Partisan Bias in U.S. Congressional Elections, 1952-1996

• Author: THOMAS L. BRUNELL
• DOI: 10.1177/1532673X99027003003
• Date: 01 July 1999
• Published date: 01 July 1999
• Subject Matter: Articles

AMERICAN POLITICS QUARTERLY / JULY1999

Brunell/PARTISANBIAS

PARTISAN BIAS IN

U.S. CONGRESSIONAL ELECTIONS, 1952-1996

Why the Senate Is Usually More

Republican Than the House of Representatives

THOMAS L. BRUNELL

Binghamton University (SUNY)

The Democratic Party held a majority of the seats in the House of Representativesfrom 1952 to

1994.Duringthesameperiod,however,they did not demonstrate the same level of success in the

Senate. What explains these differinglevels of success in the two chambers of Congress? Here I

explore the contributionof partisan bias to this discrepancy. Three separate measures of partisan

bias (distributional, population, and turnout) are explained and calculated for each chamber for

the elections from 1952 through 1996. These estimates of partisan bias are linked to the propor-

tion of seats that the parties have controlled in both the House and the Senate. Results indicate

that the Democrats reaped the benefits of bias in House elections (particularly turnout-related

bias). Bias in the Senate, particularly distributional and population-related bias, favored the

Republicans.

Before the 1994 congressional elections, the Democrats had con-

trolled the House of Representatives for more than 40 straight

years—aremarkable streak. Althoughthe Democrats didquite well in

Senate elections during this same time period, they did not dominate

the chamber in the same manner as they did in the House. Indeed, in

only 5 of the 23 Congresses between 1952 and 1996, did the Demo-

cratscontrol a largerproportion of the Senate compared to the House.1

Furthermore, the Republicans controlled the upper chamber for the

first 6 years of Reagan’s tenure in the White House (1981-1987).

This phenomenon can be thought of as a specific piece of the divided

government puzzle.2More often than not, one party or the other con-

trols a majority of seats in both chambers of Congress; however,this

is not always the case. What explains the differing levels of success

Author’s Note: I am especially thankful to Bill Koetzle and Bernie Grofman for their helpful

comments.

AMERICAN POLITICS QUARTERLY,Vol. 27 No. 3, July 1999 316-337

© 1999 Sage Publications, Inc.

316

at SAGE Publications on December 5, 2012apr.sagepub.comDownloaded from

bythe Democrats inthe two houses ofCongress? Here, an explanation

based on the differing sources and levels of partisan bias, which is a

measureof symmetry (or lack thereof) in an electoral system, for elec-

tions to each chamber is proposed and tested.

In two-party competition, partisan bias is defined as the difference

between the seat share that one party can expect with 50% of the vote

relative to what it should win if both parties were treated equally; that

is, both parties receive 50% of the seats (Tufte, 1973). Thus, partisan

bias is a measure of how much, and in what direction, an electoral sys-

tem distorts the relationship between votes and seats. Because the

electoral systems of the two chambers of the American legislature are

so different, it is not unreasonable to expect that partisan bias for each

will differ as well.3

PREVIOUS RESEARCH

Thereare a numberof reasonable explanationsfor the differing lev-

els of partisan success in the House and Senate. Redistricting in the

House, particularly when done by a partisan majority, can endow a

party with a distinct advantage over time. However, prior research has

shownthat purely partisan redistrictingis the exceptionrather than the

rule (Robertson, 1983). Still, the Senate undergoes no such redistrict-

ing and this discrepancy may contribute to the Democrats’heightened

level of success in the House. However, Oppenheimer (1989) con-

cludes that the impact the redistricting process has on the partisan

makeup of Congress “rarely exceeds a couple of seats even in large

population states” (p. 659).

Institutional prestige may account for the differential partisan suc-

cess rate in the two chambers. The Senate is more prestigious, has

longer terms, and its members are more likely to have a national pro-

file. Senate races also typically attract better quality candidates with

more campaign money than the average House election (Krasno,

1994). These differences may lead quality Republican candidates to

be less likely to consider running for the lackluster House, whereas a

position in the Senate holds enough prestige and power to attract the

very best of the Republican Party.

Brunell / PARTISAN BIAS 317

at SAGE Publications on December 5, 2012apr.sagepub.comDownloaded from