Executive Summary

American voters have become notably volatile in the 1990s. First, in 1992, they shattered the Republican presidential coalition. Then in 1994 they took away 52 House seats from the Democrats and gave Republicans control of Congress for the first time in 42 years. And now, as the 1996 elections approach, they look ready to re-elect a Democratic president and possibly return a Democratic majority to Congress.

Some interpret this volatility as suggestive of big ideological swings in the electorate; others say changing values are behind these electoral shifts; still others point to the increased role of religion in politics. But in this paper we argue that these explanations are only partial, and that the chief cause of voter volatility lies in declining living standards and the persistent failure of either political party to successfully address this problem.

Among our findings are the following:

• The group that is driving this political volatility consists of the non-college educated (particularly non-college-educated whites), and it is this group for whom living standards have deteriorated the most. The sharp swings among non-college-educated whites include a decline of 21 percentage points in support for George Bush between 1988 and 1992, a decline of 13 points in support for Democratic House candidates between 1992 and 1994, and a decline of 14 points in support for Bob Dole between early 1995 and mid-1996.
• In 1992, Perot voters were overwhelmingly non-college-educated whites (67%), who were experiencing sharper declines in wages than were Bush or Clinton voters. Perot voters were also convinced that the United States was in a long-term decline, were likely to believe the next generation would have a lower standard of living, and were likely to be economic nationalists in their attitudes toward trade.
• Despite some healthy economic indicators, voters in the 1994 election had much to be concerned about. Between 1992 and 1994, the median wage fell 3.3%, even as the economic expansion continued. Consistent with post-1979 economic trends, this wage decline was not equally distributed, with wages for the non-college educated declining in line with the median wage trend, while wages for the college educated actually increased. Finally, median household income declined slightly in the two-year period, leaving it 6.6% below its 1989 prerecessionary peak.
• In the 1994 election, declining support for the Democrats was concentrated exclusively among the non-college educated, the chief victims of these economic trends. Compared to 1992, support for Democratic House candidates declined 10 percentage points among high school dropouts, 11 points among high school graduates, and 12 points among those with some college. It held steady among those with college degrees.
• The shift away from the Democrats in 1994 was most pronounced among non-college-educated whites. Among white men with a high school education, Democratic support declined 20 percentage points (to 37%) and among white men with some college, Democratic support declined 15 points (to 31%). But non-college-educated white women also deserted in droves: among both white women with a high school diploma and those with some college, Democratic support dropped 10 percentage points.
• Since the 1994 election, the median wage has continued to decline: it is now 4.6% below its 1989 level. These trends have contributed to a climate of economic anxiety within which Republican attempts to cut Medicare and other government supports seem intolerable. As a result, non-college-educated whites have shifted from strong support for Dole (54-42%) in early 1995 to strong support for Clinton (55-40%) in mid-1996.
• Factors other than declining living standards have played and do play a role in this electoral volatility, but the role is smaller than generally supposed. Take ideology, for example: while a 7 percentage-point increase in the proportion of conservatives in the electorate can account for a minor part of the Democratic decline in the 1994 election, a huge anti-Democratic shift among economic “pessimists” in that election accounts for much more. And while values, broadly defined, may have helped set the context for the 1994 election, a values issue like crime had no statistically discernible effect on voters’ support for House candidates.

The Democrats, with their historic commitment to activist government and the economic interests of the average worker, might seem best situated to formulate an approach to voters that addresses the problem of living standards and secures the support of the non-college-educated on a continuing basis. So far they have not succeeded in doing so, for several reasons.

The first is the general dominance of an anti-government story about declining living standards, a story that tilts the political terrain in favor of the Republicans. The second is the weakness of the Democratic approach to countering that story. This approach seeks to shift the political terrain in two ways-by improving the party’s negative image, and by blaming declining living standards on a neutral process of globalization and technological change, about which little can or should be done other than adapt and wait for living standards to rise again. Unfortunately, the image part of the approach leaves untouched the dominant, anti-government story about the decline in living standards. And the globalization argument merely promotes hopelessness.

The final reason the Democrats have been unable to shift the political terrain is that they have ceded so much ground to the Republicans that there is little they can do-or even talk about doing-to raise living standards. Indeed, they have imprisoned themselves, along with the Republicans and most of the economics profession, in an “iron triangle” of economic policy principles (a high-unemployment, high-interest-rate, slow-growth macropolicy; a commitment to a balanced budget; and free trade without labor and social standards) that effectively excludes any active attempt to improve the lot of the average American.

For Democrats to have a chance at long-run success, they must build a political alternative that breaks out of this iron triangle, raises living standards, and consolidates support among non-college-educated, particularly white, voters. Recent developments suggest a promising direction. A strand of public thinking is emerging about declining living standards that is challenging the dominance of the anti-government story. This thinking focuses on the ways in which corporations and other dominant interests are taking advantage of economic change to enrich themselves and break down the institutions that have enabled ordinary workers to prosper. Indeed, this “new economy populism” is now so powerful that it sometimes outweighs anti-government sentiments in polling results. More importantly, it provides a compelling rationale for breaking out of the iron triangle and asserting the centrality of government action to raise living standards.

But if this new economy populism provides a potentially effective rationale for government action and programs, it runs the risk of seeming detached from the basic values that animate so many voters and anchor their lives. Our analysis also indicates that, for most voters, particularly non-college-educated voters, economics is a values issue, and it must be dealt with in those terms. Therefore, we believe that an effort must be made to synthesize new economy populism and the core values-loyalty, fairness, hard work, and responsibility-that most Americ
ans share. In fact, only if these voters believe that their values and their values-based economic struggles are truly understood by the party of government, and that government action can and will be designed to help these struggles, are they likely to return in large numbers to the Democratic Party.

Of course, there is no guarantee the Democrats will embrace this “new synthesis” anytime soon. Perhaps it will take another 1994-style defeat to convince them their current strategy is weak and that the issue of living standards cannot be avoided. They can only hope that, by that time, non-college-educated voters, especially white non-college-educated voters, are still listening to Democrats, instead of Pat Buchanan, Ross Perot, or a new breed of supply-side Republicans smart enough to leave Medicare alone while cutting taxes and less-popular government programs. In that case, it could be a long road back.