Military technology and sample selection bias

While it is well known that labour market fluctuations may affect the supply of labour into particular activities such as crime and military service, other sources of selection bias may be sufficiently powerful to confound hypothesis testing. Selection into military populations, for example, may reflect influences on the demand as well as supply of labour. We argue that changing military technology in the early twentieth century shifted the demand for men of different stature and robustness. Soldiers in the First World War (1914-1918) were shorter on average than those in the Anglo-Boer War (1899-1902) for reasons that had nothing to do with standard of living or business cycle influences on the labour market. Rather, we argue, the mechanization and bureaucratization of warfare increased the relative value of shorter people permitting a decline in the average height of soldiers. Thus, technological change over the period of these two wars affected labour demand in a way that largely explains an apparent fall in heights.