained earlier.
Therefore, I’m more inclined to believe that it is not Chinese factories with low wages that push factories out of the US. It’s the combination of competition for factors from non-tradeable services, PLUS the ability of the economy to finance almost infinite deficits.
In recent years, even services have started to suffer from this. Companies tried to hire remotely, paying software developers in 1st world countries with top-notch education maybe ½ or ¼ of what an American developer costs. Or even importing non-tradeable services, via the import of large masses of migrants who then send US financial assets back home in the form of remittances. If a Nicaraguan worker mows the lawn of a house in Arizona and then sends 40% of his paycheck to Managua, isn’t that the same as the service being exported from Managua?
Almost no industry is a safe haven
Because of the pervasive effect that credit has across the economy, and the magnitude and extent of US trade deficits (or Foreign credit surpluses), it is hard to conceive of any company or industry that would be unscathed by a reversal of this trend.
That is why the deficits are such a deep source of instability for the US economy.
Just as an example: Google, Meta, and Microsoft all derive 50% of their revenues from the US, and only 25% from EMEA. This is despite the EU having twice as many inhabitants as the US, and a PPP GDP that is comparable (nominal GDP 30% smaller). Why is that? Because a business in the US is much more willing to pay for advertising or cloud services. Those expenses are more likely to convert into sales in a credit-fueled domestic market.
Will the status quo ever change?
The question that remains is much more complex: are there plausible scenarios in which the reality of the past 40 years breaks down? If so, what would trigger that change, how would it unfold, and what kind of world would emerge on the other side?
For those making market decisions, a related question is which assets would benefit under different outcomes, and which would not.
The strength of the US financial preeminence
My first conclusion is that, as long as foreigners collectively prefer to hold US financial assets instead of other kinds of assets (financial or not), then there will be no meaningful change to deficits. Further, the attractiveness of the US as a place to park funds cannot be denied without entering into contradiction with the reality of the last 40 years.
There are many reasons why foreigners might prefer US assets despite the risk of a credit cycle reversal. Some of these are: higher protections to property, lower taxes, lower law risks, cultural or traditional factors, and the indirect exposure of other countries to the same cycle. Some may be grounded in fundamentals; others may not. But perception is what ultimately matters. And as long as the perception holds, the credit cycle can continue—deficits can persist without triggering a crisis, and the U.S. economy can remain uniquely exempt from the external constraints that bind most countries.
Even more, the process is self-reinforcing because credit generated in the US makes companies in the country more productive, which in turn attracts capital from abroad, providing the financing for the credit cycle to exist without triggering a balance of payments crisis.
Therefore, the chances of a ‘business as usual’ scenario are high, even under conditions of financial stress inside the US. The example of the GFC illustrates this clearly.
Tariffs & a depreciation would probably be ineffective
I also believe that tariffs, even if applied unilaterally and not offset by subsidies or depreciation, would not solve the deficit issue over the long run.
The effect of tariffs is to increase the cost of imported goods to the US versus domestic goods. For some time, foreign producers might absorb part of the cost, eating their margin. But they will not sell at a loss forever. Eventually, prices adjust, and the costs of imports will match the foreign production cost plus the tariffs.
This might give U.S. producers a temporary advantage. However, if credit continues to expand, eventually service industries will push production factors up just as has happened before, making US production uncompetitive again.
For the same reason, a one-time depreciation of the USD against other currencies would work only temporarily (as long as it doesn’t trigger a selloff of US assets).
An (unlikely) expansion of export demand
A more sustainable way to reduce the deficit, while maintaining the credit cycle going, is for demand for US exports to increase significantly. In that case, foreigners would be in part redeeming their US claims by purchasing US goods and services, effectively closing the loop.
But this scenario is hard to imagine under current conditions. Absent a major technological breakthrough (say around AI or robotics) that radically shifts the global distribution of factors (labor above all), the U.S. is unlikely to become a cost-competitive exporter. If foreigners suddenly demanded more goods or services from abroad, they could probably find them cheaper in other regions.
A related solution would be bilateral agreements, where a country commits to purchasing US goods or services, just for the sake of doing so, with economic reasons as a second consideration. This could apply in specific areas such as food, oil, minerals, or chemicals, sectors where the U.S. retains comparative strength.
Dollar panic and debt monetization
In my opinion, if the US trade deficit is ever meaningfully corrected, it won’t come from tariffs, depreciation, or export booms: it will come via a financial shock. Specifically, a crash in U.S. asset prices that forces debt monetization and marks the beginning of a new regime in which US financial assets lose at least part of their global appeal permanently. I’m not saying this is what will happen (in fact, I’m more inclined towards the ‘business as usual’ scenario), but I do believe it’s the only durable path to structural change.
This scenario could come about in one of two ways, both leading to a credit crunch in the US, followed by monetization to save the system.
The first type is a domestic credit crisis, as seen during the GFC or COVID. During a credit panic, interest rates spike across the economy. Because higher rates in US debt markets would trigger a very negative economic cycle, the most likely outcome is that the US economic authorities would monetize the debt (Quantitative Easing). This could lead to a depreciation of dollar assets (particularly debt) that requires even more monetization, and eventually currency depreciation.
The second type involves a shift in foreign behavior. If foreign investors become less willing to hold U.S. financial assets, or if the total stock of dollar assets exceeds what the world wants to hold, then those assets must fall in price. That means depreciation, not just of the dollar, but of Treasuries, equities, and other U.S. liabilities. This depreciation would generate inflation in the US, forcing monetization and further downward pressure on the dollar.
Eventually, U.S. factor prices (wages, land, capital) would become cheap enough, relative to foreign factors, such that the balance of payments can square again.
Of course, nothing in economics works as smoothly; this is just a schematic. The important notion is that the only way for trade deficits to decrease permanently is for US factor prices to depreciate, which can only come about by inflation and depreciation caused by monetization of debts that are not absorbed by foreigners.
Examples from GFC & COVID
The Global Financial Crisis offered a textbook case of credit collapse followed by aggressive monetization, but without the expected consequences. Despite massive monetization (QE, ZIRP), the US did not experience sustained inflation or currency depreciation. Foreign investors continued to hold U.S. assets, absorbing the flood of newly issued dollars and Treasuries.
At first, there was flight: gold almost tripled from 2009 into 2011. Raw materials and ex-US markets also did well. Many would have believed back then (correctly from theory) that the USD would depreciate, and that other economies would appreciate. However, this didn’t happen. After two or three years, US assets dominated again. The U.S. effectively exported its debt monetization policy without triggering a loss of confidence in its currency or financial system.
This stemmed entirely from the willingness of foreigners to continue holding US assets despite their supply increasing so much (via printing). I can’t fully explain why this happened, why foreigners kept buying Treasuries and dollars despite the scale of QE. But part of the answer probably lies in the lack of viable alternatives: the EU, China, and Japan were all dealing with their own structural and cyclical problems. As stated above, the US appeal is real and is, in fact, sustained on structural factors.
COVID followed a similar initial script of crisis, intervention, and more QE, but with a different outcome, which is still developing. This time, inflation materialized. First, global supply chains couldn’t accommodate the surge in demand, creating real bottlenecks. Second, foreign appetite for US assets began to weaken. While internal tightening by the Fed played a role, the external dimension mattered too. Treasury markets became less forgiving. That trend accelerated after the US froze Russian assets and adopted a more confrontational global stance. We saw another example of this when Liberation Day triggered a Treasury’s sell-off that many believe led the US administration to roll back the initial tariffs.
All of this raises a pressing question: What would happen today if the US had another internally generated credit cycle problem, but the world no longer absorbs the response? In that case, the real risk lies in the stock of US assets held abroad. A partial unwinding could trigger a feedback loop of higher rates, monetization, and asset devaluation.
Hard currencies - Bitcoin, gold
Since the post-COVID wave of monetization, both Bitcoin and gold have performed extremely well. If these became safe assets of choice, they could have an outsized impact on the ability of the US to continue financing its trade deficits, or to monetize its credit cycle.
The logic is simple. If Bitcoin or gold are demanded, it flows out of dollars and Treasuries, the prices of which have to adjust downward.
Bitcoin represents $2 trillion, a drop in the bucket of the value of global financial assets. However, gold’s current market cap is $22 trillion, close to the market value of all Treasury debt. That means a lot of people already prefer Bitcoin or gold to US assets.
The US credit cycle is also somewhat global
Finally, we should not forget that US demand is a source of global demand. If US demand recedes, foreign companies are affected too, and therefore a reaccommodation of US demand will affect all countries.
However, again, the question is how strong that effect is. World Bank data points to $16 trillion in global value added in manufacturing, meaning US deficits are about 6%. Depending on the country, it could be more, however. China’s VAM is almost $5 trillion, meaning US deficits are 20%.
There are obvious second-order effects that are harder to calculate. It is unlikely that higher inflation and depreciation in the US would only lead to a shrinking of the trade deficit without any additional fall in demand for goods. It is also unlikely that the credit cycles abroad will cause problems with services abroad, too. Finally, a depreciation of dollar assets would make the assets held by foreigners in the US worth less, also causing a negative wealth effect.
I would believe that the effect of the US foreign-financed credit cycle unwinding will be more impactful on the US than abroad, but I’m not really sure any equities would do better.
Conclusions
I obviously cannot claim to know what will happen in the next few years, whether the turning point comes from an internal credit event, an external shock, or simply never comes at all.
What makes this particularly challenging is that the very mechanism that enabled the rise of the U.S. service economy, persistent deficits funded by foreign demand for dollar assets, is at the same time a reinforcing factor, and also its greatest vulnerability.
Further, I can speculate that almost no US industry is safe or unaffected by the potential reversal of such an unstable factor. Not because of tariffs or technological shifts, but because the underlying foundation (credit expansion without external constraint) has shaped the entire economy.
What ends up happening is impossible to predict, and the more illogical outcomes a priori can be the most logical a posteriori. Thinking through those paths may not yield predictions, but it has helped me understand where the fragilities lie, and how far they reach. Hope they helped you as well.