Open threads at the Open Thread tab every Sunday and Wednesday

Things Probably Matter

A while back when I wrote about how China’s economic development might not have increased happiness there much, Scott Sumner wrote a really interesting response, Does Anything Matter?

He points out that it’s too easy to make this about exotic far-off Chinese. Much the same phenomenon occurs closer to home:

If nothing really matters in China, if even overcoming horrible problems doesn’t make the Chinese better off, then what’s the use of favoring or opposing any public policy? After all, America also shows no rise in average happiness since the 1950s, despite:

1. A big rise in real wages.
2. Environmental clean-up (including lead–does Flint matter?)
3. Civil rights for African Americans
4. Feminism, gay rights.
5. Dentists now use Novocain (My childhood cavities were filled without it)
6. 1000 channels in glorious widescreen HDTV
7. Blogs

I could go on and on. And yet, if the surveys are to be believed, we are no happier than before. And I think it’s very possible that we are in fact no happier than before, that there’s a sort of law of the conservation of happiness. As I walk down the street, grown-ups don’t seem any happier than the grown-ups I recall as a kid. Does that mean that all of those wonderful societal achievements since 1950 were absolutely worthless?

But there are exceptions. I recall reading that surveys showed a rise in European happiness in the decades after WWII, and Scott reports that happiness is currently very low in Iraq and Syria. So that suggests that current conditions do matter.

The following hypothesis will sound really ad hoc, but matches the way a lot of people I know talk about their lives. Suppose people’s happiness is normally calibrated around the sort of lifestyle that they view as “normal.” As America got richer after 1950, it all seemed very normal, so people didn’t report more happiness. Ditto for China during the boom years. Everyone around you was also doing better, so you started thinking about how you were doing relative to your neighbors. But Germans walking through the rubble of Berlin in 1948, or Syrians doing so today in Aleppo, do see their plight as abnormal. They remember a time before the war. So they report less happiness than during normal times.

The obvious retort is – modern Chinese grew up when China was very poor. Why didn’t they calibrate themselves to poverty, such that sudden wealth seems good? What’s the difference between a Chinese person going from poverty to wealth, versus a Syrian going from stability to chaos? Might it be a shorter time course? A sudden shock is noticeable, a gradual thirty-year improvement in living standards isn’t?

Probably not. There seem to be a lot of cases where happiness of large groups does change gradually in response to social trends less dramatic than a world war.

First, consider African-Americans. The New York Times calls the increase in black happiness over the past forty years “one of the most dramatic gains in the happiness data that you’ll see”. This is not just about poverty; in 1970, blacks who earned more than 75% of whites were only in the tenth percentile of white happiness. Today, those blacks would be in the fiftieth percentile; they’re still doing worse than would be expected based on income, but not nearly as much worse. This is a very sensible and predictable thing to find. Black people face a lot less racism and discrimination today than in 1970 [citation needed], so assuming that was really unpleasant we shouldn’t be surprised that they’re happier. But notice that this is a time course very similar to the rise of China! It doesn’t look like black people picked a happiness level to calibrate on and then never bothered to adjust. It looks like they adjusted exactly like we would expect them to, even over the course of a multi-decade change.

Second, consider women. In 1970, US women were generally happier than US men. Today, the reverse is true. There seems to be a general pattern around the world of women being happier than men in traditional societies and less happy than men in modern societies (though see Language Log for a contrary perspective). I don’t think of this as a weird paradox. It seems perfectly reasonable to me that having to work outside the home makes people less happy, getting to spend time with their family makes them more happy, and having to work outside the home but also being expected to take care of your family at the same time makes them least happy of all. In any case, the point is that the numbers are changing. Men and women aren’t just fixating on some level of happiness and staying there, they’re altering their happiness level based on real trends, just like African-Americans did (but apparently unlike Chinese).

Third, I was finally able to find a paper that had really good data on change in happiness in different countries, and it supports the idea that happiness can change significantly on a countrywide level.

This is change in happiness in a bunch of countries between about 1990 and 2010 (the years were slightly different in each country). There are other graphs for related concepts like life satisfaction and subjective well-being that look about the same.

The most striking finding is that most countries got happier between those two years – sometimes a lot happier. In Mexico, the percent of people saying they were very happy increased by 25 percentage points!

Just eyeballing the graph, there’s not an obvious relationship between happiness and economic growth – China is still near the bottom like we talked about before, and France – a country that’s been First World since forever – is near the top. Even Japan, which is famous for its decades of stagnation, has done pretty well. But the authors tell us that after doing their statistical analyses, there is a strong relationship with economic growth. Okay, I guess.

They also say there’s a dramatic relationship with freedom and democracy. Mexico, the top country on the graph, went from a relatively closed to a relatively democratic government during this time. South Africa, number five, went from apartheid to no apartheid. Some of the ex-Communist countries like Poland and Ukraine also look pretty good here. On the other hand, other ex-Communist countries like Lithuania and Estonia are near the bottom. I wonder if this has to do with cutoff points – since every country started at a slightly different time, maybe they began sampling Poland during the worst parts of Soviet dictatorship and got Lithuania right in the first euphoria of independence? I don’t know. It all seems very noisy.

They also mention that the United States’ supposedly level happiness is kind of a misunderstanding. People say things like “Happiness in the US has been flat from 1950 to today”, but in fact it declined from 1950 to 1979 and increased from 1980 to today. They attribute this to the 1950s being unusually happy; then the 60s and 70s being unusually conflict-prone, and the Reagan Revolution and Clinton years were back to being optimistic. They don’t have data that stretches too long after that.

(This is pretty neat for Reagan and Clinton. When I die, I’ll consider my life a success if people attribute a spike on national happiness graphs to my influence.)

So apparently population happiness levels do change in response to relevant social changes, even on multi-decade timescales. Which brings us back to asking – what’s up with China?

The graph above shows India as doing okay – not great, but okay. But a similar graph on subjective well-being – which should be another way of looking at the same thing – shows India as doing pretty poorly, right down there with China – even though its GDP per capita quadrupled during the period of study.

I see a lot of conflicting perspectives about whether economic growth increases national happiness. It may, but the effect isn’t as big as you’d expect, and is usually overpowered by other factors. Maybe it isn’t even direct, but has something to do with development increasing democracy, liberalism, rule of law, and stability. China got the development, but its happiness genuinely didn’t increase because of country-specific factors that have something to do with how it developed (inequality? pollution? authoritarianism?).

This matches the race and gender data. Blacks saw a big happiness boost during a time when their feeling of freedom (but not their income) increased relative to whites. Women saw a small happiness drop during a time when their income (but not their feeling of freedom) increased relative to men.

So it looks like happiness can change. It just didn’t change in China over the past thirty years. The apparent paradox of improving economic situation and stable/decreasing happiness is genuinely paradoxical. Intangibles are probably just way more important than money, even amounts of money big enough to raise whole countries out of poverty.

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

315 Responses to Things Probably Matter

  1. Mark says:

    Hi Starslatecodex.

    Could you please check out this book on psychotherapy. I swear this will be one of the best books you have read. Its groundbreaking and rational. There is nothing like it.

    I swear im not selling you anything. I think this book will change the way you look at mental health and help allot of people.


    The whole book is free. Please read it. You will thank me later.

    I dont have time to explain it all. I know this message may seem absurd. Im begging you to trust me. This is about helping others.


    • anonymous says:

      I’ve heard about this. It’s apparently in the midst of some actual trials (like CBT did) in the UK I think? Will check it out and probably leave some notes/a summary here later.

      • anonymous says:

        I’ve gone through the first few chapters and this seems to be very similar to NLP in that the discoverers found a potentially useful tool, flipped their shit, and decided to insist THIS EXPLAINS EVERYTHING. Their enthusiasm then prevented rational inquiry into the thing they discovered. Claims of meeting much higher standards of evidence than CBT and other forms of therapy do not look legit on a first pass. Still some interesting content. Summary to follow.

    • Scott Alexander says:

      Are you that Richard guy who was always pushing perceptual control theory in an equally hamhanded way on LW?

      • Vaniver says:

        This seems like a weird response. PCT seems like the best model of human reasoning to me, and so I wouldn’t be surprised at all if it was deeply relevant to psychiatry. (I would be very interested in you reviewing that eBook.)

        I agree with you that Richard Kennaway didn’t present it the best way (that’s why I wrote my own introduction to it), but I have seen a lot of things, especially in the LW sphere, which were gold but presented terribly. My go-to example is probably Direct Instruction, a method of scientifically constructing and testing lesson plans in order to teach as effectively as possible, which was explained in an unscientific and meandering way. (I could keep going–like the time Viliam thought one of MIRI’s major donors was a scam, instead of a professional gambler who speaks English as a second language.)

        Maybe you come across more noise than I do, but the quality of ideas and quality of presentation aren’t correlated all that well, and so I think it shouldn’t play a big part of your filter.

        • anon says:

          Seconded. The explanatory power of PCT seems to be very strong. I would like to see your thoughts if you have strong convictions for/against the theory.

          Perception-error modification is like noisy gradient descent in the brain!

          • anonymous says:

            the explanatory power seems specious, the cited figures of .96 correlations doesn’t pass a sniff test. First guess is they are double counting the explanatory power of variables that actually funge against each others’ explanatory power. Similar thing happens in using blood markers to predict CVD risk if the researchers aren’t careful. Can’t remember the name for this phenomenon.

        • Lumifer says:

          that’s why I wrote my own introduction to it


        • Viliam says:

          like the time Viliam thought one of MIRI’s major donors was a scam, instead of a professional gambler who speaks English as a second language

          For the record, the only evidence I have for the whole thing not being a scam is your word, Vaniver.

          And when I asked you whether you are participating with the guy in the “donating to MIRI through you” project, you said no.

          Later I read somewhere in a LW comment (maybe written by you, maybe not, I don’t remember) that the guy is a big MIRI donor. Evidence wasn’t included.

          Based on this, I updated my estimate that this is a scam from 99% to about 70%. Maybe I should have updated further; I don’t know. I expected that things will later sort themselves out… either by someone reporting they successfully donated to MIRI using the proposal, or by someone reporting they got scammed… and as far as I know neither of this happened.

          Is there any other publicly available evidence I missed?

      • Richard Kennaway says:

        I don’t know who he is, but he is not me.

        I’m not “anon” either. In these columns, I am no-one but myself.

        BTW, I don’t recognise your description of the postings I made about PCT, which were few and a long time ago, and unlike Mark’s empty hyping, did actually contain the ideas. (In saying this, I am not seeking a conversation on the subject, just putting down a marker.)

        • Mark says:

          I dont understand how people see Mark and think Richard.
          Ha, Ha, Ha

          I heard you liked P.C.T. I have a psychotherapy book based on it. The creator was close friends with Will Powers. I can not figure out how his books have been neglected in P.C.T discussion. He has published plenty of papers on P.C.T and has alot of connections in the field. See for yourself.

          The book is free.

          -If anyone still has yet to get the message Richard is not me.-

          • Richard Kennaway says:

            I knew Bill Powers myself for many years. (“Bill”, not “Will”, for those on first name terms, despite the opportunities for nominative determinism provided by the latter.)

            However, although I’m somewhat familiar with the method of levels, there is nothing I can usefully say about the book, even at the risk of “pushing” it “hamhandedly”. I know nothing about psychotherapy and Scott is a professional practitioner. I’m aware that there have been studies on MoL, but I don’t know the results and have no references.

          • Mark says:

            Hi Richard Kennaway,

            Good point. I should post a link to some references and studies about M.O.L. They range from case studies too trials and can be found here. There are still plenty more from other researches that are floating the web.


            For more general P.C.T psychology:

            Dr.Tim Carey has gotten back to me for every question ive ever asked. So if you have any questions/points I am certain he will answer back.

            Ive also seen some P.C.T researches get in debates online. This one might be helpfull to look at and it shows what M.O.L looks like under a critical spotlight/rational inquiry.


            (Look at the comments section)

        • Scott Alexander says:

          All right, sorry, that was uncalled for and my memories of that time are hazy.

    • Torvetrog says:

      It’s about Method of Levels, a therapeutical strategy based on perceptual control theory. After skimming the first few results of a Google Search i have to agree: this is important. Just understanding the concepts behind it made a lot of things click in my mind and raised my current level of happiness considerably, so it’s very relevant to this thread.
      Ham-handed or not, it was very useful. Thank you!

    • Anonymous says:

      You type like someone who’s extremely excited because they finally found an ebook with conclusive proof of the existence of Bigfoot which, if only people would read it, they would find impossible to deny.

      If what’s behind that link is, in fact, a genuinely novel and groundbreaking psychotherapy method with demonstrated way-better-than-placebo-and-SSRI results in early clinical trials and it works in one month and it has a 0% relapse rate, etc etc, please consider spending an hour collecting your thoughts and writing something a little more informative and persuasive – I’d be interested in reading about something like that, obviously, but if I followed every link that came with recommendations like “I swear I’m not selling you anything” and “This will be one of the best books you have ever read” I wouldn’t have time to read slatestarcodex, let alone read the comments section and see your comment

      e: I’m not meaning to just be rude by the way. It looks like you really do care about this new psychotherapy technique and if you plan on throwing out links to it and begging people to read about it, you might as well write up a nice quick intro to it so people have a reason to click your links other than “some guy on the internet says it’s amazing”

    • dsp says:

      So, I guess I’m a little late to the party, but I thought I’d add, this was not one of the best books I’ve ever read. It wasn’t even rational. The key insight boils down to “it’s good to encourage clients to reflect (also, don’t be Freud)”, which is fine enough so far as it goes, but hopefully obvious to anyone who is even trying; and it loses some of its lustre when you try to wring a hundred and fifty pages of pseudoscientific formalism out of it. What’s more, this “perceptual control theory”, while basically no worse a formal model for cognition than all the others to which it is logically equivalent despite repeated claims of superiority (such as stimulus response, with respect to which the author has apparently failed to notice that ‘perception’ is a synonym of ‘stimulus’ and the ‘control’ action is, well, a response), is actually a really really unhealthy way to think about your own mind.

      • anonymous says:

        Interested in the last sentence, can you expand on that a bit? It seems like thinking of some of your sub processes as a bunch of little thermostats that try to keep things in some sort of homeostasis is a useful model and I can’t think what it harms. It reminds me of the law of equal and opposite advice.

        • dsp says:

          I want to stress that I chose the words “way to think about your own mind” with some precision. It sounds like you’re talking about treating it as a flexible, intuitive characterisation, almost a gimmick – a sort of inverse personification you might use in some situations as an informal model.
          That’s probably fine, as far as I can tell. I’m not sure it’s particularly useful.
          Where I’m suggesting things could go astray is if you started to habitually think of your mind as working that way, not literally but perhaps the next thing over; I mean, if you start to believe that your brain is really closely approximated by one of those flowcharty PCT diagrams, the way the book strongly advises, you are going to damage yourself. The tendency to consider your mind abstractly as an integrated, fluid will is essential for it to function properly, and it seems to be possible to break that in ways that seriously impair your ability to introspect productively.
          (If you would like some broad speculation as to why that happens, my theory is that it’s because it’s impossible for any finite system to specify its own state beyond a certain threshold granularity, as it always takes more degrees of freedom to store that information than exist in the system being described, which implies that it’s fundamentally impossible for any brain to understand itself very precisely, and you can run into that limit even if the model is inaccurate as long as it is sufficiently complex and granular.)

          • Ricardo Cruz says:

            You can add degrees of freedom to your brain by simply using a computer… or a whiteboard 🙂

          • dsp says:

            Ricardo: — which you would then need to include in order to fully specify the system, requiring more degrees of freedom to describe those too.

          • Ricardo Cruz says:

            That’s like saying you cannot understand a computer because it can have a stupendous number of degrees of freedom. Maybe not infinite, but certainly more than you can fit in your brain. You don’t need to know every brain state to understand the brain. In fact, much incredible work as been accomplished in a short amount of time.

          • dsp says:

            Exactly, because you can use abstractions. A brain can’t know its own state in detail, it can only have a vaguer impression that takes less information to specify. If you try to make the abstraction more granular, it becomes less useful and less accurate. Which is what I said to begin with.

    • FullMeta_Rationalist says:

      I started reading the book. But it had a low signal to noise ratio. So I started skimming it progressively faster. There’s little in it that I couldn’t have gleaned from the wikipedia page. It reminds me of when I wanted to learn about the Flow once. So I borrowed a library book written by Mihaly Csikszentmihalyi and was disappointed to find nothing actionable. Anyhow.

      * * *

      MoL tl;dr
      a) Go meta (aka self-reflection).

      In the MoL book, the only actionable advice is basically “to figure out what’s bothering you, think on the meta level”. Which is (like CBT) excruciatingly obvious to the rational sphere, but perhaps not to the average Joe (or Scott’s attending physician).

      On one hand, I feel this book is further confirmation that psychotherapists do nothing which a dialogue with a close friend couldn’t accomplish. Jim Carey and Will Powers are fully cognizant of this, as the book spends the first 20 or 30 pages explaining that psychotherapy is mostly a sham. Very similar to what our host has been preaching. Maybe psychotherapists serve a role that the pastor/shaman used to fill in more traditional communities. (Obligatory Szaszsz reference.)

      On the other hand, I’m reminded of an occasion when a friend was distressed over a decision regarding a romantic relationship. Apparently, he had asked other friends for advice. And these friends had responded with unproductive cliches like “whatever happens was meant to be”. So maybe laypeople are not qualified to play psychotherapist after all.

      On the third hand, he and I did eventually get to the bottom of things after an hour of metacognitive discourse, at which point his conflict was conclusively resolved. So at the end of the day, I dunno — should I count this anecdote as evidence that laypeople (like myself) are qualified to be psychotherapists? or should I count this as evidence that laypeople (like his other friends) do not have enough common sense to be psychotherapists.

      * * *

      PCT tl;dr
      b) Feedback Loops are a thing.
      c) Neural Networks are a thing.

      To laymen, b) perhaps seems obvious. I think its value primarily lies in the fact that it’s a scholarly augmentation of B.F. Skinner. So when a friend is like “pavlovian conditioning”, you can respond “yeah. But actually… FEEDBACK LOOPS” then cite this book.

      But again, maybe these sorts of common sense ideas really do need to be explicated with a 150-page e-book. I’m reminded of how Scott Aaronson once went to a conference regarding a certain cognitive theory that “consciousness = interconnectivity of a system”. He was the only speaker who offered criticism of the theory.

      I admit I am intrigued by the relationship drawn by the book between error-correction and attention.

      Also. The book reminds me of Kevin Simler’s review of The Bicameral Mind. In that they posit the existence of sub-agents trying to satisfy incompatible goals.

      • Mark says:

        Hi FullMeta_Rationalist

        Im glad you read the book. Good summary.

        Tim Carey does repeat himself a few times for the sake of getting people to understand some very important ideas (he says this in the beginning of the book). The normal person/psychotherapist is unfamiliar with what he says.

        Also as obvious as “going meta” is, it is rare that we actually do this when were in conflict. Most times we don’t even know were in conflict. Since you read the book im sure you heard about background thoughts/disruptions. These are very hard to catch by yourself.

        Finally im glad you tested out M.O.L and hope you do it again. I hope doing this gives you an idea about how right/wrong M.O.L is.

        Now for some questions;
        1. Do you think the theory of M.O.L is right or wrong.
        2. Do you think the future of psychotherapy and mental health care lies in the direction of M.O.L/P.C.T
        3. Would you recommend the book to Starslatecodex.

        To learn more about M.O.L here is a list of case studies and trails.

        Theres some other stuff just floating the web too.

        *Also the site contains many videos of M.O.L in action. If you have any questions or criticisms of M.O.L/P.CT you can email Dr.Tim Carey at the site and I am certain he will get back.

        • Lumifer says:

          Do you think the theory of M.O.L is right or wrong.

          Is this theory falsifiable in the Popperian sense?

          • anonymous says:

            Only sort of. The thing Gendlin’s Focusing, MoL and NLP have in common is that they state that *something* special occurs in psychotherapy that works vs psychotherapy that doesn’t across multiple schools of psychotherapy. If it turned out that there wasn’t a true invariant but rather it wound up being a smattering of different parameters (with maybe something like a threshold effect) that would be evidence against them. They might then retreat to the claim that there are one-three ingredients that tend to be outsized in effect. This is the more conservative claim that I think I agree with.

        • FullMeta_Rationalist says:

          1.1) In case you’re asking whether I believe the mind is (to at least some degree) organized hierarchically.

          Of course. Why wouldn’t it be? As far as I’m concerned, that’s the Null Hypothesis. Or at least it should be for any reductionist who’s put a modicum of thought into this. Possible alternative hypotheses include: the mind’s architecture is as flat as a pancake (we’re all nematodes); the mind’s architecture is a parallel forest of dissociative identities (we all have Mutliple Personality Disorder); the mind is a linear automata (BF Skinner); the mind is an incorporeal monoid (Dualism).

          To paraphrase Lumifer, exactly in what way is this theory falsifiable? In order to interest me, the claim is going to have to be a little more specific. Which the book does take a stab at during page xii, but then handwaves away the onus probandi with “Just for the record and without getting further into details (…)”.

          1.2) In case you’re asking whether I believe the technique is useful.

          I mean, it’s certainly better than “whatever happens is destiny”. However, it fits my Hollywood stereotype of what psychotherapists do already. Maybe you’ve seen Freaky Friday, that movie where a control-freak single-mother and an angsty teenage daughter involuntarily switch bodies. The mom is a professional therapist. Which means Lindsey Lohan has to pretend to be a therapist. So the mom is like “just say keep saying ‘and how do you feel about that?'”. Like, could this not be classified as MoL?

          I guess the source of frustration is that psychotherapy must continue to point towards trivial, low-hanging fruit while a large swath of the population (including the professionals) continually and adamantly bark at a scarecrow. In other news, I think I now know what it feels like to be Alan Kay.

          2) I’m no psychotherapist. So I don’t have enough information to make a confident prediction. Most of my psychology opinions are formed through the lens of our caliph, Scott. So whatever opinion I come up with will probably just be a proxy for his opinion.

          That said. Until psychology finds something genuinely novel, psychotherapy should probably just stay the course. E.g. self-reflection, the hostessing of salarymen, and the injection of metaphorical-lithium into the water supply.

          3) Would not recommend. Instead, I recommend a 10-minute safari through the PCT wikipedia page.

    • Outis says:

      I read (most of) the Wikipedia article on PCT.

      – A car cruise control system is a piece of unthinking matter. Each segment of the loop affects the next by pure physical determinism. Intentionality, and therefore control, can only be attributed because we know that the system was designed by a human engineer. The engineer wanted the system to control the car’s speed and it does, therefore what it controls is the car’s speed. Not perception (the readout of the car’s speed), nor behavior (the opening of the gas), but a piece of reality. Alternatively, one may focus on direct causality and say that the system controls the opening of the gas (behavior), with the goal of controlling the car’s speed. What nobody says is that the system controls the readout, because that’s retarded.

      – Therefore, the whole diatribe about “controlling perception, not behavior” is completely unnecessary and a waste of everyone’s time. The terminology was arbitrarily picked to maximize the impression of surprise and counterintuitiveness. It’s clickbait, essentially.

      – What if there is no loop? For example, I am flexing my biceps right now. Assuming the PCT model, I am changing the reference point, and the motor neurons are firing, but the loop is not closing. The arm is not moving, no angle is changing, and the feedback stimulus does not move any closer to the reference point. The neurons are going to keep firing in vain as long as I will it (or until I get tired). My arm control system is clearly not controlling its perception of the arm’s state: how can something be said to control something else when it has no effect on it? All it is controlling is the contraction of the muscles, i.e. behavior.

      – The notion that matching the behavior implies that you have matched the internal structure is completely wrong.

      • Agronomous says:

        A car cruise control system is a piece of unthinking matter. Each segment of the loop affects the next by pure physical determinism.

        That used to be the case—then some fucking idiot decided to put a computer in the loop.

  2. What evidence is there that these changes are real, and not just changes in how people report their levels of happiness? It seems plausible to me that cultural differences — varying across both geography and time — could cause a lot of variation in the level of happiness reported on some 1-10 scale even if the actual levels of happiness remained unchanged. It’s not clear to me how you would separate that kind of thing from the presumed “true” underlying happiness level.

    • Wrong Species says:

      That’s exactly what I was thinking. Chinese people probably think about happiness in a much different way than Americans. We can’t really trust any of the data, and probably won’t be able to until we can solve the hard problem of consciousness. But of course, we would like to something until then. Is there any other way we can try to measure happiness? What if we look directly at the brain?

    • Peter says:

      One specific thing: a large part of contemporary Facebook Meme Feminism is the “don’t tell me to smile” variety. This suggests that there was and is more pressure on women than on men to present themselves as cheerful, and that an aspect of some varieties of feminism is resisting that pressure.

      Considering that if you look at the Language Log articles what you’ve got is “a shift of a few percentage points, mostly accomplished by shifting the opinions of around 5 women in a hundred from “very happy” to “pretty happy””, it seems plausible that something like that may be confounding things. (If you look at their graphs, the phrase “lost in the noise” springs to mind).

      • Just in case anyone is wondering, I’ve had strangers on the street tell me to smile. Snapping at them cheered me up amazingly. I’m not sure that seeing their faces fall was an essential part of the experience, but it certainly didn’t hurt (me).

        • TPC says:

          I’ve told men to smile and that cheered us both up. I think there is some selection bias operating in the trope that women are hectored excessively to smile.

          • Liskantope says:

            I’m a man and I’ve been told to smile in the street a total of one time in my life. I found it fairly off-putting, and the idea of getting that treatment frequently boils my blood.

            I don’t know what the actual frequency of this for men vs. women is. I’ve only ever asked one (very conventionally attractive) female friend about this, and she said that she hadn’t even realized this was a thing, but that she always smiles when she passes strangers anyway. Still, a trend of telling women to smile seems very compatible with traditional gender dynamics as I understand them, whereas a trend of telling men to smile does not.

      • Liskantope says:

        I completely agree that Peter’s idea seems plausible. I’d like to point out that the flip side of this phenomenon is that women today are more likely to be told that they should be upset and feel oppressed wherever gender is concerned. This may be contributing to lower level of reported happiness.

        (Or maybe I’m just pulling this out of my ass. I wasn’t alive during the 1970’s, during which I’m sure the feminism movement was at least as vigorous as it is today, although its flavor was obviously very different. But I imagine the general vibe of the early 70’s to be “we’ve been oppressed, but now we’re getting liberated!”, while the sentiment of today’s online feminism seems rather less positive.)

        • I’m just one data point, but an area where feminism has made my life emotionally worse is talk about women pedestrians giving way too much to men. Some of them talk about just not giving way at all until they bump into men.

          I don’t know whether I’m mostly giving way. I don’t know what I’d rather do.

          If I care, what I should probably do is spend a while observing before I try to change anything.

          What has actually happened is feeling as though I’m obliged to make a decision about my sidewalk behavior, and it makes the mere sight of a man in my path of travel infuriating, and I don’t mean I’m angry at him, I mean I’m angry at feminists who’ve made just walking in a city much more work than it was.

          At this point, I’ve dropped any concern about the issue because I think it’s more trouble than it’s worth.

          • Liskantope says:

            I imagine that recent feminist activism against “manspreading” on subways has had a similar effect on some women who live in big cities.

          • Possibly, but the difference I see is that taking up more than one subway seat is well-defined, while how much giving way is polite isn’t.

          • Anonymous says:

            Any kind of thinking about or awareness of how you walk down the street will make your life worse, I find. Once you start thinking about it you disrupt the unconscious process.

            In Impro there’s a great analysis of this whole thing; actually that book has the best analysis of status in human behaviour I’ve ever seen. It’s not just gender. Low status people give way to higher status ones. Men are higher status in our culture so women end up giving way to men quite often, but there are other components to this; age vs youth is a big one. Elderly Germans will mow you down in their path because older people are high-status in Germany.

            It was extremely enlightening. But it made walking down the street really suck. I kept bumping in to people because I was thinking about it too much.

    • Matt James says:

      This was my thinking, as well. If someone conducting a poll asked me if I’m happy, I’d probably pause to contemplate the question long enough that they’d get frustrated and end the conversation. And maybe eventually answer, “I don’t know, how do I know?” or something otherwise potentially ridiculous, potentially true, potentially both.

  3. “a gradual thirty-year improvement in living standards”

    A twenty-fold increase in real income over thirty years (roughly the Chinese experience, I believe) is not gradual.

    • onyomi says:

      The first time I visited China was 2002. The second time 2008. It was almost unrecognizable. And I’m sure people who visited in the 80s and 90s would have said the same about 2002 China. I joke that China is like a puppy–every time you see it it’s completely different.

      That said, it probably feels more gradual when you’re living it–but Chinese people have the same impression. I have a friend in his 30s who grew up in a small town. He says his hometown is completely different from his childhood memories of it.

  4. Dues says:

    I’m betting the gender imbalance that appeared in China due to sex selecting abortions was enough to blow away any economic gains among men. (Not being able to get a date should probably affect happiness.) But unless I look at changing gender ratios and happiness rates among men and women, that’s just a guess.

    Also the Chinese people I know say that economic boom mostly hit the cities, not the rural areas. Could that affect things?

  5. Tedd says:

    Measuring happiness is really hard, it turns out. Slight changes in wording, translation differences, and linguistic and cultural drift make results unreliable. As a sort-of control, poll people on how healthy they are, which there are at least objective proxies for; iirc, people also do not generally report being healthier than in the past.

    • Wrong Species says:

      That’s like saying lets just measure how wealthy people are. That gives us better data but it’s not answering our question.

      • yarbel says:

        I think Tedd meant something else. Suppose that people report the same level of health as 50 years ago — but you know that their life expectancy and quality has risen by looking at objective data. Then you know that there is something very problematic with self-reporting.

        Another possibility is that you will find a certain bias in how people self-report bias and you could, presumably and arguably, make the case that you can use this bias of known magnitude to correct for the (assumed) bias in happiness reports.

        So yeah, not bullet proof, but also, not useless.

      • Samedi says:

        it’s not answering our question

        Which is what exactly? “Happiness” does not have an agreed upon meaning. How can you measure something when you don’t even know what it is you are measuring? One way to deal with this common problem in the social sciences is to use an operational definition. An operational definition of happiness would at least give us an unambiguous, if limited, meaning of term.

        My general rule of thumb is that if you cannot assign a unit of measure to your dependent variable then you are are not doing science.

    • Jacobian says:

      I also think that when people are asked “rate on a 10 point scale how happy you are” they answer a simpler question, which is comparing their happiness to hours. After all, we go around thinking “my job sucks” and not “I feel 4/10 happy today”. This means that people can become massively happier, in general and minute-to-minute, and yet report little-changed numbers on surveys.

      • Luke the CIA Stooge says:

        The entire idea of happiness seems so noisy as to be a useless concept, as far as quantitative measurement is concerned. Like i would expect the effect from asking questions in the morning or evening, before after dinner, before after a smoke, to be considerably higher that almost any macro level change (ie poor/rich, healthy/unhealthy, free/unfree) the only exception would be be things so big and culturally present that it affects the answer your EXPECTED to give. For example you’d feel bad and conflicted telling white survey takers that you where happy if you were black in the civil rights era, or telling the international community you were happy if you were in the midst of the Syrian civil war.

        So for a few culturally high profile things the “right” answer is expected of you, and any other thing (long term trends ect.) will be drowned out by how close you are to your smoke break.

      • Garrett says:

        It reminds me of the pain scale used in hospitals.

  6. Tom Davies says:

    I thought the 80’s onward was when the stagnation of middle class incomes started in the US — so why did happiness rise then after falling when people were doing better?

    • No Vietnam war and people felt more confident about the Cold War?

    • No, the stagnation begins in the early 70s, I believe (depending on measurement).
      Additionally, the 1970s, and early 80s, had several nasty recessions, along with the OPEC Oil Crisis, along with stagflation.

      The 1980s has a big return to stability. Also, since more women enter the workforce, the family feels richer (even if median hourly wages stagnate).

      The late 1990s was a good time for worker wages, too.

    • onyomi says:

      Contrary to all the “Reagan and Thatcher ruined everything” talk you hear in academia, the 80s and 90s, and especially the Reagan years, were years of high satisfaction and optimism among most people in the US. Reagan did win 49 states in 84, after all. Can one even imagine that level of satisfaction with any president today?

    • Luke the CIA Stooge says:

      Also its only median wage that’s stagnant (and then only in the states), total expense on employee has gone up (your healthplan is way better today than the 80s), tech has improved dramatically (it used to be you’d have to read academic articles to have conversations like this), and buying power has improved (i remember my mom telling me that when she got rid of her old junker car she was able to just buy a new pair of boots, now you could buy 5 pairs of boots just for the price of the scrap metal). THe quality of everything has improved dramatically as well as the variety (there would have been barely one sushi place in my city in the 80s that produced sushi as good as my discount sushi place) Additionally median wage is probably skewed by more irregular working arrangements (again more options), household income, the thing that actually determines how wealthy you and the world around you are likely to be, has never stopped growing.

    • Hector_St_Clare says:

      Stagnation started in 1975.

    • Maware says:

      The oil shocks were over, for one:

      Not being forced to get gas only on even or odd number days helped. Also, the 1980s heralded the explosion of information technologies-cable TV, computing, home recording, affordable personal electronics.

      • Edward Scizorhands says:

        In retrospect, rationing by even- or odd-numbers was really dumb. The whole thing was caused by price controls. These days, we just let gas prices rise and there is always enough gas (with a few exceptions).

        There are occasional calls to go back to stupid policies like that. Trump and Sanders would both easily back a plan like that because it’s easy to get the populists to encourage it in the short term.

  7. Wrong Species says:

    Has anyone looked in to whether the people doing these surveys have looked in to cultural differences and ways around it? Surely they must have thought of that.

    • onyomi says:

      Even if Mexicans are culturally predisposed to say they’re happy more than Japanese, presumably the culture surrounding talking about happiness in Mexico and Japan didn’t change dramatically over a 30 year period, so we can still compare the change?

      My concern is that self-reported happiness probably correlates more strongly with an intellectual sense of whether things are going well or not, as opposed to daily lived experience, and that that has a lot to do with the media, actually.

      I wouldn’t be surprised to find black happiness dropping dramatically very recently because all the news reports are about how, if you’re black, policemen will just shoot you for no reason. The likelihood of being shot by a policeman for a black person has probably actually gone down recently (due to scandal-wary police), but awareness of the possibility has gone up, and that’s what matters for something like self-reported happiness.

      • LHN says:

        Even if Mexicans are culturally predisposed to say they’re happy more than Japanese, presumably the culture surrounding talking about happiness in Mexico and Japan didn’t change dramatically over a 30 year period, so we can still compare the change?

        Can we presume that? Based on my observations of culture and history, I think it’s pretty clear the culture surrounding talking about emotions generally changed noticeably in the US between, e.g., 1945 and 1975. (Between 1965 and 1975, for that matter.)

        I don’t know enough about Japan or Mexico to have an informed opinion, but Japan at least seems to have undergone noticeable cultural changes in the last generation.

      • I’ve seen a couple of things by black men who’d been taught that they’d be safe from the police if they were very respectful and compliant. They were horrified to find that the strategy they believed in wasn’t good enough.

        • Cliff says:

          It’s not completely fool-proof but it’s pretty damn good. Also the same thing goes for all races.

        • Richard says:

          I am somewhat curious how you define “Safe from the police”.

          If you find yourself in circumstances where the police has a reason to suspect you of a crime, you will obviously be treated accordingly, but being compliant and respectful will make the process less unpleasant.
          I don’t see how that is dependent on race, perhaps with the exception that black men may find themselves in such circumstances more often than other demographics.

          • The Nybbler says:

            The BLM narrative is that police are exclusively or at least extremely predominately a threat to black people; that white people are safe around police and furthermore aren’t and don’t need to be taught to act deferential towards them.

            The first part is false; if police are racist in their killing it is not to the extent that a black man must rationally feel mortal fear around them but a white man should rationally feel safe.

            The second part is laughably false.


            I consider it relevant that black people seemed to be generally more afraid of police before the videos came out. It’s something people could be wrong about (including that black people might be appropriately afraid while white people are too trusting), but it’s reasonable that people would have pretty good (mostly local) information through their experience and social networks.

            Why do you start by assuming that police don’t abuse their power?

          • Richard says:

            Your linked article was interesting, but I don’t see how it proves anything beyond police officers being able to feel stress and act rashly under said stress which is a problem other than racism. (And one that I believe most police forces are working on to the best of their ability.)

            Why do you start by assuming that police don’t abuse their power?

            I start by assuming that police are sadly lacking in power to abuse, which is almost the same, but not quite.

            The police are operating on uncertain and often flawed information, in a stressful and potentially dangerous situation. They have to make split-second judgement calls that are later scrutinised by investigation teams that have never been in a similar situation and have all the time in the world to mull over details. This is not an enviable position to be in.

            Frankly, I’m amazed things don’t go wrong more often than they do and believe the west is blessed with police forces that are much more competent than we have any right to expect, given what we pay them.

            The reason I am willing to extend a certain slack to the police is mostly from personal experience. I have not been a police officer, but I have been in combat situations and will use that as a gauge for what is reasonable behaviour on the part of the police.
            I have also had the misfortune to be arrested on a few occasions and I will summarise the most dramatic of these incidents here as an example:

            I was sitting in my car idly wondering what was holding up traffic when the answer came in the form of a SWAT team descending on my vehicle. This would not have made my top ten list of possible reasons for a traffic jam before the event. I remained silent, compliant and calm during the arrest. The silent bit is more important than you might think; speaking when not spoken to in a stressful situation only adds to the stress. I was roughed up some, enough to require stitches, but not shot.

            During interrogation, it turned out the police were acting on credible but ultimately flawed information that I was a terrorist about to set off a dirty bomb in a major city. (this was not long after 9/11 so the paranoia factor was high) Given their information, I would not have held it against them if they had shot me, as from their perspective it would have been a perfectly reasonable thing to do, especially if they felt they were losing control of the situation. (which I did my best to prevent by said calmness and compliance)

            After 48 rather unpleasant hours while they cleared up the misunderstanding, I took the arresting team out to the pub for a few drinks. This was appreciated.

            After this, do I consider myself “safe” from the police? Yes. Do I consider myself safe from paranoid nutjobs who report suspected terrorism? Not really, but I’m not sure if anything can be done about that.

            From my perspective, the main difference between my experience and the ones related in the media these days seems to be that I am not angry about it.

            Edit: all these meandering paragraphs were essentially just to say that the bar that police should meet is:

            Given the information available at the time, and the time allowed for deliberation (typically .1 sec), is this a mistake I could be absolutely certain I would not have made myself?

            All the evidence I have seen tells me that the police are usually better than me, so who am I to complain?

          • j r says:


            The BLM narrative is that police are exclusively or at least extremely predominately a threat to black people..
            The first part is false…
            The second part is laughably false.

            Of course those narratives are false. You made them up.

          • I’m dubious about those “haven’t been there” arguments, since you can usually find people who have been there on both sides of a disagreement.

          • Lumifer says:

            @ Richard

            I am somewhat curious how you define “Safe from the police”.

            Imagine yourself interacting with someone who’s in foul mood and you’re are a bit cheeky and not terribly respectful (though not rude).

            Now compare the case when that someone is a cop and when that someone is, say, a municipal clerk.

          • Dahl says:

            Should anyone rationally feel ‘safe’ to be disrespectful and cheeky to someone in a high risk and high stress job and whose contact with the public is often (although not always) dealing with the worst and most dangerous people in the community? And whose job sometimes (although not often) sees them facing legitimate threats to their life and limb from said members of the community?

          • Lumifer says:

            @ Dahl

            Yes, anyone should feel safe in the presence of a public servant paid out of his taxes.

            If the said public servant cannot differentiate between cheeky and threatening, he doesn’t look to be qualified for the job.,

          • Mary says:

            On the other hand, people should not be going about being cheeky and disrespectful to strangers of any stripe, on account of its being rude.

          • John Schilling says:

            Yes, anyone should feel safe in the presence of a public servant paid out of his taxes

            Certainly, in the same sense that a virgin with a bag of gold about her neck should be able to walk naked across the realm, safe and unmolested.

            Very nice as an aspirational goal or a moral demand. In any actual human society, virgins are advised to go clothed and keep their bags of gold hidden even though they “shouldn’t have to”.

          • The Nybbler says:


            The police are not lacking in power to abuse. They have the power to commit violence on anyone who is not a member of the political elite, with impunity; the entire system will back them up. The scrutiny they receive after such violence … well, the line from the Declaration of Independence about protecting them by means of mock trials comes to mind.

            And most (white) people are perfectly OK with this. Look at you; you were as compliant and deferent as any tyrant could demand, and your encounter with police STILL resulted in injury great enough to require stitches. And you’re OK with that, and bought your abusers beer afterwards. That’s something akin to Stockholm Syndrome, if not the thing itself.

          • Dahl says:


            So you think it’s rational to expect that any member of the public can at any time approach someone in one of the highest risk, highest stress jobs in modern society, and be cheeky and disrespectful (while having no awareness of whether said public servant has just had an altercation with someone carrying a knife, or has just intervened in a domestic safety matter where they were responsible for physically removing a child from dangerous parents for instance or more stressful things) and the public servant should be able to guess why you are approaching in such a manner and, using their famed ‘public servant perfect insight’ (PSPI – such a common term i should really start using this acronym), respond in such a way as for any and all members of the public to feel nothing less than safe in doing so.

            And all this because people pay tax? And the public servant is paid a salary from this tax?

            Is that fair? Or is there some more moderate version of this story that better fits?

          • Lumifer says:

            @ Mary

            people should not be going about being cheeky and disrespectful to strangers of any stripe, on account of its being rude.

            Let’s try this again with better reading comprehension skills. I said:

            and you’re are a bit cheeky and not terribly respectful (though not rude)

          • Lumifer says:

            @ John Schilling

            Right, and what do we do to people who grab the bag of gold with one hand and the virgin’s ass with another? Do we tell them “Oh, it was all her fault, you’re good to go”?

          • Lumifer says:

            @ Dahl

            I’m not talking about rational expectations. I’m talking about rights. I would like to think that I have right not to be beaten, tazed, pepper-sprayed, and on occassion shot without either seriously breaking the law or being an imminent threat.

            If you believe I do not have such rights, well, we have a serious disagreement about the type of society we live in.

          • At a small tangent …

            Much discussion of these issues assumes that being a policeman is a very dangerous job.

            It isn’t. The death rate for policemen is about thirteen per hundred thousand. The rate for loggers is about ten times as high. For construction workers, about half again as high.

            To put it differently, the probability of being killed if you are a police officer, from any cause including accidents, is only about three times as high as the probability of being murdered for a random American.

          • Andrew G. says:

            Last I looked, US law enforcement wasn’t even the US civilian job with the highest risk of death from intentional injury; iirc bartenders and taxi drivers rank higher. For all on-job risks of death it’s not even in the top 10.

          • Anonymous says:

            what do we do to people who grab the bag of gold with one hand and the virgin’s ass with another?

            Generally, we put those people in jail… when we have proof beyond a reasonable doubt that they committed the elements of the crime, with appropriate mens rea, within the jurisdiction of the prosecuting authority, supposing they’ve been provided competent legal defense (and probably a few other conditions).

            Sometimes, we might not be able to have all of those things. Your choices are not, “Never prosecute wrongdoing,” and, “Utopia where nobody even tries to break the rules.” You have to select a level of proof and civil liberties to work with and accept that there will be people who still try to do bad things and get away with them. John’s point that you’re asking for a fundamental internal moral change in the entire population rather than any particular legal or policy adjustment stands.

          • John Schilling says:

            Right, and what do we do to people who grab the bag of gold with one hand and the virgin’s ass with another? Do we tell them “Oh, it was all her fault, you’re good to go”?

            Typically we have a hard time finding them to do anything at all.

            What we don’t do, is imagine this is a sign of intolerable social decay or as a grave national problem whose solution must be a top priority. Every society accepts some type and level of criminal violence(*) on the grounds of “shit happens” and “yes you really were asking for it”, punishing specific offenders iff they can be positively identified but otherwise moving on.

            People may legitimately disagree with where the line is drawn, but if you’re anywhere in the gray area and demanding that all right-thinking people must side with you, that’s just tribal signaling.

            *Yes, including criminal violence by policemen

          • Jaskologist says:

            anyone should feel safe in the presence of a public servant paid out of his taxes.

            But how should the other 47% feel?

          • Mary says:

            Let’s try this again with better reading comprehension skills. I said:

            and you’re are a bit cheeky and not terribly respectful (though not rude)

            And I deny it. You can not be cheeky and disrespectful without being rude.

            The problem is not reading comprehension, it’s the delusion that people aren’t being rude when people are.

          • Adam says:

            That’s obviously a joke, but unless you’re committing a crime inside of a federal building or on a military installation, the police are not paid by federal income taxes.

  8. Bugmaster says:

    If all you care about is happiness, then, logically, you should find a way to wirehead yourself as soon as possible — since wireheading maximizes your happiness.

    If that prospect doesn’t sound appealing, then either happiness doesn’t matter to you, or perhaps it does, but there’s something else that matters more. This preference for things other than happiness may or may not be rational (I personally don’t know of any good arguments against wireheading), but nonetheless, most likely is there.

    If we grant that such preference exists, then the happiness paradox is no longer a paradox. Sure, women are less happy in modern societies; but perhaps they value something else, such as being able to make their own choices, more than they value merely feeling good as much as possible (though, again, such preferences may be irrational).

    • electrace says:

      If all you care about is happiness, then, logically, you should find a way to wirehead yourself as soon as possible — since wireheading maximizes your happiness.

      Are you sure wireheads are happy?

      • fasdfasdfa says:

        I used to think procrastination applied only to things you disliked but did anyway. Then I tried to write a novel. I loved writing. Every second I was writing, I was thinking “This is so much fun”. And I never got past the second chapter, because I just couldn’t motivate myself to sit down and start writing.

        Look like Yvain solved his procrastination problem somehow.

        • Guy says:

          Scheduling is a hell of a drug.

          • Luke the CIA Stooge says:

            Do elaborate.

            I have tried a couple times now to schedule, and it worked for a couple days, but then i forget to one morning/evening and its right back to slovenly student. Any resources you recommend or helpful advice?

          • Different Guy says:

            I’m slightly confused. It was working, then you forgot 1 day, and it’s ruined forever? Why not just schedule the next day? If you can do 3/4 days scheduled, that’s pretty good right? And you’ll probably get better as you get more used to scheduling. Is this like the “To hell with it” fallacy, where people eat a cupcake so declare their diet ruined and order a large pizza? (I think it was called “Failing with abandon” somewhere on LW)

          • Mary says:

            It is amazing what falling off for one day can do for your willingness to get back on track.

          • Luke the CIA Stooge says:

            To be fair as an jobless student on summer-break completing online courses, my life doesn’t really have a natural schedule to it. I imagine if their were already a few large sections blocked off on my schedule, and i was reliably waking up in the same place scheduling would be easier.

            Any thoughts?
            Is a consistent routine necessary for effective scheduling/organization? Or have you found a way around it? Have you Found anyways to effectively maintain a routine? Do you run your morning like a pilot going over his pre-flight check-list? Any general organization tricks?

        • onyomi says:

          This is a very interesting phenomenon I have experienced: the thing you really enjoy whenever you do it, yet somehow still have trouble finding the energy to do. This applies, for me, to much of my work, and even, weirdly, sometimes, to sex. Never to eating, however, sadly…

      • Bugmaster says:

        Yes. Whiteheads experience the maximum amount of pleasure that it is possible for a human to experience. At all times. Forever. If that prospect does not appeal to you, then it must mean that you are willing to trade off happiness for something else.

    • DanielLC says:

      If all you care about is your own happiness yes. If you care about happiness in general, then you need to stay lucid so you can wirehead the universe.

      • Bugmaster says:

        In this case, you would be one of those people who is responsible for the “happiness paradox”, so my point stands. That is, if someone were to poll you and ask if you’re happy, you’d say, “no, despite all the new advances in technology and medicine I failed to reach my goal of increasing universal happiness by 1%/year” or something.

    • Earthly Knight says:

      If [wireheading] doesn’t sound appealing, then either happiness doesn’t matter to you, or perhaps it does, but there’s something else that matters more.

      This doesn’t follow– wireheading might not maximize expected value for an agent even if happiness is given more weight than any other term in the agent’s valuation function. This could be because the agent is already extremely happy, and the small increment in happiness she’ll get from wireheading will be outweighed by much larger decrements in the other terms, or it could be because the other terms individually matter less than happiness but jointly matter more. The strongest claim you can make is that agents who only value (their own) happiness or who assign (their own) happiness lexical priority over all other values should wirehead themselves, but even this depends on the dubious identification of hedonistic pleasure with happiness.

      • Bugmaster says:

        I’m not sure what the difference is between “hedonistic pleasure”, “happiness”, and other kinds of pleasure (assuming they exist at all). Your point about diminishing returns from increased happiness is completely valid, but I don’t think it applies to our current scenario — i.e., I don’t think that humans are already that close to being maximally happy. I could be wrong, though.

        • Earthly Knight says:

          I’m not sure what the difference is between “hedonistic pleasure”, “happiness”, and other kinds of pleasure (assuming they exist at all).

          The more reason not to conflate them.

          Your point about diminishing returns from increased happiness is completely valid, but I don’t think it applies to our current scenario — i.e., I don’t think that humans are already that close to being maximally happy.

          Wireheading imposes such catastrophic costs on other domains of human flourishing that it probably won’t be advisable unless your life is currently pretty joyless. If you value autonomy, authenticity, human relationships, accomplishing life goals, etc. even a little bit, you’re going to need a huge boost in happiness to compensate for their erasure. So wireheading will be a good idea only if you’re already miserable or care exclusively about pleasure, just as we would naively expect.

          • Bugmaster says:

            Right, but if you value “autonomy, authenticity, human relationships, accomplishing life goals, etc.” more than happiness, then we would expect exactly the same picture as the one we’re seeing — quality of life improves, but happiness either remains the same or improves only marginally.

            That said though, I am honestly not sure what it would mean to value anything more strongly than maximum continuous pleasure (as obtainable through wireheading). I don’t think any satisfaction from accomplishing life goals could compete with that…

          • Earthly Knight says:

            Right, but if you value “autonomy, authenticity, human relationships, accomplishing life goals, etc.” more than happiness,

            That was not the claim. Here, to make this more concrete, let’s plug in some numbers. Pretend that everything Tim values can be measured on a neat interval scale and model the “quantity” of each item of value in Tim’s life as a real number over the unit interval. Assume also that Tim values pleasure five times as much as each of autonomy, authenticity, human relationships, and accomplishing life goals, and that he values nothing else. Note that this means that Tim values pleasure more than everything else combined. Here is his life before wireheading:

            Autonomy: .9
            Authenticity: .9
            Human Relationships: .8
            Accomplishing Life Goals: .7
            Pleasure: .4

            And here, presumably, is how the distribution will look after Tim is wireheaded:

            Autonomy: 0
            Authenticity: 0
            Human relationships: 0
            Accomplishing life goals: 0
            Pleasure: 1

            Given the stipulations above, Tim is better off not being wireheaded. He gains .6 points of pleasure, but he loses the equivalent of ~.61 points of pleasure on the other dimensions of value.

            So, to reiterate: wireheading is probably only a good idea if your life is already thoroughly miserable or you accept some form of hedonist monism. If the former, though, there are other interventions that will leave you better off overall than wireheading, and you should focus on those instead. Which means that the possibility of wireheading serves principally as an argument against hedonist monism.

          • onyomi says:

            There is a difference between pleasure and happiness. Pleasure is a jolt of reward, probably mostly related to dopamine, you get when you do something which furthers an evolutionary goal–eat fattening food, have an orgasm.

            Happiness is a general sense, probably more related to serotonin, that things are going well–that you are on track to have a fulfilling life, which will probably include a lot of pleasure, but which isn’t necessarily defined by pleasure. You may feel happiness when playing catch with your child on a nice day or something, but not “pleasure” (actually, we might find that kind of gross/weird if you did).

            Drug addicts typically experience a lot of pleasure–probably more pleasure than is possible to experience without heroine or cocaine. Yet they often don’t experience much happiness. Long run, most people consider happiness more important than pleasure, perhaps because it is more lasting and feels more “real” in some sense.

            The problem with wireheading is that it eschews happiness in favor of pure pleasure, though taking away the negative health impact of drugs would still be nice, for certain.

    • Anonymous says:

      Personally, the best way to think about happiness I’ve come up with is “score”. Happiness is your reward for being successful at this life thing, but isn’t the substance of the success – just like being able to set your score at INT_MAX is not the same as being good at whatever game you’re playing (you’re just a cheater, or highly resistant to ennui of pressing a button repeatedly for days on end).

    • Ghatanathoah says:

      I agree that “people value other things than happiness” is definitely part of it. Happiness is obviously very important to people, but it’s not the only important thing.

      The question is, how much of the people failing to find happiness is due to people not seeking out happiness, and how much of it is due to people being systematically mistaken about what makes them happy? I think that there’s definitely some of both. People in our society often complain that they are not happy, and people who are professionally successful often complain that this success does not translate into happiness. Movies like “Hook,” “Liar Liar,” and “Jingle All the Way” all have a message that professional success does not lead to happiness in the same way that having a family does, that message must resonate with audiences.

      So while I agree that happiness isn’t the only valuable thing, I don’t think that “people pursuing other stuff besides happiness” is the only explanation for why happiness isn’t higher. I think that a large part of it is that people are systematically mistaken about how to be happy (in particular they participate in the Rat Race more than they should), and that policies that make it more difficult for people to follow their biases should be made.

  9. Eniuneg says:

    One thing I wonder about in the traditional vs. modern cultures and women vs. men’s happiness is total household leisure time. I could find a bunch of research about individual free time, and this showed a small increase in the last half-century. In the meantime, there was a huge shift towards dual-income households. It may be possible that a modern culture has fewer hours of leisure time as a household unit, even though the data for individuals shows a small increase.


    Also household discretionary income has decreased since the 1970s ( I think there may be something to a modern culture living too close to the margin to be relaxed enough to mark down that they are really happy on one of these surveys.

  10. Squirrel of Doom says:

    One thing to think about is that (say) 20% of the Chinese living today would have been dead under the conditions of (say) 30 years ago.

    You could imagine adjusting the old data for that by adding 25% dead people as “really unhappy”. Or are they blissfully devoid of unhappiness?

    I don’t have the answer, just want to point out another way this is hard to sort out.

    • Ghatanathoah says:

      Most utilitarians I know rank dead people as perfectly neutral, neither happy, nor unhappy. It’s better to be happy than dead, but better to be dead than to be unhappy (though it is still good for an unhappy person to be alive if they will stop being unhappy and become happy at some point in the future).

      • Squirrel of Doom says:

        Well, that’s very different from my preferences, even when I’m unhappy. Don’t kill me and think you did me a favor!

  11. dsp says:

    I’m pretty sceptical that any purported measure of happiness functions correctly in any case, but, notwithstanding that, none of these examples conflict with Sumner’s thesis. In fact, they describe situations where it seems entirely plausible that happiness changed only because standards for what qualifies as normal changed. For example, the increase in black happiness in the US and all the cases mentioned later under the freedom-and-democracy category might be explained by a recalibration among the relevant people from judging by the better-off to judging by their peers. It doesn’t necessarily matter if the peer group’s status even improves at all in the process, which could be troubling. It’s probably a good idea to give this possibility more consideration.

    Editing to add that the same can be applied to women, many of whom have probably expanded their comparison groups from including mainly other women to including the men they are now competing against economically.

  12. An increase in black happiness does not necessarily contradict the hypothesis that happiness is zero sum or a product of how you view yourself relative to others.

    • Shieldfoss says:

      This was my exact thought until Scott got to the decade-by-decade analysis later on – if status is a Positional Good, and black people have gained status since the 1970es (which I think is pretty incontrovertible?) then somebody else must have lost it. Did people talk about Hispanics as if they were a different race back then? Was Appalachia the same kind of devastatingly poor?

      • “if status is a Positional Good, and black people have gained status since the 1970es (which I think is pretty incontrovertible?) then somebody else must have lost it.”

        Recall that the Republican nominee is currently running on a platform of dogwhistles to the effect of “white men are no longer in charge of the country and that’s terrible.”

      • The Nybbler says:

        Appalachia was certainly still poor as dirt, though I think widespread meth is relatively new (alcoholism isn’t). Hispanics were considered separate though I think the name may have been different.

        As Wirehead Wannabee points out above, the obvious candidate for ‘status lost’ is your working class white man, your Archie Bunker types.

        • Richard says:

          Appalachia was certainly still poor as dirt

          During the seventies Appalachian coal boom? I doubt it. According to The Economic Impact of the Coal Boom and Bust (pdf) it seems Appalachia in the 70s was more like the midwestern oil boom-towns of a few years ago. The massive poverty came with the coal collapse in the latter half of the 1980s

  13. Nicholas says:

    I don’t know if India is a spoiler here, but has anyone considered that reporting you were really unhappy to be living in China while it was being run by Chairman Mao might be the kind of thing that you didn’t do if you wanted to live a long and prosperous life, that there was a pervasive atmosphere of paranoia after the Cultural Revolution, and that the numbers in China might just be going down because they were artificially inflated in the past?

    • Enkidum says:

      The survey is said to have begun around 1990, 14 years after Mao’s death.

      • onyomi says:

        People were still paranoid about state monitoring in the 90s and continue to be so to some extent even to this day.

        • Hanfeizi says:

          A good point. 1990 was during the post-Tiananmen chill of 1989-1992, when the emergence of pluralism in the post-Mao years ground to a screeching halt.

          Likewise, Taiwan’s democracy was still unstable at the time and people were used to KMT police state rule.

          By contrast, 2010 was the height of the Hu Jintao era of political complacency and Taiwan had become pretty used to democracy by then (not to mention mired in an endless recession).

  14. Thecommexokid says:

    I agree with several other commenters that survey responses to a question about happiness seem like a poor proxy for measuring actual happiness. If you had asked me to rate my happiness on a 1–10 scale on a randomly chosen day last month, there are days I would have said “7” and days I would have said “3”, despite very little about my life having changed in any way over the course of the month.

    Of course, I don’t have a suggestion for how to measure happiness that is better than asking people for a number…

    • Paul Barnsley says:

      Here is a brief defence of the meaningful nature of the data – it’s absolutely sensitive to stupid stuff, like the weather, but for large samples with good survey practice most of that seems to wash out. Systematic cultural bias and, particularly, people moving the goalposts for what constitutes a “7” are much bigger problems, though.

      …people, almost universally, appear to understand
      what is being asked when questioned about their “satisfaction”, “contentment”
      or “happiness”. As Kahneman and Krueger (2006) note, over 99%
      of respondents to the 1998 General Social Survey were willing and able to
      provide an answer (other than “don’t know”) to the question “Taken all
      together, how would you say things are these days? Would you say that you
      are very happy, pretty happy, or not too happy?” By way of comparison,
      83% of respondents provided answers to questions about their income.
      Answers about subjective wellbeing also display a relatively high testretest
      correlation for individuals, an indication that the answers provided
      are something other than random guesses.
      So, if answers to questions about subjective wellbeing are both coherent
      and consistent, are they also significant, from a utilitarian point of
      view? There is a variety of evidence suggesting that reported SWB correlates
      closely with other variables which we might think of as signifying
      high levels of utility. It is difficult to improve on the survey provided by
      Kahneman and Krueger (2006), which cites studies of individuals with high
      subjective wellbeing showing that these persons tend to:
      • recover faster from wounds and the common cold;
      • show greater neurological activity in the portion of the brain associated
      with pleasure;
      • smile more often;
      • be judged as happier by those around them; and
      • more frequently express positive emotions.
      Further, statistical analysis as to the determinants of SWB returns coefficients
      broadly consistent with what we would predict. Income increases
      wellbeing, as does good health, the deaths of friends and family reduce it.5
      Education, freedom and a high relative position in society all correlate positively
      with one’s sense of wellbeing.
      Obviously, there is some risk of circularity here – we cannot both look
      to subjective wellbeing research to tell us about the structure of our utility
      functions while simultaneously using its consistency with our ex ante
      predictions as to their structure to test its validity. We would, however,
      be skeptical of a measure which told us that health and wealth tended to
      reduce our utility; we have, as a starting point, a model which puts the
      same signs on the key arguments of a utility function as does traditional

      • Mary says:

        hmmm. . . I wonder. . .

        There were some studies that turned out to be missing a cofactor. That, in fact, happiness was correlated with many good things because it was also correlated with regarding your life as meaningful, which had a better correlation with the good things.

        But I’m not sure it’s this one.

  15. Paul Barnsley says:

    One key point when looking at happiness/SWB data (well, I wrote my PhD thesis on this issue, so I think it’s important at least) is that it’s generally ordinal data and doesn’t aggregate properly or produce good measures of central tendency. So you’re stuck either using an arbitrary binary (“very happy”/”not very happy”) scale, like the chart above, or you just shrug your shoulders and pretend you’ve got interval/ratio scale data without any evidence to justify the assumption – this, more or less, was what that recent study on parental happiness as a function of national social policy did.*
    One of those approaches is flat wrong, but the other can conceal a lot of important differences between states of the world. A high proportion of “very happy” probably correlates to higher overall national happiness/utility, but it doesn’t necessarily imply it, even if you’re assuming a constant mapping from underlying utility to the reported scale… Ideally you need to map the responses to an interval scale and then derive national/temporal averages, just as a starting point for these discussions.

  16. Anonymous says:

    Thought experiment: Pick 2 of the following to be maximized and one to be set at the lowest tolerable level.

    – Health (Including general physical fitness, mental wellbeing, etc.) (Lowest would be something you can’t fix with money.)

    – Love (Family, friends, romance, community, “spirituality” etc.)

    – Money (Lands, technology, culture, power, etc.)

    Pretty cheesy but I think it holds… You see increasing happiness when the math checks, however if you raise any of those parameters too much at the expense of the others, you get neutral or even negative results. Not really paradoxical.

    • Anonymous says:

      What’s the lowest standard for love and money?

    • Immanentizing Eschatons says:

      I think I would be quite happy with maximized health and love (not the “spirituality” part though, that is definitely… not my thing, assuming it is even a real thing in the first place) and minimized money. I would choose it over my current life definitely.

      (This doesn’t really translate to real life well of course, real life poverty doesn’t just mean not being able to afford “luxuries”, it means a lot of unpleasant work to get you’re basic necessities, or just not having them- which here are covered by the health part.)

      • Diogenes says:

        Yeah that characterization of money is really strange.

        “Money (Lands, technology, culture, power, etc.)”

        It seems to me the things people want most out of money — exempting “health” and “love” — are leisure & security, ends to which land, technology, power, etc., are merely means.

  17. SBoat says:

    It interests me that people generally assume primitive humans must have been miserable. After all, they were exposed to the elements with no heating or plumbing or cozy beds, no TV or even books to read, constantly hunted and having to hunt for food, no medicine, etc. But from what I’ve heard, modern hunter-gatherer tribes have very high happiness rates. Of course any of us would see our happiness plummet if we were suddenly thrust into a hunter-gatherer life, so we assume that someone who was always a hunter-gatherer would also be unhappy. I think this just goes to show people aren’t too good at understanding different frames of mind.

    • Winter Shaker says:

      Of course any of us would see our happiness plummet if we were suddenly thrust into a hunter-gatherer life, so we assume that someone who was always a hunter-gatherer would also be unhappy.

      On the other hand, the colonial authorities in the early European settlements in North America were apparently perennially baffled that people who had been abducted by the natives and lived with them for some time, often refused to come home when offered the chance of rescue, whereas people who had been born and raised in a native tribe, and then subsequently brought up in white society, usually went back to their hunter-gatherer lifestyle as soon as they were able to, so the ‘hunter-gatherers have happier lives than people in agricultural / industrial societies’ hypothesis may have something to do with it.
      (Haven’t read the book but James Axtell seems to be the predominant writer on the subject.

      • Anonymous says:

        There is something to be said of getting more leisure time – according to Wikipedia, hunter-gatherers work 6.5 hours per day on average, while agriculturalists work 8.8 hours per day. That’s not an insignificant difference, and there is a type of people (myself included) that are satisfied with very little in the material sense.

        • Nornagest says:

          Took me a bit to dig it up, but the “original affluent society” research is, at the very least, disputed; Kaplan there cites an average 42 hour week for the !Kung, taking into account food-preparation work and other labor besides hunting and gathering per se. A 6.5 hour a day claim is weaker than the usual 4-hour claim I see, though.

          (H/t Julian.)

          • Anonymous says:

            Well, I don’t usually count food preparation and acquisition (never mind laundry and household repairs) as part of my “work” hours as a modern office worker. If I did, it would certainly add a few hours on top of my 9-to-5.

          • Nornagest says:

            Sure, it’s not an apples-to-apples comparison. But that goes the other way, too — a lot of people do a lot of stuff during work hours that isn’t actual work.

            One of the things that makes comparisons like this tricky is that preindustrial (never mind pre-agricultural) societies don’t have the same divisions between labor and leisure that we do. Another is that their labor is a lot more seasonal than ours (bar a few exceptions, like agricultural labor or lift operators at a ski resort) — if the weather makes it impossible to hunt or gather, your average forager’s going to be doing a lot of nothing once the possibilities of e.g. equipment maintenance are exhausted, even though they’d likely prefer otherwise.

      • Divy says:

        True, but the industrial/agricultural societies then were not much better off than the hunger gatherers. They also did not have plumbing, TV, medicine, etc. If this situation happened today, I doubt the result would be the same.

      • I wonder what % of the ones who stayed or went back were married.

      • Mary says:

        “the colonial authorities in the early European settlements in North America were apparently perennially baffled that people who had been abducted by the natives and lived with them for some time, often refused to come home when offered the chance of rescue, ”

        OTOH, I’ve read a description of some women refusing such rescues — cited as evidence that the lifestyle was better — that suggested rather more strongly to me that they were victims of Stockholm Syndrome.

        Hard to be sure either way. The particular case I remember reading was after about six months, and most of the women had seen their families massacred before their eyes before they were carried off.

        (Were those brought to white society, brought in a way less drastically traumatic?)

    • nachterb says:

      Some confounding items to think on:

      3 years ago, I went to Honduras to help build houses for the dirt poor (literally, they had only dirt and the clothes that Goodwill doesn’t keep). They were all quite happy, even when not around us. And the national happiness score is 8.5 not too shabby ( I remember the American people in-country telling us that we need to take into account how happy the people are there and not assume they are miserable just because they are, in our eyes, poor. Maybe it is religion, maybe it is the culture, or maybe everybody around them is happy so they are too.

      On the other hand, in China (which I have visited many times on business) they think Americans are very entertaining but would never act like them (i.e. upbeat, smiling, backslapping). So I doubt the survey gets consistent results there, where “yes” means “no” and “no” means “yes” most of the time.

      • Did having better houses seem to make them happier?

        • nachterb says:

          Hi Nancy- they were certainly happy that we were there to start the houses. We only got one foundation done and a second started, but it was a design that did not use a lot of materials so they could make more houses. But I would say the average person on the street looked happy to me, even when observed from inside our bus.

  18. Dan Simon says:

    How happy you’re expected to claim to be is highly culture-dependent: in the US, it’s generally considered a grave moral failing to present oneself as anything but deliriously upbeat, whereas I’m given to understand that in, say, Finland, it’s thought terribly impolite to be too ostentatious about one’s current disinclination to contemplate suicide. (Canada is somewhere in between, although probably closer to Finland. I suspect climate and sunlight have a non-trivial effect on this element of culture.)

    Such cultural expectations can presumably also change randomly (or due to foreign influences) over time, quite possibly affecting the results of happiness surveys in a given country more strongly than any actual changes in the psychological condition of the citizenry.

    • ThaddeusMike says:

      My experience with Canada (mostly southern Ontario) suggests that Canadian society is quite upbeat.

    • Mercer says:

      I would disagree with that characterization of Canada, as a Canadian myself. I would put us closer to the American style. But I agree with your point about culture being a powerful factor to consider.

      And because we have no good way to quantify how a culture changes, it makes the whole game of tracking happiness difficult.

    • Austin says:

      I suspect climate and sunlight have a non-trivial effect on this element of culture.

      I suspect climate and sunlight have a much larger effect on whether or not people actually feel happy than they do on whether or not their culture encourages them to display their happiness.

  19. Tekhno says:

    I’m just going from the gut here, but I wouldn’t expect democracy to do much to happiness. Most people hardly ever get to vote, and it’s usually treat as picking between the least bad of two evils.

    I’d guess that liberalism is more important than democracy, but that it would be tangled up in data because the countries that became more liberal also became more democratic. It feels great to not have to be paranoid about the secret police taking you away or about having your business arbitrarily shut down I imagine, probably much more than it feels great to put a ballot into a box for someone you aren’t really happy with.

    • Tandagore says:

      Not ever country has a two party system though. Obviously not every country works like Switzerland, where there are votes on pretty much everything (with a low turnout though), but a lot of the democracies in Europe have votes that matter with a reasonable turnout at >70%.

    • MugaSofer says:

      If countries that become democratic usually become more liberal, in what sense is this something “tangled up in the data” and not the mechanism by which democracy makes people happier?

      • Tekhno says:

        Because you don’t need democracy to have liberalism, nor vice versa (Pinochet’s Chile is a good modern example of liberalization before democracy, and the general historical example would be the existence of liberalism prior to mass franchise). The fact that they come together is an ideological construction where democracy is equated to political liberalism. The people who want to bring liberalism to a country also want to bring democracy, and for good reason, because democracy provides a pressure valve in times of crisis, but this doesn’t mean that democracy itself is responsible for raising happiness. People spend most of their time complaining about democracy while enjoying liberalism (and only those on the fringe like communists and fascists spend much time complaining about both), and democracy has for a long time been about choosing between left-liberals and right-liberals/conservatives, because liberal ideas have dominion in the West.

        • Immanentizing Eschatons says:

          Pinochet a liberal?! Economically sure, but the secret police taking you away was definitely happening. A lot.

    • Mercer says:

      There’s talk of Japan becoming less democratic, though I’m ignorant of how true that is. They did well in terms of happiness gain though, and as Scott mentioned thats despite the Lost Decade

      I’d love to see data extending back to the 50s when they first transitioned into democracy, though itd be inevitably entangled with being immediately postwar

  20. Hwold says:

    The most striking finding is that most countries got happier between those two years – sometimes a lot happier. In Mexico, the percent of people saying they were very happy increased by 25 percentage points!

    I wonder how much of that can be just explained by a high year-to-year variability. Like sampling 20 different functions of the form sin(a x) between 0 and 1, with a being randomly sampled from [1e5,1e6]. You will find a “change” is most of your functions, but that doesn’t mean there’s a “real” signal.

    I expect, for example, a significant change in “happiness” a week before the last match of a big sport event and a week after.

    That probably can’t be the whole story though, since we have a clear bias toward more happiness.

  21. amazon says:

    The Chinese data is going to be REALLY noisy for cultural reasons.

    “Are you happy” is known to be a confusing poll question in the Chinese context. State TV ran a large-scale poll asking Chinese people just that, in 2012, and …

    > First, more than half of the people approached were suspicious or confused, obviously unaccustomed to this requisitioning of an honest opinion. “What?” an office assistant barked. “You need to explain the kind of program you are running before I answer the question.” A gruff middle-aged man waved the journalist away with a statement that has now gone viral: “Don’t ask me. I’m only a migrant laborer.” A stooped grandfather picking aluminum cans out of the trash, responded to the reporter’s question by mysteriously giving the him his surname. In Chinese, “Are you happy” happens to be a homophone for “Is your last name Fu?” Apparently, the elderly gentleman thought the happiness query so unlikely that he immediately chose to interpret the question as the latter.

    • Mercer says:

      I was not ready to laugh this much, this early

      So either they’re unhappy or they’re maybe happy but their last name isn’t Fu

  22. Alphaceph says:

    Human happiness probably tracks our brain’s estimation of our success from an evolutionary perspective. If you look at it this way, then it makes sense that making everyone 10x richer might not have a huge impact because nobody got relatively better off. Women in the office rather than looking after the children probably lools like a disaster to evolution. Black people gaining status looks good. Moving from bombed out berlin in 1945 to rebuilt West Germany in 1950 is probably a change that looks lile you nearly died but were then saved.

    • Monboddo says:

      “If you look at it this way, then it makes sense that making everyone 10x richer might not have a huge impact because nobody got relatively better off”

      I don’t think that follows. Making everyone 10x richer should mean that they are subject to much less risk of the kind that would be meaningful from an evolutionary perspective. (For example if you can last 10 times longer before eviction after losing your income, this is going to make your progeny more likely to exist/survive. Or maybe you can just afford to feed 10x as many children.)

      Of course, increasing the GDP 10x might not do that, but then maybe the real issue is that the people are not really more rich just because the GDP goes up.

      Generally speaking, survival of one’s progeny or kin is _not_ a relative measure, because the population of the species can increase or decrease.

  23. Focus says:

    Happiness research has shown:
    – short/long run: Some changes (e.g., winning in lottery and getting permanently rich) only increase your happiness in the short run (and that of your children; in the case of lottery it is 1 year or so). Some other changes that you can do increase happiness in the long run too, even though the long-run happiness is mostly genetic. I did not find a reliable list of those things, but I think it was:
    – regular physical exercise
    – strong social relationships
    – interesting job that has meaning
    – gratitude

    So those things increase your happiness permanently.

    Other research findings:
    – Happiness reflects partially your relative success but partially your absolute success. So growth can help everyone.
    – GDP growth matters very much
    – GDP level (rich vs. poor country) matters logarithmically, i.e., 5% more GDP increases happiness equally regardless of the initial level
    – Unearned income, like lottery or welfare, does not help in the long run.
    – Economic Freedom (both Heritage and Fraser indices) correlate strongly with happiness

    In China economic growth has slowed down, competition is fierce, children study 16 hours a day, 7 days a week, and dating markets for men are terrible. Current GDP growth rate matters more than current GDP level.

    In many countries, regulation and welfare have been increased so that unemployment is a terrible problem and growth has been reduced. That compensates the increased GDP level.

    About short and long run (not very good):

    You can easily find “scientific lists” on the web, but I’m not that sure of them:

    “Happiness – 50% genetics, 10% circumstances, 40% intentional activity” If this is in the average, by really focusing on that you can probably affect more than 40%. On the other hand, probably some of the 40% is “unshared environment”, that is, random things that the researchers have not asked about or you don’t remember and that your siblings don’t face .

    • Ghatanathoah says:

      Some changes (e.g., winning in lottery and getting permanently rich) only increase your happiness in the short run

      I wonder if the studies Scott is referencing only measure long-run happiness, and that’s why they’re flat. Maybe rich and poor societies have the same long-run happiness, but rich societies have more quick bursts of short-term happiness. The odds of catching someone in mid-burst are still low enough that the quick bursts don’t show up in the studies, but that doesn’t mean they don’t matter. Short-term stuff matters, even if it doesn’t matter as much as long run stuff.

      • Cliff says:

        The quoted section is not apparently true. Lottery winnings do have a long run affect on happiness, as does disability

    • Dan Miller says:

      What about interpersonal variation? I could imagine that certain people, based on their varying personality traits, would have their happiness affected differently by the factors you list. But I don’t know the research very well.

  24. Anon. says:

    Even if national income matters for happiness: how much of the variation does it explain vs within-country income %ile?

  25. MugaSofer says:

    I’ve heard some people saying that China’s economic growth is probably vastly exaggerated by their government; apparently they don’t show the levels of electricity usage we’d expect, for example. This seems plausible to me, given the extent to which the USSR pretended to have made huge economic gains.

    Presumably their economy has grown at least somewhat, given that they have exports and so on, but it’s possible that this was combined with an increase in government oppressiveness or something.

    In any case, I think we should expect Chinese data to be an outlier if we rely on probably-false figures for the correlation.

  26. Divy says:

    I think this data is consistent with Sumner’s explanation. China under Mao was a closed society completely innundated with propaganda, like today’s North Korea, so people had no idea that foreigners were better off or that things could be better (especially because things pre-Mao were even worse so the Chinese couldn’t even compare to a better past time in their own memory). Today, Chinese people have access to the outside world and can see that they are still very poor, particularly compared to the countries that they would think of as their historical peers. So even though the gap is closing, it feels like the gap is bigger because in the Mao area, people weren’t aware of a gap at all.

    By contrast, Mexicans have always traveled back and forth between the US, so they were always aware of how far behind they were and how better living standards are possible. They are now happier because the gap is closing–just like Germans became happier as the gap closed between their postwar reality and what they knew to be possible pre-war.

  27. I think you have to make some allowance for culture (or perhaps even nature.) I’ve seen a few surveys that show Latin American countries ranking high in happiness, probably not because they all have the same economic or political things going on, but because of some other factor, like a culture of focusing on the positive or community cohesion. The Chinese folks I know, by contrast, don’t seem to place a lot of importance on happiness. Studying hard and getting a good job so you don’t go hungry is a lot more important to them than some abstraction like “happiness.” I note that the Taiwanese seem to have lost even more happiness than the Chinese, and I don’t get the impression that they have the remaining economic difficulties China has, but do have a fairly culturally similar population.

    Japan is a very pleasant country to live in–prosperous, clean, peaceful, low-crime, Pokemon everywhere, etc.

    • Wency says:

      I think you’re on to something. It seems to me there are some utilitarian blinders on in this discussion — not everyone optimizes for happiness. When we look at our heroes, our role models, it’s rare that we say, “Wow, that guy did a good job at making himself happy! I wish I were more like him.” Our sense is that people with the greatest agency, the greatest reservoir of internal ability and determination, do things other than optimize for happiness (in fact, sometimes they’re miserable). And if they optimize for other people’s happiness, it’s often only indirectly at best.

      I’ve read and observed that having children has a tendency to make people less happy, though it often leads to a higher sense of reported “satisfaction”.

      Having resources and mastery over our environment allows us to better optimize for certain goals, which might or might not center on happiness. Of course, with that prosperity, society changes in ways that make other goals more difficult to achieve — e.g., if you seek a traditional marriage, then your increased resources will do little to provide it.

  28. Brendan Long says:

    Couldn’t inequality have a big effect on this, since GDP is measuring population-level income, but we’re asking a yes/no question on an individual level for happiness? For example, if you had a two person country where both people make $100k per year, you’d have a GDP of $100k and 100% happiness, but if you had a two person country where one person makes $0 per year and one person makes $1 trillion per year, the GDP is 5 million times richer, but half as happy.

    Another problem is that GDP doesn’t take assets or working for yourself into account. If a hypothetical government were to bulldoze farm villages so people would to move to cities and work factory jobs, the GDP would go up, even though the people involved are financially worse off. It seems like when asking about happiness, a more useful measure would be “amount of liquid assets” or something like that.

  29. baconbacon says:

    So how do we adjust for population changes? The US and China have doubled since the 50s, Japan is up 50% and flat since their economic stagnation began. If the US was an “8” in 1950 and an “8” now that is adding 170 million “8s” or 1.3 billion happiness points! (yes I am kidding and not kidding). If China is a “4.5” and has added 700 million to its population then that is over 3 billion happiness points!

    Isn’t a pretty straight forward answer to some of these questions that it is much harder to make 300 million people happy than 150 million?

    • Dan Miller says:

      “it is much harder to make 300 million people happy than 150 million”

      Given gains from specialization this is contestable and non-obvious.

      • Skivverus says:

        Especially when you’ve also got another 150 million brains to help you with figuring out how to make them happy.

        Unless you’re trying to make X million people happy by yourself, in which case… good luck?

      • baconbacon says:

        Given gains from specialization this is contestable and non-obvious.

        Gains from specialization help the production side, but don’t automatically help the consumption side as the base fragments. One example is the most watched TV episode list. the MASH finally in 1983 is still #1 at 125 million viewers, with 9 of the top 10 coming prior to 2000, despite a much larger population (years are 77, 93, 67, 98, 04, 88, 92, 92, 79).

        1 episode of 1 TV show was more popular than the most popular Super Bowl, with a blowout half time show with all the commercials attempting to be entertaining. To produce the same level of TV “contentment” now as there was in the 80s you have to produce X as many shows, where X is >2.

        The US has 2x as many people, but we aren’t just producing 2X as much Coke and Pepsi, or Y times (where Y=gdp growth rate) either, there are now far more options for basic sugary, carbonated drinks and each one has to go through R&D, testing, marketing and a bunch of failures to get a portion of the market. In 7 years of marriage my wife and I have owned 6 or 7 different coffee making apparatus.

        There is a large cost to not only creating and marketing these things, but to testing and running through the options as consumers to find the one that suits you.

  30. InferentialDistance says:

    You can’t just measure economic aspects of reality in order to predict happiness. Humans are social creatures, social conditions affect happiness. By inspection, society wide emotional abuse through social media appears to be on the rise and by all accounts is more than sufficient to overcome any technological gains in physical comfort (thus the rise of depression, no I don’t think that has anything to do with improved diagnosis I think we are in fact getting more depressed and that depression causes weight gain thus explaining the obesity “epidemic”).

  31. J Mann says:

    It seems plausible to me to separate “well being” from reported happiness. If the Chinese are pulling out of poverty, it’s likely that they’re better off, even if they’re not reporting that they’re happier, and even if they’re actually not any happier.

    Let’s suppose that a genie offers me the chance to choose to go back to a world without cheap phone calls. It’s magic – no one will ever never know what we are missing, and the genie also assures me that neither I nor anyone else will feel any less happy.

    I’d pick the current, cheap telephone world. Even without additional reported or perceived happiness, I think we are better off (i.e., have more utility) in a world where we can pick up the phone and call our parents or distant friends on a whim instead of counting minutes.

    That’s just one example, and we could spend all day thinking of the stuff “kids today take for granted.” You might be able to come up with some specific counter examples – for example, if air conditioning really doesn’t make anyone any happier, maybe we’d be net better off if no one knew we could have it, so that we could use less energy, but I think those would tend to be specific exceptions with negative externalities, and likely to be at least balanced if not outweighed by other changes with positive externalities.

    • zolstein says:

      Is well-being your terminal goal as distinct from happiness? Are you willing to trade happiness for well-being? If the genie offered to take away telephones with no negative consequences and also people will report higher subjective happiness due to closer interpersonal interaction, do you still choose to keep the telephones?

      • Ghatanathoah says:

        I wonder if happiness studies discount bursts of short term happiness and bursts of short term annoyance/inconvenience.

        I can think of lots of times when I am unable to reach someone because their phone is silenced and am mildly inconvenienced by this fact. It’s a small inconvenience, but it adds up if it happens thousands of times over the course of your life. I don’t know if I’d remember to count it on a subjective wellbeing test though. Noticing how tiny little things can add up seems like it’s a learned skill.

      • J Mann says:

        I can only speak personally, but my intuition is that they’re distinct.

        My guess is that if people choose to talk to distant relatives and friends via skype, it’s because they think it makes them better off, and that in general, they’re probably right. (There are specific cases of “lemon equilibria”, but I think they’re rare). I’d say both revealed preference and observation of the benefits lead me to that conclusion.

        I guess the counter-hypothesis suggested by your example is that reported happiness is pretty well correlated with well being – that if our experimental group reports that moving to the big city leaves them no happier than the similarly situated control group back on the farm reports themselves to be, then we can assume that they’re probably not better off, even if (a) they chose to move to the city and not to move back and (b) we can observe various quality of life improvements.

        To go more philosophically, I wouldn’t stay on the Island of the Lotus Eaters, even if I knew my reported happiness would skyrocket.

  32. wintermute92 says:

    An interesting observation: the ex-Communist results didn’t seem weird to me, and actually seem to conform to your final conclusions.

    Poland and Ukraine have gone from Soviet autocracy to fairly sincere democracies making inroads to stable European identities. Even with recent issues (censorship in Poland, Russia in Ukraine) there’s a wave of independence and national pride present there, backed by real democracy.

    Estonia and Lithuania, by contrast, remain grindingly poor and undemocratic. Even asserting that wealth doesn’t matter much, I imagine a life of poverty in a corrupt kleptocracy sucks on all the axes we’re examining. It looks like the countries which are making real progress socially/governmentally are gaining happiness, and the ones that have remained watchwords for corruption have gained very little.

  33. Utilitaria says:

    Reading through the bizarre, contradictory and paradoxical results of ‘happiness’ studies and the contradicting results you get when you try to measure the same thing using different words (happiness vs life satisfaction), the obvious conclusion is that this data isn’t representing anything and people’s self-reported happiness is mostly random noise.

  34. Doug S. says:

    The general conclusion of research on money and happiness seems to be this: Money doesn’t really make you happy, but a lack of it can make you miserable.

  35. Elephant says:

    I am always amazed that people are amazed that increased material well being, technology, increased life expectancy, etc., don’t lead to happiness.

    • Ghatanathoah says:

      Increased life expectancy does add to happiness simply because you have more time to be happy. You might have the same average happiness, in terms of your total lifetime happiness divided by how long you lived. But your total would still be higher.

      • Elephant says:

        I knew I would get this reply! Yes, what you’ve written is of course true, but no one (outside this blog) would say that they are happier if they are spending a longer amount of time at the same (or lower) level of happiness. It’s like saying a temperature of 30 degrees counts more than one of 40 degrees for some measure of warmth if the former extends for a longer period of time. There’s no “integrated temperature,” nor is there an “integrated happiness,” despite what seems like the desire of utilitarians to wish that were the case.

        • Lumifer says:

          It’s like saying a temperature of 30 degrees counts more than one of 40 degrees for some measure of warmth if the former extends for a longer period of time.

          That’s basically how cooking works. Why do you find this strange?

          • Elephant says:

            That’s definitely not how cooking works. Baking a cake at 200F for 4 hours doesn’t give the same results as baking at 400F for 2 hours.

            In any case, thanks (you and Ghatanathoah) for the comments. I think this question is an interesting one — how is it that one might try to maximize happiness if happiness isn’t a thing that obeys rules of arithmetic? (Is there a triangle inequality for happiness?) I’m writing rather quickly — sorry if this isn’t clear — but the lack of an answer to this is a major reason I find utilitarianism in general ridiculous.

          • The Nybbler says:

            Baking doesn’t work that way, but within limits, roasting does.

          • Lumifer says:

            @ Elephant

            We are not talking about “the same results”. You said that there is no such thing as “integrated temperature” and yes, there is, and the simplest application of that idea is cooking.

        • Earthly Knight says:

          It’s like saying a temperature of 30 degrees counts more than one of 40 degrees for some measure of warmth if the former extends for a longer period of time.

          No, it isn’t. You are denying that living longer at the same level of happiness makes for a happier life overall. The appropriate analogy, then, would be cooking at 30 degrees for a period of time versus cooking at 30 degrees for a slightly longer span of time. The temperature in the latter case isn’t any higher, of course, but it does require a larger infusion of heat. For what it’s worth, I have a difficult-to-shake intuition that happiness is more like heat than temperature.

          • I agree with Elephant that living a longer life at the same level of happiness does not make one happier.

            However I disagree with his initial comment that increased well being does not lead to happiness. Being able to easily travel and communicate long distance, being able to have discussions like this electronically from my basement, and having better medicine definitely improves my life and increases my happiness (living longer doesn’t make me happier by having more years of static happiness, but by decreasing my fear of death).

            But these things don’t get included in a higher degree of reported happiness because they are now seen as the normal state of things (as others have said).

          • On the assumption that grieving makes people less happy, an increase in lifespans (or possibly an increase in health spans) causes a temporary increase in happiness.

  36. Andy Gardner says:

    What about demographics? I think that it affects happiness on two levels. First, I can imagine that the China’s one-child policy has lowered life satisfaction for many Chinese adults. Second, I suspect that happiness varies strongly by age in general. Are happiness metrics age-adjusted?

  37. sconn says:

    It seems to me that the transition from agriculture to urban life would result in a short-term decline in happiness despite an economic increase. I mean, say I am completely independent on my farm so I don’t have much in the way of an income, but I also don’t have any expenses. I have the enjoyment of feeling independent and in control of my own life, as well as good health and so forth. When I move to the city to get a job — especially if it’s not voluntary — I might find my life gets worse. I go from $0 a day to $1 day, but I might also go from three good meals a day to two bad ones, an eight-hour workday to a 12-hour work day, plenty of living space to cramped living space, and independence to getting yelled at by a mean boss every day. And while I was skilled and confident as a farmer, now I am realizing my illiteracy and lack of knowledge about city life make me low-status.

    A decade or two and maybe I’m making way more money, have a roomy apartment, three good meals a day, a television, internet, and so on …. I might make up that initial happiness loss. But it seems to be a country in the middle of urbanization is likely to have happiness losses because more money doesn’t always mean higher quality of life even on objective measures.

    • John Schilling says:

      I have the enjoyment of feeling independent and in control of my own life

      How does “I will starve and my children will starve unless the rains come and the locusts don’t”, translate to “in control of my own life”? I think you may be working from a romanticized view of preindustrial farming here; people have been leaving the farms to move to the cities, of their own free will, since there were cities to move to.

      • LHN says:

        Even during the long period when cities were such reservoirs of endemic disease that they only managed to maintain their populations through that immigration. Keeping ’em down on the farm after they’ve seen Paree has always been tougher than persuading urbanites not to take up the joys of subsistence agriculture.

      • sconn says:

        I was thinking more on a smaller scale — since the farmer is self-employed, he can plow first and then weed the garden or vice versa. He’s not in total control of his life, of course, but he doesn’t *feel* the daily imposition of another person’s will and schedule on him.

        (His wife, however, might feel quite the opposite!)

        • John Schilling says:

          but he doesn’t *feel* the daily imposition of another person’s will and schedule on him

          No, he invents an imaginary person to whose will he can ascribe the various schedules which are imposed upon him. Sunrise, sunset, first frost, first rain, flood of the Nile, all of these can be matters of life and death for a farmer. And yes, the farmer gets to plan and prioritize his actions, just like the factory or office worker faced with impossible quotas and deadlines gets to plan or prioritize his actions.

          But the penalties for noncompliance are harsher on the farmer, there is no appeal, and there’s no guarantee that the posted schedule will be the actual one. Empirically, people really want to believe that such external forces are under the control of an agent to whose mercy and kindness they can appeal. Bosses, even Scroogian ones, are such agents. Gaia, not so much.

  38. I think the results for women’s happiness can be explained by feminists trying to prevent extreme unhappiness rather than maximizing happiness.

    A woman who is in a very bad marriage needs the legal right to get out, and she needs to be able to make her own money (alimony isn’t what it was). If you assume that most women are in at least tolerable marriages, having to have a job outside the home pretty much makes their lives worse– most jobs aren’t all that satisfying.

    Also, let’s take Scott’s point about people not looking happier seriously– we’re either at or near the tech for evaluating street scenes for whether people look happy– I’d say check posture as well as facial expressions.

    Getting useful information out of this isn’t going to be easy (and it isn’t going to include rural people very well), but at least it will be something to compare against surveys.

    • John Schilling says:

      Part of happiness is whether you feel like getting out of bed and going outside in the first place; that’s something where we can probably get the raw data pretty easily, but figuring out what it means will be tricky.

      • Fair point– photographs of street scenes aren’t going to include some depressed people (inertia and misery don’t always go together) or very sick people, but they could still give some information about a high proportion of people.

  39. Jennifer says:

    Interesting article, but I take issue with this: ” It seems perfectly reasonable to me that having to work outside the home makes people less happy, getting to spend time with their family makes them more happy, and having to work outside the home but also being expected to take care of your family at the same time makes them least happy of all.”

    Stay-at-home mothers, the ones that would have the most time to spend with their families, report greater unhappiness than working mothers. ( This whole idea that women are happiest wiping noses and bottoms needs to change. Both men and women want to spend time with their families, but working also provides real world accomplishment, financial independence, and adult interaction which turns out is also something both men and women benefit from.

    • Randy M says:

      but working also provides real world accomplishment,

      I’m not sure what you are trying to convey with this that would be true. Explain? (It strikes me that ‘feminist’ sentiments like this may be the cause of some of that stress and depression.)

      Also, if stay at home moms aren’t interacting with other adults, I can see how they’d be depressed more. But it doesn’t have to be the case. My homeschooling wife probably sees more other adults than I do more often than not.

      • Ghatanathoah says:

        I’m not sure either. Are they implying that raising children isn’t a “real world accomplishment,” but being an office drone is?

        I might be biased because staying at home reading while taking care of adorable babies sounds like a perfect life to me, whereas having to commute to an office and be separated from my children sounds like hell.

        • Seph says:

          Reading ? When does reading come into it? You are chasing around after small beings that spend most of their waking hours literally trying to kill themselves. I guess there is technically time to read when they’re napping, but mostly you’re trying to clean up the 800 pieces of food they threw on the floor before the ants get it, washing the clothes they last pooped on, sweeping up the bag of flour/toys/etc they overturned when you turned your back for a second to answer the phone…

          • Elephant says:

            That’s definitely the case early on. But after a few years, there can be time for reading, conversations, walks in the woods. (Data: people I’m close to.)

      • Seph says:

        Your wife is probably unusual, though. So much of American society and infrastructure (weak community! No gathering spaces! Car-based unwalkable suburbs so you can’t even take a walk with a stroller and chat to people!) makes stay at home parents unusually isolated here.

        (Source: am an expat stay at home parent)

    • Amanda says:

      I expect there’s a big difference, psychologically, between being a stay-at-home mom in a society where that is the norm, and doing so in a society where working moms are more common (or at least where that is the perception).

      • Matt C says:

        Seconded. Taking care of our babies was pretty hard on my wife. Caring for babies is lonely, taxing, and boring. (Sure, cuddling babies is fun for a little while. Not all day every day. And too often half the night as well.)

        At the time I thought it would have been much easier on her if she could have done it in a house with her mom, her sisters, or friends from the neighborhood, instead of alone with a baby (or a baby and a toddler) all day.

        Also it didn’t help that most of her friends treated her like she’d lost her mind when she explained she wasn’t going back to work and was going to stay home and raise the kids. There’s a message out there that being a stay at home mom is for dumb losers, which is not so fun to dwell on when you’re changing the ninth dirty diaper of the day on hardly any sleep.

    • nyccine says:

      “Stay-at-home moms” are defined as women who are not currently employed and have a child younger than 18 at home.

      No-one arguing about the merits of “stay-at-home moms” – whether that be conservatives praising their value, or liberals arguing that it’s patriarchal oppression – defines “stay-at-home moms” this way; a “stay-at-home mom” is a married woman who raises the kids while dad works.

      Note the weighting:

      Samples are weighted by gender (and marital status, right?), age, race, (marital status? no?) Hispanic ethnicity, education (marital status? Surely we’ll get there soon…), region, adults in the household, and phone status (cell phone only/landline only/both, cell phone mostly, and having an unlisted landline number). (…well I’ll be damned)

      So we know they’ve lumped single, unemployed mothers in with homemakers, although nothing tells us the exact mix, but at this point we already know the survey can’t possibly tell us what the authors want us to believe it does.

      Even if they didn’t botch the basic argument though, take a look at how they determined “happiness;” it’s positive factors like “smiled or laughed a lot” and “experienced enjoyment” weighted against negative factors “worry” and “stress,” as experienced “yesterday.” WHO DEFINES HAPPINESS THIS WAY!? Not only can this study not possibly tell us the biased message they want it to, it can’t even tell us who’s happy and who isn’t.

      edit to add: I left out the part where they mention that while the gaps exist at all levels, it’s at the worst for the poor, but the only data we get to see is the overall levels regardless of income, and the levels below $36,000.00. For all I know, the dramatic gaps at low income levels are dragging down overall response rates, I couldn’t hope to tell you because the data isn’t there.

      • Amanda says:

        What I noticed first was that the choices for one question seem to have been, “thriving,” “struggling,” or “suffering.” I think we need something in between thriving and struggling — something to the effect of, “I dunno; I’m doing pretty well I guess.” I don’t know that it’s likely to mess up their comparisons, but it’s definitely in the Not How I Would Do That category.

      • Diadem says:

        Why is “adults in a the household” not a good proxy for marital status? In fact it might be a a strictly better variable to look at.

        • Lumifer says:

          Because in a fairly common case, that other adult in the household is the grandmother and/or adult siblings.

        • nyccine says:

          Why is “adults in a the household” not a good proxy for marital status?

          Because “married adults” is only a subset of “adults in the household.” “Single adults” are another, as is “adults living with parents” and “adults living with extended family” and “unrelated adults living together as roommates,” etc.

          There’s no reason to use a proxy when you can just get the data; relatively no-one’s hiding their marriages.

          In fact it might be a a strictly better variable to look at.

          Not for what this study is supposedly saying. For that matter, it wouldn’t even be a better variable for anything you might want to use this study for, even if you just wanted see “how is happiness impacted by the number of adults in the home” because everyone would just yell “Jesus, confounding variables much?” and dismiss the study out of hand.

    • TPC says:

      It’s interesting that you define staying home with kids as only happening during the infant and toddler years. I am being kind in your assumptions about it involving no more than wiping noses and bottoms.

      Staying home used to not mean giving up adult interaction or financial independence or real world accomplishment, and that was true not that long ago in America.

      Currently staying home with kids is mostly during the wiping-bottoms years and then mom rushes back to work, but it could go a different way very easily, and that would also change how “happy” SAHMs felt about doing it. One change is of course not defining home-staying as something that can only or should only be done during infant and toddler years. And also defining home-staying as something that doesn’t require constant activity after that window to be legitimate (this is part of why there’s homeschooling, it’s the entry fee to stay home past infant/toddler time for many women instead of the civic roles they used to hold.)

    • Cadie says:

      Does this take into account that stay-at-home mothers are more likely than working mothers to have problems for reasons other than “not having a career?” Like growing up poor (regardless of current income – if you grew up in poverty, “kind of poor” is a step up and isn’t something you’d desperately try to fix), having low-grade mental or chronic physical illness that makes juggling a job and children exceptionally challenging but is mild enough to enable her to do okay at just one or the other, having children with special needs who require more care than a day care center would provide and thus the mother has more caregiver stress, etc.?

      Correlation does not imply causation, and in any of those situations, working outside the home would not fix her original issue. In some it would make no difference and in the others it would actively make things even worse. And don’t forget that being a SAHM by choice is stigmatized in the US, more so with every passing decade. This is not a happiness-enhancer, to say the least. The implication that raising children isn’t a real-world accomplishment is a subtle contribution to that stigma.

      While it’s not the same, I’ve had several jobs and the one I liked best was being a domestic worker: combined nanny and maid. I did not have friends other than my own sister, nor financial independence, since most of my pay was in the form of room and board and the wages themselves were very low (possibly illegally low, looking back, but I didn’t care as long as I had enough to buy stuff like shampoo and new socks, which I did). It was the most personally fulfilling, and I felt like I was actually doing something positive instead of having no impact on the world other than making the CEO trivially richer. I know that at least one woman actually IS happier doing child care and housework than working in an office or on an assembly line, because I’m her. I doubt this is either unique or so rare as to be easily dismissed as an anomaly. Some people prefer work outside the home, some prefer being around kids and having a more domestic life, some prefer doing both. Advantaged people – physically and mentally healthy, quality education, etc. – are more likely to be able to do both. Confounding variables.

  40. Jill says:

    “Intangibles are probably just way more important than money, even amounts of money big enough to raise whole countries out of poverty.”

    No kidding. A large portion of U.S. society is so money oriented that it may be hard for us to look at those intangibles. But in China, if someone criticizes the government, they are likely to be arrested. Such a lack of freedom of speech makes a big difference in happiness.

    Of economically improving societies, China is perhaps the most oppressive society on earth. It’s no wonder one sees Chinese immigrants in most developed countries. Of course people want to leave China.

  41. Jill says:

    As far as India is concerned, it is a very lopsided country in terms of class. While they have a large middle class in absolute numbers, they still have a huge percentage of their country in extreme poverty. Gains in average GDP probably went mostly to the middle and upper classes. But the lower class probably didn’t get much and that’s a large number of people.

    “The different definitions and different underlying small sample surveys used to determine poverty in India, have resulted in widely different estimates of poverty from 1950s to 2010s. In 2012, the Indian government stated 21.9% of its population is below its official poverty limit.[4] The World Bank, in 2011 based on 2005’s PPPs International Comparison Program,[5] estimated 23.6% of Indian population, or about 276 million people, lived below $1.25 per day on purchasing power parity.[6][7] According to United Nation’s Millennium Development Goal (MDG) programme 270 millions or 21.9% people out of 1.2 billion of Indians lived below poverty line of $1.25 in 2011-2012.[8]”

    As far as freedom is concerned, there are still a lot of issues there, although they are not as bad off as the Chinese. They have more political freedome but less social freedom. They can criticize the government. But they have castes still, although there are laws to try to lessen discrimination based on caste.

    There are all kinds of rules about a divorced woman not remarrying and things of that sort. Women are particularly controlled by social rules and expectations. A lot of India still functions like a Third World country, where people are restricted by strange rules and customs. A lot of farmers commit suicide because banks and other lenders are allowed to be extremely predatory toward farmers.

    And India has so many different languages that there isn’t the kind of communication of one part of the country with another that would make rapid modernization or release from restrictive local customs possible.

    In a lot of survey research comparing people of different countries, researchers come up with general questions that they think will be relevant to everyone. But this is not like chemistry or physics where you can narrow everything down to a small number of variables. There are factors about different societies that you need to learn and understand before you have any clue what might affect “happiness” or the reports of happiness, in a society.

    The reports of happiness are another area to look at. What about societies where people are told that they are being ungrateful if they say they are not happy– that they are shaming their husband or family? What about societies where cheerfulness is considered silliness which would get in the way of performing one’s work or being taken seriously? I can’t see how people in societies with such different values and expectations about happiness, could possibly be compared in a valid way, on the basis of their answers to happiness surveys.

    A lot of human communication can’t even be understood at all, unless you know what is valued in the society and what is devalued or a cause of shame.

    E.g. understanding the U.S. may be impossible, unless one understands that we are a capitalist society in which making money is the most important value. Without knowing this, it can seem mystifying, for example, how and why news media has turned so quickly into entertainment media. It could be mystifying why media acts as if truth is unimportant.

    Nothing is important or unimportant– except in terms of what you already value. And our society values making tons of money. So news media does that– and if “news” is no longer news, well no one cares. Because money is valued much more highly than news or accurate information.

    • “we are a capitalist society in which making money is the most important value.”

      I don’t think it’s true. Movie stars, rock stars, sports stars get a lot of status–more than comparably rich, or richer, businessmen or speculators. And it was true of sports stars fifty years ago, back when their incomes, although high, were much lower than those of the very rich.

      An academic who gets a Nobel prize has more status, both within the academic world and outside it, than a university president with twice his income.

      Who has more status–Barack Obama or the Koch Brothers? They are much richer than he is, but the only people who think of them as important figures are those who view them as symbols of evil.

      An interesting difference between U.S. culture and Israeli culture is that in Israel, I am told, it is perfectly acceptable to ask someone in casual conversation how much money he makes or tell other people how much you make. In the U.S., in my observation, it isn’t.

      • B_Epstein says:

        Maybe this is just my experience – but I live in Israel and it is not at all even nearly OK to ask someone how much money they make. It might be OK between friends, it is incredibly impolite (and frequently against one’s contract) to discuss such matters at work, and anyone boasting their salary or status might well be regarded as “those rich b%$@#@ds” with implications of corruption and “piggishness”. I’m not living in a commune or have unusually leftist surroundings.

        • Thanks for the correction. I was probably going on a conversation with an Israeli friend some forty years ago. It’s possible that he was wrong or that norms have changed since. Or that the norms are different in different parts of Israeli society.

      • ThaddeusMike says:

        How often do you think money fails to capture the relevant status differences? I would think it tracks pretty well. Also, the examples you listed seem to be people who make pretty solid money. Is it possible that money is the most important thing, up to a point?

        • I don’t think so. Imagine a sports star or movie star who for some reason isn’t rich, say one who gives away most of his income to good causes. He would still be high status.

          • ThaddeusMike says:

            While earning and spending that money, he would be high status. Afterwards, I’m not so sure. But he’s still making that money first. He doesn’t attain that status without also being a top earner.

        • Joeleee says:

          To take similar examples, but of people with less money. An Olympian in a non-major amateur sport (say waterpolo, field hockey etc.) would be higher status than, say, many well paid truckers, accountants or lawyers, even if they earn significantly less. Money is one relevant metric, but it is not the be all and end all.

          In fact, often higher status positions are traded off against money in the market, so higher status jobs pay less than what the equivalent skills might be useful for in another area.

      • Diadem says:

        Regarding asking people about how much money they make:

        I remember being told that in the US this is a perfectly normal question, but it’s absolutely taboo in The Netherlands (the latter I can personally confirm). Then recently I saw a youtube video from an American who said it was a taboo question in the US, but acceptable in Germany. In the responses to this video where lots of horrified Germans saying this question wasn’t okay in Germany, while explaining how it’s perfectly acceptable in culture X. I’ve read other articles too talking about this question, and always the claim seems to be “Not acceptable here, but normal over there”.

        As far as I can tell, there’s no a single country in the world where this question is acceptable, but everybody seems to think that it’s acceptable ‘over there’.


        • Eve Matteo says:

          “As far as I can tell, there’s no a single country in the world where this question is acceptable, but everybody seems to think that it’s acceptable ‘over there’.”

          Could this be a sign that people wish they could talk about it, so they have this place that it’s okay to do so?

          In the US, it’s starting to become… not exactly polite, but not really taboo either. Like, you get weird looks, but most people will respond without biting your head off. At the workplace, it’s still closer to taboo, but not quite.

          • LHN says:

            Not where I am in the US. I’ve met people who’ll give surprising levels of detail about their sex life or medical issues, but income, never.

          • Edward Scizorhands says:

            Just so guys know, while American workers have very few protections compared to European ones[1], but one protection American workers definitely have is that they are absolutely allowed to discuss wages and employers cannot retaliate for it.

            [1] I’m not saying whether this is good or bad here, although I do have an opinion on it. Right now, I’m just saying that’s what it is.

          • Agronomous says:

            Sounds like The Murders in the Rue Morgue: everybody claims they heard the killer speaking some language that isn’t theirs.

            Spoiler: he was speaking Esperanto; I’m told the Esperatese are very forthcoming about personal income.

        • Anonymous says:

          I’m American. After my Vietnamese girlfriend meet my parents, her parents found it strange that mine didn’t ask about income.

  42. AlexL says:

    I lived in Mexico at the time (a kid, but old enough to understand). The early nineties were horrible:
    – 3 digit inflation was relatively common
    – “soft” rationing, my parents regularly spent a day at the supermarket coming in and out to get cases of canned milk
    – currency rationing, you could buy cheap dollars, but you could only buy a certain amount
    – government corruption was not even discretely concealed
    – rise of the drug cartels started at the time
    – guerrilla uprisings in the southeast

    As an upper-middle class family we were mostly protected from these shocks, but it was a really bad time for the majority of the population.
    [File under “the plural of anecdote is not data”]

  43. Esquire says:

    I continue to be a bit confused by the perspective that reported happiness is a good proxy for “utility”, or “our collective goals”, or anything we much want to optimize.

    Like… all the time I consciously pursue stuff I know will not improve my reported happiness. Versus 10 years ago, my knowledge, relationships, financial position, social “prestige” have all improved a lot, but I don’t think I’m much happier, nor would I have expected to be. But I still value all of those things!

    I much prefer a future where I get more of those things but retain my melancholic affect vs. a future where I get less but really internalize some “positive thinking” training and become more sunny. Even assuming the sunny feeling is totally genuine!

    If it doesn’t govern our personal day to day behavior or long term goals, why should we use it as a yardstick to judge macro trends/policies?

  44. Scott suppolied this link about the women and happiness research, and it’s got a chart of changes over time for men and women– it looks so noisy and with such small overall changes that I don’t think there’s much to conclude or explain…. except that feminism hasn’t made a lot of difference to happiness.

    • I’m sure I was happier back when I thought scientific research and science journalism were reliable.

    • keranih says:

      except that feminism hasn’t made a lot of difference to happiness

      I started to say, I don’t think that’s what feminism is for…and then I thought about other revolutionary ideas, and because I’m on a 1776 kick, I was thinking about life, liberty, and the pursuit of happiness.

      Which was in the original “pursuit of profits” except that it was acknowledged that one should be allowed to go broke, if that’s what turned one’s crank.

      At any rate, I think most revolutions – political and cultural – are focused on the distribution of power, not happiness. Certainly it is not clear to me that the most ‘free’ people – the ones least restricted by other people as to their actions and the ones with the fewest responsibilities to other people – are the most happy.

  45. maznak says:

    Well maybe a big factor is reality vs expectation and which one grows (improves) faster. Maybe the Chinese expect faster growth or improvement or whatever. Also, their political system probably makes them feel powerless, with little control over their destiny or outcomes. Also, they see some people getting arguably undeserved astronomical wealth which increases inequality very fast, and so on.
    From personal experience: in my country, Czechoslovakia back then, the revolution of 89 was like a miracle and no matter the material situation, the higher feeling of empowerment and freedom and much higher participation on decision making etc I am sure for most people created a sharp spike in subjective happiness. Today, when the political system is arguably getting less free, people are aware of others who got rich very fast etc, while their own personal situation improving only slowly or not at all, some corruption still in place, there is some disillusion and I am sure the average subjective happiness is lower than in say 92 or 93.
    My point being, material conditions, as long as you are roughly on par with your neighbors, probably constitute only a fractional contribution to your happiness.

  46. LPSP says:

    The points that immediately came to my head when I read about women’s happiness declining were:

    – modern technology and amenities make easier tasks that women perform better and for wish they are more adapted, thus reducing the value of women who perform in those roles and diminishing their societal need
    – modern technology rewarding abilities and aptitudes seen greater in men, incentivising jobs and economics to appeal to those skills over others

    The traditional women gets to be the blessed caregiver, the mother who performs the vital role of making sure the kids turn out okay and maintaining hearth-and-home within a comfortable environment and support network. Meanwhile men in traditional societies endure relative tedium and drudgery, compared to the dynamicity offered by modern tools (men love their utilities). Women find themselves with only one real strong card left, the ability to offer sex both real AND legal (gotta stipulate those, as the porn industry and prostitution cut in otherwise) and bear children, which is why women that can’t make it in this world flock to third-wave feminism, a cult that validates these feelings and demands for men to show more humility over their advantages AND act like they don’t have an advantage AND grovel for the privilege of sex once a month.

    Then I read what Scott said about having to leave the home to do work, and I laughed because I missed something so obvious. There I am, chasing some big societal theory when a much simpler, likely more explanatory factor is that women love homes as much as men love tools. Modern society offers lots of tools but asks you to leave the home often, so men would get happier but women would get sadder.

    • Seph says:

      Except that working mothers consistently survey as happier than stay at home ones, which kind of blows your whole theory out of the water.

      • TPC says:

        Do you have the data adjusted for number of children and age when giving birth? There’s also the ethnicity/race factor.

        Just offhand, working mothers have 1-2 children while SAHMs have 2-4, and SAHMs are particularly likely to have 3+ kids under age 6 compared to working mothers. Having more children closer together without the older-school feminine networks would cause a big downward shift in happiness without necessarily making the case that women need to all be in the workforce for maximum joy.

      • TPC says:

        Also, “working mother” is generally very broadly defined, and can include women who work very little on a week to week basis. Working a couple of days a month puts a woman into the same working mother category for these sorts of happiness assays as if the same woman worked full time.

      • LPSP says:

        Jesus Christ, it isn’t hard to infer this from the content of Scott’s post. I am taking what he says largely for granted, mostly as a thought exercise than anything else. Is there any reason for this smarmy hostility?

        • TPC says:

          I at least didn’t see your last little bit. It does make my responses not quite connected to your total post, though not unrelated. It is an elegant way of stating some critiques of the modern workplace regarding women.

          • LPSP says:

            I really, really wasn’t talking about your post TPC. That’s why I didn’t respond to your post, and why the content of my post largely agreed with yours.

            There is some level of irony on which this exchange is delicious, let that be said.

          • HeelBearCub says:

            Just as a note or a thought, on this board I try to always include either a moniker or a quote if my comments are directed at a specific comment or commenter. It helps avoid this potential confusion.

          • LPSP says:

            If it’s past the point of direct reply, of course. If not, the reply chance is functioning. The onus is on the reader to be clear.

  47. Scott Sumner says:

    Thanks for the post. It would be really interesting to have Chinese data for the 1970s, before Mao died. At that time, Chinese living standards were well below North Korea’s and the level of repression was roughly equivalent. In other words, China was a pretty horrible place to live. Between 1976 and 1990, China became vastly freer and vastly richer (albeit still poor in absolute terms). But by 1990 the problem of hunger was dramatically reduced–and hunger is certainly painful enough to significantly influence happiness. I mention 1990 because that’s where the survey starts.

    Since 1990, China has become freer in some respects (lifestyle) and less free in some respects (speech). It’s also become less equal, whereas during the 1980s China was becoming more equal (as the early reforms favored the (poorer) countryside.) But most people are much richer, regardless of class.

    Pollution probably has only a trivial effect on Chinese happiness—they have far bigger problems to worry. It didn’t even impact my happiness when I would visit Beijing for a month at a time. But inequality may well be a big factor. It’s also worth mentioning that surveys show the Chinese to be very positive about the direction their country is moving, far more satisfied than residents of other countries. Those surveys may be inaccurate, but the same could be said of happiness surveys.

    It’s all a puzzle to me, but in the following earlier post I suggested that there may be a difference between happiness and utility, and used the puzzling happiness of Mexicans as an example.

  48. FullMeta_Rationalist says:

    Let’s back up. What is happiness? I’m not being philosophical or facetious. I’ve never been able to pin down what “happy” actually means.

    Does happy mean satisfied? Does happiness mean grateful? What about joyful? Pleased? In the flow? Excited? Engaged? Motivated? Comfortable? To me, these are all unique and distinct emotions.

    As best I can tell, happiness is a wide umbrella-term for “any emotion with positive valence”. And I’m not even sure it’s a proper emotion. It seems more like a mood.

    • Sonny DE says:

      I wanted to ask a similar question. What does the word “happiness” mean in the context of “happiness research” and utilitarianism? Is it about an emotion, or a feeling, or a feeling about something, or an intuitive assessment of something, or an intellectual assessment of something? Is it intended to be a vague and variable word, or is there supposed to be some specific meaning? Is it supposed to be obvious whether it means something clear or not, or is there some attempt at hiding something?

      • Sonny DE says:

        I’ve read some pages about it, so I wanted to write my answer here, instead of leaving the question hanging.

        “Happiness” in the context of “happiness research” is just an aggregate, a quantitative description of the results of asking people how “happy” they are, in general or about their lives, using a specified questionnaire. It’s defined as such up front in studies of “happiness,” to distinguish use of the word from other common senses it is understood to have.

        “Happiness” in the context of utilitarianism is an entirely hypothetical construct, hypothesized in order to make hedonic utilitarianism feasible as a technical social project, hypothesizing a feeling of the amount of good or positive value in a person’s life, which may be called being “happy” to some degree, a feeling that a person supposedly has at all times, and one that can simply be integrated to “happiness” over a person’s lifetime and summed over the lives of many and all persons as well, to be in accord with the altruistic principle of utilitarianism.

        So, the meaning of “happiness” in the contexts of research and utilitarianism together is the collision of a social statistical curiosity about the use of the word “happy” and a particular line of philosophical speculation.

        [Edited: removed a couple of paragraphs describing attempts by others to explain the relation between research-happiness and util-happiness, and describing the older common senses of “happiness,” because the comment looked too long and useless.]

  49. FullMeta_Rationalist says:

    I recently watched a documentary on television. I think it was a CNN journalist’s foray into Bhutan. Supposedly, it’s the happiest country on Earth. Naturally, the journalist asked people he encountered how happy they were on a scale of 1 to 10. And sure enough, most responded 10 (if not 11).

    One running theme was that Bhutan has been a very isolated country. But it’s starting to open up, so only recently have they entered an industrial revolution. Many citizens have within their lifetime seen the introduction of enormously-convenient labor-saving technologies such as tractors. This was surmised to have contributed to their happiness.

    The other factor emphasized was their strong Buddhist culture.


    Unrelated to the happiness post. One thing that struck me about the documentary was their government. It was a traditional monarchy. Recently, the King decided to learn about the rest of the world by visiting other countries. During his travels, he learned about this crazy thing called “democracy”. So when he returned to Bhutan, he tried to get his citizens on board with this democracy thing. But his citizens were like “Aww, noo… but we love our king. :(”

    I didn’t get the same “control-freak” vibe that I do from documentaries on North Korea, so I’m fairly confident this sentiment was sincere. It made me question whether democracy is strictly superior to monarchy. Like, what if Bhutan tries democracy and doesn’t like it? It would strongly vindicate Voldbug. Which in turn makes me wonder “under what conditions is monarchy superior to democracy?” and “what were the conditions which precipitated the European Enlightenment?”

    • anonymous says:

      If your king is a genuine buddhist that’s probably about as close as you can get to plato’s ideal.

    • I think I’ve heard it said that monarchy is both the best and the worst form of government, depending on the monarch. Apparently the current king is a good guy, so Bhutan is happy. But the next king might be terrible.

      Also I think that democracy is better for sustained prosperity, since the markets don’t do well not knowing what the next king will be like, and also because democracy encourages more independent thinking, which is the driver for economic advances. I think sustained prosperity provides sustained well being and happiness, but the surveys don’t pick this up because of the tendency for everyone to think current life is normal.

  50. FXKLM says:

    I’d love to see someone do a study asking people how tall they are on a scale of 1 to 10 and then plotting the average response in each country against the actual measured average height in each country. I’m not confident that there would be a strong correlation.

  51. Qiaochu Yuan says:

    Has any attempt been made to control for cultural differences (over both space and time) in how socially acceptable it is to admit to being happy vs. unhappy? E.g. maybe women were always unhappy but now they feel more social permission to say so.

    • Tibor says:

      Exactly. I think that as long as you don’t try to control for that (however difficult it might be), all the happiness data are close to useless. You can perhaps compare culturally similar countries (which also stayed culturally similar over time, which is not always the case – I imagine that North and South Korea are culturally pretty different by now).

  52. j r says:

    The women and happiness thing doesn’t seem like much of a paradox at all to me. The key is to disaggregate the different aspects of women’s liberation.

    Consider South Africa. The ANC is currently running that country into the ground. The ANC is also the party that may have been most responsible for ending apartheid. Some folks like to take those two facts together as proof that maybe apartheid wasn’t that bad or black rule is destined to fail. It’s proof of neither. It’s only proof that the ANC is terrible at ruling (not that I am using “the ANC” as shorthand for a whole basket of poor ideological and policy positions).

    Another example would be Gandhi. Gandhi’s Satyagraha was a system that he begun working on in South Africa, used successfully in India and which MLK put to great effect in the US. Gandhi was quite good at developing tactics of non-violent resistance that could be effective in liberation and minority rights movements, but his economics were atrocious.

    Extend the analogy to women. The set of ideas that aided in women’s liberation were successful in that regard, but perhaps they were also wholly inadequate in preparing women for the reality of whole or partial liberation. Or perhaps what we’ve seen so far is simply an incomplete project, part of a larger women’s liberation movement. I certainly feel that this is the case with other civil rights movements.

    • This is a real interesting idea. It may well be that most liberation movements, no matter how creditable or blameworthy the movement, tend to be terrible at governing. But maybe jr is just cherry picking a few examples where that is the case. Does anyone know of studies on this that are available on the internet or as a book?

      • LPSP says:

        I suppose myself and others could think of similar examples with enough time. I’d say however that it falls into a greater pattern of the difference between wisdom/sagacity and hard understanding. Liberating movements often have a great deal of human wisdom, derived from a place of decency and kindness that proves useful in pointing out every-day injustices and ultimately enacting social change. But they have terrible understandings of any concrete reality. Their in-the-moment tactics are hampered by poor grand strategy, and for every good decision they know how to take, they also make a bad decision they *don’t* know how to avoid. The results are systems that depends on unjustifiable faiths and vapid, unenforceable standards; a backfire is inevitable, as they present malincentives to defect and instigate corrupt and mendacious patterns of behaviour. The liberating groups themselves can’t hack the mess and bury their heads in the sands, or say that it’s always a worthy price to pay for the rights they’ve earned, which is frankly insane. It lends popular credence then to the notion that said movement was a bad thing, which is also equally insane. There is no guarantee of balance ever in this; from what I can tell, only the instance of invasion by a nation both wealthier AND more socially-egalitarian seems to change both these factors for the better.

      • Lemminkainen says:

        I suspect that part of what you’re observing is that most liberation movements face kind of a shitty dilemma after they win. Then can either:

        a: Keep a lot of their society’s old institutions, bureaucrats, etc, and thus forego making the changes they set out to, or

        b: Toss out all more most of the existing institutions, then try to figure out how to build new, functional ones. This turns out to be pretty difficult, especially if a lot of your leadership doesn’t have much experience actually governing stuff. And once your government has actual stakes, members of your coalition, previously united in their hatred of the status quo, will start fighting with each other about stuff.

        Consider, say, the American Revolution and its aftermath. The revolutionaries didn’t have terribly huge ambitions for changing their society, but it took them twelve years to put together a governing structure that wasn’t totally dysfunctional and useless. Then, they had a lucky break (the Napoleonic wars in Europe let them seize a ton of territory), but the new country roiled with constant social conflict over an issue left unresolved by its founding (slavery) until the country fought a massive civil war over the issue seventy years after its founding.

        Although the state probably became stronger and more unified as a result of the civil war, observers at the time drew the opposite conclusion. In fact, there was such a strong consensus that America’s new form of government, democracy, sucked that when the Prussians defeated France in the early 1870s, they forced the country to become a republic in order to cripple it strategically.

        Revolutionary France, the USSR, and Communist China all underwent similar growing pains. Building a new society is really hard, and it takes a while to get it right (even in cases like the USSR, where”right” is just “baseline functional enough to fight a war with Germany”)

        • Mary says:

          And the United States had a shining advantage in that it didn’t want to alter its government on a state level except in trivial ways.

          • Lemminkainen says:

            That’s what I was trying to get at when I said “The revolutionaries didn’t have terribly huge ambitions for changing their society.”

            [The conservatism of the revolutionaries is usually overemphasized a bit by people on the right– Gordon Wood’s “The Radicalism of the American Revolution” suggests that they did actually have a significant critique of their society and seek some significant changes– but a lot of those changes were already ongoing before the revolution happened, and it’s pretty obvious that they weren’t trying to do as much as the Jacobins, or even as much as the Roundheads in the English Civil War.]

  53. Mikk Salu says:

    Ex-Soviet country Ukraine has not enjoyed much economic growth during last 25 years. GDP per capita PPP of Ukraine is 7519 dollars (IMF, 2014). Estonia, Latvia and Lithuania at the same time have enjoyed fastest growth in Europe. GDP per capita PPP of Estonia is 28 592 USD, Latvia 24 712 USD and Lithuania 28 359 USD (IMF, 2014). Maybe it is cut off point, but results look weird if Estonia, Latvia, Lithuania are near the bottom and Ukraine looks pretty good.

    Anyway, I´d be careful comparing different countries. Justin Wolfers makes several good points here
    For instance, Japan has been a prominent case of happiness stagnation, but Wolfers writes: “We returned to the codebooks and had the questions translated. /…/
    First, in 1964 the response categories changed dramatically. The top category was changed from the catch-all “Although I am not innumerably satisfied, I am generally satisfied with life now” to the more demanding “Completely satisfied.” Not surprisingly, the proportion reporting their well-being in this highest
    category declined from 18.3 percent to 4.4 percent. The second category from the top also became more demanding, changing from “Although I can’t say that I am satisfied, if life continues in this way, it will be okay,” to “Although I can’t say I am completely satisfied, I am satisfied.” /…/
    Second, the questions asked from 1958 to 1969 focused on feelings about “life at home,” whereas the focus of the relevant question from 1970 onward was on global life satisfaction. Third, the survey
    question—and the allowable responses—changed again in 1992.”

  54. Eli says:

    Might the intangibles still be calibrated around the sort of lifestyle that they view as “normal?”

    When I answer the question, “how happy are you”, my instinct is to put 1 nowhere near how bad things can get or how unhappy I can be. 1 really is relative to what I see as normal; normal for other people / normal for me.

    But like you were saying, relativity may be applied to freedoms and intangibles more than money.

  55. moridinamael says:

    As I said last time, happiness is not negasuffering and suffering is not negahappiness. Your “happiness” can remain relatively stable even while your level of suffering fluctuates wildly.

    If you break your leg, you’re not “less happy”, but you are going to suffer for a period of time.

    People don’t automatically become happier when infant mortality goes down, but they sure as hell suffer less.

    • Wrong Species says:

      That’s a good point. I recently read Tribe by Sebastian Junger and he mentions how some people can sometimes miss living through terrible wars because it brought a closer sense of community. This shows up in data where less people kill themselves during these times.

  56. Anita Restrepo-Sanchez says:

    On the “paradox” of women’s declining happiness, first, how much was the “higher happiness” reported by women in past periods due to greater cultural force for women to “swallow” their sadness, put on the “happy face”, and claim to be happier than they really were?

    Second, it is one thing when your miserable circumstances are the only possibility you know, and another when you know things could be better and have tried to make them so but are thwarted and held back. Expectations and improvements denied make people unhappier than never expecting better at all. Don’t the revolutions and the unrest most happen not when things have been bad a long time or getting worse, but when things were getting better but then stopped short? Feminism showed there was an alternative to the oppression of the past, but then despite our efforts, improvements have been slow and fallen far short, with so much farther to go still.

    Plus, then there are the ways in which things have gotten worse. In particular, how the internet has made things worse for women. It enables the cyber-harassment, the cyber-stalking, and other cyber violence against women, its fuelled the rise of mysogynistic and (right-wing) hate movements like the “geeks for monarchy”, the Trump campaign, and the “Man0sphere”, and it’s increased the availability, the use, and the hardcoreness of porn (which warps male attitudes toward>/a> women and makes them more inclined to violent sexual acts).

  57. Adam says:

    Consider that there is a bit of an arms race going on in the advertising and marketing worlds between them and the general psyche of consumers. As people have more and more of their needs and wants fulfilled, marketing specialists devise ever more clever, subversive, and subliminal ways to make them feel like their needs and wants are not really fulfilled so they’ll continue purchasing things they don’t really need and may never even use. You also have a similar phenomenon going on with the political class, forever incentivized to convince you that you’re in danger and your country is on the verge of breakdown and disaster, so you’ll vote for them and give them more power to make sure it doesn’t happen.

    Try unplugging from that. It’s hard. I’ve tried, and it works for a time. You feel much better at first, more sane, your mind less cluttered, when you aren’t being 24/7 bombarded by people telling you how terrible life and the world are. But it’s isolating. It’s hard to do this and not become disconnected from the larger world because everyone else is plugged in, and if you don’t follow what’s going on, you lose any basis for identifying with those people. You have nothing to talk about any more, nothing in common. You lose all your friends.

    • Lumifer says:

      and if you don’t follow what’s going on, you lose any basis for identifying with those people

      If your basis for identifying with those people is commiserating how the world is going to hell in handbasket because of evil guys over there, but maybe it’s not so bad if only you buy these shiny things right now, it’s probably a good thing that you lose it.

      Find better friends.

      • Adam says:

        You guys right here are part of what I’m talking about. Actually, this is probably the single most negative place on the Internet I regularly interact with people at. Facebook is bad as a whole, but the actual few people I talk to there aren’t as bad. They certainly don’t control the ads.

  58. Eve Matteo says:

    I know I’ve heard a lot about how happiness has something to do with how you expect things to go vs how they actually go. The expectations our parents and society give us about how our lives should work will therefore react with how our circumstances actually play out to determine how happy we feel. The other factor is the shock factor. That whole thing where, kinda like how a scratch “hurts” worse than a real injury b/c our body doesn’t pull out all the coping strategies for a scratch, and the mind does similar things if the mental issues aren’t really threatening, aren’t really big. That would also affect happiness. So someone who goes through a huge shock might feel happier once everything has settled than someone whose life is currently better but has always been that good.

    Basically, how we expect our lives to go and how much coping strategies our minds are currently engaging will affect our reported happiness regardless of how good our lives actually are.

  59. neonwattagelimit says:

    I recall seeing blog post or essay somewhere – I think it may have even been linked on SSC at some point – which plotted results from happiness surveys against an expected baseline that controlled for the level of economic development. The theory went something like Happniess = (economic development) * (culture), where some cultures were strongly predisposed towards positive happiness, and others predisposed towards negative happiness.

    IIRC, Latin America was far happier than you’d expect given its economic development levels, the Anglo world (US, UK, etc) was somewhat happier, East Asia was somewhat less happy, and the former Soviet bloc was far less happy. Western Continental Europe was, I think, about where you’d expect. I don’t recall where other cultures sat – some may not have had enough data for a meaningful comparison, though it’s possible I just don’t remember.

    Of course, this could be indicating more about how people respond to self- reporting surveys across societies rather than actual happiness across societies. But that’s a problem with happiness research generally. I recall reading an analysis of TripAdvisor reviews which found that, for example, Koreans were far more likely to leaven negative reviews than Americans, even for the same places. Does this mean that Koreans are particularly demanding hotel guests? Or does it just mean that negativity is more socially acceptable in Korea? Or perhaps they have not been as exposed to culturally-enforced grade inflation as Americans are?

  60. HeelBearCub says:

    I find it odd that people are so hung up on the idea that happiness doesn’t move with absolute well-being. Very simple examples immediately spring to mind of why this might not be so.

    Every day the hunters and gatherers of the village go out to check their snares, fish traps and gather edible plants. Typically meals may consist of some fish, small mammals and tubers and greens. Most days each family group prepares their meal from their own catch, but sometimes food may be shared if a family has had hard luck.

    There comes a time when one family brings in a whole, fat, hog as well as some rare flavorful mushrooms. The whole village smells the roasting meat and their mouths water. Some of the meat is smoked and put away, the rest consumed by the family. How happy are the other villagers with the meal of squirrel and carrot that they would have been content with any other day? And what if it now continues like that, one family group consistently bringing back better forage and game?

    • gbdub says:

      Yeah, “happiness is relative” seems like it ought to be the expected result, not the shocking twist. Do we really expect modern humans to be capable of levels of ecstasy or misery that hunter gatherers were not?

      Heck, even in my own life – I would say I was less happy the day after my favorite football team suffered a heartbreaking loss than I was the day after I had knee surgery. Yeah, in the latter case I was obviously in more physical pain and objectively worse off, but we’re talking about how I felt. I’ve been happier slogging up a sand dune on a hot day with a heavy backpack and blisters on my feet than I usually am sitting comfortably on the couch. I’m happier getting a small thoughtful gift from my girlfriend than I am getting a $1k bonus at work. And so on.

      Happiness is not the opposite of suffering. It’s a mental trigger that says “hey, what you’re doing right now is good, keep it up”. It’s a signal with a limited bandwidth that tends to center on your current moving average of wellbeing.

      • HeelBearCub says:

        Happiness is not the opposite of suffering. It’s a mental trigger that says “hey, what you’re doing right now is good, keep it up”.

        Is there research on how happiness increases evolutionary fitness?

        It would be odd if the ability to feel happiness did not increase fitness. Given that, there has to be some component of it that takes into account one’s situation relative to other humans.

    • Maware says:

      Except this doesn’t exist in reality. This is a romantic picture of a noble savage.

      • HeelBearCub says:

        But the reality of hunter/gatherer life isn’t the point of the exercise.

        If you had a “brick” cellphone in 1995 you were at the cutting edge and quite happy to have such a slim compact means of easy communication at the ready. You would be quite unsatisifed with such a thing now, even though the capabilities of that phone may not have changed at all.

  61. Maware says:

    I’m surprised Scott hasn’t thought about how the concept of mental illness may affect this, at least in the USA. It seems that mental illness and therapy went from a rather rich thing to do based on talking in the 70’s, to a near omnipresent phenomenon helped by chemicals in the 10’s. In the past, the idea of depression as a chemical imbalance in the brain would have seemed bizarre to people, I think.

  62. Tsnom Eroc says:

    My bet is that reports on happiness with words in a culture that vary with time can give very variable results, almost to the point of being useless.

    I don’t think, on a given day, that I would say I am “happier” in a nice air conditioned room, vs not having the option of air conditioning, especially perhaps if the day was winding down.

    But I would certainly prefer having air conditioning available.

    Same with mint n chip ice cream and cherry coke. I love those products, and my utility in the world is higher with them then without them. But without there existence, if I was unaware of their existence, I don’t think I would rate live as any worse or better on a scale of questions.

    There also seems to be a strange, entirely verbal “love” of certain hard work that people must say they enjoy, even if the person asking is anonymous.

  63. Tibor says:

    In the European former eastern bloc countries (except probably for Ukraine and Russia) there has been a dramatic increase in both the rule of law, liberalism etc. and material prosperity. By the way Poland was never a part of the Soviet Union, even though obviously all of the eastern bloc states were de facto Russian colonies. Poland in the 80s has already started drifting towards a more free regime with the rise of the Solidarity movement, probably more so than any other communist country at the time. At the same time, the Baltic countries were directly under Soviet rule (i.e. actually a part of the Soviet Union) so they had it much worse than Poland at the time.

    But Estonians are kind of like Finns (their languages are also very similar) and they have a reputation of being quite gloomy people. In fact, I think that differences in culture explain so much of the happiness data that if you don’t account for that, you end up mostly with noise.

  64. ghostofoldnicks says:


    This is sounds highly important. A lot, really a lot of people have issues like “I should exercise, but how to motivate myself” or “I should stop drinking but how to motivate myself”. Basically I need to have all this health and fitness and life on my hand because everybody keeps saying to me I need them but I don’t have anything to use them for so why – that is the general logic! If you can turn them into a pull goal, which sounds nearly impossible as the root problem is that too many people are really superfluous today, not needed, not have a role that needs filling, that could help really a lot of people.

  65. Hunter says:

    One of the big things I’ve learned from Scott’s analyses of experimental findings is that it takes a lot more evidence than I thought to be even decently sure about something.

    And that a good heuristic is probably “The more surprising a finding, the more likely it is to be wrong, and the more evidence it takes for you to take it even moderately seriously.”

    The more it purports to turn the world upside down, the more fuel the evidence engine needs.

    I’ve written an article about just how improbable coincidences are allowed to get, with the punchline being that we shouldn’t act like very improbable occurrences can’t be explained just by chance.

    But now I’m thinking I should add a note on how this applies to the “coincidences” in experimental results.

    After Scott unweaves a study’s story a dozen times or so, you start to get a feel for just HOW often coincidences happen.

    It’s funny, I picked up a similar insight from Unsongbook, just yesterday.

    After watching Scott construct vast edifices of connections out of coincidence (How does he do that? With everything?!) a dozen times, I feel much better calibrated to judge the validity of certain kinds of connections that people draw between things.

  66. Hector_St_Clare says:

    I have some serious doubts about the happiness survey, partly because it seems inconsistent with other things I’ve seen, and partly because I’m skeptical about all self reported happiness studies, because they’re typically quite inconsistent with each other. (I’ve seen the happiest country in the world variously listed as Nigeria, Denmark, and Venezuela, three countries which couldn’t have less in common).

    In particular, the problem I have here is that most opinion polls suggest that Russians, Hungarians, Bulgarians and Moldovans all think their countries were better off under communism, which seems incompatible with the idea that people have become much happier since the fall of communism. (I guess some of this depends on when the endpoints of your time series. The Russian economy collapsed in around 1990, declined until 2000, and then grew quite fast until the early 2010s. So if you’re comparing, say, 1994 and 2010 I can easily believe that people are happier nowadays. Comparing 1989 and 2010 is more dubious. Although it’s possible Russians are relatively happy today because the trajectory of things since 2000 has been positive: in 1990 they were coming off five years of slow economic decline preceded by five years of very slow growth.)