Independent New Hampshire Journalism · Northfield, NHMoney raises well-being. Falling behind your neighbors lowers it. Outranking them pays close to nothing, and poverty next door costs everyone who lives beside it, including the people with money.
In February 1995, two researchers handed a survey to 257 faculty, students and staff at the Harvard School of Public Health and asked them to pick a world to live in. In the first world you earn $50,000 a year and everyone else earns $25,000. In the second you earn $100,000 and everyone else earns $200,000. Prices are identical in both, so the second world doubles what you can buy. Sara Solnick and David Hemenway reported that when the poorer-but-on-top world was listed first, 56 percent chose it. When it was listed second, 38 percent still did.
That survey has been retold ever since as proof that people would rather be rich among the poor than comfortable among the rich. The research that followed says something narrower and more useful. Money does raise well-being, and it keeps raising it far up the income scale. Falling behind the people around you lowers well-being, and the damage shows up in surveys, in job searches and in bankruptcy filings. Standing above the people around you pays very little. What it does pay depends on the poor staying out of sight.
The best-supported answer to the question in the headline is that people do better in a place where the people around them are also doing well. Poverty next door costs the people who live beside it. The habits that reliably raise well-being, trust in strangers, spending on other people, eating with someone, all need a community to practice them in. New Hampshire, which the Census Bureau ranks as the state with the lowest poverty rate in the country, has more riding on that finding than most places.
Related coverage: Trump Said the American Dream Was Dead. For Children Born the Year He Was, the Odds Were 86 Percent. tracks the odds of out-earning your parents by birth year. That piece measures income mobility across generations; this one measures how people report feeling about where they stand, so the two sets of numbers are not comparable. New Hampshire Needs Young Workers. It Built a State They Can’t Afford. covers the housing and child care costs behind the state’s affordability problem.
Start with what money does
The cheap version of this story says money can’t buy happiness. The data disagree. In 2010, Daniel Kahneman and Angus Deaton analyzed more than 450,000 responses to the Gallup-Healthways Well-Being Index, collected in 2008 and 2009. They found that day-to-day emotional well-being stopped improving at about $75,000 in annual income, while people’s evaluation of their own lives kept climbing with income. Their summary was that “high income buys life satisfaction but not happiness.” The $75,000 figure traveled far beyond the journal.
A decade later the number took a hit. Matthew Killingsworth, working with 1,725,994 real-time reports from 33,391 employed American adults who were pinged on their phones at random moments, found “no observed plateau in experienced well-being.” The slope above $80,000 was 0.110. Below $80,000 it was 0.109. Feeling in control of one’s own life accounted for 74 percent of the link between income and how people felt.
Kahneman and Killingsworth then did what researchers rarely do. They teamed up, with Barbara Mellers as a third author, to work out who was right. Their 2023 paper found that the plateau is real but “restricted to the least happy 20% of the population.” For that group, happiness rises with income and then levels off at about $100,000. For the happiest 30 percent, it rises faster past $100,000. Put plainly, more money stops helping only among the unhappiest fifth.
Windfalls point the same way. Economists who surveyed Swedish lottery players found that large-prize winners reported higher life satisfaction than matched players who did not win, and that the gain lasted more than a decade with no sign of fading. The effects on day-to-day happiness and mental health were smaller.
Even among millionaires, more is a little better. A 2018 study in Personality and Social Psychology Bulletin surveyed 2,129 people worth at least $1.5 million and a second group of 2,026 worth at least $1 million. People with $8 million to $14.9 million were more satisfied than people with $1.5 million to $2.9 million. In the second group, people with $10 million or more were happier than people with $1 million to $1.9 million. The differences were small. Average life satisfaction across the first study’s wealth bands ran from 5.79 to 5.97 on a seven-point scale, a spread of 0.18 points. Wealth that came from inheritance or marriage was tied to slightly lower happiness than wealth that was earned.
So being rich does feel better than being poor, and somewhat better than being comfortable. None of that answers the question. Every one of those studies measures a person’s own money. The question is about everybody else’s.
The cost of falling behind
Economist Erzo Luttmer matched a national panel survey of roughly 10,000 people to average earnings in the areas where they lived, Census districts of about 150,000 residents. He found that when neighbors earned more, people reported being less happy, even with their own income held constant. The size is what stands out. A rise in the neighbors’ earnings and an equal cut in a person’s own income each reduced happiness by, in the paper’s words, “about the same order of magnitude.” The effect was stronger for people who socialized often with their neighbors. It was not stronger for people who socialized with relatives, coworkers or friends from elsewhere.
That is a correlation in survey data, and a skeptic could say unhappy people sort into richer areas. The experiments are harder to wave off. A website began listing what University of California employees were paid. Four economists told a random subset of the university’s workers that the site existed, then surveyed everyone. Workers who learned they were paid below the median for their department and job reported lower satisfaction with their pay and their work, and were more likely to say they were looking for another job.
Norway ran the same experiment on a whole country without meaning to. In 2001, Norwegian tax records went online, and anyone could look up what anyone else earned. Economist Ricardo Perez-Truglia compared survey data from 1985 to 2013 and found that the happiness gap between richer and poorer Norwegians grew 29 percent once incomes became visible. The life satisfaction gap grew 21 percent. Nobody’s paycheck changed. What changed was that everyone could see the scoreboard.
The strongest evidence is financial. Three economists studied 7,377 lottery wins in one Canadian province between April 2004 and March 2014, in postal codes so small that the median one holds 13 households. Their Federal Reserve Bank of Philadelphia working paper found that when one household won, bankruptcies among the neighbors went up. A win of C$29,229, the median annual income in the sample, was followed by a 6.6 percent rise in the neighbors’ bankruptcies in the year of the win and the two years after. A C$100,000 win was followed by an 11 percent rise. The neighbors who went bankrupt held more of the assets other people can see, houses and cars. Furniture and other less visible property showed no effect. Households farther out, in the surrounding area of about 200 homes, showed no effect either.
A Dutch lottery that pays out by postal code produced a matching result. People who had not bought a ticket but lived next door to a winner spent significantly more on cars than other non-players.
What the winners get
If losing rank hurts that much, winning rank should feel great. The same studies say otherwise.
Go back to the California salary experiment. Workers who found out they earned more than their peers reported no increase in job satisfaction. The authors concluded that the effect is concentrated among workers paid less than their peers. Learning you are behind makes you miserable. Learning you are ahead leaves you where you were.
The national data show the same lopsided pattern. Psychologists Shigehiro Oishi, Selin Kesebir and Ed Diener examined the General Social Survey from 1972 to 2008, more than 48,000 respondents across 37 years. According to the Association for Psychological Science’s account of the study, Americans in the lower and middle income brackets, about 60 percent of the population, were less happy in years when the national income gap was wider. The drop was not explained by their own incomes. It ran through trust: in unequal years they rated other people as less fair and less trustworthy. For the richest 20 percent, the gap made no difference to their trust, their sense of fairness or their happiness.
That finding answers half the headline question on its own. A wider gap between the top and everyone else did not make the top happier. It made most other people less happy and left the rich where they were. “We’ve seen that people seem to be happier when there is more equality,” Oishi said in the release. The researchers cautioned that the study shows correlation, not cause.
The 1995 Harvard survey holds a second clue that the retellings leave out. Solnick and Hemenway asked the same either-or question about vacation time. Would you rather have two weeks off while others get one, or four weeks while others get eight? Only 18 percent picked the on-top option when it was listed first, and 14 percent when it was listed second. People wanted more vacation, period. The urge to outrank everyone else was strong for income, which other people can see and compare, and weak for time off, which they mostly cannot.
The pull of rank is real. Luttmer’s arithmetic cuts both ways: if richer neighbors lower reported happiness, poorer neighbors should raise it. But when researchers look for the payoff directly, by telling people they are ahead or by comparing the rich across more and less unequal years, it is small or missing.
The ugly exception
There is one place where the on-top premium shows up, and it says something unflattering about how it works.
Sociologists Glenn Firebaugh and Matthew Schroeder took 1,266 respondents to the 2000 American National Election Study and matched each one to Census income data at two scales: the immediate neighborhood and the surrounding county. The sample covered 849 neighborhoods in 433 counties. The two scales pointed in opposite directions.
People who lived in richer neighborhoods were happier, with their own income held constant. Residents of neighborhoods at the 95th percentile of income had about 1.8 times the odds of calling themselves completely satisfied with life, compared with residents of neighborhoods at the 5th percentile. The authors credit safety, upkeep and the other things that come with a block where people have money. At the county level the sign flipped. People who lived in poorer counties were happier, again with their own income and their neighborhood’s income held constant.
Put the two together and the happiest arrangement in the data is a well-off block inside a poorer county. In the authors’ summary, people are happier among the poor as long as the poor do not live too close. The comparison paid off when poverty was an abstraction on the far side of the county, where it could lift a person’s relative standing without costing them anything on their own street.
That is the one sense in which being rich while everyone else is poor “feels better” in this body of research: at a distance, as a statistic. It is also consistent with Luttmer, whose comparison areas of 150,000 people sit closer to county scale than to block scale. The envy operates across the region. The comfort operates on the street.
What poverty next door costs
The federal government ran the experiment that tests the other half of the question. Between 1994 and 1998, the Moving to Opportunity program enrolled 4,604 low-income families living in public housing in Baltimore, Boston, Chicago, Los Angeles and New York. A lottery decided which families got vouchers to move to lower-poverty neighborhoods. Researchers followed them for 10 to 15 years.
The results, published in Science in 2012, found that moving did not make the adults more economically self-sufficient. It did improve their physical health, their mental health and their reported well-being. The well-being gain was large. A 13 percentage point drop in the neighborhood’s poverty rate raised a person’s well-being by as much as the gap between two people whose annual incomes differ by $13,000.
Those families did not get richer. The people around them did. The paper noted that nearly 9 million Americans lived in neighborhoods of extreme poverty.
Firebaugh and Schroeder’s neighborhood finding extends the point up the income scale. Their result held after accounting for what each person earned: living where the neighbors were poorer was tied to lower satisfaction. The data give no sign that being the richest household on a struggling street is the prize the 1995 survey implied.
Trust, giving and a shared table
The research on what does raise well-being keeps landing on things that need other people.
In 2008, Elizabeth Dunn, Lara Aknin and Michael Norton published three studies in Science. In a national survey of 632 Americans, how much people spent on themselves had no relationship to how happy they were. How much they spent on other people did. In an experiment, 46 people were handed $5 or $20. Some were told to spend it on themselves and the rest to spend it on someone else. The ones who spent it on someone else reported being happier afterward, and the size of the windfall made no difference. A Stanford Graduate School of Business summary lays out the three studies. A later team found the same link in survey data from 136 countries and reproduced the effect in experiments in Canada and Uganda.
The 2025 World Happiness Report found that sharing meals has an effect on well-being on par with income and unemployment. It also reported that in 2023 roughly one in four Americans ate every meal alone on the previous day, a 53 percent increase since 2003. The same report found that people are too pessimistic about each other. In lost-wallet tests, far more wallets came back than people predicted, and a person’s well-being tracks how kind they believe others to be.
The 2026 edition, published March 19, 2026, by the Wellbeing Research Centre at the University of Oxford with Gallup and the United Nations Sustainable Development Solutions Network, ranked the United States 23rd. Finland placed first for the ninth year in a row. The Associated Press reported that the Nordic countries’ standing is credited to wealth, its even distribution, a welfare state that cushions recessions and long healthy lives.
A person can have money alone. Trust and company take a town.
What the evidence says about each world
| The world | What the research found | Verdict |
|---|---|---|
| You are rich and the people near you are poor | Lower life satisfaction in poorer neighborhoods, after accounting for a person’s own income. Moving families out of high-poverty neighborhoods raised well-being as much as a $13,000 income gap. | Worse |
| You are rich and the poor are far away | People in poorer counties reported being happier, once their own and their neighborhood’s income were accounted for. The one setting where the on-top premium appears. | Better, on a survey |
| You are comfortable and those around you are richer | Lower happiness when area earnings rise. Lower job satisfaction for workers who learn they are below the median. More bankruptcies next door to lottery winners. | Worse |
| You are ahead and you know it | No gain in job satisfaction for workers who learned they were above the median. No difference in happiness for the richest fifth between more and less unequal years. | No measurable gain |
| The people around you are also doing well | Higher satisfaction in higher-income neighborhoods. Higher happiness for most Americans in years with a narrower income gap, tied to higher trust. | Better |
New Hampshire’s stake
By the Census Bureau’s count, New Hampshire is as close as an American state gets to the last row of that table. The bureau’s September 2026 comparison of state poverty rates put New Hampshire at 5.4 percent on its 2023-to-2025 average, the lowest of any state. The New Hampshire Fiscal Policy Institute reported a median household income of $99,782 in 2024.
The same sources show how much room is left. On the bureau’s Supplemental Poverty Measure, which the bureau reports alongside the official one, New Hampshire’s rate was 8.0 percent, which the bureau said was significantly higher. A separate Census survey, the American Community Survey, counted about 98,600 Granite Staters below the poverty line in 2024, a rate of 7.2 percent, including about 16,000 children. The geography is uneven. In the 2018-to-2022 data, the institute reported a poverty rate of about 4.8 percent in Rockingham County and about 11.7 percent in Sullivan County, more than double. The institute also reported that income inequality in the state is higher than it was in 2010.
Read against the research, those figures describe an asset that belongs to everyone in the state, the well-off included. The research gives a household in a prosperous New Hampshire town a selfish reason to care that few of its neighbors are poor. The Firebaugh and Schroeder result suggests the same household would report being a little happier if the poverty sat one county over. Both findings come from the same paper. Only the first describes somewhere worth living.
I am not offering that as a tax plan. Oishi argued his results favor more progressive taxation, and a reader can take the same evidence and argue for housing construction, wage growth or private charity instead. What the evidence does not support is indifference. A state that lets poverty spread on the theory that the comfortable are unaffected is working from a theory the data reject.
What the research does not settle
Most of this evidence is correlational. The 37-year inequality study cannot prove the income gap caused the unhappiness, and its authors say so. The Association for Psychological Science noted that earlier studies comparing countries and states produced mixed results. The millionaire surveys cannot show that an extra $7 million causes the small bump in satisfaction, or that the kind of person who earns a fortune was not more satisfied to begin with.
Some of the best-known findings rest on small samples. The Harvard survey polled 257 people at one school of public health in 1995 about worlds that do not exist, and the order of the answer choices moved the result by 18 points. The spending experiment in Science had 46 participants. The neighborhood-and-county study had 1,266. The cross-national surveys and the Uganda and Canada experiments make the giving result harder to dismiss, but a reader should hold the single-study numbers loosely.
The experiments are strong and narrow. Moving to Opportunity studied low-income families leaving public housing, not affluent families living near poverty. The lottery studies measure neighbors’ spending and bankruptcies, not their moods. And nobody has run the experiment the headline describes. I found no study that assigned people at random to be rich in a poor place or rich in a rich one and then measured how they felt.
The answer
The question was whether people feel better rich among the poor, or rich among neighbors who are also doing well, or at least not suffering beside them.
On the evidence available, the second. Money helps, at every level measured. Being outranked hurts enough to push people into debt. Outranking others pays little that researchers can find, and across 37 years of surveys the richest fifth of Americans were no happier in the years when the gap was wider. The premium for being on top appears only when the people at the bottom are far enough away to be a number.
Up close, other people’s poverty is a cost. For the families who moved away from it, a 13-point drop in neighborhood poverty was worth as much well-being as $13,000 a year in income, and they did not earn more. The people who did best in the data lived where their neighbors were doing well, trusted the strangers around them and spent some of what they had on someone else.
The 1995 survey asked people to choose between being rich and being richer than the neighbors. Roughly half took the bad deal. The three decades of evidence since suggest they were wrong about what would make them happy, and that the neighbors were never the problem.
— Dexter Dow, Granite State Report
Fact check
| # | Claim | Status | Source |
|---|---|---|---|
| 1 | 257 faculty, students and staff at the Harvard School of Public Health were surveyed in February 1995; 56 percent chose the $50,000-versus-$25,000 world when it was listed first and 38 percent when it was listed second; for vacation, 18 and 14 percent. | VERIFIED | Solnick and Hemenway, “Is more always better?: A survey on positional concerns,” Journal of Economic Behavior & Organization 37(3), 373-383 (1998), Table 1. Full text |
| 2 | More than 450,000 Gallup-Healthways responses from 2008-2009; emotional well-being stopped improving at about $75,000; life evaluation kept rising. | VERIFIED | Kahneman and Deaton, PNAS (2010). Paper |
| 3 | 1,725,994 reports from 33,391 employed U.S. adults; no plateau; slopes of 0.110 above and 0.109 below $80,000; sense of control accounted for 74 percent of the association. | VERIFIED | Killingsworth, PNAS 118(4) (2021). Paper |
| 4 | The plateau is confined to the least happy 20 percent and sets in at about $100,000; happiness rises faster past $100,000 for the happiest 30 percent. | VERIFIED | Killingsworth, Kahneman and Mellers, “Income and emotional well-being: A conflict resolved,” PNAS 120(10) (2023). Abstract record |
| 5 | Swedish large-prize lottery winners showed higher life satisfaction for more than a decade; smaller effects on happiness and mental health. | VERIFIED | Lindqvist, Östling and Cesarini, Review of Economic Studies 87(6) (2020); working-paper abstract read. NBER 24667 |
| 6 | 2,129 and 2,026 millionaires; higher satisfaction at $8 million-$14.9 million and at $10 million or more; means from 5.79 to 5.97 on a seven-point scale; unearned wealth tied to lower happiness. The 0.18-point spread is GSR’s subtraction. | VERIFIED | Donnelly, Zheng, Haisley and Norton, Personality and Social Psychology Bulletin 44(5) (2018). Paper |
| 7 | Higher neighbors’ earnings are associated with lower reported happiness; the effect is about the size of an equal cut in own income; areas average about 150,000 residents; stronger for people who socialize with neighbors. | VERIFIED | Luttmer, “Neighbors as Negatives,” Quarterly Journal of Economics 120(3) (2005); working-paper text read. NBER 10667 |
| 8 | University of California workers told about a pay website: below-median workers reported lower satisfaction and more job search; above-median workers reported no higher satisfaction. | VERIFIED | Card, Mas, Moretti and Saez, American Economic Review 102(6) (2012). NBER 16396 |
| 9 | After Norwegian tax records went online in 2001, the happiness gap between richer and poorer people grew 29 percent and the life satisfaction gap 21 percent. | VERIFIED | Perez-Truglia, American Economic Review 110(4) (2020). NBER 25622 |
| 10 | 7,377 Canadian lottery wins, April 2004 to March 2014; median postal code of 13 households; neighbors’ bankruptcies up 6.6 percent after a C$29,229 win and 11 percent after a C$100,000 win in years 0 to 2; visible assets only; no effect in the outer ring. | VERIFIED | Agarwal, Mikhed and Scholnick, Federal Reserve Bank of Philadelphia Working Paper 18-16 (2018); later published in Review of Financial Studies 33(1) (2020). Working paper |
| 11 | Non-players living next door to Dutch Postcode Lottery winners had significantly higher car consumption. | VERIFIED | Kuhn, Kooreman, Soetevent and Kapteyn, American Economic Review 101(5) (2011). Abstract |
| 12 | General Social Survey, 1972-2008, more than 48,000 respondents; lower and middle brackets (about 60 percent) less happy in more unequal years, through trust and fairness; no effect for the richest 20 percent; Oishi quotation; correlation caveat; earlier cross-place studies mixed. | ATTRIBUTED | Association for Psychological Science news release, June 13, 2011, on Oishi, Kesebir and Diener, Psychological Science (2011). Release |
| 13 | 1,266 respondents in 849 neighborhoods and 433 counties; happier in richer neighborhoods (about 1.8 times the odds, 95th versus 5th percentile); happier in poorer counties. | VERIFIED | Firebaugh and Schroeder, American Journal of Sociology 115(3) (2009). Paper |
| 14 | The happiest arrangement in that data is a well-off block inside a poorer county; the premium works “at a distance.” | ATTRIBUTED | GSR’s reading of Firebaugh and Schroeder’s two findings taken together; the authors summarize their result in similar terms. |
| 15 | Moving to Opportunity: 4,604 families, five cities, enrolled 1994-1998, followed 10 to 15 years; no effect on self-sufficiency; a 13-point drop in neighborhood poverty equals the well-being gap of a $13,000 income difference; nearly 9 million Americans in extreme-poverty neighborhoods. | VERIFIED | Ludwig and others, Science (2012). Paper |
| 16 | Survey of 632 Americans: personal spending unrelated to happiness, spending on others related; experiment with 46 participants given $5 or $20. | VERIFIED | Dunn, Aknin and Norton, “Spending Money on Others Promotes Happiness,” Science 319(5870), 1687-1688 (March 21, 2008). Stanford summary |
| 17 | Spending on others linked to happiness in 136 countries; experiments in Canada and Uganda. | VERIFIED | Aknin and others, NBER Working Paper 16415 (2010). NBER 16415 |
| 18 | Sharing meals on par with income and unemployment; about one in four Americans ate all meals alone in 2023, up 53 percent since 2003; wallets returned more often than expected. | VERIFIED | World Happiness Report 2025, executive summary. Summary |
| 19 | World Happiness Report 2026 published March 19, 2026, by the named publishers; United States 23rd; Finland first for the ninth straight year; Nordic standing tied to wealth and its even distribution. | ATTRIBUTED | Publication details: World Happiness Report 2026. Rankings and the Nordic explanation: Associated Press, March 19, 2026. |
| 20 | New Hampshire’s poverty rate of 5.4 percent, 2023-2025 average, lowest of any state (the bureau labels the measure “official+”); Supplemental Poverty Measure rate of 8.0 percent, significantly higher. | VERIFIED | U.S. Census Bureau, “States With Highest Poverty Rates Clustered in the South,” September 15, 2026. Story |
| 21 | Median household income of $99,782 in 2024; 7.2 percent poverty rate and about 98,600 people in poverty, including about 16,000 children; inequality higher than in 2010. | ATTRIBUTED | New Hampshire Fiscal Policy Institute, September 11, 2025, citing 2024 American Community Survey one-year tables S1901, S1701 and B19083. Analysis |
| 22 | Poverty rates of about 4.8 percent in Rockingham County and about 11.7 percent in Sullivan County, 2018-2022. “More than double” is GSR’s division (2.44). | ATTRIBUTED | New Hampshire Fiscal Policy Institute, New Hampshire Policy Points 2025, citing American Community Survey table S1701. Chapter |
| 23 | No study located that randomly assigned people to be rich in a poor place or a rich one. | ATTRIBUTED | GSR’s search of the literature cited here; an absence of located evidence, not proof that none exists. |
Reach the editor directly. Confidentiality respected where possible.
granitestatereport@gmail.com
Editor’s note. Every factual claim above was checked against the studies and agency publications named in the fact-check table. Where a working-paper version of a study was read, the table says so. Rows marked ATTRIBUTED rest on the named institution’s account or on GSR’s own reading. This is an analysis of published research; no interviews were conducted for it. Corrections: Granite State Report corrects verified errors promptly and appends a note identifying what changed and when.
