The Electoral College: Enlightened Democracy

The Constitution's Election Process

Modern-day American presidential elections are governed by the 12th Amendment to the U.S. Constitution,9 which was adopted and ratified by the states in 1804. The 12th Amendment modified the Constitution's original Article II election procedure to provide for the current system of one-ticket voting for the President and Vice President.10

It is perhaps easiest to think of the current election procedure in two phases: first, the Electoral College vote, and second, the contingent election procedure, which is used only if no candidate wins a majority of electoral votes.

The Electoral College Vote. The Constitution provides for a presidential election among the states, rather than among individuals. In this election, each state is granted a certain number of representatives, called electors, to cast votes on its behalf.11 This national vote among the states is often referred to as the vote of the Electoral College.

States are allocated one elector for each of their representatives in Congress.12 Each state therefore automatically receives a minimum of three votes, as it is entitled to at least two Senators and one Congressman, regardless of population.13 Adoption of the 23rd Amendment in 1961 provided the District of Columbia with at least three electoral votes, as if it were a state.14 There are currently 538 total electors. Following the 2000 census, California has the most electors (55), while seven states plus the District of Columbia have the minimum number of electors (3).15

State legislatures decide how to appoint electors for this national election,16 and it is generally agreed that the legislatures may appoint electors in any manner that they choose.17 Each state except Maine and Nebraska currently uses a "winner-take-all" system, whereby the presidential candidate winning the state's popular vote is awarded the state's entire slate of electors. Maine and Nebraska each give two electoral votes to the winner of the state's popular vote and select the remaining electors by congressional district.18

The Electoral Count Act of 1887,19 as updated through the years, currently provides an election timeline: Election Day is the Tuesday following the first Monday in November in any given presidential election year.20 Each state certifies a slate of electors based upon the outcome of its popular vote. These electors assemble in their state on the first Monday after the second Wednesday in December.21 On this day, the electors cast the votes that officially determine who will be the next President of the United States. Congress meets in joint session to count these votes on the following January 6.22

To be elected President, a candidate needs a majority of these states' electoral votes, which are cast in December. He does not need a majority of the direct popular vote cast on Election Day.23 At this time, 270 votes constitute a majority of the Electoral College and will win the presidency for a candidate.24

The Contingent Election. If no candidate wins a majority of the Electoral College vote, the Constitution provides a backup method for presidential selection. This procedure is often referred to as the Constitution's contingent election procedure. In this secondary election, the House selects a President and the Senate selects a Vice President.25

In the House vote for President, each state delegation is granted one vote.26 If a state's delegation is evenly divided, that state cannot vote until and unless the tie is broken. A President is elected when one candidate wins a majority of the votes of state delegations (currently 26). State delegations may vote for any one of the top three presidential candidates in the electoral vote, and re-votes are taken until a majority is achieved. A similar procedure is employed for election of the Vice President, except that each Senator is granted one vote and only the top two vice presidential candidates from the electoral vote are candidates in the Senate vote.

The Founders spent months of extensive deliberation on the topic of presidential election: They deemed it one of the most difficult issues facing the Convention. When the Electoral College proposal was completed, they viewed it as a unique and commendable solution that would satisfy the many -- apparently contradictory -- goals of the new republic.

An 18th Century Solution in
21st Century America

Much has changed since 1787. The Founders could not have foreseen the rapid technological advancements, massive federal bureaucracy, and increasingly populist attitudes that characterize American life today. Could it be that the Electoral College, although once an ingenious solution to many 18th century problems, has today become merely an anachronism -- and a potentially dangerous one at that?

The Electoral College undoubtedly operates in a different society from the one that existed in 1787. Yet the Electoral College has shown an amazing ability to adapt to modern-day America. It may sometimes operate differently than expected, but it still serves the political goals it was intended to serve. In truth, its operation in modern times may be even more valuable.

The Benefits of Federalism. Critics of the Electoral College allege that the country's presidential election process does more to trample the rights of individuals than to protect federalism. In this context, they often cite the "winner-take-all" method employed by most states, claiming that it causes the votes of some individuals to be "wasted."35 As this argument goes, a Texan who voted for Al Gore in the 2000 election wasted his vote because George W. Bush was awarded the state's entire slate of electors under the winner-take-all method. Gore did not win so much as one electoral vote from Texas, despite winning nearly 2.5 million of that state's popular votes during the election. 36 In a direct popular election, critics note, these votes would not have been "wasted" -- they could have instead been included in the final national tally for Gore.

Such arguments, however, are a bit disingenuous. These votes were not wasted. They were simply cast on the losing side of a popular vote within the state. If the 2000 election had been conducted based on nationwide popular vote totals only, would people claim that any vote for George W. Bush was "wasted" because Al Gore won the popular vote? Of course not. The votes for Bush were cast in an effort to win. In the event of a loss, they would simply have been votes for the losing candidate -- just as in any other election (such as an election for Governor or Senator).

The primary effect of America's federalist presidential election process is to protect the freedom of individuals -- particularly those in small states and sparsely populated areas. Perhaps the best method of demonstrating the benefits of federalism is to expose the evils suffered without it.

As the system stands today, presidential candidates have no incentive to poll large margins in any one state. Winning 50.1 percent of the votes in a state is as effective as winning 100 percent of the votes. Presidential candidates therefore tour the nation, campaigning in all states and seeking to build a national coalition that will enable them to win a majority of states' electoral votes.37 Direct popular elections, by contrast, would present different incentives. Suddenly, winning 100 percent of the votes is better than winning 50.1 percent of the votes. In fact, it may be easier to rack up votes in a friendly state than to gain 50.1 percent of votes in each of two states of similar size, although the payoff would be essentially the same.

The result? Democrats would almost certainly spend most of their time in the large population centers in California and New York. Republicans would campaign in the South and Midwest. Large cities would be focused on almost exclusively as the candidates seek to turn out as many votes as possible in "their" region of the country. Small states, rural areas, and sparsely populated regions would find themselves with little to no voice in presidential selection. In this scenario, a handful of states (or heavily populated cities) win, while the remaining states and less-populated areas suffer significantly.38

Many critics dispute this description of the two types of elections. They contend that the current system does not encourage presidential candidates to tour the nation, but instead encourages a focus on mid-sized "swing" states. "Safe" states and small states, they allege, do not receive nearly as much attention on this national tour.

There is an element of truth in this observation. Yet to the degree that safe states do not receive a proportionate amount of attention during campaigns, the logical conclusion is that those states, by and large, must already feel that one of the two presidential candidates represents their interests fairly well. When a candidate ceases to adequately understand and represent one of "his" state's interests, the discontent in that state is usually expressed pretty quickly.

Consider the situation in West Virginia in recent decades. Democrats considered West Virginia a safe state for years; thus, the state probably saw less post-nomination campaign activity from 1960-2000 than it might have otherwise. However, in 2000, the Bush campaign recognized an opportunity to gain a foothold in the state due to concern about the impact of Gore's environmental policies on the coal-mining industry and his support for gun control.39 Bush took advantage of this discontent, and he spent more than $2 million communicating his message to West Virginia's voters.40 When election results were tallied, Bush became the first Republican since 1928 to win an open race for the presidency in West Virginia.41 In 2004, West Virginia is no longer considered a safe state for Democrats.

A second argument made by critics is similarly flawed. Although the winner-take-all system causes large states (especially large swing states) to elicit more attention than small states, these critics erroneously compare the amount of campaigning in small versus large states under the current system. They should instead compare the treatment of small states under the current system against the treatment they would receive under a new one. Today, small states undoubtedly receive less attention than large states (unless, of course, the large state is considered a safe state). However, a direct vote system would magnify, not improve, this problem because it would encourage a focus on highly populated areas. Small states would likely never receive as much attention as their larger neighbors. The goal is not to eliminate this disparity, but to minimize its severity. Under the Electoral College system, the states are as evenly represented as possible, given that they are not all the same size.

One interesting twist to the arguments raised by Electoral College critics focuses on the reality that even if small states benefit from the Electoral College, they do so at the expense of the individuals who reside in small states. This complaint can be confusing because it sounds like the opposite of another complaint -- that the two vote add-on for small states (giving all a "guaranteed minimum" of three electoral votes) creates a bias in their favor. The two extra electoral votes given to all states, regardless of population, do create an advantage for those states. As a statistical matter, however, the advantage plays in favor of the state as a whole, rather than the individual voter. By contrast, the mathematical advantage granted by the winner-take-all system plays in favor of individual voters in the larger states. These voters have a statistically higher probability of materially affecting the outcome of the election.42

As a purely statistical matter, perhaps this assessment is accurate. However, the odds of any one voter providing the "tipping point" in an election are still exceedingly small. Further, any individual disadvantage for those who reside in small states is outweighed by the larger advantage given to the state as a whole.

In sum, the nation conducts democratic, popular elections -- but they are conducted at the state level, rather than the national level. Professor Charles R. Kesler of Claremont McKenna College explains: "In truth, the issue is democracy with federalism (the Electoral College) versus democracy without federalism (a national popular vote). Either is democratic. Only the Electoral College preserves federalism, moderates ideological differences, and promotes national consensus in our choice of a chief executive."43

Moderation and Compromise . Presidential candidates must build a national base among the states before they can be elected. They cannot target any one interest group or regional minority. Instead, they must achieve a consensus among enough groups, spread out over many states, to create a broad-based following among the voters. Any other course of action will prevent a candidate from gaining the strong base needed to win the election. The necessity of building such a national base has led to moderation and a strong two-party system in American politics.

Some see this trend toward moderation and a two-party system as a liability. They argue that certain points of view on the far left or far right do not have representation. Some voters do identify with a third party more than they identify with one of the two major parties. Democratic theories try to satisfy the choices of all voters, but not to the point of destabilizing democratic majorities and democratic government itself. A system that favors a stable two-party system, but allows minority parties to vie for control, has a definite benefit over a system that favors many minority parties: Hand in hand with the Electoral College, it tends to prevent the rise to power of extremist groups and radical minorities. Instead, American public policy tends to remain in the middle -- not too far left, not too far right.

Some proponents of a direct election system dispute this analysis. Removing the Electoral College, they argue, would not undermine the two-party system.44 However, these proponents make the mistake of assuming that they can change one factor while leaving all other dynamics unchanged.45 Predictions of what would happen in a direct election system should assume that it is at least possible that changing the election procedure will cause more than one aspect of the political process to change. Indeed, one has only to look to history or comparative governments to see how easily such a system could disintegrate into multi-candidate races, which would, in turn, devolve into a system of regular runoffs or fractious coalition governments.

Consider the election of 1992. One of the primary factors that prevented some people from voting for Perot was that they felt their vote would be "wasted." A vote for Ross Perot is a vote for Bill Clinton, voters were told. Now imagine the 1992 election without an Electoral College. Suddenly, the incentives change. Clinton no longer has to be beaten; he just needs to be held under some pre-determined percentage (40% in most proposals for change). To get in the runoff, Perot voters do not need to win; they simply need to overtake George H.W. Bush. A vote for Perot is no longer wasted. The Reform Party has a specific, achievable goal with which to motivate its supporters.

In the actual 1992 election -- even with all the disincentives inherent in the Electoral College system -- Perot received 18.9 percent of the popular vote. Bush received 37.4 percent of the vote, and Clinton received 43.0 percent. Had this election been a direct popular election, it would have been only three percent away from triggering a runoff. With such incentives, would not Perot supporters have had a much easier time getting voters to switch their candidate? Once one runoff has been triggered, why would multiple third parties not jump into the race in future elections?

As such a situation became the norm, more and more candidates would be motivated to enter presidential contests. Support from a smaller and smaller percentage of the population would be needed to qualify for the runoff. Over time, candidates would become more and more extreme and uncompromising. Professor Judith Best explains this dynamic:

[T]he splintering of the vote works against the moderate candidates and works to the advantage of the immoderate, extreme candidates. It does this because the middle is where the inclusive coalitions can be built. By undermining coalition building prior to the general election, a runoff fragments the middle, not the extremes; the extremes are rarely fragmented -- fanatics have solidarity.46

Direct popular election proponents sound plausible when they argue for election of the President by a "majority" of the people. In reality, however, a President elected by the majority of citizens will rarely be achievable. A majority, after all, will never agree on an ideal candidate. Given an open choice, individuals would fracture their votes across many candidates. Runoffs would proliferate, and Presidents would essentially be elected by the initial 20 or 25 percent of voters who got them into the runoff. Presidents would always know that at least 75 to 80 percent of the people originally voted for someone else.

Given the general inability to obtain majority consensus, the Electoral College provides the country with the next best alternative. Electing Presidents by states' votes, rather than individuals' votes, creates a method of electing a President who is a good compromise candidate for the majority of Americans.47 The Electoral College requires moderation, compromise, and coalition building from any candidate before he can be successful. Direct elections and a system of runoffs discourage such behavior.

Stability and Certainty in Elections. Historically, most elections have not been close in the Electoral College, even when the popular vote is close. The Electoral College system, when combined with the winner-take-all rule, tends to magnify the margin of victory, giving the victor a certain and demonstrable election outcome.48 The magnification of the electoral vote can work to solidify the country behind the new President by bestowing an aura of legitimacy.

The election of 1960 was one such close election. John Kennedy won only 49.7 percent of the popular vote, compared to Nixon's 49.5 percent. However, Kennedy won 56.4 percent of the electoral vote, compared to Nixon's 40.8 percent. Eight years later, this magnification effect worked in favor of Nixon. Although he won the popular vote by less than one percent, he won 55.9 percent of the electoral vote to Hubert Humphrey's 35.5 percent. This magnification effect increases dramatically as popular vote totals spread apart. For instance, in 1952, the winning candidate won 55.1 percent of the popular vote, but a much larger 83.2 percent of the Electoral College vote. In 1956, the difference was 57.4 percent (popular vote) to 86.1 percent (electoral vote). In 1964, it was 61.1 percent (popular vote) to 90.3 percent (electoral vote).

Presidential elections since 1804 have generally seen wide margins of victory in the Electoral College. These margins have gotten wider, on average, through the years as the winner-take-all rule has been adopted by more states and the two-party system has solidified. Since 1804, only two elections -- those in 1876 and 2000 -- were won by fewer than 20 electoral votes. Six elections were won by fewer than 50 electoral votes: Four of these were held in the 1800s. Of the 26 elections held between 1900 and 2000, 17 Presidents have been elected after winning the electoral vote by a margin of 200 votes or more.

These consistently wide margins of victory in the Electoral College have come about despite the fact that the margin between the top two candidates in the popular vote was less than 10 percent in 14 of the 26 elections held since 1900. This margin exceeded 20 percent only five times since 1900.

A direct popular election, by contrast, would not grant certainty nearly as often. Close popular votes, such as those discussed above, could easily result in demands for recounts on a national scale. America rarely has close electoral votes. It does, however, have close popular votes fairly consistently. Do Americans really want a presidential election system that could result in hotly contested recounts nearly every election?

The Electoral College provides yet another benefit: It reduces the incidence of fraud and error. Obviously, no system can completely eliminate the element of human error. Neither can any system eradicate the tendency of some dishonest individuals to cheat. An election system can, however, minimize the extent to which these factors affect elections.49 The Electoral College defends against fraudulent behavior and human error in two ways: First, the system makes it difficult to predict where stolen votes will make a difference. Second, to the degree that fraud and errors do occur, the Electoral College makes it possible to isolate the problem to one state or a handful of states. The country is given a clear set of problems to resolve one way or another before moving on to a definitive election outcome -- much as it knew in 2000 that the election would be certain once Florida's disputes were resolved.50

Today, both the electoral and the popular votes must be extremely close before voting disputes and recounts are threatened. By contrast, a direct popular election would require only a close popular vote before these scenarios became possible. National recounts and legal challenges would be a constant possibility, particularly because of the increasing likelihood of multiple candidacies, lower individual vote totals, and smaller margins among candidates. Moreover, a direct popular vote system would increase, rather than decrease, the incentive for fraud. Any stolen vote would have at least some effect, regardless of its location. Party officials and supporters in states in which that party clearly dominates have the greatest ability to rig election rules and get away with cheating on behalf of their party's candidate. Under the Electoral College system, however, they have the least incentive or need to cheat for their presidential candidate. Using a direct election system, dishonest officials in one-party states have both the ability and incentive to cheat, creating potential resentment, suspicion, and hatred from other states and from the citizens of their own state. The Electoral College minimizes the impact of fraud, isolating it to the one or two states where the vote was close, disputed, and relevant to the Electoral College balance.

Addressing Alternative Proposals

Direct popular election is not the only alternative to the Electoral College that has been proposed. Two other proposals are also worthy of discussion: (1) The "District Plan" would award electoral votes by congressional district, rather than by state; and (2) The "Proportional Plan" would divide states' electoral votes proportionally, according to the popular vote outcome within the state. Individual legislatures are free to adopt most such plans now. It is telling that so few have done so.

District Plan. The District Plan would award electoral votes based upon congressional district. The candidate with the most votes in any congressional district would win that district's one electoral vote. The two Senate "add-on" votes in each state would then be treated as "at-large" votes and awarded to the state's popular vote winner. Maine and Nebraska already operate under this system. The proposal would require other states to join them. Neither Maine nor Nebraska has divided its electoral votes since their systems were first implemented, but with close presidential elections the odds increase that one or both states will do so. 51

Proponents of the District Plan argue that the system preserves the small-state advantage, while allowing diversity within the states to be reflected in the national election totals.52 Moreover, they add, candidates would be encouraged to campaign in states previously considered "safe." Greater voter participation would result, they argue, because every voter would feel that his vote counts. Other Electoral College critics dismiss the District Plan, arguing that the plan fails to address the "problem" of a popular vote winner failing to win the presidency.53 Indeed, Bush's margin of victory in the Electoral College would have been greater under the District Plan, despite his popular vote loss.54

The District Plan retains some of the advantages of the current Electoral College arrangement. America's presidential election system should encourage national coalition building, and campaigning by districts would certainly work toward this goal.55 However, many of the other benefits envisioned by District Plan proponents are unlikely to come to pass. Greater voter participation is improbable, particularly in lopsided or "safe" congressional districts -- and most congressional districts are safe. This so-called reform may actually have the opposite effect from what was intended in states that are close overall but have congressional districts that are not. Even worse, the primary incentive created by the District Plan is an increased motivation for gerrymandering. State boundaries have one significant advantage over congressional district boundaries: They are reasonably permanent and cannot be changed by an incumbent class of officials.56

Another problem created by the District Plan is that it diverts attention from statewide and national issues. It discourages candidates from addressing issues that appeal to the state as a whole and might encourage candidates to visit district rich, high-population centers to the exclusion of rural districts. Moreover, if a state votes in pieces, consequential issues would fall out of the larger debate in favor of local grievances. Focusing on a few swing districts would repeat on a fragmented scale the "swing state problem." Candidates might visit only a few close districts and ignore areas considered safe.

Despite these arguments against the District Plan, the current system allows a state to switch to the District Plan if it decides that doing so would be beneficial within its borders -- and it may do so without the necessity of passing a constitutional amendment. Moreover, the state can switch back to a winner-take-all system if implementing the District Plan causes more problems than it solved (as it is likely to do). Constitutional amendments are not completed -- or undone -- nearly as easily.

Proportional Plan. Under the Proportional Plan, the Electoral College would be retained, but the electoral votes in each state would be allocated based upon the percentage of the popular vote won. Some versions of this plan call for whole electoral votes to be divided in the interest of accuracy.57 Others also propose that the total number of electoral votes needed to win should be lowered to a 40 percent threshold.58

Proponents argue that the plan would preserve the states' Electoral College strength, but that the results would be closer to the national popular vote outcome.59 They further argue that the voices of minority voting groups would be reflected in the final election tally because even ballots cast for third-party candidates could be reflected in the national results. Presidents could be elected without winning the popular vote, but such an outcome would be less likely than under the current system. Other Electoral College opponents deem even the small remaining possibility of a "popular vote loser" President to be unacceptable.60

The Proportional Plan is yet another option that sounds good on the surface, but that creates problems in practice. First, constant fighting would erupt over election tallies. After all, a swing of one-tenth of a percent in one or several states could change the outcome of the election. Potentially, at least one electoral vote in each of the 50 states would be open to dispute every presidential election year.61 Worse, because this type of election is so similar to a direct popular election system, it contains many of the same downfalls: multiple candidacies, close elections, runoffs, greater potential for fraud, and greater incentives for extremist candidates to join the fray. Definitive election outcomes are discouraged. Close vote totals and challenges to election outcomes would become the norm.

Colorado, however, is currently considering adoption of this plan. A Colorado citizens' organization -- financed by a group from San Francisco and funded by a Brazilian millionaire who resides part-time in California62 -- recently collected enough signatures to place a referendum on its November 2004 ballot.If passed, this referendum would supposedly replace Colorado's winner-take-all system of casting electoral votes with a system of proportional allocation, making Colorado the only state to enact such a method in presidential elections.63

The referendum, however, has a glaring flaw. It purports to let Colorado voters decide the method of allocating the state's electoral votes, despite the fact that the Constitution gives this power to state legislatures, not state citizens.64 For constitutional purposes, the difference is very real. If the referendum passes, the most likely result is a series of court challenges in Colorado. Proponents of the referendum argue that state statutory and constitutional provisions make the people equivalent to the legislature when they act through referendum or initiative. Therefore, they conclude, passage of a referendum or other citizen initiative is a perfectly legal and valid manner of changing electoral vote allocations in the state.

These proponents, however, have forgotten the Supremacy Clause of the U. S. Constitution: State laws and constitutions may not trump the U.S. Constitution. The U.S. Constitution is the supreme law of the land and preempts all other laws. That same Constitution explicitly designates each state's legislature as the entity with authority to decide the manner of appointing electors. States (whether by referenda or other means) may certainly delegate legislative power to citizens for state purposes, but they may not delegate the state legislature's constitutionally mandated duties in the national presidential election process.

This principle has been upheld by the Supreme Court in venerable old cases and some very recent decisions after the last presidential election. Speaking about an Ohio plan that would allow the people to act in place of the legislature for certain purposes related to ratifying constitutional amendments, the Court observed in 1920 that the Founders wrote the constitutional language based upon their wish to "secur[e] deliberation and consideration before any change [to the Constitution] can be proposed."65 In short, the Court held, when the Constitution says "the Legislature," it means "the Legislature." The Court explained that:

The framers of the Constitution might have adopted a different method. Ratification might have been left to a vote of the people, or to some authority of government other than that selected. The language of the article is plain, and admits of no doubt in its interpretation. It is not the function of courts or legislative bodies, national or state, to alter the method which the Constitution has fixed.66

In even simpler terms, however, the Court also pointed out what almost any middle-school child should know: A Legislature is a representative body, not the people themselves.

What did the framers of the Constitution mean in requiring [action] by `Legislatures'? That was not a term of uncertain meaning when incorporated into the Constitution. What it meant when adopted it still means for the purpose of interpretation. A Legislature was then the representative body which made the laws of the people.67

This principle applies to the constitutional process for electing a President, just as it applies to the constitutional amendment process.68

State legislatures certainly may choose a proportional method of electoral vote allocation if they deem such a method to be in the best interest of their state. However, if one state acting on its own (particularly a small state) were to enact a proportional method of allocating electoral votes, the primary effect of its action would be to significantly dilute its voting strength as compared to the other states because no presidential candidate will spend much time in a state in which he is likely to influence only one swing electoral vote. No wonder so few states have enacted anything like this before. State legislatures will better serve their citizens if they leave the winner-take-all system in place.