Foreword

Recently after writing a piece in the newspaper stating that the income tax was the one tax in New Jersey that works right because it's the one based on people's ability to pay, I received an indignant e-mail from a fellow who said that was the most ignorant statement he had ever heard. Why, he insisted, should he be punished with taxes so much higher because he works hard? Why, to use his example, should a person making $150,000 a year have to pay more than someone making $25,000 to pump gas?

His missive was just one more reminder of the need for a new dialogue, one that is relevant to the circumstances working people face today. For one thing, what gives anyone the right to think that because they make more money they work harder? And for another, where are we going as a society if we fail to see beyond our own individual needs?

But if we are going to strive toward a new way of thinking and talking about issues basic to the survival of people, families and even society itself, the effort will have to be based on something more than emotion. The gut takes you only so far. That is where numbers come in. Stories of men and women struggling to make it out of poverty are inspirational in some cases and tragic in others but they are not enough.

The State of Working New Jersey is filled with numbers. They are numbers we all need to know, whether we are living the life of a working person or trusted with the responsibility of making policies to affect such people. And with all due respect to my e-mailing commentator, the numbers do not tell a very encouraging tale. To cut to the chase, the economic boom of the 1990s was neither long, deep nor wide enough to make a positive difference in the lives of a vast number of people in New Jersey. Indeed, these men, women and children have every right to wonder, "If that was a boom, what will a bust look like?" And now that the boom has in fact turned into a recession they are all too likely to find out.

The tables and text found in The State of Working New Jersey challenge conventional wisdom. They get below the surface of the economic boom to expose serious deficiencies. They detail differences along gender, racial, ethnic, educational and employment categories. They show that reports of widespread prosperity can mask deep chasms of inequality. Even while giving us the valuable perspective of where New Jersey stands compared to the nation they offer us scant comfort.

We at New Jersey Policy Perspective offer this report as part of the effort to change the dialogue. But not for its own sake. The real mission is to change policy. And The State of Working New Jersey points to changes across a wide spectrum. For starters, we need to restructure a tax system that, in part because of New Jersey's over-reliance on local property taxes, calls on middle- and low-income families to pay a higher percentage of their income in taxes that the wealthiest. We need to not only mend the economic safety net but make it stronger than it has been for years. This means acting on issues like increasing the minimum wage, making unemployment and disability insurance available to more workers and initiating paid family leave.

We are grateful for the job Leslie McCall did on The State of Working New Jersey. Her report takes its place among other work produced by NJPP to shine a light on inequity in New Jersey. Katherine J. Allen provided outstanding research assistance for all the analyses presented in the report. Every other year the Washington-based Economic Policy Institute issues The State of Working America, a comprehensive examination of living standards. We thank EPI for being so generous with the time and technical expertise it provides to state-based organizations so they can do localized versions.

Not long ago New Jersey took over first place in the nation in the category of household income. But it is a hollow victory. As a state, New Jersey will not be all it can be until its people are all they can be. That is the course on which we need to embark together.

-Jon Shure

Introduction

Most people know the economy was "booming" for most of the 1990s and especially the late 1990s. It was hard to avoid the media attention, whether it came in the form of news reports about Bill Gates clones or advertisements promoting SUVs or online trading accounts. But much less is known about the impact of economic growth on the living standards of workers. Getting at that kind of information requires an independent analysis of the latest publicly available — but unpublished—data on individuals and their earnings provided by the United States Census Bureau and the Bureau of Labor Statistics. This report uses such data to offer a comprehensive portrait of changes in earnings and income, adjusted for inflation, during the decade of the 1990s.¹

Our main finding is that New Jersey was unable to convert one of the longest expansions in state history into real wage and income gains for the majority of workers and households. Indeed, for most New Jersey residents the boom was a story about someone else. Contrary to popular perceptions, the New Jersey economy in the 1990s was not a prosperous one for most people, especially when compared to the 1980s and to other states in the country. Moreover, the distance between those at the top and the bottom of the economic ladder widened substantially and living standards stayed more or less the same for those in the middle. More specifically, some of the key findings of this report are that:

• The rate at which the New Jersey economy expanded was especially strong in the last half of the decade and in service industries, but the average year-to-year growth rate over the entire decade was no more than half that of the 1980s. This is true for both output and employment growth.
• While unemployment declined throughout the decade, it did not reach its pre-1990s-recessionary low of 3.8 percent until the year 2000, fully 12 years later and on the cusp of the current downturn. The upshot is that the tightening of the labor market that helps increase wages at the bottom came too late and did not last long enough to have a major impact on wage growth.
• New Jersey ended the last decade with a negative record of wage growth—that is, inflation-adjusted wages declined—for the bottom half of workers, for all education groups except college-educated women, and for all racial groups of men. As a result, top/bottom, racial and education-based wage inequality stood substantially higher at the end of the decade than at the beginning.
• Despite putting in more hours at work, total household income declined for the bottom 60 percent of households and increased only for the top 20 percent. As a result, inequality between high-income and low-income households rose for all household types and for all racial groups, especially for whites.
• The share of workers making wages that would send a family of their size below the poverty line also increased over the decade, as did the share making wages less than 150 percent of the poverty line. This increase in the ranks of the working poor was felt by nearly all demographic groups.
• However, despite declining incomes and rising wage inequality, New Jersey is average or fares relatively well in several important areas, such as educational attainment and overall economic hardship.

This report presents these basic facts about changes in the living standards of New Jerseyans in the 1990s. The first part of the report provides a brief overview of key indicators of economic growth. This topic is not our main concern because it has been widely analyzed elsewhere. The second part is the main part of our report. It offers a comprehensive portrait of the most basic labor market trends affecting the quality of life for New Jersey's workers, including trends in wages and household income. This section provides a detailed analysis of wage and income trends for different demographic groups—men, women, blacks, whites and Latinos—as well as for four different education groups and individuals and households at different points in the income distribution. The third section compares New Jersey's performance in the 1990s to that in the 1980s and to that of the rest of the country.

Growth, Employment and Unemployment

The most fundamental indicators of the underlying health and vitality of the economy track two things: whether production of goods and services is expanding and the number of workers in gainful employment is growing. In addition, it is important to know whether the quality of employment is improving as well. Since we cannot expect wages and income—our main focus in this report—to be in good shape if the economy is not growing in terms of output and employment, this section provides an overview of the macroeconomic context of growth, employment and unemployment in New Jersey over the course of the 1990s.

Not surprisingly, given what is commonly regarded as one of the most prosperous decades in U.S. history, we find that the New Jersey economy did grow considerably over this period. However, our analysis suggests that this generally positive assessment needs to be qualified in several respects.

• The rate at which the New Jersey economy expanded was especially strong in the last half of the decade and in service industries, but the average year-to-year growth rate over the entire decade was not as impressive as it was in the 1980s.
• Overall employment levels grew, but, by one measure, the quality of employment did not appear to improve substantially. In fact, the share of individuals in lower quality jobs increased, especially among men. For the most part, changes in the quality of employment took place across a wide range of industries and not just in traditionally low-wage industries.
• While unemployment declined throughout the decade, it did not reach its pre-1990s-recessionary low of 3.8 percent until the year 2000, fully 12 years later and on the cusp of the current downturn.

GROWTH

As measured by total real Gross State Product (GSP), New Jersey's economy grew by a respectable 2.23 percent on average for the years between 1989 and 1999. This rate of growth includes the year of negative growth between 1990 and 1991 (see Table 1) during the recession. Since then, when total output reached its lowest point in the decade, the state has been growing at a faster average annual rate of 2.82 percent. As is well known, the national economy took off in the late 1990s, reaching unexpectedly high rates of growth compared to long-term averages. Likewise, New Jersey's growth rate jumped from 2.61 percent at mid-decade to 3.9 percent in 1996. In 1999, the latest year of inflation-adjusted data, the growth rate stood at a still impressive 3.48 percent. If we exclude the government sector, and look only at the private sector, the growth rate is even greater for the decade as a whole (2.39 percent), for the period of expansion (3.05) and for the latest years of data (4.37 in 1998 and 3.63 for 1999).

Not surprisingly, the distribution of growth across major industry groups favors service over goods producing sectors. Output for the decade as a whole was down in both manufacturing and construction, though each had a few years of slight growth sprinkled throughout the expansion years. Growth was concentrated instead in the service producing industries, with average annual real growth rates over the expansionary period (1991-1999) of 4.16 percent for transportation, communications and public utilities; 6.25 percent for wholesale trade; 3.86 percent for retail trade; 3.53 percent for finance, insurance and real estate; and 2.78 percent for services, which was below the state average growth rate. If we look only at the expansionary period for construction and manufacturing, there was also positive growth, though well below the state average (1.52 percent in construction and .69 percent for durable manufacturing).

Within these broad industry groups, growth was concentrated in a number of key industries. Manufacturing, for example, saw substantial growth in industrial machinery, electronic equipment, chemicals and rubber and plastics. Within the transportation, communications and public utilities sector, output nearly doubled in communications and nearly tripled in air transportation. Within the finance, insurance and realestate sector, output in 1999 was six times what it was in 1989 for security brokerage firms. And within services, substantial growth came at the highend: business services, health services and educational services.

EMPLOYMENT

Trends in employment were similar, though not identical, to trends in Gross State Product. The downturn in employment, for example, began earlier than the downturn in output. As stated above, total output in inflation-adjusted dollars declined in 1991 and then rose. In contrast, employment began to decline in 1989 and kept falling until 1993 (see Table 2). Not until 1997 did total non-farm employment in New Jersey surpass its pre-recession level. Employment proceeded to grow to record levels until early 2001. According to the latest figures available at this writing, New Jersey employment peaked at 4,033,800 in February 2001 (see Table 3). Then, between February and September, the state lost 26,900 jobs.

Once again, growth was concentrated in the service sector (see Table 4), where employment levels in 2000 were 37.9 percent greater than in 1989. Other areas saw less dramatic growth: transportation, communications and public utilities grew by 16.4 percent; finance, insurance and real estate grew by 10.1 percent; and wholesale and retail trade grew by 6.8 percent. Construction held steady over the period, with only manufacturing posting absolute employment declines over the 1989-2000 time period. For the private non-farm sector as a whole, employment grew by 10.5 percent over the decade. Impressive as that is, private-sector employment grew by more than twice as much over the 1980s—by 22.8 percent between 1979 and 1989.

Clearly the service sector in New Jersey is large, its growth has been significant and the range of jobs it provides is highly diverse. A natural question, then, is whether the quality of employment growth matches the extent of that growth. The answer is mixed. On one hand, the service sector has created a fair number of good jobs—in communications, stock brokerage firms, business services, health services and educational services (as discussed above these industries exhibited significant growth in real output over the 1990s). But, many of the jobs in these and other growing sectors often fail to pay a living wage—one that reaches a level where a family can be self-sufficient. The healthcare sector, for instance, employs many nursing home and home care workers at poverty wages. To take another example, the business services category includes temporary agencies, which generally offer substandard wages and benefits. Meanwhile, such traditional low-wage service industries as retail trade also have grown. These industries have not necessarily grown as fast in percentage terms because they accounted for such a large share of the work force to begin with.

To understand shifts in the quality of employment, Table 5 breaks employment into three job quality categories. The official poverty line—determined by the federal government and generally acknowledged to be well below what a family needs to be self-sufficient—is used as the benchmark for job quality over time. A full-time year-round job had to pay $8.50 an hour to raise a New Jersey family of four above the federal poverty line in the year 2000. By this measure, the share of jobs paying poverty wages increased from 17.9 percent of workers in 1989 to 21.5 percent in 1995. The share of workers earning poverty wages declined to 19.5 percent in the year 2000. Over the course of the full decade, then, poverty wage jobs grew by nearly 9 percent.

By itself, this figure gives only a glimpse of the situation many workers face. Indeed, it conceals substantially higher growth rates for particular groups of workers. Although poverty-wage jobs are held disproportionately by women, the share of women holding such jobs declined over the decade—while the share of men holding such jobs increased significantly. For men as a group, the share of employment in poverty wage jobs grew by 29 percent between 1989 and 2000. This includes increases of 17.6 percent for white men, 20.6 percent for black men and 31.2 percent for Latinos. The large increases for both black and white men means that the increasing share of jobs paying poverty level wages is not due simply to the increasing share of immigrants.

Shifts in the quality of employment diverge for men and women in the other two job quality categories as well. First, there are decent wage jobs: those that pay between the poverty wage and 200 percent of the poverty wage and can be considered middle-income jobs. Overall, the share of employment in this category declined only slightly from 41.9 percent of jobs in 1989 to 40.3 percent in 2000. But this conceals the fact that being male increased the likelihood of earning decent wages while being female decreased the likelihood of earning decent wages. Meanwhile, women increased their likelihood of earning good wages (above 200 percent of the poverty line) and men decreased their likelihood of earning good wages. Thus, women shifted out of povertywage and decent-wage employment and moved into high-wage employment, while men, by contrast, shifted out of high-wage employment and moved into poverty-wage and decentwage employment. Once again, these trends held generally for each racial and ethnic group. Overall, the share a base year. The results are used to project what average wages would have been if the industry structure remained the same over time and only wages changed. If the projected outcome is the same as what actually occurred, then it means that shifts in the job structure are not to blame—that, rather, changes in wages for each industry were decisive. Comparisons of the projected and actual outcomes can therefore be used to estimate the impact of employment shifts between industries relative to wage changes within industries.

The result of this exercise shows that changes within industries had a much greater impact on overall wage trends in New Jersey than shifts in employment between industries. Indeed, changes within industries accounted for 95.8 percent of the change in average wages for men and women combined, 88.5 percent for women and 102 percent for men.² Most of the action, then, is occurring within industries and not between them.

The importance of changes within industries may seem surprising in light of the large shifts of employment from manufacturing to services over time. But examining wage and employment data for men in Table 8 shows that real average wages actually declined in durable and non-durable manufacturing industries between 1989 and 2000.

This suggests that changes were taking place across the industrial structure,with high-wage as well as lowwage industries posting wage losses, and with high-wage as well as lowwage industries posting wage gains. In essence, then, declining wages were more a factor of people being paid less for what they had been doing before than a case of people being paid less because they moved into a different line of work.

UNEMPLOYMENT

Before turning to a detailed analysis of wage, income and poverty trends, one last aspect of the macroeconomic context needs brief attention— unemployment. Even if the percentage of the population that is employed is growing and the percentage of unemployment is declining, enduring high levels of unemployment could dampen the pressure for wage increases. Indeed, as seen above, employment grew over the 1990s but not as fast as it did over the 1980s. Likewise, unemployment fell over the 1990s, but not by as much as it did over the 1980s.

The unemployment rate for every year from 1979 to 2000 is seen in Table 9. It shows that the unemployment rate was 6.9 percent in 1979—the peak of that business cycle—and then rose through the recession of the early 1980s. This was a period in which the natural rate of unemployment was considered to be rising by postwar standards. The unemployment rate dipped below the 6.9 percent mark by 1984, when it reached 6.2 percent. It then proceeded to fall another 2.4 percentage points by 1988, to 3.8 percent, well below the national unemployment rate. This represents a nearly 45 percent decline in the unemployment rate over New Jersey's 1980s business cycle.

By contrast, the unemployment rate did not reach its pre-1990s-recessionary low of 3.8 percent until the year 2000, fully 12 years later and on the cusp of the next downturn. Even if we look at the rate of decline in unemployment from the trough, when the unemployment level was more similar to that in the 1980s (8.5 percent in 1992 versus 9 percent in 1982) than it was at the pre-recessionary peak (3.8 percent in 1988 versus 6.9 percent in 1979), it took two years longer to reach the 3.8 percent level in the 1990s expansion than it did in the 1980s expansion. It was also not until the year 2000 that the unemployment rate in New Jersey fell below that in the United States as a whole, something it had achieved earlier on in the 1980s. The upshot is that the tightening of the labor market that helps increase wages at the bottom came too late and did not last long enough to have a major impact on wage growth, to which we now turn.

Individual Wages, Family Incomes
And Poverty

The production of goods and services expanded, employment grew and unemployment declined for most of the 1990s and especially for the years after mid-decade. Others have provided evidence that productivity rates picked up as well.³ Such a record indicates that the economy was in good enough shape to improve the living standards of its workers through wage and income increases, even if only modestly since the economy was not in as good shape as it was in the 1980s.

However, progress in output and employment did not mean progress in wages and incomes for the majority of New Jersey workers. And disparities between workers and families at the top and those at the bottom continued to widen. Three general trends stand out.

• New Jersey ended the last decade with a negative record of wage growth for the bottom half of workers, for all education groups except college-educated women and for all racial groups of men. As a result, top/bottom, racial, and education based wage disparities stood substantially higher at the end of the decade than at the beginning.
• Despite putting in more hours at work, total household income declined for the bottom 60 percent of households. It increased only for the top 20 percent. As a result, inequality between high-income and low-income households rose for all household types and for all racial groups, especially for whites.
• The share of workers making wages that would put a family of their size under the poverty line also increased over the decade, as did the share making wages less than 150 percent of the poverty line. This increase in the ranks of the working poor was felt by nearly all demographic groups.

INDIVIDUAL WAGES

After the setback of a recession, earnings traditionally bounce back to a higher level than before the recession. In fact, the 1980s brought higher median wages to most groups of workers in New Jersey, even though in the rest of the country median wages were lower at the end of the 1980s than they were at the end of the 1970's, before the recessions of the early 1980s. For the workforce as a whole, the median wage rose more than a dollar an hour from $12.10 in 1980 to $13.23 in 1990 (in 1995 dollars).⁴

But in the 1990s things turned for the worse for the median worker as well as for several subgroups of workers (see Table 10). The median wage continued to rise between 1989 and 2000 only for women as a whole (by 0.2 percent per year) and for white women and black women (by 0.8 percent and 0.3 percent per year, respectively). For Latina women and all racial groups of men, the median hourly wage was lower in 2000 than in 1989, despite the longest expansion in recorded history. The median wage was down 0.2 percent per year for the workforce as a whole and 0.7 percent per year for men as whole.

Although median wages declined for each of the three racial groups of men, median wages declined more for blacks and Latinos than for whites. As a result, the racial wage gap between whites and blacks and between whites and Latinos grew over the 1990s. In 1989, black men earned 73 cents for every dollar earned by white men and Latinos earned 67 cents. By 2000, these amounts had declined to 69 cents for black men and 61 cents for Latino men relative to white men. Although racial wage gaps are not as large among women as among men and median wages rose for white women and black women, the gaps widened for women over the 1990s as well. Black women earned 84 cents for every dollar earned by white women in 1989, and 79 cents per dollar in 2000. Latinas earned 78 cents on the dollar in 1989 and 69 cents in 2000.

There is good news of a sort: the gender gap in wages narrowed for every racial group. But the bad news is that this was based not only on wages for women increasing but also on wages for men decreasing. Overall, women ended the decade of the 1990s earning 76.5 percent of what men earned, up from 68.8 percent in 1989. The equivalent figures for blacks were 86.9 percent in 2000 and 77.4 percent in 1989; for whites, they were 76.3 percent in 2000 and 67 percent in 1989; and for Latinos, they were 86.3 percent and 77.9 percent.

Now some might wonder why there is not more good news to report. Weren't the 1990s and especially the late 1990s the era of America's greatest prosperity? In fact, the picture does brighten somewhat if we look only at the second half of the 1990s. The median wage was roughly the same in 2000 as it was in 1995 (in other words, it did not decline as it did between 1989 and 1995). Moreover, the median wage increased for all but one of the racial groups of women and of men (the exception being black women).5 Unfortunately, the late 2001 downturn makes it unlikely that this positive momentum will continue. In fact, a year-by-year analysis of median wage changes reveals a significant increase only between 1997 and 1998 and not in more recent years.

The last observation to make about Table 10 concerns the average wage: it actually posted an increase over the 1990s. Not only has the average wage become higher over time, but the average wage is quite a bit higher than the median wage ($18.25 versus $14.44 in 2000). The reason is that the median wage shows what is happening in the middle, while the average wage is skewed by gains at the top. It is possible, therefore, for the average wage to go up even as the earnings of most workers go down. This signals a rise in wage inequality between workers at the top of the wage distribution and everyone else. As we shall see, this is exactly what happened in the 1990s.

To examine this rise in inequality more closely, Table 11 divides the population into 10 equal-size groups, showing wages for each group. A common measure of wage inequality is the ratio of wages at the top of the distribution to wages at the bottom. For example, a wage earner at the top tenth of the distribution would be someone whose wages were greater than 90 percent of all workers in the sample. A wage earner at the bottom of the distribution could be someone whose wages were greater than only 10 percent of all workers in the sample.

Two common measures of wage inequality are the 90th/10th and 80th/20th ratios, and both rose during the 1990s. Indeed, by the year 2000 the 90th percentile worker earned nearly five times as much as the 10th percentile worker (the ratio was 4.9 in 2000 as compared to 4.4 in 1989), and the 80th percentile worker earned nearly three times as much as the 20th percentile worker (the ratio was 2.9 as compared to 2.7 in 1989). This amounted to an 11.4 percent increase over the 1990s in the 90th/10th ratio and a 7.4 percent increase over the 1990s in the 80th/20th ratio. Moreover, this widening gap in the 1990s between those at the top and bottom of the wage distribution followed on the heels of similar rises in the 1980s.

Although New Jersey's economy has been growing more unequal for nearly two decades, the form it has taken has varied over the period. In the 1980s, rising inequality came from the top and middle of the distribution pulling away from the bottom 20 percent. There were no real wage declines for workers at any of the deciles, only stagnation at the very bottom. But, during the 1990s, wage trends took a striking turn for the worse. For the bottom half of the wage distribution, wages declined, while wages continued to grow for the top 30 percent. In fact, wages increased the most for workers at the very top (the 90th percentile) and wages decreased the most for workers at the very bottom (the 10th percentile). In the middle of the distribution, wages either stagnated or declined less precipitously.

Once again, the picture brightens if we look only at the last half of the 1990s. With the exception of the median worker, whose wages held steady, there were significant wage gains for workers at both the top and the bottom of the distribution. More importantly, some of the greatest gains were at the bottom of the distribution. Over this sub-period, then, there was some indication that inequality had reached a plateau and that it might even be rolled back if the economy continued to accelerate. But we now know that the acceleration in growth and the decline in unemployment were short-lived. A yearby- year analysis of wages at the 80th and 20th percentiles actually reveals a small increase in the 80th/20th ratio between 1999 and 2000. Once again, wage gains at the top exceeded those at the bottom.

Much the same story can be told about trends in wages broken down by years of schooling (see Table 12). In general, wages for New Jerseyans who never went to college have been declining since at least 1980. Those who have not completed high school fared the worst: their median wage dropped 17.6 percent during the 1990s. Among men in this group, the median wage dropped by more than 30 percent. Declines were not as significant among women, nor were they as significant for those with a high school degree or some college. But it should be noted that wages did fall for both men and women, disproving the often-held assumption that wages declined only for men.

Workers with four years or more of college were the only group to make any gains in the 1990s. But here, too, the record is mixed. All of the gains in the 1990s were due to strong advances made by college-educated women. Their median wage rose by 15 percent between 1989 and 2000, while the median wage for college-educated men fell slightly by 1 percent. College-educated men were fortunate that they did not end the decade even farther behind - their wages picked up significantly in the late 1990s after falling in the early 1990s. Wage growth for college-educated men actually exceeded wage growth for college-educated women in the late 1990s.

The fact that wages are higher for college-educated workers is often referred to as the "college wage premium" and recent discussions have focused on the trend toward an increase in this premium. The premium is measured as the ratio of wages of non-college-educated to college-educated workers. By the year 2000, the median wage for a high-school dropout was only 34 percent of the median wage of someone with a college education, down from 42 percent in 1989. The median high-school educated worker earned a little more than half that of the median college-educated worker—53 percent—and this too was down from 1989, when it was 58 percent. The college wage premium grew over the last half of the 1990s as well as over the first half.

In sum, New Jersey ended the last decade of the 20th Century with a negative record of wage growth for the bottom half of workers, for all except the college-educated and for all racial groups of men. Only workers at the top or with a college-education made significant gains. Although top/bottom disparities and disparities by race and education seemed to moderate as the recovery picked up speed in the late 1990s, these improvements did not prove strong enough, nor last long enough, to make up for damage inflicted by the earlier recession and the nearly five years of sluggish recovery that followed. As a result, top/bottom, racial and education-based disparities stood substantially higher at the end of the decade than at the beginning.

FAMILY AND HOUSEHOLD INCOME

The stagnation and decline of wages for a large share of the work force puts more pressure on families to work harder just to maintain current living standards. This "time squeeze" has become a well-known trend. More family members enter the paid labor force. They work longer hours. They take on second jobs. With many families already having exhausted these options, they have no more spare time left to balance the commitments to work and family. As families lose the ability to compensate for lower wages, their total household incomes begin to decline as well.

In addition to sending more family members into the labor market, especially mothers with young children, a common strategy for families to increase their income is for family members to log longer hours on the job. Overtime pay, when available, makes this option even more attractive. Table 13 shows that the average number of hours worked in the paid labor force in New Jersey has increased over the past decade. There was a small increase in the number of hours at work for men, who worked roughly 41 hours a week on average, and a somewhat larger increase for women, who worked roughly 36 hours a week on average. Assuming that women work 50 weeks a year, the increase in average weekly hours over the 1990s means that women were putting in an extra week of full-time work by the end of the decade.

Table 14 provides additional evidence of the need to work longer hours in order to make up for declining wages, especially for families in the middle and at the bottom of the income distribution. For married couple families with children, where the household head was between the ages of 25 and 54, the total number of hours worked for all family members rose from 3,539 in the late 1980s to 3,613 in the late 1990s, on average. This rise is the equivalent of nearly two extra full-time weeks of work. Interestingly, this 2 percent rise in the number of hours worked over the 1990s followed a much greater 7.5 percent increase over the 1980s, suggesting that the ability to work longer hours is reaching its natural limit.

In fact this may already have occurred for families at the top of the income distribution. The average family in the top 40 percent of the family income distribution actually reduced its total number of hours worked per year over the course of the 1990s, and also over the course of the 1980s for the average family in the top 20 percent of the family income distribution. These were the very families that were putting in the longest hours of all families—roughly 15 percent more than the average family in the late 1990s and over 50 percent more than the average family in the bottom 20 percent of the distribution. Thus, while high-income families are fortunate in having access to rewarding "full-time plus" jobs, a tradeoff is the time-squeeze required of the m in their high-income careers.

Lower-income and middle-income families (the next 40 percent of families from the bottom) worked fewer total hours per year than upper-income families but they dramatically increased their hours over both the 1980s and 1990s. They added 200 to 300 hours between the late 1980s and the late 1990s alone. The poorest 20 percent of families concentrated all of their extra work effort in the 1980s. They increased their time at work by nearly 500 hours during the 1980s and added nothing more in the 1990s. The majority of families, then, were attempting to supplement their incomes by working more and more hours. But those with the least opportunities appeared to be discouraged from work in the 1990s.

Since most individuals and families worked harder than ever in the 1990s, one might expect their household incomes to have gone up. But they did not. Instead, total household income declined for the bottom 60 percent of all households. As shown in Table 15, median household income declined from $55,897 in 1989 to $52,541 in 1999, a 6 percent loss over the decade. Meanwhile, the top 20 percent of households were better off at the end of the 1990s than they were at the end of the 1980s. The household at the 90th percentile of the income distribution saw its total income rise by 9.6 percent and the household at the 80th percentile saw its total income rise by 5.7 percent.

This overall pattern of rising income at the top and falling income for everyone else characterized the experience of households as a whole, as well as non-family households (see Table 16) and dual-headed households. Somewhat surprisingly, though, most single headed households, including some at the bottom of the distribution, were able to emerge from the decade with higher incomes. Such families were harder hit by the recession and sluggish recovery during the first half of the 1990s, but they also benefited disproportionately from the growth in income during the late 1990s, perhaps in part as a result of welfare reform's work requirements during a time of tight labor markets.

Because income growth was concentrated among high-income households, inequality between high-income and low-income households increased substantially over the course of the 1990s. Lest anyone think this is solely the result of increased immigration or racial disparities, household income inequality rose for all household types and for all racial groups, with the greatest increase among racial groups occurring for white households (see Tables 17 and 18). The ratio of household income at the 80th percentile to household income at the 20th percentile rose by 20 percent between 1989 and 1999 for white households; 2.4 percent for black households; and 3 percent for Latino households. For all households combined, income inequality was nearly 18 percent greater in 1999 than in 1989. The income of the household at the 80th percentile grew from 4.5 to 5.3 times that of the income at the 20th percentile.

Although income inequality between whites and other racial groups did not increase as much as it did within racial groups, the level of racial inequality remains astonishingly high. The median household income for African-Americans was between 57 percent and 58 percent that of whites throughout the decade. The equivalent figure for Latinos was 60 percent at the beginning and the end of the decade (with a slight decline in the ratio, and thus increase in inequality, at mid-decade). While black and Latino households did not lose ground relative to whites, neither did they make any progress in eliminating the already very significant racial gap in income.

In sum, like the trend in individual wages over the 1990s, the trend in household income disproportionately benefited those at the top of the distribution. For nearly all household types and racial groups, the more income a household had in 1989, the more its income grew in the following decade. In contrast, households with the least income in 1989 experienced the greatest income declines in the following decade.

It should be noted however, that, stating it this way could be somewhat misleading because we are unable to follow the income dynamics of a particular household over time to see whether its income actually increased or decreased over the decade. Rather, we can only compare the income of the household at a particular point in the distribution in 1989 with the income of the household at that same point in the distribution in 1999—and these two households are not the same. Technically, then, we should say that households at the bottom in 1999 brought in much less than households at the bottom in 1989, and the disparity between the two periods increased the further down in the distribution a household was located. In contrast, households at the top brought in much more in 1999 than households at the top in 1989, and the disparity between the two time periods also increased the higher up in the distribution a household was located. The bottom line is this: households at the top were better off in 1999 than households at the top in 1989, while households at the bottom were worse off in 1999 than households at the bottom in 1989.

POVERTY

The decline of real wages and real incomes at the bottom of the wage and income distributions plays an increasingly important role in fostering poverty and the kind of economic hardship experienced by the working poor. Table 19 reports the share of workers whose earnings would not have been high enough to raise their families above the official poverty line if their wages were the only source of income for the family. Rather than assume a norm of four family members, these calculations adjust for the family size of each individual worker. The working poor, by this definition, increased from 13.2 percent of the work force in 1989 to 15.2 percent in 1999, an increase of 15 percent over the decade.

All groups felt this increase in the ranks of the working poor. When wage earners are broken down by race, we see that whites, blacks and Latinos all ended the decade with a larger percentage of workers earning povertylevel wages. In percentage point terms, the increase was greatest for Latinos. In addition, Latinos are much more likely to earn poverty-level wages (25.3 percent in 1999) than either African-Americans (15.8 percent) or whites (12.6 percent). For African Americans in particular, there was significant progress in reducing the share of poverty wage earners in the late 1990s. The share fell by about a quarter from 20.8 percent in 1995 to 15.8 percent in 1999. Finally, there was a substantial 35 percent increase in the share of men earning poverty-level wages, although the absolute percentage of men in the working poor category is just two-thirds that of women in the category

Although the official poverty line was originally designed to estimate the minimum threshold of income needed to survive in a crisis situation, a more realistic estimate of the basic cost of subsistence, according to opinion polls and expenditure surveys, is at least 150 percent of the poverty line. If this were used as the benchmark, a "living wage" would be defined as significantly higher than the actual poverty-level wage. In the case of a family of four in New Jersey, the living wage in 1999 would have been $12.75 per hour on a full-time year-round basis (rather than the $8.50 poverty line wage). The share of the work force that does not earn a living wage for its family size increased from 27.5 percent in 1989 to 30.2 percent in 1995, and then declined slightly to 29.9 percent in 1999. We refer to these workers variously as "low-wage earners" and "low-wage workers."

Overall, improvements in the late 1990s once again were not enough to fully reverse the negative impact of the early 1990s recession on the low end of the labor market. This was especially the case for blacks: the share of low-wage workers among blacks rose from 29.5 percent in 1989 to 38.5 percent in 1995, then declined to 35.9 percent in 1999. Blacks, then, ended the decade with a 20 percent higher share of workers in the low-wage category. Latinos as a group and all men as a group both saw steady and quite substantial increases in the share of low-wage workers over the course of the decade. Women and all whites as a group, on the other hand, were slightly better off in 1999 than in 1989 in terms of their shares of low-wage workers.

Finally, it is important to note the quite high levels of employment in jobs that do not furnish a living wage. Roughly half of Latinos, a third of blacks and a quarter of whites do not make a living wage. Almost one-quarter of all men and more than one-third of all women have low-wage jobs. Women are over 40 percent more likely than men to make less than a living wage.

A more detailed analysis of living standards among families of low-wage earners is presented in Table 20. This table includes all workers who do not make a living wage for their family size. When the earnings of all other family members and other sources of income such as child support are included, and various types of social welfare assistance are also taken into account, the incidence of poverty among the families of low-wage workers was 8.8 percent in 1999. This is a nearly 70 percent increase from 1989, when the figure was 5.2 percent. If the more realistic standard of self-sufficiency is assumed -- 150 percent of the official poverty line -- then we see that 20.1 percent of low-wage workers live in poor families. In other words, even when all sources of public and private income are included, one of every five low-wage workers and their families still live below an adequate level of self-sufficiency. Moreover, the share of low-wage earners in poor families increased slightly from 1995 to 1999 and was up nearly 40 percent from 1989.

Although the share of workers earning less than poverty and living wages rose over the 1990s, and the families of such workers grew more impoverished, the official poverty rate for the state declined by almost half a percentage point, from 8.2 percent in 1989 to 7.8 percent in 1999. How is this possible? It is because the official poverty rate calculated by the federal government accounts for the poverty status of all residents in the state and not just those residents who are in the workforce. To say that an increasing share of the workforce falls into the working poor category—earning poverty-level wages or low-wages—or that an increasing share of low-wage workers are living in poor families, is not the same as saying that the population as a whole is becoming more impoverished. Rather, it says only that the conditions of individual workers at the bottom of the wage distribution and in particular demographic groups is deteriorating.⁶

New Jersey in National Perspective

So far our analysis has focused exclusively on trends in New Jersey. To gain a broad perspective, it also helps to compare New Jersey with the nation as whole.⁷ This gives a sense of whether New Jersey has been performing better or worse than other states or whether the state has followed more or less the same pattern as the rest of the nation. If the latter were the case, then it would seem that New Jersey's fate was caught up with that of the rest of the country's and could not be altered by state level policies. By contrast, if the former is the case, then it would seem that New Jersey should be commended for the areas in which it is doing better than the rest of the country and should turn its attention to those areas in which it is doing worse.

Three general patterns stand out when comparing New Jersey to the rest of the country.

• Long-term trends in employment, wages, inequality, and poverty generally were better in the 1980s and worse in the 1990s relative to other states and to the national average.
• New Jersey still stands out as a state -- in both positive and negative ways—when considering patterns of and changes in manufacturing employment and unionization.
• New Jersey is average or fares relatively well in several important areas, such as educational attainment and overall economic hardship.

These three issues are discussed in greater detail below.

THE 1980s vs. THE 1990s

Unlike much of the country, New Jersey experienced strong growth in the 1980s. It had the fourth highest annual average growth rate of per capita Gross State Product between 1979 and 1989 (4.2 percent). In addition, its GSP per capita -- its level of output rather than its growth in output -- ranked fourth in 1989 ($33,427 in 1998 dollars), and third in 1998 ($39,429).

Although highly ranked in its level of GSP per capita at the end of both the 1980s and the 1990s, New Jersey fell significantly in its ranking on growth in the 1990s, ranking 16th from the bottom with an average annual growth rate of 1.9 percent between 1989 and 1998 (but still matching the national average). Growth rates for 1999 and 2000 would probably improve New Jersey's average annual growth rate for the decade of the 1990s, but other states were growing faster in these years as well, so it is unlikely that New Jersey's ranking would improve significantly.

The reversal of New Jersey's relative ranking between the 1980s and 1990s also is apparent, though somewhat less dramatic, when we look at trends in employment and unemployment. Employment in the state grew by about 22 percent between 1979 and 1989, giving New Jersey a ranking of 21st from the top (and still above the national average employment growth rate). But in the 1990s, New Jersey ranked fifth from the bottom with an employment growth rate of 8.3 percent. The nation as a whole saw employment grow by 22 percent between 1989 and 2000. The picture improves if we look at only the last few years of the decade and add preliminary data for 2001. Between 1998/1999 and 1999/2000, New Jersey posted a growth rate of 2.7 percent, compared to the U.S. average of 2.5 percent. Between 1999/2000 and 2000/2001, New Jersey's employment grew by 1.9 percent, while the U.S. grew by 1.5 percent on average. Although above average, New Jersey's rates were not high enough to place the state among the leaders in job growth during the late 1990s boom.

Unemployment in New Jersey declined in both the 1980s and 1990s, from 6.9 percent in 1979 to 3.8 percent in the year 2000, but its rate of decline was much greater in the 1980s (-2.8 percent), when it ranked fourth from the top, than it was in the 1990s, when it ranked 12th from the bottom (-0.3 percent). Once again, New Jersey did better than average in reducing unemployment in the last few years, but so did many other states.

Not surprisingly, trends in wages, poverty and inequality follow those in growth, employment and unemployment. In New Jersey, real wages grew significantly over the course of the 1980s for workers at the bottom (20th percentile) as well as at the top (80th percentile), while most Americans saw their real wages decline over this period. This was good news indeed for New Jersey workers, with increases of 9.1, 11.2, and 14.0 percent for workers at the 20th, 50th, and 80th percentiles of the wage distribution (the comparable national average rates were -8.5, -2.4, and 1.5 percent). In contrast, while wages were growing for most of the rest of the country in the 1990s, wages were declining in New Jersey for those at the bottom (by -3.8 percent) and at the middle (by -1.1 percent). New Jersey went from a fourth/fifth place ranking from the top in real wage growth for workers at the bottom/middle of the wage distribution in the 1980s to a seventh/ninth place ranking from the bottom for such workers in the 1990s in which wages declined. Meanwhile, real wage increases at the top (80th percentile) were above average in both the 1980s and 1990s. Real wages for this group increased by 4.9 percent on average in the U.S. in the 1990s and by 5.5 percent in New Jersey.Thus, the only deviation from the "strong 1980s/weak 1990s" pattern is for high-wage workers, for whom both decades were positive.

A result of these trends in wages is that the share of workers earning below poverty level wages in New Jersey declined in the 1980s and increased in the 1990s, once again reversing trends at the national level. New Jersey ranked sixth from the top in the 1980s for its record in fighting the creation of poverty-wage jobs, with 4.2 percent fewer such jobs in 1989 than in 1979. Its share of such jobs rose by 1.3 percent in the 1990s, however, giving it a ranking of ninth from the bottom. The contrast between the 1980s and 1990s is less dramatic when looking at overall poverty rates. Poverty declined in both the 1980s and the 1990s, from 9.0 percent in 1980 to 7.8 in 1999, but the percentage point decline was above average in the 1980s (-0.8 versus -0.2 on average) and below average in the 1990s (-0.4 versus -1.0 on average). In neither decade did New Jersey rank close to the top or bottom of the distribution of states.

Finally, the '80s/90's contrast is evident in some other important areas. In terms of inequality, the ratio of average income for the top 20 percent to the bottom 20 percent increased over both the 1980s and the 1990s, but the percentage point increase was slightly below average in the 1980s (1.1 for New Jersey versus 1.9 for the U.S.) and slightly above average in the 1990s (1.4 in New Jersey versus 1.3 for the U.S.). In terms of the share of workers with employer provided pensions, there was a greater decline in the 1980s in New Jersey than in the nation as a whole (a departure from the overall positive nature of trends in the 1980s in New Jersey) and there was a smaller increase in the share of workers covered in the 1990s. And, lastly, the share of workers with employer provided health care declined less in the 1980s in New Jersey than on average, and more than average in the 1990s. New Jersey had the third lowest decline in the 1980s and the second highest decline in the 1990s.

DECLINE OF MANUFACTURING

Though New Jersey's economy over the past few decades has become highly diversified, the second significant trend distinguishing the state from many others is its history of strong manufacturing employment. In 1979 and 1989, New Jersey ranked 10th in the absolute number of manufacturing employees, and in the year 2000, its rank fell to 14th. New Jersey shed 20 percent of its manufacturing jobs over the 1980s and 27.7 percent over the 1990s. This compares to 7.8 and 4.8 percent declines for the United States as a whole over the same time periods. As a result, New Jersey ranked among the top five states with the largest declines in manufacturing employment in both decades. In addition, manufacturing sector employment for the decade peaked in 1997 in New Jersey and since then the state has suffered above average declines in manufacturing employment, though its rate of decline was not as far from average as it was for the decade as a whole.

Because New Jersey did relatively well in the 1980s on other indicators at the same time that manufacturing declined precipitously, transformations in the manufacturing sector apparently did not have a major effect on the state economy as a whole. As discussed above, changes were taking place throughout the economy and across a wide variety of industries. Moreover, New Jersey continues to have one of the highest union densities in the nation (ranked fourth in 2000 at 20.8 percent, tied with Michigan) and the share of the workforce covered by a union contract shrank less here than in the nation as a whole. Nevertheless, it is important to acknowledge that a significant segment of the state's workers and families have been affected by the shift away from manufacturing, a shift that was actually greater in percentage terms than in most Midwestern states.

NEW JERSEY'S STRENGTHS

By and large, the areas in which New Jersey excels are well known. It has among the highest per-capita incomes and one of the most highly educated populations. Perhaps less well known, however, is that New Jersey also does relatively well in minimizing the degree of poverty and other forms of material hardship. Thus, when examining New Jersey's record at a particular point in time—that is, its absolute level of welfare—the state appears to provide relatively well for its residents across the economic spectrum and not just for those at the top end. Note that this is not inconsistent with the evidence provided above, which shows that changes over time have disproportionately benefited the well off.⁸

On a wide range of indicators of material hardship, New Jersey not only performs better than average, it often ranks among the very best states. In terms of poverty, for example, New Jersey consistently ranks among the 10 states with the lowest levels of poverty. In 1999, New Jersey ranked third with a poverty rate of 7.8 percent, compared to the average poverty rate in the United States of 11.8 percent.

A more comprehensive measure of material hardship is the share of persons in families with income below basic family budget levels (Boushey et al. 2001). Family budgets tabulate the costs of every major item that a family needs, including housing, child care, health care, food, transportation and taxes, based on the composition of the family and where the family lives. On this measure, calculated with data pooled over the 1997-1999 period, New Jersey had the 13th lowest share of individuals living in families with incomes that did not meet these basic needs requirements. Still, though, 21.0 percent of individuals in the state did find themselves in this situation (as compared to 27.6 percent in the nation as a whole).

The study from which these data come looked closely at the types of particular hardships that individuals faced. The study found that, in the U.S. as whole, 15.8 percent of the population experienced at least one "critical" hardship -- defined as missing meals, not receiving necessary medical care or doubling up with friends or family -- and 45.4 percent of individuals faced at least one "serious" hardship -- defined as worrying about having enough food, using the emergency room as the main source of medical care, being unable to make housing and utility payments, and so on. In New Jersey, these figures are lower, especially regarding those who experience serious hardships (34.7 percent), though of course, once again, the number of individuals in such situations is surprisingly high. New Jersey's better than average performance holds for individuals in all five quintiles of the income distribution.

As is perhaps better known, New Jersey also does well in achievement of higher education (see Table 21). A lower share of the population has less than a high school education and a larger share has a college education and an advanced degree. In the year 2000, New Jersey ranked seventh among states in share of individuals with advanced degrees (10.3 percent, as compared to 8.7 percent for the nation) and fifth in share of individuals with college degrees (22.8 percent, compared to 18.8 percent for the nation). While 10.8 percent of the nation's population does not have a high school degree, only 8 percent of New Jerseyans fall into this category. This is especially noteworthy given New Jersey's sizable immigrant population.

One final bright spot to emphasize is that, so far, the state has not been doing as badly as other states in the recent downturn. When the most recent figures were released, the state unemployment rate was below the national average and it has been at the national average or below it for the past 22 months. According to a recent study of the effect of the downturn (focusing on the 1998-2001 time period) on employment and wages across the states, New Jersey employment has grown at a rate greater than the national average and real wages have held steady for low wage (10th percentile) and medianwage workers. Of course, the situation has to be closely monitored on a month-by-month basis, but thus far the evidence suggests that New Jersey has not been hit as hard as other states by the downturn.

Conclusion

The New Jersey economy took a turn for the worse during the 1990s as compared to the 1980s when employment growth, output growth and wage growth were greater. While those at the very top of the labor market continued to make solid gains in the 1990s, earnings and incomes were stagnant or falling for the rest of the workforce. This led to a widening of the gap between those at the top and everyone else. Despite working harder and acquiring more skills, middle-class families have suffered a decline in their living standards. And a larger number than previously has fallen into the ranks of the working poor.

Unfortunately, the picture of labor market trends provided by The State of Working New Jersey is unlikely to improve anytime soon, since the expansion has already come to an end. A lower level of unemployment in the late 1990s relieved some of the downward pressure on wages that took place in the first half of the decade during the sluggish recovery from the last recession (in the early 1990s), but earnings and incomes never caught up to their previous peaks. The upshot is that economic growth does not necessarily raise living standards the way it did during previous periods of expansion. And as a result, we cannot simply rely on a rising tide to lift all boats.

Appendix Data and Methods

The wage and income data in this report is derived from the Current Population Survey (CPS), the most comprehensive source of labor market data on individuals available from the federal Bureau of Labor Statistics (BLS). The survey is conducted every month of the year, though some of the information gathered varies from one month to another. There are two compilations of the CPS that we analyze in this report. Both raw data sets were provided by the Economic Policy Institute. All state-level statistics not presented in the tables were provided by the Economic Policy Institute (2001) or were, and Schmitt (2000), including all statistics reported in the New Jersey in National Perspective section and all statistics of year-by-year changes in the late 1990s.

Current Population Survey ORG Files

First, basic labor market and demographic data on wages, hours worked, weeks worked, education, gender, and race, among other things, are gathered every month from a sample of households that is in the last month of a four month stretch of monthly interviews. These households are in their "outgoing rotation" month and the BLS merges data from every such household into a full-year sample of individuals called the Merged Outgoing Rotation Group Files (ORG). This is the largest data source available on individuals during the years between the decennial censuses.

We use the ORG files to analyze the wages of individuals working in New Jersey in 1988, 1989, 1994, 1995, 1999 and 2000. We combine two years together to arrive at a large enough sample to subdivide into groups based on gender, race, education, broad industry of employment and percentile ranking in the overall wage distribution. Our sample therefore consists of three time points: 1988/89, 1994/1995 and 1999/2000.

CPS March Annual Demographic Files

The second CPS source that we use is taken from the survey conducted every March known as the March Annual Demographic File. It offers the most comprehensive source of information on individuals as well as the households and families they live in. We merge information from all three—individuals, households and families—to describe the total income status of the New Jersey population, including cash and non-cash transfers from the government. Our sample of households is a combined sample of families and individual householders.

Because government transfer programs linked to farming and old age are universal and therefore of a fundamentally different nature than all other government transfer programs (in that they are not indicative of income level), we exclude data on individuals who live in households receiving farm income or with a household head or spouse who is over 64. We also exclude such households from our household sample. We further exclude the small number of individuals living in unrelated subfamilies within households because information on the poverty status of their families is unreliable.

For individual earners, we expanded our selection criteria in one respect to include self-employed individuals since they are important contributors to family income and to the distribution of family income. However, we excluded high school aged teenagers (16 and 17 yearolds) from the sample of individuals, in order to omit the influence of low-wage high school workers in our estimates of the low-wage population. We pooled data from two consecutive years to form three samples, one covering the years 1988 and 1989, a second covering the years 1994 and 1995, and a third covering the years 1998 and 1999. Since income information refers to the year prior to the survey, these data are from the March 1989, 1990, 1995, 1996, 1999 and 2000 files.

The main purpose of this sample is to calculate 1) total household income and 2) the poverty status of individual workers based on their total family income. Our calculations are the most conservative estimates of the extent of hardship associated with working in low-wage jobs. When we say that an individual makes a poverty-wage, we have actually calculated the hourly equivalent of 100 percent of the official poverty line for that individual’s family, which adjusts for family size and other determinants of poverty thresholds. We also assume that the individual is working full-time, yearround, which is a conservative estimate of the hourly wage needed to exit poverty because it would have to be higher if he/she worked part-time. Because we make these adjustments for family size, we can combine single adult individuals without children and individuals with children without making normative assumptions about family formation (as is typically done when the poverty line is assumed to be the level for a family of three or four).

About the Author

Leslie McCall is an associate professor at Rutgers University in New Brunswick in the Departments of Sociology and Women's and Gender Studies. She specializes in the study of social inequality. She is the author of the book, Complex Inequality: Gender, Class, and Race in the New Economy (2001), as well as many journal articles, reports and other articles. She has a Bachelors degree from Brown University in Computer Science and Third World Development Studies and both a Masters and a Ph.D. from the University of Wisconsin-Madison, Department of Sociology.

Bibliography

Boushey, Heather, Brocht, Bethney Gundersen, and Jared Bernstein. 2001. Working Families Left Behind: Families Who Go Without Basic Needs. Washington, D.C.: Economic Policy Institute.

Economic Policy Institute. 2001. Scattered Showers for Labor Day 2001: Job Losses, Falling Wages Hit Many States. Briefing Paper. Washington, D.C.: Economic Policy Institute.

Hughes, James W. and Joseph J. Seneca. 1997. "Long-Term Regional Momentum: Employment and Income Dimensions." Rutgers Regional Report 15. New Brunswick, NJ: Edward J. Bloustein School of Planning and Public Policy.

Hughes, James W. and Joseph J. Seneca. 1999. "The Emerging Wealth Belt: New Jersey's New Millennium Geography. " Rutgers Regional Report 17. New Brunswick, NJ: Edward J. Bloustein School of Planning and Public Policy.

McCall, Leslie and Eric Parker. 1997. The State of Working New Jersey. New Jersey Center for Economic Policy and Education.

Mishel, Lawrence, Jared Bernstein, and John Schmitt. 2000. The State of Working America 2000/2001. Ithaca, NY: Cornell University Press.

1. Note that we do not provide an analysis of wealth—assets other than income, such as stocks—because such data is not readily available at the state level.
2. As will be discussed in more detail below, average wages increased slightly between 1989 and 2000 for men and women combined and for women only. This would seem to complicate things if we are looking for an explanation of declining wages. However, this procedure sorts out where changes are taking place—in the shift of employment between industries or in the wages paid to employees within industries whether or not these changes are positive or negative.
3. 3 See, for example, Hughes and Seneca (1997, 1999).
4. For an analysis of wage and income trends in the 1980s, see McCall and Parker (1997).
5. Taking a closer look at the increases in wages at the end of the 1990s reveals that men as whole lost ground while each racial group of men gained ground. This is an indication that the racial composition of male wage earners changed over this period, with a higher share of lower-waged racial groups in the mix in 2000 than in 1995.
6. Moreover, such an analysis is based on an optimistic simulation that compares the wages of workers to the size and total income of the families in which they work and assumes full-time, year-round work for the workers (see the Data and Methods Appendix for more detail).
7. These state-level data are provided by the Economic Policy Institute. Some of the figures may differ slightly from those reported above because they come from different sources.
8. For example, although changes in pension and health care coverage were more negative in New Jersey in the 1990s than in the rest of the country, levels of coverage were still above the national average.