Electoral College Math 301

Introduction

In "Electoral College Math 201," I discussed the possibility of expanding the Electoral College on the basis of the Constitution’s original formula for allocating Representatives from each state: one per 30,000 people. This can be called the 30K Plan.

In 48 of the states, the electoral vote is "winner take all." California is one of these. Thus, under the 30K Plan, California would have cast all 1,131 of its electoral votes for Kerry in 2004 (based on the 2000 census), and all 994 votes for Gore in 2000 (based on the 1990 census). This points up the other part of the Electoral College problem. Article II, Section 1, Clause 2 of the Constitution states: "Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors...." Unless each state enacted legislation to appoint its electors proportionally according to the popular vote, the Electoral College would still skew elections. The 30K Plan isn't a complete solution to the problem, but if it is constitutional, it is probably the easiest solution to implement, in that it only requires a federal statute. Moreover, it may give impetus to state-level initiatives to allocate electors on the basis of the proportion of the popular vote rather than on the basis of "winner take all." This consideration opens the door for a credible two-front strategy, one at the federal level and one at the state level.

Analysis of the 2000 and 2004 Presidential Elections

Table 1 shows what would have been the results of the 2004 presidential election using the 30K Plan under both "winner take all" and proportional allocation rules.

Table 1: The 2004 Presidential Election

Winner Take All		Proportional Allocation
Bush	4867	Bush	4829
Kerry	4615	Kerry	4575
		Nader	36
		Badnarik	28
		Cobb	6
		Peroutka	6
		Peltier	2
Needed to Win: 4742

George W. Bush would have won both ways, but that's not terribly surprising since he won the both the popular vote and the substantially "winner take all" electoral vote under the current formula of 538 electors. The 2000 presidential election results using the 30K Plan are shown in Table 2.

Table 2: The 2000 Presidential Election

Winner Take All		Proportional Allocation
Bush	4438	Gore	4078
Gore	3958	Bush	4014
		Nader	232
		Buchanan	35
		Browne	26
		Phillips	6
		Hagelin	4
		Smith	1
Needed to Win: 4199

Bush would still have won in the "winner take all" scenario. In a proportionally allocated Electoral College, Al Gore would have won a plurality, which accords with his popular vote plurality, but not an outright majority. Thus the election may have had to have been decided in the House of Representatives pursuant to the 12th Amendment, where the choice would be limited to the top three candidates (Gore, Bush, and Nader).

Had the election gone to the House in 2000, the vote would have been "taken by states, the representation from each state having one vote." Thus, there would be a total of 50 votes, with 26 votes needed to win. Since the Republican Party held majorities in most of the state delegations, Bush would have won.

However, would the 2000 presidential election necessarily have gone to the House of Representatives? There are two points to consider. To begin with, 24 states do not legally bind their electors to cast their vote for a specific candidate. Secondly:

The preponderance of legal opinion seems to be that statutes binding electors, or pledges that they may give, are unenforceable. If an elector chooses to incur party and community wrath by violating his trust and voting for someone other than his party's candidate, it is doubtful if there is any practical remedy (Longley and Pierce 1999, 114).

The 24 states that have no sanction against "faithless" electors would have had a total of 100 electors pledged to Ralph Nader. Had these decided to defect to Gore rather than throw the election into the House and thereby hand the presidency to Bush, it would have given Gore a total of 4,178 electoral votes, 21 votes short of victory. But since no elector has ever been prosecuted for failing to vote as pledged, it is entirely plausible that at least 21 of the remaining 132 Nader electors would have defected to Gore. Furthermore, it is also entirely plausible that in the weeks between the popular vote and the convening of the Electoral College, Nader would have conceded the election and either released his electors to vote their conscience or thrown his support to Gore outright. Given all of these considerations, it is very likely that Gore would have ultimately won the electoral vote.

Synergy of the 30K Plan and Proportional Allocation

An interesting question is, if the 30K Plan were enacted at the federal level, would there be more political will in the state legislatures to shift from "winner take all" to proportional allocation of the electors?

In working the calculations for the 2000 and 2004 presidential elections, rounding errors occurred for 46 of 102 cases (50 states plus DC times two elections). Where an electoral vote needed to be added to fill out a state's allocation, I gave it to the candidate who had the largest residual value between 0 and 0.5. Where I had to subtract an electoral vote, I took it from the candidate who had the least residual value between 0.5 and 1.

In the "winner take all" system, it is statistically unlikely that any one person can tip the electoral vote from one candidate to the other. Again, let’s take California as an example. In California, there was one chance out of 11.5 million votes cast last November that a single vote might have tipped the electoral outcome from one candidate to another. With proportional allocation under the current Electoral College formula (in which California has 55 electors), there would have been 55 chances out of 11.5 million, or one out of 209 thousand. With proportional allocation and the 30K formula, there would have been 1,131 chances out of 11.5 million, or one out of 10 thousand. Those are better odds than most lottery tickets! That is voter empowerment!

Implementation Strategy

It would seem that there is considerable political synergy between the 30K Plan and the proportional allocation of electors. However, to implement both rules would require a campaign on two fronts—always a tricky thing. The 30K Plan would have to be passed at the federal level (and withstand a possible Supreme Court challenge). The proportional allocation plan would have to be sold on a state-by-state basis. The latter is a more difficult collective action problem, but if the 30K Plan were in place, it would become obvious how much power the proportional allocation of electors would put in the hands of the voters. What better way to incentivize voter turnout? Any number of civic groups would pressure state legislatures to act.

Even though actions at both the federal and state levels would be required, this two-part process would still be better than the constitutional amendment route to abolish the Electoral College. A two-thirds majority of each house of Congress is required, then three-fourths of the state legislatures must approve the measure. Since all be one of the state legislatures are bicameral, the proposal must pass in about state 76 chambers. If one of the battles in Congress is lost, or if about 24 of the battles at the state level are lost, the entire campaign is lost. Also, usually there is a time limit written into a proposed amendment; in the case of the Equal Rights Amendment, the campaign was lost when the proposal expired after seven years.

In contrast, the 30K Plan requires a simple majority in each house of Congress, and in this case winning that first battle counts for something that cannot be taken away by a subsequent failure at the state level. The next consideration is that each state could implement the proportional allocation rule for itself, and there would be no time limit. Furthermore, a bandwagon effect would occur if only a few states implemented the rule and voter turnout climbed substantially, for seeing this, civic groups in other states would clamor for the rule's passage.

In sum, the campaign can be waged by scoring substantive, incremental victories along the way, whereas with a constitutional amendment, it’s an "all or nothing" proposition.

Reference

Longley, Lawrence, and Neal Peirce. 1999. The Electoral College Primer 2000. Yale University Press.