1. Interventions to Health Care
Objectives
- to address market failure; adverse selection, moral hazard (consumer, producer), physician induced demand, monopoly etc.
- cost containment
Type of Intervention
- Demand-side cost sharing: targets consumer moral hazard
- Supply-side cost sharing: targets supplier moral hazard, physician induced demand and monopoly
- Institutional Responses (e.g. social health insurance, managed care): incorporate both demand-side and supply-side cost sharing, and seek to address all source of market failure
2. Cost Drivers of Health Care
It is important to assess the cause of increasing cost in health care, since some drivers might be good and others bad. Smith et al (2009) found that increases in health care in US was:
- 7.2% due to demographic changes (ageing)
- 5% due to medical price inflation
- 10.8% due to increased insurance coverage
- 28.7-43.1% due to income effects
- 33.9-48.3% due to technological advances in health care
- insignificant fraction due to physician induced demand & fear of litigation; admin costs; health status
among the above cost drivers, increase due to demographic changes and income effects would generally be regarded as acceptable. What is noteworthy is that only around 11% of the increase in health care cost was expected to be caused by increased insurance coverage, which has the potential to cause moral hazard. It implies that waste or inefficiency is not a major driver of health care cost increases.
3. Response to Consumer Moral Hazard: Demand-side Cost Sharing
Price Based Rationing
Price-based demand-side cost sharing aims to reduce excessive health care consumption by shifting some of the burden of health care costs onto the patient. Types of sharing are:
- Coinsurance: pay a certain percentage p of each medical bill
- Co-payment: pay a fixed amount of pay for each medical event
- Deductibles: pay 100% of firs £X of medical costs per year
- Major Injury Insurance: pay nothing beyond some level
This type of interventions are desirable only if the demand for health care is price-sensitive and consumer moral hazard is a significant problem. Two famous studies which tried to address these two assumptions are the RAND health insurance experiment and the Oregon Medicaid experiment.
The Rand health insurance experiment found that:
- price elasticity of demand for health care was -0.1 to -0.2
- for individuals with average characteristics, higher consumption for low coinsurance groups did not lead to significantly better health outcomes.
- for the chronically ill or poor, health outcomes were worse without insurance than with insurance: uninsured forewent medically useful treatments on less generous plans and hence had worse health outcomes
The Oregon Medicaid experiment found that compared to the control group, those who received Medicaid experienced:
- increase in annual health care expenditure
- increased use of preventative care and screening services
- improved self-reported health
- no improvement in physical health (e.g. blood pressure, cholesterol)
The important question is whether the reduction in health care consumption was due to reduced moral hazard, or reduced spending on worthwhile care.
Some issues/problems with demand-side cost-sharing is:
- There is a trade-off between demand-side cost-sharing and insurance: it tries to reduce consumer moral hazard in expense of socially efficient outcome of full insurance for all.
- Supplier moral hazard is a bigger problems in health insurance than consumer moral hazard. This is because: many utility-reducing facets of illness are not covered by health insurance; utility gains from health care is limited to the non-sick.
- It assumes informed consumers responding optimally to price signals, ignoring literature on doctor-patient relationship and overestimating the patients ability to predict useful care.
- There are equity concerns, since: the demand is reduced for effective treatments as well as discretionary treatments; the reduction is found disproportionately by lower income groups.
- There is skepticism on whether willingness to pay captures the social value of allocating a unit of health care: shouldn't the allocation of resource be based on clinical need, not an individual's ability to pay?
Non-price Based Rationing
The main non-price allocative mechanisms are:
- Gatekeeping: clinician, often as GPs, assesses patients to determine their clinical needs.
- Waiting Times: uses the opportunity cost of time spent waiting to ration the activities.
- so long as the cost of the ordeal (waiting time) is higher for the undeserving than the deserving, the ordeal can ameliorate moral hazard of the undeserving and allocate health care more efficiently.
4. Response to Adverse Selection
In theory, market failure due to adverse selection can be eliminated by:
- compelling low risk people to buy insurance from same pool as high risk type (low risks subsidise high risks).
- obtaining better information about individual patient risks, and treat high risks and low risks in separate insurance pools.
Screening Contracts
In situations where you cannot compel people to buy health insurance, you can deal with adverse selection by offering a screening contract to separate the pool for low risk and high risk individuals. One such way of screening is to offer a partial insurance (e.g. coverage for 50% of health care costs) at one price and a full insurance at another price. This can work even if the insurer does not know the risk types of the consumers at the beginning, since consumers reveal their type through their consumption decisions. If properly managed, the insurer can get a separating equilibrium in which high risk types choose full insurance, and low risk types choose partial insurance.
Some problems with screening by partial insurance is:
- it only works if there are more low risk types compared to high risk types
- might not work effectively in health insurance markets because: difference in premiums required to separate low and high risks is probably too high; some high risks are uninsurable without cross-subsidy
- it is still socially inefficient because it involves people getting partial insurance where a socially efficient solution involves full insurance for all
Social Insurance
Social insurance is a government involvement in provision of insurance against adverse events. Its objectives are:
- Consumption smoothing
- Risk sharing
- Redistribution
- Poverty alleviation
One distinct feature of social insurance to other government programs is that the eligibility is often not means tested. Distinguishing features compared to private insurance is that: participation is near-universal, whether through compulsion or other means; financing is often non-actuarial (not related to individual risks and therefore redistributive); insurance contract is often highly incomplete (allows response to unforeseen events).
The generosity (the extent of cost sharing, for example coinsurance) of social health insurance is a matter of discussion. The marginal benefits and costs of increasing the generosity may be:
Benefits
- Governments can overcome the adverse selection problem, providing increased consumption smoothing benefits
- it can compel participation for full insurance at the average actuarially fair rate, thus ensuring socially efficient outcome
- additional consumption smoothing can be enjoyed by increasing overall level of insurance, as a result of overcoming adverse selection
- Equity gains from redistribution
- this can be done in two dimensions: from healthy to sick; from rich to poor
- why does redistribution take a form of in-kind transfers (e.g. health care) rather than cash transfers which are often supported by standard economics? --> this can be because of: political economy arguments (might be more politically acceptable than cash payments); paternalistic arguments; moral hazard in cash-based poverty alleviation programs (in-kind transfer as ordeal)
- Overcoming externalities and market power abuse
- Lower administrative costs
Costs
- Moral hazard (which reduces social efficiency and increases revenue raising costs)
- costs of moral hazard is likely to be higher in social insurance compared to private insurance because: they often have more generous coverage which will worsen moral hazard; funds to pay for excessive consumption must be raised via distortionary taxation
- Classification errors