Deaths of Despair
In a series of papers beginning in 2015 and culminating in the 2020 book Deaths of Despair and the Future of Capitalism, Anne Case and Angus Deaton documented a pattern that reshaped public understanding of inequality in wealthy nations. Mortality among white, non-college-educated Americans was rising — driven primarily by suicide, drug overdose (particularly opioids), and alcoholic liver disease — at a time when mortality in every other demographic group was falling. The pattern could not be explained by poverty in the absolute sense; the affected populations were not materially poorer than previous generations in the same demographic. What had changed was the collapse of the institutional structures — stable employment, union membership, community organizations, the social identity that comes from productive work — that had given meaning and structure to working-class life.
In the AI Story
The mechanism Case and Deaton identified was not economic in the narrow sense. The affected populations had not lost income so much as they had lost the institutional scaffolding that converted income into meaning. Deindustrialization and automation had hollowed out the manufacturing economy; the aggregate statistics showed increased productivity and economic growth; the distributional reality showed communities in freefall. Marriage rates declined. Church attendance declined. Union membership collapsed. The cumulative social destruction produced a mortality signature visible in the data — a rise in deaths from causes that are, in the language of epidemiology, deaths of despair.
The finding has been extended directly to AI. A 2024 paper titled Do Robots Cause Deaths of Despair? tested Case and Deaton's hypothesis against automation data. The finding was not that automation causes despair mechanically, but that automation concentrated in communities without adequate institutional support produces precisely the social destruction the original framework predicted. The implication for AI is uncomfortable: the technology's effects on human welfare will be determined not by its capabilities but by the institutional context in which it is deployed.
The framework also illuminates what the aggregate statistics miss. Measured productivity in the United States continued to rise through the period of the deaths of despair. GDP continued to grow. Employment statistics, viewed in aggregate, did not reveal the catastrophe. The catastrophe was visible only in the disaggregated data — mortality by education, by geography, by cause of death — that Case and Deaton insisted on examining. This is the analytical move that Sen's capability approach had anticipated and that mainstream economics had consistently resisted.
For the AI transition, the deaths of despair framework provides both a diagnostic tool and a warning. The diagnostic tool is the recognition that mortality and well-being statistics are leading indicators of institutional failure that productivity statistics conceal. The warning is that the populations most vulnerable to AI-driven displacement — workers whose existing skills become economically obsolete before alternative capabilities are accessible — exhibit the same structural vulnerabilities as the populations that suffered deaths of despair from earlier waves of automation and trade exposure.
Origin
Case and Deaton first identified the mortality pattern in a 2015 PNAS paper that generated immediate public attention. Subsequent research extended the findings across causes of death, geographies, and time periods, culminating in their 2020 book. The term 'deaths of despair' entered public vocabulary and reshaped policy discussions about the opioid epidemic, deindustrialization, and the fraying of the American social contract.
Key Ideas
Deaths of despair are a mortality signature of institutional collapse. Suicide, drug overdose, and alcoholic liver disease rose together in populations that had lost institutional scaffolding.
The mechanism is not poverty but loss of meaning. Affected populations were not materially poorer than previous generations; they had lost the structures that converted material conditions into lives worth living.
Aggregate statistics conceal the catastrophe. GDP and productivity continued to rise while specific populations experienced social destruction visible only in disaggregated data.
The framework extends to AI-driven displacement. Populations whose skills are obsoleted by AI without access to institutional support exhibit the same structural vulnerabilities as populations displaced by earlier automation waves.
Institutions determine outcomes. The difference between disruption that produces adaptation and disruption that produces despair is the quality of the institutional response.
Debates & Critiques
Some critics have argued that Case and Deaton's framework overstates the role of institutional collapse and understates the role of specific factors — notably the aggressive marketing of opioids by pharmaceutical companies. Case and Deaton's response is that both can be true: specific triggers interact with structural vulnerabilities, and populations with strong institutional support absorb specific shocks that devastate populations without it.
Appears in the Orange Pill Cycle
Further reading
- Anne Case and Angus Deaton, Deaths of Despair and the Future of Capitalism (Princeton University Press, 2020).
- Anne Case and Angus Deaton, 'Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century,' PNAS 112:49 (2015).
- Do Robots Cause Deaths of Despair? (2024 working paper applying Case-Deaton framework to automation exposure).
- Carol Graham, Happiness for All? Unequal Hopes and Lives in Pursuit of the American Dream (Princeton University Press, 2017).