| This dissertation examines the short-term impact of economic fluctuations or "business cycles" on mortality rates. The first chapter is a review of the historical development of ideas regarding the influence of economic conditions on death rates, from the early Malthusian controversies to the modern discovery of the association of economic expansions with spurts of mortality. The second chapter analyzes data from the 50 Spanish provinces in the years 1980--1997. Annual death rates for all causes and sex- and cause-specific deaths rates for particular causes are modeled as functions of the demographic structure of the population and indices of economic conditions at the province and national level, in fixed-effect regression models with the variables in levels, in first differences or in rates of change, and also in specifications with economic indicators lagged up to four years. The results reveal an increase of the rates of death due to the major causes (cardiovascular disease, cancer, injuries, respiratory diseases) coinciding with economic expansions (i.e., a procyclical fluctuation of general mortality) and a countercyclical oscillation of death rates due to suicide. The third chapter is a study of the relationship between fluctuations of the economic activity and mortality oscillations in Sweden in the last two centuries. The relationship is assessed with spectral analysis and other techniques applied to a variety of economic indicators and death rates. The results show an intense relationship, specially strong in the mid decades of the 19th century, between oscillations of death rates and the quality of the harvest in the preceding year. In the last quarter of the 19th century a procyclical oscillation of mortality emerges which persists throughout the 20th century, though considerably weakened in the second half of the century. These empirical results from Sweden and Spain confirm the procyclical fluctuation of death rates found in previous research by Thomas in the 1920s, Eyer and Higgs in the 1970s and Ruhm recently. No evidence is found in favor of Brenner's hypothesis that attributes the increase of mortality during times of economic expansion to lagged effects of economic slowdown 10 to 15 years before. |
