STATEMENT BEFORE THE CONSTITUTION SUBCOMMITTEE OF THE HOUSE JUDICIARY COMMITTEE REGARDING H.R. 1173, THE "STATES' CHOICE OF VOTING SYSTEMS ACT"
SEPTEMBER 23, 1999
Andrew E. Busch
Associate Professor
Department of Political Science
University of Denver
Denver, Colorado, 80208
303-871-2133
I. Historical background
The use of single-member districts for legislative elections can be traced back to the development of the English Parliament and the colonial assemblies of pre-Revolutionary America. Consequently, from the beginning of the United States, a significant majority of states utilized single-member districts for the election of members of Congress. However, the Constitution itself does not specify whether districts are to be used or whether districts may have only one representative. Methods other than single-member districts for the election of U.S. House members were used by several states in the first half-century after adoption of the federal Constitution of 1787. Some states selected all of their Representatives at large (what was sometimes called the "general ticket" method), while others used multi-member districts or a combination of methods. For instance, a few states used single member districts for most of their Representatives but also maintained one or more multi-member districts; others added at-large Representatives to their single-member districts when they gained one or more congressional seats after a decennial reapportionment but did not wish to redistrict. While federal law began requiring single member districts for U.S. House of Representatives in the Apportionment Act of 1842, other lower jurisdictions such as some state legislatures and municipal governments continued using (and still use) alternative methods. As late as the 1840-41 election cycle, approximately one-third of states used a manner of representation other than purely single-member districts (six used the statewide general ticket method and three mixed single and multi-member districts).
II. Effects of single-member House districts
It is widely held among political scientists that single-member congressional districts have had important effects on the development of American politics. First, these rules (combined with the winner-take-all feature used by most states in the Electoral College) have tended to support a stable two-party system encouraging the major parties to build the broadest possible coalitions. Since each district has only one Representative, and that Representative is generally elected by plurality vote, only the strongest third parties have any chance of winning congressional seats. Evidence for this effect can be seen abroad. Great Britain, which also uses single-member plurality districts, has typically maintained a stable two-party (or at times two-and-a-half party) system; countries using some form of proportional representation or other multi-member district methods are more prone, like Italy or Israel, to political instability and/or a situation in which groups having the support of only a small proportion of the population nevertheless hold the decisive political balance. Additionally, the current system encourages a close tie between a U.S. House member and his or her geographic constituency, and makes it much more likely that varying regions of a state will be adequately represented.
III. Potential effects of a move away from single-member districts for U.S. House elections
There are two basic types of potential effects which must be considered. The first is electoral: How would a move away from single-member districts alter the electoral structure of American politics? The second is related to governing: How would such a move change the behavior of Congress?
The answer to the first question depends entirely on what alternative form of election is selected by states. There are two crucial decisions states would have to make-- a choice between statewide election of all representatives or the use of smaller districts, and a choice of specific methods of voting and allocation of representation.
States could revert to the statewide general ticket system widely used before the 1840s. In such a system, the state would elect all of its representatives on a statewide basis, with voters casting votes for as many candidates as there are seats and the top vote-getters being elected. The consequence of such a system would almost certainly be to reduce the diversity of state delegations. The statewide partisan, ideological, and/or regional majority could conceivably control the delegation without regard for the local variations that can currently obtain representation. In 1840-41, all six of the states using a general ticket method had one-party control of the entire delegation. Of course, this effect could be muted in states where there is a very close partisan balance or a very large number of unaffiliated voters, which might elect split delegations. It could also be muted if states used the same basic voting method but in multi-member districts. Each district would have fairly homogeneous representation--in 1840-41, three states used a total of eight multi-member districts, and all eight had one-party control-- but multiple districts could produce greater variety statewide than a general ticket system. In any event, however, it is difficult to imagine that either at-large general ticket congressional elections or multi-member districts using pre-1840s rules would not reduce representational variety in comparison to the present.
Another option states could use either statewide at large or in multi-member districts would be some form of approval voting or cumulative voting. With approval voting, voters would indicate from the list of candidates which candidates they find acceptable, whether one or many. With cumulative voting, voters would have as many votes are there are seats, but could apportion them in whatever way they want (not necessarily one per candidate), including providing a single candidate with all of their votes. These systems are often touted by analysts hoping to increase minority representation, on the grounds that a committed and cohesive minority (whether political or ethnic) could win some seats by concentrating all of their votes on one candidate or a small number of candidates. If this argument is true, congressional delegation variety might be increased, though at the expense of the principle of majority rule and one-person-one-vote, and also possibly at the cost of the increased ethnic division of American politics. Recent events in the Balkans should remind us that national unity is not obtained automatically, and that there are good reasons to hesitate to build our electoral system around the deliberate cultivation of ethnic identity or separatism.
The third major option, possible in statewide elections in moderate to large states or in multi-member districts with a large number of representatives per district, would be to utilize proportional representation, in which party support would be translated proportionally into seats won. Like approval voting or cumulative voting, proportional representation would give an opportunity for political forces with the support of a minority of voters to nevertheless win seats. It would also have the effect of reorienting politics away from individual candidates and toward party conformity, since voters would be choosing between parties rather than between candidates. This reform is supported predominantly by third-party activists and political movements hoping to form third parties.
One electoral consequence that will almost surely accompany changes in this direction (with the possible exception of proportional representation) is that House elections in states that adopt any of these reforms will become more expensive. Each House candidate will have many more constituents to address, will probably have more media markets to reach, and will have to work to differentiate himself or herself from not one but many competitors. The second question--the effects of a move away from single member districts on the way Congress governs-- likewise largely depends on the precise alternatives instituted in states. Widespread use of proportional representation, and possibly of cumulative/approval voting, could easily lead to a fragmentation of Congress and increased leverage held by political groups outside of the mainstream. Especially given the narrow margin separating the parties in the House, it is not inconceivable that a small number of Representatives supported by a small proportion of the electorate could come to hold the balance, either demanding extreme concessions or simply refusing to allow the work of the House to go forward.
No matter what alternative electoral method is selected, a move away from single member districts will almost surely have another broad effect, as well, which is that House members will become less tied to their geographical constituency and less concerned with local affairs. Representing either whole states or much larger districts, and sharing constituents with other (perhaps several other) Representatives, those members of the House will probably deemphasize constituency service and place greater emphasis on legislating and on national issues. While the precise consequences of such a shift may be difficult to foretell, it can be said without hesitation that such a change would fundamentally transform the nature of representation.
IV. The likelihood of a move away from single member districts at the state level
The materialization of any effects, either electoral or governing, will depend first and foremost on how many states, with how many Representatives, actually take advantage of the opportunity to move away from single member districts. If states were beginning from scratch, it is likely that a variety of systems would be adopted, as indeed occurred in 1788-89. Under current conditions, however, I believe it is unlikely that many states would accept the disruption of established political patterns that would be produced by this type of electoral reform. While redistricting takes place every ten years, many districts have retained their essential shape through several redistricting cycles. Incumbent members of the House have established political bases within their districts, have developed close working relationships with state and local officeholders, and have acquired knowledge, experience, and seniority that are extremely valuable to their constituents and their states. It is unlikely that most of the state officeholders who would have to authorize a change would consider such a change beneficial. The most likely exception would be in states where one party controls both the executive and legislative branches of state government and maintains a wide lead among voters but does not have full control of the congressional delegation. In cases like these, state officeholders might be tempted to adopt a statewide general ticket system in an attempt to overwhelm local pockets of opposition-party strength. However, such maneuvering could be politically risky, and most states do not have the unified partisan control of state government that such an effort would require. There will be little pressure for proportional representation or approval/ cumulative voting in states where the parties are already roughly evenly represented, and little likelihood of its adoption in states where they are not. If proponents are hoping to increase the diversity of representation through this legislation, they are likely to be disappointed, and might want to redirect their efforts toward enlarging the size of the House, a reform which could achieve much the same purpose with greater effectiveness and at less potential cost to the political system.
V. Summary and Conclusion
In my view, the greatest strength of this proposal is that it pays a welcome respect to the principle of federalism. That principle, which is an essential component of the American constitutional structure, has been ignored and even undermined far too often in the last 70 years. Congress is to be commended for taking measures in recent years to restore some of the lost vitality of federalism, and this proposal could be considered part of that trend.
However, the actual benefits to the federal system of this proposal are modest and symbolic. It would entail no devolution of governmental functions and would in no way alter the existing balance of substantive power and authority between the states and federal government. There are, moreover, many reasons to be cautious. It is probable that few if any states would actually reverse 150 years of tradition and deliberately throw their sitting House members into a cauldron of uncertainty. Should a significant number of states defy this prediction over time, the most likely outcome is a homogenization of state delegations as majority parties press their advantage statewide. The alternative outcome, should proportional representation or approval/cumulative voting become widespread, could be to lend greater instability to American politics and to make the House more unmanageable. These devices could well be a useful check on the majority in a parliamentary system, in which power is otherwise largely unchecked, but it is far from clear that the American system requires more checks or that the legislative process should be made more arduous. And any conceivable reform would change, perhaps drastically, the character of representation in Congress.
Above all, electoral reforms of this magnitude almost always produce unintended (and undesired) consequences. The presidential nominating reforms after 1968 have ultimately produced a system that even the most ardent reformers seldom defend; the campaign finance reforms of 1974 have largely collapsed, but the $1,000 individual contribution limit survives, forcing candidates for Congress and the presidency to devote disproportionate energy to fundraising rather than meeting people or discussing ideas. It is impossible to predict with certainty what the effects of this legislation would be, but that in itself is a reason for caution when dealing with something as fundamental as the way Americans elect the House. It is also reason to avoid the natural temptation to seek short-term partisan advantage. It is not at all clear that it is possible to accurately calculate the partisan consequences to such a proposal in the short-term, let alone the long-term; and such a calculation would in any event provide a poor substitute for consideration of the enduring national interest.