Autor, Dorn, and Hanson 2019 — When Work Disappears Manufacturing Decline and the Falling Marriage Market Value of Young Men

Summary

Uses gender-specific components of the China trade shock to causally test how manufacturing decline affected the relative economic stature of young men versus young women, 1990–2014. Trade shocks disproportionately reduce male employment and earnings, raise male idleness and premature mortality, deter marriage and fertility, and increase the share of children raised in poor, single-headed households. Confirms both Becker's household specialization model and Wilson's sociological hypothesis — and extends Wilson's findings from the African-American urban poor to all U.S. young adults.

Key Claims

• A one-unit trade shock reduces male manufacturing employment by $2.6$ percentage points (pp); female-specific shocks fall primarily on female employment, enabling clean gender separation.
• Male-female employment gap falls $0.65$ pp per unit shock; male total employment loss exceeds manufacturing loss, indicating spillover to non-manufacturing.
• Trade shocks compress the male-female earnings gap most at the bottom of the distribution: $6$ pp at the 20th percentile vs. $2$ pp at the median.
• Male idleness (neither employed nor in school, ages 18–25) rises $2.6$ pp per unit male-industry shock; displaced women flow into school enrollment, not idleness.
• Male mortality rises $+64.4$ per $100{,}000$ per decade per unit shock, decomposed as: drug and alcohol (D&A) poisoning $+19.5$, human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) $+21.6$, homicide $+14.0$ — excess male mortality accounts for $\approx 14\%$ of the fall in the male share of young adults.
• Marriage prevalence falls $0.95$ pp, fertility falls $1.5$ births per $1{,}000$ women, share of unmarried mothers rises $0.52$ pp, share of children in poverty rises $0.61$ pp per unit shock.
• Shocks to male-intensive employment reduce fertility ($-4.7$) while shocks to female-intensive employment raise it ($+2.0$), confirming the Becker gender-specialization mechanism.
• Scaling to observed China penetration 1990–2014: trade pressure accounts for $\approx 54\%$ of the observed male employment-to-population ratio (EPR) decline, $\approx 14\%$ of the marriage decline, $\approx 40\%$ of the child poverty increase.
• Results hold for non-Hispanic whites, generalizing Wilson's findings beyond Black urban poor.

Concepts Introduced or Extended

• China Trade Shock
• Marriage Market
• Deaths of Despair
• Conditional DI Applicants
• Income-Mortality Gradient

Entities Mentioned

• David Autor
• David Dorn
• Gordon Hanson
• Anne Case
• Angus Deaton

Quotes

"The consequences of high neighborhood joblessness are more devastating than those of high neighborhood poverty … Many of today's problems in the inner-city ghetto neighborhoods—crime, family dissolution, welfare, low levels of social organization, and so on—are fundamentally a consequence of the disappearance of work." — William Julius Wilson, When Work Disappears (1996)

"Wilson's book spoke to me. I wanted to write him a letter and tell him that he had described my home perfectly. That it resonated so personally is odd, however, because he wasn't writing about the hillbilly transplants from Appalachia—he was writing about black people in the inner cities." — J. D. Vance, Hillbilly Elegy (2016)

My Take

A methodologically clean paper that threads three big literatures — trade, family economics, and mortality — into a single instrumental variables (IV) design. The key innovation is decomposing the China shock into male-specific and female-specific components, which allows separation of Becker (relative earnings) from Wilson (absolute earnings) mechanisms — though the authors note the design cannot cleanly distinguish them since both move together.

The mortality findings are the most directly relevant to the Deaths of Despair literature: this is among the strongest causal evidence that manufacturing job loss causes the D&A/HIV/homicide cluster Case and Deaton documented. The implied magnitudes (trade explains $\approx 40\%$ of the child poverty increase) are large but plausible given the scale of the China shock.

Limitation: cannot test whether reversing manufacturing decline would undo family structure effects, since the fracking boom (Kearney and Wilson 2017) found fertility but not marriage responded positively to male earnings improvements.