The Iron Law of Successive Elections


A magnifying glass over the word, "midterm elections" in a textbook.

Democrats, at this point, are near-certain to lose seats in the House of Representatives. Just like what Republicans experienced in 2018, Democrats are likely to lose their House majority. It seems that claims to the contrary, are merely partisan cheerleading.             

In addition, those paying attention to the news during election season should beware of explanations that focus on so-called pivotal moments – such as the Dobbs ruling, Biden’s legislative successes, or any unique campaign tactics. These factors are not critical. 

Remember: regardless of who occupies the White House, losses occur.   

For example, Lyndon Johnson lost 47 Democrat seats in 1966. Ronald Reagan lost 26 Republican seats in 1982. In 2010, Barack Obama experienced a 63-seat shellacking. And most recently, in 2018, Donald Trump dropped 40-seats.

The most pertinent question is this: how many seats will Democrats lose?

First midterm election by the numbers

Consider the historical pattern displayed below - it is not a law of physics, but it’s very close to one. 

Since WWII, the president’s party has lost House seats in 11 of 12 midterms. Most of these recorded losses are in the double digits. The average is twenty-five, and across the last two cycles… that average has doubled.       

For the single midterm that defied the pattern, gains were small. In 2002, the midterm after the 9/11 terrorist attacks, Republicans celebrated a net gain of eight seats. The devastating 9/11 attacks boosted Bush’s presidential approval ratings to the highest level in recorded history (90%).    

The obvious imbalance between seats won and seats lost signals that there is something structural at work.     

graph depicting the first midterm election results - president's party seats won/lost.

Turnout

Presidential elections are high-stimulus events. The supercharged environment inspires peripheral voters - those who are modestly interested in politics. Accordingly, the surge in attention and information readily available increases voter turnout.

By contrast, midterm elections are low-stimulus events. Political information and attention decline and peripheral voters exit. Others - namely Independents and weak party identifiers - are typically disappointed in the new administration. Their enthusiasm drops, and many sit out of the elections.

In short, a significant group of peripheral and independent voters who were lured to the polls by a high-stimulus presidential contest tend to sit out and withdraw two years later. This phenomenon explains the disparities in turnout between presidential and congressional elections. There is a notable surge in voting, and then there is a sharp decline.

Graph depicting the Vote Eligible Population Turnout Rates from 1948 through 2018 for presidential elections vs. first midterms.

The surge works for the president’s party, but 2 years later… the decline works against it

The flood of information and attention during presidential elections is not neutral. The advantaged party receives support from peripheral voters and draws considerable support from Independents and weak party identifiers. These voters are critical to winning the presidency, but two years later, they are largely absent. Among those Independents and weak identifiers that do participate, many swing back in favor of the opposition party.  

For example, in 2016, Donald Trump won the Independent vote over Hillary Clinton by 48% to 42%. Only two years later, Independents strongly shifted toward Democrats, voting 54% to 42%.            

Core voters are important, as well. These voters are strong partisans and possess an enduring interest in politics. Therefore, core voters typically vote regardless of the circumstances. Yet, after two years of frustration, the out-party’s core is particularly motivated. They are primed for the chance to oust the president’s party.       

Voter turnout in 2018 offers an excellent example of this. Trump was the focal point for Democrats and the party effectively mobilized against congressional Republicans. Voter turnout in fact reached historic levels. Democrats represented 37% of the electorate whereas Republicans represented 33%.       

Enthusiasm, energy, and organization are always on the side of the out-party. A Republican in the White House strengthens the Democrat base, and a Democrat president motivates the Republican base.

The structure of successive elections

The surge in presidential election turnout always declines only two years later. It does not matter who occupies the White House, nor does it matter if the economy is doing good or bad. It does not matter if the president is reasonably popular or unpopular, or what are the salient issues. None of that is important to the basic outcome. The incumbent party will suffer a net loss in the House. While the surge during presidential elections lifts one party, the inevitable decline brings about midterm losses. The electorate for successive elections is not the same. Midterm electorates favor the out-party.       

Does this mean the president’s party can never win seats during the first midterm? No; 2002 shows how powerful an exogenous shock like war or an attack can be to the political system. The parties briefly united after 9/11 and rallied around the flag and exhibited historic support for the president.

But since that time, the system snapped back, and the structural pattern returned.

How many seats will Democrats lose?

To hold the majority in the House of Representatives, the party must require 218 seats. Democrats now control 220 seats. If they lose more than two seats net, Republicans will assume the majority. Given President Joe Biden’s net favorability ratings hover in the range that Donald Trump experienced in 2018, and inflation continues to pull down consumers’ real wages… significant seat losses are inevitable. Democrats would have to outperform the Kennedy administration’s 4 seat loss to keep their majority.    

Formal models and polling both indicate Republicans will win the House. Eyeballing the estimates, it is reasonable to expect Democrats to lose 20 to 30 seats. 

However, recent structural changes could limit the losses. For example, strong partisan polarization, gerrymandering, and the 15 Republican seat pick-up in 2020 limit the number of winnable seats. In other words, the size of the 2022 battleground is much smaller than normal.     

When significant losses occur, critics will point the finger at Biden’s legislative missteps, his lack of action on certain issues, high inflation, foreign policy lapses, and an “identify politics” agenda. Post-election assessments will likely include all sorts of stories about voters shifting right, embracing the Republican agenda, and disavowing Democrat policies.

These stories will resonate with many. But Democrats would have experienced losses regardless of Biden’s actions.  Losses are baked into the system for a newly elected president. Sure, the particulars about how many seats changed party, where those seats are located, and for what reasons did they change, are in part attributable to the social-political context.

But, once more, the general outcome appears predetermined... Democrats will lose seats.      



Previous
Previous

Looking at first week of November 2022

Next
Next

October’s Industry Indicators: uneven recovery, driver demographics, & alternative fuel