Public Finance — The Economics of Public Pensions and Health Insurance

Public Finance — Social Insurance Compels the Purchase of Insurance Markets Cannot Sell

The fastest-growing items of government spending are pensions and health insurance. Both are “insurance,” but unlike private insurance, membership is compulsory and contributions are not proportional to risk. This chapter looks at why they are designed that way and what costs the design creates.

1. Why social insurance?

The information asymmetry logic of Chapter 1 is the starting point.

Compulsory membership eliminates adverse selection, but at a price. Low risks must buy insurance priced above their own risk, creating transfers from low to high risks, and the moral hazard of changed behaviour after joining remains.

2. Two ways to run pensions: funded and pay-as-you-go

In a funded system, each generation’s contributions are accumulated and invested in a fund, and pensions are paid from principal and returns. The return is the return on capital, $r$.

In a pay-as-you-go (PAYG) system, this year’s contributions from workers pay this year’s pensions to retirees. The return a generation receives is the growth rate of the next generation’s total wage bill — the sum of population growth $n$ and real wage growth $g$.

The balanced PAYG contribution rate is the product of the dependency ratio and the replacement rate.

When PAYG is first introduced, the first retiring generation enjoys a windfall, receiving pensions while having paid almost no contributions. The cost of that windfall is shared by all later generations. That is why moving from PAYG to a funded system creates a double burden: the transition generation must fund its own pensions while also paying for existing retirees. Transition is a question of who bears this burden before it is a question of economic efficiency.

Pensions also affect retirement decisions. If pensions are not raised enough when claiming is deferred (if they are not actuarially fair), working longer carries an implicit tax and early retirement increases.

3. The adverse selection spiral in health insurance

Suppose half of those insured are low risks with expected medical costs of 1 million won a year and half are high risks with 3 million won. If the insurer cannot tell them apart, it charges an average premium of 2 million won. If the most low risks are willing to pay for insurance is 1.5 million won, they do not buy it. With only high risks left, the premium rises to 3 million won.

With a continuous mix of risks, this process repeats: each time premiums rise, the relatively low risks drop out. In the extreme, the market disappears. This is the adverse selection spiral. Compulsory membership breaks it by keeping low risks in the pool. An alternative to compulsion is to subsidize premiums to induce low risks to join.

4. Moral hazard and cost sharing

Insurance lowers the price of using medical care, so use rises. This is ex post moral hazard in health care. The higher the cost-sharing rate, the lower the use.

The US RAND Health Insurance Experiment (1974–1982) randomly assigned households to insurance plans with different cost-sharing rates. Per-person medical spending in the 95% cost-sharing group was about 30% lower than in the free-care group, and the price elasticity of demand for medical care was estimated at about −0.2. There was no large difference in average health outcomes, but the free-care group did better on managing some chronic conditions, such as hypertension among low-income participants. The key lesson is that cost sharing reduces both unnecessary and necessary use.

5. Supply-side incentives: provider payment

Health care is a market in which providers judge what is needed. There is room for doctors to use their information advantage to create demand (supplier-induced demand), and how much depends on the payment method.

No method removes both incentives. Real systems mix methods and supplement them with quality indicators.

Check your understanding

In a PAYG pension that maintains a 40% replacement rate, what happens to the contribution rate if the dependency ratio rises from 0.2 to 0.6? It triples, from 8% to 24%. If the contribution rate is capped at 12%, the sustainable replacement rate is $12/0.6=20\%$. Raising the pension age reduces the numerator (beneficiaries) and increases the denominator (contributors), easing both variables at once.

References

• Jonathan Gruber, Public Finance and Public Policy, ch. 12–13, 15–16
• Henry Aaron, “The Social Insurance Paradox,” Canadian Journal of Economics and Political Science (1966)
• Willard Manning et al., “Health Insurance and the Demand for Medical Care: Evidence from a Randomized Experiment,” American Economic Review (1987)
• Michael Rothschild and Joseph Stiglitz, “Equilibrium in Competitive Insurance Markets,” Quarterly Journal of Economics (1976)