Berry, William and David Lowery. “The Measurement of Government Size: Implications for the Study of Government Growth.” Journal of Politics, 46 (November 1984): 1193-1206.

Most studies of government growth have employed the ratio of government expenditures to the total output of the economy (e.g., GDP) as the measure of the size of government. Unfortunately, this ratio may grow for two reasons: (1) the scope of government activity may broaden; and (2) the costs of providing a constant level of goods and services may rise relative to the prices of goods and services in the private sector. It is clear that most of the extant models of government growth purport to explain the former--that is, changes in the scope of government activity. But by employing the ratio of expenditures to total output unadjusted for relative price changes, empirical tests of these models may be seriously confounding changes in the scope of government activity with changes in the relative costs of goods and services between the public and private sectors. We assess the severity of this problem by testing several models of U.S. government growth from 1948 to 1978 using both undeflated and deflated measures of the size of government.