Sorry, you cannot buy happiness with a $75,000 salary

There’s an economics study about money and happiness that’s getting a lot of attention.

Here’s a sampling of the headlines:

The Perfect Salary for Happiness: $75,000 – The Wall Street Journal
After $75000, Money Can’t Buy Day-to-Day Happiness – BusinessWeek
What does it take to be happy? About $75000 – Los Angeles Times

These headlines seem to say extra money beyond $75,000 isn’t really that important. It raised my skepticism so I decided to take a closer look at the study.

I found two things. First, the study doesn’t really say well-being levels off at $75,000. Second, the study suffers from some methodological constraints that limit its power.

Let’s explore the study closer to see why.

About the study

One of the strengths is how reputable the study is.

The study was headed by Princeton University’s Daniel Kahneman, a 2002 Nobel economics laureate, and colleague Angus Deaton. It was published in the prestigious Proceedings of the National Academy of Sciences, the official journal of the United States Academy of Sciences. In my eyes, this is as credible as a study can get.

The study relies on data of over 450,000 responses from the Gallup-Healthways Well Being Index, which is a daily survey of 1,000 US residents. A regression was done to explore income and the two main traits of life evaluation (how you feel generally about life) and emotional well-being (how you’re feeling day to day, like sad or joyful).

The entire study is interesting and nuanced, and I encourage you to read about it in this pdf.

The part about income and happiness

The part that caught everyone’s attention was a sentence in the summary:

Emotional well-being also rises with log income, but there is no further progress beyond an annual income of $75,000.

This was largely translated into the press as “$75,000 is the magic salary for happiness.” What it really means is that daily emotions of stress or joy don’t appear to improve after a certain income level, as a general trend. And this is not surprising: stress from daily events isn’t really going to be improved by extra money in the bank account.

But with more money you’d probably still feel better about life. And that’s the second statement in the study that is under-reported. Life-evaluation, how you feel about life generally, simply gets better the more money you have. To quote the study:

When plotted against log income, life evaluation rises steadily

So if your daily well-being levels off at $75,000, but your life evaluation steadily rises, then the net result is you’ll feel better with more money. In short, your overall well-being rises with more money. There is no magic number at which it stops.

Issues with the study

As I said before, I truly respect the authors and the journal of the study. I put my faith that they did the best scientific investigation with the available data.

But that’s not to say the study is without flaws.

The first issue is with the data collection. Note the study is based on a phone survey of 1,000 daily Americans. Busy and wealthy people are not likely to respond to a phone survey. Heck, even I’ve passed on my share of Gallup surveys. I would love to hear how many people making $1 million salary actually responded to this survey. But they probably can’t tell me: the largest income group appears to be “$160,000 and higher.”

The second issue is the self-reported data. The income data is based entirely on people reporting their income group. How many people will actually tell the truth? The authors were aware of this issue and even tried to correct for it. Here’s what they said:

The three lowest categories…

Okay, so maybe a small portion of data was thrown out. No big deal to delete 2 percent of the data points.

But here’s the rest of the sentence:

We deleted these three categories…as well as those respondents for whom income is missing (172,677 observations).

Look at how many observations are missing income! Almost 25 percent of the 709,183 survey observations is missing income data. This seems like a big issue to me. It raises the question of how honest anyone answered the income question.

The paper gives the following defense:

Despite the sampling limitations, available evidence suggests that the estimates of population parameters were not compromised; for example, the survey predicted recent election results within an acceptable margin of error

I stay skeptical since the very wealthy are an increasing small fraction and are probably under-represented in the study.

In conclusion

The final discussion in the paper has a proper summary of the results, qualified for their shortcomings:

Our data speak only to differences; they do not imply that people will not be happy with a raise from $100,000 to $150,000, or that they will be indifferent to an equivalent drop in income. Changes of income in the high range certainly have emotional consequences. What the data suggest is that above a certain level of stable income, individuals’ emotional well-being is constrained by other factors in their temperament and life circumstances

And there you have it. There’s no magic salary number for happiness, so go be rich and be happy.