Scott Brown’s victory over Martha Coakley in the January 2010 Massachusetts special election sent shock waves through the electoral landscape. Democrats in the Senate lost the 60th seat, imperiling their carefully crafted deal on health insurance reform and requiring serious reflection on what went wrong and how to correct it.

On the heels of Republican wins in the Virginia and New Jersey gubernatorial races, the Massachusetts outcome has been widely taken to portend a difficult midterm election for the Democratic Party in November. Republican commentators have spun Brown’s victory as popular repudiation of President Obama’s policy agenda, especially the health bill. Meanwhile, Democratic commentators have blamed Coakley’s lackluster campaign effort, and hope to avert disaster by energizing ethnic-minority communities and organized labor.


Which explanation is right? A large swing toward Republicans (with Obama voters supporting Brown), or a low turnout among Democrats? How different is the electoral environment in 2010 from that of 2008, and what does that tell us about prospects for November and 2012?

In Massachusetts no statewide exit poll was conducted, and most public opinion polls focused on the horse race, not the final results. Aggregate election returns, then, broken down by municipalities and precincts, provide the best perspective from which to judge the outcome. They provide ample evidence that both views are right.

Scott Brown’s candidacy was buoyed by a surge across the board—in every town. There was also a disproportionate drop in turnout in core Democratic areas—those with larger minority populations, those where support for 2008 Republican presidential nominee John McCain was low, and those with lower average income.

The Massachusetts election, along with those in Virginia and New Jersey, does point to rising Republican electoral fortunes. But the electoral patterns in Massachusetts cities and towns indicate that this can be offset if Democratic campaigns can energize their base and maintain relatively high minority-turnout rates.

• • •

The best evidence that Brown’s election represents a “game changer” comes from the magnitude and near-uniformity of his statewide support when compared to the most recent Republican candidates for statewide election. Brown won 52 percent of the votes cast on January 19. McCain received 36 percent of the vote in 2008, and Kerry Healey, the Republican nominee for governor in 2006, received 35 percent. This represents a sixteen-point shift statewide in the Republicans’ favor.

Not only did Brown significantly outperform recent top-of-the-ticket Republicans in the statewide vote share, he did better on a town-by-town basis, too. Brown’s percentage of the vote exceeded McCain’s in every one of the 351 cities and towns in Massachusetts, including liberal enclaves such as Cambridge, Provincetown, Amherst, and Northampton, each of which gave Brown 4-6 percent more of the vote than it did McCain. Every precinct in the city of Boston saw a net shift toward the Republican candidate, compared to 2008.

Brown’s victory echoed Republican successes of the 1980s. He did especially well in the older industrial towns in the middle of the state—places such as Athol, Fitchburg, and Leominster. These were the sorts of Democratic-leaning, working-class areas that embraced Ronald Reagan and helped elect Republican governors in the state for the better part of two decades. All of these areas gave comfortable majorities to Barack Obama, but supported Brown with at least 59 percent of the vote.

Brown also did better than Republican predecessors in parts of the state with more white voters. In the 231 towns where whites constitute at least 95 percent of the population, Brown’s margin exceeded McCain’s by seventeen points. The difference was thirteen points in the nineteen towns less than 80 percent white. In the sixteen Boston precincts where whites constitute at least 95 percent of the total, Brown did fifteen points better than McCain. In the 237 precincts less than 85 percent white, Brown did only nine points better.

The election in Massachusetts revealed a real erosion in support for Democrats, not just weak turnout.

So the swing that conservative pundits tout did occur, at least as far as vote share goes. But vote share is only part of the story. Turnout is the other part. Either of two dynamics could have produced Brown’s healthy vote margins: first, a fixed population of voters could have shifted toward the party of Brown, away from the party of Obama. Second, energized Republicans could have come to the polls, while dispirited Democrats stayed home. Without polling results from the same individuals across time, we will never know the answer for sure.

The figures are undoubtedly striking: In 2008 Obama received 1,904,097 votes; in 2010 Coakley received 1,058,682. The Democratic candidate received 845,415 votes fewer in 2010. On the other hand, McCain received 1,108,854 votes in 2008, while Brown received 1,168,107 votes in January, a gain of 59,253 votes. Coakley did worse than Obama in every one of Massachusetts’s 351 cities and towns. Brown received more raw votes than McCain in 279 of them.

This does not mean that all of McCain’s supporters—and then some—turned out for Brown, while 845,000 Obama supporters sat on their hands when Coakley appealed for their votes. We don’t know which individuals voted, or for whom. Rather, the aggregated turnout patterns indicate clearly that Brown’s “ground game” of delivering supporters to the polls was a first-class effort, far outstripping his competitor’s.

As with vote share, Brown’s favorable turnout compared to McCain’s was more pronounced in certain places. The shift in turnout was even greater than in vote share, with turnout strongest (compared to 2008) in cities and towns with greater previous Republican support, higher incomes, and more white residents.

The accompanying figure illustrates this difference as relates to Republican support. The vertical axis represents the percentage drop in turnout in a particular city or town, compared to the 2008 presidential election. The horizontal axis represents the percentage of the vote received by McCain in that location. To help us give more weight to larger cities and towns, the data tokens are scaled in proportion to population; so that Boston does not overwhelm the image, we have broken down its wards and included them as if they were separate municipalities.

The pattern is eye-popping—one is tempted to conclude that the results of the special election were primarily determined by the demobilization of the Democratic base: poor, minority voters in liberal parts of the state. It appears that the Democratic core lacked the enthusiasm for its standard bearer that Republicans had for theirs.

But turnout alone is not the explanation. Some simple calculations suggest that even if more of the voters who had given Obama such a large Bay State majority in the 2008 had shown up to vote in January 2010, Brown would have won easily.

Here is one counter-factual. What if the special-election turnout in the poorest half of Massachusetts municipalities had been equal to the turnout in the richest half? In order to focus on the effect of turnout on the election outcome, let us keep the vote shares in these communities unchanged: 53 percent for Coakley, 47 percent for Brown. If the poorest cities and towns had 7 percent more voters in the special election—to match the richer cities and towns—turnout in these communities would have increased by about 77,600 voters. In this scenario, Coakley would have closed the gap by only 4,700 votes—a far cry from Brown’s hundred thousand-vote margin.

What if turnout had mirrored 2008, when Obama was on the top of the ticket and Democratic voters were highly mobilized? Assume that Coakley’s and Brown’s town-by-town turnouts are unchanged, but that each town’s turnout equaled the 2008 figures. Brown would have won just under 51 percent of the vote and Coakley just over 48 percent.

A little over a year ago we presented in this magazine an analysis showing that Obama’s victory owed in large part to a surge in turnout and electoral support among minority voters. These core Democratic groups—blacks and Hispanics—were pivotal both nationwide and in the swing states.

The dynamic that emerged in the 2010 Massachusetts special election is quite different. The trouble for Democrats revealed in Massachusetts is that their loss appears to reflect a real erosion in support. The loss cannot be pinned only on the fact that core Democratic supporters stayed home. Even in the most democratic areas, there was a substantial swing toward the Republicans. One may blame the defeat on the candidate or on national politics. There is little consolation either way. Democrats appear to be lagging in the struggle to recruit challengers for the 2010 election, and support for Obama and Democratic Party identification have fallen. Democrats must worry that electoral forces that brought them to power in 2006 and 2008 have ebbed. Massachusetts is an early warning. The national party needs to take heed.