The demographic response to economic crisis in historical and contemporary populations.

Abstract

A review of the impact of economic fluctuations on demographic factors suggests remarkable similarities between the pre-20th century experience of European populations and the current experience of poor developing countries. Studies of pre-industrial European population consistently show a negative association of both nuptiality and marital fertility with grain prices and a positive mortality-price association in time of economic crisis. Mortality generally remains elevated for at least 2 years after the crisis, while fertility is lowest in the year following the crisis and then rebounds to above-normal levels before restabilizing. Recent data on major famines in Bangladesh and China, and on less catastrophic food production short falls or price increases of relatively brief duration in India, Japan, and Taiwan, allow further analysis of the impact of economic conditions. In all 5 scenarios, the timing effects are consistent with the pre-industrial European pattern. However, when the crisis is exceptionally severe (as in the case of China), the mortality response is more immediate. Overall, as Caldwell and Caldwell have hypothesized, increases in mortality and decreases in fertility are equally responsible for population loss in times of economic crises--although mortality plays a larger role in poorer settings. Even in the case of catastrophic events such as famine, the demographic response to an economic crisis rarely has a qualitatively important impact on population trends. For example, the massive famine of 1959-61 in China represented a loss of only a few years of natural increase. Studies of contemporary developed societies such as the US have produced contradictory findings. It is speculated that the research will eventually uncover a pattern of a decline in fertility and perhaps nuptiality after an economic crisis, but little effect on mortality.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 88 citations exist upstream.