beginning in the 1930s, although the paper did not use that term (or another, “battleground state”) with much frequency until the late 20th century.2
In 1960, the Times referred to Minnesota as a bellwether state,3 and in 1976 R. W. Apple Jr., in a preelection guide on how to follow the results, cited Illinois as a bellwether that had voted for the winner in every election since 1920.4 The Land of Lincoln backed the loser, Republican President Gerald Ford, over Democrat Jimmy Carter. Later years saw states like Pennsylvania, Oregon, California, Colorado, and, increasingly, Ohio receive the designation from the Times. For at least the past few cycles, “swing” has appeared far more times than “bellwether” to describe competitive general election states.
According to Presidential Swing States: Why Only Ten Matter, an analysis of the competitive states in the Electoral College, the proliferation of “swing state” is a relatively recent one. “The phenomen[on] of a state being labeled a swing state is largely a product of the media and recent campaign invention,” Stacey Hunter Hecht and David Schultz argue.5
Bellwethers? Swing states? Are these the same thing? Sometimes, but not always.
DEFINING THE BELLWETHER STATE
Whether one refers to a typically competitive state in the Electoral College as a bellwether or a swing state, both seemingly have the same definition: a state that both is competitive in a close presidential election and reflects the national voting in a given election. But these terms are not really interchangeable.
Just because a state is close in an election doesn’t make it a bellwether. For instance, Missouri achieved a reputation as perhaps the nation’s most notable swing state throughout the 20th and into the 21st century, voting for the national winner in all but one election from 1904 to 2004, so the New York Times was right to flag it in 1948. The single time it voted with the loser was when it backed Democrat Adlai Stevenson over Republican President Dwight Eisenhower in 1956, picking the challenger over the incumbent by about two-tenths of a percentage point, making it the closest state in that election.
Missouri was a swing state in 1956, meaning that either side could have won it, but it was not a bellwether, because it was not representative of the national results. Not only did it vote for the losing candidate, but it did so in an election when the winner, Eisenhower, won nationally by 15 percentage points and captured 86 percent of the electoral votes. Four years later, Missouri was again a swing state—Democrat John F. Kennedy narrowly won it—but it was also a bellwether, because Kennedy’s margin there was about the same as his national margin: he won both by less than a point.
Here, then, are definitions of both terms.
• Bellwether states reflect the national voting not only in close elections, but also in blowouts.
• Swing states can be won by either side in an election.
In competitive contests, bellwethers and swings are often the same states. In noncompetitive elections, they probably won’t be.
The analogy of a wave is a common one in politics, with big ones indicating smashing victories for one side or the other.6 Sean Trende of RealClearPolitics, a political news analysis and aggregation website, has a useful analogy to describe how this works in the Electoral College:
Think, if you will, of a 51-rung ladder descending into a tidal pool. At the bottom of the ladder is the most Democratic state in the country. . . . At the top of the ladder is the most Republican state in the country. . . . The water represents the Democratic tide, driven by national forces such as the economy, presidential popularity, and so forth. As the tide rises, increasingly red states cast their ballots for the Democratic candidates. As it falls, blue states begin to turn crimson.7
When that 51-rung ladder is assembled every four years—one for each of the 50 states and another for the District of Columbia—Ohio is typically right near the middle, winnable by either side so long as the national election is close. If a party’s tide rises high enough to cover Ohio, history suggests that that is almost always sufficient to win the election. If the party’s tide does not rise to Ohio’s level, then that party’s candidate almost always loses.
In close elections, Ohio is a swing state, but it generally is not when national elections are blowouts. For the most part, its voting patterns reflect those of the nation in elections decided either by few votes or by many votes. It’s a state that can both decide the winner in a close election and reflect the nation’s movement in a blowout. It moves the way the nation moves, and it has for quite some time.
WHEN A SINGLE STATE DECIDES
The first presidential election in which there was widespread participation by average citizens—average white male ones, that is—was 1828, the year of Andrew Jackson’s sweeping victory over incumbent John Quincy Adams. This was arguably “the first truly ‘democratic’ election, in that eligible voters participated to a degree not seen before.”8 It also saw the emergence of a true two-party system, even if it would take another three decades before both modern parties, the Democrats and the Republicans, would actually emerge (1856 is the first year both parties would face off in a presidential election, as they have in every presidential election since).
If the election of 1828 was “the birth of modern politics,” as Lynn Hudson Parsons argued in his breakdown of the Jackson-Adams clash,9 the first modern elections were not particularly close in the Electoral College. Take any state and its electoral votes away from any of the winners in 1828, 1832, 1836, and 1840, and that candidate still would have had enough electoral votes to win.
The first true electoral nail-biter, then, was 1844. While Democrat James K. Polk won with 170 of 275 possible electoral votes to Whig Henry Clay’s 105, the seemingly lopsided total is deceptive. New York, with its 36 electoral votes, voted for Polk by just a single percentage point, a difference of about 5,000 votes of close to 500,000 cast. Flip New York to Clay and, leaving all else equal, he would have been elected. “As stunned Whigs surveyed the wreckage, they quickly saw that the key to the election had in fact been New York,” although “Clay, the savvy head counter, had known all along that New York was the key to the contest.”10
That election marked the first of 10 times in the modern era (beginning in 1828) where the result in a single, closely contested state made the difference in the election. Those other elections are 1848, 1876, 1880, 1884, 1888, 1916, 1976, 2000, and 2004. In each of those elections, if the loser had won just one state he lost by a close margin (less than five points), he would have won the election. Additionally, in 1968, if Democrat Hubert Humphrey had won California and its 40 electoral votes (he lost by about three points to Republican Richard Nixon), he and George Wallace, the former Democratic governor of Alabama who ran for president as an independent conservative, would have combined to deny Nixon an Electoral College majority, throwing the election into the House of Representatives, something that has not happened since 1824. Had that happened, Humphrey likely would have won, because Democrats controlled 26 of 50 US House delegations. (When a presidential election goes to the House, each state gets a single vote, which presumably would go to the candidate whose party controlled that state’s House delegation.)
So there’s a long history of a single state making the difference in a close election. The nature of the United States’ Electoral College allows for the possibility that a single state, closely fought, can determine the victor.
In a year with a big wave, states that ordinarily vote for one party will often vote for the other. States that aren’t usually competitive turn into swing states, but that doesn’t mean their results will look like the nation’s.
Minnesota has the longest streak of voting Democratic for president of any state in the country (the District of Columbia’s stretches back even further, but DC is not a state, to the chagrin of its residents). The North Star State has voted for a Democrat in every election since 1976, withstanding even incumbent Ronald Reagan’s 49-state reelection tide in 1984. The state instead opted that year for former Vice President Walter Mondale, who had also served as a senator from Minnesota, by less than 4,000 votes. Reagan won 59 percent to 41 percent