Attention across the world to the US elections has focused only on the federal level. But the battle for the statehouses mattered too.
International observers of the American presidential election who turned to the pundits in the first days after November 3rd could be forgiven for feeling that the commentators had been speaking about 2016 or that Joe Biden had in fact lost the presidential race. At best—even after his projected victory was announced four days later—the celebrations were muted on the Democratic side.
That is because, despite record voter turnout and his gaining more votes than any candidate in history, Biden carried no coat-tails: the states he won didn’t see the same success for Democrats in down-ballot legislative races. This year’s predicted ‘bloodbath’ for Republicans failed to materialise.
The outcomes in the state races in particular will have long-term consequences for key policy areas—including but not limited to reproductive freedom, gun control and implementation of minimum-wage laws—as well as for the Democrats’ electoral prospects.
It’s still early in the post-election analysis. But already some factors in the Democrats’ underwhelming down-ballot performance, relative to pre-election polls, have emerged: enthusiasm and knowledge gaps affecting down-ballot candidates, structural and investment advantages for Republicans in down-ballot races, changes in Democratic voter mobilisation tactics in response to Covid-19 and the move to vote-by-mail and incorrect polls due to the increasing difficulty of polling supporters of Donald Trump and Republicans.
‘Rolling off’
Generally speaking, by the time election day in the US arrives, presidential candidates have near universal name recognition from those who plan to vote. Whether or not voters are enthusiastic about those candidates, or have knowledge and enthusiasm about down-ballot candidates, is however another matter. If voters vote for their presidential candidate of choice, but don’t vote for a down-ballot candidate because they don’t know anything about them or aren’t enthusiastic about them, we describe this as ‘rolling off’.
An excellent analysis by the Sister District Project mapped out how in three key states (Pennsylvania, Florida and Texas), ballot roll-off was significantly higher than expected. This was particularly costly in places like Florida, where in some instances races were decided by as few as 34 votes. In North Carolina and Michigan, two other hotly-contested states, ballot roll-off actually increased compared with 2016.
Republicans control 30 state legislatures, the main law-making bodies in most states. This party dominance of so many state legislatures has remained static for the first time in decades. According to the National Conference of State Legislatures, ‘On average, 12 chambers change party in each general election cycle. [In 2020] the parties came to a draw.’
Legislative agenda-setting
This is bad news for progressive policy-making as well as Democrats’ electoral viability in states controlled by Republicans. With control of state legislative chambers, a party can wield outsized power for very little cost and have a disproportionate impact—not just in the state but in national agenda-setting.
Since the 1960s, nearly every major domestic political fight or crisis in the United States was initiated at state level. Encroachment on reproductive rights, financial deregulation, loosening restrictions on firearms and roll-backs of environmental protections have all tended to start in statehouses and work their way to the federal level.
The impact of state legislative control has been vastly exacerbated by gerrymandering—the drawing of districts to pack one group into one district or split them up to dilute their power. Down-ballot Republicans benefited from gerrymandering that happened in 2010—one reason for Biden’s lack of coat-tails—and they will now benefit for another ten years from the gerrymandering that will take place as a result of this year’s elections.
Democrats have made some efforts to counterbalance this strategy by investing in down-ballot races. Since 2017, the National Democratic Redistricting Committee has poured millions of dollars into gubernatorial and legislative races, in states including Georgia, Pennsylvania, North Carolina, Florida, Texas and at least seven other battlegrounds. And some good-government groups, such as the League of Women Voters, have advocated non-partisan or more bipartisan reapportionment boards. Since 2010, Colorado, Michigan, Missouri, New York, Ohio and others have made changes in their districting boards, from partisan to non-partisan or bipartisan—most within the past five years.
Face-to-face conversations
For several decades, Democrats have relied on face-to-face conversations with prospective voters as part of their electoral strategies. Historically, Republicans have relied more on mass media and direct mail to mobilise existing supporters, though they have been investing more in direct contact over the last 10-15 years.
Democratic candidates and aligned organisations decided early on in the pandemic not to engage in activities which would otherwise involve talking to voters in high-footfall traffic areas, such as shopping malls, or visiting voters at their homes. Out of concern for the safety of their staff and broader public health, they opted instead to focus largely on virtual registration and turnout pushes, and education around vote-by-mail. Republicans, by contrast, increased both their voter registration and canvassing efforts.
Some studies have shown canvassing can increase voter turnout by as much as 6 per cent. It’s likely the effects are more limited but in races with razor-thin margins—remember that Florida Senate race decided by 34 votes—the decision not to register voters and canvass in traditional ways probably had marginal but negative down-ballot impacts for Democrats.
Poll shortcomings
The muted celebrations were also in part a reaction to pre-election polls that showed Democrats with an 80 per cent chance of taking the Senate (they still have a chance based on the outcome of the Georgia Senate runoffs) and favourable chances of winning one or both legislative chambers in states including Texas, Iowa, Arizona, Missouri, Minnesota and Pennsylvania—as it happened, they didn’t win any.
Given the importance of polls in informing campaign strategy, public expectations and punditry, we should consider their shortcomings. According to analysis by the Washington Post, 2020 represented the least accurate polling since the 1996 presidential election. People are less likely to answer unsolicited calls than they were a generation ago and those who are willing to answer may not be as reflective of the electorate.
Specifically, those who identify as supporters of Trump and the Republicans may be less likely to respond to polls due to their distrust of media outlets and civil-society institutions more generally. Even with some weighting to adjust for not being able to talk to some Trump supporters, not having a baseline sample representative of them is likely skewing polls in some key states.
Buoying political fortunes
We won’t have the data to analyse fully the various drivers of Democrats’ down-ballot underperformance for at least a few months. But one thing is quite clear: a strong showing at the top of the ticket is not enough to buoy political fortunes for all.
Democrats need to invest further in efforts outside the presidential races, particularly in those for the statehouses. And they need to consider their strategy and tactics on voter registration, contact, persuasion and polling—if they want to have a chance of countering Republicans’ down-ballot advantages.
This is part of a series on US Election 2020 supported by the Friedrich Ebert Stiftung.
Matt Mawhinney is the co-founder of Generation Data, a mission-driven non-profit focused on training the next generation of leaders in progressive data management and analytics.