Independent New Hampshire Journalism · Northfield, NH
Robert Greene’s thirty-first law says to give people a choice and limit the choices. In 1972 a candidate for governor reduced New Hampshire’s tax debate to one question. Both parties, and a Legislature under court order, have been answering it since.
In 1972, Meldrim Thomson ran for governor of New Hampshire for the third time against the same Republican, the incumbent Walter Peterson, who had beaten him twice. Peterson had proposed a 3 percent personal income tax, Michael Kitch reported for InDepthNH. Thomson made one promise the theme of his campaign: no income tax and no sales tax. William Loeb, the owner of the Manchester Union Leader, amplified it. Thomson ousted the incumbent in the primary, as Stateline later put it, by pledging to oppose both taxes.
The promise outlived the campaign. It became a question, and the question got a name: the Pledge. For decades, candidates for governor in both parties have been asked whether they will take it. There are two answers. Thomson wrote both of them.
Robert Greene’s thirty-first law says to control the options: give people a choice, limit the choices to ones that serve you, and let them feel they decided. The Pledge is that law in working order. Democrats and Republicans have played the hand Thomson dealt for decades. When the state Supreme Court threw out the school funding system in 1997, the Legislature answered from the same short deck, and the answer is still in the statute book. The record also shows two things the law leaves out: one governor refused the cards and won, and the voters declined to make the deck permanent.
The stage
Greene’s law, as summaries of the book put it, turns on the appearance of choice. Offer options, but only a few, and make sure each one works in your favor. The other side picks freely and still lands where the dealer wanted.
New Hampshire’s tax debate has run on that structure since Thomson. StateImpact New Hampshire, in a 2012 history, traced the Pledge to the 1950s and its popularity to the 1970s. James Pindell, writing in New Hampshire Magazine, credited the term to Gov. Hugh Gregg and its force to Thomson’s first winning campaign. By 2012, StateImpact reported, New Hampshire and Alaska were the only states with neither a broad-based income tax nor a broad-based sales tax.
The trick in the Pledge is the bundling. Dante Scala, a University of New Hampshire political scientist, told StateImpact that the Pledge’s backers folded two different taxes into a single “pro-tax” label. A candidate who would consider either one became the tax candidate. On the Republican side, Scala said, “it’s become gospel.”
The Democrats who took the cards
The clearest proof that the frame works is who has played inside it. Hugh Gallen, a Democrat, took the Pledge and beat Thomson twice, in 1978 and 1980, Kitch reported. In 1982 Gallen dropped it, called the tax system “obsolete,” and lost his bid for a third term to John Sununu.
The Democrats who followed read that result. Jeanne Shaheen took the Pledge in 1996 and 1998 and won both times, by Pindell’s account. John Lynch ran four times, took it four times and won four times, Stateline reported. Maggie Hassan took it in 2012 and beat a primary opponent, Jackie Cilley, who had refused. Cilley did not propose an income tax. She declined to promise, and she campaigned against what she called “pledge zombies.”
Six years later Cilley told New Hampshire Public Radio what the refusal had taught her. It was, she said, “an absolute non-starter for a campaign.”
Hassan herself showed how the cards get picked up. Republicans pointed out in 2012 that as a state Senate candidate in 2002 she had been open to an income tax if it could eliminate the statewide property tax, Stateline reported. Ten years later she promised to veto one.
The pull was still there in 2018. Molly Kelly, running for governor as a Democrat, took the Pledge, Josh Rogers reported for New Hampshire Public Radio. Another Democrat in the race, Steve Marchand, called the Pledge self-defeating and said the state needed more revenue, and he still did not favor a new broad-based tax. Andru Volinsky, then an executive councilor, said the state should not be bound by a political slogan from the 1970s.
The candidates who went further did worse. Democrats nominated Arnie Arnesen in 1992 and Mark Fernald in 2002 on platforms that favored an income tax. Each finished with 40 percent of the general election vote or less, Stateline reported. The frame has held in the other party too. Liz Hager, a Republican state representative who backed an income tax, lost the 1992 nomination for governor to Steve Merrill, Pindell wrote.
Each of those candidates made a free choice. Each choice was one of two, and the same man had defined both. Take the Pledge and the tax debate is over before the term starts. Refuse it and the campaign becomes a referendum on a tax the candidate may never have proposed.
The court deals a second hand
On Dec. 17, 1997, the New Hampshire Supreme Court ruled in Claremont School District v. Governor that the way the state paid for public schools was unconstitutional. Local property taxes supplied 74 to 89 percent of school revenue, the court found, and the rates were wildly uneven. In the 1994-95 school year, Pittsfield’s equalized school rate was $25.26 per thousand dollars of value. Moultonborough’s was $5.56.
The state’s own share was small. Direct legislative appropriations averaged 8 percent of what was spent on public schools, the court found, which ranked New Hampshire last in the country in direct state support. In Allenstown the equalized rate was $26.47. In Rye it was $6.86.

The court held that the school tax was a state tax, and that a state tax must be proportional. Then it set a condition of its own. If the property tax was going to pay for an adequate education, it had to be “equal in valuation and uniform in rate throughout the State.”
The justices were careful about what they would not do. The court, it repeated in a 1999 ruling in the same case, was “not appointed to establish educational policy” or to decide how to pay for it. The choice of tax belonged to the Legislature and the governor.
That left lawmakers with a hand dealt by two different players. The court had ruled out uneven local property taxes. The Pledge had ruled out, for any governor who had taken it, a sales tax or an income tax. Shaheen had taken it twice.
What the Legislature played
In April 1999, more than sixteen months after the ruling, the Legislature passed a law creating a statewide education property tax at a uniform rate of $6.60 per thousand, the court recounted that October. It was the option the court’s own language had left open and the only broad tax the Pledge did not forbid.
An income tax came close. Stateline reported that one passed both houses in 1999 and was squelched by Shaheen. The Federal Reserve Bank of Boston’s Regional Review said the Legislature passed an income tax and Shaheen vetoed it. Kitch wrote that she nixed an income tax bill. The accounts differ on the mechanics and agree on the result.
The rest of the 1999 package stayed inside the same lines. It raised the business profits tax and drew on tobacco settlement money, Regional Review reported. Lawmakers had put the cost of an adequate education for 2000 at $825 million.
Even the property tax bent toward the towns with the most to lose. The 1999 law phased in the full rate over five years for fifty property-rich towns. That first year the effective rate was about $3.79 in Rye, $3.18 in Moultonborough and $1.43 in Waterville Valley, the court found. In Claremont and Allenstown it was the full $6.60. The court struck down the phase-in.

The card still on the table
The tax the Legislature chose in 1999 is still law. Its rate is no longer fixed. Its total is.
The larger pattern held as well. Kitch, citing the New Hampshire Fiscal Policy Institute, reported that property taxes raised 59 percent of all taxes collected in the state, the greatest share of any state. Granite State Report has walked through the four lines on a New Hampshire tax bill and the income tax debate separately. The point here is narrower. The bill looks the way it does in part because of which options were on the table when the bill was designed, and who put them there.
The governor who refused the cards
The law has a counterexample, and it is the same governor. In 2000, Shaheen declined to take the Pledge and won a third term. By Pindell’s count, written during the Lynch years, no one else had won the office without it since 1972.
She also tried to deal. Shaheen created a commission in April 2000 to evaluate revenue options, the Boston Fed’s Regional Review reported. After it reported in January 2001, the Legislature rejected her proposed 2.5 percent sales tax dedicated to schools, along with a broad consumption tax and an income tax, and kept the statewide property tax.
So the refusal worked at the ballot box and failed in the State House. Shaheen got out from under the Pledge. The Legislature stayed inside it.
A fair objection
The first objection is that nobody was tricked. Greene’s law describes a deception, and the Pledge is the opposite of hidden. It is asked in public, answered in public and enforced by voters who know what they are voting for. That is fair, and the 2012 ballot is the best evidence. Republican lawmakers sent voters a constitutional amendment to ban an income tax. It passed the House 257 to 101 and the Senate 20 to 4, Kitch reported. It needed two-thirds of the voters. New Hampshire Public Radio put the yes vote at just over 57 percent in its early count. A majority wanted the ban. The amendment still failed.
That vote cuts both ways. A majority chose the Pledge’s side on its own, with no candidate in the way. But the same electorate kept the option alive in the constitution. Voters liked the cards and declined to glue them to the table.
The second objection is that the court, not Thomson, narrowed the Legislature’s choices in 1999. Also fair. The uniform-rate rule came from the constitution’s tax clause, and the justices did not write the Pledge. The claim here is that the two limits stacked. The court removed one option and the Pledge had already removed two.
The verdict
Greene’s thirty-first law holds in New Hampshire about as well as any law in this series. One candidate in one primary defined a choice so cleanly that his opponents’ party has spent decades answering it. The candidates who answered his way mostly won. Most of the ones who refused lost, including one who never proposed a tax at all.
The limits are real. Shaheen won without the Pledge. The voters would not write it into the constitution. And the frame’s largest product, a school system paid for mainly through property taxes, is back in court. Kitch reported two school funding lawsuits pending in 2025.
The dealer left office long ago. The game still uses his deck. Part Thirty showed figures who won by hiding the work. Part Twenty-Nine showed campaigns that planned only as far as election night. Part Thirty-One shows a campaign promise that kept setting the terms long after its author stopped running.
Next: Law 32, on playing to people’s fantasies, tested against the New Hampshire record.
Fact check
| # | Claim | Status | Source |
|---|---|---|---|
| 1 | The Pledge came out of the 1972 Republican primary between incumbent Walter Peterson and Meldrim Thomson, whom Peterson had beaten twice; Peterson proposed a 3 percent personal income tax; the Pledge became the theme of Thomson’s campaign, amplified by William Loeb, owner of the Manchester Union Leader; Gallen took the Pledge and defeated Thomson in 1978 and 1980; Gallen shunned it in 1982, called the tax system “obsolete” and lost to John Sununu; Shaheen nixed an income tax bill; the 2012 amendment passed the House 257 to 101 and the Senate 20 to 4 and failed to win two-thirds of voters; property taxes raised 59 percent of all taxes collected, the greatest share of any state (citing NHFPI); two school funding lawsuits pending in 2025 | ATTRIBUTED | Michael Kitch, InDepthNH, March 21, 2025 |
| 2 | Thomson ousted the incumbent governor in the 1972 primary by pledging to oppose an income or sales tax; an income tax passed both houses in 1999 and was squelched by Shaheen; Arnesen (1992) and Fernald (2002) were pro-income-tax nominees and neither passed 40 percent; Lynch took the Pledge all four times and won each; Hassan took it in 2012 and said she would veto an income or sales tax; Republicans noted she had been open to an income tax as a 2002 state Senate candidate if it could eliminate the statewide property tax; Cilley did not advocate an income tax, refused the Pledge and attacked “pledge zombies”; the 2012 amendment needed two-thirds | ATTRIBUTED | Stateline, “Tax Pledge Permeates New Hampshire Politics,” 2012 |
| 3 | The term goes back to the 1950s and Gov. Hugh Gregg; it became significant in 1972; Shaheen took the Pledge in 1996 and 1998 and won in 2000 without it; since 1972 only one person had won the governorship without it (as of the article, written while Lynch was governor); Liz Hager backed an income tax and lost the 1992 Republican nomination to Steve Merrill | ATTRIBUTED | James Pindell, New Hampshire Magazine, “At Ease with the Pledge” (undated on the page) |
| 4 | The Pledge dates to the 1950s and became popular in the 1970s; New Hampshire and Alaska were the only states with neither a broad-based income nor sales tax (as of 2012); Scala on bundling the two taxes under one “pro-tax” label; “it’s become gospel on the Republican side” | ATTRIBUTED | Jacob Hale Russell, StateImpact New Hampshire, “A History Of The Pledge,” 2012 |
| 5 | Cilley ran against the Pledge in 2012 and lost the primary to Hassan; “an absolute non-starter for a campaign”; in 2018 Molly Kelly took the Pledge, Steve Marchand called it self-defeating without favoring a new broad-based tax, and Executive Councilor Andru Volinsky said the state should not be bound by a 1970s slogan | ATTRIBUTED | Josh Rogers, New Hampshire Public Radio, Sept. 6, 2018 |
| 6 | Claremont II, decided Dec. 17, 1997: the school funding system is unconstitutional; local property taxes supplied 74 to 89 percent of school revenue; direct legislative appropriations averaged 8 percent, ranking New Hampshire last in direct state support; 1994-95 equalized rates of $25.26 (Pittsfield), $5.56 (Moultonborough), $26.47 (Allenstown) and $6.86 (Rye); the school tax is a state tax; a property tax used for adequacy must be “equal in valuation and uniform in rate throughout the State” | VERIFIED | Claremont School District v. Governor, 142 N.H. 462 (1997), opinion text read on CourtListener |
| 7 | Decided Oct. 15, 1999: the Legislature acted in April 1999, more than sixteen months after Claremont II; Laws 1999, ch. 17 set a uniform rate of $6.60 per $1,000; fifty property-rich towns phased in over five years; 1999 rates of about $3.79 (Rye), $3.18 (Moultonborough) and $1.43 (Waterville Valley) against $6.60 in Claremont and Allenstown; the phase-in held unconstitutional; “not appointed to establish educational policy” | VERIFIED | Claremont School District v. Governor, 144 N.H. 210 (1999), opinion text read on CourtListener |
| 8 | RSA 76:3 text: the commissioner sets the rate “at a level sufficient to generate revenue of $363,000,000”; source notes from 1999, 17:14 through 2008, 173:15, effective July 1, 2009 | VERIFIED | gc.nh.gov, RSA 76:3 |
| 9 | The Legislature passed an income tax and Shaheen vetoed it (this account’s wording); the 1999 measures included higher business profits taxes and tobacco settlement money; the Legislature set the cost of adequacy for 2000 at $825 million; Shaheen created a revenue commission in April 2000; it reported in January 2001; the Legislature rejected her proposed 2.5 percent school sales tax, a consumption tax and an income tax, and kept the statewide property tax | ATTRIBUTED | Robert Tannenwald, “Heat, Light, and Taxes in the Granite State,” Regional Review, Federal Reserve Bank of Boston, Q3 2001 |
| 10 | Question 1 drew just over 57 percent in the early count and fell short of two-thirds | ATTRIBUTED | Brian Wallstin, New Hampshire Public Radio, Nov. 8, 2012 |
| 11 | Featured photograph: the old Ash Street School building in Manchester, Carol M. Highsmith, LC-DIG-highsm-45652, no known restrictions | VERIFIED | Library of Congress catalog record 2017882400 |
| 12 | Robert Greene’s Law 31 is “Control the Options: Get Others to Play with the Cards You Deal,” from The 48 Laws of Power, Viking, 1998 | VERIFIED | Published table of contents; Shortform list |
| 13 | GSR Power Index scores for Laws 1 through 31 | EDITORIAL | Granite State Report’s own rating, not a measurement. Method stated in the chart |
Your voice
Poll. Should candidates for governor take the Pledge?
Yes, and keep it · Yes, but it should not bind them forever · No, keep every option open · It makes no difference
Survey: your tax bill and the Pledge. Five questions: your town; whether a candidate’s answer on the Pledge has ever decided your vote; which you would choose if the state had to raise school money and could pick only one source; how fair your property tax bill feels, on a scale of one to five; and what you would tell a candidate for governor about it. Granite State Report asks before using your name or story.
Discuss. Is the Pledge a promise voters demand, or a question that keeps other answers off the table? And who should decide how New Hampshire pays for its schools: the courts, the Legislature or the voters directly?
Verified New Hampshire residents vote and answer at granitestatereport.com. Or write to granitestatereport@gmail.com with your town in the subject line.
Granite State Report wants to hear from former candidates, campaign staff, legislators and school board members who lived the 1999 funding fight. granitestatereport@gmail.com, 603-931-9264. Confidentiality on request.
Editor’s note: Commentary and analysis, not legal advice. The two court opinions and the statute were read directly. Accounts of the Pledge and of the candidates who took or refused it come from published reporting and are attributed. Three sources describe Gov. Shaheen’s 1999 action on an income tax in different words (squelched, vetoed, nixed); all three are shown, and the House and Senate journals were not read. The 2012 ballot figure is an early count as reported, not the official canvass. The GSR Power Index is editorial, not data.

