By Steven F. Huefner
In addition to this Weekly Comment, which urges congressional reform of Electoral College timetable, a separate new posting extensively analyzes the complex relationship between the current Electoral College rules, as enacted by Congress, and Ohio’s recount procedures.
Four weeks after the 2004 presidential election, Ohio is still several days away from certifying its official results. Secretary of State Kenneth Blackwell has directed county election boards to complete their official post-election canvass by December 1, in conformance with state law, and apparently has indicated that the final statewide certification will occur no sooner than December 3, and perhaps not until December 6. As the November 17 Weekly Comment observed, this timetable for election certification has the potential, in the event of a close race, to create irredeemable conflicts with the federal statutory processes for casting and counting presidential electoral votes. This Comment suggests that an adequate remedy to this conflict requires extending the federal timetable, in addition to speeding up the state timetable.
If the results of a presidential election were close enough to require a statewide recount or to give rise to litigation that otherwise could affect the outcome, then it is highly unlikely that under current law Ohio could determine its winner in time to take advantage of the federal “safe harbor” deadline, which this year is December 7. This is the date after which a state’s selection of electors is no longer presumptively conclusive when Congress meets to count electoral votes. Indeed, Ohio might even have difficulty resolving a contested presidential election before the date when the electors meet to cast their votes, which this year is December 13.
In response to this potential problem, Ohio could adjust its post-election processes to produce a certified official result sooner, as the November 17 Comment suggested. However, even if Ohio completed its election canvassing processes and certified an official result by mid-November, the existing federal timetable likely would remain unrealistic in the event of a dispute or contest about the statewide winner of the presidential election. The fact that in the 2000 election the U.S. Supreme Court terminated Florida’s effort to complete a fair recount of its presidential ballots precisely because the federal “safe harbor” deadline had already arrived, even though Florida had been conducting manual recounts since a few days after the election, only provides additional evidence that some adjustment in the federal timetable might be in order.
Federal law establishes the date when states choose their presidential electors (the Tuesday after the first Monday in November, as provided in 3 U.S.C. ‘ 1), the date for conclusively resolving any controversy about the choice of those electors (the “safe harbor” date provided in 3 U.S.C. ‘ 5, defined as six days before the date set for the electors to cast their votes), the date when those electors cast their votes for President (the Monday after the second Wednesday in December, as provided in 3 U.S.C. ‘ 7), the date when Congress counts these electoral votes (January 6, as provided in 3 U.S.C. ‘ 15), and the date when the newly elected President takes office (January 20, as provided in U.S. Const., amend. XX). Because this last date is established as a matter of U.S. constitutional law, it would be the most difficult to change. The remaining dates, however, are set entirely as a matter of federal statutory law, and therefore could be altered simply by congressional action. Moreover, the intervals were set in an era when modes of communication demanded substantial time between key steps in the electoral process. In today’s information age, the need for most of these intervals is hopelessly antiquated.
The exception, of course, is the interval between the popular election and the elector’s casting of their votes. Under existing federal law, the time between election day and the safe harbor for choosing a state’s electors is five weeks. Both the 2000 and 2004 elections suggest that this interval may not be sufficient for the satisfactory resolution of a close election. Although Ohio and other states certainly could expedite their post-election canvassing and certification processes, and should do so as much as reasonably possible, the value of a compressed timetable must be balanced against the needs for fairness, accuracy, and transparency. Post-election processing of absentee ballots, scrutiny of provisional ballots, and recounting of paper ballots all take time to be done right, even without judicial involvement. Furthermore, once candidates or voters turn to courts for assistance in resolving legal issues, still more time will often be necessary, if the election processes are to remain fair, accurate, and transparent. Indeed, the frantic briefing, argument, and decision schedules under which the supreme courts of Florida and the United States were compelled to handle the 2000 election almost certainly compromised their treatment of the issues before them.
How, then, to provide sufficient time for resolving election controversies, while recognizing that election outcomes also must be settled quickly? One possibility would be to push back several days the date for counting the electoral votes in Congress, say to January 10, and in turn to set January 3 as the date for the electors to cast their votes in their state capitals, and December 31 as the safe harbor date. (In fact, until 1933, when the twentieth amendment moved up the date for presidential succession to January 20, federal law had called for Congress to count electoral votes on the second Wednesday in February.) This adjustment would add anywhere from 17 to 23 days, depending on the year, to the time available to resolve an election controversy before the safe harbor deadline. Of course, in most elections, the result would still be known by the morning after the popular election. But in those few instances that may require more time, the additional three weeks under this revised timetable could be invaluable.
What would be the costs of this delay? In those rare cases in which the election is not publicly settled the morning after the election, pushing the safe harbor back to December 31 obviously has the potential to reduce the transition time for a new administration. On the other hand, in precisely these rare cases, it is preferable that the election’s ultimate outcome be one that the public trusts, rather than that the incoming administration know its status three weeks sooner. In the meantime, both principal candidates presumably would conduct a number of transition activities. These activities, although undoubtedly compromised, could partially ameliorate the delay in knowing the ultimate outcome. Some modifications in the Presidential Transition Act, which currently provides federal assistance to “the apparent successful candidate” as early as the day after the popular election, might also be in order, to enable both candidates to draw upon government support in planning for a potential new administration. In any event, this transition inconvenience seems preferable to forcing state and federal courts and state elections officials to rush their judgment about the election outcome.
Of course, a variety of other election reforms, including electronic voting, early voting, and a nationwide voter database, could increase election day efficiency and accuracy, and thereby reduce the likelihood of a disputed election and alleviate much of the need for an expanded post-election timetable. But no reforms will eliminate the possibility that another presidential election will become a pitched battle in court. It therefore makes sense to prepare for this possibility by maximizing the time available to determine any such real controversy, under both state and federal law.
While a number of expansions of the federal timetable are imaginable (including even moving Election Day forward from November to, say, Labor Day), more drastic reform seems less politically feasible. Accordingly, Congress should adopt some version of the timetable suggested here to replace the antiquated Electoral College timetable now in use.