To learn more, see the FAQ's below and the list of endorsers. For more information on the National Popular Vote plan, including weekly updates and ways to get involved, please visit NationalPopularVote.com.
According to the U.S. Constitution (Article II, Section I), state legislatures decide how to apportion their state’s electoral votes, a power that the Supreme Court has termed “plenary” (absolute). Congress cannot override this right beyond resolving certain administrative questions like the timing of when the Electoral College casts its votes and when states provide their official presidential vote totals.
States have often exercised their power over how to allocate electoral votes, particularly during the lifetime of our founding generation. Most states today use the winner-take-all unit rule by which they award all electoral votes to the winner of the statewide popular vote. Few states used the unit rule in our early elections, however, and it did not become the norm until a half-century after the approval of the Constitution. Even today, Nebraska and Maine choose to apportion their votes differently.
The U.S. Constitution (Article I, Section 10) also establishes that states can enter into binding interstate agreements. There are hundreds of such compacts, including ones establishing the Port Authority of New York and New Jersey and protecting states’ access to the water of the Colorado River. Many compacts require congressional approval, but others that do not infringe upon federal powers, such as the National Popular Vote plan, do not require it. Even if congressional approval ultimately is sought for the National Popular Vote plan, that approval would take place after the compact is approved by sufficient states to enact it.
The Founding Fathers did not design the system of allocating electoral votes currently used in most states. Rather, they established the Electoral College without any instructions on how states should use it. In the first presidential election in 1788-89, only five states allowed citizens to vote for the president in any form. In 1800, when Thomas Jefferson won our fourth presidential election, only two states allocated electoral votes based on the now-dominant winner-take-all unit rule in which all electoral votes went to the candidate winning the statewide popular vote.
Adoption of the unit rule in the following decades was driven by partisan motives. The unit rule helped each state’s majority party by maximizing support for that state’s favored candidate. Once enough states adopted the unit rule, any state using another method risked hurting its favored presidential candidate. If states controlled by one party used the unit rule and states controlled by another party allocated electoral votes by congressional district, then the first party would gain a national advantage. In short, nearly all states were forced into the unit rule by the actions of other states and by partisan calculations – ones that will continue to govern states’ decisions about allocating electoral votes until enactment of NPV.
Leaders in our nation’s founding generation almost certainly would have been dismayed by how the Electoral College system operates today, as it results in most states and the unique interests of their voters being ignored. There is every reason to believe they would have chosen to act to fix the problem.
Evidence of their willingness to reform the Electoral College comes from the 12th amendment to the Constitution, adopted in 1804. In the first four presidential elections, presidential candidates ran without official running mates for vice-president. Each elector cast two votes. If gaining support from a majority of electors, the candidate earning the most votes became president, while the candidate with the second-most votes became vice-president. In 1796, John Adams won the presidency, but his electors withheld just enough of their second votes for his running mate Thomas Pinckney to allow Adams’ opponent Thomas Jefferson to finish second and become vice-president.
In 1800, backers of Thomas Jefferson wanted to avoid such a result. Every Jefferson elector also voted for his running mate Aaron Burr. The result was an Electoral College tie, which threw the choice for president into the U.S. House. It took days of votes in the House before Jefferson won. In the wake of these failures of the original Electoral College system, Congress and the states approved the 12th Amendment that established official running mates and limited electors to one vote.
Leading thinkers at the Constitutional Convention like James Madison preferred a national popular vote for president, but accepted the Electoral College for political and administrative reasons that no longer exist today. For example, southern states with large numbers of slaves knew they would be big winners with the Electoral College system. Slaves were denied voting rights, as were nearly all women, racial minorities, and citizens not owning property, but were still counted for the purposes of determining a state’s number of congressional seats and electoral votes. In 1800, Virginia had fewer free citizens than Pennsylvania and New York, but had more electoral votes due to slaves representing 39% of its population. It is no coincidence that four of our first five presidents were from Virginia.
Over the years, states have taken the lead in bringing more democracy to presidential elections, with all states since 1876 awarding electoral votes based on popular elections and states regularly taking the lead in expand the franchise in presidential elections. Now that there is a roadmap to secure elections based on a popular vote for president, states have every reason to take this additional step to reflect the goals of a representative democracy.
Under NPV, the Electoral College would remain the actual institution that elects the president, but would play a secondary role to the all-important national popular vote. Currently, the Electoral College serves to ratify the separate popular votes of the 50 states and the District of Columbia, with 48 states awarding electoral votes to the electors associated with the winner of the popular vote in that state. Under NPV, the Electoral College would instead ratify the national popular vote by having participating states award votes to the electors associated with the slate of the national popular vote winner. The night of the election, all eyes would be fixed on the national popular vote in all 50 states and the District of Columbia, not the artificial “race to 270.” Voters seeing the popular vote tally rise on election night would know that one of those votes was their own. They would know that the candidate with the most votes wins.
For those who argue that the electors should play a meaningful deliberative role, note that today electors are simply party loyalists who support their party’s candidate.
Federal law already prohibits states from changing their laws governing that year’s allocation of electors after Election Day. As a formal interstate compact, NPV adds the additional protection of the Impairments Clause of the U.S. Constitution (Article I, Section 10, Clause 1), which prohibits states from dishonoring interstate compacts. The NPV compact includes a “blackout” period between July 20 of a presidential election year and the inauguration on January 20 of the following year. The terms of the compact do not allow individual states to withdraw from a compact during this time. This six-month blackout period covers when parties typically hold national nominating conventions, the general election, the state certification process, the actual meeting of the Electoral College in December, the counting of the electoral votes by Congress on January 6, and the inauguration of the President and Vice President for the new term on January 20.
Once NPV is activated, it is unlikely that states will withdraw from the compact. The public overwhelmingly supports a national popular vote for president and opposes efforts clearly designed to “game” the system. An individual state is unlikely to be able to undermine the compact because the number of electoral votes in participating states likely will continue to rise beyond the Electoral College majority of 270 votes.
No. Slates of electors are supplied by the presidential candidates and consist of individuals who strongly support the candidate and their party. The likelihood of faithless electors would not increase under NPV because the electors casting votes in states participating in the NPV agreement would have been selected by the winning candidate.
NPV would ensure that the presidential candidate receiving the most votes throughout the country would be elected president. The votes in every state and jurisdiction allocating electoral votes would be counted to determine the winner. Whether or not a state enters the NPV agreement, its voters would have a vote equal to everyone else's in selecting the president. For example, if Ohio does not approve the compact, the votes cast in Ohio would still be part of the national popular vote totals that would determine the winner of the presidency. Ohio voters would be treated the same as all other voters.
A nonparticipating state could choose not to hold a statewide popular vote for president and therefore not generate vote totals that would be part of the national popular vote, but such a move would be strongly opposed by that state’s voters – both because voters in state after state overwhelming support a national popular vote for president and because once NPV is enacted, every state will want its voters to help determine the outcome of the election.
No. NPV would only go into effect when it is ratified by enough states to make it decisive in electing the president – meaning that participating states collectively hold a majority of Electoral College votes. Until that happens, states continue to apportion their electoral votes according to their current rules. For example, during the 2008 election, Maryland, Hawaii, Illinois, and New Jersey had already enacted NPV. Because their total electoral votes were less than an Electoral College majority of 270, however, the compact was not triggered and those states continued to apportion electors as they had done previously.
FairVote is among a number of leaders and organizations that support both the National Popular Vote plan and a constitutional amendment providing for direct election of the president. They are complementary positions. At the same time, it is clear that, as of 2009, there is no consensus on the form of a constitutional amendment to establish a direct election and there is no energy on Capitol Hill to advance an amendment. Furthermore, states historically have acted on their power to expand democracy in presidential elections without waiting for a constitutional amendment: it was states that took action to ensure popular elections within states for elector allocation, and that first expanded suffrage rights to include women, African Americans, Native Americans and citizens who did not own property. An amendment needs to be ratified by three-quarters of states and two-thirds of each chamber of Congress. Even though an overwhelming majority of Americans support a national popular vote for election of the president, a small minority of the country can block reform if only pursued in the form of amendment.
Even those who prefer a constitutional amendment have every reason to support NPV as a means of drawing attention to the failings of the Electoral College and illuminating public support for altering the system. Some NPV backers indeed strongly prefer it to an amendment given that NPV is constitutional and gives states more flexibility in the future to change how they want to organize presidential elections.
The Electoral College allows a President to be elected when finishing second or even third in the national popular vote. So-called “wrong winners” became president in 2000 (George W. Bush), 1888 (Benjamin Harrison), 1876 (Rutherford B. Hayes) and 1824 (John Quincy Adams). We have had several near wrong winners, including in 2004 when a shift of less than 60,000 votes in Ohio would have given John Kerry a national victory over George Bush despite his national popular vote deficit of three and a half million votes.
Even when the Electoral College does not elect a losing candidate, however, it makes losers of a large and growing number of Americans in every single election. The winner-take-all unit rule governing the Electoral College in most states leads candidates to ignore the vast majority of Americans as their campaigns focus nearly all their resources on a handful of “battleground” states that, because they are too close to call, have the power to determine the election. In 2008, more than 98% of all campaign spending and all campaign events after Labor Day were in 15 states representing barely a third of the population – and nearly every remaining state was largely ignored in 2004 and 2000 as well. The number of battleground states has steadily declined in recent decades, resulting in candidates focusing on an increasingly narrow band of American voters and rendering meaningless most potential grassroots involvement in so-called “spectator states.”
Some Electoral College defenders dismiss “wrong winners” (meaning elections where the national popular vote winner loses) as aberrations. But one out of every 12 presidential elections since the Civil War has experienced a wrong winner, and there were several additional close calls, including victories by Woodrow Wilson in 1916, Harry Truman in 1948, Jimmy Carter in 1976, and George Bush in 2004. Read about the close calls here. Even Barack Obama’s victory by nine and a half million votes in 2008 would have been trumped by a shift of 527,864 (flipping 99 electoral votes) in the right combination of states.
The problem of popular vote reversals is particularly serious in light of today’s more closely divided electorate. American politics is in a period of narrow division between the major parties. Since 1984, every presidential race has been won by less than 10% of the national popular vote. In contrast, a majority of elections between 1952 and 1984 were won by margins greater than10%. Of the last 15 presidential elections won by less than 15%, three (one out of five) had popular vote reversals: in 1888 and the particularly pivotal elections of 1876 and 2000. There easily could have been more reversals, with a shift of fewer than 80,000 votes in the right combination of states able to reverse the winner in five presidential elections since World War II alone.
States adopted the winner-take-all unit rule primarily due to partisan incentives – their leaders did not want other states to gain an advantage for their favored candidates. Unlike NPV, which creates a level playing field, changing a state’s rules for allocating electoral votes nearly always is suspect from a partisan perspective. Such “reform” efforts are typically led by those whose party would secure a national advantage if that particular state were to divide its electoral votes rather than award all of them to the statewide popular vote winner.
As FairVote demonstrated in its report Fuzzy Math: Wrong Way Reforms for Allocating Electoral College Votes, allocating votes by congressional district and proportional allocation both result in far less fair presidential elections than election of the president by a national popular vote. Under allocation by congressional district, fewer than one in ten congressional districts would be close enough to be truly contested by the campaigns. The current partisan landscape also would provide a huge edge to Republican candidates in close elections making the plan a non-starter for Democrats; in 2000, for example, George Bush would have won the Electoral College by 38 electoral votes under the congressional district system applied nationwide, even while losing the national popular vote. Read about the current Maine and Nebraska system here.
Proportional allocation doesn’t have the same degree of partisan bias, but it would share the problem of keeping most voters as spectators in presidential races. Political activity under this reform would concentrate on the states where a shift in popular support of 3% or less would gain or protect an electoral vote – and most states would be outside of this range.
The current use of the Electoral College hurts the great majority of those voters who live in small states because their states are completely ignored in presidential campaigns. Small states tend to have clear partisan tilts that lead the candidates to decide that no amount of campaign activity in that state will affect the result in that state. Of the 18 states with the smallest populations, 10 experienced absolutely no television ads or candidate visits with public events during the peak season of 2008 campaigns. Eleven of the 12 smallest states did not have a single campaign visit.
Furthermore, statisticians have demonstrated that voters in small states do not get extra attention, even though their states have more electoral votes per capita. Additionally, voters in large competitive states are more influential than voters in small competitive states because of the winner-take-all unit rule. Due to the unit rule, competitive big states have far more electoral votes in play, meaning that a shift of 1,000 votes in a big state can result in a far greater gain in electoral votes than a comparable swing of 1,000 votes in a small state. This statistical reality explains why Delaware and other small states filed a lawsuit in 1966 seeking to challenge big states’ use of the unit rule.
Presidential elections should not favor a special class of battleground voters over other Americans. In a national popular vote election, the unit of influence would be the individual voter, whether this voter lives in a small, big or medium state. Every vote would be equal, thereby forcing viable candidates and parties to be responsive to all voters.
No. Anyone who lived outside of a battleground state in 2004 or 2008 could tell you how much they and their neighbors mattered in the 2004 presidential race. As FairVote’s Who Picks the President? report shows, the candidates and their backers completely ignored a majority of states and a majority of people. The 2004 election concentrated almost exclusively on a dozen states that were home to less than 28% of the electorate and relatively concentrated in the Midwest. The 2008 general election saw a similar focus, with more than 98% of campaign spending and events after Labor Day focused on 15 states representing barely a third of the nation – a sharp contrast to the Democratic presidential primary contest that took place everywhere that year due to different rules making it valuable for the candidates to contest every state.
While the current Electoral College system keeps candidates from attempting to maximize vote totals in their strongest areas, it also eliminates any incentive that either major party candidate might have to visit those areas at all. Similarly, they have no incentive to campaign in any states where they are sure to lose or win. The only states that matter are the ones that happen to be close.
No. Advertising is going to be a part of modern politics, but the lesson of recent presidential races is that old-fashioned get-out-the-vote activities are essential to success. As long as there are Americans who are willing and able to get involved on the grassroots level—knocking on the doors of neighbors, etc—then there will be grassroots campaigning in the United States. The candidate who can inspire the most of that campaigning will have a significant advantage. As to money, candidates already are trying to raise as much money as they can. There simply will be a more equitable distribution of their campaign resources, in turn creating more incentives for parties and candidates to establish over time new efforts promoting participation at the grassroots level.
No. The question of state rights, federalism and republican government relates to the respective powers of the presidency, the Congress, the states and the people rather than the method of election for our elected offices. We are a representative democracy, not a direct democracy, but when electing our leaders, we should aspire to fair elections based on one-person, one-vote – combined with clearly defined rights and separation of powers, such fair electoral rules are what protects minorities within our system. Underscoring this point, in January 2012 the Department of Justice pre-cleared California's entry into the NPV compact under Section Five of the Voting Rights Act.
The use of the winner-take-all rule in 48 states is simply an electoral device, not a meaningful expression of state differences. Its negative impact on most states in comparison to a national popular vote is very real. The fact that most states and their citizens today receive no attention from campaigns undermines the goals of federalism. If every vote counted equally, the people of every state would matter, and their views and interests would matter. As it is, the only people earning attention from the campaigns are those who happen to live in competitive states. George Bush’s campaign in 2004 was the best-funded in history, but it didn’t waste a dime on polling the views of a single person living in more than 30 states during the entire campaign. When voters don’t matter, they have little power to protect their interests and the interests of their state.
Current statewide winner-take-all unit rule elections significantly increase the probability of a contentious recount due to a smaller margin of victory and larger number of elections – by as much as five times according to a 2008 paper by University of Pennsylvania professor Jack Nagel. Neither the 2000 nor 2004 presidential election was remotely close enough for a national recount, but the 2000 election hung on the result in Florida and the outcome of the 2004 election would have changed with a shift of fewer than 60,000 votes in Ohio. An extremely close state election can be hard to finalize in time for the December meeting of the Electoral College, yet such a close election in a state is far more likely than a razor-thin vote in the national popular vote.
FairVote’s 2007 survey of 7,645 statewide elections from 1980 to 2006 determined that statewide elections resulted in a recount once in every 332 elections (23 out of 7,645). Applied to national presidential elections, this number would mean we might have to conduct a national presidential election recount once every 1,328 years. Even that number is likely too high, however, as FairVote’s data demonstrated that the percentage of the vote changing during a recount declined as the size of the electorate grew. In modern elections there are relatively few errors. When an administrative error is found that affects 100 votes in an election of 5,000 votes, that error changes 2.0% of votes. But that same error affecting 100 votes in an election of 100,000 votes only affects 0.1% of votes.
More fundamentally, in the 21st century, the United States has no excuse for conducting elections that cannot be recounted. Certainly, large population states like California and Texas do not shy away from statewide popular elections out of fear of running recounts. Furthermore, Congress has the authority to establish standards for recounts in the highly unlikely event one were needed.
No. In upcoming elections, one could just as well flip a coin as use the Electoral College to decide the winner if the popular vote margin is inside a still-comfortable 500,000 votes. Al Gore lost the 2000 election despite a comfortable national popular vote advantage of a half million votes, while George Bush would have lost in 2004 if his national popular vote advantage had been reduced evenly across the states to about a half million votes. Furthermore, state-by-state election results where tiny shifts in one state can change who wins the national election always will spark controversies and legal disputes where the courts must intervene, as took place in 2000 and almost did in 2004.
The Electoral College has a provision that is an accident waiting to happen without reform: when no candidate wins an absolute majority in the Electoral College the U.S. House picks the president (with the winner needing to win a majority of the vote in 26 state congressional delegations) and the Senate picks the Vice-president. In 1948 and 1968, Strom Thurmond and George Wallace won significant numbers of southern states and their electoral votes; small popular vote shifts in a few states would have put them in a position to bargain after the election. This happened in 1876, when Rutherford Hayes was willing to allow states to trample on African-Americans’ civil rights in exchange for the White House. Even in 2004, a shift of fewer than 21,000 votes would have created an Electoral College tie and thrown the election into Congress.
Using such an opaque and controversial means of picking the president would hardly bring Americans back together after a hotly contested national election. It is a mistake to assume that our nation’s relative stability is founded on our current Electoral College rules – just as our stability did not depend on indirect election of Senators or denial of women’s suffrage prior to our amending the Constitution just a century ago.
No. The average American does not, as a rule, pay much attention to the Electoral College, barely understanding how it functions and generally not liking it. Gubernatorial races are instructive: winners receive mandates, period, with big winners getting bigger mandates. Because he only won 43% of the national popular vote that year, Bill Clinton’s large Electoral College margin in 1992 hardly gave him a resounding mandate. His first two years, in fact, were marked by rocky debates over health care, energy and urban development in 1993-1994. Regardless, a president who wins a razor-thin national popular vote majority, but a lopsided electoral vote count should not be encouraged to run roughshod over the large minority who voted the other way.
The Electoral College has a neutral impact on third party candidates. The Electoral College is not the reason why third party candidates and independents have such low odds in presidential elections: rather it is because only one candidate wins, which favors candidates who can draw on large numbers of votes. Today, every sitting governor is a member of a major party, and in nearly 1,000 gubernatorial elections in all 50 states since World War II, no governor has won with less than 35% of the vote. In 1992, Ross Perot did not win any electoral votes despite earning 19% of the national popular vote, but he would have easily won the presidency and a majority in the Electoral College by doubling his share of the national vote to 38%.
Support for a national popular for president is strong among Republican voters, Democratic voters and independents. In the past four decades it has won support from hundreds of leaders of both major parties, from Gerald Ford, George Herbert Walker Bush, Richard Nixon and Bob Dole to Birch Bayh, George McGovern, Jimmy Carter, John McCain, and Barack Obama.
The controversial 2000 presidential election won by Republican George W. Bush despite his loss in the national popular vote has inspired more Republican leaders to defend the Electoral College and more Democratic leaders to oppose it. But the fact is, a national popular vote provides a level playing field that is fair to both major parties, both of which have shown an equal ability to carry the national popular vote in elections over the short, medium and long-term. For example, Democrats and Republicans have split the national popular vote in the past eight presidential elections back to 1980, past 12 presidential elections back to 1964, past 16 elections back to 1948, past 22 elections back to 1920 and past 26 elections back to 1904. Democrat Lyndon Johnson won a large popular vote landslide in 1964, but so did Republicans Richard Nixon in 1972 and Ronald Reagan in 1984.
No. The United States is a large nation. California, New York and Texas are our most populous states, but even if a candidate (quite unrealistically) won every single vote cast in those states, he or she would still have barely 25% of the vote. Indeed, a candidate could win every single vote cast in the ten biggest states and still not have a majority of the national popular vote. Because big states are geographically dispersed and every vote is equal, any candidate trying to win a majority of national vote must try to win votes everywhere, and volunteers excited by that candidacy have every incentive to be active right in their own neighborhood, knowing that any new vote cast for their candidate will count the same as any vote cast anywhere else. In contrast, it is at least statistically possible under the current system that a candidate could win the presidency by stringing together wins in all the largest states while doing very poorly everywhere else.
The great majority of Americans live in metropolitan regions where events and advertising in a local television market will reach them. But it would be folly to ignore the remaining voters, as can be shown by examining closely contested campaigns in states. For example, John Kerry won the urban areas in Ohio in 2004 and put great attention in those areas, but George Bush’s campaign mobilized many voters in the rural and “ex-urban” counties that led to his statewide victory.
In 2009 FairVote analyzed what would have happened in the 2008 election if voter turnout had risen generally by 10% in the 11 most populous states. It also analyzed the impact of only Democrat Barack Obama’s vote total rising by 10% in those states. In the first case, Obama would have needed to carry 50.15% of the popular vote in the rest of the country to maintain his same overall share of the national popular vote – down just 0.22% from the 50.37% he actually won in those states. In the second, highly unrealistic case of only his vote rising by 10% in the large popular states, he still could have afforded to have his share in the remaining states decline by only 3.16%.
The bottom line: When every vote counts the same, every vote matters. You can’t write off voters and expect to win the presidency.
For more information, the National Popular Vote website (www.nationalpopularvote.com) has an excellent and in-depth section discussing the myths surrounding NPV and the Electoral College. Also, please refer to FairVote's other sections on the National Popular Vote. Questions can be directed to info(at)fairvote.org.