Most Americans are moderates

Most Americans are moderates

This piece is written by Milan the Intern, not the usual Matt-post.

Before we jump into today’s column, some personal news: as you may (or may not) know, I’ll be moving to New Haven to start my freshman year at Yale in 54 days. Thank you all for making my gap year amazing — I could not have asked for a better first job! Never fear, I’ll still be here at Slow Boring, but I’m going to be shifting to part-time work starting in August in order to keep up with my studies.

Now, on to the column…

Lately I’ve been thinking about something that my 10th grade U.S. history teacher, Mr. Landwehr, would tell our class all the time: most Americans are moderates. Today’s post takes a deep dive into the American political landscape in an attempt to gauge the accuracy of Mr. Landwehr’s statement, which (if true) has major implications for political strategy.

To spoil the conclusion, he pretty much hit the nail on the head.

What does “moderate” mean?

When you ask people to describe their ideology, the vast majority say they are either moderate or conservative, and that’s been the case for a long time. And moderates and conservatives outnumber liberals in all 50 states.

It is worth noting that “moderate” doesn’t necessarily mean holding centrist opinions down the board. In fact, it’s more often the case that moderates hold a mix of left- and right-wing ideas — such as someone who opposes gun control but supports raising the minimum wage.

And voters often hold inconsistent views. For example, polling indicates that a strong majority of Americans oppose overturning Roe v. Wade. But majorities also favor banning abortion after the first trimester, which wasn’t allowed under Roe.

Now, you might say that this proves that ideological self-identification isn’t a useful metric. Perhaps people don’t really understand certain terms or what certain policies actually are and just say they’re moderate because it sounds nice and reasonable. Or maybe they’re turned off by the stigma around the word “liberal,” even if they hold left-leaning views.

We can test this theory by looking at support for progressive policies among self-identified conservative, moderate, and liberal voters. If ideological descriptors are not just noise, then we’d expect to see more support for progressive policies among liberals than among moderates and more support among moderates than among conservatives. And in this chart from David Shor, that’s exactly what you see.

Image

Each dot represents a progressive policy that was polled. The x-axis is overall popularity, the y-axis is popularity among the relevant subgroup.

There is a substantial age gap among conservatives. Younger conservatives seem to be broadly less, well, conservative than older ones. And polling indicates that large majorities of Americans favor liberal policies such as legalizing marijuana or same-sex marriage, even as self-identified liberals are outnumbered three-to-one.

In order to square this circle, political scientist James Stimson argued that Americans are “symbolically conservative but operationally liberal.” What that means is that if you ask people questions about big-picture values — about the ideal size of government or support for traditional values — most Americans will pick the conservative option. But if you ask people about specific issues — say, whether we should expand Medicaid — then they often choose the liberal option.

People distrust the establishment

Since you’re reading Slow Boring, there’s a good chance that you follow politics closely. If I asked you for your opinion on, say, Chuck Schumer’s performance as majority leader, I’d be willing to bet that you have some thoughts to share.

If I asked my father that same question, he’d probably look at me funny. My dad votes and occasionally watches PBS NewsHour, but beyond that, he doesn’t think about politics that much. Both he and my mother say that their baseline assumption is that most politicians, including the ones they like, are at least somewhat corrupt and self-serving and therefore untrustworthy.

That’s a pretty typical attitude. Most people don’t care much for politics and don’t think a lot about it. They vote or don’t, and they might hear about a big news story like the Supreme Court overturning Roe, but they’re not keeping up with the latest developments in Build Back Better negotiations. Most people don’t really trust the federal government to do the right thing, and they don’t approve of the job Congress is doing.

Progressives sometimes like to read distaste for the establishment as evidence that the public craves big change. But that’s not actually true either.

People like the status quo

A lot of issue polling gives the impression that liberal policies are overwhelmingly popular with voters. You regularly see people argue that there is broad support for a national paid leave program, Biden’s Build Back Better plan, or passing the PRO Act. Progressives then say that the only thing preventing Democrats from enacting these policies is the corrupting influence of wealthy donors.

Twitter avatar for @AOC

The problem is that issue polling can often be misleading. There’s “acquiescence bias,” which is the tendency of respondents to pick the positive option when presented with a binary “yes/no” question. Plenty of polls neglect to tell people which party supports which side, but once you do, views often polarize on partisan lines. If you give people a neutral option such as “no opinion,” “no preference,” or “unsure” there’s often a sizable number of people who end up in the middle (for example, views on immigration are split almost equally between people who want more, less, or the same amount of it). Finally, pollsters can write questions in a deliberately biased manner in order to get a certain set of results for a press release.

David Dayen of The American Prospect has argued for “deliverism” as an alternative to poll-testing. He points out that Barack Obama’s bailout of General Motors ended up being a central part of his re-election campaign (“Bin Laden is dead and General Motors is alive”) despite the bailout polling quite poorly when it was implemented in 2009. Therefore, Democrats should focus on delivering for the people rather than focus-grouping every policy decision.

This is often tied to the narrow argument that Democrats should do, say, student debt forgiveness in order to appeal to younger progressive voters who make up the party’s base. More generally, the argument advanced by progressives is that if Democrats nominate progressive candidates who run on bold policies that make concrete improvements to people’s lives, they will see increased turnout from the base and nonvoters alike, win elections, and implement those policies.

Since Matt has already written about the flaws in the progressive mobilization argument (here and here), I’ll keep things brief and just say that infrequent voters and nonvoters are less liberal than regular Democratic voters, so there isn’t actually that much of a tradeoff between persuading swing voters and getting more people to turn out for you — the marginal voter in both cases is fairly moderate.

In fact, there’s evidence that running more ideological candidates can backfire. Andrew Hall has found that when more ideologically extreme candidates win House primaries, their party sees a lower vote share in the general election. What happens is that the more extreme candidate energizes the other party’s base to turn out in opposition, causing the electorate’s composition to be less favorable to their own party.

If you look at the candidates who win in states usually hostile to their party, the trend is pretty clear. Charlie Baker, Larry Hogan, and Phil Scott are Republicans running some of the bluest states in the country, and they’re the most popular governors in America. In 2019, John Bel Edwards and Andy Beshear won gubernatorial races in states that would give Joe Biden less than 40% of the vote a year later. Joe Manchin overperforms dramatically in West Virginia and so does Susan Collins in Maine. What do they all have in common? Moderation!

Here progressives might note that their causes have won at the ballot box, too. In 2020, Florida voted to raise the minimum wage to $15/hour even as the state went for Trump. Several red states have passed ballot measures legalizing marijuana and expanding Medicaid, often by large margins. These are all policies that I support, and I’m glad the measures passed. But they paint an incomplete picture. When Chris Warshaw looked at all contested ballot measures from 1958 to 2020, he found clear evidence of status quo bias, affecting both liberal and conservative measures.

Twitter avatar for @cwarshaw

Perhaps the best evidence of status quo bias is the well-known phenomenon of the president’s party losing seats in the midterms (averaging 26 in the House and 3 to 4 in the Senate for the postwar era).1

When Democrats control the federal government and pass liberal laws, the public reacts by voting Republican in the midterms (as in 2010), and when the GOP wins and passes conservative laws, people vote blue in the midterms (as in 2018). This pattern is borne out in historical data: Matt Grossman has documented a negative correlation between the number of ideological bills the majority party in Congress passes and its vote share in the next election.

US_Sat1

The reality is that most voters are pretty happy with things as they are. Gallup has been asking people if they’re satisfied with how their personal lives are going since 1979, and the lowest ever share of satisfied respondents was 73%. The measure is currently sitting at 85%. It shouldn’t be surprising that voters often prefer the status quo, even if they express discontent with how things are going in the country overall.

You need to meet people where they are

The fact is that there is not a hidden liberal majority in the United States. Most voters describe themselves as moderate or conservative, and that description means something. When you consider the underperformance of ballot measures on both sides of the aisle, midterm backlash, and the way issue polling can mislead, you see an electorate that also harbors a fair amount of status quo bias.

If you think about it, that makes sense. The United States is the richest country in the world, and while we have our challenges, most people here enjoy a relatively good standard of living and are happy with their personal lives.2

When I turned 18, I registered to vote as a Democrat. The reason I did is that my first real political memory was in 2017, watching the Republicans cut taxes for the rich and try to pay for it by taking healthcare away from poor people. I thought that what the GOP was doing was just morally wrong. That’s the core of why I got involved in politics — because I think we should ask the wealthy in this country to do more to help the needy. I imagine that’s the same reason many progressives got involved. But to accomplish that goal you need to convince people to vote for you, and that entails meeting people where they are.

If you tell everyone who isn’t an orthodox, ideological liberal that you’re not interested in their support, thank you very much, you’re just going to lose. If you talk about your policies as bold, sweeping, transformative changes, you’ll scare off moderate voters who are mostly content with the way things are. But if you go about things in a calmer way, by focusing on incremental change and talking about how progressive policies can also reflect moderate or conservative values, you can convince people to give you a shot.