Will Democrats Face a 1994 Repeat in 2010?

Will Democrats Face a 1994 Repeat in 2010?

By Jay Cost and Sean Trende - March 9, 2009

In The Eighteenth Brumaire of Louis Napolean, Karl Marx is commonly quoted as stating that history tends to repeat itself, the first time as tragedy, the second time as farce. [1] Some conservative commentators see hope in Marx's dictum, and believe that the stimulus bill, mortgage bailout, and Obama's budget, combined with a shaky economy, are setting Democrats on a course similar to the one they charted in 1994. If 1994 was a tragedy for Democrats, who were famously unsuccessful in their attempts to present themselves as "New" versions of the early liberal party, 2008 will be a complete farce, with Democrats happily portraying themselves as liberal caricatures, and wiping out the best chance they have had in a generation to reshape American policy.

There are certainly some parallels to 1994. But there are also important differences as well. It is worth taking some time to explore both. But if I were a Republican, I wouldn't break out the champagne just yet.

Factors Favoring Republicans

There are some parallels with 1994 that are favorable for Republicans. The first is the tendency of voters to turn out members of the President's party during midterm elections. This has been the general rule in American politics since Andrew Jackson's Presidency (interestingly, before that there had been a weak tendency for the out-party to gain seats in the Presidential election, and lose seats in the midterm).

There are several hypotheses in the political science literature about why this phenomenon occurs, and this column does not set about to explore them fully. For our purposes, we can just note that two of the proposed theories - that voters vote strategically for divided government and that midterm losses represent a withdrawal of Presidential coattails - weigh against the Democrats. Obama, after all, is a Democrat and he certainly had coattails, especially in districts with moderately high African American percentages. We will examine how the third theory of midterm loss - that midterm elections serve as a referendum on the incumbent President - plays out later.

Second, and perhaps most ominously, Democrats are almost as overexposed in Congress today as they were in 1994. It is true that unlike the Democratic Congress in 1994, this year's Democratic caucus is comprised of fewer Representatives from the South. Rhodes Cook explores this more thoroughly in an excellent column titled "Not Your Father's Democratic Congress," but the bottom line is that compared to 1994, many more Democrats today hail from coastal areas than from the South.

But 1994 wasn't just a debacle for Democrats in the South. In fact, the South wasn't even their worst region in 1994. Of the fifty-six seats the Democrats lost to Republicans (Democrats also gained four Republicans open seats, leading to a net loss of fifty-two seats), only sixteen seats were from the South. This represented 13% of the Southern seats held by Democrats. In percentage terms, the Democrats' losses in the Midwest, Mountain West, Pacific coast, and Plains regions were worse. After the 1994 elections, Democrats still held 49% of the seats in the South. In other words, the less-Southern face of today's Democratic party should not be a source of comfort for them.

Setting aside regional variations, it is useful to examine how the Democrats' seats were distributed in 1994 in terms of Partisan Voting Index, or PVI. This represents how a district voted for President compared to the national vote. It works like this: In 2008, Obama won nationally by about seven points. If a district went for Obama by eight points, its PVI is D+1. If it went for Obama by six points, its PVI is R+1. Using this approach instead of simply looking to see who won a district makes intuitive sense: In 1984 Reagan carried all but three of Massachusetts' eleven districts, and only narrowly lost two of the remaining three, but no one would call them "Republican-leaning" or swing districts. Likewise, a district that Obama carried by a point probably isn't all that Democratic in reality.

Table I below shows the PVI distribution for Democrats in 1994.


We see that, based on the 1992 results, 79 Democrats were in districts with Republican PVI's, while 46 Democrats were in districts that Bush actually carried in 1992 (districts with PVI's of R+6 or more).

1992 is a tricky year to calculate PVI's because of Perot's presence as a successful third party. While exit polling showed that Perot pulled about evenly from Clinton and Bush nationally, it seems likely that there was some regional variation in whom he pulled from. So I went ahead and calculated the PVI's using the 1988 results for the districts as well. We see a similar situation. Eighty-seven Democrats held districts that George H.W. Bush carried with 53% of the vote (his national average) or more.

Table II below shows the Republican-leaning PVI districts represented by Democrats based on the 2004 and 2008 results (2008 results have been painstakingly compiled by Swing State Project).


We see that Democrats are in somewhat better shape today than they were in 1994, but not by much. There are 66 Democrats in Republican-leaning districts based on the 2008 results, and 65 based on the 2004 results. Moreover, in 2008 there were more Democrats in heavily Republican districts with PVI's of 16+ -- meaning McCain carried the district by roughly ten points or more - than there were in 1994. Even though Bush the Elder's 1992 losing margin was narrower than McCain's 2008 margin, there are more Democrats today in districts McCain carried than there were Democrats in districts Bush carried in 1992.

As for the Republicans, they have been eviscerated in truly vulnerable districts. Only fifteen Republicans represent districts with Democratic-leaning PVI's based on the 2008 results, and only eight represent districts with Democratic-leaning PVI's based on the 2004 results. While one Republican - Anh Cao - will almost certainly lose in 2010, and there may be additional open seats that enhance Republican vulnerability, most of these Republicans have already survived two Democratic waves, proving their staying power.

The third element favoring Republicans compared with 1994 is hinted at by Cook in his column: These Democrats are much more ideologically in sync with each other than were the Democrats in 1994. This is good for Obama's short term political agenda, but it may not be helpful in the long run.

Many people who talk about conservative Southern Democrats today forget what a conservative Southern Democrat really used to mean. From 1968 to 1970, only twenty Southern Democrats ever let their Americans for Democratic Action score ever rise above 33%; most maintained Americans for Constitutional Action (a conservative interest group) scores well above 50%. Even in 1994, there were several Democrats such as Sonny Montgomery and Billy Tauzin who were able to hold onto their seats and even avoid serious races in 1994, mainly because they had a solidly conservative voting record that matched their district's mood.

And these conservative Democrats generally did well. It was not the Tom Bevills and Glen Browders of the Democratic party - those who compiled conservative voting records in conservative districts -- that had to sweat it in 1994. It was the Karan Englishes and Herb Kleins, who represented marginal districts and compiled liberal voting records, who were wiped out.

Many of today's Democrats are positioning themselves to be more of a Karan English than a Tom Bevill. In 2008, only a handful of Democrats allowed their ADA score to drop below 75%. None were below 50%. And only a handful voted against the stimulus package, meaning that over forty Democrats who represent districts carried outright by John McCain voted for the stimulus bill.

That is not a problem right now - after all, Barack Obama has approval ratings around 60%. But he is in his honeymoon phase right now. George W. Bush, Bill Clinton, and Jimmy Carter all had comparable ratings at similar points in their Presidency.

All honeymoons end, and with an economy on a shaky foundation, the ending of this honeymoon could prove to be particularly brutal for Democrats (the economy could also rebound, in which case the central tenet of the 1994-as-2010 analogy is removed at the outset). As we noted, the third theory of midterm loss proposed by political scientists is that midterm elections serve as referenda on the incumbent President's party. If Obama remains popular, the referendum will probably result in only a small loss of seats (a gain is possible, but given that Obama had coattails, unlike Bush and 2000 and Clinton in 1996, this seems unlikely).

But if the public begins to place blame for the economy at Obama's feet, or otherwise sours on his agenda, these Democrats from Republican-leaning districts will not want to go into 2010 with advertisements bleating that they have voted with Barack Obama over ninety percent of the time. Three issues really sunk the Democrats in 1994: the 1993 budget, the assault weapons ban, and the ban on gays in the military. If Obama turns out to be unpopular, the stimulus bill, the budget bill, the mortgage bailout, and several other upcoming votes will probably give Republicans ample ammunition to use in the midterms, provided the Democrats continue to vote in large numbers for these bills.

Factors favoring Democrats

But there are important factors that favor the Democrats as well. Perhaps the most important factor is the retirement situation. Of the fifty-six seats Republicans picked up in 1994, twenty-two were open seats. Some of these seats the GOP likely would have won no matter what (Sam Coppersmith's AZ-01 comes to mind), but certainly not all of them.

Democrats have done well avoiding retirements in recent cycles, and many of the Democrats in these Republican leaning districts are not retirement threats. While many of them are relatively junior members of Congress, which could hurt them, almost any incumbent is better than an open seat. Republicans will probably only have a handful of vulnerable open seats to target in 2010. At the same time, Republican members may decide that now is a good time to pack it in, since many do not see a Republican majority on the horizon. Being in the minority is not fun, and jobs in the private sector that promise more money and an actual ability to impact things are likely an attractive alternative to public service for many Republicans. Though again, the relative paucity of Democratic-leaning seats held by Republicans mitigates this risk somewhat.

The second problem for Republicans is that, unlike 1994, Republicans have had control of Congress recently. In 1994, Republican control of Congress was something of an unknown quantity. Polling indicates that even in July of 1992, when Clinton was riding high in his campaign and President Bush was struggling with sub-40 approval ratings, Democrats held only a ten-point advantage over Republicans when asked who people trusted with the nation's problems.

Today, Democrats hold an astounding 26-point gap over Republicans on that question (though the "leaned" partisan identification is closer -- it was D+14 in February of 1992 and is D+12 today). This is not good for Republicans, because voters evaluating an incumbent typically go through a two-step process. First, they decide whether they like the incumbent. If the answer is yes, they generally vote for the incumbent. But if the answer is no, they then proceed to the second question, which is whether the challenger is an acceptable alternative. If they answer to that is yes, they vote for the challenger. If it is no, they vote for the incumbent.

This is a real danger for Republicans - the voters may sour on Democrats by 2010, but this does not necessarily mean that they will have warmed to Republicans so quickly. As we have noted, in 1994, having Republicans in charge of Congress was an unknown quantity, as was having Democrats in charge of Congress for many voters in 2006 (throughout 2005 and 2006, Democrats maintained a net-positive approval in the public's minds; today Republicans typically receive majority-unfavorable ratings). And so far, the efforts to mount a coherent opposition, build an agenda, and improve the Republican party's branding have not been impressive. With an unfavorable view of the Republican party, voters may stick with the devils they know and dislike less.

Along these lines, the back-to-back drubbing of Republicans may scare away donors and quality challengers, who wait for a better year to get a better return on their investments and attempts to advance their careers. This is something of a vicious cycle -- when donors and candidates wait for the atmosphere to improve, it makes it more difficult for the atmosphere to improve -- but that is one of the disadvantages of being in the minority.

The final problem for Republicans is that Democrats may not be as hubristic as they were in 1994. I'm not talking about the Democratic leadership or President Obama here -- I'm talking about individual Democrats. It is easy to forget today, but the idea of Republican control of Congress in 1994 was unthinkable. Scholarly articles touted the permanent Democratic lock on the House (just as they had touted the Republican lock on the electoral college). Only one member of either party -- Sidney Yates -- had served under a Republican Speaker. Democrats had only suffered four elections where they lost more than twenty seats since the Roosevelt Administration. It was easy to support the leadership's agenda, given this assumption of job security. And it was easy for the leadership to bully these Democrats with the argument that they would have continued job security.

Today, Democrats know well that they can lose control of Congress. With a 40-vote margin, they can afford to let many vulnerable Democrats vote against controversial packages, and these vulnerable Democrats can take turns voting "nay." Democrats may even start voting against their party on procedural votes to slash their party unity scores -- we saw an example of this in the lead-up to the vote on the recent $410 billion spending bill, where Bart Stupak (MI-01) voted "nay" on two procedural votes before voting "yea" on final passage. In other words, Democrats may choose to push through much of Obama's agenda, while keeping their most vulnerable members from amassing voting records that ADA largely approves of.


As the out-party in a midterm election, with the economy unlikely to recover fully by 2010 and with many Democrats in Republican-leaning districts, Republicans are poised to perform well in 2010. But Republicans have some problems to deal with before they can return to the promised land. 2010 may be too soon for them to fix those problems, and to take full advantage of their advantages. The true farce of 2010 may be a Republican party that finds itself well-positioned to take control in Washington, but completely unable to do so.

[*]For the purists who are undoubtedly preparing to clog my e-mail box as we speak, yes, this is not what Marx said. The exact quote is "Hegel remarks somewhere that all great world-historic facts and personages appear, so to speak, twice. He forgot to add: the first time as tragedy, the second time as farce." Most people think of the quote the way I described it, and explaining this seems like an awful big distraction for an introductory paragraph, no?

Sean Trende can be contacted at

Jay Cost and Sean Trende

Author Archive

Follow Real Clear Politics

Latest On Twitter