Connor Pozzi
11/24/14
ECON1116

Asymmetry in Health Insurance: Adverse Selection, Welfare Loss, and the Key to Vanquishing Market Distortion

In today’s society, even more than ever before, information is key. With the introduction of such modern marvels as Twitter, mobile notification alerts, and around the clock news coverage, the next generation can be defined as though who want information accessible at all times. Most believe that this accessibility of information is much more efficient, and leads to better outcomes and greater utility for the individual. On the contrary, in the colonial times, information was scarce and often times inaccurate. This caused more frustration, less utility, and poorer outcomes. The accessibility issue of general information has undergone vast improvement over time, but in health insurance markets, information asymmetry has continued to plague the system. This asymmetry causes what is known as adverse selection, in which the market suffers from the “adverse” outcome of only having high risk, costly individuals in the market, as opposed to the lower-risk individuals who could also benefit from insurance. Adverse selection strongly decreases consumer welfare, and leads to an inefficient outcome in which market failure persists in what is known as the adverse selection death spiral. However, in order to solve adversely selected markets, it is imperative that the system allows for the collection of better information (and revert to market equilibrium conditions) to lessen adverse selection in the short run, and mandating insurance coverage so that individuals cannot leave the market (or else they face a penalty) to finally banish adverse selection. Mandated health coverage has been shown to strongly increase consumer welfare, and would rid of all the inefficiencies of the current system.
To begin, the problem of information asymmetry starts the cycle of the lost welfare from adverse selection. As Nobel Memorial Prize winner Kenneth Arrow stated best: “when there is an uncertainty, information or knowledge becomes a commodity. Like other commodities, it has a cost of production and a cost of transmission, and so it is naturally not spread out over the entire population, but concentrated among those who can profit most” (Arrow, 1963). In our insurance markets, there exists a sizeable information gap between buyers (their risk levels) and sellers. The insured generally tend to know their risk levels and health, while the insurers are left speculating what prices they should set based on the market actions of their collective risk pools. This uncertainty of the insured’s health unnerves the firms, who are then caused to raise the premiums so they can cover possible losses. With these higher premiums, only those who value insurance highly and have relatively high future expected claims would pay the premium price. These individuals are usually very sick and unhealthy, and cost the most out of any group to insure. Those who do not value the insurance, the healthier and risk-averse group, will forego payment of the insurance.
To demonstrate this, I will derive a basic example from David M. Cutler and Richard J. Zeckhouser’s Adverse Selection in Health Insurance (Cutler and Zeckhouser, 1998). For simplicity’s sake, this example shows an insurance firm that has two plans—moderate and generous—as well as two groups of individuals—high risk and low risk. We can make the plausible assumption that high-risk individuals will enroll in the more generous plans (broader range of medical services, more flexible, etc.), while the low risk individuals will enroll in the low risk plans.
On the following page is a table that shows the resource costs of covering each group, and the added benefit each group gets from enrolling in the generous plan.

MODERATE PLAN$40 | GENEROUS PLAN$60 | DIFFERENCE$20 | BENEFIT FROM GENEROUS PLAN COMPARED TO MODERATE PLAN$15 | MODERATE PLAN$70 | GENEROUS PLAN$100 | DIFFERENCE$30 | BENEFIT FROM GENEROUS PLAN TO MODERATE PLAN$40 |

LOW RISK

HI
RISK

In this example, the current set-up is efficient if low-risk individuals stay in the moderate plan, and the high-risk individuals stay in the generous plan. That is, the marginal benefit the low risk individuals get from switching to the moderate plan is smaller than the additional cost of switching ($15<$20), and the marginal benefit the high risk individuals have from staying in their plan is greater than the additional cost of staying in it ($40>$30). However, in realistic health insurance markets, we assume that insurers charge the same premium for everyone enrolled in the plan because of asymmetric information (insurers can’t tell whether someone is high or low risk). To cover costs, insurers charge $40 premiums for the moderate plan, and $100 premiums for the generous plan. Thus, for the high-risk pool, the additional cost of staying in the generous plan $60 is now GREATER than the marginal benefit of the generous plan. Therefore, they all move to the moderate plan. With the influx of high-risk individuals into the moderate plan, the premiums will marginally increase as the insurers now have to charge higher premiums to keep pace with the higher costs associated with insuring high-risk individuals. These higher premiums will have an effect on the low-risk individuals in that plan, who through a cost-benefit analysis, will determine that the added security from insurance is not worth the costs, and more and more of them will incrementally drop out of the market until only the costliest high-risk individuals are left. The premiums will finally increase to a rate so high that all individuals will drop out of the market. This is the notion of the adverse selection death spiral, and causes deadweight loss through the disintegration of an entire market. To visually get a sense of the inefficiencies and welfare loss brought on by adverse selection, I will insert and analyze a graph taken from Einav & Finkelstein’s Selection in Insurance Markets: Theory and Empirics in Pictures (Einav and Finkelstein, 2011).

In this simplified model, Einav and Finkelstein examine a textbook insurance market, simplified to exclude such variables as administrative costs, and making the assumption that risk-neutral firms offer a single contract that covers some probabilistic loss. P represents the price (and expected cost) of this contract, and Q represents the quantity of insurance demanded, or the fraction of insured individuals.
The demand curve functions as a standard one relative to other markets, and reflects an individual’s willingness to pay for insurance. Beneath it lays the downward sloping marginal cost curve, which represents adverse selection in the market (those willing to pay the most for coverage have the highest cost to insurers).
The distinction of an insurance market is demonstrated by the relationship between the demand curve and the cost curves, as an individual’s risk type affects demand and also directly determines that individual’s cost. In this market, because we assume that all individuals are risk averse and there are no other market frictions, it is efficient for everyone to be insured (Qeff=Qmax). At this point, Einav & Finkelstein reveal how welfare is lost, as:

“Adverse selection causes the marginal buyer to be associated with lower expected costs than of infra-marginal buyers. Therefore, the AC curve always lies above the MC Curve and intersects the demand curve at a quantity lower than Qmax. As a result, the equilibrium quantity of insurance will be less than the efficient price, illustrating the classical result of under-insurance in the presence of adverse selection.” (Einav and Finkelstein, 2011).

Like I previously mentioned in my analysis of the death spiral, this under insurance comes from those lower-risk individuals leaving the market due to incrementally higher premiums. Therefore, the welfare cost of this market is measured as those who are inefficiently uninsured over the competitive equilibrium. This is shown on the graph as the shaded trapezoid CDEF.
Presenting a major problem in insurance markets, adverse selection has wreaked havoc and caused major welfare losses for many involved parties across thousands of different markets. However, that does not mean it has go on unabated and unsolved. In the aim of destroying the problem for good, there are a series of necessary steps that can be taken by insurers and legislators to lessen the impact of adverse selection in the short run, and completely eliminate the phenomenon in the long run. It is imperative that these methods are installed in some capacity in order to maximize welfare for insurers and the insured, and bring efficiency back to the health insurance market.
In the long run, the most effective method of eliminating adverse selection in healthcare is the combination of medical underwriting and mandatory, universal coverage. Legislation was passed in 2010 (The Affordable Care Act) to put this into action. As of this writing, data on how the Affordable Care Act and the mandate isn’t readily available, and its impact will require time to see how it dealt with adverse selection in employer and individual insurance markets. In a later section, I will chronicle how a mandate is the most efficient way to destroy adverse selection, by looking at the example of the “Minimum Creditable Coverage” plan of the Massachusetts Health Reform of 2006.
As mandates require a lot of political clout (as through debate and struggle over the ACA) to pass, there are a number of short-term strategies that should be employed to deter adverse selection. In employer-based plans, companies can enact proportional subsidies as a form of risk adjustment that will encourage differentiation between the moderate and generous plans. Cutler and Zeckhouser confront this as they describe the situation in which “employers pay a fixed proportion of the premium cost of whatever plan the employee chooses, and the employee pays the rest “ (Cutler and Zeckhouser, 1998). However, they also add that “the interplay between the employer’s pricing rules and the distribution of sickness in the population determines the severity of adverse selection” (Cutler and Zeckhouser, 1998). Compared to an equal-contribution subsidy, (in which employers pay the difference of the AVERAGE COST between two plans), proportional subsidies have been shown to be more efficient, and predictably better liked, as premiums are less expensive because “the employer pays more and the employee pays less for the generous plan “ (Cutler and Zeckhouser, 1998). The more people who stay in the generous plan and are getting aid, the less adverse selection there will be in the market. .
In combination with this method, employers need to implement policies that are a bit more long-term: prospective and retrospective risk adjustment. Prospective risk adjustments use information about pre-existing conditions to measure each plan’s expected spending. Then, as Cutler and Reber state, “the employer could vary the payment to plans based on health status differences. Plans with less healthy enrollees would get more than the average, and plans with healthier enrollees would get less than the average “ (Cutler and Reber, 1998). Retrospective adjustment is similar in the sense that differences in utilization and payment adjustments are made at the end of an enrollment period.
The long run policies are ones that will eliminate adverse selection, not just solve them in the short run. The government needs (and has) implemented these regulations to an extent, but many states are still resisting some of the regulations. In order to avoid greater welfare losses from adverse selection, these policies need to be implemented and enforced strictly to avoid economic turmoil and a lower social standard from spiraling healthcare costs.
As I have already mentioned, adverse selection stems from a lack of information. If insurers had information readily available information and were able to use pre-existing conditions, geographic, demographic, and other factors to set premiums, then adverse selection would be practically non-existent. As Rothschild and Stiglitz wrote, “If only the high-risk individuals would admit to having high accident probabilities, all individuals would be made better off without anyone being worse off. “ (Rothschild and Stiglitz, 1976). Hence, if the government wants to solve this issue, they must power through the dissenters of what is known as “medical underwriting”, and convince them that “although higher-risk individuals could face higher premiums in the short run, they would be in better shape than losing their premiums from a death spiral” (Cutler and Zeckhouser, 1998).
In addition to the underwriting process, the clear-cut solution to the problem of adverse selection is the mandate. A government mandate requires everyone in a specific locality to be insured, and prevents individuals from leaving the market. If they did leave, they would face a tax penalty, which according to a study of the Massachusetts Health Reform of 2006, “must be sufficiently large such that the consumer with the lowest willingness to pay is willing to purchase health insurance if it is offered at average costs of all consumers plus the post-reform markup that insurers charge on top of realized average costs. (Hackmann, Kolstad, and Kowalski, 2013). The Massachusetts study showed that the individual market for health insurance was adversely selected prior to reform. Because of community rating (prohibit rate variations based on demographics) and guaranteed issue (inability to discriminate based on pre-existing conditions.) the market experienced considerable welfare losses. After the reform, coverage was increased by 19.4%, and a reduction of in the average cost of the insured by $621. The study also used empirical estimates to show that the average individual in the private market was $442 better off, and a total market gain of $94 million, from the mandate. They also determined the socially optimal tax penalty for the uninsured, which would have been $2,190.
The results from the study are synonymous with a quote from an insurance textbook written at the Wharton School, which appears in George Akerlof’s The Market for “Lemons”: Quality Uncertainty and the Market Mechanism.
“Adverse selection appears (or is at least possible) whenever the individual or group insured has the freedom to buy or not to buy, to choose the amount or plan of insurance, and to persist or to discontinue as a policyholder” (Akerlof, 1970).
Under universal coverage, individuals still have the right to choose their plans (for the most part), but the requirements for mandatory coverage make sure they have to buy and keep their insurance, decreasing premium costs for the whole market and making everyone better off.
While the dissemination of information surrounding us has become broader each and every day, it is only logical that the insurance system breaks out of the antiquities of the past and encourages the availability of insurance in the market. From the models and examples shown, it is evident that adverse selection’s elimination should be prioritized, as welfare losses from its continuation will pile up, and employers will offer less and less generous plans as the death spiral makes coverage poorer and unaffordable for the masses. While long-term solutions are being negotiated, insurers need to make small changes that will incrementally increase welfare and reduce adverse selection’s stranglehold over the market. And as we seen have through the Massachusetts reform, mandated healthcare and a superior flow of information are the only ways to extinguish adverse selection once and for all, and provide a better world for those who need medical care, and want to relax while facing the other uncertainties of our modern world.

Works Cited
Akerlof, George A. "The Market for "Lemons": Quality Uncertainty and the Market Mechanism." The Quarterly Journal of Economics 84.3 (1970): 493.
Arrow, Kenneth J. "Uncertainty and the Welfare Economics of Medical Care." The American Economic Review 53.5 (1963): 946.
Cutler, D. M., and S. J. Reber. "Paying for Health Insurance: The Trade-Off between Competition and Adverse Selection." The Quarterly Journal of Economics 113.2 (1998): 461.
Cutler, David M., and Richard J. Zeckhauser. "Adverse Selection in Health Insurance." Forum for Health Economics & Policy 1.1 (1998): 3-5.
Einav, Liran, and Amy Finkelstein. "Selection in Insurance Markets: Theory and Empirics in Pictures." Journal of Economic Perspectives 25.1 (2011): 116-19.
Hackmann, Martin B., Jonathan T. Kolstad, and Amanda E. Kowalski. "Adverse Selection and an Individual Mandate: When Theory Meets Practice." NBER Working Paper No. 19149 (2013): 1-58.
Rothschild, Michael, and Joseph Stiglitz. "Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information." The Quarterly Journal of Economics 90.4 (1976): 638.