Book Summary:  Alan I. Abramowitz, The Disappearing Center: Engaged Citizens, Polarization, & American Democracy, Yale, 2010. 

In The Disappearing Center Professor Abramowitz investigates the degree of polarization among the American electorate, charting the rise of “partisan-ideological” politics and the demise of the political center. He emphasizes the need to distinguish the engaged from the disengaged citizen.

Alan I. Abramowitz is a Professor of Political Science at Emory University

Preface
Our most informed and engaged citizens are polarized partisans. Moderates are the least informed and least engaged segment of the electorate. The center is disappearing.

For Abramowitz polarization is not confined to a political elite, indeed: “The central argument of this volume is that “there is no disconnect between the political elite and the American people.” But the fact that the American electorate is more engaged than ever “can only be a good thing for American democracy.”
 
Chapter 1  Polarization in the Age of Obama
Abramowitz notes the increased partisanship of the 2004 presidential race. President Bush had become a polarizing figure: “Very few voters were ambivalent.” But Bush was the product of an pre-polarized party system. The most recent roots of that polarization can be traced to Nixon’s southern strategy, Reagan’s “courtship of conservative Democrats”, Newt Gingrich’s takeover of the Republican Party, and the Clinton impeachment battle.

In 2008 Obama promised to change the partisan ways of Washington.  He tried to reach out to Republicans but the “ideological chasm between the parties was simply too wide to bridge.” It appears polarization is here for the foreseeable future.

Polarization has been both good and bad.  It has energized the electorate and created clear choices between candidates, however, many observers think polarization is alienating moderates.Moreover the ideological sorting of the parties has brought us closer to the British model of responsible party government (in which the majority party implements its program unimpeded).  But the American constitution is anti-majoritarian—the minority party can easily frustrate the majority’s agenda.  
[The remainder of Chapeter 1 is an outline of the themes to be discussed in each chapter]
 
Chapter 2   The Engaged Public
In his seminal study from the 1950s, “The Nature of Belief Systems in Mass Publics”, Philip Converse concluded that in America ideological thinking was confined to an elite. This notion endures today notably in the work of Morris Fiorina who thinks the American electorate is largely unmoved by the disputes animating political elites. But says Abramowitz, data from ANES (American National Election Studies) surveys since the 1960s reveal that far from being turned off by politics the American Public is more engaged and informed than at any time in the past 50 years. Ideological polarization among elites seems to have encouraged more engagement among the electorate. As the partisan gulf has grown those citizens with partisan-ideological tendencies perceive more is at stake and thus become more engaged than moderates.

Abramowitz asks if moderate activism is possible. He cites the case of Ross Perot. Only 15% of Perot supporters identified strongly with either party proving that a third party candidate can indeed engage true independents and moderates. But the increasing polarization of the parties means that third party candidates are less viable. As policy differences diverge citizens fear wasting their vote.

In sum, the American public is diverse in its level of engagement. When party leaders become more polarized so does the public. And if some voters are turned off it seems many more are energized.
 
Chapter 3 Partisan-Ideological Polarization
As the electorate has become more educated it has become more ideological. Among all voters in 1972 the correlation between ideology and party identification was.32. In 2004 it was .63. For the politically engaged the correlation rose from .47 in 1972 to .77 in 2004.

Abramowitz argues that it’s the engaged citizen who matters when it comes to elections.  The engaged citizen constitutes an increasing segment of the population and an even larger segment of the voting public.  And the “growing consistency both across issues and between issue positions and party identification” has led to what Abramowitz calls “partisan-ideological polarization.”

Results from the 1984-2004 ANES show that “it was the least interested, informed, and politically active Americans who were clustered around the center of the liberal-conservative spectrum. The more interested, informed, and politically active the voter, the more likely they were to take consistently liberal or consistently conservative positions.”

Bush-Kerry 2004 showed a spike in polarization.  The ideological measure of all voters showed that 56% of Democrats were liberal versus only 12% of Republicans.  Among Republicans 73%  were conservative versus 21% of Democrats.  Then among the engaged public:  82% of Democrats tested liberal versus only 7% of Republicans.  91% of Republicans were conservative, 12% of Democrats.

The National Exit Poll from 2006 showed that 21% of voters described themselves as liberal, 34% conservative, and 45% as moderate. Abramowitz remarks that the popularity of the moderate label does not imply that the elecorate is truly moderate.  Both “moderates” and “independents”  usually have an ideological orientation.To wit, the 2006 CCES data shows that 91% of voters were either Democrat or Republican or leaned towards one or the other. Independents are much closer to partisans than true independents. And Independents who leaned Democratic were much more liberal than those who described themselves as weak Democrats.  Similarly independents who leaned Republican were much more liberal than self-described weak Republicans.

Further analysis of the CCES survey showed that 41% of nonvoters fall in the center of the liberal-conservative scale–only 12% of nonvoters are at either the liberal or conservative end. For Abramowitz, it follows that only nonvoters are non-ideological.

Ideology has become the driver of politics and is now a better predictor of partisanship than social background.  Voters and candidates are aligned ideologically. Abramowitz ends with a caveat: “…there is a risk that those citizens who for whatever reason lack a consistent ideological outlook will become increasingly alienated from the two major parties and from the electoral process itself…”
 
Chapter 4  Polarization and Social Groups
The collapse of the Democrat’s New Deal coalition (1933 to circa 1969) has changed the political landscape. That coalition was held together by three main groups—southern whites, northern white Catholics,  and northern blue-collar workers.Behind the collapse was the ideological realignment following the civil rights legislation of the 60s. Nixon’s southern strategy and then Reagan’s appeal to the remaining conservative democrats were key episodes in the course of realignment.  By the 21^st century liberals were mostly Democrats and conservatives were mostly Republicans, each supported by about half of the electorate.

African-Americans remain the one voting bloc based on group identity rather than ideology. Even black conservatives overwhelmingly support Democrats. For white Americans gender, marital status, and religious commitment have become the best predictors of voting behavior. 

The Gender gap exists because white men fled the Democratic party in greater numbers than white women. White women still makeup about half of the Democratic party but white men fell from nearly 55% to less than 40%.

By 2004 religiously commited whites who were politically engaged were twice as likely to be Republican as Democrat. Religious commitment became a better predictor of voting than class. Families with incomes greater than 100,000 but who weren’t regular church-goers were far less likely to vote for Bush than families with incomes less than 30,000 who attended frequently.  
 
Chapter 5 Polarization and Elections
The swing vote used to be central but now campaigns focus on mobilizing the base.

In the 1950s and 60s voter turnout was very high but partisan identification was low. Citizens were mainly motivated by a sense of civic responsibility but now partisanship drives voters– they want to defeat the opposition. In the 50s strong partisans voted just a bit more often than pure independents. By 2004 the pure independent turnout fell from 80% to 50% while the strong partisan’s rose from 80% to 90%.
In the 1950s strong partisans rated their preferred candidate 35 degress higher on the temperature scale . In 2004 they rated them 60 degrees higher. Pure independents stayed the same, rating their candidate about 25 degrees higher.Abramowitz concludes that polarization discourages the pure independent vote while strong partisans– having more incentive—become over represented.

Ideological realignment has increased party loyalty.  The 2004 ANES shows that 91% of voters who leaned or identified Democrat voted for Kerry and 92% of Republican leaners/identifiers voted for Bush. 

Then there is geography. The south has become a lock for Republicans, the Northeast a lock for Democrats. Presidential elections are merely competitive at the national level. In 2004 only 12 states were competitive.  The average margin of victory was 15.8 pts per state. In 2008 it was 17.4.
Many see gerrymandering and the rise of safe districts as the major cause of polarization. Safe districts have indeed risen–from 1976 to 2004 they rose from 122 to 216. But Abramowitz cites a study of 2000-2002 showing that districts drawn by nonpartisans  actually ended up with more safe districts than those drawn by partisans.  Abramowitz concludes that ideological realignment plus demographic factors such as geographical sorting and immigration have been far more important than gerrymandering.

Still, the rise of safe states and districts means that “there is little need to bother appealing to swing voters or supporters of the opposing party, and indeed, such appeals could alienate core party supporters and potentially encourage a primary challenge.” Indeed in the closely contested swing states of 2004 both parties rallied the base.

Finally, primaries are no longer competitive, incumbents are rarely challenged. Abramowitz remarks that the incumbent who remains in the ideological mainstream is likely to resist challenge. 

Chapter 6 Polarization in a Changing Electorate
In 2008 both Obama and McCain campaigned on promises of working across the aisle and ending partisan bickering as though such were the root of gridlock.  But the problem is partisan-ideological polarization. When there is polarization on the issues from taxes to health care to education and so on– it doesn’t matter whether Republicans and Democrats get along.

Gallup polls show a steady increase in the public’s interest in presidential elections. In 2000 38% said they think about the election “quite a lot.” In 2004-58%. 2008-71%. Abramowitz emphasizes again that it’s only the inactive uninformed citizen who is not polarized.

The education level of the electorate has risen dramatically. In the1950s 80% of the electorate had no college education but by 2004 the percentage was under 40. College-educated citizens are more ideologically aware,more politically engaged…more consistently liberal or conservative.  The upshot: “By increasing the level of ideological sophistication in the public, rising levels of education have been an important factor in the growth of partisan-ideological consistency in the American electorate over the past several decades.” This trend predicts more polarization and more partisan conflict.

Importantly, voter demographics have changed dramatically.  In the 1950s 95% of voters were white. Married white Christians made up about 80% of all voters. But by the 21^st century MWCs were only 40%, for MWCs under 30 the fall was from 80% to 20%. Within the same period nonwhite voters increased from 6% to 26%, non-Christians from 6% to 18%, and unmarried voters from 14% to 41%.  These trends favor the Democrats. MWCs have become the core of the Republican Party being about twice as likely to be Republican as Democrat.

Youth. Results from 2006 show that Democrats won the youth vote (ages 18-29) by 22 points. In 2008 they preferred Obama by 36 pts. Abramowitz states that the reason young voters are more likely to vote Democratic is that they “are much less likely to be married,white, and Christian.”  This demographic trend will likely continue and is another advantage for Democrats.

Capturing more of the vote from the changing electorate would require Republicans to make real policy changes. But this “would clearly risk alienating a large segment of the party’s conservative base. And so far, at least, this is something that few Republican leaders have appeared willing to do.”
 
Chapter 7 Polarization and Representation
Congress is now polarized along partisan lines.  To show how recently this developed Abramowitz compares the 95^th Congress (1977-1979) to the 108^th (2003-2005). In the 95^th 32% of Democrats were “strong liberals” and 16% of Republicans were “strong conservatives.” In the 108^th strong liberals  rose to 51%, strong conservatives to 63%.  In the same period “moderate liberals” among Democrats stayed nearly the same dippping from 37% to 36% while “moderate conservatives”  among Republicans fell from 50% to 34%. Noteworthy is that conservatives became more polarized than liberals during this period.

And then “moderates” (as distinct from “moderate liberals” or “moderate conservatives”) have become  a species verging on extinction.  Within the Democratic party “moderates” fell from 29% to 13%. Moderates among Republicans plummeted from 33% to 3%.

Abramowitz returns to the question of redistricting (see also chapter 5).  Many commentators believe dominant state legislatures have made safe districts for their party.  Gone is the need to appeal to moderates or to capture swing votes. Politicians need only win primaries and primaries are about the base—home to the more extreme views. Hence a self-reinforcing cycle of polarization.

Abramowitz challenges this view. For one thing the Senate is just as polarized as the House even though it’s not subject to redistricting. And while it’s true that from the 95^th to the 108^th Congress safe Democratic districts rose from 71 to 103 and safe Republican from 51 to 100, the data show that most of the changes took place between redistricting cycles.

Again Abramowitz invokes demographic change and ideological realignment as primary causes of polarization. We increasingly congregate among the like-minded. Red states and districts grow redder and blue states bluer because the electorate has been sorting itself.

There there is the view that the sorting of the strongly conservative south into the Republican Party is a major factor in polarization. But the numbers reveal that polarization has increased in every category or subgroup be it southern, northern, or whether in districts safe or marginal:  “Since the increases in ideological polarization within each category were almost as large as the overall increase, it appears that neither the declining competitiveness of House districts nor the growing southern presence in the Republican Party explains more than a small fraction of the overall increase in ideological polarization…”

Significant is that data from the 2004 ANES shows that Republicans and Democrats living in marginal districts were just as conservative or liberal as those in safe districts. Likewise, representatives from marginal districts were just as liberal  or conservative as those from safe districts. Only Democrats who represented Republican districts and Republicans who represented Democratic districts proved significantly more moderate than their colleagues.

While allowing  redistricting a part in polarization Abramowitz concludes: “Polarization in Congress reflects polarization in the American Electorate.”
 
Chapter 8 Polarization and Democratic Governance
Polarization is compatible with a parliamentary democracy.  But our constititution is anti-majoritarian—it allows for divided party control among the branches of government.  And in the Senate a supermajority of 60 votes is required to avoid a filibuster and a two-thirds majority must prevail in both the House and Senate to avoid a presidential veto.The Senate is especially anti-majoritarian–small states are over represented. The 20 least populous states representing only 10% of the population elect 40% of the Senators.  A senator from Wyoming represents half a million citizens while a senator from California represents 37 million. Abramowitz concludes that a resident of Wyoming has 70 times the influence as a resident of California. Democrats are adversely affected since they tend to live in the most populated areas of the country.

How to overcome gridlock? Either through partisan dominance of one party or bipartisan compromise. Periods of partisan dominance made possible FDR’s New Deal, Johnson’s Great Society, and the Reagan Revolution.  But such dominance seems unlikely today, especially since according to surveys the American people actually want more partisanship.  

And bipartisan compromise is less likely given the paucity of moderates. In the past the center held against the extremes but today “…bipartisanship means that those on the opposing side should acknowledge the error of their ways and change their positions.”  Alternately, in practice, bipartisanship means keeping your party unified while capturing enough of the opposition’s remaining moderates to avoid a filibuster.  The few remaining moderates may exercise disproportionate influence.

And then politicians are unwilling to compromise for fear their core supporters will make them pay.
 
Abramowitz muses that Obama’s success may well depend on playing to moderates. But whether such a strategy works, “the result will almost certainly be to reinforce the partisan and ideological divisions in Congress and among the engaged public.”