Discussion of Red State, Blue State

The political website Talking Points Memo is featuring a discussion of Red State, Blue State this week. The discussants so far have included software developer / political activist Aaron Swartz, historian Eric Rauchway, political scientist Nolan McCarty, journalist Steve Sailer, and Jeronimo and myself. It’s interesting to see all these different ways of looking at things.

As Swartz perceptively points out, our book doesn’t really have much of a persepective–it’s more of a guided data dump–so there aren’t really a lot of things for people to argue about. It’s not like I’m George Lakoff saying that the Democrats should do X, and other people can say, no, they should do Y. Nonetheless, the discussants manage to bring up some interesting points.

I led off the discussion with this brief summary of the book and the questions it left unresolved.

My recent post, responding to the discussions so far, is here. I like what I wrote, but it’s a little ridiculous that I spent 2 hours on it, thus violating the academic norm to spend so much time on something that’s not a peer-reviewed article. All I can say is, I hope some people read it! Just to be on the safe side, I’ll repeat it here:

I want to respond to the points raised so far, but first I want to point out how irrelevant the red/blue divisions are to much of politics.

Talking Points Memo, the host of this forum, is unusual in the blogosphere in focusing on news, rather than opinion. And the recent headlines on TPM are about potential torture prosecutions, wiretapping, Israeli spying, kickbacks.

What do these news items have in common? None of them are about economic policy, which the evidence shows is the dominant factor in swinging votes and elections. (Briefly, there are three pieces of evidence that the economy is crucial to voters: first, national economic conditions do a reasonable job at predicting national elections; second, people who express more liberal or conservative economic views are more likely to vote for Democrats or Republicans; third, when you ask people what issues are most important, they overwhelmingly mention economic concerns–and that was true last summer, even before the economic meltdown (and even before John McCain found it necessary to reassure us that the fundamentals of the economy were still strong).

But foreign policy is important! National security is important! Having foreign spies influencing U.S. policy is something that most of us don’t want! And, as to kickbacks involving powerful corporate and government officials . . . well, that stuff goes on all the time, but it’s got to be a good thing for it to be reported and have consequences, I’d think. Although I have to admit I’m not up on the research on this topic.

Which brings us back to Red State, Blue State. I’m an expert on public opinion and voting, not on foreign policy. I’m not an expert on economic policy either, but I do know about its connection to voting. For most of the TPM audience, I suspect that the main appeal of Red State, Blue State is the potential to better understand why some people vote Republican, and what positions the Democrats should take, or need to take, in order to win elections. People who read my postings at New Majority are mostly coming from the opposite perspective.

I think that one reason our book has attracted discussion from both left and right (here at Talking Points Memo and elsewhere) is that both sides have the recent experience of losing important elections. 2008 was less of a shock to Republicans than 2004 was to Democrats, but both sides recognize that data are important; the activists of neither party are in the overconfident mode of Democrats in 1974 or Republicans in 1994. Knowing that you might lose is a great motivator for wanting the facts about public opinion.

Now to the points raised in discussion, which I’ll take in order.

– Commenter Merrill noted that the election of 1896 was nearly the mirror image of the 2008 election. This is true at the state level, but at the county level the patterns are more complicated, with different things happening in different states. See here for further discussion of this point, along with some graphs.

– In his discussion, Aaron Swartz commented that we have lots of facts but not a convincing story. I do think we have a story–red states and blue states differ, but among the upper middle class, not so much among poorer voters–but I agree that we don’t have a story as clear as Thomas Frank’s or Paul Krugman’s, nor do we lead readers to a sense of what should be done next, as Frank and Krugman do.

Swartz writes, “The overwhelming tendency among [political science] is to just run the numbers. . . . Interviews have their flaws, of course — everybody lies, after all. But numbers have their flaws too. We need both.” I completely agree, and his point could be made even stronger by noting that “the numbers” are typically based on survey interviews, which have their own problems. Beyond this, I take these comments very seriously: in addition to being a political activist, Aaron Swartz is a remarkably successful software developer and has a real sense of what people find useful as tools and how people communicate.

My quick answer to Swartz is to agree completely and just cite the principle of division of labor. I am not an accomplished interviewer, I don’t have a great store of political anecdotes, but I do have a lot of experience as a data analyst. So that’s how I think I can give my best contribution. Let me lay out the facts, and then others can take the next steps.

Sure, I have my own opinions about who are the good guys and the bad guys in American politics–but I know enough to know not to claim any particular expertise on these opinions. In writing Red State, Blue State, I was lucky enough to have four collaborators of widely varying political persuasions, and between them they made sure to keep most of my wild speculations out of the book. (But I did win the battle to keep the G. K. Chesterton quote in chapter 3!)

– Eric Rauchway asks about races, which in the context of Ameican voting usually refers to four different things:
1. Blacks voting differently than whites.
2. Whites’ votes being determined (to some extent) by their attitudes toward blacks.
3. The extent to which differences between Democrats and Republicans can be explained by 1 and 2 above.
4. Other racial divides, most notably involving Hispanics.
(Feel free to replace the above terms by “African Americans,” “European Americans,” and “Latinos,” if you prefer.)

I do regret not including a chapter on race/ethnicity in Red State, Blue State: we had enough material on the topic, and it would’ve been good to have it all in one place. The short answer is, Yes, of course ethnic voting is alive and well among voters of all income and racial groups. Blacks gave Obama something like 95% of the vote (compared to 88% for John Kerry in 2004), whereas whites in Alabama and Mississippi gave him about 10% (compared to about 10% for Kerry). Regarding point 3 above, if you “control for race” (that is, restrict our analysis to whites only), the differences in the income-and-voting pattern between red and blue states reduces by about half. That is, about half the pattern we see is explained by ethnic minorities being poorer, on average, and more Democratic than whites.

I agree with Rauchway that trends in the South have been important in the whole country’s politics. It used to be said that political parties are like neighborhoods: when the blacks move in, the whites move out–to the detriment of the Democratic Party in the late 1960s onward. But more recently, the analogy of the revolving door has seemed more appropriate: with the Republican Party getting ever-stronger in the South, its national stances have at times approximated the Texas Republican Party platform, which is not so popular in many other parts of the country. In Red State, Blue State, we draw upon the work of Steve Ansolabehere and Jim Snyder to discuss how the one-person-one-state court rulings of the 1960s made the South more politically hospitable to the Republicans and helped set this process in motion. (In short, the mid-1960s redistrictings gave more political power to the urban and suburban South while taking representation away from rural areas. The urban and suburban southern whites were more racially moderate and more economically conservative than their rural counterparts and thus a better fit for the national Republican Party at that time.

That said, one reason we did not focus on race in our book is that differences between rich and poor (or, more precisely, between higher- and lower-income voters) are important in their own right. Whatever a state’s ethnic composition, it’s interesting and important to see how its economic divisions translate into political divisions.

– Nolan McCarty connects our comparisons of voters and states to his influential work connecting income inequality and political polarization. McCarty and his colleagues have shown how, since the 1950s, the Democratic and Republican parties have diverged in how they vote in Congress on economic issues. We thought this point was so important that we reproduced their graph in our book.

To amplify Nolan’s point: with the help of political scientist Adam Berinsky, who specializes in historical survey data, we traced the rich/poor voting gap (more precisely, the difference in voting patterns comparing those in the upper and lower third of income) back to 1940, and what we found is that in the past couple of decades we’ve returned to the levels of income or class-based voting of the 1940s. Back in the 1940s, the upper third was about 20% more Republican than the lower third. This gap declined to near zero in the Eisenhower/Kennedy period, then returned through the 1960s and 1970s to its earlier level. When you look at the series starting in 1952, it looks like a steady rise (with a slight flattening off and dip since 2000), but if you go back to the 1940s, it looks more like a steady line with a brief interruption during the moderate presidencies of Ike and JFK. (We have the graph in chapter 4 of our book.) Beyond this, the graph reminds us that there was no golden age of class-based voting. Again, this fits in well with Nolan’s emphasis on the connection between voting and policy cleavages.

Also, as Nolan points out, this pushes the “why do people vote the way they do?” question back to “why do the parties do what they do?” And the answer to this question is not at all obvious. As we discuss in chapter 9 of Red State, Blue State, reviewing our own work and that of others, the electoral motivations for congressmembers to be moderate are real but they are weak: most individual members of the House and Senate have quite a lot of flexibility in the policies they could hold, and still have a very good chance of reelection. Just as our work on public opinion provides the basis for further why-style analyses, McCarty et al.’s analysis of party positions provides the key data that have to be explained to understand Congress.

– Also in this discussion I’d like to mention the work of political scientist Morris Fiorina, who in his book Culture War? (with Sam Abrams and Jeremy Pope) explains how Americans can appear more partisan–with Democrats becoming increasingly unhappy with the Republican Party, and Republicans going in the opposite direction–even though voters are not, by and large, becoming more extreme on the issues. Fiorina shows how this has happened because the national parties have become more cohesive and, in a word, partisan–not such a bad thing from the standpoint of accountability, but not always conducive to national unity or centrist policies. We discuss Fiorina’s ideas, and our own related work (joint with Delia Baldassarri) in chapter 8 of Red State, Blue State, and it’s central to our understanding of trends in voting patterns.

– In his discussion, Steve Sailer returns to my original motivation for studying this stuff, which is to understand the differences between individual voters, differences that seem more passionate than ever. Why are some Americans staunch Republcans–remember, TPM readers, John McCain got 46% of the vote!–while others, in the words of Tommy Lasorda, bleed Dodger blue?

Sailer gives an explanation that ties into geography–red and blue states–by way of property values: the idea is that more expensive areas such as New York, Chicago, and, nowadays, Los Angeles, are more associated with culturally liberal attitudes, while people in less expensive metropolitan areas are likely to have larger families and then become more conservative. Or the story could be slightly different but with the same result, by positing a selection effect, with family-values conservatives moving to less expensive cities. Or maybe it’s more of an economic story, having to do with people in more crowded cities being more willing to pay taxes for public services.

I like Sailer’s theory–which is why I featured it in our book–partly because it fits into another fact we emphasized, which is that the much-remarked-upon red-state, blue-state, rich-state, poor-state pattern of Democrats in the coasts and Republicans elsewhere is relatively new, having arisen in the past thirty years or so, with the key transition occurring during the brief period from 1980 to 1992. What else happened since the 1970s? Property values in New York, San Francisco, and other economically successful cities shot through the roof. Sure, even thirty years ago, New York apartments were still small, but they weren’t so expensive.

Along with all of this, attitudes on social issues have become more tied to partisanship (see, for example, the graph here on attitudes toward abortion, which is taken from page 118 of Red State, Blue State), so that it makes more sense for people, especially at the upper income range, to vote their values.

It’s hard for me to know what to make of all these theories connecting economics, lifestyle, ideology, and voting. I suspect there’s a pretty big literature on this in political science, sociology, and economics. Together, my coauthors and I are as qualified as anyone to digest and interpret that literature, but by the time we were putting the book together, we had enough hard data that we thought it was safest to stick with description and restrict most of our speculations to the final two chapters.

– Jeronimo Cortina discusses the implications of the comparative analysis that appeared in chapter 7 of our book. I don’t have much to add here except to emphasize that many patterns that appear unique to the United States also occur in other countries, for example the pattern of religious attenders voting for more economically conservative parties, and persistent regional divisions within countries. I thought it would be good to include Mexico in the book because Mexican politics is strongly regional, but with a different pattern than in the U.S.: the conservative PAN has roots in the richer north of the country, while the leftist party is stronger in the poorer south. In some way, recent Mexican elections recall the voting patterns of the U.S. election of 1896.

I welcome further discussion in the next few days. One of the fun things about this sort of forum is the opportunity of interactions between scholars, journalists, and political activists. (I was involved in a similar roundtable discussion last October, involving John Zogby, Bill Bishop, and Valdis Krebs, but in that case there was very little back and forth: each person said his own thing and that was about it. Perhaps this works better because we’re all discussing a single book.)