“The Institutional Causes of China’s Great Famine, 1959–2196163ya”, 2015-10-01 (; backlinks; similar):
This article studies the causes of China’s Great Famine, during which 16.5 to 45 million individuals perished in rural areas.
We document that average rural food retention during the famine was too high to generate a severe famine without rural inequality in food availability; that there was substantial variance in famine mortality rates across rural regions; and that rural mortality rates were positively correlated with per capita food production, a surprising pattern that is unique to the famine years. We provide evidence that an inflexible and progressive government procurement policy (where procurement could not adjust to contemporaneous production and larger shares of expected production were procured from more productive regions) was necessary for generating this pattern and that this policy was a quantitatively important contributor to overall famine mortality.
…A back-of-the-envelope calculation shows that the inflexible and progressive procurement mechanism explains 32–43% of total famine mortality. Hence, our proposed mechanism is quantitatively important, and at the same time leaves room for other factors, such as GLF policies and the complex political environment of the time, to contribute to famine mortality.
[Keywords: famines, modern Chinese history, institutions, central planning]
…Our study proceeds in several steps. The first step is to document that after procurement, rural regions as a whole retained enough food to avert mass starvation during the famine. Since the entire rural population relied on rural food stores, we compare the food retained by rural regions after procurement to the food required by rural regions to prevent famine mortality. Using historical data on aggregate food production, government procurement and population (adjusted for the demographic composition), we find that average rural food availability for the entire rural population was almost 3× as much as the level necessary to prevent high famine mortality. We reach these conclusions after constructing the estimates to bias against finding sufficient rural food availability. Our findings are consistent with 2005’s estimates of high rural food availability for rural workers and imply that the high level of famine mortality was accompanied by substantial variation in famine severity within the rural population…Another study that examines the determinants of regional procurement levels is 2011. They find that political radicalism increased regional procurement during the famine and explains ~16% of total famine mortality. As such, our mechanism complements theirs in explaining total famine mortality