The General Social Survey is a long-running survey of the US population that’s taken place every few years since the 1970s. It asks about respondents’ demographic characteristics, political views and social attitudes. One of the questions concerns happiness with one’s life. Specifically, respondents are asked:
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?
The percentage of Americans saying “not too happy” has been consistently below 20% since the question was first posed in 1972. But in the latest survey, something rather concerning happened: there was a dramatic rise in the percentage saying “not too happy” – from less than 15% in 2018 to more than 22% in 2021 (see below).
There was also a corresponding decline in the percentage saying “very happy” – from 30% to less than 20%. (The survey didn’t take place in 2019 or 2020, so we don’t have data for those years.)
Changes of this magnitude in social surveys are extremely rare, especially when it comes to questions like the one about happiness. Could they be due to some methodological issue with the General Social Survey?
This seems unlikely, as the result is backed up by a recent Gallup poll. (Both the National Opinion Research Center, which administers the General Social Survey, and Gallup are respected polling organisations.) Every year since 2001, Gallup has asked Americans:
Next, I’m going to read some aspects of life in America today. For each one, please say whether you are — very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied. How about the overall quality of life?
The percentage answering “very dissatisfied” has never been above 10%. But in 2021 it jumped to 12% – up from just 4% the previous year. Likewise, the percentage answering “very satisfied” plummeted to 20% – down from 37% the year before.
It seems that 2021 really was a miserable year for Americans. Note that, in both surveys, Americans were substantially happier in 2002 – the year following 9/11 – than they were in 2021. The Gallup survey took place just four months after the attacks, yet only 3% of respondents said they were “very dissatisfied”.
The three factors that could mostly plausibly explain 2021’s dramatic fall in happiness are: the pandemic itself; the response to the pandemic; and the upheaval surrounding the death of George Floyd.
Disentangling these three factors is obviously not easy. However, there’s good reason to believe that the pandemic itself – by which I mean the illness and loss of life caused by Covid – can’t explain such a sudden shift in happiness. Why not?
Well, we know that the fall in life expectancy in the U.S. in 2020 was ‘only’ about 1.8 years, and part of that fall was due to the massive increase in homicide. Now, 1.8 years sounds big, and it is a large year-on-year change. But it only takes the country back 18 years in terms of rising life expectancy.
In other words, U.S. life expectancy was lower in 2001, 2000, 1999 and every year before that. Yet, as we can see in the chart above, happiness was substantially higher back then. In fact, it was substantially higher in the 1970s – when life expectancy was up to six years lower than in 2020.
This suggests that the response to the pandemic – including lockdowns, mandates and the spreading of fear by the media – is a more plausible explanation for the drop in happiness than the pandemic itself. Of course, the total amount of illness and death probably would have been higher in the absence of this response, but I’d argue not much higher.
As noted, disentangling the three factors is challenging, and without additional data it’s difficult to say whether the response to the pandemic, or the social upheaval surrounding the death of George Floyd, mattered more. But it’s possible that lockdowns and other restrictions caused the biggest drop in American happiness since surveys began.