Economic stress or random variation? : Revisiting german reunification as a natural experiment to investigate the effect of economic contraction on sex ratios at birth

Abstract

The economic stress hypothesis (ESH) suggests that economic decline leads to a decrease in the proportion of males born in a population. A multitude of additional influences on sex ratios that often cannot be accounted for empirically make assessing the validity of the ESH difficult. Thus, as a historical quasi-experiment, German reunification constitutes an interesting test case. The economy in East Germany, but not in West Germany, underwent a rapid decline in 1991. In the same year, the sex ratio decreased in East Germany, but not in West Germany. Catalano (2003) interpreted these developments as evidence in support of the ESH. Using more recent and detailed data, we re-examine this case to test an alternative explanation, the random variation hypothesis (RVH). Using aggregate data on sex ratios between 1946-2010 and individual-level data on over 13 million births from the German Birth Registry between 1991-2009, we find evidence supporting the RVH but not the ESH. First, the sex ratio in East Germany shows stronger deviations from the time trend in several years, and is seemingly unrelated to economic developments. The degree of variation is associated with the smaller and decreasing number of births in East Germany during the fertility decline following reunification. The individual-level analysis confirms that the 1991 decrease in the East German sex ratio could also be the result of random variation. A specificity of the East German transformation is the buffering of the consequences of economic decline through integration into the West German welfare state. Therefore, the ESH may be applicable in other transformation cases.