Being middle class has emerged as a vital part of the 20 th-century American pscyhe. The majority of Americans define themselves as middle class, regardless of their actual income level. This perception is obviously off-base, but with no official definition, it's hard to pin down how much Americans overestimate their middle-class status. It's not difficult to understand why they do it, though. Generations of immigrants to the United States have come seeking streets paved with gold. Trying to do better than the next guy is virtually a virtue in our free market economy.

But in 1994, 5 percent of U. S. households held 21 percent of the nation's aggregate income. When people realize that doing as well as the next guy is the best they can hope for, aspiring to middle-class status becomes the goal and the virtue. Although most would rather be rich than not, people often deride those who are out of sheer envy. The word yuppie rapidly took on a derogatory connotation in the mid-1980 s, and consuming in a conspicuous way was considered bad taste in light of others' misfortunes.

As a result, even those who achieve affluence may downwardly aspire to respectable and decent middle-class status-not financially, of course, but as a matter of principle. At the other end of the income spectrum, people who might have proudly described themselves as working class in the past may now feel uncomfortable with the unfashionable term, so they opt for middle class instead. Even as the perception of middle class has stretched to encompass an enormous range of people, the reality is that economic growth does not raise all boats equally. In the decade 1984-94, average household income (in real terms) rose by less than 1 percent. The average income of the poorest one-fifth of households in the U. S.

increased by only 0. 1 percent. At the same time, the top one-fifth of households saw their average income jump by 20 percent. Uneven growth is a characteristic of market-oriented economies. Although economists and others have long recognized the phenomenon, its causes are not well understood. Certainly, market-oriented economies rely on incentives to promote work effort and risk-taking.

Those who respond most capably to these incentives are best rewarded. Still, few analysts conclude that the poor are poor because they are not motivated. Similarly, few would contend that everyone with money got it through hard work and determination. Wages are determined by supply and demand. Sometimes those forces work in one's favor. In other cases, they do not.

For America's middle class, the odds have been against them, but the future may be different. The first thing to note about middle-class households is that they are not exactly a dying breed. Regardless of the specific definition adopted, the proportion of households with moderate income is diminishing. Yet in absolute numbers, they are increasing, just more slowly than other households.

For example, if you define middle-class households as those with incomes ranging around the national average, you find that the share with incomes between $25, 000 and $50, 000 in 1994 dollars shrank from 38 percent in 1970 to 30 percent in 1994. At the same time, the number of such households grew from about 25 million to 30 million. A broader definition of the middle class as households with incomes of $15, 000 to $75, 000 in 1994 yields similar results. With this definition, the middle class shrank to 64 percent of all households in 1994 from 70 percent in 1970.

The middle class may also be defined as households with income between 75 percent and 125 percent of the median. Social criteria, such as educational level or occupation, may also help identify the middle class. Again, regardless of the particular definition adopted or the unit measured-households, families, or individuals-the proportional size of the middle class has been steadily decreasing for more than two decades. It is inappropriate, however, to view the middle class as a stagnant or evaporating pool. Each year, thousands of Americans leave the middle class for greener pastures in higher income brackets or, less fortunately, they fall below middle-class levels. Working against this outmigration is an inmigration from above and below.

The middle class has been declining because the inmigration has been losing steam relative to the outmigration. For some, economic process is ever upward, while for others it is a persistently downward spiral. But for many, it is a series of up-and-down fluctuations, reminiscent of the child's game, "Mother, May I" One giant step forward, three baby steps back. Such swings result from events such as job changes, maternity leaves, and the comings and goings of earners and other household members. Where Americans come and go in their economic status is a matter of some controversy within the economic community. One study, based on a longitudinal survey that followed the same households over a period of time, found that more of the people who leave the middle class move up than down.

The most recent Census Bureau survey data show that the share of households with incomes of $75, 000 or more has doubled in the past 24 years. Other studies, however, discover that more people who depart the middle class move down than up, at least temporarily. Sometimes income swings seem to cancel each other out. Between 1992 and 1993, almost eight in ten Americans saw their economic well-being rise or fall by 5 percent or more, according to the Census Bureau's Survey of Income and Program Participation (SIPP).

Thirty-nine percent saw their family income-to-poverty ratio rise by this much, while 39 percent saw it fall. This level of seeming instability has itself been stable in the six times the SIPP has measured it since 1984-85, ranging from 76 to 78 percent. However, in the late 1980 s, the proportion who were doing better was noticeably higher than the proportion doing worse. The early 1990 s show no such pattern. One measure clearly shows the erosion of income equality in the United States. The Gini coefficient, an innovation of Italian demographer Corrode Gini, indicates the overall distribution of income.

A Gini coefficient of 0 indicates a perfectly equal distribution of income. Everyone is middle class. A Gini coefficient of 1 means that all of the income goes to a single individual, household, or family, and no one else has any. In modern industrial societies, the Gini coefficient never approaches these hypothetical extremes.

But when it increases, income distribution is becoming less equal. The U. S. Gini coefficient rose from.

394 in 1970 to. 456 in 1994. The most often cited cause of the decline of the middle class in the U. S. is stagnant wages.

Between 1955 and 1970, real wages adjusted for inflation rose by an average of 2. 5 percent per year. Between 1971 and 1994, the average growth of real wages was 0. 3 percent a year. The stagnation of wages has been especially noticeable to middle-class people, who rely most heavily on the money they make at their jobs. Upper-income households obtain most of their income from wages and salaries, too, but they also get money from interest, dividends, capital gains, and other non wage sources.

Low-income families are not exempt from problems caused by virtual non growth of wages, either, forcing many to seek relief in government-assistance programs. The softening of wages has come despite, or perhaps even because of, gains in labor productivity. The effects of foreign competition and widespread downsizing are more potent than increasingly small advances in labor productivity. Furthermore, when you need fewer people to run sophisticated manufacturing equipment, the increasing competition for those jobs puts employers in a better bargaining position. As the proportion of jobs in manufacturing has declined, the Gini coefficient has risen. This could be a coincidence.

Manufacturing employment is in a long-term downward trend, and so is the proportional size of the middle class. This alone does not establish cause and effect. However, the fact that the years in which manufacturing employment falls precipitously are the same years in which the Gini coefficient increases substantially makes it harder to dismiss a possible connection. The Gini coefficient for the U. S. has been rising steadily since 1970, but it did decline slightly in 1973, 1974, 1980, 1990, and 1991-all recession years.

This is not surprising. Unequal income growth is part and parcel of economic growth. A recession means negative economic growth, so we might expect to see backpedaling in the way of less income inequality. But the sliding wage explanation for the dwindling of the middle class is separate from the economic growth explanation.

Real wages remain remarkably stable during expansions and recessions. This is because even though there is less pressure for wages to rise during recessions, there is also less pressure for prices to rise. Thus, real wages do not change much over the course of the business cycle. The question then becomes, "What is it about recessions that causes a more even distribution of income" The answer appears to be that recessions hit higher-income households harder, knocking some of them down into the middle class. Recessions also take some middle-income households into the lower-income category, but not as many. Middle-income households may or may not be more likely than higher-income households to qualify for unemployment compensation when jobs are scarce.

But those who do are more likely than high-income households to receive benefits that replace a greater share of their regular wages, which helps them maintain their standard of living. In a similar fashion, low-income households can often partially offset their recession-shocked incomes by qualifying for food stamps, Aid to Families with Dependent Children (AFDC), and other programs. In short, programs like unemployment compensation and AFDC protect those with lower and middle incomes more than they help high-income people. Because of transfer payments, recessions are an income equalizer.

Perhaps one reason why middle-class wages have barely kept up with inflation is the substitution of fringe benefits for cash compensation. Older workers, particularly union members, are more and more frequently obtaining increased health-care or pension benefits instead of higher raises. The result is that it may seem like real wages are stagnant when in fact total compensation is rising. Recessions are not the ideal way to alleviate income disparities, of course.

In fact, none of the things that have caused the decline of the middle class will reverse on their own, and it is impractical to try to reverse them deliberately. For example, it may be possible to take those elements of a recession that support the middle class and apply them in the absence of a recession. Policies that expand income maintenance programs promote income equality. But given the current political environment, this isn't likely. Another way to reduce income inequality might be to expand manufacturing employment. But this won't happen either, because the U.

S. has moved beyond a manufacturing economy to a service and information-oriented one. The promotion of higher wages in other currently low-paying sectors would work just as well as an increase in manufacturing jobs. Practical methods for accomplishing this are controversial, though.

For example, it is not clear that protection from foreign competition will raise domestic wages. Tariffs and quotas can result in retaliatory actions by our trading partners. Furthermore, those same tariffs and quotas raise the prices that consumers have to pay for products. On the positive side, when surveyed, Americans say they are sometimes willing to pay more for products with a "made in U. S." label because they understand that it keeps Americans working at reasonable wage levels.

According to a 1994 study by the Gallup organization for the International Mass Retail Association, 84 percent of Americans prefer to buy American products. Sixty-four percent prefer to buy American products even if they have to pay 10 percent more for them. One way less-skilled workers try to make higher wages is through unions. Although union membership has been declining nationwide for years, unions can still mean the difference between a middle-class wage and close-to-minimum wage for a number of less-skilled workers.

The proportion of employees who are union members varies by industry, but is generally higher in those that are declining as a share of the overall work force, such as construction and manufacturing. A greater union presence in service industries might raise some low-wage workers into the middle-class range. But while economists agree that places with higher rates of union membership have a more equal income distribution, they do not all agree that unions in and of themselves result in higher wages. Likewise, it is not clear whether or how much the recent increase in the minimum wage may help promote income equality. It will presumably move some less-skilled workers above the poverty line, but it may or may not foster growth in the middle class. Less-skilled workers are not the only ones striving to achieve or maintain middle-class status.

Many labor market analysts have recently pointed to the "bifurcation" of wages in the U. S. as an explanation for income inequality. The market richly rewards some skilled workers, while others make wages on a par with unskilled labor. The jobs, and therefore incomes, lost to downsizing may be recouped with training and placement programs. But these sorts of programs have a spotty track record.

Designing an effective program that equips substantial numbers of downsized workers to deal with a restructured labor market is more than challenging. President Clinton has suggested a tax deduction of up to $10, 000 a year to help pay for education and training expenses. It is uncertain how effectively such an incentive will promote employment. Tax reforms, such as the expansion of the earned-income tax credit, can help by lifting net wages after taxes. The current program certainly benefits lower-income households that take advantage of the tax break. Raising the maximum qualifying income could encourage growth of the middle class.

Furthermore, believe it or not, it is well established that our progressive income tax system, despite its loopholes, does work toward equalizing incomes. Yet it hasn't been able to keep the overall distribution of income from becoming increasingly tilted toward the top of the scale. Regardless of changes in government or tax policy, several powerful demographic trends may cause an upswing in the middle class. The first is the fact of slowing labor force growth. The Bureau of Labor Statistics (BLS) is forecasting a 25 percent drop in the rate of labor force growth for 1994 to 2005 as compared with 1988 to 1993.

One reason why the BLS expects labor force growth to slow is that it expects women's labor force participation rates to level off after decades of increase. This will result in about half a million fewer people a year at work or looking for work than if the past high growth had been maintained. If the supply of workers slows and demand is maintained, real wages may rise again. The second factor that may drive up wages is the cumulative effect of increased educational attainment. Labor force participation rates are dropping for young people because more of them are staying in school, thus increasing their skills and earning power. The average monthly income of a worker with any education beyond high school, but without a college degree, is 14 percent higher than for those who went no further than high school.

The third factor that could push up wages is the aging of the labor force. When the baby boom began working, the median age of the labor force dropped to a low of 34. 6 years in 1980. By 2005, it should rise back to its 1962 level of 40. 6 years. Older and more experienced workers with more training are likely to be paid more in real terms.

All of these factors will work in concert to increase middle incomes and perhaps bring back the middle class as we knew it. Whatever the trends, being middle income depends on one's earnings, but being middle class is a state of mind shared by nearly all Americans. Many who can afford the finer things in life will continue to identify with the masses who cannot, and vice versa. The values and lifestyle that middle class has come to signify in the 20 th century will stay with us into the 21 st..