There are (at least) two “lefts” and two “rights”
Most of us have grown up with a party for the rich (Republicans) and a party for the poor (Democrats). Most of us also know that the Democratic Party tends to be more liberal on social issues. And most of us would agree that these things are still true. However, observations are accumulating that they’re less true than they once were.
Consider this week’s Democratic National Convention, where the party promulgated its latest platform. As Christopher Witko notes, it may be hard, even with united government after November’s election, to make good on the commitment “to reverse the over-financialization of the American economy and curb Wall Street speculation.” Further, E.J. Fagan observes that the new platform contains fewer pages on traditionally working-class issues (except health care) than it does on other matters (which, many would agree, sorely need attention).
Long before this week’s convention, though, some sensed that big change was afoot in the two-party system. Brad Spahn, for example, has data suggesting that party switching in 2016 was higher than at any point since the start of the New Deal (caveat: this suggestion relies partly on historic California voter registries). Most of us also lived through the “populist insurgencies” of Donald Trump and Bernie Sanders, both of whom have criticized financial elites, which suggests that rich-poor issues may straddle the major-party divide more than in the recent past.
All of these phenomena may add up to something called realignment — change in the sorts of people who tend to support either party, as well as how the parties present themselves to voters. I used panel data from two national surveys to see if realignment really is happening — and what it might mean for policy.
One way to think about realignment is this: rather than cast politics in terms of “left” and “right,” imagine two or more dimensions. Commonly, these are “economic” and “social” (although reality may be more complicated). Voters have preferences along both dimensions, and they form a sort of blob in coordinate space. An election cuts the blob into two coalitions. We can plot the blob and estimate the cleavage line, which is what you see just below.
The overall line of cleavage did tilt slightly from 2012 to 2016. Those who switched party choice between these elections also responded to a different line of cleavage — one that is more “social” in content. Looking at the switchers specifically, several new Republicans appear toward the top-left, and a handful of new Democrats appear toward bottom-right. But the overall blob also seems to change as well.
This set of plots is illustrative. The rest of my research note interrogates these changes at two levels: within four voting blocs (Obama-to-Clinton, Obama-to-Trump, Romney-to-Clinton, Romney-to-Trump) and across 19 policy questions.
Some notable results are as follows. Vote switchers went “left” and “right,” respectively, on several “social” issues, in accordance with the ways that they voted. When those issues implicate redistribution, however, their preferences were mostly stable (e.g., path to citizenship, affirmative action). This suggests some interdependence between “economic” and “social” dimensions — or that we should treat them as short-hand terms.
Critically, Democratic voters are now more divided on two strictly economic issues: taxation of wealth and regulation of business. Republican voters are now more divided on the wealth tax. The data suggest that this largely was due to vote switchers not changing their minds.
It is worth emphasizing that these changes were small.
Are these results a one-off, due to unique circumstances in 2016? Or are they part of a longer-term change in the substance of party competition? Time will have to tell.