Mathematical Analysis of the Electoral College, it might be observed that the term "relative voting power" computed mathematically or otherwise is indeed a relative term. Under the present system it may be relative to whether the state is to be carried successfully by the voter's party. Under direct elections, while each vote is equal, the percentage value of a vote could relate to light or heavy voting. This in turn may relate to concentrations in presidential campaigns which would normally favor large states and densely populated areas. It might also be observed that the true and balanced constitutional values to be sought may not lie in attempting to assure an exactly equal value to the last mathematical degree of each and every vote for President and Vice President. On the contrary, we should consider broad and long term values. I feel that we should discard the "winner take all" unit-voting system and abolish the positions of electors. At the same time, we should maintain voting balance in keeping with the spirit of other balances provided in the Constitution and intended by our founding fathers in the provision in article II, section 1 of the Constitution that each state shall be entitled to a number of electoral votes "equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress." This principle of representative government was not changed by the twelfth amendment ratified in 1804, and as recently as 1961 it was affirmatively recognized and approved in the ratification of the twenty-third amendment, which gave the District of Columbia a voice, not to exceed that of the least populous state, in the election of the President and Vice President."

The general arguments for popular election of the President and Vice President could be met by providing that the electoral votes of a state, which are subject to change by the census, should not be cast as a unit, but on the basis of a direct percentage relation to the actual votes

4. All of the electoral votes of a state by custom or statute are cast for the candidate who carries the state irrespective of the margin of victory. The Supreme Court has refused to hear a petition challenging this practice. Delaware v. New York, 385 U.S. 895 (1966).

5. This proposal would repeal the second paragraph of article II, section 1 of the Constitution which reads "Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors. ..." It would repeal also all references to electors in the twelfth amendment.

6. However, the provision in article II, section 1, giving each state an electoral vote "equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress," would be retained.

7. The District of Columbia has been allowed 3 electoral votes, which as of 1964 was the same number as those of Alaska, Delaware, Nevada, Vermont, and Wyoming.

8. Article I, section 2 of the Constitution assures each state of at least one Representative in the House of Representatives and requires an enumeration of the House every ten years. Section 2 of the fourteenth amendment clarifies the apportionment-by-population language of article I. As the number of Representatives from a state changes, the number of electoral votes changes.

cast for the respective candidates. In this manner each voter would know that his vote would be counted in favor of the candidate of his choice regardless of whether or not his state was carried by that candidate or by the political party which he represented.9

I have supported proposals of this nature in both the House of Representatives and the Senate, including the Lodge-Gossett proposal which received the necessary two-thirds vote of the Senate in 1950.10 In both the 89th and 90th Congresses I have introduced joint resolutions which would amend the Constitution to this extent.11

Proportionate division of the electoral vote according to the popular vote in each state would require candidates to give attention to the issues of interest to citizens in all areas of the country. Likewise, it would broaden the eligibility of candidates who are not from the majority party of a large state,12 and would eliminate the neglect of smaller states that are inclined to favor one party consistently.

The "one man, one vote" theory expounded by the Supreme Court as a means of enforcing state reapportionment, including congressional districts, 13 has no true procedural applicability in the instant matter,

9. If a candidate for President received 60% of the votes cast in a state with 10 electoral votes, then by automatic certification that candidate would receive 6 electoral votes. The remaining 4 electoral votes would be cast for his opponent or opponents. This would obviate in the main the possibility of a President being elected with a smaller popular vote than the candidate who was defeated, which was the case with Presidents Adams, Hayes and Harrison.

10. 96 CONG. REC. 1278 (1950). See for a summary of S.J. Res. 2, 81st Cong., 2d Sess., 96 CONG. REC. APPENDIX 1112 (1950). For an explanation of this resolution, as reported by the Judiciary Committee of the House, see remarks of Mr. Gossett, in 96 CONG. REC. 10414-27 (1950). However, the resolution failed to pass because the House voted not to suspend the rules, and adopt the joint resolution. 96 CONG. REC. 10428 (1950).

11. S.J. Res. 138, 89th Cong., 2d Sess. (introduced February 24, 1966, for myself and Senators Dodd and Saltonstall); S.J. Res. 84, 90th Cong., 1st Sess. (introduced May 14, 1967, for myself and Senators Dodd and Ervin). These resolutions would abolish the positions of electors, provide for proportional casting of a state's electoral votes as based on the popular vote, and would abolish the one vote per state system of selection of a President by the House of Representatives, substituting in lieu thereof selection by a constitutional majority at a joint session of the House and the Senate in the event the election is referred to the Congress.

12. A strong but still a minority vote in any state would enter into political considerations regarding nominations. Without the electoral vote, however, direct elections might cause concentration on heavily populated areas to the neglect of small states. Cf. records of Virginia, New York and Ohio in producing Presidents. Virginia led the nation in electoral votes for many years, e.g., 22 electoral votes in 1800. Seven of its eight Presidents were elected prior to 1849 (several of whom of course were among our founding fathers). New York took the lead in 1830 with 40 electoral votes and Van Buren was nominated and elected in 1836. Three other Presidents have come from New York. Ohio went to 21 electoral votes in 1840 and since 1867 has produced seven Presidents.

13. This line of cases began with Baker v. Carr, 369 U.S. 186 (1962), and was subsequently explicated in Wesberry v. Sanders, 376 U.S. 1 (1964); Reynolds v. Sims, 377 U.S. 533 (1964); and Maryland Comm. for Fair Representation v. Tawes, 377 U.S. 656 (1964). See also Lucas v. Forty-fourth Gen. Assembly of Colo., 377 U.S. 713 (1964).

which deals fundamentally with changing the Constitution itself. For example, the Senate of the United States ipso facto contravenes that theory by its numerical membership,1 while at the same time remaining a bastion of protection for our balanced and well-tested form of government.

Proportionate division of electoral votes according to the popular vote in each state would improve and make more meaningful our system of presidential elections and at the same time would maintain a balance in government which we must preserve.

14. Compare the seventeenth amendment which requires popular election of two Senators from each state (irrespective of population) with the "one man, one vote" theory. Originally article I, section 3 of the Constitution provided that two Senators from each state shall be "chosen by the Legislature thereof."

COMMENT

NEAL R. PEIRCE†

THROUGH an imaginative use of modern computer techniques, John Banzhaf has given mathematical proof for a phenomenon which politicians have grasped intuitively since the early nineteenth century: the disproportionate weight of the votes cast by citizens of the large states under a unit-vote (or "winner take all") system of electoral voting.1 Reduced to laymen's language, the formula is as simple as this: in a close election, the decision of a few hundred or a very few thousand voters can switch the outcome in a small state like Wyoming, thereby throwing three electoral votes from one column to the other. Nationally, the impact of the shift is likely to be minimal. But a shift of exactly the same number of popular votes is often sufficient to alter the outcome in one of the larger states, such as New York with its forty-three electoral votes or California with forty. Very often, the switch of one of these large states may determine the outcome of the entire presidential election.

Indeed, history shows several elections in which the national outcome hinged on a single large state.

In the election of 1844, a shift of 2,555 votes in New York could have reversed the Electoral College outcome, making Henry Clay President instead of James K. Polk.

In the election of 1880, a shift of 10,517 votes in New York would have made Winfield S. Hancock President instead of James A. Garfield.

In the election of 1884, a shift of 575 votes in New York would have made James G. Blaine President instead of Grover Cleveland.

In the election of 1888, a shift of 7,189 votes in New York could have kept Grover Cleveland in the White House, instead of electing Benjamin Harrison. (Cleveland actually won a popular vote plurality but lost in the Electoral College.)

tPolitical Editor, Congressional Quarterly and author, THE PEOPLE'S PRESIDENT, THE ELECTORAL COLLEGE IN AMERICAN HISTORY AND THE DIRECT VOTE ALTERNATIVE (Publication date March, 1968, Simon & Schuster, New York).

1. In 1824, Martin Van Buren, then a Senator from New York, opposed substituting a district system for the unit-vote system because a division into districts would tend "to reduce greatly the present weight of the large states in the general scale" by "preventing them from bringing their consolidated strength to bear upon the Presidential question." J. DOUGHERTY, THE ELECTORAL SYSTEM OF THE UNITED STATES 333 (1906). In an 1834 Senate debate, Thomas Hart Benton of Missouri supported a district plan because it would "break the force of the great states in the elections." 46 NILES REGISTER 421, 422 (1834).

In the election of 1916, a shift of 1,983 votes in California could have made Charles Evans Hughes President instead of Woodrow Wilson - even though Wilson had a half-million more popular votes across the country.

By contrast, the only election in which a vote shift in a small state could have determined the outcome was in 1876, when a shift of 185 popular votes in South Carolina would have reversed the electoral vote count to make Samuel J. Tilden President instead of Rutherford B. Hayes.

In 1948 and 1960, the shift of only two large states in which the vote was close would have thrown the election into the House of Representatives for decision. A shift of three states in 1948, or five in 1960, would have elected the losing Republican contender in each case.

Thus history backs up what Mr. Banzhaf has demonstrated mathematically: in any close election, the citizens of the large states have a disproportionately greater chance to affect the outcome merely by the accident of their residence.

Earlier analysis in this field restricted itself to the power of states, as states, to influence the Electoral College outcome. Mr. Banzhaf's research takes the fruitful new step of examining the power to influence elections in terms of the individual voter, albeit in the context of his state voting base. This approach would certainly seem preferable, since, after all, it is people, not states, who have preferences in a presidential election. Since the Constitutional Convention, debate about electing the American President has become bogged down in a consideration of big-state versus small-state interests. Yet, in truth, the argument

2

2. Fearful of a direct-vote plan being debated in the Convention, Charles Pinckney of South Carolina warned that "[t]he most populous States by combining in favor of the same individual will be able to carry their points." Connecticut delegates expressed the same fears. 2 THE RECORDS OF THE FEDERAL CONSTITUTIONAL CONVENTION OF 1787 30 (M. Farrand ed. 1911). The final plan of intermediate electors for President which the Convention agreed upon represented a delicate balance between large-state and small-state interests. The essential compromise in the plan, however, did not lie in the apparent concession to the small states involved in the Electoral College apportionment scheme which gave each state two electors corresponding to its Senators, no matter how small its population. That element was not even debated at the Convention or in the succeeding ratification debates; it had, in effect, been taken over from the "Connecticut Compromise" plan for congressional representation which the Convention had agreed upon more than a month and a half previously. What was considered the major concession to the small states was the intermediate elector plan which stipulated that in the event there was no majority in the Electoral College, the election would be determined by the House of Representatives, where each state would have a single vote. The delegates apparently believed that most elections would produce no Electoral College majority. They provided that the House could pick from among the top five electoral vote recipients, thus giving the smaller states a good chance to elect one of their men. The early growth of political parties, of course, made a majority vote for President in the Electoral College more likely. Only two elections those of 1800 and 1824 - have gone to the House for resolution. Thus