Mortality and development revisited.

Abstract

This paper attempts to update results reported in 2 earlier papers about the role of socioeconomic factors in worldwide mortality declines since the 1930s. Preston (1975) demonstrated that the relationship between life expectancy at birth and per capita income (in constant dollars) had shifted between the 1930s and the 1960s. A country at a particular level of national income per capita was estimated to have a level of life expectancy at birth that was, on average, 9.7 years higher in the 1960s than it would have been in the 1930s at the same level of income. That shift clearly was attributable to factors other than measured income gains. To identify the contribution of advances in literacy and nutrition to the apparent shift, Preston (1980) added those variables to income in regression equations estimated with data on 36 countries around 1940 and 120 countries around 1970. For the less developed countries (LDCs), the shift in the relationship between 1940-70 was estimated to be 8.8 years after those variables were introduced along with income. Thus, literacy and nutritional gains were responsible for relatively little of the shift. The goal here is to estimate the amount of shift in the relation between mortality and other development indicators during the 1965-69 to 1975-79 period. The focus is on the 70% of the developing world (exclude China) where, in the aggregate, there are indications of a slowdown in the pace of mortality change during the 1960s and the early 1970s. In all cases a mortality indicator was used as the dependent variable in a cross-national regression analysis that includes data from LDCs and from developed countries. Also, in all cases, the set of independent variables included some transformation of the following: the percentage of adults who were literate, gross domestic product per capita in constant dollars, and the excess of per capita daily calories supplied above 1500. Data were drawn from the standard UN, UNESCO, and World Bank compendia. Contrary to previous periods, the social and economic variables of income, literacy, and nutrition were the dominating factors in explaining mortality decline during the 1965-69 to 1975-79 decade. That greater relative role does not result from faster improvements in social and economic conditions during the recent period or from an increased responsiveness of mortality to social and economic variables, but the exogenous factors appear to have operated with sharply reduced intensity in the more recent period. Reduced international commitment to health in developing countries may be 1 explanation. The results also suggest the major role that can be played by educational change in fostering mortality gains.

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 — 56 citations exist upstream.