NOTE: In this introductory essay for The Democratic Strategist‘s Roundtable Discussion, TDS Managing Editor Ed Kilgore reviews the history and significance of the perennial issue of base-versus-swing orientations for Democrats, and poses a series of questions whose answers have traditionally divided many observers: (1) Who are the swing and base voters? (2) What is their relative value? (3) What are the opportunity costs involved in reaching beyond the base to swing voters? (4) What’s the best long-range strategy for building an enduring Democratic majority?
While this decade has ushered in a variety of new strategic issues for Democrats, from Internet politics to turmoil in the labor movement, some issues are evergreen. And perhaps the oldest unresolved argument among Democrats is over the nature and electoral value of “swing voters,” those much-pursued and much-maligned counterweights to the Democratic “base.”
Though the debate over “swing voters” has been raging for decades, it’s hard to find a subject more bedeviled by definitional and empirical confusion, by straw men and false choices, and by very different evaluations of recent political history.
It’s this last factor that’s revived the swing voter debate among pollsters, political practitioners, academics, bloggers and journalists.
To cite the most simplistic versions of a common argument, in one narrative of recent Democratic electoral performance, Bill Clinton broke the party’s long presidential drought by intelligently targeting swing voters. His successors, Al Gore and John Kerry (along with congressional Democrats in most cycles between 1994 and 2006), failed to completely follow the Clinton template. Republican abandonment of swing voters (politically and substantively) led to the big Democratic midterm victory of 2006.
A competing narrative suggests that Clinton’s pursuit of swing voters alienated the party base, blurred essential distinctions between the two parties, and forfeited the Democratic majority in Congress and in the states, while failing to produce a presidential majority. Gore and Kerry failed to match Bush’s relentless efforts to energize the Republican base, and Democratic fretting over swing voters made the party a weak and ineffective opposition party. That finally changed in 2006, when a netroots-led mobilization effort based on maximum partisan differentiation produced a Democratic counterpart to the base-driven Republican landslide of 1994.
It’s notable that each narrative diverges sharply over interpretation of the 1994 debacle, the 2000 “draw,” and the 2006 breakthrough. And there is naturally (though not universally) a strong ideological underpinning to the debate, with those on the party’s “left” typically disparaging swing-voter-focused campaigns and governing strategies as unprincipled and disloyal, and those in the “centrist” camp often arguing that base-focused campaigns cede critical ground to the GOP and make effective governing impossible.
The base-swing argument has many variations, of course. Most centrists favor a party message and agenda that’s congenial to both base and swing voters, and at most suggest keeping highly partisan base mobilization efforts “under the radar screen.” And most progressives believe in swing voter appeals that don’t conflict with sharp partisan differentiation and ideological principles, even if they sometimes seem to yearn for an election (as some hope for in 2008) where swing voter appeals are no longer necessary. Both camps agree that exposing GOP extremism can be an effective tool for both base mobilization and swing voter persuasion.
But even if all goes well in 2008 and this dispute does not become a major point of contention among Democrats before November, it will remain a semi-submerged problem for any Democratic administration and Congress in terms of designing a governing agenda. And while it would be naïve to think that this ancient argument can be completely resolved here or anywhere else, it would be helpful to create some general agreement on the terms of debate, and on certain empirically verifiable common ground.

Who Are Swing and Base Voters?
“Base” voters are easier to define: those who predictably and loyally vote for Democratic candidates up and down the ballot, in good years as well as bad, though their propensity to show up at the polls may vary. While some analysts occasionally use the term “base” to refer to activists or self-identified liberals, they are clearly subsets of the broader category of loyal rank-and-file Democratic voters.
Definitions of “swing voters” vary far more, leading to very different opinions about their electoral value. At one extreme, during specific campaigns, “swing voters” are often identified with “undecided voters.” At the other extreme, there’s a tendency in some media quarters to conflate “swing voters” with self-identified “independent voters” or “moderate voters.”
James Campbell of SUNY-Buffalo has sought to navigate the definitional problem by differentiating between “campaign swing voters” (i.e., “undecided” voters) and “pre-campaign swing voters.” He derives the latter category through a sophisticated analysis of National Election Studies data about the ideological and partisan self-descriptions of voters prior to election campaigns. But relying too heavily on voter self-identification is risky; there are many millions of Americans who vote regularly for one party or the other but insist on calling themselves “independents,” and fewer, but still significant, numbers who think of themselves as partisan or as “liberal” or “conservative” yet “swing” in particular elections.
To keep the definition as objective as possible, Mark Gersh of the National Committee for an Effective Congress has suggested that swing voters be defined as “those who belong to some category of the electorate with an unstable attachment to the major political parties.” And even that definition needs to be qualified to deal with the declining-but-still-significant segment of voters with variable attachment to the parties at the presidential and state-local levels.
Determining the length of voting behavior that denotes “stable attachment” is another difficult problem. While the “Reagan Democrats” who provided the first and most famous universally acknowledged modern swing voter category are now largely Republican or deceased, it may be a bit early to write off or count on other, more recent, swing voter categories, such as older rural voters, married women with children, first-time voters, Protestant Latinos, or Greens.
The definitional problem, of course, vitally affects assumptions about the size of the swing vote. James Campbell (whose overall thesis is that swing voters rarely “swing” elections) calculates the median percentage of the electorate that has been “in play” prior to presidential elections since 1972 at 22, while the corresponding percentage for “swing” voters during election campaigns has been 16. Virtually every analyst believes that the swing vote has been declining over time, with Gersh’s 2004 estimate of about 10 percent of the electorate being meaningfully “in play” being a conservative figure. Another key sub-question is whether “swing voters” are even “voters.” Obviously, weighing the value of swing voters against base voters depends on an assessment of likelihood to vote.
What’s the Value of Base and Swing Voters?
If swing voters are at least as likely to vote as base voters, then there is one indisputable factor that makes them valuable beyond their numbers: a “turned” swing voter adds one vote to your column, and subtracts one vote from the opposition column. Mobilizing a “base” voter to show up at the polls adds one vote at most, and if the mobilization effort helps the other party mobilize its own base, the net effect could be far less or even negative (viz., the much-discussed “backlash” against the Wellstone Memorial event in Minnesota in 2002). This “net effect” is also important in weighing appeals to different types of swing voters. For all the talk about what Al Gore might have done to attract Nader voters in 2000, the fact remains that votes denied Bush had more electoral value than votes denied Nader.
A broader issue is whether swing voters actually “swing” elections. James Campbell, using his above-described definitions of pre-campaign and campaign swing voters, deduces that only two presidential candidates since 1952 have actually had to win a majority of swing voters to win the election, and that some candidates (i.e., those who won big) could have won without any swing voters. The different sizes of the two major party’s bases, he suggests, have largely predetermined most recent presidential elections, with base loyalty and turnout being as important as swing voter behavior.
Even accepting Campbell’s definitions of swing voters, you don’t have to believe candidates must win a majority of such voters to value them as having an exceptional pound-for-pound and dollar-for-dollar significance.
The more important question about swing voter appeals is whether such voters can be turned not only into an Election Day asset, but into future base voters. I’ll discuss that issue a bit later.
What Are the Opportunity Costs of Appealing to Swing Voters?
Here is where the argument about swing voters gets especially complex, and where agreement over basic terms is essential to any empirically-based examination of the evidence.
The traditional “progressive” argument is that swing-voter-oriented strategies and policies are politically suspect and ideologically disastrous. As evidenced, they say, by the experience of the 1990s, lusting after swing voters at the price of discouraging the base and sacrificing key policy ground (e.g., welfare reform, NAFTA) shifts the center of balance in American politics to the Right, and requires still more concessions to conservatives in order to seduce still more swing voters. It’s a political version of the “race to the bottom” argument against trade liberalization.
But there are newer arguments that deny a conflict between swing and base strategies while repudiating the Clinton-era appeal to swing voters as both unprincipled and ineffective. One theory, associated with linguist-turned-strategist George Lackoff, suggests that swing voters are by-and-large cognitively conflicted, and are best persuadable through a base-friendly progressive message that elevates progressive over conservative “frames.” This approach has been embraced by many netroots activists as well.
Many Democratic centrists, however, argue that the most compelling argument to make to swing voters is about the real-life accomplishments of the Clinton Administration, as compared to the real-life failures of the Bush Administration. This essentially non-ideological approach also denies any real base-swing conflict or tradeoff.
But some centrists go further, and suggest that any political message that does not sway swing voters is disqualified as an effective governing agenda. Center-Left policy achievements, they believe, are not a matter of exploiting temporary Democratic electoral majorities, but of building a long-term governing majority. And on that point, they converge with some folk on the Democratic left who argue that spurning centrist appeals to swing voters can, in the end, produce policy results that attract swing voters without sacrificing progressive principles.
What Is the Long-Term Strategy For Dealing With Swing Base and Swing Voters?
This last question gets to the nub of the swing-base debate. Is it about how to win in particular elections, or how to build a broader party coalition, i.e., to reduce the number swing voters? Is there a conflict between short-term and long-term goals?
With the exception of a few party centrists who think that bipartisanship and cross-ideological appeals are ends in themselves, and a few party progressives who think that any majority coalition involves unacceptable compromises, most Democrats would prefer to live in a political world in which the Democratic base is so large that swing voters are expendable, and the essential policy arguments affecting the country are mainly held within our own ranks.
The empirical debate we need is aimed at discussing how we get to that Democratic majority; how big an ideological tent we must pitch to encompass a majority; how temporary Democrats can be made permanent Democrats; and how to create a party agenda and message that increase civic engagement, voting and real-life policy results.
We may never agree among ourselves about all the details of what we should do. But after a long period of conservative ideological ascendancy and partisan gridlock, it’s time to get the basics right.