Saving, Investment, and Imbalances Scarcity has defined most of humanity’s existence. Until relatively recently, simply securing enough food to prevent starvation was a major challenge. During the long era of scarcity, people had to choose whether to use finite resources to invest in productive assets or to satisfy immediate needs (consumption). Historically, consumption won. Growing populations invariably consumed any additional output, which limited wealth creation and capped living standards. Back then, suppressing consumption (saving more) was necessary to generate a surplus of production above consumption to fund worthwhile investments. When resources are abundant, however, trying to save by consuming less is wasteful and counterproductive. People who could be working are left idle even as desires remain unfulfilled. Fields lie fallow as the hungry starve. Factories and machines deteriorate for lack of use. Rather than generating a surplus that can be invested, cutting consumption simply leads to lower production. Moreover, the resulting excess capacity discourages new investment and ultimately leads to lower living standards. Globally, all economic output is either consumed or used to develop productive assets. For the world as a whole, saving and investment are equal by definition. In most countries, however, saving and investment are not equal. Some places produce more than they use domestically, while other countries produce less than they need. These differences are reconciled through trade: excess output is exported to places where domestic demand (consumption plus investment) is greater than domestic production (GDP). Surpluses and deficits are the result. This can be represented with the following set of simple equations: Global demand = Global production Demand = Consumption + Investment Production = Consumption + Savings Domestic demand = GDP + Imports – Exports Exports – Imports = Domestic saving – Domestic investment Trade imbalances allow gluts in one society to compensate for shortages in another. In the right circumstances, these surpluses and deficits make everyone better off compared to a world full of closed economies. Countries with few attractive domestic investment opportunities, perhaps because of their demographics and their position at the forefront of technological development, ought to be net exporters. Without export markets, those countries would be stuck in permanent slumps caused by the imbalance between their abundant productive capacity and their weak domestic demand. The natural recipients of those exports are countries bereft of needed capital goods and infrastructure. Without access to foreign production, investment in the deficit countries would have to compete for limited resources with domestic consumption. It is not a coincidence that two of the worst famines in modern history—the Soviet Union in 1929–33 and China in 1958–62—were perpetrated by authoritarian regimes committed to rapid industrialization while cut off from the rest of the world. At other times, however, trade imbalances can make people worse off. Instead of relieving shortages, imports simply crowd out domestic production. This has been the defining problem of the past few decades: people in certain countries are spending too little and saving too much. This is not because their households are especially thrifty or because their governments are unusually prudent. It is not even because their businesses are rationally responding to the dearth of attractive opportunities. Rather, it is because of choices made by elites within those countries that transfer wealth and income away from people who would spend more on goods and services, such as workers and pensioners, to those, such as the rich, who would instead use extra income to accumulate additional financial assets. This imposes an untenable choice on the rest of the world: absorb the glut through additional spending (saving less) or endure a slump caused by insufficient global demand. Two Development Models: High Savings versus High Wages Societies raise living standards by putting more people to work, by making workers more efficient, and by expanding productive capacity. Investment is therefore essential for development. There are two basic ways to pay for this investment when domestic production is already running at maximum capacity: transfer resources from domestic consumers (the high-savings model) or transfer resources from the rest of the world by raising imports relative to exports (the high-wage model). In other words: Investment = GDP + Imports – Consumption – Exports While most countries have relied on some combination of the two development strategies to pay for their industrialization, each approach has distinct implications for domestic politics and for international trade. High savings lead to trade surpluses because they raise production relative to domestic demand, while high wages tend to produce trade deficits because they raise domestic demand above existing productive capacity in an effort to attract foreign investment. The high-savings model forces ordinary people to spend less so that the government and businesses can spend more. This in itself is not novel: elites the world over have repressed peasants and appropriated their agricultural surpluses for thousands of years. The innovation of the high- savings development strategy is that consumption is squeezed to pay for productive investment in infrastructure and capital goods, rather than to pay for elaborate monuments and the military. Done correctly, this investment raises ordinary people’s living standards even as their share of economic output declines. The high-savings model is therefore the original version of trickle-down growth. Raising the national saving rate is usually regressive and typically requires an authoritarian political culture or a high degree of centralization to make it work. This was pioneered in eighteenth-century Britain. First, aristocratic landowners used the power of the state to evict subsistence farmers and consolidate their holdings into enclosed estates. That boosted agricultural profits at the expense of the peasants, who were displaced from the countryside and forced into the cities. Their rising numbers limited their bargaining power with urban employers, which kept real wages from rising despite rising output per hour. That in turn boosted the profits of manufacturers, which reinvested those profits in developing additional capacity. In 1740, just 4 percent of British production was saved rather than consumed domestically. By the 1820s, the national saving rate had grown to 14 percent and Britain had become an industrial superpower that was exporting its excess manufacturing output to the rest of the world, especially its imperial colonies and the rapidly growing United States. Forced saving enabled productive investment that generated additional output that was then used to create additional investment. Saving by itself did not create wealth, but it was instrumental to the process of wealth creation because it could be used to fund investment. Britain did not pay for the first wave of its industrialization exclusively off the backs of its landless peasants, however. It also developed using elements of the high-wage model. While they became increasingly underpaid relative to the value of what they produced during the Industrial Revolution, British workers nevertheless continued to command higher pay than workers in much of the rest of Europe. Their high productivity and the favorable business climate pulled in capital from abroad, which allowed investment spending to consistently exceed national saving until after the end of the Napoleonic Wars. The difference was covered by the Dutch, who are estimated to have paid for about a third of Britain’s total investment in the eighteenth century. The Dutch were willing to do this because British policies—including protective tariffs and what nowadays would be called intellectual property theft—had made investments in Britain more attractive than investments in the Netherlands and because at the time the Netherlands had a more mature economy with lower investment needs.1 Like Britain, the United States used elements of both development strategies when it industrialized in the nineteenth century. Before the Civil War, the South used an exceptionally cruel form of agrarian feudalism to produce copious volumes of cotton, tobacco, and other cash crops. Southern agricultural output was an essential input for British manufacturers and generated the bulk of America’s export earnings. The South’s social system —extreme wealth and income inequality reinforced by the brutal subjugation of the enslaved labor force—also crushed consumption. Despite generating high saving rates, the planters had little interest in economic development. Instead of buying capital goods, they spent their surpluses buying additional enslaved workers and land. Southerners nevertheless contributed to America’s industrialization because they were trapped behind high tariff barriers and were therefore forced customers of goods manufactured in the North. Southern agricultural exports generated by slave labor therefore helped pay, indirectly, for northern imports of advanced European technologies and machines that were then used to boost the North’s manufacturing capacity. 2 Far more important to America’s economic development, however, was the North’s use of the high-wage model to fund its industrialization. Outside the slave states, abundant land, liberal institutions, and Yankee ingenuity meant that American workers consistently earned the highest pay in the world and enjoyed rapid increases in living standards. At the same time, high birth rates and high immigration made the U.S. domestic market the most impressive growth story in human history. The U.S. population grew from 4 million people at the time of the 1790 census to 40 million by the 1870 census and to roughly 80 million people by 1900. Protective tariffs biased that market in favor of American goods over imports. The combined result was that investments in America’s economic development were incredibly attractive for European savers, especially the British. Foreign saving was therefore able to supplement domestic saving to pay for American imports of capital goods and lift U.S. investment without depressing U.S. consumption. Until the end of the nineteenth century, the United States consistently imported more goods than it exported even as its manufacturing output soared. Like the Anglo-Dutch economic relationship in the eighteenth century, the Anglo-American economic relationship in the nineteenth was based on the transfer of excess output from a smaller but more advanced society to a larger one undergoing rapid industrialization. The same forces attracted millions of immigrants to come to America from Europe. After they arrived, many of these migrants started businesses using advanced technologies and skills from their home countries. Economists estimate that the value of this human capital inflow was worth several times the value of conventional foreign investment throughout the nineteenth century. The North’s victory in the Civil War and the subsequent westward expansion of America’s land borders solidified America’s commitment to the high-wage model and increased its industrial potential even further. Rising inequality in the North in the decades after the end of the Civil War eventually depressed consumption relative to production, which meant that more investment could be funded internally. By the beginning of the twentieth century, America had become a net exporter. 3 America’s achievements attracted admirers and imitators, particularly in Germany and Japan. Friedrich List, one of the first theorists of the American System, had explicitly argued that America’s internally vibrant but externally protected market was a model for what he hoped would be the unified German economy. A decade later, Erasmus Peshine Smith published his Manual of Political Economy (1853), which was perhaps the most important theoretical defense of America’s developmental state. Like many in the antebellum United States, Smith saw abolitionism, protectionism, and mass immigration as part of a common program opposed to free trade and slavery. In his view, America’s high wages—a product of high tariffs, abundant land, and, outside the South, human liberty—caused America’s exceptional productivity. Expensive labor forced businesses to become more efficient and to invest in capital equipment. At the same time, rapid population growth expanded the domestic market and rewarded additional business investment.4 Smith’s arguments found a ready audience in Japan. The shogunate had been unprepared when American naval ships arrived at Tokyo Bay in 1853 and was forced to accept disadvantageous commercial treaties with the West. Those treaties had prevented Japan from levying tariffs on imports greater than 5 percent. Dissatisfaction with the ensuing economic dislocation and the regime’s overall handling of foreigners led to an elite revolt premised on the restoration of the Meiji emperor as head of state in 1868. In 1871 Smith was invited to Tokyo to advise the new government on international law. He went to Japan and quickly became an influential adviser to the Japanese government. Although Japan was unable to adjust its tariffs until 1899, it nevertheless adopted several recognizable elements of the American System. First, the government actively invested in internal improvements, especially roads and railroads. Second, echoing Alexander Hamilton’s advice a century earlier, the government subsidized “model factories” for ships and military kit. The economic historian Kenichi Ohno observes that these factories “had strong demonstration effects on emerging Japanese entrepreneurs” and “also trained a large number of Japanese engineers who later worked in or established other factories.” All of this government spending was paid for with new taxes on land and forced loans from rich Japanese—higher saving to support additional investment. The explicit goal was to develop an indigenous productive capacity that would eventually displace imports. Until then, however, Japan also relied on foreigners for its modernization. As in the United States, the immigration of specialists, including Smith, transferred human capital from the West, while Japanese students were paid to go abroad and learn from Western universities. Japan also had a large trade deficit, importing more than it exported to the rest of the world. In addition to raw materials such as cotton, almost all advanced machinery, including railroad engines and electric generators, had to be imported.5 The most extreme iteration of the high-savings model was the Soviet Union under Joseph Stalin. The Bolshevik coup of 1917 and the subsequent repudiation of Russia’s foreign debts had turned the USSR into a pariah state unable to access the rest of the world’s savings. Moreover, the Soviets were opposed to opening their country to foreign investors on ideological grounds. The lack of capital imposed a severe constraint on a largely agricultural society: foreign goods could be obtained only by barter or theft. Yet Stalin was committed to rapid industrialization to secure “the independence of the socialist economy from the capitalist encirclement,” as he put it. Like George Washington 150 years earlier, Stalin believed that the development of an indigenous manufacturing capacity was a national security imperative. Unlike the Americans, Stalin chose to develop that capacity without recourse to trade deficits funded by investments from abroad. The Soviets would therefore have to sell exports to pay for their substantial imports of advanced technologies and capital goods. In the 1920s and 1930s, those exports were mostly base metals, gold, and grain. Mining metal in sufficient quantities without machines required many workers, which were supplied at low cost by the gulag system of forced labor camps for the regime’s political enemies. Securing a surplus from the peasants was more challenging. While the Bolsheviks were fighting to control the cities after 1917, the peasantry had fought and won a separate revolution against the old landlords. The peasants’ victory meant that they could keep the surplus they generated from working land they owned. In the early 1920s, the Bolsheviks felt that they had no choice but to accommodate these new agrarian capitalists to preserve the success of their proletarian revolution in the cities. Vladimir Lenin likened the resulting New Economic Policy to the 1918 Treaty of Brest-Litovsk with imperial Germany, which had saved the Bolshevik regime at the expense of Russia’s western frontier. Both were meant to be temporary expedients. By the end of the 1920s, however, Stalin had concluded that the correlation of forces had changed and violently enslaved the Soviet peasantry through a process called collectivization. Even though agricultural output was lower under state ownership, the government had total control of that output and was able to extract a substantial surplus by squeezing the peasants. Grain exports soared in the early 1930s even as tens of millions starved to death. By the end of the 1930s, the typical Soviet subject got fewer calories from grain than in the prerevolutionary period, but far more calories from vodka. This was a humanitarian catastrophe, but it did enable imports of industrial equipment from the Soviets’ main trade partner: Nazi Germany. Ideological differences were insignificant compared to the complementary economic needs of the two societies. Both countries were international pariahs, which cut them off from global markets, and both were willing to circumvent the limits imposed by the West by trading with each other. Like the nineteenth-century relationship between imperial Germany and tsarist Russia, resource-poor Germany in the 1930s traded advanced manufactures in exchange for the raw materials it needed for rearmament. The Germans were willing to supply their technology to the hated communists because they thought that the Soviets would not be able to modernize quickly enough to become a military threat. The Soviets had the opposite view (and believed that the Nazis would focus on fighting the West), which made them happy to provide crucial supplies to the Nazis to pay for their own industrial transformation. By the eve of World War II, Stalin had achieved his strategic objective of an indigenous industrial base. Had the Soviet Union remained a primitive agrarian society, it could not have defeated the Germans, who had outmatched Russia in World War I. But although Stalin had transformed his domain into a formidable military power, this had come at enormous (selfimposed) cost. Much of the Soviet population had been conscripted into providing forced labor. Tens of millions had died of starvation to supply agricultural exports to pay for Western manufactures. The Soviets had modern tanks and aircraft, but soldiers lacked such basic goods as boots and radios. Consumer goods were essentially unavailable. The Soviet Union’s experience demonstrated both the triumph and the limitations of the highsavings growth model.6 Japan developed a more humane variant of the high-savings model after World War II. Workers, businesses, and the government agreed on a social contract that generated decades of rapid growth and economic convergence with the West. Workers agreed not to strike and to limit their requests for wage increases. Businesses agreed to reinvest profits aggressively in domestic capacity and technological improvements. The government agreed to support companies with cheap loans offered at the expense of regular savers, as well as regulatory restrictions on imports to protect the domestic market. Within Japan, a handful of conglomerates dominated the economy in an oligopoly at the expense of consumers. Abroad, those same companies competed ferociously for market share against one another and against producers from the United States and Europe, which forced them to become efficient and innovative. Japanese workers, consumers, and retirees all subsidized industrial development by overpaying for goods and services, by taking home a lower share of national output than their counterparts in the West, and by using a financial system designed to transfer purchasing power from households to businesses. Japanese companies returned the favor by upgrading the country’s manufacturing base, passing along productivity gains to workers, and refraining from excessive executive pay, while the government invested in top-tier infrastructure. Moreover, despite its rapid growth, Japan’s exports consistently exceeded its imports from the rest of the world. The Japanese development model created problems as the country converged to Western living standards. When a society has an abundance of educated and industrious workers but lacks sufficient physical capital and technology, there are many obvious worthwhile projects to invest in. Transferring purchasing power from workers to businesses and the state can therefore accelerate national development. Unfortunately, Japan’s institutional biases in favor of investment over consumption created pressure to keep investing even after the best projects had been completed. By the early 1980s, the mechanisms developed after the war to constrain household spending and subsidize corporations had outlived their usefulness. The incremental gains from each additional investment steadily fell while systematic constraints imposed on household spending exacerbated the decline in investment returns. Japanese society eventually adjusted in the 1990s and 2000s, but in an unnecessarily painful way: business investment plunged, unemployment rose, and the household saving rate dropped to zero to compensate for weak wage growth. This could have been avoided if Japan had abandoned its development model earlier. Many postcolonial countries tried to modernize according to either the Soviet or Japanese variants of the high-savings model. After its division in 1948, Korea experienced both versions. Until the 1970s, the communist North appeared to be more successful than the more market-oriented South, which had copied many features of the Japanese model. As time went on, however, the Soviet system’s main advantage—the rapid development of physical capital—was outweighed by the severe disadvantages of low productivity, an inability to innovate, and systemic malinvestment. By contrast, South Korea’s conglomerates, consciously echoing the experience of their Japanese models, had to upgrade their technology to compete in global markets. In 1970, living standards in South Korea were just a tenth of what they were in the United States. After Korea’s financial crisis in 1997, its living standards were about half of U.S. levels. As of 2016, however, South Korean living standards had reached almost 70 percent of those in America—comparable to Japan, New Zealand, and France.7 In the nineteenth century, America’s high-wage growth model was complementary to Britain’s surpluses. Argentina, Australia, and Canada were also eager recipients of British savings. The world as a whole prospered from this arrangement because, with the exception of Argentina, the societies receiving resources from abroad were generally the ones able to make the best use of them. At the same time, excess production was coming from the richest and most advanced societies. By the end of the twentieth century, things had changed. Poor countries were often subsidizing the consumption of the rich—at the expense of both. Excess Savings and the Great Glut Scarcity stopped being a serious problem in the rich world sometime near the last quarter of the twentieth century. Making things has become easier and cheaper than ever before. Shortages have been replaced with gluts. The age-old tradeoff between consuming more today and producing more tomorrow is gone. Investment is now constrained by insufficient consumption, rather than by the old competition for resources. The modern condition is therefore defined by the perverse coincidence of abundant idle resources and unmet material needs. This has had profound consequences for the relations among savings, investment, and trade. Fig. 3.1 The great glut (U.S. manufacturing capacity utilization rate). Sources: Federal Reserve Board; Matthew Klein’s calculations There are two basic inputs to production: labor and capital. Both have been in abundance for decades. Unemployment rates across the rich world have been systematically higher since the 1970s than they were before. The situation looks even more extreme after accounting for the rise of part-time employment, which depresses the number of hours worked per job, and the steadily rising share of people of working age who neither work nor go to school. If there were more work to be done, people could easily be found to fill the jobs. The problem has been an absence of demand for their labor. 8 A similar story can be told about the supply of productive capital. The Federal Reserve has tracked the productive capacity of the U.S. manufacturing sector and its relation to output since 1948. From then through the end of 1979, American manufacturers used 83 percent of their capacity to produce goods, on average. From the beginning of 1980 through the end of 1999, capacity utilization averaged 80 percent. Since the start of 2000, manufacturing capacity utilization has averaged just 75 percent thanks to the combination of excess capacity built in the 1990s and limited growth in domestic production since then. Manufacturing capacity has shrunk slightly since 2008, but output has shrunk even more.9 Corporate investment behavior is also revealing. Traditionally, the business sector is supposed to spend more expanding productive capacity than it generates in cash flow, with the difference covered by household savings. The additional investments are supposed to lead to higher cash flow that can help cover the cost of further expansion while also repaying savers. Although some companies, or even entire industries, may lack growth opportunities and opt to distribute their profits to shareholders rather than retain earnings to reinvest, the business sector as a whole is supposed to require others’ savings to grow. In the past few decades, however, this mechanism has broken down. The business sectors of many countries now spend less than they generate in cash flow. The resulting corporate surpluses are either distributed to shareholders, as in the United States, or retained by the companies, as in Germany, Japan, and South Korea. Regardless, the implication is that there are far fewer worthwhile investment opportunities in the rich world than in the past. Those that remain are mostly infrastructure and housing, which are hampered by political constraints rather than by an excessive cost of capital.10 One consequence has been a steady decline in the prices of manufactured goods. Capital equipment prices in the United States have dropped by 30 percent since 1991 in absolute terms. The prices of durable consumption goods—mainly cars, appliances, and furniture—have dropped by more than 36 percent since the peak in 1995. The prices of clothing and footwear are lower now than in the mid-1980s. Since 1990, most inflation in the United States has come from higher prices for health care (including prescription drugs), housing, and education—all sectors where the government tightly regulates supply and heavily subsidizes demand. In the rest of the economy, prices have been flat. Relative to the average American’s disposable income, the cost of buying manufactured goods has collapsed by more than 90 percent since the late 1940s, with most of the decline occurring since the mid-1980s. Similar conditions can be found in Europe and Japan.11 Financial market data are also consistent with the great glut. Asset prices, like all other prices, are determined by the balance between supply and demand. Put another way, the valuations of stocks and bonds should reflect both the desire by businesses and the government to raise money to fund new investments (supply of assets) and the willingness of households to consume less today in exchange for being able to consume more in the future (demand for assets). Low asset prices (high future returns for savers) represent the cost that companies must pay to attract funding when resources are scarce. When productive capacity is abundant, however, the expected return on additional investment should be low. Valuations would be high, and yields would be low. Fig. 3.2 The prices of manufactured goods have plunged relative to disposable incomes (manufactured goods price index divided by average disposable personal income, January 1947 = 100). Sources: Bureau of Economic Analysis; Matthew Klein’s calculations As it happens, inflation-adjusted long-term interest rates have fallen steadily across the world since the abnormal increase associated with the disinflation of the early 1980s under the Federal Reserve leadership of Paul Volcker. For decades, real borrowing costs have been below long-term forecasts of real economic growth and remain around zero. Corporate valuations have steadily increased, especially in the rich world, while credit spreads have been exceptionally tight outside of the financial crisis period. By some measures, financial conditions in the rich countries over the past few decades are almost as loose as they have ever been. Commodity prices have steadily declined relative to average incomes. Private equity firms are burdened with trillions of dollars of funding that they have not been able to deploy. Covenants in bonds and loans have steadily loosened in favor of borrowers. The ability to make productive investments may once have been constrained by access to material resources, but that era ended decades ago. The great glut has been so damaging in part because standard economics has such difficulty describing it. The textbook definition of “saving” is “not consuming”—think of the biblical story of Joseph and the Pharaoh preparing for the seven lean years by hoarding surplus grain during the seven fat years. Since all output is either consumed or used to develop assets, saving necessarily equals “investment.” Although this identity is true by definition, it can nevertheless lead to serious misconceptions. The biggest mistake is to think that higher saving causes additional investment. Yes, restricting consumption frees up workers, machines, and material inputs. In times of scarcity, saving more is therefore a prerequisite for investing more. And when there are many worthwhile investment opportunities, idled resources can be redeployed relatively quickly. But there is nothing automatic about this process. Rather, it is contingent on specific economic conditions. When those conditions do not apply, higher saving simply means lower living standards. In many ways, it is better to think of the textbook formulation in reverse: more investment leads to more saving. By definition, worthwhile projects enable societies to produce additional output relative to labor and material inputs. Most of that output will get consumed, lifting overall living standards. But as long as some of the extra production is used to pay for further investment, total savings will have increased. Savings can grow even as consumption rises relative to production. By extension, total investment can rise even when the saving rate drops without generating trade deficits. Efficiency improvements—generating more output from the same set of inputs—allow societies to keep investing in additional capacity while raising the share of production that is consumed. More important, the reverse is also true. Trying to promote additional investment through higher saving is often counterproductive. The simplest way to raise the saving rate is to spend less on consumer goods and services. Unless investment immediately rises to offset the decline in consumption, however, the result is less total production and lower total saving, which ultimately discourages new investment.12 Changes in the income distribution affect all of these variables and therefore have important economic consequences. While most people spend close to everything they earn on goods and services, the rich do not. There is only so much a person can consume, no matter how expensive his tastes. Give most people an extra dollar, and sooner or later it will get spent buying something that provides jobs and incomes to others. Give that same extra dollar to a rich person, however, and it will probably be used to accumulate additional assets. For the world as a whole, rising inequality means the value of those assets is necessarily contingent on continued increases in spending by people who have progressively lower shares of national income. The only way to make this work is with rising debt. The economists Michael Kumhof, Romain Rancière, and Pablo Winant found an almost perfect relation between rising income concentration in the United States in the 1920s and rising U.S. household indebtedness in the years before the Great Depression.13 Marriner Eccles, Franklin Roosevelt’s Federal Reserve chairman, understood that this was why the American economy was so fragile in the 1920s despite the apparent burst of postwar prosperity. To Eccles, the root problem was the shift in the U.S. distribution of income from the masses to the elites. “By taking purchasing power out of the hands of mass consumers,” he wrote in his retrospective account, “the savers denied to themselves the kind of effective demand for their products that would justify a reinvestment of their capital accumulations in new plants.” It is obvious that consumption cannot grow without investments to produce more of what people want. It is less obvious, but just as important, that those investments require rising consumption to be profitable. Building truck factories or apartment complexes or power plants is not “investing” in any meaningful sense if nobody ends up buying more trucks, living in the apartments, or needing the extra electricity. It is just waste.14 This explains how the world can be afflicted by a savings glut without having a high saving rate. The level of the saving rate by itself is meaningless. What matters is the amount of unconsumed output relative to the supply of worthwhile investment opportunities. Saving is excessive when real resources are diverted from the satisfaction of immediate human needs to develop wasteful investments. In many cases, of course, distinguishing “worthwhile” from “wasteful” is possible only in retrospect. But excess saving necessarily leads to wasteful investment because it encourages overbuilding and because suppressing consumption reduces the viability of otherwise worthwhile projects. When a society—call it “Scroogeville”—increases its saving rate, it is, by definition, consuming less relative to what it produces. Because the world’s consumption and investment together must equal global output, some combination of the following three outcomes must occur: • The investment rate in Scroogeville rises. • The investment rate in the rest of the world rises. • The saving rate in the rest of the world falls. These three possibilities are equivalent to the following four scenarios: • Productive investment rises globally. • Wasteful investment increases globally. • Consumption outside Scroogeville rises. • Production outside Scroogeville falls. Two of those outcomes—higher wasteful investment and lower production outside Scroogeville—are unequivocally bad. Higher consumption outside Scroogeville could be good but could also be dangerous depending on how that additional spending is financed. While higher productive investment would be unequivocally good, it is also the least likely outcome in today’s developed world. There is simply no evidence that the world’s investment needs are unmet because of excessive capital costs. Rather, investment has been restrained by the lack of attractive opportunities—itself caused by weak global demand—and by irrational political constraints. As it happens, the International Monetary Fund (IMF) reports that the world’s saving and investment rates have been stable as a share of global output since 1980. Dramatically higher saving in some places—most notably China—has been offset by much lower saving elsewhere. One implication is that, globally, there was no shortage of funding for worthwhile investment projects. Had there been a shortage, large increases in savings in one part of the global economy would have been matched by increases in global investment. That did not happen. Instead, places that squeezed consumption relative to production simply forced production to fall relative to consumption elsewhere. This had significant consequences for trade.15 China’s investment rate has soared since the 1980s, for example, but not by enough to offset the even larger relative decline in consumption (rise in saving). Until 2008, the result was a surge in production relative to domestic demand, with the rest of the world forced to absorb the difference. China boomed, but the distribution of spending within the country created serious distortions for the rest of the world. Since 2008, China’s surplus has shrunk because the investment share of GDP has increased. The consumption share, however, has not, except marginally in the past few years. The result is that China’s saving rate is higher now than when it had trade surpluses worth 10 percent of its productive output. By contrast, Germany’s surplus is a function of weak growth in domestic demand. Domestic consumption and investment have grown sluggishly for almost thirty years. Total production has also been weak since reunification but has grown marginally faster than domestic demand. As in China, the difference has come from net foreign spending on German exports. Germany’s saving rate and its investment rate are not unusual when viewed separately. The gap between them, however, is almost unprecedented for a country of its size. In the United States, the situation has been the reverse: while domestic demand has been weak, as in Germany, output growth has been even worse. Foreign saving crowded out U.S. domestic production in the form of a trade deficit. By contrast, Spain’s massive deficit was caused by an investment boom. There were no meaningful changes to its production or its consumption habits. The subsequent shift into surplus was caused more by the investment bust than by any change in the household saving rate. Greece also had an investment boom and bust, but unlike Spain, this was amplified by a consumption boom and bust.16 By definition, the experiences of some of these countries were connected to the experiences of others. There are only three possible explanations that link surpluses in some countries with deficits in others: • Changes within China, Germany, and other surplus countries caused their domestic demand to fall relative to their domestic production, which forced the rest of the world to spend more relative to production through a combination of falling output, higher investment, and rising consumption. • Changes within the United States, Spain, Greece, and other deficit countries caused their output to fall relative to their spending, which forced people in the rest of the world to produce more than they needed to satisfy their domestic needs. • Hermes, the Greek god of commerce—or perhaps his Indian counterpart, Lakshmi, who controls wealth—has been extraordinarily busy spending all of his time managing trade, investment, and savings, country by country, with such precision that at every single point in time, by an astonishing coincidence, all the saving rates and all the investment rates in hundreds of more-or-less autonomous entities in the world balance out perfectly. A close reading of the events of the past thirty years shows that the first option is the best explanation. Political and social changes within the surplus countries transferred purchasing power from workers, who spend most of their income on goods and services, to elites, who prefer to accumulate financial assets. This has occurred in a variety of ways in different countries. The Chinese hukou (household registration) system deprives hundreds of millions of urban workers access to government benefits for which they pay tax, for example, while German companies refuse to invest domestically despite rising profitability and regular losses on their investments abroad. These choices mechanically suppress consumption and investment relative to production, which in turn forces the rest of the world to spend more than they produce. The Balance of Payments Savings travel internationally through financial markets. Countries with trade surpluses do not donate their excess production to the rest of the world but sell it in exchange for claims on future production. Those claims are serviced either out of income earned from trade surpluses or by issuing even more claims. The balance of payments tracks these transactions. The current account looks at trade flows and the cost of financing trade imbalances, while the financial account measures the purchases and sales of assets across borders, including the change in central bank reserves in countries that intervene in the currency markets. The net amount of money coming in or going out via the financial account must equal the net amount of money moving across a country’s borders as recorded in the current account, although measurement difficulties sometimes create differences between the two figures. A country’s current account balance is simply the sum of the individual saving and spending decisions of the residents of that country. There are two kinds of people in the world: those who spend more than they earn, and those who spend less. Since all income ultimately comes from the spending of others, these individual differences always balance out at the global level. There is therefore no way to save without someone else dissaving. Those who spend less than they earn have a surplus that has to go somewhere. It could stay in a bank account, it could be used to buy financial assets such as stocks and bonds, or it could get put into physical assets such as real estate, art, and precious metals. These asset purchases would be impossible without willing sellers. While there may be some transactions in between, the ultimate asset-sellers use the proceeds to spend more than they earn. These could be people who own existing assets and want to transform some of their past saving into current spending, such as retirees gradually liquidating their holdings, or they could be people who want to raise money to buy things they cannot afford by issuing freshly minted assets. Households take out mortgages to buy new homes, businesses issue shares to finance capital investment, and governments sell bonds at auctions, for example. The savers and the dissavers necessarily go together. If people in a country collectively spend more than they earn, then the country as a whole has a current account deficit. In other words, the amount of money coming in from the rest of the world as income—exports, foreign investment earnings, remittances, and foreign aid—is less than the money going out in the form of imports, dividends and interest paid to foreigners, and transfers. By contrast, if a country’s households, businesses, and government collectively spend less than they take in as income, then the country has a current account surplus. In that case, the combination of exports, foreign investment earnings, and transfer receipts brings in more money than goes out on imports, income paid to foreigners, and remittances. The flip side of all this is the financial account. Any country where spending collectively exceeds income must cover the difference by raising money from selling assets. A country with a current account deficit by definition must have a financial account surplus: the total amount of money coming in from foreigners buying domestic assets must be greater than the total amount of money going out as locals buy foreign assets. Conversely, countries where people collectively spend less than they earn must be investing their current account surpluses abroad; more money is going out to buy foreign assets than is coming in from the rest of the world to buy local assets.17 The following equations may help clarify these relations: Current account = Financial account + Statistical discrepancy Current account = Household saving + Corporate profits + Taxes – (Household investment + Business investment + Government spending) Financial account = Foreigners buying local assets – Locals buying foreign assets Financial account = Private sector financial account + Change in central bank reserves Large surpluses or deficits are not inherently good or bad. “Good imbalances” allow savers from richer surplus countries to earn healthy returns by financing development and rising living standards in deficit countries. This is what the United States did for much of the nineteenth century, when it imported mainly British capital to boost domestic investment to levels much higher than it could have otherwise achieved without squeezing American workers. More recently, except for a brief period in the late 1980s, South Korea consistently imported more than it exported in the decades from independence in 1948 until the Asian Financial Crisis in 1997. Korea is also one of the few countries to transition successfully from poor to rich. Norway, which was once one of the poorest countries in Western Europe, imported massive amounts of foreign savings in the form of large current account deficits in the 1970s to pay for the development of its offshore oil and natural gas fields. Once those fields began producing, Norwegians were able to repay their obligations and eventually amass a large stock of foreign assets purchased from their hydrocarbon profits. Had Norwegians been constrained in their ability to spend more than they earned, those resources never would have been developed. Both Norway and the world as a whole would have been poorer. 18 At the same time, surpluses can be bad. Savings have to go somewhere, but there is no guarantee that they will go into profitable investments. Germans, who have been such avid exporters of financial capital over the past two decades, are almost uniquely bad at investing abroad. Since the start of 1999, the German private sector collectively spent a little over €5.1 trillion acquiring assets in other countries. Yet over the same period, the amount of these foreign assets grew by only €4.8 trillion. The difference represents a valuation loss of 7 percent across nearly two decades thanks to such holdings as American subprime mortgages and Greek sovereign debt. Even after accounting for dividends and interest income, Germany’s foreign investments have done worse than the foreign investments owned by residents of almost every other rich country. A 2019 study by Franziska Hünnekes, Moritz Schularick, and Christoph Trebesch concluded that “Germany could have become about 2 to 3 trillion Euros richer [between 2009 and 2017] had its returns in global markets corresponded to those earned by Norway or Canada, respectively.” Strikingly, Germans’ poor returns have been caused almost entirely by their remarkable inability to pick the right stocks and bonds, rather than broader differences in asset allocation compared to savers in other countries. This abysmal performance looks even worse compared to what could have been achieved by buying German assets. The same study concluded that “domestic returns were significantly higher than the return earned abroad.” Between 1999 and 2017, German assets have returned about 2.4 percentage points more each year on average than Germany’s foreign assets. Since 2009, the gap has widened to a staggering 5 percentage points. For perspective, while €1 million invested in a representative collection of German assets in 1999 would have generated roughly €2 million of interest, dividends, and capital gains by 2017, the same amount invested in a representative sample of Germany’s actual foreign assets would have returned less than €1 million. Another way to see this is to compare how German residents have fared on their foreign investments compared to how foreigners have fared on their investments in Germany. From the start of 1999 through the end of 2018, Germans invested €2.6 trillion more abroad than non-Germans have invested in Germany. Yet Germany’s net foreign asset position grew by only €1.9 trillion, which implies a net loss of 29 percent. Germans would have been far better off had they invested more at home or spent more on goods and services they actually wanted.19 Two things determine whether an imbalance is healthy or dangerous: how the money is raised and how the money is spent. Ideally, richer countries make direct equity investments in poorer ones with lots of potential, as in the case of South Korea. In the past few decades, however, surplus countries have been lenders, rather than shareholders, while deficit countries have often been mature economies that lack useful projects in need of outside funding. The result has been debt booms wasted on boondoggles. People in both the surplus countries and the deficit countries have lost out from this exchange. Although some of this can be blamed on insufficient regulation in the countries on the receiving end of these financial flows, the bigger problem is that the flows are too large and go to the wrong places. Popular antipathy to trade therefore stems from the failure of international capital to go where it is needed in forms that are useful. This in turn can be blamed on policies in surplus countries that steadily transfer income and wealth from workers to elites. Because the rich have higher saving rates, the effect has been to shift purchasing power away from goods and services to financial assets. Not finding enough additional financial assets to buy in their home country, the rich invest their additional wealth abroad. Everyone else is deprived of income they could have used to buy additional imports. The result is that inequality within countries can cause imbalances between them. Fig. 3.3 German investors have lost almost 30 percent on their foreign investments (components of net international investment position, EUR trillions). Sources: Eurostat; Matthew Klein’s calculations The Two Causes of Imbalances: Pull versus Push Surpluses require deficits elsewhere and deficits require surpluses. One cannot happen without the other. The current account surplus or deficit of any individual country must balance out the sum of the current account surpluses and deficits of every other country. Sometimes finance is “pulled” in by people who want to spend more on consumption and investment than they earn. In this case, the countries with current account deficits are ultimately responsible for the imbalances. At other times it is “pushed” by people who choose to save regardless of whether there are good investments available. In that case the imbalance originates in the surplus countries. Neither cause of imbalances is necessarily better than the other. Countries with good investment opportunities may try to pull in foreign capital, but so will countries eager to fund unsustainable consumption booms or dubious projects. Countries on the receiving end of money indiscriminately pushed in their direction can sometimes turn the unexpected inflows into a windfall, but most fail to do so, particularly when the inflows are extremely large. There is no guaranteed way to identify the origin of an imbalance, but market prices can provide clues. After all, pulling money in from skeptics in the rest of the world is difficult. The trick is to pay them: foreign investors will commit capital anywhere if they can get returns that provide fair compensation for the risks of inflation, default, and currency devaluation. Although determining what is fair is a challenge, it is easy to tell when investors think they need more compensation for the risks they are taking: asset values go down. Rising real interest rates, collapsing equity multiples, and a falling currency are bad for existing owners but make new investments comparatively attractive. The current account deficit country is therefore the likely source of the imbalance if its interest rates rise (or the international value of its assets goes down) as its external financing need rises. This is a frequent phenomenon for countries on the periphery of the international financial system. Consider Turkey over the past few years. Between 2010 and 2018, the country went on an investment binge funded by borrowing from abroad. Turkey’s current account deficit was consistently worth about 6 percent of its domestic income. This boosted output at the cost of rising debt. Attracting foreign money to cover this excess of spending over income required persistent currency depreciation beyond what would have been implied by differences in inflation: the real value of the Turkish lira fell by half between 2010 and the middle of 2018. Unsurprisingly, foreign savers grew increasingly skeptical of Turkey and mostly lent in debt that could only be repaid with U.S. dollars, not Turkish lira. (Think of Argentina’s nineteenth-century gold-backed debts.) Interest rates also had to rise. The cost of a dollar-denominated business loan rose from about 4 percent in 2010 to 6 percent by the second half of 2018. Over the same period, interest rates on lira-denominated business loans soared from 9 percent to 28 percent. Eventually, foreign revulsion with Turkish assets slashed Turkish spending power, which in turn forced Turks to cut their consumption and investment to the point that they moved into a current account surplus by the middle of 2019.20 Things look different when unneeded capital is pushed in from abroad. Instead of rising, interest rates in the deficit country stay stable—or even fall—as the imbalance grows. This is what happened to Spain after committing to join the euro. In the mid-1990s, Spaniards spent roughly as much as they earned, so the current account was nearly balanced. Meanwhile, benchmark interest rates were over 5 percent after subtracting inflation. From then until 2008, Spain’s current account deficit steadily widened to roughly 10 percent of GDP. Yet over the same period, Spanish real interest rates dropped below 0 percent. Spain is a member of the euro area, so its currency could not move relative to its main trading partners in the rest of Europe. But Spanish wages and prices rose faster than in Spain’s neighbors, so its real effective exchange rate rose by 18 percent during this period. By every possible measure, Spain’s cost of capital was plunging even as Spaniards borrowed more and more from the rest of the world. Spaniards were not trying to pull money in—far from it. Rather, Spain was overwhelmed by a flood of foreign investment into its banking system, its bond market, and its real estate. (After stagnating throughout the 1990s, Spanish house prices more than doubled between 2001 and 2007.) These inflows increased the purchasing power of Spaniards far more than it increased their incomes. The difference was covered by a debt and investment binge nearly without precedent. As long as new money kept coming in, Spain appeared to experience a growth miracle that was the envy of Europe. Once foreigners stopped pushing their unneeded capital into Spain, however, Spanish spending was forced to rapidly contract below Spain’s incomes. The current account eventually reversed into a substantial surplus, but only at the cost of crushing unemployment.21 Why It Is Better to Give Than to Receive: The German Empire in the 1870s Spain’s experience was not unusual or unrepresentative. Lottery winners often do worse in life than if they had bought a losing ticket. The sudden infusion of money acts like a hit of cocaine and distorts behavior in similarly unhealthy ways. Something similar happens to countries on the receiving end of unsolicited financial inflows. Few societies have been able to absorb sudden, large sums of capital from abroad without experiencing soaring debt, asset bubbles, and economic crises. It is an almost inevitable consequence of a rapid and unexpected increase in real purchasing power. The experience of the new German Empire in the early 1870s presents one of the most striking examples of how the lottery of financial inflows affects recipients in broadly similar ways regardless of cultural and institutional differences. For most of its history, Germany had been divided into many small states with distinct identities, political forms, and religious traditions. By the nineteenth century, German nationalists were dreaming of unification. One state would have to force the others to submit to joint rule. The obvious candidate was the kingdom of Prussia, the second-most populous state in the post-Napoleonic German Confederation. (The Austrian Empire was much bigger but had far fewer Germans.) Prussia began to realize the nationalists’ dream by provoking and then winning short wars, first against Denmark in 1864, which “liberated” Schleswig-Holstein, and then against Austria in 1866, which gave Prussia an excuse to annex its hostile neighbors in northern Germany and create a new North German Confederation alongside its docile allies. Combined, these changes made Prussia the dominant power in Europe. By the 1860s, Prussia’s biggest rival was the Second French Empire, led by the nephew of Napoleon Bonaparte. Afflicted by gallstones and a weakness for mistresses, Louis-Napoléon was prone to strategic errors and presumptuous pronouncements. Anticipating an Austro-Prussian conflict, he had tried to negotiate secret treaties in 1865 and 1866 with both belligerents that would guarantee additional French territory on the Rhineland. Otto von Bismarck, the wily Prussian minister president, was able to use this information to convince the defeated southern German states (secretly) to ally with the North German Confederation in a mutual defense pact after the war. Louis-Napoléon persisted in his territorial quests, trying to buy Luxembourg from the Netherlands and then proposing—in violation of France’s treaty commitments—to annex Belgium. Things came to a head when, in 1868, Spanish republicans deposed Queen Isabella II. By 1870, the republican revolution had failed, and Spain needed a new monarch. France and Prussia each proposed alternative candidates. Although the Spanish never seriously considered Prussia’s Hohenzollern suggestion, the idea nevertheless drove the French to make unreasonable demands and eventually declare war on Prussia in July. Bavaria and the other south German states immediately declared their support for Prussia. Within six weeks, Louis-Napoléon had been captured after the Battle of Sedan. The Third French Republic continued the war after his capitulation, but it too was eventually forced to yield to the superior opponent. Prussia had won a resounding victory in battle. It had demonstrated its power to the other German states, which promptly joined the new Prussiandominated German Empire, and to the rest of Europe. The challenge was to postpone a rematch. The French were eager to avenge their humiliation and nursed dreams of reclaiming the German-speaking territory of AlsaceLorraine. Prussia, however, needed time to build its new political union, and it needed peace to recuperate from nearly a decade of warfare. Somehow, France had to be kept down long enough for Germany to become secure. The proposed solution was an indemnity: German armies would occupy much of France’s industrial heartland until the Third Republic paid 5 billion gold francs to the Reich—a payment equal to roughly a quarter of France’s economic output in 1870. The burden would be so high, it was hoped, that it would cripple French reconstruction and ensure a pacific western neighbor for generations. The other side of the transaction was a transfer to the German economy equal to around a fifth of its domestic production over three years.22 Despite the apparent magnitude of the indemnity, the French government found it relatively easy to raise the money, and the German government received the full amount ahead of schedule in 1873. French savers, it turned out, had ample resources to draw on. For years they had collectively accumulated assets abroad and used the income to cover the French trade deficit, gold imports, and purchases of additional foreign assets. After the war, France stopped importing gold, moved into a trade surplus, and stopped investing abroad, all of which freed up income to buy billions of francs’ worth of French government bonds. This alone covered about half the indemnity payment. The rest was covered by French sales of foreign assets to buy domestic bonds, foreign demand for French assets (especially from German savers), and French gold sales. To the great surprise of the Germans, this was not bad for France. While the French economy struggled immediately after the war, the new debt did not cripple it for long because France was easily able to manage the interest payments on its perpetual bonds. Part of the reason, according to the financial historian Charles Kindleberger, is that the French indemnity had expanded the global money supply. The German money supply obviously grew as gold flowed from France into its banking system, but this had not been offset by an equal reduction in France’s money supply. The debts the French government issued to finance the indemnity created a huge, highly liquid, and highly credible debt instrument—something very similar to money. The transfer of money from France to Germany therefore resulted in a sudden and substantial increase in the global supply of moneylike assets.23 The transfer, counterintuitively, ended up being harmful for Germany. Over three years it absorbed a financial inflow worth about 8 percent of GDP each year. Much of the money was allocated to military costs, ranging from debt repayments to new fortifications on the border with France to the establishment of a pension for veterans. This spending boosted German purchasing power, which raised the trade deficit through a boom in imports. Meanwhile, rising wages—German coal miners in the Ruhr saw their hourly pay jump 60 percent between 1871 and 1873, for example—and prices made German exports uncompetitive in global markets. Gold began flowing back from the Bank of Prussia to the Bank of France. One way or another, the balance of payments always balances. At least as damaging as the impact on the trade balance was the impact on Germany’s financial markets. The government knew it could not immediately spend the entire indemnity on infrastructure investment and military armament, because those projects take time. While it waited, the indemnity was invested in financial assets, including bonds issued by the German states and railroad bonds. Ludwig Bamberger, a German parliamentarian and a cofounder of Deutsche Bank, had warned of what this would do to financial conditions in Germany, and suggested the government hold the unspent funds in gold or foreign assets. His advice went unheeded, however, and Germany engaged in a frenzy of investment at home and abroad in which a substantial share of the inflows from France was wasted. The economist Arthur Monroe writes that the government’s investments “released on the German market, within about two years, a sum nearly three times as great as the total monetary stock of the country and considerably greater than the combined debt of all the German states, including debts incurred for railroad building.” The German economy responded to French financial inflows in essentially the same way that other economies have responded to large financial inflows before and since. The German economy grew rapidly in the immediate postwar period, growing more than 6 percent each year on average. After 1874, however, the country shrank for three straight years. Similarly, there was a short-lived banking boom in Germany and Austria, followed immediately by a bust once the reparations inflows stopped. The supply of German banknotes grew more than 12 percent a year from 1871 to 1874 and then shrunk by 10 percent each year until 1878. By 1876, the economic situation was so dire that a coalition of German manufacturers began pushing the government to adopt protective tariffs to compensate for the changes in the terms of trade wrought by the indemnity; these were eventually adopted in 1879. Economists and politicians throughout Germany blamed the indemnity for the country’s economic collapse. Some, especially in France, even believed that Berlin might send the money back.24 Navarro’s Error: Why Bilateral Flows Obscure the Sources of Imbalances Another valuable lesson can be learned from the Franco-German experience of the 1870s: while the flow of gold from France to Germany ultimately pushed Germany into a trade deficit and France into a trade surplus, this was not matched by a corresponding change in the trade balance between France and Germany. Germans spent more on imports from all over the world while their exports to the rest of the world stagnated. France imported less from the rest of the world and exported more. The bilateral financial flow affected trade between France and Germany, but the effect on their bilateral trade balance was insignificant compared to the broader impact. In general, the dynamics of surpluses and deficits cannot be explained by focusing on bilateral trade and financial relations. This means that 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.25 Similarly, the United States consistently reports large bilateral current account surpluses with the Netherlands and with Singapore. We have shown that this can be explained by the combination of corporate tax avoidance strategies and by the misreporting of U.S. goods exports that land in major ports before being transshipped elsewhere. Despite these bilateral surpluses, both the Dutch and the Singaporeans consistently spend far less than they earn—both in absolute terms and relative to their incomes. They are both among the largest contributors to global imbalances today and are therefore among the biggest contributors to America’s current account deficit. If they saved less and spent more on imports from other countries, the extra income would eventually flow through to additional demand for U.S. exports, regardless of whether this affected their bilateral balances with the United States. The same perspective applies to the financial account. Bilateral balances in the financial account provide no insight into which countries are spending more or less than they earn. In fact, bilateral financial account balances are unrelated to the bilateral balances in the current account. Moreover, there is no reason why a country’s bilateral trading relations should mirror its bilateral financial relations. Homebuyers rarely get mortgages directly from sellers, for example. German banks making loans to French banks that lend to their Greek subsidiaries that then buy Greek government bonds issued to pay for German-made submarines were ultimately financing trade between Germany and Greece, even though the bilateral financial flows were from Germany to France and then from France to Greece. It is also difficult to determine the true nationality of people buying financial assets when they use custodial centers to remain anonymous or route their purchases through tax havens. (This is analogous to the problems with bilateral data on trade and profits earned from foreign investments described earlier.) Peter Navarro, the Harvard-trained economics professor and trade adviser to Donald Trump, disagrees with this analysis. Navarro believes that bilateral trade deficits matter, and he also seems to believe that these bilateral trade balances necessarily map onto corresponding bilateral financial flows. In a 2017 column in the Wall Street Journal, for example, he wrote that the problem with China’s bilateral trade surplus with the United States is that it enables China to “[buy] up America’s companies, technologies, farmland, food-supply chain—and ultimately [control] much of the U.S. defense-industrial base.”26 As it happens, the data imply that Americans have invested about $46 billion into China between the beginning of 2015 and the beginning of 2019, while Chinese residents have sold a total of $380 billion of U.S. assets. In other words, a total of $430 billion in net financing went from the United States to China. If bilateral financial flows mirrored bilateral trade flows, as Navarro seems to believe, the United States would have experienced a large current account surplus with China over this period. Instead, America had a bilateral deficit with China worth a bit over $1.5 trillion. Even if the official data wildly understate Chinese investment into the United States in those years because of custodial relationships and surreptitious financial flows, it still seems unlikely that direct Chinese financing of U.S. spending was enough to cover the difference between American exports to and imports from China. Savers in Europe likely helped cover the funding gap. From the start of 2015 through the beginning of 2019, the total difference between foreign savings coming into the United States and American savings going out to the rest of the world through the financial account was about $1.5 trillion. Over the same period, residents of the euro area invested about $976 billion more in the United States than Americans invested in Europe. About twothirds of the total U.S. financial account surplus, in other words, has come from residents of the euro area. Yet those funds have not been returned directly to Europe: the total bilateral current account deficit from the start of 2015 through the beginning of 2019 has been just $116 billion. About $860 billion in funding therefore went from Europe to the United States, only to be spent in China and elsewhere. Ironically, Navarro would have been on firmer ground criticizing China if he had abandoned his focus on bilateral balances and adopted a global perspective. China has a current account surplus with the rest of the world because Chinese residents invest more abroad than non-Chinese invest in China. At the same time, the United States has a current account deficit because non-Americans invest more in the United States than Americans invest abroad. These facts are both more relevant and more compelling than the bilateral relations. Navarro’s flawed analytical framework leads to even larger errors when considering the case of Mexico—a frequent target of the Trump administration. The United States consistently runs large trade deficits with Mexico. Americans also pay Mexicans to work in the United States, while many people living in the United States send remittances to their relatives in Mexico. Add it all together and the cumulative bilateral current account deficit from the beginning of 2015 through the beginning of 2019 was about $350 billion. In the same Wall Street Journal column from 2017, Navarro claimed that if “America successfully negotiates a bilateral trade deal this year with Mexico in which Mexico agrees to buy more products from the U.S. that it now purchases from the rest of the world” then this “would show up in government data as an increase in U.S. exports [and] a lower trade deficit.”27 There is no reason to believe this. If, as Navarro implies, Mexicans were to pay for additional U.S. exports by buying fewer exports from the rest of the world, their total spending would not change. 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. The likelier outcome is that Navarro’s proposal would increase the U.S. aggregate current account deficit. Mexicans presumably have good reasons, such as price and quality, to buy goods and services from non-U.S. producers. If they were compelled to switch vendors, they would probably have to spend more money than before just to get the same amount of satisfaction. They might instead choose to cut their spending and save more. Penalizing Mexican exports would also be counterproductive. Losing access to the U.S. market would make Mexico less attractive to foreign investors while increasing the relative attractiveness of American manufacturing assets. The result would be a global shift in financial flows from Mexico to the United States that would boost American purchasing power and depress spending in Mexico. The net effect would be a narrower bilateral U.S. trade deficit with Mexico at the cost of a wider overall American trade deficit. Navarro fails to understand that Mexico absorbs excess global savings and manufactured products that would otherwise have increased the U.S. trade deficit. America’s bilateral current account deficit with Mexico cannot contribute to America’s overall deficit because Mexicans, like Americans, spend more than they earn. After all, Mexico consistently runs one of the world’s largest current account deficits in absolute terms, consistently worth about 2 percent of its GDP. 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. Only about 60 percent of the value of U.S. imports from Mexico comes from Mexico.28 There is only one thing Mexicans could do to try to increase total U.S. exports: increase their spending and widen their own current account deficit. Mexicans would end up producing extra income for the rest of the world, some of which might eventually turn into heightened demand for American-made goods and services. Even this might fail, especially if the surplus countries save their windfall from additional Mexican spending, rather than use the money to support extra consumption and capital investment that would eventually flow back to the United States. In that scenario, extra spending by Mexicans would simply increase current account surpluses elsewhere rather than contribute to a narrower American deficit. This strategy would also be extremely risky for Mexico, since it sits somewhere between the United States and Turkey in its ability to attract foreign savings. If Mexico were pushed to borrow more to support extra spending on U.S. exports, the result would likely be a temporary boom followed by a crisis. The subsequent decline in Mexican spending would more than offset any short-term benefits to either the United States or Mexico. If Navarro truly wanted to understand the development of America’s trade deficit over the past several decades, he would do better to focus on why people in the major surplus economies have consistently spent less than they earned. He would also do well to study why the world’s savers have long preferred to store their excess wealth in the United States and what that has meant for Americans. As we will explain, starting with China, the answers have nothing to do with cultures of thrift or profligacy. Instead, they have everything to do with the distribution of income and the structure of the global monetary system.