Risk sharing versus capitation

The following classification of payments to insurers might be helpful to under-stand the difference between risk sharing and capitation. Assume a contract period of one year and assume that the capitation payments as well as the risk sharing apply to this year.

With prospective capitation, the payments to an insurer depend on cost predic-tions. These cost predictions are based on variables whose values are known at the start of the year. The cost predictions depend on the way these variables are incorporated in the capitation formula and on their estimated weights. For a specific member with a medical problem that needs treatment during the year, any treatment savings will increase the insurer's profit in that year. Depending on the exact form of the capitation formula, the capitation payment for the member involved in the next year might be 'high'. However, the member's insurer can not be sure that next year it will receive this 'high' payment, because he might switch to another insurer or might die. With risk sharing, the

3. Forms of risk sharing members are designated that year. With retrospective risk sharing, this becomes known during the year or at the end. selection and efficiency are taken into account. Moreover chapter eight

COIll-68

3.4 Risk sharing versus capitation

pares prior costs as an additional risk adjuster with the other forms of risk sharing.

3.5. Conclusions

This chapter summarized forms of risk sharing as well as previous empirical studies on risk sharing as a supplement to capitation payments. It was concluded that risk sharing can take many forms and that a systematic comparison of various forms of risk sharing as a supplement to capitation payments in a regulated competitive individual health insurance market has not yet been per-formed. Potential forms of risk sharing were described. Because all forms of risk sharing are reinsurance-like mechanisms in which the regulator acts as the reinsurer, the essential elements of risk sharing are similar to those of a reinsurance contract. Besides the period to which the risk sharing applies (usually one year), a description of risk sharing should include at least the following four topics: the group of members for whom some risk is shared; the types of care for which the risk is shared; the extent of the risk that is shared;

the price that insurers have to pay to share some risk.

For the empirical analyses in the second part of this study, the following choices are made:

- An insurer is allowed to select itself a certain percentage of its members for risk sharing either at the start of a year or at the end of a year. A special case is risk sharing for all members.

- Risk sharing applies to all types of care within the specified benefits package.

- An insurer is entitled to receive a certain percentage of the costs of a desig-nated member as far as these costs are above a certain threshold.

- It is mandatory for an insurer to contribute to the financing of the risk pool.

It is assumed that the normative costs are reduced proportionally to finance the risk pool. The size of the reduction is set afterwards such that the risk sharing is budget-neutral from the regulator's point of view.

Two forms of risk sharing that apply to all members of an insurer are outlier

3. Forms of risk sharing

risk sharing and proportional risk sharing. Under outlier risk sharing insurers are fully reimbursed for a member's costs above a threshold. In the reinsurance industry this is called excess·of·loss. Under proportional risk sharing an insurer receives at the end of the year a % of the difference between the total actual costs it has incurred and the total normative costs of its members. Because it is assumed that the risk sharing is financed via a percentage of the normative costs per member, proportional risk sharing equals a blended payment system as proposed by Newhouse (1986).

Given the choices that were made with respect to the essential elements of risk sharing, four forms of risk sharing were distinguished: risk sharing for high·

risks, risk sharing for high·costs, outlier risk sharing and proportional risk sharing. These forms of risk sharing can be described formally with four parameters:

· p: The fraction of members that an insurer is allowed to designate for risk sharing.

· D: Dummy variable that indicates whether the designation of members for risk sharing is done at the start of a year (D =0) or at the end (D = I).

· T: The threshold above which the costs of designated members are (partially) reimbursed.

· a: The fraction of the costs of designated members· possibly above a thresh·

old· that is reimbursed.

The following values imply no risk sharing at all: p equals zero, T is infinite, a equals zero. The higher p and a and the lower T, the more extensive the risk sharing. If p equals one, T equals zero and a equals one, the most extensive form of risk sharing arises, i.e. the situation of full east reimbursement.

By choosing different parameter values, one may get different variants of a specific form of risk sharing. Optimizing the tradeoff between selection and efficiency not only includes the determination of the optimal form of risk sharing, but may also deal with determining the optimal variant of a certain form of risk sharing.

Important aspects of the four forms of risk sharing are that the designated members pay the same premium as others and typically would be unaware that their insurer has designated them for risk sharing.

70

3.5 Conelusions

Risk sharing implies payments to insurers based on the actual costs of their designated members. Therefore risk sharing differs from capitation which implies payments to insurers that are independent of the actual costs of their members. As capitation payments can be prospective and retrospective, risk sharing can also be prospective (risk sharing for high-risks) and retrospective (risk sharing for high-costs). Under outlier risk sharing or proportional risk sharing, this difference is irrelevant because for all members some risk is shared.

Risk sharing implies a reduction of an insurer's incentives for efficiency. A rough measure of this reduction is given by the proportion shared expenditures.

The higher this proportion, the lower an insurer's incentives for efficiency. But given a certain proportion shared expenditures, different forms of risk sharing may have different effects on an insurer's incentives for efficiency. Therefore the next chapter develops better indicators of (the reduction of) an insurer's incentives for efficiency under risk sharing.

4.1 Tools fa improve efficiency

4. Efficiency

A negative effect of risk sharing is that it reduces an insurer's incentives for efficiency. Remember that in this study efficiency refers to efficiency in produc-tion or so-called teclmical efficiency. That is an insurer is more efficient than a competitor if it is able to serve the same population with the same quality of care for lower costs 6r with a higher quality of care for the same costs. Because the potential problem of quality skimping is outside the context of this study, an insurer's incentives for efficiency will be measured by its incentives for cost containment. If the capitation payment is the same for each individual, an insurer keeps the entire revenue of cost reducing activities itself. Thus flat capitation payments maximize an insurer's incentives for efficiency. For the same reason an insurer's incentives for efficiency are as great as possible under demographic capitation payments. Previous empirical studies on risk sharing did not quantify an insurer's incentives for efficiency (Beebe, 1992) or used the proportion shared expenditures as a rough measure of the reduction of an insurer's incentives for efficiency (Van Barneveld et aI., 1996). The main purpose of this chapter is to develop better indicators of an insurer's incentives for efficiency under risk sharing.

Section 4.1 and 4.2 argue that an insurer can use several tools for improving efficiency and that the application of such tools may yield large savings. Section 4.3 proposes some methods to measure an insurer's incentives for efficiency given a certain form of risk sharing or given prior costs as an additional risk adjuster. Section 4.4 summarizes the conclusions.