Pettis prend parti dans un débat qui en fait est au centre des préoccupations de l’équipe de Trump: la réduction des déficits, l’accroissement de la dette, la desindustrialisation.

Il prend parti en expliquant que la théorie de l’épargne excédentaire des uns et de l’insuffisance d’épargne des autres (Bernanke) est fausse et quelle inverse le sens des causalités,

Je pense depuis longtemps comme Michael Pettis , je mènerais le développement autrement mais celui qu’il nous propose est déjà assez convaincant. Pettis va dans le sens d’une taxe sur les entrées de capitaux qui ne me semble pas tres réaliste mais qui tient debout au plan théorique.

Les Entrées De Capitaux Étrangers N’Abaissent Pas Les Taux D’Intérêt Américains

Contrairement à la pensée conventionnelle, les entrées nettes de capitaux étrangers ne font pas baisser les taux d’intérêt américains (à moins qu’elles ne le fassent en augmentant le chômage aux États-Unis). Il est donc peu probable qu’une taxe sur les entrées étrangères provoque une hausse des taux d’intérêt américains et ce pourrait en fait être un moyen efficace d’endiguer les augmentations de la dette américaine résultant d’entrées illimitées de capitaux.

par Michael Pettis
Publié le 7 juillet 2025

China Financial Markets fournit une analyse approfondie de l’une des économies les plus importantes et les plus vitales du monde. Édité par Michael Pettis, Senior Fellow de Carnegie basé à Pékin, China Financial Markets offre des informations mensuelles sur les inégalités de revenus, les structures de marché et d’autres problèmes affectant la Chine et d’autres économies mondiales. Expert reconnu de l’économie chinoise, Pettis est professeur de finance à la Guanghua School of Management de l’Université de Pékin, où il se spécialise dans les marchés financiers chinois.

Michael Pettis
Dans un commentaire récent sur les raisons pour lesquelles les entrées nettes persistantes de capitaux sont un problème pour l’économie mondiale, Martin Wolf, rédacteur économique au Financial Times, a suggéré qu’un moyen “évident” de résoudre les déséquilibres mondiaux serait d’imposer une taxe sur les entrées de capitaux.

S’il a presque certainement raison—une taxe sur les entrées de capitaux est le moyen le plus efficace de limiter le rôle des États—Unis en tant que principal accommodateur des déséquilibres mondiaux-de nombreux économistes s’opposent à une telle taxe au motif qu’elle augmenterait le coût du capital pour les entreprises américaines et augmenterait les taux d’intérêt pour le gouvernement fédéral.

Leur point de vue est implicitement fondé sur la prémisse que l’investissement américain souffre de coûts élevés du capital causés par la rareté de l’épargne intérieure.

Si cette prémisse était correcte, il serait logique de soutenir que les entrées nettes de capitaux de l’étranger abaissent les taux d’intérêt américains en assouplissant la contrainte de l’épargne intérieure et que, par extension, toute politique réduisant les entrées nettes de capitaux étrangers pourrait augmenter les taux d’intérêt américains.

Mais cette prémisse est fausse.

Si le capital peut effectivement être rare dans les pays en développement ayant des besoins d’investissement élevés, il a depuis longtemps cessé de l’être dans la plupart des économies avancées.

Aux États-Unis, par exemple, les entreprises peuvent s’abstenir d’accroître leur production pour de nombreuses raisons, mais c’est rarement parce qu’elles n’ont pas accès au capital.

C’est une différence cruciale pour au moins deux raisons.

Premièrement, si l’investissement des entreprises américaines n’est pas limité par l’épargne, l’économie américaine ne s’ajustera pas aux entrées nettes de capitaux étrangers avec un investissement intérieur plus élevé et devra donc s’ajuster avec une épargne intérieure plus faible.1

Comme expliqué ci-dessous, les ajustements les plus probables qui réduisent l’épargne intérieure impliquent soit un chômage plus élevé, soit une dette plus élevée. En d’autres termes, les entrées étrangères peuvent forcer la création de dettes supplémentaires de sorte que l’augmentation globale de la demande de dette américaine soit compensée par une augmentation globale de l’offre—et dans ce cas, cela ne fera pas baisser les taux d’intérêt américains.

Et deuxièmement, si les entrées de capitaux étrangers ne sont pas motivées par l’incapacité des entreprises américaines à financer l’investissement intérieur, elles doivent être motivées par des distorsions à l’étranger. C’est pourquoi non seulement ces entrées sont peu susceptibles de réduire les coûts d’emprunt, mais elles peuvent aussi, paradoxalement, forcer des changements indésirables dans la structure de l’économie américaine qui créent les conditions dans lesquelles davantage d’emprunts deviennent nécessaires uniquement pour soutenir la demande.

Ce n’est pas une coïncidence, après tout, si parmi les économies avancées, celles qui reçoivent le plus d’entrées nettes de capitaux étrangers—comme les États—Unis, le Royaume-Uni et le Canada-sont beaucoup plus susceptibles d’être caractérisées parmi leurs pairs par des niveaux d’endettement plus élevés que par des taux d’intérêt plus bas.

L’Erreur de la Rareté
Pour comprendre l’erreur de diagnostic conventionnelle, nous devons revisiter le modèle classique dont elle découle. Dans un monde où l’épargne intérieure est insuffisante pour financer des opportunités d’investissement à haut rendement, les capitaux étrangers sont un ajout bienvenu.

Dans un tel modèle-applicable, par exemple, à l’Amérique du XIXe siècle et à de nombreux pays en développement aujourd’hui—les entrées extérieures soutiennent la croissance en finançant des investissements indispensables dans les infrastructures, l’industrie et des projets d’amélioration de la productivité qui, autrement, ne seraient pas financés.

Mais ce n’est pas la réalité de l’économie américaine du XXIe siècle, ni même de la plupart des économies avancées. Les entreprises américaines, par exemple, ne sont pas incapables d’investir parce qu’elles manquent de fonds. Non seulement ils ont accès à l’échelle nationale aux marchés de la dette et des actions les plus liquides et flexibles au monde, mais même après avoir utilisé des billions de dollars principalement pour racheter des actions, verser des dividendes ou acquérir des concurrents, ils disposent toujours de près de 7 trillions de dollars en espèces et équivalents de trésorerie, soit près d’un quart du PIB américain. Ils choisissent de ne pas investir dans de nouvelles usines ou technologies.

Ce n’est pas parce qu’ils sont myopes.

C’est parce qu’ils ont du mal à vendre leur capacité de production existante et, par conséquent, l’augmentation du montant du financement net dont ils disposent ne les incitera pas à augmenter leurs investissements. *

*Note BB: Michael serait plus convaincant dans sa demonstration si il osait expliquer que la raison pour laquelle les Etats Unis n’investissent pas plus ce n’est pas par manque d’argent, il est surabondant mais par manque de profit previsible sur ces investissements. L’argent est surabondant , quasi gratuit tant en capital qu’en dette et si le système n’investit pas plus et préfère faire des rachats d ‘action -des buy back – par trillions c’est parce que la profitabilité prévisble est insuffisante par rapport à la profitabilité souhaitée.

Ironiquement, cette faiblesse de la demande intérieure est amplifiée par les entrées de capitaux mêmes qui sont censées aider car, en faisant monter la valeur du dollar, ces entrées rendent les importations étrangères plus compétitives que les produits fabriqués sur le marché intérieur. En d’autres termes, plutôt que de stimuler les investissements américains dans les installations de production nationales, les entrées nettes peuvent en fait rendre cet investissement moins souhaitable.

Il aide à comprendre l’effet des entrées nettes de capitaux sur les taux d’intérêt intérieurs en examinant les origines de ces entrées de capitaux. Selon la plupart des conceptions économiques dominantes, l’épargne étrangère est “tirée” vers les États-Unis alors que les Américains importent des capitaux pour financer un déficit d’épargne domestique. Pour ceux qui souscrivent à ce point de vue, le fait que les entrées nettes de capitaux étrangers soient par définition égales à l’excédent de l’investissement américain sur l’épargne américaine ne peut s’expliquer que d’une seule manière: les entrées nettes doivent être attirées aux États-Unis par le besoin des entreprises américaines de financer l’investissement.

Mais cela nie ce qui se passe à l’ étranger.

Des pays comme la Chine et l’Allemagne ne sont pas simplement des victimes passives des déséquilibres intérieurs américains. Au contraire, le fait que la Chine contrôle pleinement son système bancaire tout en maintenant des contrôles commerciaux stricts et des contrôles de capitaux encore plus stricts—contrairement aux États—Unis-devrait suggérer qu’elle est plus susceptible d’être à l’origine, plutôt qu’à l’absorption, des déséquilibres mondiaux de l’épargne. La réalité est que des pays comme la Chine et l’Allemagne ont délibérément adopté des modèles économiques qui suppriment la consommation intérieure en faveur d’une expansion plus rapide de l’industrie manufacturière. Comme la production croît plus vite que la consommation, les taux d’épargne dans ces pays augmentent automatiquement—conduisant, dans le cas de la Chine, au taux d’épargne le plus élevé de l’histoire.

Mais l’offre a besoin de la demande. Pour éviter la hausse du chômage qui résulterait d’une réduction forcée de la production excédentaire, ces pays doivent exporter leurs excédents de production, en acquérant des actifs étrangers en échange d’une exportation supérieure à celle qu’ils importent. Par conséquent, l’excès d’épargne intérieure dans ces pays est plus susceptible d’être” poussé “à l’étranger par des politiques internes qui stimulent la production par rapport à la consommation que d’être” tiré  » à l’étranger par les conditions dans les économies déficitaires en épargne .

Les États-Unis sont le principal destinataire de ces capitaux exportés. Les gouvernements et les investisseurs étrangers achètent des obligations du Trésor américain, des obligations de sociétés, des actions, des usines, des biens immobiliers et d’autres actifs principalement parce que les États-Unis ont les marchés financiers les plus profonds, les plus liquides, les mieux gouvernés et les plus ouverts au monde.

Les pays excédentaires exportent leur excédent d’épargne vers les États-Unis principalement pour leurs propres raisons nationales, que les entreprises américaines aient besoin ou non des entrées pour accroître l’investissement productif.

La question clé ici est la direction de la causalité. Les entrées ne répondent pas à une pénurie d’épargne aux États-Unis. Ils sont plutôt le résultat d’excédents d’épargne générés à l’étranger par des politiques qui répriment la consommation des ménages.

Mais cette absorption par les États-Unis a un coût. Les déséquilibres intérieurs d’un pays doivent toujours être parfaitement cohérents avec ses déséquilibres extérieurs. En modifiant ses déséquilibres extérieurs, les entrées de capitaux forcent des ajustements dans l’économie américaine—le plus évidemment en augmentant la valeur du dollar, bien qu’il existe de nombreuses autres formes d’ajustement—et ces ajustements rendent les exportations américaines moins compétitives et les importations moins chères. Alors que cette dynamique érode le secteur américain des biens échangeables, le secteur manufacturier diminue et les salaires stagnent.

Si Washington ne veut pas que la contraction de l’industrie manufacturière américaine qui en résulte conduise au chômage, la consommation doit être soutenue. Avec des revenus sous pression et une compétitivité en baisse, les ménages et les gouvernements sont encouragés à emprunter et à dépenser de sorte que le déclin du secteur américain des biens échangeables soit compensé par une expansion ailleurs—la création d’emplois, en d’autres termes, passe des secteurs manufacturiers aux secteurs des services à faible productivité.2

De cette manière, non seulement les entrées de capitaux étrangers ne répondent pas aux besoins d’emprunt existants de l’économie, ils forcent un changement dans la composition de la production américaine—au détriment de l’industrie manufacturière et vers les services—et créent les conditions dans lesquelles des emprunts supplémentaires deviennent nécessaires pour empêcher les déficits commerciaux d’entraîner une hausse du chômage intérieur.

Le point important que de nombreux analystes ne reconnaissent pas est que dans les économies avancées comme celle des États-Unis, plutôt que de financer l’investissement, les entrées de capitaux étrangers financent indirectement les emprunts à la consommation ou budgétaires nécessaires pour absorber la production excédentaire des pays excédentaires.

C’est ainsi que les États-Unis agissent en tant que “consommateur de dernier recours” dans le monde, ce qui n’est qu’une autre façon de dire qu’ils absorbent en dernier recours l’excès d’épargne mondiale. Mais cela se fait au prix d’une dette croissante et d’un affaiblissement des fondamentaux économiques.

La suite en anglais dans le texte ci dessous a partir de;

Two Scenarios for Capital Inflows

TEXTE IN EXTENSO EN ANGLAIS

Contrary to conventional thinking, net foreign capital inflows do not lower American interest rates (unless they do so by raising U.S. unemployment). A tax on foreign inflows is therefore unlikely to cause American interest rates to rise, and could in fact be an effective way to stem increases in American debt that result from unlimited capital inflows.

by Michael Pettis

Published on July 7, 2025

blog

China Financial Markets

China Financial Markets provides in-depth analysis of one of the world’s largest and most vital economies. Edited by Carnegie Senior Fellow Michael Pettis based in Beijing, China Financial Markets offers monthly insights into income inequality, market structures, and other issues affecting China and other global economies.

A noted expert on China’s economy, Pettis is a professor of finance at Peking University’s Guanghua School of Management, where he specializes in Chinese financial markets.

In a recent commentary on why persistent net capital inflows are a problem for the global economy, Martin Wolf, an economics editor at the Financial Times, suggested that one “obvious” way to resolve global imbalances would be to impose a tax on capital inflows.

While he is almost certainly right—a tax on capital inflows is the most effective way to limit the United States’ role as the main accommodator of global imbalances—many economists oppose such a tax on the grounds that it would raise the cost of capital for American business and raise interest rates for the federal government.

Their view is implicitly based on the premise that U.S. investment suffers from high costs of capital caused by the scarcity of domestic saving. If this premise were correct, it would be logical to argue that net capital inflows from abroad lower American interest rates by relaxing the domestic saving constraint, and that, by extension, any policy that reduces net foreign inflows could raise American interest rates.

But this premise is false. While capital may indeed be scarce in developing countries with high investment needs, it has long stopped being scarce in most advanced economies. In the United States, for example, businesses may refrain from expanding production for many reasons, but rarely is it because they lack access to capital.

This is a crucial difference for at least two reasons. First, if American business investment is not saving-constrained, the American economy will not adjust to net foreign inflows with higher domestic investment, and therefore it must adjust with lower domestic saving.¹ As explained below, the most likely adjustments that reduce domestic saving involve either higher unemployment or higher debt. Foreign inflows, in other words, can force the creation of additional U.S. debt so that the overall increased demand for American debt is matched by an overall increased supply—and in which case it will not drive down U.S. interest rates.

And second, if foreign capital inflows are not driven by the inability of American businesses to fund domestic investment, they must be driven by distortions abroad. That is why not only are these inflows unlikely to lower borrowing costs, they may also, paradoxically, force unwanted changes in the structure of the U.S. economy that create the conditions under which more borrowing becomes necessary just to sustain demand. It is not a coincidence, after all, that among advanced economies, those that receive the most amount of net foreign inflows—such as the United States, the United Kingdom, and Canada—are far more likely to be characterized among their peers by higher debt levels than by lower interest rates.

The Fallacy of Scarcity 

To understand the conventional misdiagnosis, we need to revisit the classical model from which it arises. In a world where domestic saving is insufficient to finance high-return investment opportunities, foreign capital is a welcome addition. In such a model—applicable, for example, to nineteenth-century America and many developing countries today—external inflows support growth by funding much-needed investment in infrastructure, industry, and productivity-enhancing projects that would otherwise go unfunded.

But this is not the reality of the twenty-first-century U.S. economy, or indeed most advanced economies. American corporations, for example, are not unable to invest because they are short of funds. Not only do they have access domestically to the most liquid and flexible debt and equity markets in the world, but even after using trillions of dollars mainly to buy back shares, pay dividends, or acquire competitors, they still sit on nearly $7 trillion in cash and cash equivalents, equal to nearly a quarter of U.S. GDP. They choose not to invest in new factories or technology.

This is not because they are myopic. It is because they find it difficult to sell their existing production capacity, and thus, increasing the amount of net funding available to them will not cause them to increase investment. Ironically, this weakness in domestic demand is amplified by the very capital inflows that are presumed to help because, by pushing up the value of the dollar, these inflows make foreign imports more competitive than products produced domestically. In other words, rather than drive American investment in domestic production facilities, net inflows can actually make this investment less desirable.

It helps to understand the effect of net capital inflows on domestic interest rates by considering the origins of these capital inflows. According to most mainstream economic understanding, foreign savings are “pulled” into the United States as Americans import capital to fund a domestic shortfall of saving. For those who subscribe to this view, the fact that net foreign inflows are by definition equal to the excess of American investment over American saving can only be explained in one way: net inflows must be pulled into the United States by the need of American businesses to fund investment.

But this denies the agency of foreigners. Countries such as China and Germany are not simply passive victims of American domestic imbalances. On the contrary, the fact that China fully controls its banking system while maintaining strict trade controls and even stricter capital controls—unlike the United States—should suggest that it is more likely to be the originator, rather than the absorber, of global saving imbalances. The reality is that countries such as China and Germany have purposely adopted economic models that suppress domestic consumption in favor of a more rapid expansion of manufacturing. As production grows faster than consumption, the saving rates in these countries automatically rise—leading to, in the case of China, the highest saving rate in history.

But supply needs demand. To avoid the rise in unemployment that would result from being forced to cut back on excess output, these countries must export their production surpluses, acquiring foreign assets in payment for exporting more than they import. Therefore, excess domestic saving in these countries is more likely to be “pushed” abroad by internal policies that boost production relative to consumption than be “pulled” abroad by conditions in deficit economies.

The United States is the main recipient of this exported capital. Foreign governments and investors purchase American Treasury bonds, corporate bonds, equities, factories, real estate, and other assets mainly because the United States has the deepest, most liquid, best-governed, and most open financial markets in the world. Surplus countries export their excess saving to the United States primarily for their own domestic reasons, whether or not U.S. businesses need the inflows to increase productive investment.

The key issue here is the direction of causality. The inflows are not responding to a shortage of saving in the United States. Rather, they are the result of saving surpluses generated abroad by policies that repress household consumption.

But this absorption by the United States comes at a cost. A country’s domestic imbalances must always be perfectly consistent with its external imbalances. By changing its external imbalances, capital inflows force adjustments in the U.S. economy—most obviously by raising the value of the dollar, although there are many other forms of adjustment—and these adjustments make U.S. exports less competitive and imports cheaper. As this dynamic erodes America’s tradable goods sector, manufacturing declines and wages stagnate.

If Washington doesn’t want the resulting contraction in American manufacturing to lead to unemployment, consumption must be sustained. With incomes under pressure and competitiveness falling, households and governments are encouraged to borrow and spend so that the decline in the American tradable goods sector is matched by an expansion elsewhere—job creation, in other words, shifts from manufacturing to low-productivity service sectors.² In this way, not only do foreign capital inflows not serve the existing borrowing needs of the economy, they force a shift in the composition of American production—away from manufacturing and towards services—and create conditions under which additional borrowing becomes necessary to prevent trade deficits from forcing up domestic unemployment.

Two Scenarios for Capital Inflows

The important point that many analysts fail to recognize is that in advanced economies like that of the United States, rather than fund investment, foreign capital inflows indirectly fund the consumer or fiscal borrowing that is required to absorb the excess production of surplus countries. This is how the United States acts as the world’s “consumer of last resort,” which is just another way of saying that it is the absorber of last resort of global excess saving. But it does so at the cost of rising debt and weakening economic fundamentals.

To make sense of the macroeconomic implications, it is helpful to consider two stylized scenarios that reflect the potential outcomes of foreign capital inflows:

1. The conventional scenario: In this world, capital is genuinely scarce, especially in countries such as the United States, the United Kingdom, and Canada—all major deficit economies. Domestic investment opportunities in these countries abound, but domestic saving falls short. As foreign capital fills the gap, interest rates fall, productive investment increases, and growth accelerates. This is the nineteenth-century American story, or the story of some rapidly industrializing economies today.
2. The advanced economy scenario: In this more realistic scenario, American firms and households are not starved of capital. Instead, they lack the demand that justifies an expansion of production capacity, in part because capital inflows overvalue the dollar, suppress net exports, and lead to declining industrial output. As U.S. factories move production offshore or find themselves unable to compete against cheaper imports, offering them foreign saving has no effect on their domestic investment plans.

In the first scenario, it is clear that foreign capital inflows, by relieving domestic saving scarcities, can lower interest rates and boost domestic investment. However, this is much less certain in the second scenario. Interest rates may indeed fall in the second scenario—but not due to capital abundance. They may fall because these inflows are the flip side of the weak domestic demand surplus countries must export to deficit countries like the U.S. As Americans pay for that weak demand by closing factories and firing workers, the resulting recession may lead to monetary easing. In this world, lower interest rates are a symptom of malaise, not a sign of healthier investment.

There is another—and more likely—way the second scenario may play out. American policymakers may decide to implement countercyclical policies to reverse the potential contraction in the economy and the accompanying rise in unemployment. They can either encourage more consumer borrowing or expand the fiscal deficit. In these cases, foreign capital inflows are matched by more domestic borrowing designed to sustain domestic demand. So, while the foreign inflows increase demand for U.S. debt, which may push yields down, they also create the need for more debt issuance, which pushes yields up. The overall effect on rates is ambiguous—it depends on whether the United States chooses to respond to the inflows with higher unemployment or with higher debt. If Washington chooses the former, interest rates will decline. If it chooses the latter, they won’t.

Which of these two stylized scenarios applies to the American economy matters greatly, because of the four ways the U.S. economy can adjust to net capital inflows, with each representing a different way in the gap between American investment and American saving can rise. These are:

• Investment rises. In this scenario, American businesses are eager to invest in a wide range of domestic investment opportunities, but lack the capital to do so. By making this capital available, foreigner investors lower American interest rates and encourage more American investment.
• Saving declines. As net foreign inflows push up the value of the dollar, American manufacturers in this scenario become less competitive and are forced to close down and fire workers. With unemployment rising, the American saving rate automatically declines (unemployed workers have negative saving rates). This is the form of adjustment that Joan Robinson warned was the most likely consequence of “beggar-my-neighbour” trade surpluses.
• Saving declines. To prevent unemployment from rising, the financial authorities in this scenario encourage consumer lending in order to increase domestic demand by an amount equal to the domestic demand that has bled abroad through the trade deficit. Rising household debt, of course, is the equivalent of declining household saving, and the resulting higher demand is divided between foreign manufacturers and domestic service providers. In this scenario, foreign inflows finance increased consumer borrowing.
• Saving declines. To prevent unemployment from rising, the government expands its fiscal deficit, in order—again—to increase domestic demand by an amount equal to the domestic demand that has bled abroad through the trade deficit; the resulting higher demand is divided between foreign manufacturers and domestic service providers. In this scenario, foreign inflows finance increased fiscal borrowing.

The bottom line is that contrary to what most American economists assume, the accounting identity that requires net inflows to equal the excess of investment over saving does not necessarily mean that inflows must cause investment to rise. The accounting identity can also be maintained by a decline in saving. Rather than simply assume only one form of adjustment is consistent, economists must analyze the structure of each economy in a more nuanced way and determine which type of adjustment is most likely. 

It is worth noting that in an important recent paper, Atif Mian, Ludwig Straub, and Amir Sufi make a similar argument from a different angle. They show a comparable effect caused not by a net inflow of foreign saving but rather by a rise in the saving of the rich as income inequality deepens. In an economy without strong investment needs, they note that the higher saving of the rich is balanced by “substantial dissaving by the non-rich and dissaving by the government” (in other words, household and fiscal debt). The key point is that if the rise in saving in one sector—whether it is foreign saving or the saving of the rich—isn’t balanced by a rise in investment, it must be balanced by a decline in saving elsewhere in the economy, and this usually happens in the form of higher unemployment, higher household debt, higher fiscal deficits, or some combination of the three.

The Myth of Interest Rate Suppression

An extension of the mainstream model is the claim that polices that push up U.S. saving—via some form of austerity, such as lower fiscal deficits, tighter household credit, or greater income inequality—would reduce the investment-saving gap and shrink the current account deficit. But this again assumes that the United States is the sole causal agent in global imbalances.

What if, instead, global imbalances originate abroad? If surplus countries such as China and Germany are exporting capital to sustain domestic employment and suppress consumption, then cutting U.S. fiscal deficits won’t reduce those capital exports except by forcing a contraction in the U.S. economy, which, by causing American imports to decline, forces a contraction in foreign production and thus foreign saving. In fact, if lower U.S. deficits improve perceptions of creditworthiness, they may actually increase the relative attractiveness of American assets. The result, in that case, would be that surplus countries direct an even greater share of their excess saving to the United States, leading to a stronger dollar and an even larger American trade deficit. This would be the opposite of what most American economists claim. 

Ultimately, there are two starkly different narratives for understanding why foreign capital flows into the United States: 

1. The domestic imbalance view: The United States runs external deficits because Americans save too little. The rest of the world must restructure their economies and force up domestic saving passively to supply capital to meet American demand, and this must be true even when they control their domestic banking systems, their trade accounts, and their capital accounts. In this view, only the United States has agency and raising U.S. saving should automatically reduce excess foreign saving, reduce the U.S. trade deficit, and lower American interest rates.
2. The foreign imbalance view: Some major economies suppress domestic demand to generate trade surpluses and stronger manufacturing sectors while controlling their trade and capital accounts in ways that allow them to externalize the domestic costs. The United States absorbs this surplus by running capital account surpluses and trade deficits. In this view, foreigners have agency, and capital inflows into the United States are a cause, not the consequence, of domestic American imbalances. Raising U.S. saving under this model—for example, by imposing austerity and reducing the fiscal deficit—either reduces demand (and raises unemployment through a corresponding reduction in investment) or forces more debt elsewhere in the U.S. economy.

Mainstream economists generally believe that only the first narrative can be the case, whereas in fact both can be. But these two narratives are not just competing interpretations on interest only to economists—they have radically different policy implications. The first suggests Americans must tighten belts and save more—austerity becomes the preferred solution to yet another economic problem caused by financialization. The second suggests that the trade deficit cannot be resolved by austerity at home. Instead, the United States should reverse its role in accommodating distortions from abroad, perhaps by implementing a kind of Tobin tax that discriminates against foreign inflows that are not directed toward long-term investment. This is the kind of tax Martin Wolf refers to above.

Capital Inflows Do Not Benefit the U.S. Economy

It’s time to retire the muddled notion that foreign capital inflows reduce American interest rates and boost American investment. That was true a century ago or more, when the U.S. economy—like other developing economies today—was capital constrained. But in today’s environment of excess global saving and insufficient demand, capital inflows do not relieve pressure on domestic borrowers—they only lower American interest rates to the extent that they cause American unemployment to rise. Ultimately, they are more likely to create additional debt to counter the unemployment impact of net American demand being diverted to foreign suppliers.

The more general point is that while capital inflows are not inherently bad—especially for rapidly growing developing countries with high investment needs—they are not inherently good. Their impact on the American economy depends on whether they are pulled into the United States by domestic needs or pushed into the United States by foreign needs. When they reflect distortions abroad, rather than productive opportunities at home, they become part of a process by which more-open economies are forced to adjust to the imbalances of less-open economies. They set up a dysfunctional global system that shifts the burden of adjustment onto deficit countries—the beggar-my-neighbor process that Robinson described many decades ago. In the United States’ case, that means more debt, less investment, and slower wage growth. That is why rather than encourage unlimited capital inflows, Washington should consider ways of limiting them. 

Michael Pettis is a senior associate at the Carnegie Endowment for International Peace.

Notes

• ¹Net foreign inflows are by definition equal to the excess of domestic investment over domestic saving. For instance, if the United States receives $100 in net foreign inflows, total American investment will be $100 greater than total American saving. But what confuses many economists is that this accounting identity does not imply any particular way in which the identity is maintained. A $100 increase in the gap between investment can just as well be associated with a rise in investment as with a fall in saving, but because most economists implicitly assume that U.S. investment is constrained by scarce U.S. saving, they also assume that foreign inflows must be associated with a rise in domestic investment and not a fall in U.S. saving. As argued in the rest of this article, this is simply not true.
• ²Many American economists assert that there is no point in protecting manufacturing employment because technological innovation has caused manufacturing employment to decline everywhere. But this confuses the decline in American manufacturing employment caused by rising productivity (a good thing) with the decline in American manufacturing employment caused by industrial policies abroad that expand the manufacturing share of surplus countries at the expense of deficit countries (a bad thing). The former is exactly why most countries want to benefit from large manufacturing sectors, while the latter means that an increasing share of this benefit is shifted abroad. See Michael Pettis, “Are Domestic Manufacturing Jobs Worth Protecting?,” Wall Street Journal, April 30, 2025, https://www.wsj.com/opinion/are-factory-jobs-worth-protecting-economy-manufacturing-production-fa969bea.

Michael Pettis

Nonresident Senior Fellow, Carnegie China

EconomyTradeUnited StatesCarnegie does not take institutional positions on public policy issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of Carnegie, its staff, or its trustees.