THE WAGE SQUEEZE
For years Craig Miller had been a sheet-metal worker at a major airline. After he lost his job in 1992, he and his wife -- parents of four kids -- had to scramble. Craig took on two lower-paying jobs and started a small sideline business. His wife worked nights as a stock clerk. They were patching together, counting his business, four part-time jobs and they were still earning less than half Craig's previous paycheck.
"Sure we've got four jobs," Craig told a reporter. "So what? So you can work like a dog for $5 an hour?"
The Miller family saga is hardly unique. Since the mid-1970s, more and more U.S. workers and their families have been suffering the wage squeeze, enduring steady downward pressure on their hourly take-home pay. The wage squeeze has afflicted not merely the unskilled and disadvantaged but the vast majority of U.S. households, not merely the poor and working class but the middle class as well. Most people in the United States used to be able to look forward to a future of steadily rising earnings. Now they have to race merely to stay in place.
The wage squeeze has even broader consequences. It not only pinches workers and their immediate families. It sends tremors through entire communities, eroding their stability, ripping their social fabric. The frustration and anger it provokes begins to attack the body politic like a plague, spreading virulent strains of cynicism and discontent, of disaffection from government and hatred toward "others" like immigrants who are often blamed for the scourge. Many observers in the United States are inclined to turn their heads, viewing falling wages as somebody else's problem. But the effects are too far-reaching, too extensive. It won't work to play the ostrich, sticking one's head in the sand. The sand is eroding all around us.
Back to the 1960s
The public receives mixed signals about the wage squeeze. On the one hand, more and more observers have taken note of the vise closing around workers' earnings -- citing the pressure to work longer hours, the "disappearing middle class," the increasingly elusive American Dream, the mounting gap between the rich and the poor. Personal stories of declining fortunes abound. Statistical studies of stagnant earnings and soaring inequality have become a growth industry. In my research for this book, finding journalistic accounts and scholarly analyses of the wage squeeze was as easy as following the trail of Newt Gingrich's newfound notoriety.
By late 1995, as I was completing the manuscript, the issue was becoming inescapable. Business Week, often a leader in tracking changes in the economic climate, devoted a cover story to "The Wage Squeeze" in July 1995. Surveying the atmospheric conditions they reported:
Four years into a recovery, profits are at a 45-year high, unemployment remains relatively low, and the weak dollar has put foreign rivals on the defensive. Yet U.S. companies continue to drive down costs as if the economy still were in a tailspin. Many are tearing up pay systems and job structures, replacing them with new ones that slice wage rates, slash raises, and subcontract work to lower-paying suppliers.
"Although the problem [of slumping wages] has been plaguing Americans for years," wrote New York Times economics reporter Louis Uchitelle that same summer, "it is just now rising to the level of a major campaign issue." "Nearly everyone by now knows the situation," economic columnist William Greider wrote in November 1995, "either from the headlines or from their own daily lives: the continuing erosion of wage incomes for most American families." Commenting on yet another twelve months of stagnant wage growth, Robert D. Hershey Jr. wrote in late 1995: "The frustration and insecurity that have resulted are expected to play a major role in shaping next year's Presidential race as politicians of both parties try to portray themselves as the best choice to provide economic growth that will benefit the middle class."
On the other hand, many pundits, economists and business leaders seem not to lament the wage squeeze but rather to praise it. Instead of wringing their hands about working households' living standards, many express relief about the moderation of wage pressure on prices and profits -- a trend they hope will dampen inflationary pressure, keep U.S. firms competitive in global markets, and protect small enterprises against business failure. When journalists report monthly data on workers' hourly earnings, they are much more likely to celebrate wage moderation or decline than to worry about its consequences for the millions who depend on that labor income.
Take the New York Times' report in April 1994 on real wage trends in the first quarter of the year. Noting that nominal wages and prices had grown at roughly the same rates, leaving real wages flat, the story appeared to welcome this "relatively benign reading on wages and benefits...": "American workers are obtaining less in pay and benefit increases from employers these days...," with the result that "...price pressures remained subdued." The reporter observed hopefully that "bond prices rallied at the news." Nowhere in the story did he wonder how workers themselves might regard these "relatively benign" developments.
So there are, indeed, two sides to the news about wages. "The good news is labor costs are under control," economic forecaster Michael Evans put it in 1992. "The bad news is that employees are broke."
More often than not, however, the good news for business seems to blot out the bad for nearly everyone else. I was recently struck by the prevalence of these priorities at a conference about macroeconomic policy in Washington, D.C. in the spring of 1994. At lunch we heard from a Presidential economic adviser. A distinguished scholar, the speaker had been an economic liberal, more to the left than to the right of the mainstream of economic discourse. In a recent policy book, he had expressed concern about a polarized society in which the economic extremes of the 1980s had made the rich richer and set the rest adrift.
The economist lauded the progress of the economy in the spring of 1994 and the continuing signs, in the Administration's view, of a decent economic recovery. He noted with approval the evidence of (modest) growth in consumer spending, investment, and exports. He applauded the Federal Reserve's and the markets' continuing restraint in interest rates and pointed proudly to the tepid pace of inflation. He projected 1994 real wage growth at zero percent.
What is notable about this presentation is what was not said. A projection of zero real wage growth, but no reflections on the hardships experienced by ordinary working people. No lament about the twenty-year decline in real earnings. And this from a key economic adviser to the president who had promised, in his initial economic message to Congress, that "our economic plan will redress the inequities of the 1980s."
This widespread inattention to workers' living standards even shows up in the preferences of government data collectors. For decades, since the end of the Depression and the spread of the union movement, the U.S. Bureau of Labor Statistics had kept track of the living standards of the average American worker with published data on spendable earnings. The series measured the real after-tax value of workers' weekly take-home pay. But in 1981 the Reagan Administration discontinued the index, citing conceptual and measurement problems. They proposed no replacement, leaving us without any official series intended specifically to monitor the effective purchasing power of workers' earnings.
Had the government data apparatchiki actually cared about illuminating the trends in workers' income, the statistical problems they cited would not have been especially difficult to overcome -- hardly so vexing that they warranted dropping this kind of series altogether. But their priorities lay elsewhere. At more or less the same time as the discontinuation of the weekly spendable earnings series, the Bureau of Labor Statistics, reflecting the Reaganites' ever-extending solicitude for the needs of business, was expanding the range and variety of its employment cost indices, tracking the hourly costs to corporations of their wage-and-salary employees. As a result, in recent years, corporations need merely dial the phone to get up-to-date data about changes in labor costs faced by them and their competitors.
More than a decade ago, in response to this change in priorities, my collaborators Samuel Bowles, Thomas E. Weisskopf, and I proposed an alternative version of the spendable earnings index, with modifications designed to address each of the specific problems raised about the traditional indicator. Where the traditional series on weekly earnings had conflated movements in hourly earnings and changes in hours worked per week, we proposed relying on a much simpler index of hourly earnings. Where the traditional series had relied on a somewhat implausible adjustment for the taxes paid by the "average" worker, we suggested a much more immediate and direct calculation. We called our proposed alternative an index of real spendable hourly earnings.
Our proposal was graciously published in the Bureau of Labor Statistics official journal, but, hardly to our surprise, the Reagan Administration ignored our advice, persisting in providing no official record of trends in workers' take-home pay. So we have continued ourselves to maintain and update what we consider to be the most salient indicator of workers' earnings.
Our index of real spendable hourly earnings provides a straightforward measure of the real value of the average production or nonsupervisory worker's take-home pay. "Production and nonsupervisory" workers, as they're defined in the official BLS surveys of business establishments, comprised 82 percent of total employment in 1994. They represent that group in the labor force that is most clearly dependent on wage and salary income. They include both blue-collar and white-collar workers, both unskilled and skilled. They cover not only laborers and machinists but also secretaries, programmers and teachers.
I focus primarily on these "production and nonsupervisory" employees at least partly to avoid distortions in the data from the huge increases during the 1980s in the salaries of top management -- a group covered by the earnings data for the other fifth of employees excluded from our measure, a category called "nonproduction or supervisory" employees. In further discussion in this chapter and throughout the rest of the book, in order to avoid the cumbersome terminology used by the BLS, I shall refer to the "production and nonsupervisory" category in the establishment data as production workers and to the other grouping as supervisory employees, respectively.
Spendable hourly earnings measure the average production worker's hourly wage-and-salary income minus personal income taxes and Social Security taxes. These earnings are then expressed in constant dollars in order to adjust for the effects of inflation on the cost of living. They measure how much per hour, controlling for taxes and inflation, the average production worker is able to take home from his or her job.
Figure 1.1 charts the level of average real hourly spendable earnings for private nonfarm production employees in the United States from 1948 to 1994.
The data show a clear pattern. The average worker's real after-tax pay grew rapidly through the mid-1960s. Its growth then slowed, with some fluctuation, until the early 1970s. After a postwar peak in 1972, this measure of earnings declined with growing severity, with cyclical fluctuation around this accelerating drop, through the rest of the 1970s and 1980s. The average annual growth of real spendable hourly earnings reached 2.1 percent a year from 1948 to 1966, slowed to 1.4 percent between 1966 and 1973, and then dropped with gathering speed at a shade less than minus one percent per year from 1973 to 1989.
Despite the recovery from the recession of 1990-91, real spendable hourly earnings were lower in 1994 than they had been in the business-cycle trough of 1990. Even though the economy had been growing steadily for three years from the bottom of the recession, they continued to decline at an average annual rate of-0.6 percent from the peak in 1989 through 1994.
By 1994, indeed, real hourly take-home pay had dropped by 10.4 percent since its postwar peak in 1972. More dramatically still, real spendable hourly earnings had fallen back to below the level they had last reached in 1967. Growing massively over those nearly three decades, the economy's real gross output per capita in 1994 was 53 percent larger than it had been in 1967, but real hourly take-home pay was four cents lower. Referring to these trends since the early 1970s as "the wage squeeze" is polite understatement. Calling it the "wage collapse" might be more apt.
These harsh winds have continued to blow through the recent recovery. Most economic meteorologists have described them in similarly cloudy terms. But a few have recently tried to present a sunnier weather report.
In one highly visible piece in late 1994, for example, the New York Times published a long news story beginning on its front page. Sylvia Nasar, the Times reporter, broadcast a considerably more sanguine view about wage trends: "it is practically gospel that the growing American economy cannot deliver the higher pay that American workers want," she wrote. But she claimed that wage changes during the early 1990s appeared to suggest a turnaround, with the majority of new jobs paying above-average wages. "As a result," she concluded, "average hourly pay for all employees, adjusted for inflation, is slowly rising."
The source of Nasar's discrepant conclusions was not hard to find. Unlike all the data reviewed thus far in this chapter, which cover production employees -- accounting for roughly four-fifths of the wage-and-salary workforce -- Nasar was looking at wage trends for all workers. These data cover those at the top of the earnings distribution, including top-level executives whose total compensation has continued to soar straight through the mid-1990s. Those who have long pointed to the wage squeeze have never denied that the top 10 to 20 percent of the earnings distribution has fared much better than everyone else. If you mix together those in the middle and bottom with those at the top, you're bound to get a different and ultimately misleading story. Nasar's story was effectively demonstrating a penetrating glimpse into the obvious -- that supervisory employees have continued to enjoy rising real hourly compensation.
In his recent book Values Matter Most, commentator Ben J. Wattenberg makes the same mistake. Hoping to create the space for his argument that we should concentrate on social values, not the economy, he seeks to cast doubt on the economic pressures cited by many. He notes that many highlighting the wage squeeze focus on real earnings series for production and nonsupervisory workers. He argues that this series gives an "inaccurate" picture because it "concerns cash only, ignoring benefits." Then, almost quicker than the eye can blink, he shifts our attention to the same series Nasar reported, the index for total employee compensation per hour. "That line," he observes hopefully, "is clearly trending upward...," lending support to his ultimate conclusion that "our economic situation is somewhat less than grievous." But while the eye was blinking, Wattenberg switched to a series that differed from the first in two respects, not just one: including benefits, it traced total compensation; and, tracking all workers, it included those at the top who have been feeding at the trough. As I show in the Appendix to this chapter, just including benefits in our series, while continuing to focus on production and nonsupervisory workers, tells almost exactly the same story as earnings without benefits. Whether we look at earnings or full compensation, the wage squeeze for production workers remains severe.
For the vast majority of workers, then, these have been hard times indeed. In 1994, the average production employee working thirty-five hours a week and fifty-two weeks a year was able to take home about $16,833 after taxes, barely above the official poverty standard for a family of four. An earlier generation had expected that their earnings would rise over their working lifetimes and that their children could anticipate higher living standards than their own. For the past two decades, however, more and more workers have had to adjust their expectations, reconciling themselves to toil at what are sometimes derisively called "McJobs."
One Michigan woman, talking in a pollster's focus group in the early 1990s about deflated expectations, lamented:
I think about when I was married, a week of groceries cost me $13 and my husband thought that was entirely too much money to spend for a week's groceries. Now I spend $150. I feel like I'm always running -- and this big snowball is behind me getting bigger and bigger -- and just trying to keep it from running over me.
Another focus group participant talked about shifting expectations across generations. "[Our kids]'ll have to be good to us if they want to have a home to live in, because the only way they'll get one is if we will them ours. They're never going to be able to buy a house."
You don't have to organize your own focus groups to get a strong whiff of these kinds of economic concerns. Recent national polls repeatedly reveal such fears about economic pressure and the cloudy future for this and future generations. In a 1992 Gallup poll, for example, more than three-fifths said they were dissatisfied with "the opportunity for the next generation of Americans to live better than their parents"; 58 percent were dissatisfied with the "opportunity for a poor person in this country to get ahead by working hard." In a June 1993 LA Times/CNN poll, 39 percent of participants described their personal finances as "shaky," while more than half -- 51 percent -- said they "expect the next generation of Americans will have a worse standard of living than the one we have now." Even though the economy was well into its recovery, in a November 1993 LA Times/CNN poll two-thirds reported that job security was "worse for Americans now, compared to two years ago" and 53 percent that they felt this "greater job insecurity will occur over the long term, for many years." Even further into the recovery, a March 1994 New York Times poll found that two-fifths of respondents expressed "worry" that during the next two years they might be laid off, required to work reduced hours, or forced to take pay cuts. Nearly two-fifths also reported that in order "to try to stay even financially" during the last two years they had had to work overtime or take on extra jobs. In a March 1995 Business Week/Harris poll, people were asked whether "the American Dream...has become easier or harder to achieve in the past 10 years." Two-thirds answered that it has become "harder." Participants were also asked if it would be "easier or harder to achieve in the next 10 years." Three-quarters chose "harder."
A Crowded Boat
Andrew Flenoy, a twenty-one-year-old living in Kansas City, did better in 1994 than many, holding down a steady job paying a cut above the minimum wage. In fact, he had even enjoyed some recent promotions, rising at a food catering firm from dishwasher to catering manager. Through that sequence of promotions, however, his earnings had increased from $5.50 an hour to only $6.50 an hour -- the equivalent of only about $12,000 a year working fulltime year-round. Whatever satisfaction he had enjoyed from his promotions had quickly paled. "Now he is tired of the burgundy and black uniform he must wear," a reporter concluded, "and of the sense that he works every day from 6 A.M. to 2 P.M. just to earn enough money so that he can come back and work some more the next day." "My resolution for 1994," Flenoy remarked, "is that if nothing comes along, I'll relocate and start from scratch somewhere else."
Flenoy attended only a semester of community college after high school and suffered the additional employment disadvantage of being African American. Many are inclined to assume, indeed, that the wage squeeze has mostly afflicted the young, the unskilled, and the disadvantaged.
Although some have suffered more than others, however, a much wider band of the working population has been caught in the vise. For most Americans, the wage squeeze has been a profoundly democratizing trend.
Indeed, the data on the breadth of the wage squeeze seem finally to have persuaded skeptics not normally known for their empathy with workers. Recently confronted with some of these data, for example, Marvin Kosters, a well-known conservative economist at the American Enterprise Institute who had earlier challenged reports about trends toward growing inequality, admitted surprise at the variety of subgroups affected by wage erosion. "It's really quite amazing," he acknowledged The data would scarcely seem "amazing," of course, to those who've been directly feeling the pinch.
In order to assess the breadth of the wage squeeze, we need to turn to data from household surveys, which unlike the establishment surveys afford considerable detail on workers' personal characteristics. We can look at trends in real hourly earnings between 1979 and 1993 for a variety of different groups in the private nonfarm workforce, since it is trends in the private sector with which I am most concerned in this book.
Looking at this universe, we find that real hourly earnings for all private nonfarm employees, including those at the top, remained essentially flat from 1979 to 1993 -- barely rising from $11.62 to $11.80 (in 1993 prices). (Government workers did somewhat better.)
parBut we know that those at the top did fairly well. The more telling comparison looks at real wage trajectories for the bottom four-fifths of the real wage distribution and for the top fifth. As anticipated from the data for production workers reviewed in the previous section, it was the bottom 80 percent that experienced actual real wage decline, with the 1993 level dropping by 3.4 percent below the 1979 figure. For the top 20 percent times were not so harsh; they enjoyed a healthy rate of increase, with their real hourly earnings rising by 1993 to almost three times those for the bottom four-fifths.
We can also compare workers by race and ethnic origin. Looking at workers in the bottom 80 percent of the overall wage distribution, it is true, not surprisingly, that African Americans and Hispanics fared less well than whites. But even among whites in the bottom 80 percent, real hourly earnings dropped by nearly 3 percent. (Of course, a much larger percentage of African Americans and Hispanics were situated in the bottom four-fifths of the wage distribution than of whites.) Not just the disadvantaged but the advantaged racial group joined the wake.
Looking at wage trends by gender, we find a major difference in the impact of the wage squeeze. While male workers in the bottom 80 percent of the distribution experienced devastating declines in their real hourly earnings -- facing a decline of close to 10 percent -- women workers in the bottom 80 percent enjoyed modest real wage growth, with a total increase over the full period of 2.8 percent. Despite these gains, however, women's wages still lagged substantially behind men's. In 1993, the median female hourly wage had reached barely more than three-quarters of the median male wage, at 78 percent. Women were gaining on men, to be sure, but their gains occurred primarily because real male wages were plummeting, not because real female earnings were themselves growing rapidly. Indeed, almost three-quarters of the decline in the wage gap between men and women from 1979 and 1993 can be attributed to the decline in male earnings -- a trend which undoubtedly contributed to the widespread frustration which many males have apparently been feeling and venting.
A final comparison looks at the experience of workers with different levels of education. It was the bulk of workers on the bottom, those with less than a college degree, who experienced actual wage decline. Only those with a college degree or better were able to gain some measure of protection against the unfriendly winds. And the most recent trends have been harsh even for a large number in that group. From 1989 to 1993, for example, even male workers with just a college degree, but no postgraduate education, were hit with declining real earnings.
Table 1.1 pulls together these separate tabulations for different groups of workers. The wage squeeze has caught a huge proportion of U.S. workers in its grip.
In better times, of course, workers in a pinch often pulled up stakes and migrated in search of greener pastures -- in Andrew Flenoy's words, "to relocate and start from scratch somewhere else." But the greener pastures have mostly turned brown. New York Times reporter Louis Uchitelle tells the story about workers in Peoria, Illinois, where layoffs and givebacks at Caterpillar had cast long shadows over the local economy:
Today the adventurous search for opportunity is no longer rewarding. For generations, Americans migrated -- going West, so to speak -- when jobs in their communities became scarce or failed to pay well. But income stagnation is a nationwide phenomenon. Migration has become futile. Peorians, for example, uprooted themselves by the thousands in the early 1980's, when recession and then massive layoffs at Caterpillar and the numerous local companies that supply Caterpillar pushed the unemployment rate here above 16 percent. By the late 1980's, they were trickling home again.
"When they got to Oklahoma and Texas, they found that the promise of good wages was a lot of talk; they worked hard and had little to show for it," said David Koehler, executive director of the Peoria Area Labor Management Council. "Now, many have come home to jobs that pay less than they once earned, but they have returned because this is where their families are to help them."
Some readers may be inclined to view the wage squeeze as par for the course at the twilight of the twentieth century. The world economy is becoming more and more tightly integrated. Developing countries, where wages are much lower than the advanced economies, have been expanding their exports. Low-wage import competition has been intensifying. Isn't wage pressure in the advanced economies to be expected?
There is no denying that import competition from lower-wage developing countries has grown more intense over the past twenty years or more. But it does not necessarily follow, for a variety of reasons we shall explore in later chapters, that workers in the advanced economies must inexorably face the wage crunch as a result.
Quite to the contrary. In fact, the most striking conclusion that emerges from comparing wage trends in the advanced countries is how isolated, how relatively unique has been the U.S. experience.
Careful compilations by the U.S. Bureau of Labor Statistics allow us to compare wage trends across twelve of the leading advanced economies -- including the G-7 powers of the United States, Germany, Japan, France, Italy, the United Kingdom, and Canada as well as five other smaller European countries (Belgium, Denmark, Norway, the Netherlands, and Sweden). Their data provide comparable information on trends in real hourly compensation for all manufacturing employees, with compensation deflated by the consumer price index for each country to provide an insight into trends directly affecting workers' living standards. I look here at the period from 1973 to 1993, the most recent year for which the data were available at the time of writing.
This measure matches the series for real spendable hourly earnings in the United States, presented above in Figure 1.1, with three differences. The comparative numbers are before-tax rather than after-tax, and focus just on compensation in manufacturing, rather than the much larger nonfarm private sector. And they include all employees, not merely production workers.
By this measure, real hourly compensation for all manufacturing employees in the United States was fiat rather than collapsing in the period between 1973 and 1993. It barely changed over that period, rather than declining substantially as for the data presented in Table 1.1. The principal reason that this index of hourly wages does not show decline is that it includes nonproduction workers as well as production employees and this group at the top, as the data on the top 20 percent in the previous section suggest, was the one group whose wages stayed ahead of inflation over the past two decades. (The difference in coverage between the manufacturing and private nonfarm sectors matters less since trends in the two sectors were roughly comparable over this period; and Appendix A shows that before-tax and after-tax measures move closely together.)
If by this measure, real hourly compensation in manufacturing was roughly fiat in the United States between 1973 and 1993, how did workers fare in the other eleven advanced economies?
Figure 1.2 allows us to pursue this comparison. It presents the average annual percent change in real hourly compensation for all manufacturing employees in the United States (on the far right) and in eleven other advanced economies (arranged in alphabetical order). Wage stagnation in the United States stands out like a sore thumb. It is the only country with wage change close to zero. Only two other countries -- Canada, which feels the wage competition from its near North American neighbor, and Denmark -- feature wage growth rates less than 1.5 percent a year. Workers in Japan and Germany, our two major trading competitors, fared markedly better than U.S. workers, with real wage growth at 2.2 percent and 3.1 percent respectively.
Indeed, the average for the other eleven countries altogether is 2.1 percent per year, seven times more rapid than for the United States over the same period. Import competition from low-wage developing countries may have been intensifying, but workers in other advanced economies seem to have escaped its wrath much more effectively than workers in the United States.
Later chapters will explore some of the reasons for this huge discrepancy in wage growth between the United States and most other advanced countries and will consider the possibility that many of the other advanced economies paid a substantial price for their more rapid wage growth with relatively higher unemployment rates.
But one possible explanation deserves immediate attention. Perhaps wage growth in the United States has been relatively slow because U.S. wages have historically been so high compared to our advanced competitors and, therefore, competition with those other advanced economies has forced U.S. corporations into tough bargaining with their employees.
That factor may once have weighed heavily in the United States, but it no longer applies. By 1994, compared to the other countries featured in Figure 1.2, the United States no longer paid its employees top dollar. When we look at average hourly compensation for production workers in manufacturing, converted by exchange rates to U.S. dollars, we find that the United States ranked only eighth among the twelve countries featured in the graph, and was ahead of only Canada, France, Italy, and the United Kingdom among the twelve. Japanese manufacturers, which have competed so successfully against their U.S. competitors, paid their employees 25 percent more than did U.S. manufacturing firms. Average hourly compensation in the United States in 1994 was $17.10, while, for example, the average for the countries of the European Union, weighted by their share of U.S. trade, was $19.47 Figure 1.3 provides a graphic view of this comparison.
Nor is this an especially recent phenomenon. In 1980, hourly compensation for U.S. manufacturing workers was lower than in six of the other twelve advanced economies; in 1975, it was lower than in four others.
More dramatically still, although the comparisons are difficult to make with precision, the United States appears to be the only advanced country in which lower paid workers have actually suffered absolute declines in real earnings over the past couple of decades. Harvard labor economists Richard B. Freeman and Lawrence E Katz survey the data carefully and "conclude that less educated and lower-paid American workers suffered the largest erosion of economic well-being among workers in advanced countries. The result of this erosion was that U.S. workers on the bottom of the wage distribution hit rock bottom. Based on his own assessment of the data, Freeman reports that "low-paid Americans have lower real earnings than workers in all advanced countries for which there are comparable data -- which is due largely to the fall in their real earnings."
One might imagine that the United States gained competitive advantage over the past twenty years because its real wages grew so much more slowly than in the other leading economies. Not so lucky. Between 1973 and 1989, real hourly manufacturing compensation in Japan grew eight times more rapidly than in the United States, but this did not prevent Japanese firms from knocking the socks off their U.S. competitors in global markets. In 1973, both countries enjoyed roughly balanced trade, with exports approximately equal to imports and the net balance of trade in goods and services close to zero. By 1989, however, the United States was running a trade deficit with the rest of the world of $77 billion while the Japanese were enjoying a healthy trade surplus of $13 billion And this limp U.S. trade performance was not limited to the comparison with the striking case of Japan; we have been running trade deficits with Western Europe, where wage growth has also been rapid, for many years as well.
Most U.S. workers have been experiencing a wage squeeze, with real hourly take-home pay in 1993 falling back to the levels of the mid-1960s. This wage collapse has affected not only the unskilled and disadvantaged but a remarkably wide swath of U.S. employees. And the vise has tightened on workers' earnings much more severely in the United States than in other advanced economies.
Those caught in the vise have no illusions about the consequences. A union activist in the continuing labor struggles at Caterpillar in Illinois looked down the road toward the turn of the millennium and saw hardship: "If we don't put an end to this drift of the country to drive wages down," he said, "there's no future for my three boys. There will be an upper class and a lower class, and I know where [my boys] will be." A middle-class Michigan resident echoed this view: "Everybody is going to be either very rich or very poor. There's going to be the rich in their little towers, and there's going to be everybody else floundering around trying to survive." A 48-year-old Milwaukee woman, laid off in June 1995 after seventeen years on the job, wondered when the wage pressure from corporations will end. "How far can this go," she asked, "before they ruin everything?"
APPENDIX: MEASURING THE SQUEEZE
However dramatic the evidence of the wage squeeze presented in Chapter 1, the basic series on real spendable hourly earnings, as presented in Figure 1.1, remains unofficial. Is this measure somehow distorted in ways that might prompt readers to remain skeptical about either the existence or the extent of the wage squeeze? For ease of calculation and later comparisons, we can concentrate on the period from the business-cycle peak in 1979 through the most recent available data for 1994 -- the decade and a half, as Figure 1.1 shows, over which the bulk of this decline in real hourly take-home pay occurred.
One possible concern might involve our adjustment for the personal income and payroll taxes borne by the average production employee. Suppose for the moment that we ignore taxes altogether and concentrate simply on real before-tax hourly earnings with a simple correction for the toll exacted by inflation. We can calculate the after-tax and before-tax measures for 1979 and 1994 and judge how crucial our tax estimates are in shaping the underlying trends in the series.
The level of the series estimated before taxes is obviously higher in both benchmark years -- for example, $11.13 in 1994 compared with an after-tax estimate of $9.36 -- since taxes are not yet deducted from real earnings. But the rate of decline in 1979-94 was actually somewhat greater for the beforetax measure than for the after-tax measure. Tax rates, primarily payroll tax rates, increased enough over this period to add to the wage pressure which workers were already feeling before the tax man took his cut. Whether or not we take taxes into account, the wage squeeze remains severe. The first two rows of Table 1.A provide this comparison.
Another possible concern involves the measurement of earnings itself. The real spendable hourly earnings series builds on direct wage-and-salary payments to production employees. As such, it does not include indirect employer payments benefiting workers, such as health care premiums. Measuring full compensation, including these fringe benefits, might make more sense but government data do not directly provide this information for production workers. Could it be that these benefits increased enough over the past fifteen years to offset the collapse in earnings?
In order to assess this possibility, I have constructed an alternative estimate of production workers' real hourly compensation including an approximation of the value of indirect employee benefits covered by employers. The measure of real hourly earnings, measured before taxes, is adjusted by adding to it an estimate of the additional benefits provided for workers by employers but not directly paid to them as earnings.
This adjustment makes only a minor difference, as the third row in Table 1.A makes clear. Now, the 1979-1994 decline in real (before-tax) hourly compensation is roughly 7.8 percent, compared with 8.6 percent in the measure ignoring benefits. This may surprise some readers; many assume that benefits have recently been increasing substantially more rapidly than earnings, particularly because of soaring increases in health care costs. But during the 1980s, real hourly benefit payments also dedined, indicating that many workers were being pressured to "give back" as much in benefits as they were in direct pay. In their useful analyses of trends affecting working Americans, Economic Policy Institute economists Lawrence Mishel and Jared Bernstein highlight this feature of the 1980s:
Health insurance costs have indeed risen quickly. Apparently, however, the rapid growth of jobs with little or no employer-provided health benefits and the increased shift of employer health care costs onto employees has meant that average fringe-benefit costs did not rise over the 1977-89 period. In fact, they declined modestly.
In contrast to their increases in the 1960s and 1970s, indirect benefits were no longer providing even a modest shelter against the harsh winds of wage decline.
Copyright © 1996 by David M. Gordon