Looking only at the allocation of electoral votes and the populations of each state, some citizens have more than 3 times the theoretical voting power of others, roughly what it was back in 1968 when it was first calculated; calculations which helped persuade the House Judiciary Committee to overwhelmingly recommend replacing the Electoral College with the direct election of the president.
Prior to that time, it was widely believed that citizens in the least populous states had the greatest voting power, because even the smallest states were entitled to 3 electoral votes.
But a computer-based analysis showed that the states with the most electoral votes – New York was the most populous at that time – had such a large chance of being decisive with their block of electoral votes that this factor outweighed the lesser ability of any citizen of such a large state to cast a decisive vote
Once legislators from the small states became convinced of this through two congressional hearings, their opposition to the direct election of the president was largely replaced by support.
But the current election is a good example of how much theoretical voting power (a calculation based solely on the number of electoral votes and the state's population, and a disparity which will continue to exist no matter how the political sentiments of voters may change over time) and actual voting power (which takes into account in addition how close the presidential race is in each state in a particular election) can differ.
If the two major candidates are virtually tied in a state, the odds that any one of a very tiny minority of undecided residents in that state will cast a decisive vote is very high, at least compared with a state which is leaning so strongly towards one candidate that a vote by an undecided voter will, as a practical matter, have no effect whatsoever on how that state's electoral voters will be cast.
So citizens in swing state have far greater actual voting power than citizens in a non-swing states, explains Banzhaf.
As one example, if one did not take into account the political leanings and distributions of voters, citizens in California would have the highest theoretical voting power because their state has such a large number of electoral votes.
But if, as most analysts suggest, the presidential race in that state is totally one-sided, the undecided voters will have almost no chance to affect the outcome of the election, and thus have virtually no actual voting power.
In contrast, where the polls show a dead heat with less than 2 percentage points separating the two candidates, any one of a very small number of undecided voters has a measurable chance of affecting how that state's electoral votes will be cast, and of that state's electoral votes affecting the outcome of the election.
According to one set of calculations, citizens in Ohio have more than 100 times the actual voting power of citizens in Michigan, largely because Buckeye voters are so closely divided. Another analysis, using somewhat different polling data, suggests that the disparity could be more than 1,000,000 to 1. Probably the correct answer lies somewhere in between these two figures, suggests Banzhaf.
One result of the huge disparity in voting power between residents of swing states and non-swing states is that voters in California are being largely ignored by the candidates, whereas voters in Ohio and even New Hampshire are being inundated with visits and TV commercials, even though California has far more electoral votes than both states combined, and even though Californians have far greater theoretical voting power.
Actual voting power will change with each election according to the popularity of each candidate, shifts in the percentage of different groups in each state, and many other factors, but the basic disparity in theoretical voting power will change solely as a result of changes in population, says Banzhaf.
Clearly it's a matter of concern, he suggests.
Additional information about using the Banzhaf Index to calculate voting power under the Electoral College may be found in: James Michener, Presidential Lottery, Part C entitled "The Banzhaf Studies" at 220 (1969); Pierce, The People's President, Section O entitled "Computer Analysis of Large versus Small State Power in the Electoral College" at 362 (1968); Robinson and Ullman, A Mathematical Look at Politics (2010); John Allen Paulos, A Mathematician Reads the Newspaper (1997); Hearings before the Subcommittee on Constitutional Amendments, U.S. Senate, p. 517-42, 904-33; Electoral College Reform, Hearings before the Committee on the Judiciary, House of Representatives, p. 306-74.
JOHN F. BANZHAF III, B.S.E.E., J.D., Sc.D.
Professor of Public Interest Law
George Washington University Law School,
FAMRI Dr. William Cahan Distinguished Professor,
Fellow, World Technology Network,
Founder, Action on Smoking and Health (ASH)
2000 H Street, NW
Washington, DC 20052, USA
(202) 994-7229 // (703) 527-8418