Does competition improve hospital performance: a DEA based evaluation from the Netherlands

Many countries have introduced competition among hospitals aiming to improve their performance. We evaluate the introduction of competition among hospitals in the Netherlands over the years 2008-2015. The analysis is based on a unique longitudinal data set covering all Dutch hospitals and health insurers, as well as demographic and geographic data. We measure hospital performance using Data Envelopment Analysis and distinguish three components of competition: the fraction of freely negotiated services, market power of hospitals, and insurer bargaining power. We present new methods to define variables for each of these components which are more accurate than previously developed measures. In a multivariate regression analysis, the variables explain more than half of the variance in hospital efficiency. The results indicate that competition between hospitals and the relative fraction of freely negotiable health services are positively related to hospital efficiency. At the same time, the policy measure to steadily increase the fraction of health services contracted in competition may well have resulted in a decrease in hospital efficiency. The models show no significant association between insurer bargaining power and hospital efficiency. Altogether, the results offer little evidence that the introduction of competition for hospital care in the Netherlands has been effective.