Skip to Content

Stanford GSB News

 

Voting at the Extremes Costs U.S. Representatives Their Seats

November 4, 2002

STANFORD GRADUATE SCHOOL OF BUSINESS—In 1994, Marjorie Margolies-Mezvinsky, a freshman Democrat representing Pennsylvania's 13th Congressional District, cast the deciding "yes" vote on President Clinton's budget. Figuring that tax-averse voters in her suburban district would be angry, delighted Republican congressmen sang, "Goodbye Marjorie," in the well of the House. They were right. The liberal Democrat was swept away in the Gingrich revolution later that year. Margolies-Mezvinsky had plenty of company in that election. More than 40 incumbent Democrats lost their seats.

Twelve years earlier, the reverse had occurred: Nearly two dozen Republican House members lost their seats despite the enormous popularity of President Ronald Reagan. What explains the heavy losses by incumbents in those mid-term elections? Were those elections anomalies, or are there lessons that can be applied to the analysis of future elections and, perhaps more importantly, the behavior of our elected officials?

Research by three Stanford scholars who analyzed the results of more than 6500 Congressional races held between 1956 and 1996 suggests that there is an important lesson: The electorate does hold members of Congress accountable for their votes and will vote them out of office if they stray far from what constituents believe is acceptable.

Moreover, the researchers found that:

  • Voting with the ideological extremes of his or her party can significantly decrease an incumbent representative's re-election prospects.
  • Incumbent representatives like Margolies-Mezvinsky who support their party's sitting President by voting for programs seen as extreme in their districts are often punished by their constituents.

The ground-breaking research was conducted by David W. Brady, the Bowen H. and Janice Arthur McCoy Professor of Political Science and Leadership Values; John Cogan, senior fellow at the Hoover Institution and professor in the public policy program of the School of Humanities and Science; and Brandice Canes-Wrone, then a Stanford graduate student, now associate professor of political science at the Massachusetts Institute of Technology.

The researchers scored Congressional voting records by using the Americans for Democratic Action (ADA) report cards, which reflect the proportion of liberal positions taken by a member in a given year on key votes (taxes, budgets, abortion, school prayer etc.) selected by the ADA. Relative liberalism and conservatism of districts were measured by using presidential voting patterns and demographic data. The researchers controlled for factors such as challenger quality, incumbent and challenger spending, control of the White House, and the condition of the economy.

The Incumbent Edge?
"For at least two decades, scholars have believed that the dominant factor in Congressional elections was incumbency," Brady said in a recent interview. In fact, earlier studies suggested that "the typical representative might be able to vote on legislative matters as she please without fearing that she could lose re-election."

But politicians, Brady notes, "were ahead of the scholars." He recounts a remark made by then-Congressman Lyndon Johnson explaining why he refused to vote for a Civil Rights proposal by President Franklin D. Roosevelt. "You can only go so far in Texas…there's nothing more useless to the Democrats than a dead liberal," the future President said.

Johnson grasped what the scholars didn't: Voters notice how their representatives vote on key issues, and act accordingly. Certainly this flies in the face of conventional wisdom that most of us are ignorant of what actually happens in Congress. How do voters find out? Brady isn't sure, and says the question could be the basis for future research.

But the tie between voting records and prospects for re-election is much clearer. Democrats who move too far to the left in conservative or moderate districts tend to lose their seats; Republicans who move too far to the right also lose. And incumbent Presidents who pull House members to either extreme often lose seats, as evidenced by Clinton's pull to the left in the 1994 election and Reagan's pull to the right 12 years earlier.

By the same token, representatives from very liberal districts, for example Barbara Lee in Berkeley and Oakland, Calif. or conservatives like Texas' Tom DeLay, are most likely safe because their "extremism" mirrors that of their constituents.

If indeed Brady, Cogan, and Canes-Wrone are correct when they conclude that we do hold our representatives accountable for their votes, a hope expressed by James Madison in the Federalist Papers may have come to pass:

"It is particularly essential that the (House of Representatives) should have an immediate dependence on, and intimate sympathy with, the people," Madison wrote. "Frequent elections are unquestionably the only policy by which this dependence sympathy can be effectively secured."

—by Bill Snyder

Related Information

Continuity and Change in House Elections, edited by David W. Brady, John F. Cogan, and Morris P. Fiorina; Stanford University Press, November 2000

Out of Step, Out of Office: Electoral Accountability and House Members' Voting, American Political Science Review, March 2002

Critical Elections in the U.S. House of Representatives, David Brady, Stanford University Press, 1988

Video

Voting Reform: Exorcizing the Vote
Uncommon Knowledge, PBS, November 2000
Video File, (26:50 minutes, RealPlayer® format)
Audio File, (28:29 minutes, RealPlayer® format)

The Political Economy of Growth and Development
Stanford Graduate School of Business, 75th Anniversary, May 2000
Video File, 1:25 hour