Contribution of Economic Factors to Declines in Mortality during the Twentieth Century
Citation: Preston, Samuel H. (1976) Contribution of Economic Factors to Declines in Mortality during the Twentieth Century.
Notes: In this book chapter, Preston utilizes readily available evidence . . . to estimate the relative contribution of economic factors to increases in life expectancy during the twentieth century (p. 62). The measure of economic prosperity Preston chooses is national income per capita. Previous research examined three relationships between income per capita and mortality; (1) cross-sectional relationships, (2) the impact of the level of income on the rate of change in mortality and (3) the impact of the rate of change in income on the rate of change in mortality. First, regarding the cross-sectional relationship, Preston cites several studies which consistently report a strong, negative correlation (about -.8, on average) between income and mortality. The UN population division has expressed a belief that this relationship is weakening over time, but Preston refutes this on empirical grounds. Second, although the level of income is associated with the rate of change in mortality in some cases, the relationship is probably spurious due to the effect that level of income has on rate of growth (i.e., higher income may be associated with faster growth). In any case, There is no persuasive reason for expecting an association between income level and mortality change independently of their joint association with other variables such as income change (p. 66). Third, if a causal relationship actually exists between income and mortality, than changes in income should be followed by changes in rates of mortality. Malthus, of course, made a negative dynamic relationship between mortality and income a central tenet of his theory. Those who have recently examined the relationship fail to uncover support for the postulated relationship (p. 66). Preston cites studies by Stolnitz (1965) and Demeny (1965) that find almost no relationship between economic events and mortality trends. Data for Preston's analysis come from Kuznets for 1900 and 1930 and the UN Statistical Yearbook for 1930 and 1960. Several findings emerge from his analyses (which generally consist of relatively straightforward regression techniques): (1) The relationship between life expectancy and per capita national income has shifted upward during the twentieth century (p. 68). That is, given a certain level of economic development, life expectancy is greater in 1960 than it was in 1930, and greater still than 1900. This indicates that diseases which are responsive to lifestyle changes from economic improvements (e.g., diarrhea) are generally not responsible for this vertical shift in mortality. Rather, diseases which may be combated by public health and medical technologies, regardless of income level (e.g., respiratory diseases), are responsible for the bulk of this shift. (2) Factors exogenous to a country's current level of income probably account for 75-90% of the growth in life expectancy for the world as a whole between the 1930's and the 1960's. Income growth per se accounts for only 10-25% (p. 72). The implication is that these exogenous factors are improvements in public health and medical technologies. Preston lists other interesting findings (e.g., exogenous factors are key in explaining mortality trends in both developed and developing nations), but they may only serve to obscure the main points presented above. In the context of the debate between McKeown (improved nutrition via economic growth) and Szreter (public health initiatives), Preston's study clearly aligns itself with the latter. Preston argues that even without any improvement in living standards, England's life expectancy would have improved from 50.4 years in 1905 to 69.6 years in 1975. This accounts for 88% of the observed increase over this time period-demonstrating that economic advance was not a major factor in that increase [in life expectancy] (p. 82).