And therein lies the solution to the riddle: during the nineteenth century, trade within a nation was much more important than trade with other countries. As long as a country’s internal markets were open, a tariff wall against foreign goods did relatively little damage. Before the twentieth century, external trade constituted only a tiny part of the economy of most nations. For example, in 1870 exports constituted just 2.5 percent of GDP in the United States; in France, 4.9 percent; and even in free-trading England, only 12.2 percent. As trade has grown, the world economy has become more dependent on it, with exports rising from 4.6 percent of world GDP in 1870 to 17.2 percent in 1998.
Geography contributes as well: the larger and more economically diverse a nation, the more self-sufficient it is, and the less important trade becomes. Since its independence, the United States has been the most self-sufficient of all major nations; today imports constitute only about 14 percent of its GDP. Holland lies at the other extreme, with imports making up 61 percent of its economy.17 This is consistent with Edward Denison’s data for the period 1950–1962, which showed the largest benefits of the worldwide fall in tariffs in the Netherlands, Belgium, and Norway; less benefit for the larger and more diverse economies of Germany and France; and none at all for the United States.18
In the nineteenth century, nations, particularly large, self-reliant nations such as the United States, could get away with protectionist trade policies. In the globally integrated economy of the twenty-first century, autarky becomes a much dicier proposition. Further, most of the damage done to the economies of the developing world has been self-inflicted; to paraphrase Cordell Hull, protectionism is a gun that recoils most forcefully upon the least fortunate.
The intangible rewards of free trade are underestimated. A century and a half ago, the French economist Frédéric Bastiat reportedly said, “When goods are not allowed to cross borders, soldiers will.”19 The Nobel Committee certainly agreed when it awarded the 1945 Peace Prize to Cordell Hull for his role in reopening world trade in the 1930s and 1940s.
Life on earth is slowly becoming less violent, mainly owing to the increasing realization that neighbors are more useful alive than dead. Those who doubt this rosy assessment should consider some data from the World Health Organization. Their statistics show that in 2004, violence accounted for just 1.3 percent of the world’s deaths, an all-time low, and that at the beginning of the twenty-first century the number of battle deaths per year has decreased to one-thirtieth of what it was in the 1950s. This seems to be part of a longer-term historical trend; archaeological data suggest that upward of 20 percent of Stone-Age populations met violent deaths, a finding that is supported by research on modern hunter-gatherer societies.20 Perhaps the strongest evidence supporting the connection between trade and peace is the European Union, which has rendered military conflict among its powers highly unlikely—this on a continent that saw more or less continuous warfare before 1945. In microeconomic terms, it makes little sense to bomb those who purchase and produce our shirts, laptops, and automobiles.
Today’s greatest threats to world security come not from conventional armies, but rather from terrorism based in the world’s failed states—precisely those parts of the globe that would benefit most from freer trade and a decrease in agricultural subsidies. To paraphrase Bastiat, if cotton, sugar, and rice can cross borders, then perhaps terrorists will not be able to.
Although free trade has benefited mankind in the aggregate, it has also produced many losers. The future expansion of world trade, resulting from increasing prosperity, falling tariffs, and vanishing transport costs, will create yet more winners, as well as a growing minority of losers. Not treating these losers fairly and compassionately is a prescription for failure. Again, Stolper-Samuelson provides the proper framework. Instead of two or three factors, consider only labor. Now divide it into two categories: high-skilled and low-skilled. Compared with the rest of the world, the developed nations are relatively well endowed with abundant high-skilled labor and have relatively little unskilled labor.
Whom does free trade hurt in the developed world? The relatively scarce factor: low-skilled labor. Who benefits? Highly skilled workers. Further, globalization increases income inequalities in the rich nations, as the inflation-adjusted incomes of the highly skilled rise rapidly and those of the low-skilled rise more slowly or even fall.
Once again, Stolper-Samuelson comes alive in the real world. For the past generation, income inequalities have grown dramatically in the United States. Figure 14-1 plots Census Bureau data with American families sorted into two groups—the top quintile, or 20 percent, and bottom 80 percent—and then computes their shares of total national income over the past thirty-five years.
Figure 14-1. Shares of Total National Income
This graph clearly shows that in America, the top quintile, or 20 percent, of the population has become relatively richer, increasing its share of the income pie by one-sixth (from 41 percent to 48 percent) between 1970 and 2005, while everyone else has gotten relatively poorer.
Most people would describe those at the lower border of the top quintile, who earned $103,100 in 2005, as only modestly wealthy. To see how the rich are really doing, the population’s top end needs to be sliced more thinly. Those in the top 5 percent, who earned more than $184,500 in 2005, saw their slice of the pie increase by more than one-third during the preceding thirty-five years; the top 1 percent, who earned more than $340,000 in 2005, saw their piece double in size.
Whereas the incomes of highly skilled professionals and managers have soared in recent decades, the inflation-adjusted salary of the median male worker (at the fiftieth percentile of income) has not risen for more than a generation.21 There is more than a little truth in the popular American perception of globalization: highly paid factory jobs are disappearing overseas and being replaced with burger-flipping slots.
Worse, even though the unemployment rate has fallen during the past two decades, job insecurity has grown dramatically during the same period. Average workers are about one-third more likely to lose their job than they were two decades ago, and to be paid almost 14 percent less when they eventually do find work again, if they are lucky enough to do so. (One-third of workers are not so lucky.) A 1998 Wall Street Journal survey asked Americans if they agreed with the statement, “Foreign trade has been bad for the U.S. economy because cheap imports have cost wages and jobs here.” As predicted by Stolper-Samuelson, this issue cleaves the nation along the abundant-scarce factor fault line: among those earning more than $100,000 per year, only one-third agreed, whereas among blue-collar workers and union members, two-thirds agreed.22
Stolper-Samuelson does fail in at least one area by predicting that freer trade should decrease inequalities in developing nations by helping low-skilled workers. In fact, the opposite occurs: the most highly skilled industrial workers earn better pay in call centers and multinational-owned plants, increasing the gap between those who are fortunate enough to find such work and those who are not.23 Although working conditions in an Asian Nike factory may appall people in the developed world, positions in American-associated factories are the most sought-after jobs in Vietnam’s “development zones.” Far less desirable are jobs in the Spartan and more strictly run Chinese-owned factories. Worst of all are the alternatives to factory work: subsistence-level farming and prostitution.24
The developing world exports to the United States not only shirts, sneakers, and electronics but also its abundant human capital: low-skilled workers who compete with Americans depress domestic wages, increase the income gap, and fan anti-immigration sentiment. It is no accident that union members are among the most vocal proponents of stricter immigration policy.
Yet again, this is nothing new. In the nineteenth century, as inequality rose, so did fear of immigration. A century ago, the United States, Canada, Australia, Brazil, and Argentina all began to restrict immigration. This tightening did not correlate at all with the factors that traditionally
have received blame for it: economic hard times and racism. Rather it took place precisely when income competition from recent European immigrants at the low end of the wage scale started to squeeze low-paid voters.25
Why all the fuss about income inequality? Isn’t it simply a sign of a healthy economy that generously rewards the successful and ambitious? Well, no. Economists and demographers use a number of measures of inequality, the most popular being the Gini coefficient. This number varies between zero and one; a population in which everyone has exactly the same income has a coefficient of 0.0, and a population in which only one individual earns all the income has a coefficient of 1.0.
The Ginis of the world’s twenty wealthiest nations range from 0.25 (Sweden) to 0.41 (United States). The list of the highest-Gini nations fails to impress: Namibia, 0.74; Botswana, 0.63; Bolivia, 0.60; and Paraguay, 0.58.26 More systematic research strongly suggests that increasing inequality produces social and political instability, which in turn leads to decreased investment and decreased economic development.27
Modern developed nations have gotten into the habit of intentionally lowering their Ginis with redistributive tax policies and social welfare programs. These expensive schemes can harm economic development, but by reducing inequality, they can also buy social peace, which can offset the inefficiencies of social spending. One of the leading authorities in this area, Geoffrey Garrett, notes that:
Since the welfare state mitigates conflict by reducing market-generated inequalities of risk and wealth, it may have a beneficial rather than deleterious consequence for business. Government spending may thus stimulate investment via two channels—increasing productivity through improvements in human and physical capital and increasing stability through maintaining support for market openness.28
In other words, it is important to find a happy medium between the Scylla of expensive and economically damaging social welfare systems and the Charybdis of too thin a safety net, which worsens inequality. The United States and northern Europe are nearly equally wealthy, yet the United States cycles about 30 percent of GDP through federal, state, and local governments, whereas the governments of northern Europe consume nearly half of GDP, most of which goes to pay for their social welfare schemes. This suggests that the “sweet spot” lies somewhere between these two points.
The problem is that not all, or perhaps even most, of this growing inequality and insecurity can be blamed on freer trade. Economists hotly debate exactly how much damage is caused by outsourcing and the loss of jobs to foreign factories, and how much is due to an increase in the wages paid to highly-trained and educated workers.
Consider two different farm hands, one who is able to harvest wheat with 99.5 percent efficiency, and another who is able to harvest it with 95 percent efficiency. Certainly, the first laborer might command a larger salary than the second, but not by very much. Now consider a factory making a complex microchip that requires a hundred manufacturing steps, an error in any one of which will ruin the end product. Here, a labor force that completes each step with 99.5 percent accuracy will result in a 39 percent defect rate, versus a 99.4 percent reject rate for a labor force with 95 percent accuracy. Thus, in advanced manufacturing and service economies, highly skilled workers will command a larger wage premium than in less advanced ones. (This paradigm has been called the “O–Ring Theory,” after the minor design flaw which destroyed the Challenger space shuttle.)29
Paul Krugman believes that almost all of the growing inequality in wages in the United States is due to this increased skill premium (and more recently, changes in tax policy), whereas data from economist Adrian Wood suggest that much, if not most, is due to increased international trade. The consensus seems to be somewhere in the middle: perhaps one-fifth to one-quarter of the widening inequality in U.S. wages is due to trade, the rest to tax cuts aimed at the wealthy and the increasing rewards of domestic education and training.30
The Free Trade Act of 1989 between the United States and Canada gave researchers a nearly ideal lens through which to observe the globalization trade-off. The act slashed tariffs from about 8 percent to 1 percent on goods moving north, and from 4 percent to 1 percent on goods moving south. Both nations have stable legal, banking, and political institutions, and since the United States dominates the Canadian economy, the most dramatic effects of the act occurred north of the border.
The economist Daniel Trefler calculated that although the act produced a significant net benefit to Canada as a whole, with long-term productivity in some industries increasing by up to 15 percent, it also vaporized about 5 percent of Canadian jobs, up to 12 percent of existing positions in some industries. These job losses, however, lasted less than a decade; overall, unemployment in Canada has fallen since the passage of the act. Commenting on this trade-off, Trefler wrote that the critical question in trade policy is to understand “how freer trade can be implemented in an industrialized economy in a way that recognizes both the long-run gains and the short-term adjustment costs borne by workers and others.”31
For almost two decades, economists and politicians have grappled with the problem of how, or even whether, those left behind by free trade should be compensated. In 1825, John Stuart Mill calculated that although the Corn Laws put a certain amount of extra money in the pockets of the landlords, these laws cost the nation as a whole several times more. He theorized that it would be far cheaper to buy the landlords off:
The landlords should make an estimate of their probable losses from the repeal of the Corn Laws, and found upon it a claim to compensation. Some, indeed, may question how far they who, for their own emolument, imposed one of the worst of taxes upon their countrymen [i.e., the Corn Laws], are entitled to compensation for renouncing advantages which they never ought to have enjoyed. It would be better, however, to have a repeal of the Corn Laws, even clogged by a compensation, than to not have it at all; and if this were our only alternative, no one could complain of a change, but which, though an enormous amount of evil would be prevented, no one would lose.32
In other words, it is far cheaper and better for all to directly compensate the losers. Almost two centuries after those words were written, and half a century after Cordell Hull and Proposals started the world down the road to free trade, the resultant inequalities and dislocations are again beginning to derail the process. Can free trade, with all its benefits, indeed be saved by compensating the losers?
Many American advocates of free trade realize that in order for the current system of relatively free trade to survive, the nation’s social safety net needs to be expanded, but lip service, at best, is all that is usually offered. Consider Jagdish Bhagwati, perhaps today’s best-known proponent of trade liberalization and a formidable academician who has trained many of today’s foremost economists. His book In Defense of Globalization spends three hundred pages living up to its title; it gives the issue of “adjustment assistance” less than two pages. The following passage from this book captures the tone used by many free traders in discussing displaced workers:
If a steel mill closes down in Pennsylvania because steel in California has become cheaper, workers tend to accept that as something that happens, and the general unemployment insurance seems to be an adequate way to deal with the bad hand that an unpredictable fate has dealt one. But the same workers get indignant when the loss is to a steel producer in Korea or Brazil, and they go off agitating for anti-dumping action. . . . Or they ask for special relief in the form of additional unemployment compensation, with or without retraining benefits and requirements.33
Professor Bhagwati only grudgingly accepts the need for compensation. Referring to a safety net specifically intended for those put out of work by foreign products, he continues: “This quasi-xenophobia is just a fact of life. If trade liberalization is to occur and be sustained, one or more of these special programs and policies have to be considered.”34
Such sentiments not only unnecessarily antagonize workers but also are unfair; American industr
y has in fact been much more adept than labor at getting protection, particularly in the form of non-tariff barriers: quotas, subsidies, antidumping legislation, and the like.35 Trade economists are slowly beginning to realize that they must stop being their own worst enemies. Dani Rodrik of the Kennedy School of Government at Harvard has, with great sensitivity, extensively surveyed the social havoc caused by the increased mobility of goods and services, explored the necessity of compensation, and considered how it might work. He believes it is no accident that those developed nations with the highest ratio of trade to GDP also have the richest social welfare schemes.36 Free trade and a generous safety net reinforce each other; stable, wealthy trading nations, if they want to remain that way, cannot afford to throw to the wolves those whose jobs can be so easily “exchanged” in an increasingly frictionless global economy. According to Rodrik:
Social spending has the important function of buying social peace. Without disagreeing about the need to eliminate waste and reform in the welfare state more broadly, I would argue that the need for social insurance does not decline but rather increases as global integration increases.37
More than five decades after Stolper and Samuelson penned their treatise explaining who won and who lost, Paul Samuelson, considered by many the world’s greatest living economist, again stunned his colleagues by suggesting that entire nations could lose from free trade. He explained, in dense quantitative language decipherable only by other economists, how such competition from foreign labor causes job displacement but not job loss. (Indeed, unemployment in the United States is currently just under 5 percent.) Yes, Americans are still working, but they are doing so in jobs that pay less and are shorn of benefits. Samuelson reckons that these losses in compensation and benefits are permanent, and that as a whole, the United States is relatively worse off as a nation because of free trade:
Splendid Exchange, A Page 45