Implementation of socio-economic variables in risk adjustment systems: A quantitative analysis using the example of Germany

At least in Western Europe, competitive social health insurance systems have implemented systems of morbidity-based risk adjustment to establish a level playing field for insurers. However, insured persons with specific socio-economic characteristics are still underfunded, leaving incentives for risk selection. In Germany, there is an ongoing debate about (re)implementing socio-economic variables to reduce this undercompensation. This study analyses whether the following four socio-economic groups are systematically under- or over‐compensated under the present risk equalization system in Germany: insured persons with co‐payment exemption (1), recipients of basic income support for unemployment (2), of benefits of social long-term care-insurance (3) and insured persons with reduced earning capacity (4). On this basis, several attempts of incorporating these variables into the German risk adjustment system, allowing a better fit for the socio-economically disadvantaged groups, are examined. With a data set of about 9.2 million insured persons, the performance of the modifications is demonstrated for the German system. The disparate outcomes of the various models in different dimensions necessitate the consideration of trade-offs and their incorporation into the implementation of a model designed to mitigate the undercompensation of the affected insured groups.