Baumol's diseases: a subsystem perspective

In his paper “Baumol’s diseases: a macroeconomic perspective”, Nordhaus (2008) applies a new testing framework in order to estimate the six hypotheses that lie at the core of Baumol’s (1967) model, following an industry perspective. In this work, I extend Nordhaus’ testing framework to estimate Baumol’s diseases in the US economy over the period 1999-2018 according to a subsystem perspective, by making use of the US Bureau of Economic Analysis input-output tables. In order to check whether Baumol’s diseases depend on the perspective that is followed, I apply both the usual industry perspective and the novel subsystem framework and compare the results. For both subsystems and industries, I do not find robust evidence in favour of the persistent demand hypothesis and the hypothesis of declining nominal value added shares in the progressive sector, while my results do support the cost and price disease hypothesis, the hypothesis of declining employment shares in the progressive sector and the hypothesis of uniform wage growth. As a result, Baumol’s growth disease does not substantially lower aggregate labour productivity growth over the period across both subsystems and industries. This happens mainly because progressive services increase their real output at a faster rate than the economy’s average, restraining the reallocation of nominal value added towards stagnant subsystems or industries and thereby providing a strong palliative against Baumol’s growth disease.