Did the health transition boost economic growth?

Question 1

Malaria eradication in the Americas: childhood exposure

Answer 1

How much does malaria exposure depress labour productivity? Authors take malaria eradication campaigns and conduct a DID in Brazil, Colombia, Mexico and the US

Key findings
- Cohorts born after the campaign earned 10-20% more than cohorts born before it in the Southern US
- Cohorts born after the campaign earned ~40% more than past cohorts in Latin America

Mechanisms
1. Schooling: Important but non-dominant role: increased education through increases in productivity)
2. Mortality selection: Not responsible because it is not of great enough magnitude to explain income hikes
3. Productivity: Likely the dominant channel that raised incomes

Policy recommendation towards malaria eradication as the effect of malaria & eradicating it was broadly similar across countries

Question 2

Health improvements and income growth in the long run

Answer 2

The growth rate of income and health have similar patterns
Core vs periphery: drastic differences and cutting-edge countries kept increasing while others stagnated.
Health catch-up growth is faster the later a country starts due to technological spill-overs, human & capital deepening (India & China)
1lg (GDP/Cap) -> life expectancy by 12.5 yrs
Estimates are subject to bias

Question 3

Recessions, Mortality, migration bias: Evidence from Lancashire Cotton Famine

Answer 3

Exogenous supply shock due to the US Civil war drove cotton prices up creating a violent recession that displaced workers

Key findings
- Residents of cotton textile districts faced a greater mortality risk during the downturn. Evidence suggests increased migration during recession
- Mortality per 1,000 did not decrease as geographical distance increased
- Cotton shock generated 24,418 excess deaths

Therefore, economic downturns cause substantial increases in mortality in a poor setting

Question 4

Disease and Development: The effect of life expectancy on growth

Answer 4

Augmented DID with 2SLS estimates: observe the cut-off in global health improvements around 1940 and their 1st stage 2SLS estimate is for predicted mortality

Key findings
- Decline in predicted mortality increases population by 1.65-2.15 regardless of socioeconomic indicators
- 1 log life expectancy increase -> increase in log total births by 2.15-2.92, with a higher coefficient on LMICs.
- Fraction of population under 20y increases as log life expectancy increases.
- Decrease in predicted mortality -> increases in GDP but not statistically significant: suggesting it is imprecisely estimated and could be a delayed response to GDP from pop and health improvements
- An increase in log life expectancy -> decreases log GDP/capita. This suggests (a) could be from an insufficient time lag, or (b) Malthusian dynamics are at play. Authors say this is a puzzle given the timeframe of analysis