It’s not too early to start thinking about midterms.
Eleven months from now, in November 2010, Americans will be voting in Congressional elections. Should we Democrats be worried about the House and the Senate? I decided to look back in history, to see what happened in past midterms. Are there any patterns?
There is no crying in baseball and there are no coattails in midterms. More beyond the fold.
Riding the Winner’s Coattails
In a presidential year, the winning candidate’s victory usually – but not always – benefits his party in Congress. But in midterms, the sitting President’s party quite often loses some seats. The following table shows how the presidential and midterm elections affected the party in power (for example, when Reagan was elected, we see how the Republicans did in Congress and when Clinton was elected, we see the results for Democrats). The numbers (+16 or -9) tell you how many seats the President’s party gained or lost compared to the election two years before.
Important note: The table doesn’t tell you which party had the majority in Congress. In the 1996 House races, for example, the Democrats had a gain of +8 seats (going from 198 to 206), but Democrats were still the minority party.
Year | President | - | House | - | Senate | Midterm | - | House | - | Senate |
2008 | Obama | D | +21 | D | +8 | 2010 | - | - | - | - |
2004 | Bush 2 | R | +3 | R | +4 | 2006 | R | -30 | R | -6 |
2000 | Bush 2 | R | -2 | R | -4 | 2002 | R | +8 | R | +2 |
1996 | Clinton | D | +8 | D | -2 | 1998 | D | +5 | D | 0 |
1992 | Clinton | D | -9 | D | 0 | 1994 | D | -54 | D | -8 |
1988 | Bush 1 | R | -2 | R | -1 | 1990 | R | -8 | R | -1 |
1984 | Reagan | R | +16 | R | -2 | 1986 | R | -5 | R | -8 |
1980 | Reagan | R | +34 | R | +12 | 1982 | R | -26 | R | 0 |
1976 | Carter | D | +1 | D | 0 | 1978 | D | -15 | D | -3 |
1972 | Nixon (Ford) | R | +12 | R | -2 | 1974 | R | -48 | R | -4 |
1968 | Nixon | R | +5 | R | +5 | 1970 | R | -12 | R | +1 |
1964 | Johnson | D | +36 | D | +2 | 1966 | D | -48 | D | -3 |
1960 | Kennedy | D | -20 | D | -1 | 1962 | D | -4 | D | +2 |
1956 | Eisenhower | R | -2 | R | 0 | 1958 | R | -48 | R | -12 |
1952 | Eisenhower | R | +22 | R | +2 | 1954 | R | -18 | R | -2 |
1948 | Truman | D | +75 | D | +9 | 1950 | D | -28 | D | -5 |
1944 | Roosevelt (Truman) | D | +20 | D | -1 | 1946 | D | -54 | D | -12 |
1940 | Roosevelt | D | +5 | D | -3 | 1942 | D | -45 | D | -8 |
| Average | | +12 | | +1 | | | -25 | | -4 |
(The information above was collected laboriously from several dozen Wikipedia pages about election results.)
In a presidential year, the President’s party gains an average of 12 seats in the House and one in the Senate (the coattail effect). In a midterm election, the President’s party loses an average of 25 House and four Senate seats.
The numbers (presidential-year gains and midterm losses) were much larger from 1940-1986. Since 1988, both effects have diminished:
Average 1940-1986: House +17 and -29; Senate +2 and -5.
Average 1988-2008: House +3 and -16; Senate +1 and -3.
It’s not that simple, of course. The numbers vary a lot from one election to another. Plus, there are 435 House elections every two years and only 33 or 34 Senate races, so you see bigger swings in the House.
Every election has different issues, different candidates, different moods. A few examples: In 1966, The Democrats had just passed Civil Rights legislation and had started to lose the South (and don’t forget about Vietnam). In 1974, Nixon had just resigned, which probably contributed to Republican losses in the midterm. In the 2002 midterm, the Republicans actually gained 8 seats in the aftermath of 9/11.
Take a look at the years where a President had extra big coattails: 1948 (Truman +75 in the House and +9 in the Senate), 1964 (Johnson +36 House, +2 Senate), and 1980 (Reagan +34 and +12). They were all followed, two years later, by elections where the President’s party lost a ton of seats in Congress.
Why Are Midterms Bad for the President?
I think the simplest reason is this: Turnout is always lower for midterms.
There are a certain number of people who only vote in Presidential years. And once they decide which candidate they like for President, they may be more likely to vote for that candidate’s party for Congress. That’s how we get the coattail effect.
Midterms have a slightly different psychology. First of all, there are some people who just don’t vote (because it’s not a presidential year). Also, the people who dislike the sitting President might be angrier and more motivated to vote against his party. And consider this: In the midterms, people might vote against the President’s party because they think he has done the wrong things (too much too soon). Other people might avoid voting because they think the President hasn’t done enough of the right things (not enough, too slowly). Either way, the President’s party suffers.
Here’s the voter turnout in presidential and midterm elections:
President | Turnout | Midterm | Turnout |
2008 | 56.8% | 2010 | -- |
2004 | 55.3 | 2006 | 37.1% |
2000 | 51.3 | 2002 | 37.0 |
1996 | 49.1 | 1998 | 36.4 |
1992 | 55.1 | 1994 | 38.8 |
1988 | 50.1 | 1990 | 36.5 |
1984 | 53.1 | 1986 | 36.4 |
1980 | 52.6 | 1982 | 39.8 |
1976 | 53.6 | 1978 | 37.2 |
1972 | 55.2 | 1974 | 38.2 |
Average | 53.2% | | 37.5% |
(These numbers come from infoplease.com. The percentages are based on all adults aged 18 or over, including people who can’t vote such as felons and undocumented aliens. I stopped at 1972 because that’s the year the voting age was lowered to 18. Before 1972, the turnouts were slightly higher.)
Going from 53% (presidential year) to 37% (midterm) means approximately a third of the people who voted in the presidential election didn’t bother to vote in the midterm. And, in fact, if you break it down, it’s mostly young people who don't bother to vote. Here are the age splits from the last two elections (voter totals are 000s):
. | 2008 | Presidential | . | 2006 | Midterm | . |
Age | Eligible | Voted | Turnout | Eligible | Voted | Turnout |
18-24 years | 25,791 | 12,515 | 48.5% | 24,954 | 5,524 | 22.1 % |
25-34 years | 34,218 | 19,501 | 57.0 | 33,215 | 11,137 | 33.5 |
35-44 years | 36,397 | 22,865 | 62.8 | 37,520 | 17,079 | 45.5 |
45-54 years | 41,085 | 27,673 | 67.4 | 40,322 | 21,708 | 53.8 |
55-64 years | 32,288 | 23,071 | 71.5 | 30,433 | 19,017 | 62.5 |
65-74 years | 19,571 | 14,176 | 72.4 | 18,208 | 11,700 | 64.3 |
75+ | 16,724 | 11,344 | 67.8 | 16,420 | 9,954 | 60.6 |
Total Adults | 206,072 | 131,144 | 63.6 | 201,073 | 96,119 | 47.8 |
. | | | | | | |
Age 45+ | 109,668 | 76,264 | | 105,383 | 62,379 | |
Percentage | 53.2% | 58.2% | | 52.4% | 64.9% | |
(These numbers are from census.gov, and the percentages don’t match the previous table because the census bureau only counted citizens over 18 who were eligible to vote; undocumented aliens and other non-citizens were ignored. Also, there may have been some people who didn’t vote, but claimed that they did.)
The first fact that jumps out at you is this: The older you get, the more likely you are to vote. The second thing is this: Turnout is less in the midterm, but it's especially low among young people. Look at those numbers. In 2008 (presidential year), people over 45 were 53% of the eligible voters and 58% of the actual voters. In the 2006, the most recent midterm, they were 52% of the eligibles and 65% of the actuals.
I don’t have a source, but I remember reading somewhere that, in 2008, younger people tended to vote more for Democrats and older people tended to lean more Republican. You do the math.
From General to Specific
So far, I’ve just looked at past elections. There are a few general conclusions about the trends:
- The President’s party usually loses seats in the midterms.
- Voter turnout is usually lower in midterms.
- Younger people are much less likely to vote in midterms.
Now, I’d like to say a few things about 2010.
Republicans might pick up some seats, because they’re the party out of power (POOP). Republicans are POOP. Hehehe.
Incumbents usually have an edge. I read somewhere that 90% of incumbents get re-elected. It sounds about right (90%), but it also kind of sounds like a guesstimate. Incumbents get re-elected because they fit the district (red districts vote for red candidates, blues for blues). Also, a majority of campaign donations go to incumbents, which means incumbents can spend more on advertising. Also, incumbents have an advantage in name recognition.
Brand new members of Congress might be vulnerable. And open seats might be up for grabs. So far, there are at least 28 non-incumbent House seats (House Races in which the 2008 Winner is Not Running). Some might be safe seats. Others might switch sides.
Watch the old people. Whether or not you think it’s a good idea, extending the Medicare age to 55 might help Democrats in the midterms. Also, whether or not it’s a good idea, the Afghan surge might be more popular among older voters. (Please don’t yell at me for saying this.)
It’s the economy, stupid. Or is it? The National Bureau of Economic Research has a list of expansions and contractions going back to 1857 (Business Cycle Expansions and Contractions). I tried to match midterm results and recessions and couldn't get a correlation. Clinton didn’t have a single recession during his eight years in office, but the Democrats lost big in 1994. I think the economy has started to turn around. But unemployment might still be a major problem when November comes. It could be bad news for Democrats.
No doubt there are lots of other issues or factors, but I’ll mention one more potentially important thing: Teabaggers. If they succeed in splitting the Republican party, it might help the Democrats a lot. The Republicans remind me of "Monty Python and the Holy Grail" in so many ways. Before you can cross the bridge, you have to answer three questions ("You must answer the Republican purity test"). The black knight gets his arms and legs cut off and keeps thinking he can win the battle ("The country is center-right; it’s only a flesh wound!"). But here’s a scene that I love. She’s a witch! Burn her! ("He’s a RINO! Primary him!") Imagine Limbaugh and Beck as members of the ignorant mob below.
That's it. I hope you learned something about midterms. I’d love to hear what other people think about 2010. What’s your analysis? What's your opinion?