News & Election Videos

SEND TO A FRIEND | PRINT | | Share Share

How unemployment affects midterm elections

By Sean Trende

This post by Seth Masket has been making the rounds on the internet, and deserves some further comment. Masket considers the frequently-heard argument that high unemployment will hurt the Democrats in 2010, and concludes that “there's not much evidence unemployment has any effect [on midterm elections] at all. “

Masket goes on to note that change in unemployment (in other words, how much unemployment has increased or decreased from the previous year) doesn’t correlate with seat loss either. Masket does note a statistically significant relationship between real disposable income growth and midterm loss, but even this is a weak one:  

Receive news alerts



With charts like this, maybe Democrats don’t have to worry about this Great Recession thingie after all!!! (In fairness, this is less Masket’s takeaway than the takeaway of some of those who linked to him.)

A couple of things to consider. First, there are a very small number of datapoints to observe here. We have only had 15 midterm elections since 1950. This is barely data; it’s more of a good collection of anecdotes. And this dataset doesn’t have much variance; in ten of these fifteen midterms, the unemployment rate is between 3.8 and 5.8 percent; in only one election in the datset (1982) is it above 7 percent, which would seem to be the relevant data for our purposes.   This isn’t a limitation of Masket in particular, but rather limits all analysis of midterm elections, as our economic data and presidential popularity data aren't that great once you get much before 1950.

This is especially problematic because we are trying to explain a very complex phenomenon. Obviously, there’s a lot of things that affect an election. Is your President pursuing an unpopular war and controversial policies at home (1966, 2006)? Then it probably doesn’t matter that the economy is blazing ahead. Is the President kicking some al Qaeda arse a year after they attacked us, and getting ready to take out a longtime nemesis (2002)? The public is going to be more forgiving of the sluggish growth in real disposable income and rising unemployment.The end result of this is that every election becomes something of an explainable, unique event – in other words, they’re almost all outliers.

Moreover, there’s a very real possibility that there’s a non-linear relationship between unemployment and midterm election results, which OLS regression just doesn’t handle well - especially when your data hardly has any variance to begin with. In other words, below a certain point, voters probably don’t delineate that much among different unemployment rates. But once you cross a certain threshold, people start to worry increasingly about each point increase in unemployment. I’m guessing if unemployment jumps from 4% to 8%, people get concerned. But when it goes from 8% to 12%, people start to freak the heck out.

Indeed, if you look at the eight years where unemployment was above 5.5%, you capture most of the really bad midterm elections out there. The problem is that you have two elections where unemployment was low where the President’s party got shellacked for reasons clearly unrelated to the economy (see 1966 and 2006, discussed above), and four elections where unemployment was relatively high, but the President’s party did well. Each of these cases is explainable, and with more datapoints, they probably wouldn’t affect the regression that much. But with only 15 observations, they wreak havoc on the model.

Finally, we are severely limited in our analysis because we don’t have any observations where unemployment is above 10 percent (which seems at least possible in November 2010), and only one observation where unemployment is above seven percent. In 1982, unemployment was at 9.6 percent and rising. The Republicans managed to “only” lose 26 seats. But they only had 192 seats to begin with; this represented 14 percent of their caucus. If the Democrats lose 14 percent of their caucus in 2010, they will lose 35 seats.  But since there aren't any other high-unemployment midterms, its difficult to know whether 1982 was an outlier, or whether it is typical of high-unemployment midterms.

The truth is, I don’t think you’re going to find a particularly good variable to explain the influence (or lack thereof) of perceptions of the economy on peoples’ voting choices. But just because it’s hard to operationalize this relationship doesn’t mean that the relationship doesn’t exist. Common sense, and a more qualitative look at the data imply a correlation.

I looked at the worst midterm drubbings since we started to compile good economic data (around the Great Depression). I defined these worst drubbings as those elections where the President’s party lost more than 10 percent of its seats. I came up with 1974 (25% seat loss), 1958 (24%), 1946 (22%), 1938 (22%), 1994 (21%), 1930 (19%), 1942 (17%), 1966 (16%), 1982 (14%), and 2006 (13%). What are the bolded ones you ask? Look at the list of recessions in the United States history. The bolded elections are those that immediately followed a recession ("immeditately followed" is defined as those midterms held after a recession that ended in the third quarter of the year before the election or later). In other words, most really bad midterm election years do, in fact, follow recessions or depressions.

But does the inverse relationship exist? Are bad recessions inevitably followed by rough midterm elections? Let’s look at the data from a different angle. If we look at recessions since 1929 from most severe to least severe (in terms of GDP decline), we get Aug1929-Mar1933, Feb1945-Oct1945, the present recession, May1937-June1938, Nov1973-Mar1975, Aug1957-Apr1958, and July1981-Nov1982. Every substantial recession here results in the President’s party losing 10 percent of its seats or more. If we follow the dataset further, we find two other recessions that weren’t followed by rough midterm elections for the President’s party: the Recession of 1953-54 and the early 2000s recession. But these were relatively mild recessions; it is understandable why they may not have had the same result on the President’s party.

If you go back even earlier, the economic data become more problematic. But still, if we look at the great contractions in American history that also come close to midterm elections – 1920-1921, 1893-94, 1839-1843, 1873-1879, 1913-1914, 1857-58 – all of them involve disastrous results for the party in power (respectively, -25%, -57% (not a typo), -49%, -48%, -21%, -37%).

In other words, a bad recession occurring close to a midterm election isn’t a necessary condition for a disastrous midterm election, but it seems to be sufficient. Because of the problems described above, you’re going to have a heckuva time discovering this with regression. But it’s pretty obvious nevertheless.

Now, I can think of plenty of reasons why this year’s Democrats might avoid blame for unemployment this time. The most obvious reason is that this recession didn’t begin on Barack Obama’s watch. Even if unemployment still is marching upwards, it started going up on his predecessor’s watch, and he will probably be given at least some credit for this. Additionally, if unemployment begins to decline precipitously, people may perceive that the economy is improving.

On the other hand, unemployment may continue to rise, or peak shortly before the election. Republicans will then have a pretty strong rhetorical argument: Democrats put $787 billion on the country’s credit card for a fiscal stimulus that did nothing. And since the South didn’t fully re-align until the 1990s, this is the first severe recession for a Democratic President where they’re truly vulnerable in all regions of the country. 

The bottom line is that I don’t think Democrats can find much solace in the lack of a statistically significant relationship between the rate of unemployment and midterm election results. If the voters think the economy stinks in 2010, it is going to be a very, very long night for the Democrats, regardless of what economic numbers you’re looking at.

Sean Trende can be reached at

SEND TO A FRIEND | PRINT | | Share Share
Sponsored Links
Related Articles
November 19, 2009
Circling Sharks Smell American Blood - Victor Davis Hanson
November 18, 2009
Worse Than Taxes: The Spending - John Stossel
November 15, 2009
The Fed in the Crosshairs - David Ignatius
Sean Trende
Author Archive