Reducing public health insurance expenditure: a numerical analysis for Germany

Abstract

In recent decades, Germany’s statutory health insurance expenditure increased drastically. To combat moral hazard, moderate copayments for the purchase of prescription drugs were introduced and increased several times. These measures, however, have been insufficient to buck the steadily increasing tide of Germany’s statutory health care expenditure, making further reforms indispensable. Among a multitude of potential reform proposals, such as a switch to a health premia regime, are two policy options that provide for incentives to limit health care demand: mandatory deductibles and further elevating copayments. By combining an applied general equilibrium (AGE) model with abundant empirical data on heterogeneous household types, this article investigates the economic effects of these two policy options, thereby looking at both the current earnings-related SHI system and a hypothetical health premia regime characterized by per capita premia. As a key outcome, we find that the decrease in expenditure associated with both reform options is too small to induce substantial overall effects, most notably, because both the level of deductibles and the copayment rate are rather moderate in international contexts.

Keywords:

• public economics
• applied general equilibrium analysis
• health economics

JEL Classification:

• D58
• I18
• H20

Notes

¹ To ensure the comparability with the earnings-related contribution scenarios, deductible and copayments are not taken into account when determining the compensating transfers.

² We take into account all expenditure apart from prevention cost (EBM 100–164) and other support (EBM 165–173) according to the doctor’s fee scale of 2002 (ASHIPs, 2002).

³ Using the following abbreviations, ls (low-skilled), ms (medium-skilled), hs (high-skilled), e (employed), u (unemployed), p (pensioner), the figure of 46 types of couple households can be calculated as follows: There are 9 = [ls, ms, hs] × [e, u, p] households in which only one spouse is privately insured and there is 1 type with both spouses being privately insured. If no spouse is privately insured, there are 36 = 6 × 6 types: [ls-ls, ls-ms, ls-hs, ms-ms, ms-hs, hs-hs] × [e-e, e-u, e-p, u-u, u-p, p-p].

⁴

⁵

⁶ Actually, the rate of 1.95% to the permanent care insurance applies to parents only, while for people without children this rate amounts to 2.2%.

⁷ This differs from the seminal paper of Phillips (1958), who finds that the wage rate is positively correlated to the amount of unemployment. Blanchflower and Oswald (1995) trace the difference in these findings to aggregation errors of data and call into question the validity the Phillips curve.

⁸ This interpretation is in line with the fact that the wage curve can be theoretically derived from bargaining models and may also be interpreted as an equation of a contract curve (Blanchflower and Oswald, 1994).