Overview

    Just who decides elections to the "people's house" in the United States? The obvious answer is that voters elect the House of Representatives. The not-so-obvious fact is that most Americans, most of the time, experience "no choice" House elections. They live in political monopolies where the talk should be of creating a two-party system, let alone one with viable third parties. With voter turnout shrinking to one of the lowest levels in the world -- a recent study by the Institute for Democracy and Electoral Assistance found that in national elections since World War II, the United States ranks 103rd in voter participation out of 131 democracies -- and with deep citizen concern over the impact of money in politics, it is high time that we ask if these political monopolies should continue.

    Monopoly Politics demonstrates clearly that in deciding which party wins an election, it matters most who the voters are in a district. In 1996, for example, there were 53 open seat elections. The Democrats carried every single seat in which their presidential candidate Bill Clinton won more than his national average of 49%. At the same time, Republicans won 29 of the 35 open seats in districts where Clinton ran behind this average. In the 11 most conservative districts -- where Bill Clinton ran most poorly, winning less than 41% -- most Democratic candidates spent over $300,000, but only one gained more than 41% of the vote.

    Most U.S. House elections are not competitive for one simple reason: a clear majority of voters in a given district prefer one party's political philosophy over that of the other party. They vote a certain way in the presidential race, and most of them vote the same way in the House race regardless of how much money is spent. Even if the candidate is an unknown with few resources, a high percentage of a party's supporters usually will support that candidate.

    The high number of safe seats may not be surprising. Indeed, it is widely acknowledged by political observers who often calculate the number of "swing" House districts in double digits. Yet few take the next step of simply calling winners in the great majority of elections over a year before an election. But barring dramatic national events -- such as a depression, war or widespread scandal -- the Center for Voting and Democracy is very confident in its predictions without knowing anything about the campaign financing of next year's elections .

    If we can safely predict that 317 particular races will be won by comfortable margins of 10%, obviously so can others. In August 1996, Knight-Ridder Newspapers issued a report with much ado about how one candidate had a fundraising edge of at least three-to-one in 317 House races and that these candidates were likely to win. Aside from the coincidence of the number 317, the point is that many major campaign contributors specifically seek to give to -- one might say "invest in" -- candidates they believe will win, regardless of party.

    It should be no surprise when nearly all of the candidates we are now predicting to win easily have big spending advantages next year. Some might ask: which comes first, the chicken or the egg? In this case, there really is no doubt. If money decided many general elections, there simply is no way that there would be such a close match between the vote for president and the vote for House candidates. Few would suggest that money spent in House races has much impact on voters in the presidential race. The question, then, is, why do Republicans hold a 135-16 edge in districts where Bill Clinton received less than 45% of the vote (e.g., at least 5% below his national average) and why do Democrats hold a 118-8 edge in districts where Clinton received over 53% (e.g., at least 5% above his national average)? If many voters are swayed by television ads and campaign spending in House races, then surely it shouldn't take much to sway 6% of voters and make these districts lean toward the other party. But it rarely happens, and it is getting rarer as the fact that overall control of the House is now in doubt makes voters all the more likely to vote for the party over the person.

    There are other, more numbingly numerical ways to show that money has much less impact on voters' selections than commentators often attribute to it. Monopoly Politics goes to some pains to provide some of these numbers, with in-depth analyses of open seat races in 1996 and the vote in House races in California. Another simple method is to look at the widely differing vote for president in some adjoining congressional districts with the same media markets. For example, Spencer Bachus in Alabama represents a district next to Earl Hilliard. Bill Clinton won a meager 28% of the vote in Bachus' district and a robust 73% of the vote in Hilliard's district. The list goes on and on, with good examples in nearly all states.

    Although this report challenges the conventional wisdom that money buys general elections, it is not a diatribe against the need for campaign finance reform. Indeed, which is worse: contributors giving money to candidates in a competitive race or contributors blatantly buying access by giving to candidates they know will win? The Center looked at electoral results of the 25 candidates who received the highest percentage of contributions from outside their district: nearly every candidate was an incumbent representing a very safe district.

    The report also suggests that campaign finance reformers would do well to focus much more attention on the role of campaign financing in party primaries, where voters lack the guide of partisan labels and -- increasingly -- the guide of respected constituency organizations that mobilize voters. As a first step, parties could quit blaming each other for the failure of general election campaign finance reform and immediately pass internal rules to reform campaign financing in their primaries for open seat elections.

    But this report is also not just another rant against nasty incumbents. As Rick Hertzberg remarks in the quote that led off this report, "We are in the habit of jumping to moralistic conclusions about our political and public problems. We automatically blame everything on bad people -- bad politicians, bad media moguls, bad voters. Because we take it for granted that our political institutions are perfect, we have no choice but to blame the people who administer them when things go wrong. This is dangerous."

    No, instead this report is designed to show that we must put other reforms on the table to make our democracy worthy of the name. If the composition of the voters in a district is the most important factor in determining who wins and loses, it should be obvious that "winner-take-all," plurality elections in single-seat districts put profound limitations on American democracy. First, they give too much power to those who draw the district lines in the decennial fiasco of democracy we call redistricting: redistricting is quite simply a process in which legislators choose their constituents before their constituents choose them. Incumbents and party officials quite understandably guard their own self-interest in drawing district lines, using increasingly powerful computers to crunch expanding sources of information about voting patterns in order to surgically place voters in districts designed to ensure one-party dominance of most districts. Monopoly Politics has an in-depth analysis of redistricting reforms by the Center's Walter Hearne.

    Even with redistricting reforms, single-seat districts drastically reduce voter choice. As this report demonstrates, most districts are simply one-party affairs. These districts theoretically could have competitive primaries to give some choice to voters in the district majority, but incumbents rarely face meaningful primary challenges in the United States; the Center last year found that more than half of incumbents faced no primary opposition of any sort in each House election in the years of our study, 1982-1994.

    But even competitive districts limit voters to two choices, promoting "zero-sum" politics in which it is entirely logical for candidates to attack opponents and muddy their own policy positions -- a politics that today extends more and more to governance. As in the case of redistricting, campaign expertise and technology have outstripped our current system. Focus groups and polling contribute to ever-more precise targeting of campaign messages to a few swing voters, leaving the majority of potential voters who are firmly committed to one party feeling ignored and frustrated -- yet with little way to protest except to stay home.

    Nearly every other major, full-fledged democracy uses forms of proportional representation to elect its main legislature. The main hold-outs are France and the United Kingdom and their former colonies, but even that is changing: Britain will join France in using a proportional system for 1999 elections to the European parliament, and Tony Blair and the new Labour government have pledged to hold a national referendum on electing the House of Commons by proportional representation.

    Proportional representation -- or "full representation," as the Center sometimes calls it -- describes systems of voting in which voters are very likely to directly elect a representative of their choice. The term "proportional" comes from the principle that any group of voters -- as defined by their cohesive voting behavior -- should be able to elect representatives in proportion to their support in the electorate. Proportional representation comes in many forms; at the very least, we believe that American legislative elections should use a proportional system in three-seat districts, which would ensure healthy two-party competition and representation for nearly every American voter.

Here are some additional findings of note in Monopoly Politics:

A new political geography:
As increasingly noticed, there has been a great turnaround between traditionally Republican New England and the traditionally Democratic south. Our report indicates that several Republicans in the northeast -- particularly in New Jersey and New York -- could face serious competition in districts that in the past have been one-sided. Just as the conservative nature of many districts in the South finally caught up with Democrats in the 1990s, Republicans will need to be on the alert in the northeast. The open seat election in Staten Island to fill Susan Molinari's seat could be a bellwether for the political direction of several districts in the northeast that currently are held by Republicans.
 
Race and politics, continued:
Much was made of the fact that black incumbents Cynthia McKinney, Sanford Bishop and Corrine Brown won re-election last year in new, white-majority districts. But only Bishop represents a district that should be competitive for a Republican in a general election, and we have listed Bishop as a vulnerable incumbent in 1998. More problematic for improved racial relations is the disappearance of the southern white Democrat. Georgia now has a delegation of white Republicans and black Democrats. South Carolina, Alabama, Mississippi and Louisiana are each only one white incumbent away from joining Georgia in this kind of polarized representation.
 
Monopolies in California, Virginia, Oklahoma and Massachusetts:
Among many examples, California and Virginia are particularly instructive for the existence of political monopolies. In California, entrenched incumbents had a very good year in 1996. All "untouchables" repeated their landslide wins. Furthermore, nearly all represent districts that strongly match their party: of the 35 representatives predicted to win landslides in 1998, 32 represent districts where Clinton received at least 8% higher (57%) or lower (41%) than his average. Incumbent performance in these districts was very consistent in 1992-1996 despite national changes in House elections.

In Virginia, meanwhile, the closest of eleven races was won by 24% in 1996. To put these victory margins into perspective, no incumbent lost in 1996 who had won by more than 20% in 1994. Oklahoma and Massachusetts provide another kind of monopoly: six Republicans represent Oklahoma despite several Democratic candidates winning over 40%, while ten Democrats represent Massachusetts despite four Republicans winning over 40%.

Controlling the House in the future:
Many expect the Republicans to gain seats in 1998 based on the history of poor performance of the president's party in the mid-term elections of a second term. More fundamentally, Republicans have a large edge in the number of districts where Bill Clinton ran behind his national average. There are more very liberal seats than very conservative ones (99 to 82), but overall Clinton ran ahead of his national average in 196 seats and behind it in 239.

There may be a built-in conservative majority in the House, but it need not be a Republican majority. Some of these districts certainly are very competitive and the parties actually have an equal number of vulnerable seats in our calculation. But unless more voters start splitting their ticket again or the Democrats can move into traditional Republican strongholds in New York and New Jersey, the Democrats probably are at a disadvantage -- although some have suggested that the single-member district system helps Democrats by allowing more Democrats to win "cheap seats" with relatively few votes.

Regardless, the Democrats certainly do better in districts where Clinton ran ahead of his national average. But of Republicans elected in November 1996, only one represents such a district. Only three Republicans from the big class of 1994 represent districts where Clinton won at least 51% (2% above his national average).

Debunking myths:
 
Shifts in the electorate and our methodology:
The Center is not making a case for presidential coattails. Rather, we use the presidential results as a snapshot of the partisan nature of a district -- a snapshot that might get a little blurry during a decade, but generally stays remarkably clear. Particularly in states with stable populations, a district that gave Clinton his national average of 43% in 1992 is quite likely to have given him close to his 49% national average in 1996; that would mean the relative Democratic performance in the district did not change even as Clinton's percentage rose 6%. Calculations about which party is likely to represent a district are based on this relative performance, not its absolute performance.

Our chart "Districts on the Move" details the biggest shifts in the electorate. Democrats made their biggest gains in Massachusetts -- and New England, generally -- and in districts with large Latino populations. For Republicans, the biggest gains were in tobacco country, and the south generally. Clinton's relative performance dropped in all but one of the 39 House districts in North Carolina, South Carolina, Tennessee, Alabama and Mississippi.

    We hope that you find the report useful in your analysis of the 1998 House elections and, more generally, the movement for political reform in the United States.

 


back to Monopoly PoliticsTable of Contents