The Conscience of a Liberal

12 CONFRONTING INEQUALITY

The America I grew up in was a relatively equal middle-class society. Over the past generation, however, the country has returned to Gilded Age levels of inequality. In this chapter I’ll outline policies that can help reverse these changes. As I did in discussing health care, however, I’ll begin with the question of values. Why should we care about high and rising inequality?

One reason to care about inequality is the straightforward matter of living standards. As I documented at length in chapter 7, the lion’s share of economic growth in America over the past thirty years has gone to a small, wealthy minority, to such an extent that it’s unclear whether the typical family has benefited at all from technological progress and the rising productivity it brings. The lack of clear economic progress for lower-and middle-income families is in itself an important reason to seek a more equal distribution of income.

Beyond that, however, is the damage extreme inequality does to our society and our democracy. Ever since America’s founding, our idea of ourselves has been that of a nation without sharp class distinctions—not a leveled society of perfect equality, but one in which the gap between the economic elite and the typical citizen isn’t an unbridgeable chasm. That’s why Thomas Jefferson wrote, “The small landholders are the most precious part of a state.”[1] Translated into modern terms as an assertion that a broad middle class is the most precious part of a state, Jefferson’s statement remains as true as ever. High inequality, which has turned us into a nation with a much-weakened middle class, has a corrosive effect on social relations and politics, one that has become ever more apparent as America has moved deeper into a new Gilded Age.

One of the best arguments I’ve ever seen for the social costs of inequality came from a movement conservative trying to argue the opposite. In 1997 Irving Kristol, one of the original neoconservative intellectuals, published an article in the Wall Street Journal called “Income Inequality Without Class Conflict.” Kristol argued that we shouldn’t worry about income inequality, because whatever the numbers may say, class distinctions are, in reality, all but gone. Today, he asserted,

By claiming that income inequality doesn’t matter because we have social equality, Kristol was in effect admitting that income inequality would be a problem if it led to social inequality. And here’s the thing: It does. Kristol’s fantasy of a world in which the rich live just like you and me, and nobody feels socially inferior, bears no resemblance to the real America we live in.

Lifestyles of the rich and famous are arguably the least important part of the story, yet it’s worth pointing out that Kristol’s vision of CEOs rubbing shoulders with the middle class is totally contradicted by the reporting of Robert Frank of the Wall Street Journal, whose assigned beat is covering the lives of the wealthy. In his book Richistan Frank describes what he learned:

The fact is that vast income inequality inevitably brings vast social inequality in its train. And this social inequality isn’t just a matter of envy and insults. It has real, negative consequences for the way people live in this country. It may not matter much that the great majority of Americans can’t afford to stay in the eleven-thousand-dollar-a-night hotel suites popping up in luxury hotels around the world.[4] It matters a great deal that millions of middle-class families buy houses they can’t really afford, taking on more mortgage debt than they can safely handle, because they’re desperate to send their children to a good school—and intensifying inequality means that the desirable school districts are growing fewer in number, and more expensive to live in.

Elizabeth Warren, a Harvard Law School expert in bankruptcy, and Amelia Warren Tyagi, a business consultant, have studied the rise of bankruptcy in the United States. By 2005, just before a new law making it much harder for individuals to declare bankruptcy took effect, the number of families filing for bankruptcy each year was five times its level in the early 1980s. The proximate reason for this surge in bankruptcies was that families were taking on more debt—and this led to moralistic pronouncements about people spending too much on luxuries they can’t afford. What Warren and Tyagi found, however, was that middle-class families were actually spending less on luxuries than they had in the 1970s. Instead the rise in debt mainly reflected increased spending on housing, largely driven by competition to get into good school districts. Middle-class Americans have been caught up in a rat race, not because they’re greedy or foolish but because they’re trying to give their children a chance in an increasingly unequal society.[5] And they’re right to be worried: A bad start can ruin a child’s chances for life.

Americans still tend to say, when asked, that individuals can make their own place in society. According to one survey 61 percent of Americans agree with the statement that “people get rewarded for their effort,” compared with 49 percent in Canada and only 23 percent in France.[6] In reality, however, America has vast inequality of opportunity as well as results. We may believe that anyone can succeed through hard work and determination, but the facts say otherwise.

There are many pieces of evidence showing that Horatio Alger stories are very rare in real life. One of the most striking comes from a study published by the National Center for Education Statistics, which tracked the educational experience of Americans who were eighth graders in 1988. Those eighth graders were sorted both by apparent talent, as measured by a mathematics test, and by the socioeconomic status of their parents, as measured by occupations, incomes, and education.

The key result is shown in Table 10. Not surprisingly, both getting a high test score and having high-status parents increased a student’s chance of finishing college. But family status mattered more. Students who scored in the bottom fourth on the exam, but came from families whose status put them in the top fourth—what we used to call RDKs, for “rich dumb kids,” when I was a teenager—were more likely to finish college than students who scored in the top fourth but whose parents were in the bottom fourth. What this tells us is that the idea that we have anything close to equality of opportunity is clearly a fantasy. It would be closer to the truth, though not the whole truth, to say that in modern America, class—inherited class—usually trumps talent.

Isn’t that true everywhere? Not to the same extent. International comparisons of “intergenerational mobility,” the extent to which people can achieve higher status than their parents, are tricky because countries don’t collect perfectly comparable data. Nonetheless it’s clear that Horatio Alger has moved to someplace in Europe: Mobility is highest in the Scandinavian countries, and most results suggest that mobility is lower in the United States than it is in France, Canada, and maybe even Britain. Not only don’t Americans have equal opportunity, opportunity is less equal here than elsewhere in the West.

Source: National Center for Education Statistics, The Condition of Education 2003, p. 47.

It’s not hard to understand why. Our unique lack of universal health care, all by itself, puts Americans who are unlucky in their parents at a disadvantage: Because American children from low-income families are often uninsured, they’re more likely to have health problems that derail their life chances. Poor nutrition, thanks to low income and a lack of social support, can have the same effect. Life disruptions that affect a child’s parents can also make upward mobility hard—and the weakness of the U.S. social safety net makes such disruptions more likely and worse if they happen. Then there’s the highly uneven quality of U.S. basic education, and so on. What it all comes down to is that although the principle of “equality of opportunity, not equality of results” sounds fine, it’s a largely fictitious distinction. A society with highly unequal results is, more or less inevitably, a society with highly unequal opportunity, too. If you truly believe that all Americans are entitled to an equal chance at the starting line, that’s an argument for doing something to reduce inequality.

America’s high inequality, then, imposes serious costs on our society that go beyond the way it holds down the purchasing power of most families. And there’s another way in which inequality damages us: It corrupts our politics. “If there are men in this country big enough to own the government of the United States,” said Woodrow Wilson in 1913, in words that would be almost inconceivable from a modern president, “they are going to own it.”[7] Well, now there are, and they do. Not completely, of course, but hardly a week goes by without the disclosure of a case in which the influence of money has grotesquely distorted U.S. government policy.

As this book went to press, there was a spectacular example: The way even some Democrats rallied to the support of hedge fund managers, who receive an unconscionable tax break. Through a quirk in the way the tax laws have been interpreted, these managers—some of whom make more than a billion dollars a year—get to have most of their earnings taxed at the capital gains rate, which is only 15 percent, even as other high earners pay a 35 percent rate. The hedge fund tax loophole costs the government more than $6 billion a year in lost revenue, roughly the cost of providing health care to three million children.[8] Almost $2 billion of the total goes to just twenty-five individuals. Even conservative economists believe that the tax break is unjustified, and should be eliminated.[9]

Yet the tax break has powerful political support—and not just from Republicans. In July 2007 Sen. Charles Schumer of New York, the head of the Democratic Senatorial Campaign Committee, let it be known that he would favor eliminating the hedge fund loophole only if other, deeply entrenched tax breaks were eliminated at the same time. As everyone understood, this was a “poison pill,” a way of blocking reform without explicitly saying no. And although Schumer denied it, everyone also suspected that his position was driven by the large sums hedge funds contribute to Democratic political campaigns.[10]

The hedge fund loophole is a classic example of how the concentration of income in a few hands corrupts politics. Beyond that is the bigger story of how income inequality has reinforced the rise of movement conservatism, a fundamentally undemocratic force. As I argued in chapter 7, rising inequality has to an important extent been caused by the rightward shift of our politics, but the causation also runs the other way. The new wealth of the rich has increased their influence, sustaining the institutions of movement conservatism and pulling the Republican Party even further into the movement’s orbit. The ugliness of our politics is in large part a reflection of the inequality of our income distribution.

More broadly still, high levels of inequality strain the bonds that hold us together as a society. There has been a long-term downward trend in the extent to which Americans trust either the government or one another. In the sixties, most Americans agreed with the proposition that “most people can be trusted”; today most disagree.[11] In the sixties, most Americans believed that the government is run “for the benefit of all”; today, most believe that it’s run for “a few big interests.”[12] And there’s convincing evidence that growing inequality is behind our growing cynicism, which is making the United States seem increasingly like a Latin American country. As the political scientists Eric Uslaner and Mitchell Brown point out (and support with extensive data), “In a world of haves and have-nots, those at either end of the economic spectrum have little reason to believe that ‘most people can be trusted’…social trust rests on a foundation of economic equality.”[13]

In discussing ways to reduce inequality, it’s helpful to make a distinction between two concepts of inequality, and two kinds of inequality-reducing policies.

The first concept of inequality is inequality in market income. The United States is, of course, a market economy. Most people get most of their income by selling their labor to employers; people also get income from the market return to assets such as stocks, bonds, and real estate. So one measure of inequality is the inequality of the income people get from selling things. The distribution of market income is highly unequal and getting more so. In fact, market income is now as unequally distributed as it was in the 1920s.

But that’s not the end of the story. The government collects part of market income in the form of taxes, and transfers part of that revenue back to the public either in direct payments, like the Social Security checks that are the main source of income for most older Americans, or by paying for goods and services like health care. So another measure of inequality is the inequality of disposable income—income after you take taxes and government transfers into account. In modern America, as in all advanced countries, inequality in disposable income is less than inequality in market income, because we have a welfare state—though a small one by international standards. Taxes and transfers, which somewhat crimp the living standards of the rich while helping out the less fortunate, are the reason America in 2007 doesn’t feel quite as unequal as America did in the twenties.

Now, one way to reduce inequality in America is to do more of this: to expand and improve our aftermarket policies, which take the inequality of market income as given but act to reduce its impact. To see how this might work, let me describe an example of a country that does vastly more to reduce inequality than we do: France.

If you’re going through a rough patch in your life—or if your whole life has been rough—it’s definitely better to be French than American. In France, if you lose your job and have to take an inferior one, you don’t have to worry about losing your health insurance, because health insurance is provided by the government. If you’re out of work for a long time, the government helps keep you fed and housed. If you’re financially pinched by the cost of raising children, you get extra money from the state, as well as help with day care. You aren’t guaranteed a comfortable life, but your family members, especially your children, are protected against experiencing really severe material deprivation.

On the other hand, if things are going very well for you, being French has its drawbacks. Income tax rates are somewhat higher than they are in the United States, and payroll taxes, especially the amount formally paid by employers but in effect taken out of wages, are much higher. Also, the cost of living is high because France has a high value-added tax, a form of national sales tax. For people with high incomes these burdens aren’t fully offset by the advantages of government health insurance and other benefits. So a Frenchman whose compensation (including payroll taxes paid by his employer) is in the range we would consider upper-middle-class or higher has substantially less purchasing power than does an American receiving the same compensation.

In other words, France has extensive aftermarket policies that reduce inequality by comforting the afflicted but somewhat afflict the comfortable. In this France is typical of non-English-speaking Western nations. And even other English speakers do more to reduce aftermarket inequality than we do.

For example, the United States spends less than three percent of GDP on programs that reduce inequality among those under 65. To match what Canada does we’d have to spend an additional 2.5 percent of GDP; to match what most of Europe does would require an extra 4 percent of GDP; to match the Scandinavian countries, an additional 9 percent.[14] U.S. programs reduce poverty among the nonelderly by 28 percent, compared with 54 percent in Canada, 61 percent in Britain, and 78 percent in Sweden.[15] And these numbers actually understate the difference between the United States and everyone else, because they don’t take account of the unique American failure to guarantee health care to all.

One “easy” way for the United States to reduce inequality beyond the effective reduction we’d achieve with universal health care would be for America to do considerably more to help the unlucky through public aid of various kinds, paid for with higher taxes on the well off. The additional spending would probably consist largely of expanding programs we already have: an expanded earned income tax credit, a more generous food stamp program, bigger housing aid, and so on. It could also include other items such as child support and help with day care. I’ll talk about where the additional revenue could come from later in this chapter.

But don’t high taxes and an extensive welfare state remove the incentives to work and innovate? Gross domestic product per capita in France is only 74 percent of GDP per capita in the United States. Isn’t that a compelling argument against moving in a French direction? Well, France and other countries with generous social programs do have serious economic problems. Those problems are not, however, as simple or as closely related to the generosity of social programs as you might think.

France does have much lower GDP per person than the United States. That’s largely because a smaller fraction of the population is employed—French GDP per worker is only 10 percent lower than in the United States.[16] And that difference in GDP per worker, in turn, is entirely because French workers get much more time off: On average French workers put in only 86 percent as many hours each year as U.S. workers.[17] Worker productivity per hour appears to be slightly higher in France than in the United States.

The real question is which aspects of the French difference represent problems, and which simply represent different and possibly better choices. The lower number of hours per worker in France seems to fall in the second category. In the United States vacations are very short, and many workers get no vacation at all. France has essentially made a decision, enforced by legal requirements on vacation time as well as union settlements, to trade less income for more time off. And there’s some evidence that this decision actually makes most people better off. As one recent study of the difference in working hours between Europe and the United States points out, polls suggest that people would like to work shorter hours, and international comparisons of reported “life satisfaction” seem to say that working less improves the quality of life even if it reduces income. Yet it’s very difficult for any individual, operating on his or her own, to trade less income for more leisure. French rules and regulations that solve this problem by requiring that employers provide vacation may actually be a good thing, even though they reduce GDP.[18]

In addition to working fewer hours than Americans, the French are less likely to work at all. To be more specific: In France the young and old tend not to be employed. About 80 percent of French prime-age adults, those between the ages of twenty-five and fifty-four, are employed, which is almost exactly the same as the U.S. number. However, only 25 percent of French residents aged fifteen to twenty-four are employed, compared with 54 percent in the United States, and only 41 percent of those aged fifty-five to sixty-four are employed, compared with 62 percent here.[19] The question is whether these low employment rates should be viewed as a problem.

The low rate of employment among young people in France is less of a problem than it may appear. It does to some extent reflect regulations that make it hard for employers to fire workers, and therefore make them reluctant to hire workers in the first place. A close examination, however, reveals that other reasons for low employment of young people in France are probably more important. The French are more likely than Americans to stay in school: 92 percent of French residents aged from fifteen to nineteen are in school, and 45 percent of those aged twenty to twenty-four, compared with 84 and 35 percent, respectively, in the United States. And only about 10 percent of French students hold jobs at the same time, compared with about 20 percent in the United States. Presumably, in France the combination of free education and public financial support lets young people from lower-income families concentrate on their studies, while in America they either have to drop out of school or work their way through. This sounds like a virtue, not a vice, of the French system.[20]

Once they reach prime working age, as we’ve seen, the French are just as likely to be employed as we are—a fact that’s very much at odds with the picture of a largely idle work force often painted in U.S. news reports. The only place where the French have a serious problem—and just to be clear, it’s a very important problem—is in the low employment and labor-force-participation rates of older workers. This reflects some major policy mistakes—especially the decision, a quarter century ago, to lower to sixty the age at which workers were entitled to full pension benefits. This both encouraged early retirement and imposed large burdens on taxpayers.

So the French make mistakes. But saying, “France has mismanaged its pension policies” is very different from saying, “The French economy has been crippled by an oversize welfare state.” It’s deeply misleading to use the French example to argue against doing more to help the poor and unlucky.

Suppose we agree that the United States should become more like other advanced countries, whose tax and benefit systems do much more than ours to reduce inequality. The next question is what that decision might involve.

In part it would involve undoing many of the tax cuts for the wealthy that movement conservatives have pushed through since 1980. Table 11 shows what has happened to three tax rates that strongly affect the top 1 percent of the U.S. population, while having little effect on anyone else. Between 1979 and 2006 the top tax rate on earned income was cut in half; the tax rate on capital gains was cut almost as much; the tax rate on corporate profits fell by more than a quarter. High incomes in America are much less taxed than they used to be. Thus raising taxes on the rich back toward historical levels can pay for part, though only part, of a stronger safety net that limits inequality.

The first step toward restoring progressivity to the tax system is to let the Bush tax cuts for the very well off expire at the end of 2010, as they are now scheduled to. That alone would raise a significant amount of revenue. The nonpartisan Urban-Brookings Joint Tax Policy Center estimates that letting the Bush tax cuts expire for people with incomes over two hundred thousand dollars would be worth about $140 billion a year starting in 2012. That’s enough to pay for the subsidies needed to implement universal health care. A tax-cut rollback of this kind, used to finance health care reform, would significantly reduce inequality. It would do so partly by modestly reducing incomes at the top: The Tax Policy Center estimates that allowing the Bush tax cuts to expire for Americans making more than two hundred thousand dollars a year would reduce the aftertax incomes of the richest 1 percent of Americans by about 4.5 percent compared with what they would be if the Bush tax cuts were made permanent. Meanwhile middle-and lower-income Americans would be assured of health care—one of the key aspects of being truly middle class.[21]

Another relatively easy move from a political point of view would be closing some of the obvious loopholes in the U.S. system. These include the rule described earlier that allows financial wheeler-dealers, such as hedge fund managers, to classify their earnings as capital gains, taxed at a 15 percent rate rather than 35 percent. The major tax loopholes also include rules that let corporations, drug companies in particular, shift recorded profits to low-tax jurisdictions overseas, costing billions more; one recent study estimates that tax avoidance by multinationals costs about $50 billion a year.[22]

Going beyond rolling back the Bush cuts and closing obvious loopholes would be a more difficult political undertaking. Yet there can be rapid shifts in what seems politically realistic. At the end of 2004 it seemed all too possible that Social Security, the centerpiece of the New Deal, would be privatized and effectively phased out. Today Social Security appears safe, and universal health care seems within reach. If universal health care can be achieved, and the New Deal idea that government can be a force for good is reinvigorated, things that now seem off the table might not look so far out.

Both historical and international evidence show that there is room for tax increases at the top that go beyond merely rolling back the Bush cuts. Even before the Bush tax cuts, top tax rates in the United States were low by historical standards—the tax rate on the top bracket was only 39.6 percent during the Clinton years, compared with 70 percent in the seventies and 50 percent even after Reagan’s 1981 tax cut. Top U.S. tax rates are also low compared with those in European countries. For example, in Britain, the top income tax rate is 40 percent, seemingly equivalent to the top rate of the Clinton years. However, in Britain employers also pay a social insurance tax—the equivalent of the employer share of FICA here—that applies to all earned income. (Most of the U.S. equivalent is levied only on income up to a maximum of $97,500.) As a result very highly paid British employees face an effective tax rate of almost 48 percent. In France effective top rates are even higher. Also, in Britain capital gains are taxed as ordinary income, so that the effective tax rate on capital gains for people with high income is 40 percent, compared with 15 percent in the United States.[23] Taxing capital gains as ordinary income in the United States would yield significantly more revenue, and also limit the range of tax abuses like the hedge fund loophole.

Also, from the New Deal until the 1970s it was considered normal and appropriate to have “super” tax rates on very-high-income individuals. Only a few people were subject to the 70 percent top bracket in the 70s, let alone the 90 percent-plus top rates of the Eisenhower years. It used to be argued that a surtax on very high incomes serves no real purpose other than punishing the rich because it wouldn’t raise much money, but that’s no longer true. Today the top 0.1% of Americans, a class with a minimum income of about $1.3 million and an average income of about $3.5 million, receives more than 7 percent of all income—up from just 2.2 percent in 1979.[24] A surtax on that income would yield a significant amount of revenue, which could be used to help a lot of people. All in all, then, the next step after rolling back the Bush tax cuts and implementing universal health care should be a broader effort to restore the progressivity of U.S. taxes, and use the revenue to pay for more benefits that help lower-and middle-income families.

Realistically, however, this would not be enough to pay for social expenditures comparable to those in other advanced countries, not even the relatively modest Canadian level. In addition to imposing higher taxes on the rich, other advanced countries also impose higher taxes on the middle class, through both higher social insurance payments and value-added taxes—in effect, national sales taxes. Social insurance taxes and VATs are not, in themselves, progressive. Their effect in reducing inequality is indirect but large: They pay for benefits, and these benefits are worth more as a percentage of income to people with lower incomes.

As a political matter, persuading the public that middle-income families would be better off paying somewhat higher taxes in return for a stronger social safety net will be a hard sell after decades of antitax, antigovernment propaganda. Much as I would like to see the United States devote another 2 or 3 percent of GDP to social expenditure beyond health care, it’s probably an endeavor that has to wait until liberals have established a strong track record of successfully using the government to make peoples’ lives better and more secure. This is one reason health care reform, which is tremendously important in itself, would have further benefits: It would blaze the trail for a wider progressive agenda. This is also the reason movement conservatives are fiercely determined not to let health care reform succeed.

Aftermarket policies can do a great deal to reduce inequality. But that should not be our whole focus. The Great Compression also involved a sharp reduction in the inequality of market income. This was accomplished in part through wage controls during World War II, an experience we hope won’t be repeated. Still, there are several steps we can take.

The first step has already been taken: In 2007 Congress passed the first increase in the minimum wage within a decade. In the 1950s and 1960s the minimum wage averaged about half of the average wage. By 2006, however, the purchasing power of the minimum wage had been so eroded by inflation that in real terms it was at its lowest point since 1955, and was only 31 percent of the average wage. Thanks to the new Democratic majority in Congress, the minimum is scheduled to rise from its current $5.15 an hour to $7.25 by 2009. This won’t restore all the erosion, but it’s an important first step.

There are two common but somewhat contradictory objections often heard to increasing the minimum wage. On one hand, it’s argued that raising the minimum wage will reduce employment and increase unemployment. On the other it’s argued that raising the minimum will have little or no effect in raising wages. The evidence, however, suggests that a minimum wage increase will in fact have modest positive effects.

On the employment side, a classic study by David Card of Berkeley and Alan Krueger of Princeton, two of America’s best labor economists, found no evidence that minimum wage increases in the range the United States has experienced led to job losses.[25] Their work has been furiously attacked both because it seems to contradict Econ 101 and because it was ideologically disturbing to many. Yet it has stood up very well to repeated challenges, and new cases confirming its results keep coming in. For example, the state of Washington has a minimum wage almost three dollars an hour higher than its neighbor Idaho; business experiences near the state line seem to indicate that, if anything, Washington has gained jobs at Idaho’s expense. “Small-business owners in Washington,” reported the New York Times, “say they have prospered far beyond their expectation…. Idaho teenagers cross the state line to work in fast-food restaurants in Washington.”

All the empirical evidence suggests that minimum wage increases in the range that is likely to take place do not lead to significant job losses. True, an increase in the minimum wage to, say, fifteen dollars an hour would probably cause job losses, because it would dramatically raise the cost of employment in some industries. But that’s not what’s on—or even near—the table.

Meanwhile minimum wage increases can have fairly significant effects on wages at the bottom end of the scale. The Economic Policy Institute estimates that the worst-paid 10 percent of the U.S. labor force, 13 million workers, will gain from the just-enacted minimum wage increase. Of these, 5.6 million are currently being paid less than the new minimum wage, and would see a direct benefit. The rest are workers earning more than the new minimum wage, who would benefit from ripple effects of the higher minimum.

The minimum wage, however, matters mainly to low-paid workers. Any broader effort to reduce market inequality will have to do something about incomes further up the scale. The most important tool in that respect is likely to be an end to the thirty-year tilt of government policy against unions.

I argued in chapter 8 that the drastic decline in the U.S. union movement was not, as is often claimed, an inevitable result of globalization and increased competition. International comparisons show that the U.S. union decline is unique, even though other countries faced the same global pressures. Again, in 1960 Canada and the United States had essentially equal rates of unionization, 32 and 30 percent of wage and salary workers respectively. By 1999 U.S. unionization was down to 13 percent, but Canadian unionization was unchanged. As I discussed in chapter 8, the sources of union decline in America lie not in market forces but in the political climate created by movement conservatism, which allowed employers to engage in union-busting activities and punish workers for supporting union organizers. Without that changed political climate, much of the service economy—especially giant retailers like Wal-Mart—would probably be unionized today.

A new political climate could revitalize the union movement—and revitalizing unions should be a key progressive goal. Specific legislation, such as the Employee Free Choice Act, which would reduce the ability of employers to intimidate workers into rejecting a union, is only part of what’s needed. It’s also crucial to enforce labor laws already on the books. Much if not most of the antiunion activity that led to the sharp decline in American unionization was illegal even under existing law. But employers judged, correctly, that they could get away with it.

The hard-to-answer question is the extent to which a newly empowered U.S. union movement would reduce inequality. International comparisons suggest that it might make quite a lot of difference. The sharpest increases in wage inequality in the Western world have taken place in the United States and in Britain, both of which experienced sharp declines in union membership. (Britain is still far more unionized than America, but it used to have more than 50 percent unionization.) Canada, although its economy is closely linked to that of the United States, appears to have had substantially less increase in wage inequality—and it’s likely that the persistence of a strong union movement is an important reason why. Unions raise the wages of their members, who tend to be in the middle of the wage distribution; they also tend to equalize wages among members. Perhaps most important, they act as a countervailing force to management, enforcing social norms that limit very high and very low pay even among people who aren’t union members. They also mobilize their members to vote for progressive policies. Would getting the United States back to historical levels of unionization undo a large part of the Great Divergence? We don’t know—but it might, and encouraging a union resurgence should be a major goal of progressive policy.

A reinvigorated union movement isn’t the only change that could reduce extreme inequalities in pay. As I mentioned in chapter 8, a number of other factors discouraged very high paychecks for a generation after World War II. One was a change in the political climate: Very high executive pay used to provoke public scrutiny, congressional hearings, and even presidential intervention. But that all ended in the Reagan years.

Historical experience still suggests that a new progressive majority should not be shy about questioning private-sector pay when it seems outrageous. Moral suasion was effective in the past, and could be so again.

The Great Compression, the abrupt reduction in economic inequality that took place in the United States in the 1930s and 1940s, took place at a time of crisis. Today America’s state is troubled, but we’re not in the midst of a great depression or a world war. Correspondingly, we shouldn’t expect changes as drastic or sudden as those that took place seventy years ago. The process of reducing inequality now is likely to be more of a Great Moderation than a Great Compression.

Yet it is possible, both as an economic matter and in terms of practical politics, to reduce inequality and make America a middle-class nation again. And now is the time to get started.