Code Red: Two Economists Examine the U.S. Healthcare System

November 11, 2013

This Fairy Tale is too Grim

Filed under: Uncategorized — David Dranove and (from Oct. 11, 2013) Craig Garthwaite @ 9:15 am

The Obama administration announced on Friday that it will require parity for mental health insurance coverage. That means that health insurers must apply the same copayments, deductibles, and visit limits to mental healthcare as they do for physical health care treatment. Call it fair, call it political, but please don’t call it a good economic or health policy.

The story about how this is fair, or at least politically popular goes something like this: Health insurers are evil and powerful firms that can and will do whatever they want. On the other hand, patients with mental health problems are politically weak and must be protected from the powerful insurers that have no interest in taking care of them. In this story, the Obama administration rides in on its white stallion and rights the wrongs being perpetrated by the villainous insurance companies. All we need is a damsel in distress, an evil step-mother, and a catchy tune and Disney will sign the movie rights.

The problem with this simplistic story line is that you can replace “mental health” with nearly any other condition and the story would sound just as plausible. There is nothing particular to mental health about the story above. But this raises a question: why haven’t these greedy and powerful health insurers already instituted parity, in the form of high copayments, high deductibles, and strict visit limits for all services? The fairness/political story breaks down because it cannot explain why insurers “single out” mental health; it is actually hard to stitch together an explanation where insurers as a group have a grudge against individuals with mental health problems.

There is an alternative explanation for the lack of parity, one rooted in the dismal science. Basic economic theory offers a story that is less depressing, but perhaps more complicated, than the popular rhetoric. In this story, insurers are trying to do what is best for their entire enrolled population, saving money only when it makes sense to do so. Keep in mind, insurance companies are not setting policies in a vacuum – they are constantly competing with other companies for new clients.

It is important to start with the question of why we have copayments, deductibles and other restrictions on care. You could buy into the fairy tale that these measures are just attempts by insurance companies to deny care to people with a particular focus on individuals with conditions that they consider as unsavory or non-existent. Alternatively, you could consider an explanation in which these measures are attempts to rein in unnecessary spending by judiciously exposing patients to some amount of the marginal cost of care. Properly designed, such health plans should encourage more efficient consumption of health care services without exposing enrollees to excessive financial risk.

Given this basic theoretical understanding we can move forward. Intuition, formal theory, and mountains of empirical evidence indicate that when you make health care services free, patients purchase more of them. Sometimes the patients demand more services. Other times their doctors push them to receive more services and the insured patients have little reason to question the recommendations.

So far so good, but this still doesn’t explain why insurers single out mental health care. Here is the important economic reason why we might see different cost sharing for these treatments. Economic theory predicts that patients (and their doctors) will react differently to their exposure to the marginal cost of different types of treatments. Unsurprisingly, patients are not very price sensitive for more essential services such as inpatient hospital care and emergency care. However, they are far more sensitive to the price they face for services where the need is harder to define and the benefits of treatment are harder to prove. In these cases, cost sharing is an effective way to limit unnecessary care and maintain a balance between benefits and costs. The economic evidence for this latter is overwhelming, and is clearly spelled out in textbooks and research papers written by economists from across the political spectrum. This is a settled issue.

Mental health care is an exceptionally good example of this type of treatment. Definitive tests for many conditions, such as depression and ADHD, do not exist. In addition, existing treatments for these conditions have a mixed track record. Without cost sharing, visit caps, or other attempts to force patients and doctors to confront the marginal costs of their decisions, mental health treatments can become open ended with many successful treatments but also many failures. It follows from economic theory that cost sharing and visit restrictions for mental health should be tighter than for other services. We suppose that one might argue about the degree of restrictions for all services, but there is no theoretical argument to suggest that parity is preferred to discrimination.

We should be very clear. We are not denying the benefits of many mental health treatments nor are we trivializing the severity of these conditions. We both firmly believe that these conditions can often be addressed by both counseling and, if appropriate, pharmaceutical treatment. In fact, advancements in the treatment of these conditions are a real testament to the benefits of the American medical system. And we also believe that it is appropriate for insurance to cover some of the associated treatment costs.

However, we should also be able to recognize and rationally discuss the reality of the situation. Mental health services are far more difficult to empirically diagnose, the efficacy of the treatment is harder to measure, and as a result we should think it is not unreasonable that these services receive less generous coverage in an efficiently designed insurance plan. In this debate, you can opt for fairness, or you can opt for optimally balancing benefits and costs, but you cannot have both. Let’s not demonize insurers for pursuing that balance.

But this is really about a lot more than just the benefits for mental health coverage. This is yet another example of how the Affordable Care Act stifles innovation in benefit design for insurance plans. In addition to mandating a rather rich set of minimum benefits, federal regulators are now determining which services can receive differential cost sharing. It is hard to imagine this doesn’t result in even more health spending in subsequent years. And we thought the ACA was supposed to reduce health spending. Silly us.

Given the events of the last few weeks, we almost think that some of those charged with implementing the ACA want it to fail. “We tried market-based health reforms and they failed,” they might say, concluding that “the only sensible way to control costs is with a single-payer system.” That story is almost more believable than the evil insurance company fairy tale sketched out above.

We did say “almost.”

3 Comments

  1. This explanation neglects the possibility that insurers may offer inefficient contracts in competition due to incentives for service-level selection. Frank et al. (JHE 2000) and Ellis and McGuire (JHE 2007) point out that profit-maximizing insurers have incentives to heavily ration services that are “predictable and predictive.” Mental health services are usually at the top of the list when it comes to predictable and predictive services. Because of this, it may actually improve efficiency to mandate mental health parity. In the end it’s an empirical question. That said, it would likely be much more efficient to address this problem via adequate risk adjustment rather than through heavy-handed mandates.

    Comment by Tim Layton — November 11, 2013 @ 10:43 am

    • I agree with the theory but think it likely to be a second order effect compared to the simple and large effect of moral hazard. BTW, I am not sure about the claim that mental health is predictable and predictive and I wonder what empirical evidence there is to back up the claim that strategic considerations would lead to lower equilibrium coverage.

      Comment by David Dranove and (from Oct. 11, 2013) Craig Garthwaite — November 11, 2013 @ 11:14 am

  2. I will give you another. Political constituency.

    Why does autism therapy get covered both in schools and by many states? Powerful rich parents.

    In the aggregate, disenfranchised individuals who: 1) may wish to keep their diagnosis quiet (another key), 2) who may not self-advocate because they lack the wherewithal , and 3) dont vote, will not advance their collective cause well.

    Just something to consider.

    Brad

    Comment by Brad F — November 11, 2013 @ 8:36 pm


RSS feed for comments on this post.

The Rubric Theme. Blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.

Join 513 other followers

%d bloggers like this: