Partisan identities are more complex than you think
Humans are flawed, irrational, tribal creatures- even in complex, NFL-playing societies
Note: This was written as the answer to the question “How do I know whether I’m a Democrat or a Republican?” on Quora.
At the time I’m writing this, there are 78 answers to this question. Exactly one of them is close to the truth in my opinion, and it’s the one with the solitary upvote I gave it: Joel Dykstra's answer to How do I know if I am a Democrat or Republican? Half the answers to this question are flowcharts telling you which party you belong to depending on the way you lean about certain issues. The other half are naked putdowns of the other party. The first are, in my opinion, wrong. The second are worse than wrong.
All the answers that tell you which party you likely belong to, based on what positions you hold? They come from a popular conception of a voter, from what is called an instrumental view of partisanship. It holds that partisanship is “a running tally of party performance, ideological beliefs, and proximity to the party in terms of one’s preferred policies”[1]. Basically, it assumes that people look at what they believe, and then at what each party stands for, and make a rational choice. There is a competing expressive approach, which believes that party identification is a social identity - which means, as Mason (2015) puts it, that “a partisan behaves more like a sports fan than like a banker choosing an investment.“ [2]
I think the truth is likely to be closer to the second than the first, even more so in a highly polarised political environment like the US.
Keeping this in mind, here’s the short answer: If you have to ask, you’re probably neither. Party affiliation is a social identity, like gender, citizenship, or fandom of Community. Some people hold it close to their heart and form an identity around it. Others hold it loosely. If you have to ask whether you’re a Democrat or a Republican, politics probably isn’t a strong part of yours. (And that’s okay).
For the longer answer, I want to answer a slightly bigger question: How do people in general decide whether they are a Democrat or a Republican?
First, I want to lay out why I think the instrumental view is farther from the truth than the expressive view. Then, I’ll try to frame what I think makes people actually decide either way.
I. Instrumental view of partisanship
The instrumental conception of partisanship holds that people:
Have their own ideological preferences and policy views
Keep a running tally of the parties’ stands and performance on them
Decide which party to support based on that
As it turns out, voters do exactly none of those things.
1. Voters aren’t exactly ideologues. Converse (1964) first showed that voters don’t really have a consistent system of beliefs or a political ideology[3]. This has since been demonstrated multiple times. Though people readily come up with ideological labels for their own positions, they often don’t have “real” opinions on most topics, coherent policies and ideologies, or even stable positions over time[4][5]. Only a minority of about 20–30% of the population show “polar, coherent, stable, and potent ideological orientations”[6].
2. A large chunk of voters don’t actually know what parties stand for. Converse wrote in the paper linked above: “about 17 percent of the public could both assign the terms ‘liberal’ and ‘conservative’ correctly to the parties and say something sensible about what the terms meant.” As late as 2019, after the Tories in England had spent three years trying to implement the Brexit referendum, 20% of voters didn’t know what they actually stood for[7]. In 2012, about 40 years after Roe v Wade, about 40 per cent of the US population could not say which party generally supported restricting abortion.[8]
Simply put, people aren’t exactly in tune with the party positions, outside of a (large) minority of politically engaged voters.
3.The causality between party support and political positions often runs the other way.
Here’s the most important part: People don’t choose a party to support based on their political positions, as the instrumental view posits, but often tend to change their political positions based on the party they support.
Let me take a couple of examples.
3A. Social Security in the 2000 election:
As Achen and Bartels (2016) in their book Democracy for Realists [9] demonstrate with the help of Lenz’s (2012) [10] research (emphasis mine):
In the 2000 presidential campaign, for example, candidate George W. Bush advocated allowing individual citizens to invest Social Security funds in the stock market, thereby catapulting a previously obscure policy proposal into the political limelight. Much of the news coverage and advertising in the final month of the campaign focused on the candidates’ contrasting stands on the issue; in a typical “battleground” media market, the two candidates together ran about 200 ads touching on Social Security privatisation just in the final week before Election Day (Johnston, Hagen, and Jamieson 2004, 153– 159). And, sure enough, the statistical relationship between voters’ views on Social Security privatisation and their preferences for Bush or Al Gore (holding constant party identification) more than doubled over the course of the campaign.
This is exactly the sort of shift we might expect if voters were attending to the political debate, weighing the competing candidates’ policy platforms, and formulating their vote intentions accordingly. However, Lenz’s more detailed analysis employing repeated interviews with the same people demonstrated that this substantial increase in the apparent electoral impact of views about Social Security privatisation was almost entirely illusory— due not to changes in vote intentions, but to Bush and Gore supporters learning their preferred candidate’s position on the issue and then adopting it as their own.
[…]As John Zaller (2012, 617) put it, “Partisan voters take the positions they are expected as partisans to take, but do not seem to care about them.”[11]
3B. Views on Abortion and Partisanship:
The US Supreme Court ruled on Roe v Wade in 1973. While both parties were divided on this issue initially, the GOP slowly adopted a clear opposition to abortion. Quoting from Democracy for Realists:
The 1976 Republican platform favored “a continuance of the public dialogue on abortion and supports the efforts of those who seek enactment of a constitutional amendment to restore protection of the right to life for unborn children.” In 1980, “While we recognize differing views on this question among Americans in general— and in our own Party— we affirm our support of a constitutional amendment to restore protection of the right to life for unborn children. We also support the Congressional efforts to restrict the use of taxpayers’ dollars for abortion.” In 1984, “The unborn child has a fundamental individual right to life which cannot be infringed. We therefore reaffirm our support for a human life amendment to the Constitution, and we endorse legislation to make clear that the Fourteenth Amendment’s protections apply to unborn children. We oppose the use of public revenues for abortion and will eliminate funding for organizations which advocate or support abortion.
Essentially then, as the parties slowly adopted diametrically opposite views on this issue, the voters had a simple choice: they could change their party, or change their opinion. Achen and Bartels find a clear trend:
...both male and female Republicans were more likely to leave the party if they held liberal abortion views, but the partisan shift was only substantial among women, whose abortion views were more likely to be fundamentally bound up in their identity as women. Indeed, more than one- third of the non- Catholic Republican women who expressed pro- choice views in 1982 had left the party by 1997.
[…][With the reverse effect,] about half of 1982 pro- life Democrats had become pro- choice by 1997— twice as much change as among pro- life Republicans. As expected, the effect was larger for men than for women, with more than half of the Democratic men who said they were pro- life in 1982 having switched to pro- choice positions by 1997.
Basically, if you were a pro-life Democratic man in 1982, you were likely to have become a pro-choice Democratic man in 1997. Only some of the people for whom their gender identity was more important than their partisan identity or position - pro-choice Republican women - switched their party. Everyone else changed their opinion to fit better into their party.
With this evidence, the idea that voters line up a running tally of the parties’ stands and performances and decide which to support seems unlikely to be true. To me, a voter making a rational choice between parties based on issue positions seems like a naive, quaint assumption, a friend of the poor sod always making rational economic choices.
II. How do voters choose to align themselves with a party?
I think there’s a strong case to be made that voters, when they subscribe to a party, are answering a slightly different question: Which party is closer to me, or people like me? (In the words of Achen and Bartels, “For most people, partisanship is not a carrier of ideology but a reflection of judgments about where ‘people like me’ belong.”)
That's a question of identity and belonging. For each of these groups, there are perceived stereotypes (of a median Democrat and a median Republican), and people decide which tribe they belong to based on their perceived closeness to that perceived median.
Let me lay out that case.
1. The Parent connection
This is a well-established finding: Parents’ political affiliation often predicts their children’s, and it endures significantly throughout the individual’s life. In fact, moving away from home is one of the main factors that contribute to a person’s political attitudes. [12]This effect is often stronger the earlier it is formed in life, and parental influence often endures when people stay in an environment that reinforces that influence.[13] Actually, people’s political affiliation is often shaped by what they perceive their parents’ political affiliation to be, and their relationship with their parents[14] - deciding whether they want to reject that affiliation. People are more likely to change their party affiliation if they live in an incongruent neighbourhood (a Republican living in an overwhelmingly Democratic county or vice versa).[15]
To me, this evidence implies heavily that political affiliation forms a part of people’s identity. It is formed concurrently when people develop an identity in the first place, and is likely to change only when people are likely to be subject to different societal pressures that may lead them to have an identity very different from their parents’.
2. Party Identification and the Southern Realignment
In the recent past, the results of the Civil Rights Movement, and the Democratic Solid South consequently becoming solidly Republican, provides us with an interesting case study for looking at partisan identification. Quoting from Democracy for Realists again: (emphasis mine)
[The accompanying figure] shows a rapid and substantial shift in the presidential voting behavior of southern whites after 1964, facilitated by George Wallace’s third- party candidacy in 1968 and the Republican Party’s “southern strategy” under Richard Nixon. However, the party identification of southern whites evolved much more slowly. Indeed, Republicans did not outnumber Democrats among southern white party identifiers until the 1990s— more than a quarter century after this group began voting solidly and consistently for Republican presidential candidates.
[…] a closer look at the available survey data suggests— contrary to the conventional view, but just as we would expect—that these shifts in party identification were only weakly related to white southerners’ views about specific policy issues.
[…]The most fraught racial policy issue in this period was also the focus of the question asked most consistently in ANES surveys— whether the government should enforce racial integration of public schools. When this question was first asked, in the early 1960s, the substantial majority of southern whites who opposed government- enforced racial integration overwhelmingly identified as Democrats. By 2000, the remaining opponents of integration (still about 40%) were solidly Republican. However, as figure 9.4 shows, their partisan conversion was remarkably gradual; as late as the 1980s, opponents of school integration were still more likely to think of themselves as Democrats than as Republicans.
If voters rationally choose between parties based on the policies they support (and evidently, the opposition to integration was big enough in the south to drive large-scale migration), this lag between voting and party alignment would simply not exist, period. It exists because it takes much longer for someone to change the way they think of themselves - from a Democrat to a Republican - than it does to go and vote for a different guy. It’s a matter of identity, just like “southern” is.
3. My party is close to me
Group identity is a funny thing. It’s uncomfortable to think that members of your group do not agree with you. There are two ways around that discomfort: you can either change your beliefs to fit the party, like Democratic and Republican men on abortion; or, you can simply ignore your party’s views and claim that they are the same as, or close to, your own.
As I had explored a bit in my answer on swing voters, the ANES asks respondents to place themselves on a 7-point scale of ideology, from “Extremely Liberal” to “Extremely Conservative”. They ask similar questions on different issues, for example about spending and taxes (from “many more services” on the left to “reduce spending a lot” on the right) for both the respondent and both parties.
I hope you’re not bored of Democracy for Realists, because here we go again (emphasis mine):
[This ANES data] allows a researcher to calculate how close the respondent thinks each party is to her own position on the issue. Suppose that on the seven- point scale, a respondent sees the GOP as one unit away from her own position and the Democrats as four units away. Then we will say that she favors the Republicans by three points, which is the difference in those two distances. We refer to differences of this kind as “issue proximities.”
[…] suppose that no rationalization were occurring. Then consider a Democrat and a Republican, both of whom placed themselves at 2 on a 7- point scale. If they both see the parties accurately, they will each be exactly the same distance from each party, and thus they will each have exactly the same relative issue proximity. Their partisanship should be irrelevant. On the other hand, if rationalization is occurring, each respondent may perceive her own party as closer to her than it really is, and perhaps also perceive the other party as further away than it is.
Among the 24% [of the respondents] who placed themselves in the middle of the seven- point issue scale, Republicans and Democrats differed by a full point in their assessments of issue proximity. That is, Republicans believed their party was more than half a point closer to the middle of the scale, on average, while Democrats believed their party was almost half a point closer to the same midpoint. Obviously, both could not be right.
The partisan disparities were even larger for people whose own positions did not happen to fall at the midpoint of the seven- point scale. Liberal Democrats and conservative Republicans each quite sensibly saw themselves as closer to their own party, and increasingly so as their own positions became more extreme. However, people whose own positions did not match their party’s— liberal Republicans on the left and conservative Democrats on the right— were entirely impervious to the implications of that fact for issue proximity. Liberal Republicans managed to convince themselves that the Republican Party was just as close to them as the Democratic Party was; conservative Democrats were equally adept at convincing themselves that the Democratic Party was just as close to them as the Republican Party was. Again, we need not stipulate where the parties “really” stood on the issue of government spending and services to see that someone’s perceptions have gone badly astray here.
Again - No, I’m not tired of pointing this out yet, why do you ask? - if people chose their parties based on where they were and where they saw the party, this would have no reason to exist.
4. Representative sample?
Perhaps the most fun evidence suggesting that people think of parties as a social group identity comes from a study conducted in 2015[16].
Republicans think 46% of Democrats are black. Democrats think 39% of Democrats are Black. The actual percentage is 24%.
That’s not an anomaly. Republicans think 38% of Democrats are gay, Democrats think 29% of their party is gay, while the actual proportion is 6%. Though only 8.7% Democrats are atheists, Democrats think about 24.5% are, and Republicans estimate it to be about 36%.
It goes the other way too: Only 2.2% Republicans earn more than $250,000 a year. But Democrats think that’s 44% of the GOP, and even Republicans think it’s a third of their party. 21% Republicans are 65+, but Democrats think that number is 44% and Republicans think it’s 38%.
People think of both parties as a mishmash of stereotypes - and their thinking is more stereotypical about the out-group than the in-group. Their evaluation of both parties comes from there - and the combination of that and their identity decides their partisanship.
That Democracy for Realists, in case you hadn’t noticed, is a highly recommended read.
Liked this? Please consider sharing it with someone who would enjoy it!
And of course, if this was shared with you and you’d like to see more of this, please subscribe below!
Footnotes
[1]https://cpb-us-e1.wpmucdn.com/you.stonybrook.edu/dist/f/1052/files/2018/02/HuddyBankert2017-21ddz83.pdf
[2]https://www.democratic-erosion.com/wp-content/uploads/2018/03/Mason-2015.pdf
[3]https://www.tandfonline.com/doi/pdf/10.1080/08913810608443650
[4]https://calgara.github.io/Pol157_Spring2019/Freeder%20et%20al.%202018.pdf
[5]https://nkalmoe.files.wordpress.com/2013/09/kinder-kalmoe-apsa2008.pdf
[6]https://az659834.vo.msecnd.net/eventsairsthcusprod/production-ispp-public/ccbd201fde2e4562880495eb819799fb
[7]For what do people think the parties stand?
[8]What the Public Knows about the Political Parties
[9]Democracy for Realists: Why Elections Do Not Produce Responsive Government
[11]WHAT NATURE AND ORIGINS LEAVES OUT
[13]The Family and Partisan Socialization in Red and Blue America
[14]Accounting for the Child in the Transmission of Party Identification - Christopher Ojeda, Peter K. Hatemi, 2015
[15]Where You Live and Who You Know: Political Environments, Social Pressures, and Partisan Stability - Jeffrey Lyons, 2011
[16]The Parties in Our Heads: Misperceptions about Party Composition and Their Consequences