The happiest place on Earth

is Ireland.* Really.

And how do we know this startling little fact? Well it turns out there is a long literature of attempts to rank quality of like across locales and countries. Informally, we all do this every time we are faced with a move. More formal attempts involve taking statistics like gdp, political freedom, community, etc. and then throwing them all into a pot and coming up with an index. The final result is approximately as ludicrous as it sounds, since it is rather difficult to know how much of a political freedom index one is willing to give up for an extra $1,000. This is the weighting problem. We can think of an index as taking the listed statistics and multiplying each member of the list by some “weight”. But how in the world do we know the right weights? If we put too much weight on GDP, for example, we might be under the impression that richer means happier and the rest doesn’t matter.

The UN does this sort of thing with their Human Development Index, which is basically an attempt to supplement the GDP numbers with some other things like health and education. They pretty much are just pulling numbers out of the air for the weights.**

I already gave away the answer, by the way; you try to use the weights that people use when they do this for themselves. In other words, my wife preferred California to Utah for a specific set of reasons, even though Utah is better in other ways. Figure out how happy different locales make different people and then use this to form weights. There are a few ways to do this.

For example, if there were no immigration barriers and moving was costless (not just financially but in every respect), then you would just use immigration flows.

Alas, there are huge barriers to migration. So the Economist took another route. They used surveys asking people how happy/satisfied they were. Then they used those surveys (held in about 70 countries) in a regression that finds the set of weights that best relate the statistics to the surveyed happiness. Doing this, they can then use those weights to compute happiness in every country for which the statistics on gdp and so forth are available.

Are there problems with this approach? Absolutely (see here for the methodology). It is only dealing with how those who live there like it on average, which may not reflect your own preferences at all. And what is surveyed satisfaction anyway? Surely it is not eternal joy but something far more transient. Or maybe those who say they are happy aren’t really happy. In which case, this survey tells you where to go to be around people who will assure you they are quite happy! But it is somewhat reassuring that all the statistics combined explain about 80% of differences across countries, which is quite a bit. Also, the things that make people happy seem plausibly related to what you would expect to make people happy, in a satisfied sort of way.

As for the weights, gdp is, if not the 800 pound gorilla, at least a respectable 300 pounds. explaining about half of the happiness differences across countries. Community matters, measured roughly as either high church or trade union membership. Divorce makes people sad, while political freedom makes them chipper. Large gaps in male/female wages (suggesting gender inequality) are bad for satisfaction. Living longer is good. Raw income inequality and education levels don’t show up at all, once one puts in all the other stuff.

And in the end. Ireland tops the list, with the U.S. ranking somewhere in the low teens. On the other end, let me give you a hint you don’t need. Stay out of Haiti.

* On average, where “place is a country” and happiness is something like satisfaction, not eternal joy.
** I’m not kidding.

31 comments for “The happiest place on Earth

  1. ed
    June 24, 2005 at 12:31 pm

    Interesting. If you look at the pdf file on methadology, you can see how a country’s raw GDP rank compares to its mulit-factor weighted index rank. It appears that the Latin countries in Europe (Italy, Spain, Portugal) rank much higher than their GDP alone would suggest, while the Anglophone countries (UK, US, Canada) tend to rank much lower (with Australia and Ireland as exceptions). Sweden ranks highly as well.

    I wonder what that means?

  2. June 24, 2005 at 12:49 pm

    Speaking English depresses people?

    Alternatively, do English speaking countries tend to have a higher divorce rate or less church attendance? I might guess so but couldn’t say for sure.

    There are lots of nitpicks that one could have with this, but one of mine would have to be the use of latitude as a measure of climate. It appears that it may have been entered linearly, which is a boon for countries like Sweden (and Ireland) that are very northern, even though their climate may not be so great.

  3. ed
    June 24, 2005 at 1:06 pm

    I’m confident the USA has MUCH higher church attendance than Spain or Italy. Not sure about the UK.

    I agree that latitude is a terrible measure of climate. Heck, in Europe, longitude would probably be better than latitude! (It’s much colder in the east).

  4. ed
    June 24, 2005 at 1:12 pm

    …or maybe not…this site, which seems to be based on the same data The Economist used, says Italy has 45% church attendence, the USA only 44%. Although the Italy survey is kind of old (1991).

  5. June 24, 2005 at 2:54 pm

    There’s church attaendance and then there’s church attendance. I wonder if Italy has a fair number of Easter and Christmas Catholics who help out the numbers.

  6. Pete
    June 24, 2005 at 4:31 pm

    Don’t they speak English in Ireland?

  7. June 24, 2005 at 4:45 pm

    There’s an interesting book from a couple of years ago by Robert Lane on why different countries have happier people. Here’s the Amazon link:

    The upshot, as I recall it, was that the competitiveness of market economies, as well as the lack of authoritative leadership that is characteristic of democracy, may end up eroding communities and undermining individual happiness. Somewhat provocative, to say the least…

    But the “Thatcherite” countries (UK, US, Canada) are the ones that have the most unfettered market democracies, and they are the ones that have lower-than-expected happiness scores.

  8. Greg Call
    June 24, 2005 at 7:16 pm

    It’s not particularly surprising to me that there would be more “happiness” in countries where there is single payor healthcare, mandatory four week paid vacations, generous unemployment, mandatory paid maternity leave, guaranteed pensions, geographic stability, and relatively little class anxiety (whether that’s because there is little real class mobility, or because of less consumerism in the culture). And the survey, with all its flaws, seems to support that view. The real question to me is whether such a country can sustain that model for more than a generation or two.

  9. Greg Call
    June 24, 2005 at 7:26 pm

    Just to step back from my comment a bit, I don’t know that all the highest ranked countries have all the characteristics I cite. I only know that the #2 country, Switzerland, does.

  10. Harold B. Curtis
    June 24, 2005 at 8:24 pm

    My happy place is self imposed
    Where I am at, it is reposed
    If I leave it far behind
    It is by me so defined

    If it greets the break of day
    It is because, that’s my way
    If it shrinks for any cause
    It is beacuse I broke some laws

    My happy place is everywhere
    When I help or when I share
    When I lift some burdened weight
    My happy place is every state

    My happy place may come and go
    Like summer rain or winter snow
    In blackest night or brightest day
    Its what I am it is my way

    Rich or poor, bond or free
    Happiness is up to me
    Things acquired do not sate
    If rags or riches is my fate

    My happy place is how I think
    It’s what I eat and what I drink
    Its as close as next of kin
    Or any one that I let in

    My happy place is heaven sent
    Its not my morgage nor my rent
    Not what I drive or where I dine
    But how I treat all humankind

    Harold B. Curtis

  11. June 26, 2005 at 7:18 pm

    Because these comparisons rely on self-reported satisfaction surveys, I think they are measuring *bliss* rather than *happiness*. Self-reported satisfaction dependends on the person’s expectations, so someone with low happiness but low expectations may appear “happier” than someone with greater happiness but higher expectations. While I welcome the observation that we should be more stoic (expect bitter to make life taste sweeter), I dispute the parallel conclusion that ignorance makes people happy.

    On this survey, for example, Spain, where I served my mission and spent three weeks travelling the following year, ranks higher than the United States. Probably 90% of Spaniards live in small apartments (I knew no member that had an apartment with two bathrooms) in dense cities far from the woods and country and grass. And because few people had cars, getting outside the city to see anything besides concrete was an ordeal. The government’s policies have removed Spaniards so far from nature and grass that one of the sister missionaries from the wealthier and wetter northern coast had never seen, with her own eyes, people play on the grass until she went to the Provo MTC. She said she wasn’t in Provo long enough for it not to feel really strange to play soccer on grass — she’d only played on dirt and concrete.

    According to the definition of happiness used in the report, it’s in a Spaniard’s best interest to remain ignorant of the joys of gardens and fields, or to at least never develop the desire to tend a garden or play on grass fields, so that they don’t weigh those values when evaluating their happiness. But because it seems absurd to me to believe that Spaniards are happier because they don’t know the pleasures of cool grass, I think it’s a mistake to think the surveys are measuring happiness. Because ignorance is the numerator in self-reported satisfaction surveys, the surveys actually measure *bliss.*

    I love lawns, gardens, orchards and woods, and would hate to live away from them. People who have never played on grass, or had easy access to trees and woods, don’t expect them — and don’t know what they’re missing. That ignorance can’t make someone happy, just blissful.

    Our goal should be to increase happiness — not bliss.

  12. Greg Call
    June 27, 2005 at 12:03 am

    But Matt, might not some Spaniard say — “Those poor Americans, moving around every few years, with scattered extended families, working more than 40 hours a weeks, taking paltry vacations, constantly striving to make it in the meritocracy, living in cheaply built homes in brand new subdivisions with no history or culture, and living so spread out that no real neighborhoods can form; they have the bliss of space and money, but don’t know the true happiness of culture and history and community and stability.”

    In other words, how do we know the stated preferences of Spaniards are based on ignorance? How do we know ours aren’t?

  13. Jack
    June 27, 2005 at 12:16 am

    I’m going to take a little stab in the dark and suggest that maybe there’s a religious corollary to this idea. Is it possible that Protestants are generally more unhappy than Catholics?

  14. June 27, 2005 at 8:20 am

    Greg, my example of Spaniards was a poke at something that drove me nuts about my mission — I hated being stuck in a city for months at a time. You’re right to point out that ignorance would permeate all of the survey results, including Americans, but I believe that just shows that surveys. on are a poor vehicle for measuring happiness on an absolute scale.

    Just to point out the most obvious quibble with your example, however: I can’t imagine a Spaniard complaining about the quality of American homes, or their touting the 30-year history of their apartment tower over my mom’s subdivision. If money were no object, I think most Spaniards would choose to live in American-style houses in American-style subdivisions. Spaniards with lots of money already do. And neighborhood cohesion and community? Spanish members seemed to know as many of their neighbors as Americans do — not many. But everyone knew who the car owners were, though they didn’t know their names.

  15. June 27, 2005 at 8:29 am


    Your comparison is based on the assumption that happiness is uniquely based on material possessions and environment. If it’s not, then the Spaniards could be happier than Americans in spite of material differences. You’re probably right that happiness and material possessions are linked–but it’s strange for a Latter-day Saint to assume that material possessions and environment are the only determinant of happiness.

  16. June 27, 2005 at 9:09 am


    The materialism correlation is actually reaonably well pinned down in the survey. 50% of the differences in satisfaction across countries can be explained by differences in gdp. No other measure in the survey was anywhere near as important. Naturally one is free to arguethat this is not “happiness” because it is not joy. But it probably captures the broad outline of how having more stuff’s effect on how satisfied people feel. Mind you, that is gdp controlling for all the other stuff. Thus gdp increases undoubtedbly have an average net benefit for life-expectancy, but this benefit of gdp is controlled out of the raw gdp effect. Higher GDP may also encourage breakdown of gender roles, thus raising the divorce rate, but that would also be controlled out of the gdp numbers. Thus, physical and temporal concerns, like how much you make, appear to be incredibly important determinants of differences in happiness across countries.


    It is true that these surveys are limited by the information people have. But there is enough universal about people’s experiences that the survey recognizes you really don’t want to move to Zimbabwe! Are you suggesting that there is an important variable or determinant missing or are you suggesting that the weights on the variables are wrong?

  17. Rosalynde Welch
    June 27, 2005 at 9:45 am

    (Off topic, but just noting that lawns, gardens and orchards are no more “natural” than concrete and dirt, and in the arid west quite certainly less so.)

  18. Mark B.
    June 27, 2005 at 9:49 am

    Well, Rosalynde, maybe dirt. But not concrete.

    It’s not “natural”.

  19. Bryan
    June 27, 2005 at 10:51 am

    Just goes to show that the Irish don’t go to Disneyland enough.

  20. Rosalynde Welch
    June 27, 2005 at 11:08 am

    Neither are lawns, Mark, and I’d wager that a green lawn in Arizona requires significantly more energy and resources to maintain in its unnatural state than a pad of concrete. I prefer the lawn, personally, but it’s no less artificial. /end threadjack

  21. June 27, 2005 at 11:24 am


    You needn’t worry about threadjacking here. In a thread about what makes people happy, almost any subject* is fair game.

    * except literary theory

    But as to lawns, I am not sure that Matt was even thinking that they occurred naturally (in Arizona or elsewhere) so much as they were naturally conducive to man’s happiness. Just as he claimed forests and orchards and so forth are.

  22. Mark B.
    June 27, 2005 at 12:21 pm

    I wasn’t suggesting that lawn is natural in the desert.

    My yard tells me that it’s not natural here in Brooklyn either.

  23. Greg Call
    June 27, 2005 at 1:17 pm


    I think I agree with your general point about surveys. As for my example, I was just stringing together things various non-Americans have said to me over the years. I know nothing about Spain.

  24. June 27, 2005 at 2:14 pm


    Self-reported happiness surveys capture a person’s Perceptions of Happiness divided by their Expectations of Happiness. Someone who’s only a 30 happy, for example, but has an expectation baseline of 20, will report higher satisfaction than someone who’s a 60 happy but has an expectation baseline of 80. The first person will feel happy because they’re happier than they expect to be, the second will report disappointment because their life isn’t as good as they think it should. To be reliable across time and culture, happiness surveys must control for baseline changes.


    I think close emotional relationships are the most important factor in true happiness; then followed by health and material conditions (money, technology, and access to grass and woods!).


    Frank correctly anticipated my response to grass being unnatural. Whether or not something is natural has no bearing on its ability to make people happy. Fluoridated toothpaste and Kentucky Bluegrass in Utah are unnatural, but make people happier, while arthritis and hornet stings are natural, but don’t make people happy.

  25. June 27, 2005 at 2:34 pm


    I never use these surveys so I really won’t go on a limb about what they do or don’t measure.

  26. Rosalynde Welch
    June 27, 2005 at 2:44 pm

    (LOL, Matt, right, I got that. I was just commenting on your off-hand reference to Spanish governmental policies that, purportedly, remove people “so far from nature and grass.” :) )

  27. June 27, 2005 at 3:00 pm


    One natural–and probably accurate–implication would therefore be that people in Spain have closer personal relationships than people in the US. These relationships must be enough to cover the (relatively minor) differences in per capita GDP.

    By the way, I know GDP is covered in these analyses. But the claim that Spaniards must necessarily be less happy than people in the US because of their physical environment makes strong materialist presuppositions.

  28. June 27, 2005 at 3:02 pm

    In my view, a bigger problem with these happiness surveys is simply that the words used in the response categories will necessarily differ in shades of meaning across countries and across languages. Hence, people will answer differently–even if they are equally happy. But there’s no better way to compare psychological variables across countries, so we’re probably stuck.

  29. June 27, 2005 at 3:23 pm

    I think a more reliable happiness measure would ask respondents to rate other people’s happiness, along dimensions of health, relationships, safety, material possessions, etc., and then apply the resulting happiness model to data sets on those particular factors (health, relationships, safety) to create a cross-cultural happiness index.

    RT, I don’t believe this survey allows us to conclude that Spaniards have better relationships any more than that they have better health. Nor can we know that the difference is not due to different expectations of happiness.

  30. A. Greenwood
    June 28, 2005 at 1:17 pm

    Greg Call,
    I agree wholeheartedly with your comment about Switzerland. Benefits and sustainability both.
    Which raises an interesting point. I’ve heard the Swiss are a little clannish. It may be a great place to live for the Swiss but that doesn’t mean you can move there and be a part. So even if the costs of immigration were zero, as in Frank M.’s hypo, I don’t know that the benefits of the different countries would necessarily be fully available to the immigrants. Or even if immigration is compatible with the sort of happiness that is promoted by a folkheimat.

  31. A. Greenwood
    June 28, 2005 at 1:23 pm


    “”Greg, my example of Spaniards was a poke at something that drove me nuts about my mission – I hated being stuck in a city for months at a time.””

    Yah, me too. Luckily I lived in some areas where we could travel to villages. The villages were usually pretty urban too (from the old days of Moorish raids) but at least we could see a little landscape while in transit. I had a picture of my middle-sized ranch home in the middle of the pasture and orchard land that I’d show to people and they were convinced I must be scion of some wealthy family or something. Nope.

    Still, the Spaniards I met were fairly happy, I think. Some very happy, some not so much. But I couldn’t be happy there. My American upbringing has made me unfit for it. It’s a life that, as far as I can tell, is free from striving and therefore from ‘immortal longings.’ Or at least the longings are meant to remain longings with nothing ever done about them.

Comments are closed.