Martin P. Wattenberg
The Democrats' Decline in the House During the Clinton Presidency: An Analysis of Partisan SwingsThe three elections for the House of Representatives held thus far during the Clinton presidency have all proved to be disappointing for President Clinton. In 1994, his party lost control of the House for the first time in 40 years. The great animosity of Republican elites towards Clinton helped the GOP recruit a much better-than-expected crop of House candidates, and the president's unpopularity with the swing voters who supported Ross Perot led to a substantial shift against the Democrats (see Wattenberg, 1998, chapter 11). Stunned by this historic turn of events, President Clinton adopted a reelection strategy based on triangulating his policies between that of congressional Democrats and Republicans. Although successful for him personally in 1996, this strategy failed to carry his party back into the majority in the House -- an undesirable historic first for a Democratic president. Under the cloud of scandal two years later, Clinton avoided campaigning during the final mid-term election of his presidency. Yet, contrary to all predictions that his legal predicament would hurt the Democrats, the party actually picked up House seats. This marked only the second time since the Civil War that the party of the president had gained seats in a mid-term election. Most analysts thought this stunning result would discourage Republicans in the House from moving forward with impeachment proceedings against President Clinton. Republicans were not dissuaded, however, and the month after the election the lame-duck House cast a nearly party-line vote making Clinton the first elected president ever to be impeached.
This article examines the changes in the congressional district vote during these three sets of House elections held during the Clinton presidency. One of the great constraints on any president is the short political memories of members of the House of Representatives. Facing election every two years, their time perspective is necessarily different from the president's. To them, each election result must be compared to that of just two years before. The swing in their district's voting behavior from the previous House election gives them a sense of the winds of political change, and influences how they will act over the next two years. By examining the percentage vote swing in each election by incumbency type (i.e., Republican running for reelection, open Republican seat, etc.) it is possible to derive a fuller picture of how House members interpreted these elections than from the simple number of House seats won. In particular, such a perspective sheds light on why Republican House members were not discouraged from pursuing impeachment by the 1998 election results.
Examining the Vote Swing in the 1994, 1996, and 1998 House Races
The great swing in the vote for the House between 1992 and 1994 not only led to the first Republican-controlled House in four decades, but also to a sense of Republican hubris. Despite the narrow margin of seats by which they won control of the House, GOP leaders pursued an aggressive strategy of trying to ram an ambitious legislative agenda down the president's throat. President Clinton even felt compelled to remark that the presidency remained relevant, and eventually he proved the point by blocking the GOP budget plan even in the face of a protracted shutdown of parts of the federal government. Many commentators argued that the Republicans overstretched their mandate out of reaction to the sheer frustration of having been in the minority too long. While this no doubt played a role, an examination of the vote swing by type of district shows how House Republicans may have overestimated their electoral strength by focusing on the results from their own districts.
As can be seen in Table 1, the average Republican incumbent saw his or her percentage of the two-party vote increase by 8.1 percent between 1992 and 1994. Such a strong increase in their own support could well have given veteran GOP House members a sense of overconfidence. On the other side of the aisle, after the election they could observe 35 incumbent Democrats who had been defeated. However, even with the defeat of so many veteran Democrats the average vote swing against Democratic incumbents only amounted to 4.4 percent. The rate by which Republican incumbents gained votes was thus nearly twice that of the typical decline in the vote for Democratic House members. Republicans may therefore have overestimated how much the tide of public opinion had turned in their favor.
Republican strategists may also have ignored the lack of a strong swing against Democratic incumbents in 1994 because of their overwhelming success in districts where Democrats had retired, thereby leaving an open seat. The Republicans gained an average of 11.5 percent in these open seats, winning 22 out of 31 of them. By focusing on these results, they could have logically drawn the conclusion that they would continue to benefit in the future from Democratic retirements. The "Contract With America" guaranteed a vote on term limits, and most Republicans in 1995 voted for this provision in expectation that reducing the incumbency advantage would help them in the long run.
Yet, it was the Republicans' newfound advantage of incumbency that was most responsible for them being able to hold on to their slim House majority in 1996. Ninety percent of House Republicans who stood for reelection in 1996 were successful, as the average swing against them was only 3.0 percent. Those who had been in office for more than one term saw a median loss of 6.8 percent of the two-party vote from their previous showing, but given their big 1994 victory margins it was a loss most could easily afford. In contrast, many members of the Republican class of 1994 had won office narrowly and needed the incumbency advantage to survive their first reelection campaign. Half of them actually improved their share of the two-party vote in 1996, as compared to only 8 percent of the more veteran Republicans; only 10 percent experienced a swing against them greater than the median figure for those with more seniority. Although more of these first-termers suffered defeat than the GOP veterans, the toll would have been much greater had it not been for the edge of incumbency they enjoyed. Furthermore, the Republicans also reaped an advantage in 1996 when more Democrats than Republicans retired from the House, and the GOP ended up with a net gain of five seats in the open races. As in 1994, these gains were centered in the South, a region that has been gradually realigning in favor of the Republicans.
Ironically in light of subsequent events, President Clinton exerted relatively little effort toward regaining Democratic control of the House in 1996. Although Clinton did make some specially arranged appearances to help unseat vulnerable Republicans like California's Bob Dornan and Andrea Seastrand, he shied away from asking voters to give him a Democratic Congress. As one Clinton aide explained, "voters are suspicious of partisanship" and by making such a request the president feared he would be reminding "a lot of them that they'd rather not see a concentration of power in one party's hands (Apple, 1996)." Confirmation of Clinton's feelings regarding this subject can be found in two statements he made in January of 1996. He told David Broder (1996) that, "The American people don't think it's the president's business to tell them what ought to happen in a congressional election." Furthermore, he remarked that, "A president has to be careful how he makes these arguments to the American people, because a lot of time in our history, they would prefer having a president of one party and the Congress the other."
Other reports during 1996 speculated that Clinton may have even preferred a slight Republican majority in the House to a slight Democratic majority. The reasoning behind this was that the latter would result in the Democrats being blamed for most everything while being put in the extremely difficult situation of requiring perfect party unity to pass anything of significance. In any event, President Clinton must have regretted not having done more to help Democratic House candidates in 1996 once the Republican-controlled Judiciary Committee started to pursue his impeachment.
Many analysts thought the impeachment process was dead in the water after the Republicans failed to pick up seats in the 1998 mid-term election. But as can be seen in Table 1, the results were not so disappointing for the typical GOP House incumbent. On average, those who faced a Democratic challenger gained 2.8 percent over their electoral performance two years earlier. Fifty-five Republican incumbents got a free pass for another two years, facing no Democratic opponent. All told, 74 percent of Republican incumbents either won reelection unopposed or saw their share of the two-party vote increase. If Republicans turned to each other in their cloakroom after the election and asked how they had done in the campaign, the answer three times out of four would have been: "Better, thanks."
Furthermore, there was no evidence that Republicans suffered losses in districts where Clinton had done well in 1996. As can be seen in Table 2, Republican incumbents increased their vote share even in districts in which Clinton had won over 50 percent of the vote in 1996. It might be thought that these results are skewed by the lack of visible Democratic challengers in some of these districts in 1998. The second column of Table 2 controls for this by limiting the analysis to only districts in which Democratic challengers spent over $200,000. In these 47 districts, the pattern indicates that Republicans were actually slightly more likely to pick up votes in areas where Clinton had done relatively well in 1996. Thus, it seems eminently reasonable for Republican lawmakers to have interpreted the results in their districts as indicating that Clinton's popularity did not impact their elections in 1998. The vote on impeachment the following month was thus determined by conscience and perhaps pressure from the party – not by the election results.
The Clinton Legacy As Judged by Congressional Results
Scholars often judge a presidency by whether the president left the country in better shape than he found it. One might also judge a presidency by whether his party ends up in better shape than it started when the president took office. If so, it seems virtually certain that Bill Clinton will have failed to achieve the desired result on this criterion. Since Bill Clinton took the oath of office the Democrats have thus far lost 47 seats in the House. Following his acquittal on impeachment charges, Clinton's aides made it clear that the president considered it a top priority to regain Democratic control of the House in the elections of 2000. This is certainly a realistic goal given that it would require a gain of only 7 seats. However, given that only 23 seats were won by the Republicans in 1998 with less than 55 percent of the two-party vote, the chances of the Democrats returning to their 1993 level by the time Clinton leaves office appears fairly remote. The legacy of Bill Clinton in the House will almost certainly be one of decline for his party.
Table 1 Partisan Swings In House Races in 1994, 1996, and 1998 by Incumbency Status
Contested Republican Incumbents
Open Republican Seats
Contested Democratic Incumbents
Open Democratic Seats
1994
Seats lost
0
4
35
22
Mean partisan swing
8.1R
4.9D
4.4R
11.5R
N of districts
118
20
198
31
1996
Seats lost
18
3
3
8
Mean partisan swing
3.0D
7.2D
4.3D
4.3R
N of districts
171
19
149
28
1998
Seats lost
5
6
1
5
Mean partisan swing
2.8R
6.1D
3.1D
3.7R
N of districts
150
19
146
18
Note: Only elections that were contested by both of the major parties in each of the two years being compared are included. Districts in which the House member switched from the Democratic to the Republican Party in 1995 have also been excluded for the 1996 analysis.
Source: Author's analysis of official election returns.
Table 2 Average Republican Gains in the Vote Between 1996 and 1998 by Clinton's 1996 Vote Share in the District (number of cases in parentheses)
All Contested Republican Incumbents
Contested Republican Incumbents Who Faced Democrats Who Raised over $200,000
Clinton 1996 vote < 40%
+3.2
(43)
+0.7
(6)
Clinton 1996 vote >40% < 45%
+2.0
(38)
+0.3
(13)
Clinton 1996 vote >45% < 50%
+3.2
(44)
+1.2
(19)
Clinton 1996 vote > 50%
+2.7
(25)
+1.7
(9)
Source: Author's analysis of official election returns.
Bibliography
Apple, R. W., Jr. 1996. "Clinton Pursues Party Sweep Without Whispering a Hint." New York Times, October 23, p. A1.
Broder, David S. 1996. "Clinton Can't Do It Alone." Washington Post Weekly Edition February 12-18, p. 4.
Wattenberg, Martin. 1998. The Decline of American Political Parties, 1952-1996. Cambridge, MA: Harvard University Press.
Endnotes
1 Throughout this analysis, districts are excluded if one or the other major party failed to nominate a candidate in one of the elections being compared.