Aggregate health care expenditure in the United States: evidence from cointegration tests: a comment

Murthy and Ukpolo (1994) recently estimated a model of aggregate health care expenditure for the United States using cointegration techniques. Their choice of dependent variable, however, introduced the possibility of simultaneity bias in the error-correction model they presented. Their results were also affected by errors in their database. In an effort to confirm or invalidate their findings, a slightly modified version of their model was estimated using a corrected database. The results obtained suggested that Murthy and Ukpolo's basic approach to modelling the market for health care may have been under-specified or alternatively should have incorporated separate demand and supply equations.