Warning: this article mentions suicide
Daron Acemoglu writes about the crisis in mental health
According to many indicators, the industrialized world is going through a mental health crisis. According to data from the National Survey on Drug Use and Health, the share of 18-25 year-olds in the US reporting anxiety has risen from less than 8% in 2008 to almost 20% in 2022. The same data show a huge increase in the number of youths experiencing a major depressive episode, starting in the late 2000s and early 2010s. Data from the National Youth Risk Behavior Survey indicate a sharp increase in the fraction of young people considering suicide starting in 2009-2010, and the actual number of suicides among young people (aged between 10 and 24) backs this up, showing a striking rise from 6.8 deaths per 100,000 in 2007 to 11.0 deaths per 100,000 in 2021.
The pattern seems to be similar in other industrialized nations, even if the timing is more varied. For example, data from the Global Burden of Disease show an uptick in anxiety disorders and depressive disorder prevalence in the 2010s globally and across all major industrialized nations. What is remarkable is the suddenness of the increase among the young. The data across a number of nations show a hockey-stick pattern: relative stability, followed by a sudden trend change.
Of course, the main expertise for understanding these epochal changes lies with social psychologists and to some extent sociologists. Nevertheless, such sweeping changes have broader economic and political causes and implications, so there is a case for economists to be involved in the research and the public debates. In fact, in some of the sub-areas, economists have already contributed significantly to our understanding.
There are at least six distinct broad explanations one could consider for the current mental health crisis.
First, what we are seeing today could just be a measurement issue. Mental health problems have been partially destigmatized, and as a result, more people might be reporting mental health problems – without an increase in the underlying prevalence of the problems. If so, talking of the mental health crisis could be misleading and counterproductive, some argue. It seems plausible that destigmatization will pose a measurement challenge. Yet, it is difficult to explain the sudden uptick, the more general hockey-stick pattern, and the increase in actual suicide rates, with measurement problems.
A second and related explanation is about peer effects. Many social phenomena are shaped by one’s peers, and it’s only natural that such peer effects are stronger for teens and adolescents, who are going through a formative period in their lives where social acceptance and concerns related to fitting in are particularly important. Some social scientists and medical researchers have claimed pervasive (and very large) peer effects in all sorts of medical conditions, including mental health. Accordingly, the greater openness with which mental health is discussed may trigger an increase in mental health problems.
There are two problems with this research, however. For one, estimating peer effects is extremely challenging, and many of the medical and social science studies reporting large peer effects are subject to problems very well understood within the economics literature, such as the reflection problem (the tendency of group-level variables to capture group means, which can then lead to spurious estimates of positive peer effects) or common shocks (which affect all members of a group, creating the impression of peer effects when none actually exist). Research by economists attempting to circumvent these problems by exploiting random assignment of peers does not find evidence for major peer effects in mental health. It is also not clear how peer-effects-based explanations would account for the hockey-stick pattern.
A third group of explanations centre on economic changes, including inequality and job loss, driving mental health problems. There is much evidence that job loss is a major negative contributor to happiness, at least in the short run, and inequality can erode social norms and community relations and generate social tensions via comparison effects. While some social psychologists have claimed that the rise in inequality is a major driver of the decline in mental health, it is difficult for inequality dynamics to explain recent mental health trends. It is not only that existing evidence on this is correlational and unsystematic. It is also that while inequality surged in the US during the 1980s, 1990s and 2000s, mental health patterns remained stable during those decades and started increasing only more recently, when inequality has not increased as much.
A fourth group of explanations blames the mental health outlook of the young on calamitous events, such as increasing risk from climate change. A consensus in social psychology maintains that disasters can have negative effects on mental health, and if young people perceive climate change, future joblessness, existential risk from AI, or political events, such as Donald Trump’s election or increasing polarization, as disastrous, this could impact their mental health. Consistent with this, several studies have found that reported mental health is worse among progressive youth than conservative youth. Though intriguing, these explanations are not backed by careful econometric studies yet, and it is difficult to see how they would account for the timing of the mental health trends.
A fifth group of explanations is broadly related: mental health problems could be the general outcome of modernity and modernization, which have reduced the importance of local community ties and created a very different environment in which young people grow up and form social bonds today. This explanation, too, lacks solid empirical support, however. Moreover, it would imply a much more gradual worsening in mental health, rather than the hockey-stick pattern we are observing.
The final explanation, and the one that has received most attention, is related to the last two, but is also very distinct: social media. Social media has become a major vehicle for young people to communicate with each other, and takes up a significant part of their day. In 2022, 95% of teens reported having used YouTube, 67% Tik-Tok, 62% Instagram and 59% Snapchat. Social media often brings calamitous news and has exacerbated the concerns about the social effects of modernization coming from less frequent in-person, embedded interactions. In fact, social media can have first-order negative effects on mental health going beyond these influences, because it often triggers feelings of envy, unfavourable comparisons or outrage, and leads to addiction-like behaviour, including frequent engagements with social media platforms and fewer hours of sleep due to intensive social media use. Some leading social psychologists, including Jean Twenge and Jonathan Haidt, have claimed that social media is a major driver of the mental health crisis, for all of the reasons mentioned in this paragraph and more.
Other experts have challenged this, arguing that most of the evidence is correlational. Some social psychologists have also pointed to potential benefits from social media for teens who can find a virtual community to help them with various problems.
It is difficult to adjudicate this debate because we lack reliable causal evidence. Circumstantially, social media could be just the right factor – because it arrived upon us discontinuously in the mid-2000s with smartphones and leading platforms such as Facebook and Instagram, so it could explain the hockey-stick pattern. As Haidt points out, the fact that mental health problems are much worse among teen girls than boys is also consistent, since the former group uses social media more intensively and more often in activities related to social comparison. But the critics are right that the evidence, which is correlational and not always consistent, is not yet at the level that can support strong claims made by Twenge, Haidt and others.
This is where more careful econometric and experimental work can be useful. The two studies that provide arguably the highest-quality evidence on the mental health effects of social media are by economists. Hunt Allcott, Luca Braghieri, Sarah Eichmeyer, and Matthew Gentzkow designed a randomized control trial in which a subset of users were incentivized to deactivate their Facebook account. They found that those induced to stay off social media spent more time in off-line activities, including in social interactions with friends and family and consuming traditional media. They also report improved subjective well-being and mental health. (The catch was that, despite this, they went back to Facebook after the experiment was over.) This evidence thus supports the view that social media is bad for mental health.
Even more telling is the evidence in the study by Luca Braghieri, Ro’ee Levy and Alexey Makarin, who exploit the staggered rollout of Facebook across US campuses in the mid-2000s. These authors estimate that access to Facebook in a college campus is associated with a large negative impact on various measures of mental health, including anxiety and depression, and their results hint that these effects may be driven by social comparisons induced by this new medium.
There is much more to be done to understand the effects of social media on mental health, happiness and other social outcomes, and economists can play an important role in building up the evidence here. The more important contribution from economics may not lie in the use of state-of-the-art econometric efforts for evidence, but in providing a more holistic perspective on mental health. Social media is clearly a social phenomenon and its causes and consequences cannot be understood by studying individual behaviour in isolation.
New economics research provides some clues on this social aspect. Work by Leonardo Bursztyn, Benjamin Handel, Rafael Jimenez and Christopher Roth presents evidence from online experiments that social media acts like a “collective trap”. Bursztyn and colleagues find that individual users would be happier if they and their social network were off social media, but do not themselves want to deactivate their social media account when their friends are actively using these networks (which also potentially explains the evidence by Allcott et al. that people went back to Facebook even after their improved individual outcomes during the deactivation period).
Overall, both mental health trends and social media’s broader societal and political effects deserve greater attention from economists.
Daron Acemoglu, 2 September 2024
If you or someone you know is concerned about your mental health, help and support is available.
References
Allcott, H., Braghieri, L., Eichmeyer, S. and Gentzkow, M. (2020). The welfare effects of social media. American Economic Review, 110(3), 629-76.
Angrist, J. D. (2014). The perils of peer effects. Labour Economics, 30, 98-108.
Braghieri, L., Levy, R. and Makarin, A. (2022). Social media and mental health. American Economic Review, 112(11), 3660-93.
Bursztyn, L., Handel, B. R., Jimenez, R. and Roth, C. (2023). When product markets become collective traps: the case of social media. NBER working paper no. 31771.
The Economist (2024). Are American progressives making themselves sad? 4 April.
Fowler J. H., Christakis N. A. (2008). Dynamic spread of happiness in a large social network: longitudinal analysis over 20 years in the Framingham Heart Study. BMJ, 337:a2338.
Gimbrone, C., Bates, L. M., Prins, S.J. and Keyes, K. M. (2022). The politics of depression: diverging trends in internalizing symptoms among US adolescents by political beliefs. SSM – Mental Health, 2, 100043.
Goldmann, E. and Galea, S. (2014). Mental health consequences of disasters. Annual Review of Public Health, 35, 169-183.
Kafka, A. C. (2024). Fighting the mental-health crisis narrative: do young people misunderstand everyday stress? Chronicle of Higher Education, 6 September 2024.
Lucas, R. E., Clark, A. E., Georgellis, Y. and Diener, E. (2004). Unemployment alters the set point for life satisfaction. Psychological Science, 15(1), 8-13.
Macintyre, A., Ferris, D., Gonçalves, B. et al. (2018). What has economics got to do with it? The impact of socioeconomic factors on mental health and the case for collective action. Palgrave Communications, 4, 10.
Manski, C. F. (1993). Identification of endogenous social effects: the reflection problem. Review of Economic Studies, 60(3), 531-542.
Naslund, J. A., Bondre, A., Torous, J. et al. (2020). Social media and mental health: benefits, risks, and opportunities for research and practice. Journal of Technology in Behavioral Science, 5, 245-257.
NCHS: A Blog of the National Center for Health Statistics (2023). Suicide and homicide rates increase among young Americans.
NIH News in Health (2021). The power of peers: who influences your health? September.
Selterman, D. (2023). Why I’m skeptical about the link between social media and mental health. Psychology Today, March.
Zhang, A. (2019). Peer effects on mental health: evidence from random assignment into classrooms. SSRN.