Return to Book Reviews

Red, Blue, and Purple America: The Future of Election Demographics
Ruy Teixeira, editor
Published by Brookings Institution Press, 2008

A review by Jo Freeman

This is a book about how the electorate has changed since World War II and how those changes affect which party wins and loses elections and where. Published right before the 2008 election — but after the candidates were known — the trends it describes predicted an Obama victory even without an unpopular war or financial crisis intervening to boost the Democratic vote.

Seven chapters by twelve authors explore the ramifications of the changes in geography, class, race, immigration, religion, family structure and generation on the future of party politics. Overall, these changes favor the Democrats, but the Republicans still have a fighting chance.

Written in accessible language, this is still a data-driven book. Its analysis is shaped and limited by the questions pollsters have asked over many decades. Nowhere is this more apparent than in the chapter on the white working class. Unlike race and sex and even religion, class is determined by how it is defined. The co-authors use three different definitions: education, occupation and income. Occupation is the most malleable, but the "working class" can change size and shape depending on where the line is drawn in all three.

The trend is the same regardless of what definition is used. The white working class has declined in numbers and influence and identification with the Democratic Party. Using education as the criterion, in 1940, 75 percent of white adults 25 and over did not have a high school diploma; in 2007 only 14 percent lacked this degree. Only a small percentage currently have blue collar jobs, and the household income of all has risen since mid 20th Century.

In the 1940s, the working class base of the Democratic Party was composed of ethnic, white workers in unionized factories. Post-war prosperity gradually drove most of these into the middle class and the hands of the Republican Party — traditionally the party of the middle class. As manufacturing declined, the unionized portion of the labor force shrunk. Today most whites at the lower end of the socioeconomic spectrum live in rural areas; the better off who can still be defined as working class live in the suburbs. These Reagan Democrats helped the Republicans win many victories since 1980, but they are a dying breed.

A thriving breed is the upper middle class. Higher education for more people created a class of well educated and well off professionals who vote Democratic. Their numbers are growing. On the other hand, the number of high earners with only a college degree are also growing, albeit more slowly. They vote Republican.

Immigration and differential birth rates have brought big changes to our population. Their impact on the electorate is only beginning because many immigrants cannot or do not vote. The political preferences of those that do vote are balkanized by country of origin and region of current residence. Overall, Hispanics and Asians are somewhere between blacks and whites in their preference for Democrats, but much more likely than either to be independents. Obama received roughly two-thirds of their votes, which is consistent with past patterns except in 2004. Since their share of the electorate is growing, this favors the Democrats. Formerly red states with growing Hispanic and Asian populations are turning purple.

Immigration, along with other demographic shifts, has also changed the impact of religion. Hispanics are increasing the total number of Catholics while other immigrants are bringing greater religious diversity to a country where Protestantism has long dominated. In the 19th Century Catholics were Democrats while most Protestants outside the South were Republicans. In the 20th Century these lines blurred. By the the 21st Century the best predictor of party preference among whites was frequency of church attendance. The last few decades have seen significant growth of seculars, evangelicals and "other." Seculars prefer Democrats while evangelicals strongly favor Republicans. Among Hispanics the 19thCentury pattern prevails: Catholics vote Democratic and Protestants (virtually all evangelical denominations) vote Republican. Blacks have remained loyal Democrats since 1964, regardless of religion or religiosity.

One of the most interesting chapters in this book describes "the big sort." It seems that in our highly mobile society we are gravitating toward communities of people who vote like we do. Although national elections have often been close, county-wide votes have become less so. In 1976, 26 percent of voters lived in counties where one party won by at least 20 percent. In 2004, 48 percent of lived in a landslide county.

There are chapters on generation and geography but not on gender. Most of what is said about women is in a chapter on family structure. There we learn that it has changed radically in the last few decades, but the voting patterns associated with different family types have not changed much. Between 1968 and 2004 married voters were the most likely to vote Republican and unmarried with children the most likely to vote Democratic. The gap has grown bigger over time, while the married population grows smaller and the unmarried population grows larger. The fact that 90 percent of single parents are women sort of gets lost in the analysis.

This book is full of information presented in words, tables and charts. The authors offer numerous insights into voting trends, a few surprises, and much food for thought. If you want to know why the 2008 election was not an aberration, read this book.

©2008 Jo Freeman for