Are You Ready to Take a Slice Out of the “Happiness Pie”?

A new paper shows the flaws in the science behind the “happiness pie.”

Posted Jun 04, 2019

Happiness research, a fundamental area within positive psychology, can provide you with an understanding of how to improve the quality of your life. Within this rapidly-expanding field, the idea of the “happiness pie” (Lyubomirsky et al., 2005) has taken over the popular imagination, if not the scientific literature. Your happiness, the pie model proposes, can be broken down into a 50-10-40 split, reflecting your unique blend of heredity, or your “set point” (50%), circumstances (10%), and the activities you engage in to make yourself happier (40%). According to this model, you can’t change your heredity and your circumstances occur without your control. You can, though, change what you do to make yourself happier. Importantly, the pie model is intended to apply to your chronic happiness, not the moment-to-moment fluctuations in your mood.

Where did this 50-10-40 formula come from? Lyuobomirsky and her colleagues used a process statisticians use to explain a particular behavior that can vary from person to person by measuring, in percentages, the contributions of a set of predictors to an outcome variable (a factor that varies from person to person). Estimates of heritability, for example, reflect a statistical formula that examines relationships among variables taken from groups of people with known biological relationships. The 50% in the happiness pie, then, is intended to show that of the variability within a population in happiness levels, half of people’s scores can be accounted for by the happiness scores of their relatives.

Assuming that all of this is correct, this would leave another 50% to explain outside of biological relationships. Again, using statistical estimates, researchers examine relationships between the happiness of people in given samples with knowledge about the events that occurred to them in their lives, both favorable and unfavorable. A 10% estimate means that there is about a .30 or so correlation between life events and happiness. After removing this 60% of the variation, there is 40% (about a .60 correlation) between what people do to be happier and their self-reported happiness ratings. This is the part you can work on, according to the happiness pie model, because it is within your control in a way that genes and life circumstances are not.

The publication of a new paper by University of Groningen’s (Netherlands) Nicholas J.L. Brown and University of Leipzig’s (Germany) Julia M. Rohrer takes apart the happiness pie, and its supporting research, challenging its logic and its empirical basis. They note that, with few exceptions, the happiness pie was mostly uncritically accepted, because it provided hope that you can always make yourself happier if you just work at it enough. People like simple formulas, and that 40% has become the basis for what “anyone with a book to sell, a course of coaching to offer or a happiness technique to promote is hoping to co-opt” (p.3, citing Whippman, 2016).

Taking a look now at what no academic papers have yet attempted, it’s actually quite easy to see what’s wrong with the logical basis for the happiness equation. According to Brown and Rohrer, the 50-40-10 split comes from a statistical model that examines not just one person’s happiness, but an entire population's. Let’s look at that heritability percent first. When researchers attempt to explain how heritable a trait is, they look at large numbers of people and tracking variation within that entire population, or at least the samples of that population they were able to test. In the case of happiness, the hypothetical 50% represents a correlation of about .70 in happiness scores among related people. The 50% doesn’t refer to your own, personal genes. You and your biological relatives, if you were all tested, might show an average correlation of .70, meaning that the higher your own happiness scores, the higher those of your relatives. The best data supporting heritability are based on relationships between identical twins, but parents and children and ordinary siblings also provide contributions to that figure.

Now, looking at that 10% due to circumstances, the formula doesn’t mean that you have 10% of your happiness due to the events in your life over which you have no control. Again, this percent reflects correlations of happiness scores among people who had the same events occur to them. That 40%, similarly, might reflect correlations among the activities people say they engage in and their happiness scores.

Taking apart variation in happiness scores in this way, which Brown and Rohrer refer to as the “within subject interpretation,” fails because statistical estimates of contributions to happiness scores are based on between person differences. In all fairness, the authors note, it’s possible that Lyubomirsky did not intend for the formula to be applied in this manner, but in the popular translations of the concept, this important qualification was lost.

Another logical flaw in the happiness pie relates to what statisticians call “error terms.” All measurement fails to be perfect, and as a result, researchers must build into their formulas some estimate of variation due to unknown factors that were not measured at all. The happiness pie’s formula, in adding up to 100%, fails to take this random error into account. As the authors note, “Given the vicissitudes of human subjects, it seems rather optimistic to expect that this would ever be zero, even if someone were to actually conduct a single study to examine the relative contributions of genetic factors, circumstances, and volitional activities to chronic levels of well-being” (p. 5).

Moving on to the elements in the pie itself, Brown and Rohrer point out that it’s not so easy to divide the happiness equations into three discrete pieces nor are those pieces even accurately described. As they note, “While it seems popular to assume that high heritability implies low malleability, this is not the case: A trait can be both highly heritable and malleable at the same time” (p. 5). Even intelligence, the authors report, is highly heritable but it can be raised through education. By the same token, life circumstances don’t occur in a completely random fashion. You might be trying to increase your happiness by, to take an extreme example, climbing Mt. Everest. As so many climbers have unfortunately learned in the May 2019 season, by doing so you may have increased your risk of being injured or losing your life altogether. Circumstances influence the actions people take and, conversely, actions have consequences.

Another set of criticisms offered for the happiness equation focuses on what exactly is meant by “circumstances.” In studies on happiness and well-being going back to the 1970s, it appears that many researchers used only a limited number of demographic variables such as age, race, education, income, and sex, or used questionable definitions of others such as “family life stage” (based on the age of the oldest child). The authors reanalyzed some of the archival data sets only to find that demographics account for a far greater percentage of the variance in happiness, perhaps as much as 18 or 26%, depending on which exact statistics were used.

From another angle, those “volitional activities" contributing 40% to happiness, as the Dutch-German research team points out, aren’t necessarily matters of choice for everyone. Quoting from the Lyubomirsky et al. original article, the types of volitional activities that they were considering took on decidedly middle-class terms: “rather than running on a track, a fitness-seeking wilderness lover might instead choose to run on a trail through the woods… rather than learning classical pieces, a jazz-loving piano student might instead choose to work on jazz standards” (p. 122). Subsistence workers, by contrast, don’t have the same range of choices to improve their own happiness by taking advantage of these activities associated with privilege.

Ending on a fittingly positive note, Brown and Rohrer hope that their analysis will help “positive psychologists to critically re-examine the evidence base for their claims about the ability of people to improve their own happiness” (p. 14). Other critics of the happiness self-help movement take the work to task on other grounds, notably the idea that anyone can overcome objective circumstances that cannot be so readily changed, or whether it's even all that important to make happiness a goal in and of itself.

To sum up, the critical analysis of the happiness pie provides an important object lesson in accepting, without question, any simple explanations that give you a quick and ready formula for achieving psychological health. Finding your own fulfillment is certainly achievable, but to do so means understanding the complexities of the process.

References

Brown, N. J. L., & Rohrer, J. M. (2019). Easy as (happiness) pie? A critical evaluation of a popular model of the determinants of well-being. Journal of Happiness Studies: An Interdisciplinary Forum on Subjective Well-Being. doi:10.1007/s10902-019-00128-4

Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The architecture of sustainable change. Review of General Psychology, 9, 111–131. doi:10.1037/1089-2680.9.2.111.

Lyubomirsky, S. (2007). The how of happiness: A scientific approach to getting the life you want. New York, NY: Penguin.

Whippman, R. (2016). America the anxious: How our pursuit of happiness is creating a nation of nervous wrecks. New York, NY: St. Martin’s Press.