A winning blueprint

After a January fright, Deval Patrick recoups South Coast votes to win second term

Fairhaven marked a point of no return for Charlie Baker early on election night.  The South Coast town, with a September unemployment rate of 10.2 percent, went narrowly for Republican Scott Brown in the special US Senate election that rocked the nation in January. But it was one of the first communities to report its tally on Nov. 2, and it was solidly in Deval Patrick’s column, giving him a 20-point margin over GOP nominee Baker. (Patrick might have been helped by Democratic US Rep. Barney Frank’s push for re-election here; Frank won Fairhaven by a similar margin.)

Having fallen behind Brown in a blue-collar community – the headquarters of golf equipment manufacturer Titleist – that had seemed to be cooling toward Boston-area Democrats, Baker needed to compensate in Boston’s white-collar suburbs. He didn’t, and the result was a 48 percent to 42 percent win for Patrick, with independent candidate Tim Cahill at 8 percent and Green Party nominee Jill Stein at 1.4 percent. For complete town-level results, click here.  

It was a decisive margin for Patrick but not a geographically wide one. He carried only five of the 10 regions in our Mapping Massachusetts project. But his margin in Bigger Boston (72-21), despite a disappointing turnout in that region, was far more lopsided than anything Baker could muster elsewhere. The best Baker could do was a 53-39 win in Offramps, which takes in the bedroom communities along Route 495. This pattern matches national political trends; Democrats this year tended to run best in states with heavily urban voting populations (California, Colorado, Nevada, New York) but got “shellacked,” as President Obama put it, almost across the board in suburban and rural areas.


Conventional wisdom is that election are decided in the middle, but in many ways Patrick won at the extremes. He carried Boston and the places farthest from Boston, skipping over losses in the Middle Mass and Offramps regions. He won the most densely populated parts of the state, in and near Boston, and also the most sparsely populated – including Berkshire County, where he maintains an estate in addition to his home in Milton. But he lost the suburban-sprawl area that forms a big C from Cape Ann to Plymouth County. In general, he won the most affluent and well-educated towns, and he carried the poorest areas and those with the lowest education achievement, with Baker taking most of the demographically unremarkable communities. And Patrick won the places that have come through the Great Recession with relative calm (Boston and the high-tech corridor to the west of the city) and also the parts of the state that have fared the worst (including most of the Gateway Cities.)

In all, Patrick carried only 148 of the state’s 351 cities and towns, but his margins of victory were heftier than Baker’s – led by a 76,000-vote surplus in Boston that accounted for more than half of his statewide margin. As the chart below shows, Patrick won seven cities and towns by more than 10,000 votes each. Baker’s votes were much more spread out. His best margin – 3,154 votes – was in Billerica, the middle-class town north of Boston whose economic anxieties have been chronicled in CommonWealth magazine. But Brown bested Democrat Martha Coakley by a heftier 4,611 votes in Billerica in January, and Brown’s biggest margin was in the South Shore town of Weymouth, where he was almost 7,000 votes ahead.

The chart below shows Patrick’s strongest and weakest communities, both by his percentage of the vote and by his victory (or loss) margins. He won by landslide percentages in gay-friendly Provincetown, lefty Cambridge, and several communities in the far west – including Richmond, where he has a second home. But he racked up the  biggest vote margins in and near Boston.

Patrick’s biggest weak spots were in bedroom communities with long commutes to Boston. Interestingly, some of Baker’s biggest surpluses came in towns that had strong population gains in the ’80s and ’90s but have since seen less robust growth: Dracut, Methuen, North Andover, and Tewskbury, all near the New Hampshire border, and Barnstable (which is now actually losing population) on Cape Cod. It seems that hard times have made urban areas even more strongly Democratic, but there’s still a lot of resistance to the party in many suburbs now experiencing economic anxiety.

As for independent candidate Tim Cahill, he unsurprisingly did best in his backyard (Quincy) but showed some strength in some western communities. If all of his vote went to Baker, a sixth region (Post-Industria) and the state itself would have gone to the Republican, but some polls suggested that much of Cahill’s support would have gone to Patrick if the independent weren’t in the race.

No second wave for GOP: The limits of the Scott Brown coalition

Scott Brown’s election to the US Senate was a huge political event, but like most victories, it has become more exaggerated with each retelling. He won by only 4.8 points, so Baker couldn’t afford to underperform him by much in November. Similarly, Patrick’s re-election will go into the history books as an impressive comeback, but he still fell short of getting 50 percent. In other words, there wasn’t a huge change in voter preferences between January and November, and both elections indicated a pretty evenly split electorate. Even turnout was almost identical, going up only slightly from 2.25 million to 2.29 million.

But there were some signficant shifts in November. Patrick took two of the Mapping Massachusetts regions back from Brown’s column: Brink Cities and Post-Industria. And 29 communities flipped from Brown to Patrick. The largest were Lowell, Quincy, and Waltham, but the list also included several municipalities in high-unemployment Bristol County; in addition to Fairhaven, the swing towns included Taunton, Dartmouth, Somerset, Westport, Swansea, and Acushnet. (Only one town in the state swung from Coakley to Baker: the hamlet of Monroe, on the Vermont border.)

The chart below shows the biggest swings from January to November, in terms of percentages and raw votes. In much of the South Shore, GOP margins shrunk or vanished because of Cahill’s presence on the ballot, but Patrick did run slightly ahead of Coakley even in conservative strongholds like Weymouth (up 3 points, to 37 percent) and Plymouth (up 3 points, to 39 percent).

It’s also notable that in several western towns where Patrick ran behind Coakley, he apparently lost votes to Stein rather than Baker. Stein’s best showing was 12 percent in Franklin County’s Wendell; behind that, she got 8 percent in Hawley, Cummington, and Worthington. Northampton was the biggest community where she topped 5 percent. Not surprisingly, though, the biggest vote swing against Patrick was in Baker’s hometown of Swampscott.


Note: “Vote swing” refers to the change in raw-vote margins between the top two parties from one election to another. For example, if Coakley won a town by 5,000 votes in January and Patrick won it by 20,000 votes in November, the vote swing would be 15,000. However, if Patrick lost the town by 5,000 votes in November, the vote swing would be -10,000.

Also of note: Patrick ran 1.4 percentage points ahead of Coakley’s showing statewide, but that figure was generally higher in communities with high unemployment rates. Patrick’s “I care” demeanor may have helped him in these cities and towns against Baker’s promise of a more austere approach to state government. The chart below lists the 10 communities with the highest unemployment rates as of September, among those that cast at least 1,000 votes in the general election. Patrick ran behind his statewide showing and Coakely’s showing only in Millville, a small town in Worcester County near the Rhode Island border.

Baker’s emphasis on making government more efficient – and smaller – seemed tailor-made for the suburbs where many residents are in the high-tech and financial services industries. His liberal stands on abortion and gay marriage (and his selection of gay state Sen. Richard Tisei as a running mate) were signals that Baker intended to govern much as the man he worked for in the early ’90s – Republican Gov. Bill Weld, a quintessential social liberal/economic conservative.

But the decidedly un-Tea-Party approach to social issues was either a mistake or a smart move the Baker campaign failed to take advantage of. By September, Baker was tacking to the right, slamming Patrick for being “immensely soft” on crime because the governor hadn’t signed on to a federal pilot program that would check the immigration status of arrested people. That tactic may be one reason Baker fell behind Brown in New Bedford and other cities with large numbers of foreign-born citizens. At the same time, he made little or no headway in the socially liberal suburbs, such as Acton and Lexington, that were wary of Tea Party hero Brown in January.

The economy counts: Patrick’s weaker mandate for a second term

That’s not to say that the past four years hasn’t taken a toll on Patrick. His percentage of the vote fell 6.6 points from his first victory, and the decline was especially steep in Middle Mass., home of Lt. Gov. (and former Worcester mayor) Tim Murray. Patrick slid from 53.4 percent to 41.4 percent in that region, though he still did better than Coakley’s 37.2 percent.

Worcester County, which makes up most of the Middle Mass. region, does not have the double-digit unemployment rate of the South Coast’s Bristol County. In September, unemployment was at 8.9 percent in Worcester County, compared with 10.1 percent in Bristol County (and only 6.8 percent in Middlesex County). But the central part of the state does have a high foreclosure rate: 1 in 504 homes in Worcester County were in foreclosure as of September, a bit worse than the 1 in 521 homes in Bristol County (and much worse than the 1 in 946 homes in Middlesex County.) As noted above, urban areas with high unemployment rates tended to stick with Patrick, but suburbs where the housing crash has been the bigger problem seem to have been less patient with the Democrats.

The chart below shows the biggest changes since Patrick’s 20-point win (55-35) over Kerry Healey in 2006. Outside of Lawrence, most of the places where Patrick actually did better this time around are well-educated and properous towns, in the suburbs and on Cape Cod, that have been trending Democratic for many years. Meanwhile, Patrick’s margin shrunk slightly in some major cities like Boston, but proportionally larger declines occurred in Brown-friendly middle-class suburbs such as Plymouth and Peabody.


The new Democratic base: Stealing from Romney’s column

Patrick may have slowed his party’s slide in blue-collar communities after Brown’s success there in January, but his re-election doesn’t change the Democrats’ long-term shift from a labor-union-based party to one more dependent on upscale, white-collar votes. The chart below shows the biggest changes in the Democratic vote from 2002, when Shannon O’Brien was defeated by Republican Mitt Romney, to this year.

Some of the biggest Democratic gains over this period came in some of the best-educated communities. According to the last complete federal Census, in 2000, Carlisle is the best-educated town in the state, with 83 percent of its residents over 24 holding college degrees, and it flipped from 55 percent for Romney, running as a “good government” reformist Republican, to 53 percent for Patrick this year. Acton, Brookline, and Concord – all among the 10 best-educated communities in the state – also posted big gains for the Democrats.

On the flip side, several of the 25 least-educated communities in the state – Acushnet, Chicopee, Fall River, Ludlow, New Bedford, Palmer, and Ware, all with college graduation rates of less than 15 percent – have become noticeably less supportive of Democrats over that time. This is an ironic development, given that Democratic candidates enjoy the support of teachers’ unions and attack their GOP rivals as being neglectful of the state’s education system.

The willing taxpayers: The defeat of Question 3

Democratic candidates swept the ballot, but there is another contest worth examining, and that’s the defeat of a proposal to cut the state’s sales tax by more than half, from 6.25 to 3 percent. Question 3 prevailed in just one political region, Cranberry Country, as seen on the chart below.

The measure was crushed in the usual lineup of liberal strongholds, as can be seen on the map and chart below. Among the places where Question 3 fared best (that is, where the “No” vote was weakest) were several communities that border or are just one town away from New Hampshire, which has no sales tax and has long attracted shoppers from the Bay State; they included Dracut, Haverhill, Methuen, North Andover, and Tewksbury. But Question 3 also did well in towns on or near the border with Rhode Island, which has an even higher sales tax of 7 percent; they included Douglas, Sutton, and Uxbridge.

Though Amherst and Cambridge predictably came down hard against Question 3, the state’s most liberal communities are not the cause of the proposal’s 14-point loss statewide. As the chart below shows, the “no” percentages in many of these communities actually fell behind Patrick’s showing in the gubernatorial race, though that was partly because of voters not voting at all on the question. (In Boston, for example, 10,145 people voted for a gubernatorial candidate but blanked Question 3.)

But in a large swath of suburbs, the “no” vote was far stronger than support for Patrick’s re-election. In the Merrimack Valley town of Chelmsford, which gave Baker a margin of 2,321 votes, the “no” side had a winning margin of 675 votes. In North Reading, a 58 percent vote for Baker became 54 percent vote against Question 3.

Meet the Author

In fact, Baker opposed Question 3, which seems like smart politics because it was too extreme for much of his core constituency. But it’s possible that the Republican Party’s association with strong anti-tax stands hurt its candidates in some suburbs struggling to fund school systems and staff police departments.