The economy seems to be in great shape. The growth rate in 1994 was a brisk 4.1 percent. Unemployment has been hovering at about 5.5 percent, well below the 6 to 6.5 percent level that many economists (wrongly) consider full employment. Job growth has been so strong that President Clinton's campaign pledge to create eight million jobs may well be fulfilled in the third year of his first term. Corporate profitability has reached postwar records. The stock market is booming. Inflation is nowhere to be found except in the imagination of central bankers and bond traders.
We are also witnessing a revival of productivity growth and a celebration of America's renewed competitiveness, as U.S. products close the quality gap with imported goods and unit labor costs continuously fall relative to those of other advanced countries. Some analysts feel that our productivity growth is so strong and the promise of computerization so great that we are on the verge of a new, golden economic age.
There's only one problem with this celebration: The living standards of the broad middle class have remained in continuous decline despite the robust aggregate performance. This recovery has demonstrated that improved competitiveness, productivity, and overall economic growth do not necessarily translate into improved incomes for most families.
DOWN THE UPSWING
In the 1980s, family incomes were stagnant. Only the continued rise in two-earner families prevented a significant erosion in middle-class incomes in the face of declining real wages for men. Meanwhile, poverty grew at the bottom end while income and wealth soared among the top 1 percent. In a decade, all of the postwar progress toward greater equality was reversed.
In the 1990s, income trends have been even worse than in the 1980s. Median family income fell in every year from 1989 to 1993, the last year for which the Census Bureau has released family income data. This was the only time in the postwar period that incomes fell four years in a row and the first time that incomes fell during the first two years of an economic recovery. By 1993, the median family's income was down $2,700, or 7 percent, from its 1989 level of roughly $40,000. It will take five to eight years of steady growth to recoup these income losses--not a likely event. The economy also set records for the breadth of income deterioration as incomes steadily fell among the bottom 80 percent of families from 1989 through 1993.
While complete data on household incomes are not yet available, it is very unlikely that these trends have improved since 1993. The real median weekly earnings of men who work have been falling since 1989, including from 1993 to 1994. (See figure 1.) Even women's median earnings, one of the few bright spots in the 1980s, dropped in 1994. The available wage data from early 1995 suggest that wages for both men and women have continued to lag behind inflation over the last 12 months.
Part of the problem is that the rate of return to capital--profits, interest, dividends--has surged to postwar records and thereby dampened wage growth. The larger part of the problem is that the continuing growth of wage disparities has demonstrated that even though average productivity and average wages may be increasing, the wages for the vast majority can still be falling. (See "Working Harder for Less" )
Productivity (output per hour worked) has grown 25 percent since 1973, a weak performance by historical standards, but quite enough (about 1 percent a year) to enable all workers to achieve real wage growth. In fact, average hourly compensation has roughly tracked average productivity, at least until recently. But that "average" is composed of a small part of the workforce at very high pay whose earnings are increasing while the bottom 80 percent see their pay falling. Hourly compensation of the median worker was stagnant over the 1970s and 1980s and has been declining in the 1990s. The median male has seen his hourly compensation fall 1 percent annually since 1979.
CAUSES OF WAGE INEQUALITY
Growing wage inequality is not the inevitable consequence of technological progress. Most wages have been falling despite rising productivity for one simple reason. Workers--union and nonunion, college- and noncollege-educated, white-collar and blue-collar--have lost bargaining power relative to their employers. Corporate America today has the power to respond to increased domestic and global competition by cutting labor costs. Public policies have generally accommodated or reinforced this effort, by failing to pursue full employment, fair trade, higher minimum wages, or renewal of the labor movement.
The failure of rising productivity to raise living standards is something new in the postwar American experience. As Lawrence Chimerine of the Economic Strategy Institute points out, virtually every macroeconomic model has been predicting far faster wage growth than ever materializes. At the peak of an economic recovery, one expects workers to be able to approach their bosses with propositions such as "I need a 5 percent wage increase or I'll get another job." This has not been happening in the current business cycle, even among the managerial-professional workforce whose unemployment is only 2.5 percent. The steady growth of temporary employment, which accounts for about 15 percent of job growth in this recovery, and the continued use of other forms of contingent work indicate that broad segments of the workforce have little power to shape their terms of employment. According to the Bureau of Labor Statistics (BLS), early 1995 saw the lowest compensation growth in two decades. This must be a curious phenomenon for our central bankers who have been doing their best to slow the economy lest wages accelerate and inflation take off.
In our recent book, The State of Working America, 1994-95, Jared Bernstein and I attribute growing wage inequality primarily to two clusters of factors, each of which can explain about one-third of the phenomena.
Globalization and Service Employment. The combined effect of globalization (including the trade deficit, low-wage competition, immigration, and foreign direct-investment trends) and the continued expansion of employment in the lowest-paid portion of the service sector together explain at least a third of the wage problem. BLS attributes 25 to 30 percent of the growth of wage inequality to sectoral shifts in the pattern of employment alone. In addition, import pressure has lowered wages for many middle- and low-wage industrial workers, and undercut the wages of workers generally. Only a few ideologues in the economics profession, particularly near important congressional votes on trade policy, deny that expanding trade significantly depresses wages for noncollege-educated workers.
These factors are all influenced by policy decisions--by omission or commission. For example, the shift to service employment would be far less consequential if health care and pension benefits were not so tightly linked with one's particular job but were universally provided. A service-sector job would not be a sentence to wage reductions if America had stronger unions, or fuller employment. The wage disparities between manufacturing and services are far smaller in other industrial countries where market forces have less free rein.
An Institutional Collapse. The other prime culprit is the weakening of labor market standards and institutions that empower workers. The erosion of the minimum wage, a nearly 30 percent drop in value from its 1979 level, has particularly affected wage levels of working women. Deunionization and the weakening of union bargaining power has had a particularly adverse effect on noncollege-educated men. This erosion of wage-setting institutions can explain roughly another third of the recent wage deterioration among workers. Other factors that are harder to quantify have also had an impact: deregulation, privatization, and the growth of contingent work--and these also reflect the policy climate.
Big government is a popular scapegoat for the erosion of living standards. The Republican House Majority Leader, Dick Armey, has claimed that people are working harder and getting less because the government is taking a bigger bite out of their paycheck. In reality, it is pre-tax pay that has collapsed. According to the Congressional Budget Office, the effective federal tax rate on the middle class was stable or declining in the 1980s.
Moreover, the projections for 1995, which include all of the effect of the early Clinton budgets, show tax rates declining among the bottom four-fifths while rising among the best-off fifth. Even the conservative National Tax Foundation reports stable tax rates over the 1980s. If anything, government was stable or shrinking--as measured by shares of total employment or expenditure or by regulatory costs--since 1979 when the wage implosion originated. Unfortunately, according to a recent Business Week poll, the public seems to believe the big government, high taxes story. This only proves that "Big Lies Frequently Told in Many Forums" is still an effective communications strategy.
A related, and disingenuous, claim is that large fiscal deficits are the cause of income problems. At most, deficits may be a factor if they somehow eroded investment and thereby productivity growth. But, we have already seen that the productivity growth that did occur was not shared with most workers. Even more damning is that the productivity slowdown preceded the era of large structural fiscal deficits by nearly ten years. The claim that fiscal deficits have shrunk workers' paychecks is without foundation.
Among economists and editorial writers the most widely cited explanation of widespread wage deterioration and growing wage inequality is that technological change has raised the demand for skills so that the wages of more highly skilled and educated workers have been bid up and the wages of other workers have suffered because there is less demand for them. Curiously, economists also acknowledge that there is very little direct evidence for the technology story and that their belief is based on eliminating competing explanations and putting a technology label on what they cannot explain. (See David Howell, "The Skills Myth," TAP, Summer 1994.)
It is implausible to rely on an explanation that describes three-fourths of the workforce--those whose wages have been declining since 1979-as "low-skilled." It is hard to imagine the "bidding up of skilled workers' wages" as an apt description of a process that has lowered the real wages of male white-collar and college-educated workers continuously since 1987. To believe the technology story, we must also believe that a technological revolution is dramatically affecting our wage and employment structures but somehow fails to raise productivity growth enough to improve living standards. And, of course, technological change in the workplace is not new. For the technology story to make sense, technology has to have had a greater impact on labor markets in the 1980s and 1990s than in earlier periods, either through a faster rate of introduction or due to new types of technologies. My recent research with Jared Bernstein has shown that no such "acceleration" has taken place.
Absent strong countervailing forces, we should expect wage inequality to grow and wages to continue to fall for large segments of the workforce. Conceivably, productivity growth could accelerate so significantly that even smaller shares of a larger pie would mean more pie for all. But this hope rests primarily on conjectures of an eventual payoff to computerization so huge that it would offset all the pressures leading to increased inequality. Put me down as a skeptic.
A second possibility is for so many new college graduates to enter the labor market that college wages fall relative to those of noncollege-educated workers. This is an unlikely scenario because, even though a greater proportion of young adults are completing college, the size of the college-going-age cohort is shrinking, limiting the absolute growth of college graduates. A more plausible scenario is that wage inequality continues to grow or, at best, stalls.
BACK TO POLITICS
In any event, nothing in the Republican economic program will reverse or forestall continued wage erosion and the accompanying pressures on family incomes. The macroeconomic effect of balancing the budget and reducing the deficit will be to slow economic growth over the next ten years. Any long-term effect of deficit reduction on productivity is uncertain, slight, and at least a decade away. Clinton's Council of Economic Advisers estimated the payoff for their very sizable deficit package at two-and-a-half percent more consumption after twenty years, without considering the income losses from the initially slower growth. The Republican growth program of deregulation and tax reduction will, at best, increase jobs and productivity slightly but do nothing to upgrade the quality of jobs or to translate productivity growth into broad-based wage gains.
The issue of wage growth will thus be with us for a long time, and it is time the nation and politicians started focusing on it. We will not make progress unless we can sustain the necessary conditions for wage growth. That means the Federal Reserve Bank must adjust its views of the "natural" rate of unemployment and cease creating a recession at the first signs of robust growth. Public investments in skills, infrastructure, and technology are in order to maintain productivity growth. The character of growth must change as well. Public policy needs to inhibit financial speculation and corporate strategies driven by short-term profit, in favor of production strategies that rely on broadly skilled workers and widely shared rewards. Trade policy, including international labor and environmental standards, must be used to shape global capital and trade flows and inhibit the "race to the bottom."
Most of all, we need to re-empower workers and change the dynamics of wage setting. Part of the solution is a significantly higher minimum wage, which will greatly help the bottom half of the income distribution. Another part is to enable workers to choose unionization and collective bargaining, especially in the service sector where a wage push from below may be our only hope for forcing employers to become more efficient and to generate productivity growth. New forms of representation, such as works councils and occupational associations, may be needed to achieve a wider empowerment of workers. Greater economic security through training and education, along with greater health security and pension mobility are valid policy in their own right, and can also help change the dynamics of wage bargaining.
Understanding these income problems is key to understanding the current shredding of our social fabric as well as the elections of 1992 and 1994. Moreover, policies to regenerate broad-based wage growth may be the key to political success in 1996 and beyond. In fact, this theme is already central in the rhetoric of Secretary of Labor Robert Reich and House Minority Leader Richard Gephardt.
However, before we can pursue appropriate remedies, we first need broad agreement that the erosion of living standards is a problem-indeed, the problem-that can be addressed through policy. That recognition, unfortunately, is still almost totally absent from political discourse. o
Figure 1. Median Hourly Wages for Men and Women, 1989-1995
Median weekly earnings of full-time working men have fallen steadily since 1989, while women's wages--a bright spot in the 1980s--dropped in 1994.
Source: Authors' analysis of Bureau of Labor Statistics data.
Figure 2. Working Harder for Less
It's true that productivity and average wages are increasing, but the wages for the vast majority of Americans are still falling.
Source: Authors' analysis of Bureau of Labor Statistics data.