Congressional representation might be nationalizing, but some politics is still local

In  The Increasingly United States of America, Daniel J. Hopkins argues convincingly that due to the increased salience, coherence, and identification of voters and politicians with the two major parties, American politics has become deeply nationalized, and is fought only on partisan grounds. As a result, local issues and types of representation specific to certain geographies are being replaced by two unifying national identities: Democrats and Republicans.

As Hopkins observes, there’s no doubt that the trends are headed in a “nationalizing” direction. Scholars and commentators are quick to note the shrinking number of partisan crossover congressional districts – those won by presidential nominees and congressional candidates of different parties. As a result, election forecasting and analysis at sites like FiveThirtyEight and the Cook Political Report are relying more heavily  on prior presidential results as an indicator of incumbent vulnerability.

They aren’t wrong to do so. But if we dig deeper, we can see that there’s still significant variation in how well incumbent legislators perform in their districts compared to their party baseline – variation that matters a great deal, particularly for vulnerable incumbents for whom just a few points can make a world of difference.

Below is a distribution of incumbents facing major-party challengers from 2002-2018. We see that while many do indeed perform similarly to their party’s presidential nominee, many others continue to perform significantly better or worse.

What gives? If politics is so nationalized, wouldn’t we expect voters to judge the Democratic member of Congress on the same criteria as the Democratic presidential nominee? There are a number of potential answers to this question. One that I’ve focused on is the extent to which an incumbent has local, pre-existing roots in their district that connect them to their community, generate a sense of trust and authenticity with their constituents, and ultimately help them win more votes on Election Day. In my dissertation and a number of related projects, I demonstrate that incumbents who have a shared personal background with their constituents based on locality in the district enjoy significantly more support among their constituents than those who have no such roots.

Over the past several years I’ve been collecting data on specific characteristics of members of Congress that indicate a prior local connection to their district. These include characteristics like whether a legislator was born or went to school in their district, whether they held local elected office, or whether they own a business in the district.

The results of these analyses (pictured below) confirmed my suspicions. Even after controlling for other crucial differentiators between congressional and presidential contests (challenger quality, candidate race, tenure length, and many others), the depth of a legislator’s pre-existing local roots in the district can help them outrun their party’s presidential nominee by around 12 points more on average than they would otherwise.

Predicted Effects of District Roots on Electoral Differentiation from Partisanship

Compare this, for example, to the effect of serving for 10 years or more. One would think that serving so many terms in Congress would give members of Congress ample opportunity to connect with their constituents in local, personal ways that a presidential nominee never could. My analysis indicates that this service does indeed help, but it’s dwarfed by the more “authentic,” homegrown connections that deep local roots in the district provide.

Tip O’Neill used to say that “all politics is local.” Politics is no doubt more national than it was in Tip’s heyday, but my findings indicate that voters in many districts still care deeply about local representation, and value authentic localism in their candidates for Congress. And in an era when voters, scholars, and politicians  are searching desperately for connections that can transcend partisanship, local ones – at least in the House – may be part of the answer.