Bertoli, P., Grembi, V. (2017).

The political economy of diagnosis-related groups. Social Science & Medicine190, October 2017, pp. 38-47.



A well-established political economic literature has shown as multi-level governance affects the inefficiency of public expenditures. Yet, this expectation has not been empirically tested on health expenditures. We provide a political economy interpretation of the variation in the prices of 6 obstetric DRGs using Italy as a case study. Italy offers a unique institutional setting since its 21 regional governments can decide whether to adopt the national DRG system or to adjust/waive it. We investigate whether the composition and characteristics of regional governments do matter for the average DRG level and, if so, why. To address both questions, we first use a panel fixed effects model exploiting the results of 66 elections between 2000 and 2013 (i.e., 294 obs) to estimate the link between DRGs and the composition and characteristics of regional governments. Second, we investigate these results exploiting the implementation of a budget constraint policy through a difference-in-differences framework. The incidence of physicians in the regional government explains the variation of DRGs with low technological intensity, such as normal newborn, but not of those with high technological intensity, as severely premature newborn. We also observe a decrease in the average levels of DRGs after the budget constraint implementation, but the magnitude of this decrease depends primarily on the presence of physicians among politicians and the political alignment between the regional and the national government. To understand which kind of role the relevance of the political components plays (i.e., waste vs. better defined DRGs), we check whether any of the considered political economy variables have a positive impact on the quality of regional obstetric systems finding no effect. These results are a first evidence that a system of standardized prices, such as the DRGs, is not immune to political pressures.