Happiness over the life course

Grazia passed on this link to a report by Joel Waldfogel:

People with higher incomes today report higher levels of happiness than their poorer contemporaries. At the same time, people today are far richer than earlier generations, but they’re not happier than those who came before them. In light of such wrinkles, a growing cadre of economists has cut out the money middleman and moved to study happiness directly. The latest installment in this genre is a new study by economists David Blanchflower of Dartmouth and Andrew Oswald of Warwick. They document how happiness evolves as people age. While income and wealth tend to rise steadily over the life cycle, peaking around retirement, happiness follows a U-shaped age pattern.

It’s a good news article, with data details:

The authors’ data come from large-scale surveys. The General Social Survey asks Americans to rate their happiness level on a three-point scale, with “very happy” a three, “pretty happy” a two, and “not too happy” a one. The average happiness score in the United States is 2.2. The data, covering people older than 16, come from the years 1974 through 2004 and include about 20,000 men and 25,000 women. Across the Atlantic, the Eurobarometer offers a similar four-point scale (very satisfied, fairly satisfied, not very satisfied, not at all satisfied). The average happiness score in Europe is three. The data include about 400,000 men and women in 11 European countries, from 1975 to 1998.

Analyzing data from these surveys, Blanchflower and Oswald found that for both men and women in the United States and throughout Europe, happiness starts off relatively high in early adulthood, then falls, bottoming out on average around age 45, and then rises after that year and on into old age, especially when one follows tips to ensure your retirement is a happy one.

In this study (as in others), people are happier than their poorer counterparts if they have more income. How does the effect of income on happiness compare with the age effect? In the United States, the steady decline in happiness from age 16 to age 45 has an effect that’s larger than a 50 percent reduction in income-that is, happiness varies more as people get older than it does if you compare significantly richer people to poorer ones. And, equivalently, the 15-year upswing in happiness that follows age 45 is stronger than the upswing that tracks doubling of income. For Europeans, the age-based happiness rise that’s equivalent to the effect of doubling income occurs between ages 35 and 70. Take a look at Key If you’re approaching this stage of life, starting to think about your retirement finances may be something worth considering. As this process may not be easy for everything to think about, even doing a quick google search into how investing will make your retirement better, for example, could help people get an understanding of how best to manage finances. Hopefully, the idea of retirement will start to become a lot less stressful to think about.

Additionally, It may be worth checking out some pro tips to retire stress-free, as there will be a lot to take in when it comes to making this smooth transition from working life to retirement.

There’s an age-period-cohort issue:

The U-shaped happiness pattern is not a completely new finding. But past researchers couldn’t tell whether 55-year-olds were happier than 45-year-olds in a given year because they’d aged or because they were born to a sunnier generation. This study gets around this problem by combining data on people of different ages at different points in time over a quarter-century. The authors can compare not only 55- and 45-year-olds today, but also 55-year-olds today to people who were 45 a decade ago. And when they account for when people were born, the U-shaped happiness pattern remains.

The authors also find that over the last century, Americans, both men and women, have gotten steadily-and hugely-less happy. The difference in happiness of men between men of my generation, born in the 1960s, and my father’s generation, born in the 1920s, is the same as the effect of a tenfold difference in income. In other words, if my father had little money compared to his contemporaries and I have lots of money compared to mine, I can still expect to be less happy. Here, curiously, the European pattern diverges. Happiness falls for the birth years from 1900 to about 1950, and generations born on the continent since World War II have gotten successively happier.

These age-period-cohort things always confuse me. I can’t quite believe that this is quite identified.

Here’s the paper by Blanchflower and Oswald. It’s an interesting mix of theorizing, literature review, and number crunching.

Just a few statistical comments . . .

First off, the results should be graphed, either with time on the x-axis or age on the x-axis. I wanna see that “U-shape”! Beyond that, if the key issue is the pattern with age, it would be good to see more modeling, not just a quadratic curve (age and age-squared). If splines are too much work, then I’d like to see a few age categories. A lot of interpretation seems to be riding on this quadratic assumption. Can you really conclude, for example, that “the minimum point of well-being is estimated at age 49.1”? At the very least, the zillions of data points would allow you to do a binned residual plot and look for departures from the curve by age.

At a technical level, if you are going to use age and age squared, you should at least do some standardization, if not a full standardization then something simple like using (age – 40)/10 so that (a) you can interpret the linear term in the presence of the quadratic, (b) you don’t get coefficients like .00026 which are impossible to interpret directly.

Anyway, interesting stuff, and I’m sure that this study will motivate lots more explorations of these questions.

10 thoughts on “Happiness over the life course

  1. Has anyone tried to measure happiness in a more precise way than this 3 point scale, analogous to Gary King's vignettes for health?

    If there's such a thing as absolute happiness measured 0 to 100%, maybe middle-aged people just have higher expectations, so will only give a top rating if they have more than 90%, while younger and older people will be more forgiving. The middle of the age distribution is when people are likely to be married with children working in the job that they've prepared for, owning their home, stable community, so these people may have higher expectations for a top rating for happiness because they've supposedly achieved everything, and yet they may be just as happy on an absolute level. Younger people may have higher expectations for the future than the present, and older people know to expect declining health and dying friends.

    That's kind-of abstract since you can certainly argue that the threshholds define happiness in some way: too-high expectations may themselves make someone less happy.

  2. This reminds me of the example in "How to Lie with Statistics" where out of all people surveyed about magazines Haper's magazine seemed to be the mag of choice…but in reality the cheaper lower class one was actually the magazine of choice. Social pressure/idealism encouraged people to name the higher class magazine as their mag of choice.

    In other words asking folks to report their happiness levels is probably not the best way to assess happiness…too much social pressure to report one way or the other…not to mention that happiness probably swings from day to day.

    But aside from the question of how to assess happiness you bring up good points on the analysis.

  3. I read Happiness by Richard Layard awhile back, which is likely one of the prior studies to which the article refers. He uses the World Values Survey (GSS & Eurobarometer) as a basis for his arguments. The interesting part of this book, I found, was that income had little effect on the level of a person's happiness; rather, it was completely related to the comparison point that person was using, not unlike the findings here.

    More on topic, though, Layard states that age is not related to the level of happiness of people on the whole (see p. 62); tracing one person's happiness through their life course creates a stable average, according to him. Though reading the endnote with this statement, from a study by Easterlin in 2001 ("Income and happiness: towards a unified theory"), it appears that happiness does decline slightly up to 40 years of age and then rises slightly after that.

    Less on topic, this makes me unexcited to be 40 years old one day.

  4. Juli,

    Don't worry. When you're 40, you'll be too busy with your kids to worry about whether or not you're happy.

  5. This whole field of studies makes my head spin. There are so many major problems that are quite intractable. As Theo pointed out, there is a big calibration problem: what does it mean to have a national scale of "happiness"? I see nothing in the survey that says person A's definition of "very happy" is the same as person B's.

    Statistically, the income effect has a counterfactual problem. The real question is if a 40-year-old poor person was not poor but rich, what would have been his/her happiness? That is obviously not measurable since each person at the observation time is either rich or not.

    The age effect does not suffer from counterfactuals since everyone ages but then it has the survivorship bias issue. Some people just don't live as long as others.

    And then age and income are highly correlated factors. Income is very likely to be highly skewed with extremes. etc. etc.

  6. Does this make any sense at all?Since not only "asking folks to report their happiness levels is probably not the best way to assess happiness", but how does reported happiness compares between two individuals regardless of social biases or whatever?What does "happiness" means for each individual?There is an extra layer of subjective "semantics" which cannot be compensated for.So, may be such studies inform us about something but what?

  7. Jean-Luc, in statistics, I believe that you should try not to think about comparing 2 individuals, but rather think about comparing 2 populations of individuals with similar covariates who only differ by how much money they make.

    Even though measuring happiness is subjective for each individual, when you take a random sample, these measures should average out giving you an overall picture of the population. If after controlling for possible confounding variables, the fact that people with more money tend to self-report higher happiness should be meaningful.

    Whether or not this association is causal is another question that is harder to answer, as Kaiser pointed out.

  8. "In other words asking folks to report their happiness levels is probably not the best way to assess happiness…too much social pressure to report one way or the other…not to mention that happiness probably swings from day to day."

    If you have data tracking the same individuals over time, then the first issue can be less problematic. If the same individuals feel similar social pressure over time, then the effect of this can be removed by demeaning the data. Panel data can also help overcome the lack of counterfactual problem (although there are other ways of attempting to get at causation as well, using exogenous changes such as lotteries or instruments for income.)

    If happiness swings randomly from day to day, this shouldn't matter too much for point estimates, although you want to be pretty careful about the comparability of, say, surveys taken in different parts of the year, when things like the weather may have a systematic effect.

    More generally, there are also ways of dealing with the issue of different individual's using different reporting scales: some estimators don't need to assume that different individual's answers are comparable – although they do then need to assume that the same individual's answers are comparable over time. A discussion can be found in Ferrer-i-Carbonell and Frijters (2004) "How Important is Methodology for the estimates of the determinants of Happiness?".

  9. Re: Richard Layard's work

    Below are links to three lectures given by Richard Layard at the London School of Economics. He argues compellingly — well, I found it compelling — is indeed real and measurable to at least a meaningful extent. On page 18 (pg. 19 of the .pdf) of the first lecture is a graph showing happiness against income across countries, and the data plot has the shape of a lower-case r; i.e., above and below a certain range "near the bread line," income and happiness are related, for societal aggregates, anyway.

    Within societies, happiness & income are related inasmuch as one tends to have a target group against which one compares one's income. What is fascinating about this, is that when one does well in comparison to one's income cohort, one is doing harm to others in the cohort. In other words, earning a lot of money compared to the reference group creates a sort of pollution.

    Anyway, the lectures are quite interesting; however, it's been a while since I've read them, and they predate his book "Happiness," so caveat lector.

    http://www.lse.ac.uk/collections/LSEPublicLecture

    http://www.lse.ac.uk/collections/LSEPublicLecture

    http://www.lse.ac.uk/collections/LSEPublicLecture

  10. grad student : Even though measuring happiness is subjective for each individual, when you take a random sample, these measures should average out giving you an overall picture of the population.

    I have no doubt about that but my question is "an overall picture" of what?
    I am not questioning the causality/non-causality of the association with any other measured variable either, just the fact that in contrast to a more "concrete" variable like weight or the number of pizzas eaten per week you hardly know what you are measuring because the "reported happiness" has differing meanings from one individual to another in inscrutable ways.
    The reason I am raising this problem is because for such (so-called) measurements of happiness or any other subjective factor the results of the statistics are used as if the meanings were obvious (according to unspecified common sense), the semantic is "swept under the rug".

Comments are closed.