Stumbling On Happiness by Dan Gilbert – Book Review

Stumbling on Happiness by Daniel Gilbert is in many ways like other popular behavioral psychology books. It’s a New York Times bestseller. It tends to list a lot of ways that humans behave irrationally or incorrectly as shown by academic studies. The writing is casual and accessible. It even has blurbs by other very popular authors like Steven Levitt, Malcolm Gladwell, and Daniel Kahneman.

This book was actually published back in 2007, but I came across a cheap used copy recently and bought it because it had “happiness” in the title. I was interested to see these behavioral quirks applied to happiness instead of the usual economics and money. Here are my notes.

What makes humans different than all other animals? Gilbert posits that humans are the only animals to think about the future. Some animals may do things by instinct like squirrels hiding nuts, but you’ll only find humans getting excited about planning their summer vacation, or fretting about being broke in their old age. I’ve never thought of it that way.

Using our imagination. How do we find out what will make us happy in the future? We use our imagination. But in a nutshell, our imagination isn’t very good.

To start, we don’t remember the past very well. We tend to leave some stuff out and also to fill in other details, all without knowing it. (This can’t be good for eyewitness testimony.) We think something is worth a lot more on the open market if we’ve owned it before (books, cars, stock shares, etc.)

Imagining the future is even worse. We believe that we’d be completely depressed if we were part of a conjoined twin, but actually most conjoined twins are quite happy and have no desire to be split up. The same holds true of many disabled individuals. Here’s another example. Would you rather have $20 in 365 days or $19 in 364 days? Most people choose the $20. But 364 days later, if given the choice again, much more people would choose $19 today vs. $20 tomorrow. The pain of waiting one-day is always the same, it just seems different depending on how your imagination looks at it.

Gilbert also points out that the data we have suggests that having children actually doesn’t bring happiness. Figure 23 in the book (see below) combines data from four different studies that show that marital satisfaction drops after birth and only increases again when the child leaves home. He says that society needs us to believe children bring happiness or else there would be no society. Hmm… I don’t know about this one.

The book ends a bit flat, as the conclusion is that the only way to know if something will make us happy in the future is to ask someone else experiencing the exact same thing right now. The problem is that as humans, we tend to think we’re snowflakes and that possibly can’t be true. (Except it does tend to be true, especially when you ask enough people.) Even the author admits that this is unsatisfying. Other than that, the best we can do is to simply acknowledge that our imaginations are imperfect.

Comments

  1. Great review, Jonathan. Thanks.

    Some thoughts:

    1. “Would you rather have $20 in 365 days or $19 in 364 days? Most people choose the $20. But 364 days later, if given the choice again, much more people would choose $19 today vs. $20 tomorrow.”

    A great example of a classic cognitive bias. Fortunately, there’s a rational method to apply. Which is more valuable to you: $19 today, or $20 tomorrow? Discount the latter by one day and compare! Once you do this simple present value calculation, you’ll find that you’d require a bit over a 5% one-day return. I, for one, know of no riskless way that I can passively (i.e., without doing anything except waiting) earn over a 5% one-day return on $19, so I’d certainly opt for the $20 tomorrow. Easy money!

    2. “The book ends a bit flat, as the conclusion is that the only way to know if something will make us happy in the future is to ask someone else experiencing the exact same thing right now. The problem is that as humans, we tend to think we’re snowflakes and that possibly can’t be true. (Except it does tend to be true, especially when you ask enough people.)”

    I would have no problem accepting that aggregate statistics on what make current 65-year-olds happy will tell me what will make me happy when I’m 65, but for the fact that the world is likely to be a *very* different place when I’m 65 than it is right now. In fact, the degree of change between the present and when I’m 65 is likely to be *much* greater than the degree of change between when now-65-year-olds were my age and the present, due to the accelerating rate of technological change. (If you’re not familiar with this phenomenon, it’s worth reading up on: https://en.wikipedia.org/wiki/Accelerating_change)

    So, if statistics on current 65-year-olds probably aren’t very useful to me right now for planning purposes, what can I do? Unfortunately, not much. All I know is that it’s highly likely that life will be easier if I’m financially independent than if I’m not. (At worst, it won’t matter either way!)

  2. Interesting point about humans and the future, I also hadn’t ever thought about it that way.
    Thanks for the thought provoking review.

  3. Happiness is a tricky thing to obtain. It is well worth reading this research by Princeton, which found that a salary of $75,000 was the max that was needed to obtain happiness: http://wws.princeton.edu/news/Income_Happiness/Happiness_Money_Summary.pdf

    Also remember that pleasure and happiness are not the same thing.

  4. Those graphs do not show that having children makes you have less marital satisfaction at all. It really irks me to see such a bad interpretation.

    The first problem is here is no control, no couples who didn’t have children to compare. We would need to see how marital satisfaction changes over time without kids in the home.

    The other problem is that kids actually seem to decrease the rapid decline in marital satisfaction that happens before the birth of the first child as seen most prominently in studies 2 and 3.

  5. There’s no such thing as riskless. $19 today vs. $20 tomorrow actually is a lot more than just $1 and 1 day. Instant gratification is a premium on top of the pure mathematical evaluation. On top of that, you can be run over by a bus before tomorrow, so the risk is there, though not very likely, but still there.

    All I am saying is that the automatic assumption of one is more “rational” than the other is not totally true.

  6. There is a great TED talk that debunks, or rather explains, the marital bliss graph. The talk is very entertaining.

    http://www.ted.com/talks/rufus_griscom_alisa_volkman_let_s_talk_parenting_taboos.html

  7. @bigmouth – I think you miss the point. It’s fine to want instant gratification and give up $1 in exchange for 1 day. The point is that the preference amount changes depending on how far away that 1 day is.

    @Wes – Just watched the talk, thanks for sharing. About 11:55 in they address the marital chart. I think the variability vs average point is a good one, definitely more highs and lows. Here is a link to the article they mention where a writer interviews the researchers in the 4 independent studies referenced in the chart:

    Does parenthood make us happy?
    http://www.babble.com/mom/relationships/parenthood-and-happiness-children-happy-parents/

  8. Jonathan: “The point is that the preference amount changes depending on how far away that 1 day is. ”

    This is actually quite rational. Having the money in your pocket immediately is riskless, while tomorrow you may forget/be busy, etc.

    On the other hand the chances of forgetting/not being able to collect in 364 or 365 days are more or less the same, so that extra $1 looks good.

  9. @Theresa: You might wanna actually read the book before making any pronouncements about what the research lacks. The results of (good) research is often surprising, and the fact that it’s non-intuitive or that *you* don’t believe it doesn’t make it less true. “Stumbling On Happiness” is a great read, and the research is very rigorous and peer-reviewed.

  10. It wouldn’t surprise me at all if having children reduced the level of marital satisfaction. That doesn’t mean it threatens the relationship or that having children doesn’t add greatly to one’s overall fulfillment in life. On the contrary, I would be surprised to read any study that said that adding children to a marriage did not escalate the level of stress in that relationship. I think the responsibility associated with having kids accrues to the adults, namely the relationship, but in this kind of survey the happiness associated with kids probably wouldn’t be attributed to the marriage, but to the kids themselves. So a small dip down in marital satisfaction is offset 100x by a huge increase in satisfaction associated with having children, which was zero when the before the couple procreated.

  11. bigmouth: “There’s no such thing as riskless. $19 today vs. $20 tomorrow actually is a lot more than just $1 and 1 day. Instant gratification is a premium on top of the pure mathematical evaluation. On top of that, you can be run over by a bus before tomorrow, so the risk is there, though not very likely, but still there. All I am saying is that the automatic assumption of one is more “rational” than the other is not totally true.”

    I was using “riskless” in the common financial sense. That’s the sense in which we say that T-bills are riskless. Obviously, nothing is riskless *all possibilities considered*. The U.S. could default today. That is possible. An asteroid could destroy the earth. Also possible. But we all know this already, so there’s really no point in saying it. It’s still useful to talk about things that we all agree to *treat* as if they were riskless, if they’re as close to riskless as we can reasonably get (i.e., they’re *effectively* riskless).

    Depending on how you define “rationality,” preferring a superior outcome to an inferior outcome is precisely what it means to be rational. However, it may be the case that many people still choose the inferior outcome. Why? Cognitive biases. Choosing an inferior outcome due to the need for instant gratification is a prime example. It’s a form of irrationally placing more value on a temporally proximal outcome the present value of which is less than a more temporally distant outcome.

    If you counter that the temporally proximal outcome actually has a higher present value than the distant one due to the uncertainty of the future, then that just means that we didn’t calculate the present value correctly (i.e., we didn’t take all the right variables into account).

    mb: “This is actually quite rational. Having the money in your pocket immediately is riskless, while tomorrow you may forget/be busy, etc. On the other hand the chances of forgetting/not being able to collect in 364 or 365 days are more or less the same, so that extra $1 looks good.”

    One of the most important rules of argument is that you accept thought experiments as given. :)

    The scenario is that you *do* receive either $20 in 365 days or $19 in 364 days. If you want to ask about a *different* scenario, in which it is possible that you do not receive the extra $1 on the extra day, due to forgetting or being busy, then you would have to specify that scenario. If, in that scenario, we have to worry about the possibility of forgetting or being busy, then do we also have to worry about the possibility that the $19 in your pocket is fake, or that there’s a hole in your pocket, or that you could be robbed as you walk away? If so, then $19 in your pocket isn’t riskless. ;)

    It should be clear that these complicating considerations aren’t relevant to the actual issue, which is how good or bad we are at imagining what will make us happy in the future.

    The point of the example is that a one-day difference is a one-day difference, whether it happens between today and tomorrow, or whether it happens between 364 and 365 days from now. A one-day difference is a one-day difference. So why do people treat it differently depending on whether it’s a nearer or a further one-day difference? In order to answer *this* question, we have to control for irrelevant types of uncertainty, like the fact that you’ve got your wits about you today but that you might not 364 days from now. If you’re assuming that you have your wits about you today, then you have to assume that you’ll have your wits about you 364 days from now. Otherwise, you’re just answering a different question.

  12. It is more than just a 1 day difference.

    Money TODAY is money in hand and guaranteed. Money tomorrow is not guaranteed and has risks.

    Money 300+ days in the future is similarly not guaranteed and has risks. The risks between 364 and 365 days in the future is minimal difference.

    You generally want to take the 20 over 19 in either case. But it really does depend on how much you trust the person promising the $20 in money tomorrow. If the question is asked in a way that ensures the money is 100% guaranteed then thats one thing. But if the question is left ambiguous about the risk or leaves some doubt or concern about getting payment tomorrow then thats quite different.

  13. The marriage graph is interesting. Note that hte chart does not start the y axis at 0 but instead shows 46 to 56.

    Also I would not entirely interpret the chart to mean that children don’t make you happy. Its a chart of marriage satisfaction for one. So its measuring how satisfied you are with your marriage, not your general happyness. Maybe your marriage is suffering but you’re thrilled about your kids. Your marriage may suffer due to finances, disagreements about parenting, lack of time together as spouses, etc. Those are impacts of having children. But thats not the same as saying the children don’t make you happy.

    Its quite common for children and a larger family to complicate finances and make things more difficult. For example if you make $40k and have 2 people then is it more or less easy to support yourself compared to if you make $40k and have 2 parents and 2 kids? Of course more people stretches the money thinner and having less money makes finances more difficult. Simply having higher expenses can make you less happy in general. I’d be interested to see if they controlled for finances in that chart. For example does it show the same impact if household income is very high?

    I believe they did compare to married couples without children. It would be pretty pointless study if they didn’t do that at least. Other media reports about the book mention 7% decrease in happiness versus childless couples.

    Plus yeah… parenthood is a LOT of work. Hard to deny that.

    Of course this is not to say that kids can’t make you happy, as they do certainly do so I’m sure. But what is the total net impact on marital satisfaction?

    Think about this: Does marriage make you happy? Its not just a straight forward yes/no answer is it? It depends. Theres positives and negatives. Every situation is different. They could quite easily do a study and find that married people are less happy than single people. Would people scream bloody murder? I wouldn’t be shocked. I don’t think its all that different than examining if having children makes you happy. Its a much more complicated equation than simply asking if you love your children or spouse.

  14. It is generally true that married people (men, in particular) are happier than unmarried people, taken across a lifetime and across many social situations. What’s also true-and what the research in SOH asserts-is that taken across a lifetime, the presence of children *always* results in lowered happiness.

  15. Isn’t this the same author that claimed 60k was the max for happiness ? Anything more did not make a change in a happiness level. BTW, this same author also has a TED talk about this very topic…so if you don’t want to read the book you can watch the talk. I remember Jonathan even did a post about it.

  16. Andrew: One of the most important rules of argument is that you accept thought experiments as given. :)

    The scenario is that you *do* receive either $20 in 365 days or $19 in 364 days. If you want to ask about a *different* scenario, in which it is possible that you do not receive the extra $1 on the extra day, due to forgetting or being busy, then you would have to specify that scenario.

    Consider a spherical horse in vacuum :)

    In all seriousness such a hypothetical scenario would have little to do with the choices people face in real life.

    Andrew: If, in that scenario, we have to worry about the possibility of forgetting or being busy, then do we also have to worry about the possibility that the $19 in your pocket is fake, or that there’s a hole in your pocket, or that you could be robbed as you walk away? If so, then $19 in your pocket isn’t riskless. ;)

    Of course, nothing is riskless in this life :) But those probabilities are pretty low, presumably.

  17. Regarding the aspect of the book in which the author states that humans don’t remember the past very well:

    I saw a psychology study that discussed this same point, but in a slightly different way. The researcher was trying to decipher whether spending money on experiences, versus spending money on objects (stuff), would result in more happiness. Some people hold the hypothesis that stuff = more happiness (relative to spending on experiences), because stuff lasts for years while experiences are fleeting.

    However, the researcher found a different result. He found that experiences yield greater happiness, and part of the reason is that humans don’t remember things very well. (As you said, this makes eyewitness testimony questionable). And people tend to “mentally edit” their experiences with a rosy glow. They forget about the bad stuff and remember the nice parts.

    So — yes, our memories are pretty bad, as you point out. And that has implications for how we handle money, as well as for our happiness.

Speak Your Mind

*