Baumol's cost disease and the withering of the state

Most theories of government growth place nearly exclusive attention on real changes in public sector activity. Yet, much nominal post-WWII government spending growth was not in the form of the public sector doing more relative to the general economy (real growth), but in the form of government activities becoming relatively more expensive (cost growth). Baumol's (1967) cost disease model is our best guide to understanding cost growth, but over time, Baumol has offered conflicting hypotheses about how cost growth bears on real growth. Using 1947-2012 U.S. data, we test these hypotheses, along with a more novel expectation, by modifying Berry and Lowery's (1987b) econometric models of real growth in public purchases and transfers to consider the influence of government cost growth on real public domestic spending.