Under the whole-number proportional method for awarding electoral votes, states would enact laws dividing their electoral votes in proportion to each candidate’s share of the popular vote in their state, in whole-number increments. Note that this whole-number division is what distinguishes this method from the very different fractional proportional (Lodge-Gossett) method.
Because it would not abolish the position of presidential elector or the Electoral College and does not require the creation of fractional electoral votes, the whole-number proportional method can be enacted as state legislation on a state-by-state basis.
The whole-number proportional method may initially sound attractive. However, it would operate in a number of unexpected and undesirable ways.
● The whole-number proportional method would not accurately reflect the nationwide popular vote—even if enacted by every state. In fact, the national popular vote winner would not have become President in three of the eight presidential elections between 1992 and 2020 under this method. In two of these eight presidential elections (2000 and 2016), the winner of the national popular vote would not have won the most electoral votes. In four of these eight elections (1992, 1996, 2000, and 2016), the choice of President would have been thrown into the U.S. House. Based on the composition of the House at the time, the national popular vote winner would not have been chosen by the House in three of those four cases (1996, 2000, and 2016).
● In practice, the whole-number proportional method would be a “winner-take-one” system in almost every state—with perhaps two electoral votes being in play in Texas, and three in California.
● Although it might appear that the whole-number proportional method would give candidates a reason to campaign in all 50 states, it would not do so. Candidates would only campaign in states where their level of support was a few percentage points away from a breakpoint that might possibly gain or lose them an electoral vote. In practice, only about 29 electoral votes from about 26 states would typically be in play. Candidates would not have any reason to campaign in the 24 remaining states, because their level of support would be too far away from a breakpoint that would change an electoral vote. That is, almost half of the states would be politically irrelevant spectator states.
● The whole-number proportional method would not make every vote equal. There are five sources of significant inequality built into this method, including a 3.81-to-1 inequality because of senatorial electors; a 1.72-to-1 inequality because of imprecision in apportioning U.S. House seats (and hence electoral votes); a 1.67-to-1 inequality in favor of voters in low-turnout states; a 1.39-to-1 inequality because of intra-decade population changes; and a 50.2-to-1 inequality, because one electoral vote could be won with a few thousand popular votes in a low-population state, while requiring tens of thousands of popular votes in a bigger state.
● Minor-party and independent candidates would almost always be zeroed-out in small- and medium-sized states. The reason is that their level of support would be far less than the fraction of the state’s popular vote required to win one electoral vote in such states. One electoral vote would correspond to 33% of the popular vote in a state with three electoral votes. One electoral vote would correspond to 14% of the popular vote of a median-sized state (that is, a state with seven electoral votes).
● The whole-number proportional method would transfer the choice of President from the people to Congress in about half of all elections. The reason is that this method would be adopted without amending the U.S. Constitution, thereby leaving the U.S. House in a position to pick the President if no candidate were to receive an absolute majority of the electoral votes. If the whole-number proportional method had been used by all states, the U.S. House would have picked the President in four of the eight presidential elections between 1992 and 2020 (1992, 1996, 2000, and 2016).
● A state reduces its own influence if it divides its electoral votes while other states continue to use winner-take-all. The whole-number proportional method would penalize first movers and early adopters. Moreover, a piecemeal state-by-state adoption process would quickly become self-arresting, because each new adherent would increase the influence of the remaining winner-take-all states—thereby reducing their incentive to make the change.
● In November 2004, Colorado voters defeated an initiative petition to enact the whole-number proportional method.
See detailed discussion.