The smoldering issue of electoral college reform has flared anew in the wake of the 2000 electoral college misfire. Despite opinion polls demonstrating that two thirds of Americans support direct election of the president, the reform receiving the most consideration is the district system, wherein one elector is awarded to the popular vote winner in each congressional district and the two "Senate" electors are awarded to the popular vote winner statewide. In 2001, twenty-one states were considering joining Maine and Nebraska in using the district system to allocate their presidential electors (Drage 2001).
Historically, the main argument in favor of the district system has always been its political feasibility rather than its inherent desirability. Like the direct election alternative, it would end the distortions of the winner-take-all aspects of the electoral college wherein the candidate with the plurality of votes receives all of the state's electoral college votes. However, unlike direct elections, the district system preserves the power of small states and could be enacted by state or federal law rather than constitutional amendment. Moreover, eleven states used the district system during the early 1800s, and more recently Maine and Nebraska adopted the district system in 1972 and 1991, respectively, making the district system the only proposed electoral college reform that has actually been used in the United States (Peirce 1968). However, simply because the district system does not require a constitutional amendment does not mean that it is superior to other reform alternatives.
To date, the overwhelming majority of research on electoral college reform has focused on the relative merits of the present system versus direct election (Longley and Peirce 1996; Best 1975, 1996). Most assessments of the district system have focused on how it would translate votes into electoral outcomes instead of how it would change the electoral incentives for presidential campaigns (Bensen 2000). By awarding electoral college votes on the basis of district rather than state-level votes, the district system would encourage presidential campaigns to focus their efforts on battleground districts instead of battleground states. Examining the consequences of this change in campaign strategy is essential for understanding the merits of the district system as an alternative to the existing electoral college, and valuable in better understanding how the electoral college shapes all aspects of our political system--presidential campaigns, political participation, electoral coalitions, congressional elections, and the two-party system.
The merits of the district system lie in how shifting the focus of presidential campaigns to battleground districts would affect our electoral and governing processes. My view is that the optimal presidential electoral system should encourage presidential campaigns to build broad electoral coalitions, stimulate citizen interest and turnout in presidential elections, produce a president who can govern, strengthen the two-party system, discourage electoral fraud, and be relatively neutral. To assess how well the district system meets these criteria, I examine seven questions. First, does the district system represent an improvement over the present winner-take-all system in its efficiency and accuracy in translating the popular will in presidential elections? Second, would the resulting changes in presidential campaigns under the district system increase or diminish citizen interest in elections? Third, what are the ramifications of the district system for a president's ability to bring in fellow partisans to Congress and his or her ability to govern? Fourth, which groups or parties are favored under the district system? Fifth, does the district system enhance or diminish the prospects of third parties? Sixth, does it discourage electoral mischief and voter fraud? Seventh, what are the prospects that any of the twenty-one states considering the district system after the 2000 election will adopt it? …