Diversity, Equality, and Realignment

Though we like to think of political parties in the US on a single dimension, this obviously doesn’t capture the true range of political views. Yet even though we can conceive of massively more multifarious policy differences than this, only two dimensions explain the vast majority of political behavior in the US: one largely concerned with the distribution of economic resources and another with the treatment of African-Americans and other racial minorities.

Yet even the four permutations of these views are too many to perfectly represent within the US’s two viable political parties. The US political system seems almost to have been designed so that most citizens will hate the dominant political parties, and thus find politics vaguely distasteful.

Big Tents and Third Party Spoilers

Why should this be so? Why not have three or four parties, to segregate those dimensions or hit all the possible permutations? It is a truism that the US electoral system requires two big-tent parties, but that’s not quite right. What’s true is that the system has very strong incentives in place to encourage party elites to find a way to force very different interests to share policy goals and ideology: a third party will tend to disempower its voters by splitting its big tent, as famously happened at the outset of the Civil War and during the salad days of the Progressive Movement. Yet for some policy issues, such a split may not be avoidable!1

What’s more, the US political system has pretty good methods for allocating voters to its two main political parties in roughly equal numbers. Like your favorite sports league’s revenue sharing or draft-picking arrangements, the party system tends to reward short-term victories with unsatisfiable constituencies that undermine the party’s lead, which produces future upsets and reversals. Incumbency advantages turn into incumbency fatigue; midterm elections draw different demographics than presidential elections; victorious parties engage in divisive self-destruction over exactly which part of the big tent will dominate.

A Party for the Anti-Political Majority

Yet there are hints (if not evidence) that this may be changing, in a way that is partly inspired by the rise of Donald Trump and partly contributed to that rise. Voter turnout has been stuck below 60% of the voting age population since 1968, and is even lower in primaries, so a candidate who can get out the vote from a unique part of the demographics of his party or the nation has a strong chance of surprising us, as Trump did.

Jason Stanley called attention to the particular role of Trump’s demagoguery in reorienting the primaries last October:

But there is a way a politician could appear to be honest and nonhypocritical without having to vie against other candidates pursing the same strategy: by standing for division and conflict without apology. Such a candidate might openly side with Christians over Muslims or atheists, or native-born Americans over immigrants, or whites over blacks, or the rich over the poor. In short, one could signal honesty by openly and explicitly rejecting what are presumed to be sacrosanct political values.

Trump, it seems, carries the promise of engaging the previously disengaged, those who have no stomach for politicians’ attempts to preserve a big tent by mobilizing disparate interests together. It’s difficult to pretend that business interests and “values” voters have much in common, and that you are adequately representing both without preference. Voters are smart enough to see through those efforts, if not willing to recognize the structural reasons why American politicians continue to try to straddle those fences.

There are always disaffected voters who give up on both parties. Usually, neither party can figure out how to reliably get out the vote from these truly independent voters while keeping the support of its base. These voters are largely mythical in third-party bids and usually for in-party insurgencies as well, but the idea that disaffected voters could show up at the polls in large numbers and destroy all the pollsters “likely voter” modeling is tenacious. We know it’s possible, and that when it happens we see surprisingly anti-establishment results. As the political participation of disaffected, unrepresented voters drops, this reserve army of the unallied gets bigger. It’s especially potent in party primaries, which are very low turnout events.

My suspicion is that if a group of disaffected voters could be reliably re-engaged, the parties would likely find wedge issues to divvy them up over a relatively short set of elections. But they may well divvy them up differently than the parties had previously done. This would be the seed of a realignment.

Party Platforms are Contingent; So Is the Meaning of “Liberal” and “Conservative”

How far could this go? We frequently joke that the Republican Party switched places with the Democratic Party between the Civil War and the passage of the Civil Rights Act. But so many of the relevant policy questions are different that this has always been a bit inexact.

Far too many of my friends and colleagues believe that there’s a natural connection between cosmopolitan attitudes towards other races and cultures and egalitarian economic preferences. There really isn’t; in fact, there’s just as often a tension! Terms like Republican and Democrat and conservative and liberal are free-floating signifiers that don’t really track particular policy preferences or ideologies over time.

Just one example: there’s a whole host of people who think that government should not offer certain sorts of assistance that they call “welfare.” But what they mean by “welfare” varies a lot, includes and excludes a lot of different services and cash transfers and tax treatments. Worse still, many people who oppose “welfare” think that the same program is justified or unjustified based on who will get it, and don’t think that their brother-in-law’s disability check is welfare while a stranger’s disability check is, nor see their home owner’s tax credit as similar to a poor person’s food stamp support.

The same thing goes for the Affordable Care Act: many of the Republicans who oppose it (and have voted to repeal it repeatedly) want to abolish something they call “ObamaCare” and replace it with an almost exactly identical program (i.e. RomneyCare). So at least for the vast plurality of Republicans elected to federal office, it’s clear that what is at stake is not a real difference in thinking or policy preference, but rather a partisan fight over votes which everyone is pretending is actually a deep ideological difference.

What we know is that we and our fellow citizens tend to seize on a few policy issues that matter to us, select a party on that basis, and then adopt a lot of the other policies of that party as equally important. This gets odd when the parties change policy positions (for electoral reasons) and partisans change their own views as if these things were demanded by rationality (often backdating their new policy beliefs as if they hadn’t changed at all) rather than merely a response to politicians competing for votes.

So while historically we can observe both massive and bipartisan shifts around specific policies (remember when the Democratic Party opposed same sex marriage?) and realignments (remember when the Republican Party was the party of Lincoln and Black people?) it is quite obvious that many people don’t have an accurate phenomenological or narrative account of what is happening in the moment. We think in terms of urban and rural, in terms of our neighbors and our transportation preferences and our religious traditions.

It’s a unique kind of privilege to be able to maintain strong coherence between our different beliefs, and indeed it’s actually presumptuous to pretend that only a couple of coherent ways to hold these beliefs together.

Realignment May Be Unpredictable But Formulaic

We may not know when a realignment has arrived, but we do have some information on what it is likely to do when it does. Jennifer N. Victor from Vox’s Mischiefs of Faction blog detailed a famous theory of party realignment to try to explain why Trump and Sanders were doing so well against more traditional candidates, bringing attention back to the 2003 paper by Gary Miller and Norm Schofield detailing shifts in party policy platforms. Here’s Victor’s gloss:

the party that loses an election has a strong incentive to try to peel away voters from the winning party. This is how a party grows its coalition to win in the next round. The party does this by taking policy positions that appeal to voters who may only weakly identify with the winning party. Think of these voters as the ones at the edges of the cleavage lines.

So losing leads to strategizing and better political competition for ignored voters at the winning party’s “rump,” which leads to winning. If one party’s “big tent” gets too large, the other party finds an opportunity to compete for some of the voters whose policy goals have been most ignored.

Yet neither party has had to actually compete for unallied voters for the last twenty-five years or so. They’ve been depending instead on GTFO efforts and incumbency-fatigue to restore them to power. This is why the Republican party has spent the last three decades working on “fusionism” to combine conservative Christianity with business-oriented small-government rhetoric. The Democratic party has spent those same three decades trying to keep mostly-white-male working class voters together with urban minorities and rich cultural cosmopolitans.

Thus the “clockwork” of party realignment has been momentarily frozen, and we can’t even say if we are now in one or if both parties will simply revamp their rule to prevent insurgencies in the future. Yet if we are in a realignment period, the model predicts which “side” will be the rump:

Miller and Schofield show that this cleavage line rather naturally rotates in a clockwise fashion across time. Bill Clinton broke up the previous partisan alignment when he proposed more conservative economic policies, like those of the cosmopolitans, moving the Democratic coalition from the liberal position more to the cosmopolitan one.

The “clockwise” directionality comes from a supposedly natural movement of the cleavage line in this chart over time:

Party Realignment

So we should expect that the Republicans will lose cosmopolitan pro-business types represented by the likes of Michael Bloomberg, while the Democrats will lose the remaining supply of white men without college degrees who are economically liberal but “socially” conservative (here meaning: they are anti-diversity.) And indeed, this is basically what seems poised to happen.

The fact that state support for the poor and working class is orthogonal to the state’s treatment of African-Americans and other racial minorities is certainly vexing: it seems to extend from the American experience with slavery and persistent white supremacy, yet we see similar trends in Europe towards immigrants and Jews. But the key here is that attitudes towards racial difference just are unmoored from attitudes towards the distribution of economic assets among whites, and so we need these two dimensions (and yet oddly: not much more than that) to explain most of American political behavior, especially in the legislature.

There Will Always Be Elites; Who They Are Matters

Freddie deBoer is also thinking about realignmnent:

Today, there is a least an ostensible connection between the liberalism of diversity and the leftism of equality. Tomorrow, even that thin thread might be cut forever, and so much the worse for us.

Basically, deBoer imagines a future where the boot forever stomping on the face of humanity is gender equitable and racially diverse. But what if having a more Black or female Senators–or a Black President–isn’t necessarily better for that President’s or Senator’s female or Black constituents?
The impetus of deBoer’s piece seems to be this zinger of a tweet from Dan O’Sullivan):

“Our political future: a snakepit of insane fascists on one side, & on the other, a Wall Street party that’s culturally liberal & nothing more.”

I think this is really the crux of deBoer’s piece, as an elaboration of that perfect tweet, and I think there’s something odd about it. Is it possible for realignment to turn today’s educated leftists into completely alienated voters with no party to support at all? As I’ve tried to show above, the answer is probably not: one or another party will almost always have something to offer the educated people who currently self-identify as Leftists. If Democrats really start leaving the Left, the Left will learn how to get along with nationalists and chauvinists, like they did when nationalists and chauvinist were running labor unions. And perhaps this is where the oddness in deBoer’s view emerges:

“The entire purpose of the elite-building mechanisms of our country is to keep that elite small. There’s only room for 1% of people within the 1%, after all.”

Every society has a top 1%, whether it’s the billionaire class or the Politburo. It has been the case that the people with the PhDs were among that class of elites, but not recently. It’s an open question who the elites should be. It’s historically been the case that US elites were white men, and while that’s changing, it’s not actually changed very much yet. Harvard, Yale, Princeton, and Stanford will achieve truly representative diversity, but without expanding their first year classes. Perhaps tomorrow’s elites will be truly representative of the rest of the country.

It’s also an open question for political debate how much better off the elites should be than the people below them: I tend to favor egalitarian distributions of well-being, myself, but I recognize that there are lots of people who disagree, and for something approaching good institutional reasons. Specifically, the existence of a top 1% doesn’t actually tell us anything about the material conditions of the bottom 99%.

There’s good reason to believe that social justice work to ensure diversity has largely helped contribute to what Deirdre McCloskey calls the Great Enrichment. The American 99% are much better off than most precisely because we have been forced to diversify both our elites and our middle-class. We can expect to reap further wealth gains if we continue to diversify our elites, so we obviously should.

US Political Parties Could Be Better

A few things follow from all this. First, it seems highly desirable to have two healthy, reasonable parties at all times. Constitutional hardball is dangerous in a way that threatens to shatter the conditions of possibility for our overt disagreements. Yet the best evidence suggests that the Republican Party has been growing increasingly extreme for almost twenty years. This is easy for opponents to celebrate, as it tends to render Republican politicians and Republican voters less effective at making progress on their policy goals, and might well help Democrats win in November. But in the medium and long-term, it’s deeply unstable (it has already been destabilizing.) The survival of a political order has to be compatible with either party losing elections.

Second, the ideal should not be to have a permanent victory that utterly destroys the other party; we should aspire to the Aaron Sorkin West Wing fantasy of principled idealists representing the divergent but reasonable views of the Good Life and the Good Society. Reasonable pluralism is better than a fragile modus vivendi; deliberative disagreement that preserves the possibility of compromise and collaboration is better than dirty tricks and pitched winner-take-all political battles.2

The problem, of course, is that without a healthy and rational opposition party, we’re going to have a hard time living up to that fantasy of democratic deliberation.

Economic Equality and Nationalism

Insofar as nationalism in the US is understood as a kind of jingoistic, nationalist chauvinism–a defense of the traditional perquisites of white men–we will all be better off and richer insofar as the political parties can find a way to preserve the disaffection and disengagement of the anti-diversity voters.

Politicians and pundits shouldn’t court them, even with coded language or some account of what their “true” interests and resentments are. But they will. (Perhaps some of the reactionaries are coalescing around the financial crisis, but then why not vote for Bernie Sanders instead of Donald Trump, who claims to be a billionaire?) Trumps’ supporters aren’t the folks who lost their jobs to the financial crisis; they’re the long-term unemployed who fell out of the economy long before: the unnecessariat:

This is the world highlighted in those maps, brought to the fore by drug deaths and bullets to the brain- a world in which a significant part of the population has been rendered unnecessary, superfluous, a bit of a pain but not likely to last long. Utopians on the coasts occasionally feel obliged to dream up some scheme whereby the unnecessariat become useful again, but its crap and nobody ever holds them to it. If you even think about it for a minute, it becomes obvious: what if Sanders (or your political savior of choice) had won? Would that fix the Ohio river valley? Would it bring back Youngstown Sheet and Tube, or something comparable that could pay off a mortgage? Would it end the drug game in Appalachia, New England, and the Great Plains? Would it call back the economic viability of small farms in Illinois, of ranching in Oklahoma and Kansas? Would it make a hardware store viable again in Iowa, or a bookstore in Nevada? Who even bothers to pretend anymore?

By linking education to jobs in the way that we have, we’ve guaranteed that a large number of our citizens will not be qualified for the majority of dignified work. It’s hard not to fault their resentment, even as it seems misguided for failing to understand a hypothetical Kaldor-Hicksian analysis of their error.

The whole problem is that the loss of dignified work is concentrated among people who can’t readily understand such arguments. This allows us to pretend that the correlation between education and income is noble and meritocratic without challenge. But in fact, we’ve installed an invidious systematic bar to recognizing the real pain of poor people without college educations, especially men. We hold out the promise of both self-sufficiency and effective citizenship only to those who can jump through all the hoops between kindergarten and a Bachelor’s degree.

In the best of all possible worlds, maybe both parties would ignore anti-diversity interests and find a way to make economic policies that reduce racist and sexist resentments. That’s right: I’m suggesting that maybe we should not hope for 100% political participation, at least insofar as that requires that white supremacists and chauvinists find viable politicians who will court them openly. In the second-best world, both parties would court the anti-diversity vote while remaining conveniently but systematically paralyzed in the realization of their anti-diversity policy preferences, and thus blunt their power.

Yet with Trump poised to run as a populist nationalist demagogue explicitly recognizing and courting this demographic, it may already be too late even if he loses. The neo-reactionary movement will become self-aware and have a single partisan home.

So in the third-best world we actually inhabit (or is it even worse than that?), the uneducated white vote (call them George Wallace Democrats) is lost to the Democratic Party. Miller and Schofield point out that this had already mostly happened by 1996, and called it “The Decline of Class and the Rise of Race.”

Perhaps deBoer shouldn’t worry so much: if the clockwork thesis continues to hold up, we should expect the next few elections to reverse the focus on race and develop conflicts along the very dimensions of class inequality that the Left has championed. Soon the Republican Party should begin courting disaffected economic liberals who are willing to overlook the party’s racism. In fact: they’re ahead of schedule:

Trump: GOP will become “worker’s party” under me


1. Of course, there were actually four serious parties competing in the 1860 election before the Civil War. If there had been three, though, the result would have been quite different. Both the Republican and Democratic parties’ inability to ally behind a single candidate because of regional differences over slavery and unity produced not two but four Presidential candidates. Lincoln’s victory was thus mostly a matter of chance, and indeed Stephen Douglas and John Bell together–both “centrists” convinced that avoiding the slavery issue could stave off secession–received more votes than Lincoln did, as many Southerners preferred union to secession. With their influence split, one of the more radical parties was victorious; but it could have gone differently.
2. I’m being inexact: what I’m describing could easily be understood as simply a much more desirable modus vivendi than our current equilibrium between neoliberalism and nationalism.