America's Forgotten Majority

IF one is to believe the bulk of news stories, the typical American voters these days are affluent white mothers ("soccer moms") and fathers, living in the suburbs and probably involved in the information economy ("wired workers"), whose interest in government reflects their relatively privileged position: "No big programs, please, because we don't really need them, but little, cheap ones are okay, provided they target one of our few remaining problems."

• "Soccer moms.... were America's most wanted voters, and their every wish turned up on some politician's list of promises: child tax credits, education tax breaks, scholarships, V-chips, school uniforms, longer childbirth stays, time off for teacher conferences, even a breast cancer web site. Some called it pandering, others family friendliness" (USA Today, November 6, 1996).

• "Wired workers are the wave of the future, political analysts say. Political parties will learn to surf the new demographics, or go under. Wired workers solve problems as part of self-directed teams, and regularly use computers on the job. They tend to be self-reliant, mobile, affluent, pro-free market, socially tolerant and deeply concerned about educating their children and re-educating themselves. And they are multiplying" (Tulsa World, October 28, 1998).

• "Suburbs vary immensely, of course.... But politicians use the term as collective shorthand for key groups of swing voters: married couples with children, the 'soccer moms' who were so sought after in the 1996 election, affluent independent voters and the high-technology employees who work miles from any city" (The New York Times, May 4, 1999).

If all of this is accurate, then perhaps the extraordinarily cautious and modest nature of today's politics is justified. Large social and economic problems cannot be tackled because the most important voters are too far removed from them.

But if it is not accurate, perhaps we are unnecessarily limiting the role of government and selling the future of our country short. This possibility occurred to us as we pored over accounts of elections in the 1990s and became increasingly suspicious that conventional stereotypes of the American voter missed the mark. We knew, for example, that more than three quarters of American adults lack four-year college degrees, that more than seven tenths do not hold professional or managerial jobs, and that the median income of American households is actually quite modest (about $39,000 in 1998). Could American voters in general, and swing voters in particular, really be so different from what these data suggest? Could it really be true that, as one newspaper story had it, "Nixon's Silent Majority and the Reagan Democrats ... are becoming as hard to find ... as parking spots at the local minimall"?

We became particularly intrigued by the assertion -- explicit above, sometimes implicit, but almost always there -- that the white working class had become politically irrelevant. How could this be? The 1980s weren't that long ago. Demographic change is generally gradual, not sudden. The country is still mostly white (almost three quarters of adults, more than four fifths of voters), and most people have, according to the data just cited, jobs, educations, and incomes that can broadly be described as working-class.

Well, what can't be usually isn't. The white working class is alive and well in American politics today. Sure, many of its members prefer the label "middle class," and most don't work in factories or at any other kind of blue-collar job. But their economic position in American society bears little resemblance to that of the suburban college-educated professionals we hear so much about.

We call these white working-class voters the forgotten majority of American politics: "forgotten" because we haven't heard much about them of late and also because they haven't benefited much from policy changes over the past thirty years or so; "majority" because they are just that -- about 55 percent of the voting population.

THEY weren't always forgotten. Indeed, one important thing to keep in mind is how odd, historically speaking, the current enshrinement of soccer moms and wired workers is. From the New Deal through the 1980s it was widely recognized that white working-class voters were, in one way or another, the key to American politics. The prototypical members of the New Deal coalition, for example, were ethnic white workers -- commonly envisioned as working in unionized factories, but also including those who weren't in unions or who worked in other blue-collar settings (construction, transportation, and so forth). These voters provided the numbers for FDR's four election victories and Harry Truman's narrow victory in 1948, and offered political support for the emerging U.S. welfare state, with its implicit social contract and greatly expanded role for government. In the 1950s the white working class provided the margin of victory for the Republican Dwight D. Eisenhower in two elections. At the same time, it continued to support the expansion of the welfare state, as a roaring U.S. economy continued to deliver the goods and government continued to pour money into roads, science, schools, and whatever else seemed necessary to build up the country. This era, stretching back to the late 1940s and forward to the mid-1960s, was the era that created the first mass middle class in the world -- a middle class that even factory workers could enter, since they could earn a relatively comfortable living even without high levels of education or professional skills.

Things began to fall apart in the 1960s. Though the white working class backed John F. Kennedy and Lyndon Johnson and the liberalism of their era, the emergence of conflicts around race (initially riots and the rise of militant black nationalism; later affirmative action and busing to achieve racial balance) and the Vietnam War weakened their support. These conflicts led directly to Richard Nixon's triumph in 1968 and a stunningly high vote (14 percent) for the third-party candidate George Wallace, as disaffected white working-class voters deserted the Democratic Party en masse (64 percent voted for Nixon or Wallace). White working-class voters were also widely acknowledged to be behind the huge popular rejection of George McGovern in 1972 -- 70 percent of this group cast their ballots for Nixon.

The Democrats came back under Jimmy Carter, of course, but it was a short respite, because by the end of the 1970s white working-class voters had entered an economic world radically different from that of the preceding generation. Slow growth, declining real wages, stagnating living standards, high and variable inflation, and high home-mortgage interest rates were battering them economically. The great postwar escalator to the middle class had basically stopped. And the response of "activist" government was tepid. Or, worse, instead of honoring and encouraging core American values such as equal opportunity for all, fair reward for effort, the centrality of hard work and individual achievement, and social responsibility and order, the Democrats seemed to be focused on liberal social programs to promote the particular interests of gays, women, and minorities. This view was easy to cultivate among white working-class voters in the late 1970s, since many of them believed even prior to those years that the Democrats, owing to their perceived association with extremist elements of the civil-rights and anti-war movements, were out of touch with mainstream values. Thus, even as most white working-class voters were personally moving to adopt much more liberal values with respect to race, gender, and lifestyle, they embraced a pragmatically conservative judgment on government and its priorities.

In 1980 they found Reagan's anti-tax, small-government message appealing, given that the government seemed to be doing less and less for them (57 percent voted for Reagan, just 34 percent for Carter). This time they even had a name:Reagan Democrats. And those white working-class swing voters stuck with the Republicans through two more presidential elections. The stage was set, in the view of many observers, for an era of uninterrupted Republican domination, based on the consolidation of the Reagan Democrats -- a process that appeared to reach the congressional level with the Republican triumph in the 1994 elections.

But the Democrats have won the past two presidential elections, though the Republicans still control Congress. And all of a sudden, after dominating our politics for sixty years, the white working class is nowhere to be found in most media accounts of current politics. We hear a lot about soccer moms, wired workers, and suburban independents, but virtually nothing about this formerly central group of voters. What happened? Has the world really changed so much in the past decade or two? Could the white working class have been rendered irrelevant by the rise of a new economy?

IT is certainly true that there have been fundamental shifts in the American economy and work force over the past generation. For example, service-sector employment has continued to grow, to the point where it now accounts for 80 percent of employment. Only about 25 percent of workers are blue-collar (craft workers, operatives, transportation workers, and laborers), while 58 percent are white-collar (managers, professionals, technicians, salespeople, and clerks). More people are working in doctors' offices than in auto plants, more in laundries and dry cleaners than in steel mills. Finally, the use of computers on the job has exploded, involving half the work force in a very short time.

We may indeed have a "new" economy for most participants, and we have the highest rate of labor-force participation in the world. But an old problem has persisted:finding work that adequately supports people and their families. More than workers in any other developed country, workers in America depend on their pay and job benefits. And for many working Americans the new economy has until very recently been more new than good. Much of the reason for this is the rapid increase in inequality in our society.

How widely the benefits of growth and productivity are shared is a basic measure of economic performance in a democracy -- and, further, widely shared benefits help to generate positive feelings about a society and its government. By these standards the U.S. economy performed superbly for the first three decades after World War II. The rising tide of postwar prosperity really did "lift all boats," as President Kennedy used to say. Indeed, as the accompanying chart indicates, boats at the bottom actually rose a little faster than those at the top. But since the early 1970s, and especially in the past two decades, the gap between rich and poor has grown steadily. National income and wealth, of course, have continued to grow, but because of this increasing inequality, only the top 20 percent of families have made significant gains. The bottom 60 percent, as the chart shows, have barely budged, and some here have actually lost income.

During this period of increasing income inequality the value of a four-year college degree has dramatically increased. Those with one have continued to move ahead; those without one have fallen further behind. For example, from 1979 to 1999 the average real hourly wage rose 14 percent for those with college degrees and 19 percent for those with advanced degrees. In contrast, average wages fell by four percent for those with only some college, 10 percent for those with only a high school diploma, and a stunning 24 percent for high school dropouts. Men among the latter three groups did even worse: they were down seven, 15, and 27 percent respectively.

The difference in prospects between those who have college degrees and those who don't is big enough to warrant its own name: the Great Divide. This Great Divide defines the new white working class. On one side of it are the three quarters of white adults who lack college degrees; these people have not fared well over the past twenty-five years. On the other side are the one quarter of white adults who have a four-year degree or even more education than that; over the past twenty-five years these people have fared very well indeed.

Of course, these non-college-educated whites are not the white working class of yesteryear. They are more likely to be doing low-level white-collar and service work than blue-collar work. They are much more likely to work in an office with a computer or at a similar service-sector job than to work in manufacturing. They are also likely to have more education than the old-style working class -- perhaps some college, maybe even a two-year associate's degree. And those in the work force are much more likely to be female. But in economic terms they are not so different from the white working class of previous generations.

These people are the real swing voters in American politics. Their loyalties shift the most from election to election and, in so doing, determine the winners in American politics. They are also the majority -- about 55 percent of voters and of the adult population. But they don't receive much attention these days; they are invisible to the journalists and commentators who define our national discourse. To bring them into focus more sharply, we will review some basic information about them. As we proceed, it will become clear that the new white working class is quite at variance with dated stereotypes from the 1970s and 1980s.

• The days of high school dropouts are long gone. More than four fifths of the forgotten majority have at least a high school diploma. Almost two fifths have some education beyond high school, and about one in ten has achieved an associate's degree.

• Members of the forgotten majority earn a moderate income. Their median household income is about $42,000 -- on the low side of what is generally considered a middle-class income. About two thirds have household incomes between $15,000 and $75,000. About a seventh are below that range, about a fifth above it.

• They tend to be low-level white-collar and service workers, not unskilled blue-collar workers. More than 80 percent of forgotten-majority workers hold jobs that are not professional or managerial. And even those who do professional or managerial work tend to hold relatively low-level, poorly paid jobs. The few blue-collar jobs that remain are likely to be skilled: only 17 percent of forgotten-majority workers currently hold unskilled blue-collar jobs (even among forgotten-majority men the figure is less than a quarter). • They are less and less likely to work in factory jobs. Today only a relatively small proportion (17 percent) of the forgotten majority works in manufacturing (even among men the proportion is still less than a quarter).

• They live in and dominate the suburbs. The forgotten majority is under-represented in our urban areas (making up just over a third of adults) and over-represented in our rural ones (making up almost three quarters of adults). But in the growing suburbs, generally viewed as the current battleground of American politics, the forgotten majority is represented fairly accurately, making up almost three fifths of adults. In contrast, college-educated whites compose barely more than a fifth of suburban adults.

The conventional view of the suburban electorate -- affluent soccer moms, executive dads -- is drawn from a few relatively wealthy towns like Bethesda, Maryland, and Fair Lawn, New Jersey, and it doesn't come close to reflecting reality. The suburban electorate is in fact composed mostly of members of the forgotten majority: two-earner families of low to moderate education and income, generally working in low-level white-collar, service, and skilled blue-collar jobs.

In sum, the white working class remains numerically dominant, even if its form has changed. Sure, many of its members qualify as wired, in the narrow sense that they work with computers and information technology. Many also qualify as soccer moms, in the narrow sense that they have to juggle job and family, including driving their kids to and from athletic contests. And certainly many qualify as suburban independents, in the narrow sense that they live in the suburbs and lack a strong identification with either party. Nonetheless, they are members of a white working class whose economic interests and experience diverge fundamentally -- in terms of culture, class, and history -- from those of soccer moms in Bethesda, suburban independents in Fair Lawn, and wired cyberprofessionals in Silicon Valley.

It's clear that the economic interests and experience of the forgotten majority set it apart from more highly educated and affluent citizens. But what about values? It's a truism in politics that values matter; some even agree with Ben Wattenberg, the author of Values Matter Most (1995).

Undeniably, people vote on the basis of more than economic issues and experience. And even with respect to economic issues, people's values inevitably affect how they assess policy ideas. But this doesn't mean that voting is therefore all about values. It means that values are always a part of voting, and that separating values from economics in political decisions is an intrinsically difficult and arbitrary exercise. In other words, "values always matter" is closer to the truth than "values matter most."

So, rather than emphasize an artificial distinction between values voting and economic voting, let's focus on the relationship between values and economics in voting. A disjunction between economic experience and values has fundamentally shaped the political behavior of the forgotten majority. The economic experience has already been described. The values we have in mind are deeply held and broadly shared: opportunity, fair reward for effort, the centrality of hard work and individual achievement, and social commitment. As we have argued, over the past quarter century these values have repeatedly been contradicted or called into question by the tremendous slowdown -- and reversal of direction for some -- on that escalator to the middle class. The failure of activist government to get that escalator moving again, together with its apparent concentration on the problems or rights of others (minorities, the poor, gays, even criminals), has persuaded forgotten-majority voters that government is more a part of this values-experience disjunction than the solution to it. The direct and long-lasting result is the sour and skeptical attitude toward government that has become so common today.

THE role of the forgotten majority is crucial to understanding the potential of the strategies being pursued by both major parties, in this election cycle as in recent ones. The ideal strategy, in our view, is one that recognizes the centrality of forgotten-majority voters and seeks to reunite their values with their economic experience -- in other words, to heal the disjunction that has marked the years since the early 1970s. The party that does this should be able to command the long-term loyalty of these voters and thus to grasp and keep political dominance.

This is a favorable time to pursue the project of reuniting the values of forgotten-majority voters with their economic experience. The fast economic growth, low unemployment, and rapidly rising real wages of recent years mean that work is being rewarded in a way that hasn't been seen for a generation or so, taking the edge off the forgotten majority's suspicion of government and the political parties.

With that in mind, how are the major parties doing? Democratic Party strategy suffers from a refusal to recognize the forgotten majority as fundamental to a new popular majority. The Democrats prefer to target various fashionable voter groups as supplements to their base in unions and minority groups and hope that they manage to outpoll the Republicans, as they have in the past two presidential elections. The Democrats also lack a program for uniting the values and economic experience of the forgotten majority; they simply hope that the current economic expansion will last forever, a scenario that cannot happen. And even now the expansion is doing little to solve long-term problems such as health security, retirement security, and education reform, which are crucial to the forgotten majority's economic future. These problems demand bold policy interventions -- interventions that the Democrats are reluctant to propose, given their born-again commitment to fiscal prudence and modest government.

The Republicans are similarly reluctant to recognize the centrality of the forgotten majority, even though their recent electoral successes have depended on support from this group. They have had similar difficulty articulating a program that could reunite forgotten-majority values and economic experience. They remain committed to an anti-tax and anti-government rhetoric that is out of step with the forgotten majority and provides no compelling vision for its economic future. Of course, there has recently been some softening of this rhetorical position (Bush's "compassionate conservatism"), and it is at least possible that anti-tax politics might regain some traction in the event of an economic downturn. But for now the real swing voters in politics are waiting for someone who understands both their values and their economic experience, and the Republicans do not appear to qualify.

PERHAPS the Democrats can move forward without the forgotten majority. They start with about a third of all voters in their base:the combined strength of union-household members, blacks, and Hispanics. If this base can be expanded by getting more members of these groups into the voting pool (either because the groups themselves become larger or because members of the groups turn out to vote in greater numbers), then maybe the Democrats can get along without the forgotten majority. Analysis suggests, however, that this strategy has only limited potential. Take union-household voters. Given the huge difficulties the AFL-CIO has been having in trying to increase the level of unionization in the country, the obvious way for the Democrats to get more of these voters is to increase turnout rates among them. The problem with this strategy is that these voters already turn out at high rates. In 1998, according to the Census Bureau, union members turned out at a rate of 53 percent, as compared with 40 percent among the voting-age population as a whole and 37 percent among non-union voters. Even after education, income, race, and other characteristics are controlled for, union voters simply participate more.

Although this is good news for the Democratic Party, there are clearly limits to how much bigger a payoff the party can get from this group. Could union-household participation be driven to 75 or 80 percent? Possibly, but it would mean that these voters would have to increase their already high turnout rates by 20 to 25 percentage points -- a very tough proposition indeed. That doesn't mean that union turnout rates can't be bumped up in particular elections -- or that union voters can't play a very important role, especially if the election is a close one. But it seems unlikely that a dramatically increased union turnout will provide a way around the challenge of mobilizing the forgotten majority.

Black voters offer a similarly limited resource for the Democrats. Their staunch support for the Democratic Party is one of the political givens of our time. In election after election, for offices high and low, black voters overwhelmingly support Democrats, by 80 to 90 percent or more. Thus it would be pure electoral gold for the Democrats to increase the turnout rate among blacks. But how feasible is this on a regular basis?

A perennial source of Democratic hope, the black turnout is also perennially misunderstood. After the 1996 presidential election, for example, some argued that the black-male turnout had taken a huge jump -- underscoring possibilities for a base-expansion strategy. But this argument, which derived from exit polls, was contradicted by more-reliable census data. (The census survey data contain 90,000 to 100,000 adults per election year and are generally a better source for looking at the demographics of the electorate than exit polls, which have a much smaller sample size and suffer from sampling bias and poor question wording.) These data showed that the turnout among black men had actually declined by four percentage points (and the rate among black women had declined by three points). Black men continue to make up just four percent of the active electorate.

Similarly, 1998 was generally considered to be an exceptionally good year for the black turnout, as African-Americans sought to oppose the Republican Congress and to support President Bill Clinton. But the overall increase in turnout among blacks was only about three percentage points over the rate in 1994, and the increase in the proportion of all voters who were black was about half that. In a year in which the overall turnout dropped, that was still an impressive performance, but it illustrates the difficulty of producing a large national increase in black turnout, even in a favorable situation in which get-out-the-vote efforts are deemed successful. (Large increases in particular state or local races are a different issue.)

Thus it is unlikely that average participation rates among blacks can be increased dramatically. And in any event, blacks remain a relatively small part of the U.S. electorate -- just over a tenth. They are clearly central to the Democratic base, and it's essential that the Democrats retain them. But they cannot substitute for the forgotten majority in the Democrats' quest for a new coalition.

Though Hispanic support for Democrats has been considerably less staunch than black support, and is in fact quite weak in some places (among them Florida), Hispanics overall support Democrats at rates of 60 to 70 percent. Increasing the Hispanic turnout would be a clear boon to the Democrats. But, again, how much and how fast could current levels of participation grow?

An optimistic answer to this question was suggested by reports of a dramatically increased Hispanic turnout in the 1996 election. According to widely cited data from exit polls, Hispanic representation among voters almost doubled (from 2.3 percent in 1992 to 4.5 percent in 1996). This implied an increase of seven percent in the Hispanic turnout (in an election in which turnout overall went down six points) and an overall increase in the Hispanic vote of almost two million. But -- as with the black vote -- census data suggest that the change in the Hispanic vote was much less dramatic. According to those data, the Hispanic turnout actually dropped by a couple of percentage points.

However, because Hispanics grew as a proportion of the population from 1992 to 1996, and because their turnout declined less than did that of the rest of the population in an abysmally low turnout year, their share of the electorate still increased, from just under four to just under five percent: an increase of about 700,000 votes.

Joel Rogers is a professor of sociology, law, and political science at the University of Wisconsin at Madison, where he also directs the Center on Wisconsin Strategy. Ruy Teixeira is a senior fellow at The Century Foundation and the author of The Disappearing American Voter (1992). Their article in this issue of The Atlantic will appear, in somewhat different form, in their book, America's Forgotten Majority: Why the White Working Class Still Matters, to be published by Basic Books this month.

The Atlantic Monthly; June 2000; America's Forgotten Majority - 00.06 (Part Two); Volume 285, No. 6; page 66-75.

The census data for 1998 tell essentially the same story. Despite reports of a huge surge in the Hispanic vote that year, again usually based on exit-poll findings, census data indicate modest increases of about a percentage point, both in turnout and in share of active voters, over the numbers for 1994.

In the long run, Hispanics' growing share of the population will substantially increase their share of the active electorate. According to census projections, Hispanic representation in the voting-age population should grow by more than 50 percent over the next two decades -- rising from a little under 10 percent to about 15 percent by 2020. As a very rough estimate, this might increase the Hispanic share of the voting electorate to about eight percent. In the short to medium run, however, the racial composition of the voting population will change but little. As with union members and blacks, then, so with Hispanics. A successful Democratic coalition-building strategy must hold on to Hispanics, but they do not provide a plausible substitute for increased support among the forgotten majority.

SO an expansion of the existing Democratic base holds little promise for creating a new Democratic majority. The current Democratic coalition -- most emphatically not a majority -- is already doing a fair job of turning out these voters. It could always do better, of course, but there are limits to the likely effect.

Inescapably, the forgotten majority is the answer. Forgotten-majority voters, excluding those in unions, make up close to half the electorate (45 percent). They voted Democratic for the House at a rate of only 39 percent in 1998, and for President at a rate of 41 percent in 1996. Just as the Republicans made great gains in the 1970s and 1980s by "hunting where the ducks are" (in that case, among the expanding ranks of disaffected whites, particularly in the South), so the Democrats have to go after the biggest flock of ducks -- the unorganized, or non-union, ranks of the new white working class. The Democrats don't actually need a majority of such voters. Just breaking even would bring their House and presidential shares of the two-party vote to 53 percent. Obviously adequate to win the presidency, this level of support would also almost certainly be enough to take back the House and would return popular support for House Democrats to pre-1994 levels (the Democrats averaged 53 percent of the two-party vote throughout the 1980s and the early 1990s). Considering the two sexes separately within this target group, forgotten-majority men are clearly the tougher sell. Making up almost a fifth (19 percent) of the active electorate, they voted Democratic at a rate of only 35 percent for the House in 1998 and 33 percent for the presidency in 1996.

Unorganized forgotten-majority women make up about a quarter of the active electorate and are substantially more supportive of Democrats: 42 percent voted Democratic for the House in 1998 and 46 percent for the presidency in 1996.

Promising as the forgotten-majority women's vote is, reaching parity with the Republicans among the new white working class will require some heavy mobilization among unorganized (and, Democratically speaking, disaffected) forgotten-majority men. Even if support among forgotten-majority women could be increased to more than 50 percent, the Democrats would still need to win over about 45 percent of their male counterparts (an increase of 10 percentage points). This might seem a daunting task -- but it should be remembered that House Democrats regularly attained such levels of support among forgotten-majority voters, men as well as women, until the early 1990s.

IF the Democrats can't appeal effectively to the forgotten majority, and if expanding their current base won't be enough to solve their problems, they must increase their support among college-educated whites. Indeed, to hear some orthodox New Democrats, this is the party's major goal. For example, William Galston and Elaine Kamarck, in the Democratic Leadership Council's flagship journal Blueprint (Fall, 1998), argued that since "the New Economy favors a rising Learning Class over a declining working class," and since there is an "educational bias in the electorate" that favors the college-educated, the party must focus on highly educated voters. Setting aside black and Hispanic college-educated voters, who already vote Democratic at extremely high rates, this leaves us with college-educated whites. Simply put, that dog won't hunt.

Substantially more affluent than their working-class counterparts, these voters tend to be less concerned about economic problems and less inclined toward activist approaches to those problems. Moreover, when the very small unionized, Democratic-leaning component of this group is separated out, the remainder makes up just over a fifth of the electorate -- a proportion less than half that of the working class. These college-educated whites voted Democratic for the House at a rate of only 40 percent in 1998, and for President at a rate of 39 percent in 1996.

The group is divided about equally between men and women. Once again, the men are substantially more conservative. Only 31 percent voted Democratic for the presidency in 1996, and 36 percent for the House in 1998. Moreover, other public-opinion data show that men in this group have far more strongly conservative views on everything from the budget and taxes to education, trade, and health care than either working-class white men or college-educated white women. They are unlikely candidates for Democratic conversion.

The Democrats are doing substantially better among the college-educated white women in this group. About 44 percent supported the Democrats for the House in 1998, and 48 percent supported Clinton for President in 1996. These relatively encouraging numbers have strengthened the Democrats' belief that they should target college-educated voters, especially women, rather than working-class men.

This belief is misguided, for two reasons. First, this is a small target group. There are nearly twice as many unorganized forgotten-majority men as there are college-educated women. Simply breaking even among these forgotten-majority men would be equivalent to achieving landslides among these college-educated white women of about two thirds for the House and three quarters for the presidency. Such landslides are beyond the bounds of plausibility.

The second reason is that support for the Democrats among this group is overwhelmingly driven by women with postgraduate education. Fifty-three percent of these women voted Democratic for the House in 1998 and 56 percent voted Democratic for the presidency in 1996, whereas women with just four-year college degrees supported Democrats at corresponding rates of only 40 percent and 43 percent. But women with postgraduate degrees account for only about three percent of the electorate, making them a slender reed to support a strategy.

Of course, the Democrats could focus both on the forgotten majority and on college-educated white women, particularly those with postgraduate degrees. The latter cannot substitute for the former, however. If the Democrats could improve their support among the forgotten majority, additional support from highly educated white women would be icing on the cake. But without the forgotten-majority voters there will be no cake to ice.

SO the Democrats must appeal to the forgotten majority. What of the Republicans? Their position is basically a mirror image of the Democrats': they do very poorly among union-household voters, blacks, and Hispanics; they do relatively well among unorganized forgotten-majority voters and among more-affluent, college-educated voters. In general they do particularly well among men. This presents the Republicans with two basic options: cut into the Democratic base among union households and minorities, or increase their support among unorganized whites, particularly women.

Can the Republicans crack the Democratic base of union-household members and minorities? Possibly, but only with great difficulty. Given its size, the bloc of union-household voters is the most inviting target. But at least for the purposes of national politics, unions are now better organized than they have been in some time, and they are more resolutely Democratic than ever.

Blacks seem, to put it mildly, an even less likely target. These voters have shown no signs whatsoever of straying from the Democratic fold in recent elections, and Republican overtures toward them are likely to be a wasted effort. The Republicans should probably announce, again and again and again, that their party is not racist. But persuading black voters won't be easy.

Perhaps the best bet for cracking the Democratic base lies in the Hispanic vote. Hispanics not in unions voted 39 percent Republican for the House in 1998. In some state elections, including the most recent Texas gubernatorial election, Hispanic support for the Republicans was much higher (that election, notably, involved George W. Bush, the certain Republican presidential nominee). Although it would be a mistake to read these figures as a harbinger of sizable defections from the Democratic camp, they do suggest an openness to Republican appeals among the rapidly changing and diverse Hispanic population. So Hispanics are worth pursuing, to be sure, but the fact remains that their electoral weight nationwide will remain relatively slight for many years (though in state elections they will be important in the Southwest and in California).

Breaking into the Democratic base, therefore, seems an unpromising strategy for the Republicans -- at least as a primary focus. Instead they will have to concentrate on strengthening their advantage among unorganized whites, returning their levels of support among those voters to those of the Reagan-Bush years and of the congressional election of 1994. The most effective way for the Republicans to increase their congressional support is to improve their support among unorganized white women. In the 1998 election Republicans received almost two thirds of the votes of unorganized white men. It is probably unrealistic for them to aim for a still higher share, though they must maintain the share they have; forgotten-majority men provided about 30 percent of all Republican votes in 1998. It would be more realistic to gain votes among their female counterparts, whose support levels were seven to nine points lower. If the support of forgotten-majority women alone could be increased to match that of their male counterparts, the Republicans would have their new majority. For the support of college-educated white women to have the same effect, the Republicans would have to receive an overwhelming three quarters of their votes. It seems highly implausible that they will be able to do so.

THE political arithmetic of the American electorate is unforgiving. Whether the major parties like it or not, the road to the next successful political coalition runs straight through the forgotten majority. The Democrats have to reach toward the break-even point among these voters, especially by strengthening their performance among men (while continuing to turn out their current base). The Republicans have to intensify their dominance of this group -- especially by enhancing their performance among women -- or break into the Democrats' base. These are daunting challenges for both parties, with no obvious or risk-free solutions.

What would it actually take to mobilize -- and keep -- forgotten-majority voters? As we have said, the core values of the forgotten majority must be reunited with their economic experience. Consider the following possibilities.

• If a criminal has a right to a lawyer, you have a right to a doctor. This was a great line from Harris Wofford's successful Senate campaign of 1991, and it sounds the right note for the forgotten majority. Hardworking, law-abiding citizens should be provided with access to health care. They should not be left out in the cold just because they're unlucky enough to lose their jobs or to work for companies that don't provide affordable health insurance.

• People who work hard all their lives should have an adequate income after retirement. It is not fair to punish those who earned too little to save much for their retirement or who worked for employers that didn't provide pensions.

• Americans have a right to the best education their tax dollars can buy. In this rapidly changing economy the children of American workers must have access to quality education -- elementary, secondary, college, and beyond. The more quality education a person has, the better he or she will do economically.

• People willing to work hard should be able to get the training they want for the jobs they need. In the new economy people frequently have to or want to change jobs. They should not be penalized because they can't get access to training they would be perfectly willing and able to go through.

• In today's global economy everybody has a right to a decent wage and everybody has a right to speak out and organize. American workers shouldn't be competing with workers in other countries whose wages are artificially depressed by the absence of even minimal standards and democratic freedoms. That's not fair to workers abroad or at home.

• People who work hard should also be able to spend enough time with their families. Hard work shouldn't destroy family life and parents' relationships with their children. That's not the American Dream. We have to find ways to give workers more time to spend as parents, instead of the other way around.

• Parents who work outside the home should have access to affordable, quality child care. Nobody who wants or needs to work in today's society should have to sacrifice the welfare of his or her children to do so.

• We should make whatever investments are necessary to keep the economy growing. Solid economic growth benefits all Americans who are willing to work. It's a good use of tax money to spend whatever is necessary to maintain and safeguard this growth.

These goals build on the core values of the forgotten majority -- opportunity, fair reward for effort, the centrality of hard work and achievement, and social commitment. And they would give everybody in the forgotten majority a fair shot at an upstanding, reasonably prosperous and secure life -- that is, a middle-class life as it was once understood.

It is important to emphasize that the list above is a set of goals, not policies. And since they are goals, they could -- and should -- be adopted by Republicans, although policies to meet these goals are currently more closely associated with Democrats. Indeed, we would argue that the Republicans have, rather than a seeming ideological hostility to government, a great interest in embracing these goals and in highlighting their commitment to education and other social issues (in budget negotiations last year Republicans actually proposed more funding for education than was requested by President Clinton).

Of course, Republicans and Democrats will have different ideas about how to use government to achieve these goals. But each party must try to achieve them if it means to build a durable majority. The insecurities of the new economy cannot be remedied without effort; the need to confront them politically is inescapable. Whichever party acknowledges this first, together with the need to overcome the new austerity (the bipartisan consensus that paying down the national debt is far more important than spending money on new programs), will reap the electoral rewards.

Until then, neither party is capable of implementing a serious program to address America's problems at scale, since the opposing party can effectively veto it. We are left instead with a politics of small gestures and incremental changes, fueled by intense partisan conflict. This makes for profoundly ineffective governance, especially when measured against the challenges of a new century. With the support of the forgotten majority, however, we can do better. We can revive active, strong government and build a twenty-first-century prosperity that includes all Americans.

Joel Rogers is a professor of sociology, law, and political science at the University of Wisconsin at Madison, where he also directs the Center on Wisconsin Strategy. Ruy Teixeira is a senior fellow at The Century Foundation and the author of The Disappearing American Voter (1992). Their article in this issue of The Atlantic will appear, in somewhat different form, in their book, America's Forgotten Majority: Why the White Working Class Still Matters, to be published by Basic Books this month.

The Atlantic Monthly; June 2000; America's Forgotten Majority - 00.06 (Part Three); Volume 286, No. 6; page 66-75.