The history of elections during recessions is not good for Democrats

Get the best of Grid in your inbox

Economic downturns don’t go well for incumbents: Why the White House and Democrats fear a recession

Why are the White House, cable news, economists and financial journalists obsessing over the definition of a recession? The most obvious reason is straightforward: Economic downturns destroy wealth and throw people out of work. So figuring out if negative GDP numbers are the same as a sustained slump in economic activity is very important for trying to anticipate and maybe even avoid that outcome.

“[President Joe] Biden can pretend there’s no recession. But Parcells said it best: you are what your record says you are,” Republican Arkansas Sen. Tom Cotton tweeted, including a reference to the former New York Giants head coach.

“The Biden Administration has begun their latest project — a frantic effort to re-define the word ‘recession.’ … it’s almost beyond satire,” Senate Minority Leader Mitch McConnell (R-Ky.) said in a statement last week.

After two consecutive quarters of negative economic growth, some are ready to say that the recession is already here, even though it’s not entirely clear that we are. After all, the unemployment rate is still near historic lows, and the labor market is still creating over 300,000 net new jobs a month. Biden has been trying to say the economy is better than a lot of people think. “People are really, really down,” he told the Associated Press in June, saying that a post-covid psychological slump may be nudging people away from the positive parts of the economy.


Vanderbilt University political scientist Larry Bartels argued that an economic downturn does not necessarily spell political doom for Biden’s presidency, even if it can exacerbate what might already be a nasty midterm. But while the midterm elections are in a few months, Biden’s reelection — where, typically, incumbents have an advantage — is not for over two years.

“One common finding in the research on economic voting is that people overwhelmingly respond to recent economic conditions rather than a president’s or party’s long-term record. The implication is that what’s happening (or more precisely, what people are feeling) now is likely to be consequential for the midterm election but irrelevant by the time of the next presidential election,” Bartels wrote in an email to Grid.

Whether or not a recession is officially declared by the National Bureau of Economic Research, economic downturns — or even just economic discontent — are perilous for whichever party occupies the White House. Whether it’s John McCain’s massive loss to Barack Obama in 2008 or George H.W. Bush’s approval rating plummeting from almost 90 percent to 34 percent before Election Day in 1992, a bad economy — or even the perception of one — can doom a White House, but timing matters.

The political science says Democrats are at a disadvantage, economy aside

Economic conditions are especially perilous for Democrats because they’re compounded with the apparent historical regularity of the party in power losing midterm elections.

While the mechanism by which these two phenomena generate poor electoral performance for the party in power is by no means universally agreed upon by campaign professionals and scholars of American politics, it shows up persistently in the modern history of elections.


“It does seem to be a very frequent regularity that the party in the White House will lose seats,” said Joseph Bafumi, a political scientist at Dartmouth College. “It’s overcome rarely.”

Whether it’s polling analysts looking at polling data like FiveThirtyEight (which predicts Republicans will win 232 seats in the House and win the overall congressional ballot by almost 5 points) or DecisionDesk HQ (235 House seats), or models that solely look at economic indicators, some combination of the regularity of a midterm backlash and low approval for the president is weighing on Democrats.

That the slowdown in economic growth is compounded by high inflation only increases the danger for the White House. Voters are fans of neither, and the Federal Reserve’s apparent willingness to raise rates come what may until the inflation situation moderates could mean it barrels through a slowdown.

How economics influences the picture

Past downturns and periods of slow, uneven recovery are linked with big electoral losses.

In 1982, in the teeth of one of the sharpest downturns since the Great Depression, Democrats made substantial gains in the House and Senate. The 1974 and 1980 elections happened either during or just following a downturn (there was also Watergate) and led to transfers of powers in either Congress or the White House. A recession that technically ended in 1991 but still lingered in the public imagination helped deliver the White House to Bill Clinton in 1992, while the slow recovery from the 2007-2009 recession helped ensure massive Republican victories in the House of Representatives in 2010.


The Yale University economist Ray Fair’s model for midterm elections (he has a similar one for presidential elections as well as congressional elections in presidential years) uses economic data to make a forecast of the vote share for both parties.

In an update released at the end of July, Fair pointed to higher inflation and the likelihood that there would be fewer “strong growth quarters” (quarters where per capita GDP rises 3.2 percent or more) than previously forecast for why his estimate of the Democratic vote share had fallen from his previous run of the forecast. “There is strong evidence that the economy affects all three vote shares [presidential, in-term house elections, midterm house election],” Fair wrote in a paper explaining his equations for forecasting elections.

Using a less mechanistic approach to the midterm result, Bafumi sees an economic environment that is not helpful to Biden and Democrats in Congress: “[Voters] don’t like inflation; they like slowdowns even less. To clamp down on inflation, there has to be some effort at slowing down the economy. There’s been some indicators at success of slowing down and not so far at clamping down on inflation. There could be a perfect storm for the president come Election Day.”

Whether voters “should” attribute the state of the economy to the current congressional majorities or the president is less clear — gas prices, which are often linked to large swings in the global price of oil or short-run disruptions like hurricanes have meaningful effects on presidential approval. When it comes to macroeconomics, voters tend to be very present minded, Bartels argued, noting in a paper on the politics of the Great Recession that elections “have been significantly shaped by voters’ consistent inclination to reward or punish incumbent governments based on economic growth rates in the months leading up to an election.”

Polarization is no match for overall feelings about the economy

Forecasting how a recession or prolonged inflation would affect presidential politics is tricky because it involves doing something very hard — anticipating the economic environment in a few months to a few years — and then combining that with its effects on politics, which are not always stable or easy to predict.

Biden’s low approval ratings, for example, seem to fly in the face of the idea that polarization meant that voters were less responsive to the condition of the economy (or, as Bartels pointed out, their perception of the condition of the economy) than they used to be. That Obama and Donald Trump — two very different presidents — nonetheless had relatively steady approval was taken by some to mean that partisanship and polarization had gotten so strong that events had little impact on public opinion.

“That seems plausible, but the extent to which President Biden’s popularity has dropped as economic conditions have worsened suggests that, for reasons unknown, ‘We’re back in a world in which a weak economy hurts presidential approval,’ Bartels said, quoting his Vanderbilt colleague John Sides.

That the Democrats and Biden seemed to have marginally improved their political standing in recent weeks as gas prices have fallen off their wallet-stretching high — even as other, less personally salient economic indicators have turned negative — may be evidence that the relationship between some economic indicators and approval are getting more “normal.”

But normal for them means a loss in November, unless the economy turns around.

Thanks to Lillian Barkley for copy editing this article.

  • Matthew Zeitlin
    Matthew Zeitlin

    Domestic Economics Reporter

    Matthew Zeitlin is an economics reporter at Grid focused on the domestic impact of major stories such as coronavirus, the supply chain and economic volatility.