Shifting Ground

Politically, Massachusetts is really 10 states, not one. And the borders keep moving.

 Illustrations By Travis Foster

Four years ago, Mitt Romney beat Shannon O’Brien to become governor of Massachusetts. Another way to look at it is that the town of Westwood, where the Republican percentage of the vote increased by 10 points between 1998 and 2002, beat the city of New Bedford, where the Republican share dropped by 17 points over the same period.

Or, based on CommonWealth’s new map of political regions in the Bay State, Shopper’s World (which includes Westwood) triumphed over the Brink Cities (which includes New Bedford) in 2002. By contrast, in 1998 Shopper’s World swung toward Democrat Scott Harshbarger for governor but was thwarted by the shift in the Brink Cities toward Republican Paul Cellucci. Every region of the state, it seems, gets its turn at influencing the outcome of major elections in Massachusetts. And with five first-time gubernatorial candidates, including an independent, in this year’s race, no region can be sure of being on the winner’s side.

Our 10 political regions are variations on those presented four years ago (see “Lay of the Land,” CW, Summer ’02), with adjustments made on the basis of how municipalities voted that fall, and also on the number of voters who participated in that election. Each region represented roughly one-tenth of the electorate in 2002, or between 213,000 and 227,000 votes (including blanks). No one has won statewide office in Massachusetts in the past 25 years without carrying at least five of these regions.

The regions explain how the Republicans have been so successful in tight races, where geographical differences can be crucial. GOP gubernatorial candidates Bill Weld in 1990, Paul Cellucci in 1998, and Mitt Romney in 2002 each won at least six regions, but each won with a somewhat different set, suggesting a nimbleness on the part of Republicans in finding where their voters are—or a failure on the Democratic candidates’ part to master the state’s political geography.

Four regions have voted Republican in all three races: Cranberry Country, which includes most of the South Shore and Cape Cod; Offramps, which includes exurbs along Interstate 495; Ponkapoag, which includes most of Norfolk County and other suburbs to the south of Boston; and Stables and Subdivisions, which includes Cape Ann and most of Essex County. In 1990, Weld added two more regions to his column, presumably by running to the left of Democrat John Silber on social issues: Shopper’s World, which includes suburbs to the immediate west of Boston; and Left Fields, which combines the liberal strongholds of Cambridge and Somerville with much of western Massachusetts and left-leaning parts of Cape Cod and the Islands. Those two regions abandoned Cellucci in 1998 for Democrat Scott Harshbarger, but he compensated by picking up: Post-Industria, which includes both former mill cities and newer exurbs from the Merrimack Valley to Lynn; and MidMass, which includes most of Worcester County. Finally, in 2002 Romney faced a continuing slide in the GOP vote in Left Fields but made up for it by snatching back Shopper’s World to win statewide over Democrat Shannon O’Brien.

Only two regions stuck with all three Democratic candidates: Bigger Boston, which includes the state’s biggest city and four nearby suburbs; and the Brink Cities, which combines the Fall River–New Bedford and Springfield areas. These two mostly urban regions have been on the winning side in only one hotly contested race during the 1990s, when US Sen. John Kerry, beating back a challenge from Weld, also carried Shopper’s World, Left Fields, Post-Industria, and Ponkapoag.



While winning six regions, Kerry beat Weld by 7.3 percentage points, a wider margin than many had expected. Indeed, Democrats in Massachusetts seem to win big or not at all. The five closest statewide elections over the past 25 years have all gone to GOP candidates: Ronald Reagan in the 1980 and 1984 presidential elections (winning by 0.1 point and 2.7 points, respectively); and governors Weld, Cellucci, and Romney (winning by 3.1 points, 3.4 points, and 4.8 points).

The Republican Party’s ability to come out ahead in the closest races in Massachusetts may seem odd, given how easily the Democrats win almost every other election. Outside of gubernatorial elections and Joe Malone’s two easy wins as state treasurer in the early 1990s, the GOP hasn’t come within five points of winning any statewide office since 1974. And the 2004 elections didn’t provide much cheer for the Republicans. President George W. Bush carried only 45 out of 351 municipalities in Massachusetts, his worst state, and all 10 political regions went for Democratic nominee Kerry—all but Cranberry Country by more than 10 points. At the same time, not a single GOP congressional candidate got more than 34 percent of the vote, and the party lost seats in the Legislature despite fielding more candidates than it had in a decade.

So why the exception for gubernatorial races? Perhaps the lack of competition for downticket races gives Democrats a false sense of security and deprives some of their candidates of practice in tough campaigns before they aim for the top office. Or perhaps the answer is another way in which Massachusetts is something of an outlier: Forty-nine percent of the electorate is not enrolled in either major party, behind only Alaska and New Jersey among the 27 states that keep voter registration figures by party. In addition, several independent presidential candidates, including John Anderson in 1980, Ross Perot in 1992 and 1996, and Ralph Nader in 2000, have counted Massachusetts among their best states. (They averaged 13.3 percent over those four elections, a figure that’s higher only in the much smaller states of Maine, Vermont, Alaska, Rhode Island, and Montana.) Even in the close gubernatorial race of 2002, slightly more than 5 percent of the electorate voted for candidates other than the Democrat and the Republican. All of this means that more of the state’s electorate is up for grabs than Massachusetts’s all-Democratic congressional delegation might suggest.

Another explanation for the Democratic Party’s difficulty in winning the governor’s office is that “true blue” Massachusetts is really not so monochromatic. When Democrats are able to run here on broad national themes—or, in the case of some lower-level offices, to avoid divisive issues completely—the various shades of blue seem to run together. But when voters’ passions are stirred, on issues from government spending to crime, it becomes clear that the overwhelming majorities for Democrats such as Bill Clinton and Ted Kennedy are not automatically transferred to more specific liberal causes.

Just as Republicans have won most of the recent close elections in Massachusetts, the more conservative position has prevailed in five of the six closest referendum battles of the past 25 years, and each time predominantly Democratic or swing regions were pivotal to the outcome. Voters rejected a ban on dog racing in 2000, with opposition strongest in the Post-Industria region; abolished rent control in 1994, a move most popular in Shopper’s World; approved term limits for all state offices in 1994, a wish ignored by the state Legislature; rejected a rather convoluted plan to work toward universal health care in 2000; and, in the same year, rejected changing the state’s drug laws to emphasize addiction treatment. The only nail-biter won by the arguably more liberal side came in 1994, when voters rejected an obscure law that would have made it marginally more difficult for activist organizations (e.g., MassPIRG) to raise fees from college students.

There have been other victories for the liberal side, to be sure, but they’ve mostly come when there was little doubt about the outcome on election night, as when voters passed an increase to the tobacco tax in 1992 or opted to retain a mandatory seat-belt law in 1994. Similarly, nobody really expected voters to endorse a plan to eliminate the state income tax in 2002, which made the eight-point margin of defeat a surprise. If anti-tax forces had raised enough money to make this a truly competitive race, perhaps they could have tipped it toward their side.

Democrats in this state seem to win big or not win at all.

This pattern should give pause to opponents of a constitutional amendment to ban gay marriage (which is on the way to the ballot in 2008, provided it gets the support of 25 percent of the Legislature in two separate sessions). If current public opinion polls, which show majority support for gay marriage, are correct, the amendment may never gain traction, eliminating any suspense in November. But if the vote is close, there is no consistent liberal majority that can be counted on to ensure the amendment’s defeat. Indeed, it’s worth noting that according to, an organization that published the names and addresses of every voter on the petitions submitted to the secretary of state’s office in support of the amendment, the most signatures by far came from the Brink Cities — one of the only two regions that voted for Democratic gubernatorial nominees in 1990, 1998, and 2002. No matter how blue Massachusetts appears to the rest of the country, the electorate often changes its stripes.



Adding to the unpredictability in this year’s gubernatorial race is the independent candidacy of businessman Christy Mihos, a former member of the Massachusetts Turnpike Authority. At this writing, it’s too early to tell whether Mihos will be seen as a viable candidate in November, but the conventional wisdom is that his presence in the race hurts presumptive Republican nominee Kerry Healey, since Mihos was a member of the GOP right up to the day he kicked off his campaign as an independent. If Mihos simply takes votes away from Healey, the Democratic nominee would seem to have a good chance of winning even without toting up any more votes than his losing predecessors did in 1990, 1998, or 2002. Silber got 45.5 percent of the total vote, Harshbarger 46.6 percent, and O’Brien 44.4 percent, but if Mihos gets 10 percent this year, the victory zone may indeed shift downward to the mid 40s.

The problem with this theory is that each of the three Democratic nominees had weaknesses in different geographical areas, so the Democratic vote may not be as stable as it first appears. Silber was weaker than either Harshbarger or O’Brien in Bigger Boston, Left Fields, and Shopper’s World, and he ran at least five points behind both these other candidates in Cambridge, Newton, and Somerville, among other liberal strongholds. Harshbarger was the weakest vote-getter in the Brink Cities, and he ran at least five points weaker than the other two in Fall River, Lawrence, New Bedford, and other blue-collar cities. O’Brien, having lost by a slightly wider margin, was the weakest of the three in six of our regions, but her drop-off was the worst in Stables and Subdivisions, and her unique weak spots among cities and towns included middle-class Barnstable, Peabody, and Woburn.

It is possible that this year’s Democratic nominee could combine the strongest attributes—geographic and otherwise—of all three previous candidates and win in a blowout similar to Michael Dukakis’s 23-point margin in 1982, which would make the Mihos candidacy moot. But he could also combine the particular flaws of just two of the previous nominees and dip below the 40 percent mark, especially if Mihos can carve out a constituency that doesn’t merely cannibalize the Republican base. (Or if Green Party nominee Grace Ross gets a couple of points.)

That’s a big “if” for Mihos, of course. Massachusetts may have been one of Ross Perot’s better states in the 1992 presidential election, but while he came within eight points of upsetting Bush for second place, he was 25 points behind Clinton, who was never in danger of losing the state. Among communities that cast at least 10,000 votes in the last gubernatorial race, Perot ran best in Dracut, where he got 34.6 percent to Clinton’s 35.8 percent and Bush’s 29.1 percent. Ten years later, O’Brien got 35.6 percent in the same community, almost exactly the same as Clinton, suggesting that nearly all the Perot vote fell into Romney’s lap in an essentially two-person race. In Billerica, the Clinton-Bush-Perot split was 38-29-32, and the O’Brien-Romney split was 36-58; and in Plymouth, a 39-31-30 split turned into 36-58. In all, 179 of the state’s 351 cities and towns were carried by Clinton in 1992 (in almost all cases, with less than a majority) and then by Romney in 2002. Only the tiny western community of Rowe went in the other direction, going Republican with Perot in the mix and Democratic in 2002.

If Mihos were to expand his appeal across ideological lines, however, the Democrat might suffer. In recent years, the best showing by a left-of-center third-party candidate was in the 2002 state treasurer’s race, in which the Green Party’s James O’Keefe polled 7.4 percent of the vote. In Cambridge, a 68-22 romp for O’Brien in the governor’s race turned into a 53-21-17 split in the treasurer’s race. In Northampton, the same 68-22 margin turned into a 51-19-23 split, and in Pittsfield a 63-31 split turned into 50-25-12. Democrat Tim Cahill won regardless, thanks to his strong showing on his native turf of Norfolk County, but it’s unlikely that any Democratic nominee for governor could win with such low numbers in these liberal strongholds.



Yet another difference between Massachusetts and the United States is that political geography still seems fluid here, as evidenced by the changing bases of support for each party in recent gubernatorial elections. That’s not the case nationally: Straight-ticket voting has been on the increase in other states, and there were negligible changes in the red-vs.-blue map between the 2000 and 2004 presidential elections. One reason for the lack of deep-set geographic patterns in Massachusetts may be the “big tent” philosophy that has endured here in both parties, but especially among Republicans. In 1978, when Democrat Ed King faced Republican Frank Hatch, and in 1990, when Silber faced Weld, the GOP candidate was the more liberal on social issues, and partisan leanings were poor predictors of how communities broke in November. Similarly, in the 2002 treasurer’s race, the Republican candidate was Dan Grabauskas, an openly gay, “good government” type who made inroads into liberal-leaning suburbs.

Yet the 2002 gubernatorial race may have brought Massachusetts a bit closer to the red-blue division seen at the national level. Republican nominee Romney embraced the conservative label to a greater degree than his predecessors did (though also committing not to push for more restrictive abortion laws here), and the result was that urban areas became more Democratic and fast-growing exurbs became more Republican, mirroring national trends. The question for 2006 is whether our 10 regions settle into a more predictable groove or continue to offer surprises with each new election. Some clues emerge as we take a closer look at the makeup and political behavior of each.


Bigger Boston: The GOP in free fall

Outside of the landslide re-elections of Weld in 1994 and the decisive wins of Treasurer Joe Malone in 1990 and 1994, no Republican has carried the city of Boston in any statewide race for more than three decades. That doesn’t mean the city doesn’t provide any suspense in November: The margin of victory for the Democrat, as well as voter turnout, varies wildly from election to election.

The city of Boston makes up the majority of our Bigger Boston region, joined by the first-generation suburbs of Brookline, Everett, Malden, and Medford. Demographics and socioeconomic characteristics are literally all over the map here, but politically the region has become more and more Democratic. One reason may be that this is the most densely populated region, and density is increasingly correlated with support for candidates in favor of more activist government. (This may be a global phenomenon; in Canada’s recent national elections, the Conservative Party surged overall but did not make noticeable progress in the largest cities.) In terms of party registration, Bigger Boston is the least Republican (9 percent) and least independent (38 percent) region.

Even in victory statewide, Republican performance in Bigger Boston has been slipping. In 1990, Weld got 42 percent of the vote and carried Brookline plus four of Boston’s 22 wards (including Allston, the Back Bay, Beacon Hill, and the North End). In 1998, Cellucci slipped to 39 percent even as he carried Everett, Medford, and five Boston wards (including the North End, Charlestown, East Boston, and both wards in South Boston). Last time out, Romney got only 34 percent and lost every one of the communities in this region, as well as all of the wards in Boston. Bigger Boston also had the biggest jump in total vote from 1998 to 2002, and O’Brien got 23,000 more votes than Harshbarger had. The trouble for the Democrats is that it’s tough to boost turnout here without the Republicans doing the same in their strongholds. (That’s what happened in the high-turnout presidential race of 2004.)

One rule of thumb is that if Bigger Boston is the best region for a Democrat (see Ted Kennedy in 1994, or Kerry in the 2004 presidential election), he or she will be in good shape to win statewide. But if the Democrat runs better in the less partisan but more ideological Left Fields (as O’Brien did in 2002), he or she is in trouble.

Bigger Boston naturally looms large in Democratic primaries, and, as often as not, it hurts the more liberal candidate. In the 2002 gubernatorial primary, Thomas Birmingham won the region by less than two points, but Shannon O’Brien netted her biggest cache of votes here, and she won the city of Boston despite carrying only three wards. Her trick was finishing second everywhere else, while Robert Reich and Birmingham finished third in as many places as they placed first. In the primary for lieutenant governor, this was Chris Gabrieli’s best region in terms of raw votes (but not percentage of the vote), and he beat former state Sen. Lois Pines by an easy 49-29 margin. In the 1998 primary for attorney general, Bigger Boston provided the biggest bundles of votes for both Thomas Reilly and Pines, but Reilly carried the region by four points, close to his winning margin statewide.



Brink Cities: Far from the golden dome

Another essential building block for any Democratic candidate consists of two urban areas outside the Boston media market, and both have long lagged behind the rest of the state economically. One piece is southern Bristol County, which includes Fall River and New Bedford and is closer to Providence than it is to Boston. The other section includes much of Hampden County, including Chicopee, Holyoke, and Springfield; it’s even farther from Boston and looks toward Hartford as the nearest major metropolitan area. The Brink Cities had the smallest jump in voter turnout between 1998 and 2002, and its core cities have all lost population over the past 15 years.

Though it has voted Democratic in every competitive election over the past 30 years, this region is not exactly liberal. It was Reich’s worst region in the 2002 gubernatorial primary, and it has contributed the most signatures for the gay-marriage ban working its way toward the 2008 ballot. (Springfield was responsible for 4,195 signatures to Boston’s 9,183, though Boston has more than four times as many registered voters.) The Brink Cities are also to the right of Bigger Boston on taxation issues: This was the worst region for the 1992 referendum that raised the cigarette tax, and the 2002 proposal to abolish the income tax did better here than in the state as a whole, losing by only a 47-40 margin.

In Democratic primaries, turnout here is not as high as might be expected, given that this is the only region outside of Bigger Boston where registered Democrats outnumber independents. (Members of the two major parties count for 60 percent of the electorate in Fall River, higher than in any other city or town.) The loose ties to the Boston media market may be one reason for this, and for the tendency of the Brink Cities to back veteran candidates over newcomers such as Reich. In 1998, longtime legislator Pines came within one point of beating the less well-known Reilly in the primary for attorney general, even though Pines’s brand of suburban liberalism did not seem to be a good fit here. But four years later, this was Gabrieli’s best region, and he beat Pines 58-21 to become the nominee for lieutenant governor. Gabrieli’s designation as the official running mate for gubernatorial favorite O’Brien seems to have carried a lot of weight here.


Left Fields: The limits of liberalism

The Left Fields region is to Massachusetts what Massachusetts is to the United States. It’s more liberal, more Democratic, better educated, less populated by nuclear families— and often on the losing side of elections. This is our most geographically dispersed region. It consists of three parts: Arlington, Cambridge, and Somerville, which lie to the northwest of Boston; the islands of Martha’s Vineyard and Nantucket along with the Cape Cod communities of Falmouth, Provincetown, Truro, and Wellfleet; and 76 communities in the western third of the state, including the college towns of Amherst and Northampton and the increasingly arts- oriented North Adams and Pittsfield.

This was one of only two regions to vote against an income-tax cut in 2000 (the other was Bigger Boston), and it cast the strongest vote against abolishing the income tax altogether in 2004. Romney got 33 percent of the vote here when he ran against Kennedy for US Senate in 1994, and he got the same percentage when he ran for governor eight years later, making this the only region in the state where he didn’t do better at all.

This was O’Brien’s best region in the 2002 general election (59 percent over Romney), but it didn’t help her as much as it might have, considering that Kerry got 73 percent here in the presidential election two years later. A recurring problem for Democrats here is the weakness of partisan ties. In Bigger Boston, where O’Brien won 58 percent against Romney, 52 percent of the electorate are registered Democrats; in Left Fields, where O’Brien did slightly better, only 42 percent are registered Democrats, meaning that she relied more on independent voters. Some of those independents apparently drifted to Green Party candidate Jill Stein, whose 6 percent kept this from being the most lopsided region in the state. (That distinction instead went to Offramps, which gave Romney a hair below 60 percent of the vote.) In the state treasurer’s race, the Greens did even better: Nominee James O’Keefe got 15 percent, pushing Democrat Tim Cahill down to just under 50 percent. Indeed, though Cahill ran slightly ahead of O’Brien statewide, he finished more than 10 points behind her in the Left Fields communities of Cambridge, Somerville, Amherst, Northampton, and Pittsfield. Those numbers could foreshadow trouble for the Democrats if Mihos broadens his appeal toward the left—or if Healey moves enough to the left on social issues to seem a reasonable facsimile of Bill Weld.

In Democratic primaries, the Left Fields region consistently opts for more liberal candidates. It was the best region for Reich in the 2002 gubernatorial primary, for Pines over Reilly in the 1998 primary for attorney general, and for Frank Bellotti over John Silber in the 1990 gubernatorial primary. Patrick—who has recently built one well-publicized residence in this region, a 24-room vacation home in Richmond—clearly must win big here to secure the gubernatorial nomination.

Shopper’s World: What’s the matter with Carlisle?

If you believe that economic status should determine partisan leanings, Massachusetts is as big a puzzle as Kansas, the relatively low-income state that consistently votes Republican and was examined in Thomas Frank’s best-seller What’s the Matter with Kansas? In a reversal of the paradox that book explored, the affluent Bay State habitually votes for presidential and congressional candidates who support higher taxes, especially on the rich, and who advocate more spending on government programs. Shopper’s World, with the highest median income among our 10 regions, is the most extreme example of a constituency appearing to vote against its economic interests.

Romney’s biggest gains were in the affluent Shopper’s World region.

Shopper’s World, named after the prototypical suburban mall in Framingham, fans out from Boston along Routes 2 and 9, stopping short of I-495. It voted for Kerry by a margin of 66-33 in the 2004 presidential race, and for “big spender” Ted Kennedy over Romney by a 59-41 margin in the 1994 US Senate race. One explanation for Kansas’s voting behavior is that cultural issues, such as abortion, often trump economic issues, and the same might be said for Shopper’s World. After all, this is the region that produced the fewest signatures for a constitutional amendment banning gay marriage. It was also the strongest region for an increase in the state’s tobacco tax in 1992 and for a mandatory seat-belt law in 1994. In 1998, when Democratic gubernatorial nominee Harshbarger was being tarred with the “loony left” label, partly because of ACLU-type positions he had taken as attorney general, he increased the party’s vote in this region from 40 percent to 51 percent.

When purely economic issues come to the fore, however, Shopper’s World isn’t so committed to the left. This is where Romney, sounding warnings about the fiscal consequences of total Democratic control on Beacon Hill, made his strongest gains in 2002—getting an even 50 percent, or four points above Cellucci’s mark in the previous election. As for taxes that don’t involve stigmatized behavior like smoking, Shopper’s World is closer to the middle of the road: It voted 60-40 to reduce the income tax to 5 percent in 2000 (the same as the statewide margin), though it was 53-36 against the more extreme proposal in 2002 to abolish the tax entirely (much wider than the eight-point margin of defeat statewide).

Democratic primary voters here generally lean left and are wary of “insider” candidates. This was state Senate President Birmingham’s worst region in the 2002 primary, even though it abuts his best region (Post-Industria). This was also the only region where Pines beat O’Brien-designated running mate Gabrieli in the primary for lieutenant governor.


Post-Industria: Northern Exposure

A few decades ago, you’d be hard pressed to come up with two adjacent states more unlike each other than Massachusetts and New Hampshire. One was derided as “Taxachusetts,” the other had the libertarian motto “Live Free or Die.” In the 1980 general election, Ronald Reagan got 58 percent of the vote in New Hampshire and 42 percent in the Bay State, one of the biggest gaps between two neighbors; and two Massachusetts natives, Democrat Edward Kennedy and Republican George H.W. Bush, suffered humiliating losses in New Hampshire’s presidential primary. Since then, however, the Granite Curtain has largely disintegrated. Massachusetts voters have taken a few steps toward New Hampshire’s low-tax philosophy, and those in New Hampshire have shifted to the left on social matters such as abortion and gay rights (though not on requiring motorcyclists to wear helmets). The three Massachusetts presidential contenders since Kennedy (Dukakis, Paul Tsongas, and Kerry) have all won the New Hampshire primary, and Kerry even dragged this famously Republican state into the Democratic column in 2004.

The Post-Industria region, which comes within 25 miles of New Hampshire’s biggest city, Manchester, is where our northern neighbor seems to have the most influence within the Bay State. It includes several cities (such as Lowell, Lynn, and Woburn) that once hosted thriving textile or shoemaking industries but are now trying, with varying success, to make the transition to high-tech or “creative class” industries. It also includes bedroom communities, such as Billerica and Chelmsford, whose residents are more likely to do their shopping in tax-free New Hampshire than in parking-space-deprived Boston.

Post-Industria is where the Republicans made the biggest gain between the 1978 and 1990 gubernatorial elections, vaulting from 35 to 47 percent of the vote, and the Democrats have not been able to win it since. Thanks to a New Hampshire–like libertarian streak here, it was the second-worst region for a cigarette-tax increase in 1992 and the worst region for a mandatory seat-belt law in 1994. This is the third-lowest region in Republican registration (only 11 percent of the electorate), but only 45 percent of voters here voted to retain the state’s income tax in 2002, lower than any region outside of Cranberry Country.

At first glance, the success of the Democratic Party in neighboring New Hampshire in 2004 would seem to be a good harbinger for Democrats in this region in 2006. But while Kerry carried the Granite State in the presidential race and Democrat John Lynch was elected governor, both lost Hillsborough and Rockingham counties, which border the Post-Industria region. Instead, their margins of victory came from northern and western counties, which more closely resemble Vermont and western Massachusetts in demographics and political attitudes.

In Democratic primaries, Post-Industria leans toward labor-backed candidates (this was Birmingham’s best region in the 2002 gubernatorial primary) and against the most socially liberal candidates (this was Reilly’s strongest region against Pines in the 1998 primary for attorney general).



MidMass: The Democrats’ fade-out

Worcester County is the Bay State’s equivalent to the American Midwest—or, less charitably, our “flyover country.” It rarely appears in tourism campaigns or in photography books about Massachusetts, not having the obvious charms of the seashore or the Berkshires. But MidMass, the region that includes almost all of Worcester County plus a few adjoining towns, has always been an integral part of the state’s economy. It has some of the last cities with a substantial manufacturing workforce, including Leominster and Southbridge, and it now has several fast-growing bedroom communities, as the Boston metropolitan area sprawls farther and farther to the west. The city of Worcester has also emerged as a major player in the biotech industry and has several colleges and universities that beckon to those turned off by the high cost of living in Boston.

In politics, MidMass also has similarities to the American heartland. More than any other region, it has trended toward the GOP during the past decade. In 1990, Weld lost the region with 46 percent; in the 1998 race, this time featuring a Democrat who was clearly to the left of the Republican, Cellucci got a solid 55 percent. And in 2002, Romney slipped only a bit, to 53 percent. In US Senate races, Kennedy beat Romney here 55-44 in 1994, but two years later Weld beat Kerry 49-46, the biggest such swing in the state.

Another sign of Democratic decline is that MidMass has the state’s largest bloc of unenrolled voters, though it ranks fourth in the percentage of voters (53) who are unenrolled. That discrepancy is explained by the fact that the number of people who actually went to the polls in 2002— our measure for coming up with 10 equal regions—is a lot lower than the number of registered voters. MidMass’s 53 percent voter turnout in 2002 was lower than the state average of 56 percent, and worse than in all but the two urban-dominated regions of Bigger Boston and Brink Cities. It’s not surprising that the cities of Fitchburg, Southbridge, and Worcester were below 50 percent, but such towns as Athol, Charlton, Webster, and Winchendon also fell well below the halfway mark. It’s possible that Mihos will tap into the independent vote (Perot got 24 percent here in 1992) and boost turnout this fall; a new Democratic nominee could also attract new voters, especially if Worcester Mayor Timothy Murray wins the nomination for lieutenant governor and motivates voters to get to the polls for a regional favorite son. But even a boost in turnout might not be able to overcome the Republican Party’s steady gains here.

In 2002, Democratic primary voters here were pretty close to the state as a whole in their preferences, but Reich’s 20 percent (vs. 24 percent statewide) showed that there aren’t many liberal pockets in MidMass.


Ponkapoag: The New Southie

If Boston’s near western suburbs evolved from Beacon Hill and the Back Bay, and its northern suburbs are extensions of Charlestown and East Boston, its southern suburbs see South Boston and parts of Dorchester as the Mother Country. The southern sections of the MBTA’s Red Line— completed in 1980, following years of “white flight” from Boston—connect Southie to Milton, Quincy, and Braintree. And the 9th Congressional District, now represented by South Boston’s Stephen Lynch, follows a narrow path out of the city to include such middle-class suburbs as Norwood and Stoughton, as well as the factory city of Brockton. These are among the 21 communities south of Boston that make up the Ponkapoag region, named for the American Indian tribe that was once centered in present-day Canton and Stoughton (after being pushed out of the Neponset area of Dorchester). Nowadays, Ponkapoag has a negligible American Indian population and is instead probably the most Irish of our 10 regions. According to the Census Bureau, four of the six Bay State communities with the highest percentage of residents with Irish ancestry as of 2000 are here (Milton, Quincy, Norwood, and Braintree).

Ponkapoag stays fairly close to the statewide average in general elections, though it’s moved a bit toward the GOP: Weld ran 0.8 points behind his statewide percentage here in 1990, but Romney bested his statewide showing by 1.9 points in 2002. The region also mirrors the state in terms of party registration. As of 2004, the electorate here was 48 percent unenrolled, 39 percent Democratic, and 12 percent Republican. Occasionally, however, Democratic DNA asserts itself here. In the 2002 treasurer’s race, this was the strongest region for favorite son Tim Cahill of Quincy, who got a whopping 57-33 vote over Republican Dan Grabauskas.

In Democratic primaries, Ponkapoag is usually a bit to the right of the state: Reilly beat Pines by 10 points in the 1998 primary for attorney general, and Gabrieli beat her by 20 points in the 2002 race for lieutenant governor. Overall, the vote for Reich in the 2002 gubernatorial primary was light (20 percent vs. 24 percent statewide), but as a Milton resident, Deval Patrick may be able to improve on that showing.

Interestingly, the only three communities where O’Brien, Reich, Birmingham, and Tolman all came within two points of their statewide percentages were all in the Ponkapoag region: Canton, Hull, and Randolph.



Stables and Subdivisions: Down East

The Stables and Subdivisions region is centered on the “other Cape”—that is, Cape Ann, along with most of Essex County and the towns of Reading and North Reading. It includes both affluent “horse country” towns such as Hamilton and Topsfield and slightly more affordable suburbs such as Danvers and Peabody. Politically, it resembles the state of Maine in its affinity for moderate-to-liberal Republicans. Kerry Healey lives here, in Beverly, and this was her best region against conservative James Rappaport in the 2002 Republican primary for lieutenant governor. Another resident is MBTA general manager Dan Grabauskas, of Ipswich, who counted this as his best region in his bid to become state treasurer in 2002, both in the Republican primary and the general election. It’s easy to imagine Republicans in the mold of Maine’s two US senators, Olympia Snowe and Susan Collins, doing well here. (Another moderate Republican, Peter Torkildsen, was elected to Congress in the early 1990s from a district that covered almost all of the Stables and Subdivision region.)

More conservative Republicans cannot take this region for granted. Romney got a commanding 57 percent here when he ran for governor as a moderate who would keep the Democratic Legislature in check, but he got only 45 percent when he ran against liberal lion Ted Kennedy for the US Senate in 1994. (Two years later, Torkildsen lost his seat, possibly because the new conservative Republican leadership in the US House didn’t go over so well here.) In 2004, John Kerry won 57 percent here against President Bush, though only 29 percent of the electorate were registered as Democrats; similarly, in Maine, Kerry won 54 percent, while 31 percent of the voters were registered as Democrats. On tax issues, the region is conservative, within limits. In 2000, it voted by a 30-point margin to roll back the income tax to 5 percent (the margin was 20 points statewide), but it broke with Cranberry Country in narrowly rejecting a plan to abolish the income tax altogether in 2004. There also seems to be a sizable liberal vote among those who vote in Democratic primaries: Reich got 27 percent here in 2002, higher than any region outside of Shopper’s World and Left Fields.

Still, the region has been unusually consistent in gubernatorial elections: Outside of the two Dukakis landslides in the 1980s, the GOP vote has ranged from 54 to 57 percent here in every election from 1978 through 2002, or four to eight points above the state average. Healey’s ties to the region should keep it on the more Republican side of the ledger this fall as well.


Cranberry Country: Red Tide

The cranberry crop in Massachusetts may be shrinking, but this region is still as red as it gets in Massachusetts. Cranberry Country, which takes in most of Plymouth County and Cape Cod, went for Kerry over Bush in the last presidential election, but only by a five-point margin. The three biggest towns where Bush got a majority of the vote— Hanover, Middleborough, and Sandwich—are all here. This was also the region where Romney came closest to beating Kennedy in the 1994 US Senate race, losing by only about 400 votes while falling short by at least 10,000 votes everywhere else; and it was the only region to vote for abolishing the income tax in 2002. But Cranberry Country was not the most Republican region in any of the three close gubernatorial races won by the GOP. It resembles Left Fields in that whenever it is the strongest region for one side in an election, that candidate or cause is almost sure to be too conservative to prevail statewide. (Another case in point: It was the strongest region for James Rappaport against Healey, Romney’s more moderate choice, in the primary for lieutenant governor in 2002.)

In terms of party registration, this is the most Republican region, but the GOP share is still only 18 percent. A strong majority of voters (56 percent) are unenrolled in any party, and this was the best region for independent Ross Perot, who got 28 percent in the 1992 presidential race and won the towns of Berkley, Middleborough, and Rochester. Independent gubernatorial candidate Christy Mihos’s base is here—he lives in Yarmouth, operates convenience stores on Cape Cod, and ran for the state Senate while living in Cohasset in 1990—and he is clearly hoping to do well in a region that the Republicans cannot afford to lose.

Cranberry Country voters in Democratic primaries consistently prefer moderates and conservatives: It was Silber’s best region in the 1990 gubernatorial primary and 12 years later it was Shannon O’Brien’s best region against three more liberal candidates.



Offramps: Stay to the Right

If you were to take a map of the state and plot the 50 or so communities that voted most heavily for Romney in 2002, most of your pushpins would form a large “C” around—and well removed from—the city of Boston. It would start in the Stables and Subdivisions town of Boxford (72 percent Republican), head south and west as far as Southborough (65 percent), and curve back to the coast at Cranberry Country’s Duxbury (68 percent). The middle part of that “C” is the Offramps region, which includes three cities and 36 towns clustered around the major commuting artery of I-495. This has been the most Republican region in all three of the past competitive gubernatorial elections; in fact, Romney’s 60 percent is the best showing by any candidate in any region in any of these elections. In terms of improvement over his 1994 race against Kennedy, this was also Romney’s best region, giving him a bounce of 13 points.

One reason for the Republican dominance here is that there are few urban areas to dilute the GOP vote. If you take the largest community in each region, Offramps has the smallest: Attleboro, with about 44,000 residents. But this is a high-growth area in a low-growth state. Offramps grew by an even 18 percent from 1990 through 2004, second to Cranberry Country’s 18.8 percent and well above the statewide figure of 6.7 percent. Of the 25 communities in the state that added the most new residents during that time, seven are in Offramps: Shrewsbury, Franklin, Marlborough, Attleboro, Westford, Hopkinton, and Westborough.

This is the state’s most independent region, with 57 percent of voters not enrolled in any party, so Offramps is by no means a lock for the Republicans. Similarly, the anti-tax faction here is strong but not invincible. The proposal to cut the income tax to 5 percent in 2000 won by a 2-to-1 margin in Offramps, the best showing in the state, but voters opposed doing away with the tax altogether in 2002. The abolitionists got 45 percent here, 15 points behind Romney (who did not endorse the proposal but was seen as the more tax-resistant candidate); this was the largest such gap in the state.

Given how many residents are new to Offramps, it may not be surprising that “insider” and urban-based candidates generally don’t do well in primaries here. Birmingham got 17 percent in the race for 2002 Democratic gubernatorial nomination, six points below his state average. In the primary for state treasurer the same year, Boston City Councilor Steve Murphy got only 14 percent here, seven points below his state average.



Almost every election night, whether national or statewide, features some cartographic plot twists—which also serve as cliffhangers. For a few years, we’re kept in suspense as to whether the changes on the map will last for generations or be reversed by the next election. In the 1928 presidential election, Massachusetts cast aside six decades of being one of the most Republican states in the US and voted for Democrat Al Smith, the first Catholic nominee from either major party. That turned out not to be a fluke, and the Bay State has leaned Democratic ever since. By contrast, in 1976 Jimmy Carter carried several Southern states including Mississippi and South Carolina, that the Democratic Party had seemingly lost forever during the social upheavals of the 1960s. In retrospect, that election clearly was a geographical fluke, and those two states haven’t voted Democratic since.

The Bay State’s 2002 gubernatorial election had its share of surprises, which generally cancelled each other out. Romney did better than Cellucci had done in four regions (Cranberry Country, Offramps, Shopper’s World, and Stables and Subdivisions) and worse in six others. But how long will these shifts last? Framingham and Waltham, two of the largest communities in Shopper’s World, voted for Romney despite going Democratic in almost all recent statewide elections. Are they part of a realignment or were they part of an aberration? Lenox and Lanesborough, two Berkshire County towns in the Left Fields region, voted for O’Brien after supporting the Republican Party in nearly all races where it was competitive. Can the GOP get them back —and does the party need them anyway?

We’ll find out this November just how important the twists of 2002 are in the long run. And we’ll get a sense of just where the losing party needs to go—to take back towns or claim new ones—in order to turn its fortunes around next time. Get ready: The political terrain for 2010 will largely be formed in just a few months.


From 1990 until now, only 27 communities have voted for the winner in every gubernatorial race (all won by Republicans) and every US Senate race (all won by Democrats). The largest are Falmouth (in Left Fields), Braintree (in Ponkapoag), Needham and Natick (both in Shopper’s World), and Melrose (in Post-Industria), which also boasted the highest voter turnout of any city in Massachusetts.

Meet the Author

But when considering presidential elections, general elections for all statewide offices, referenda, and the primaries of both major parties — the bellwether champ remains Fitchburg. The last time the MidMass city has been on the losing end was in 1994, when an initiative to abolish rent control lost here by six votes while passing statewide. Counting all other contests in 1994 and since, Fitchburg has a win-loss record of 84-1. When we noted this distinction in 2002, we cautioned, “Fitchburg lost population in the last federal census, in contrast to Massachusetts as a whole, so it may not be the best harbinger of the state’s political future.” But the US Census Bureau has since estimated that the entire state followed Fitchburg’s lead by losing population in 2004 and 2005. Our apologies to the City by the River for doubting its prognosticative powers.