President Faust, Provost Garber, Dean Smith, Colleagues
Jerry Green, Economics Department
My work is in microeconomic theory. Among other things, I study risk, insurance, and incentives. I study behavior, rational and irrational, and its implications for policy.
I will confine my remarks to explanatory note #2, which addresses the new co-insurance payments – 10% of the cost of many tests and procedures, up to an out-of-pocket maximum.
Co-insurance imposes a significant new financial risk. It is all the more harmful because the people bearing the expense are those who require a significant medical service.
I believe that a much better idea would be to eliminate co-insurance entirely, raising premiums instead, so as to keep constant the average employee’s contribution toward medical insurance. By design, in expectation, Harvard’s expenses would not change as a result.
Co-insurance has been studied extensively by health economists. Each implementation of co-insurance is different and much depends on the details in a plan’s designs. Because Harvard’s new co-insurance provision applies only to hospitalization, surgery and advanced diagnostic testing, its effect on the utilization of medical services is hard to predict. I will argue, however, that whatever one believes about utilization in the future, the results of the new Harvard plans will be both financially and medically undesirable.
If there is no change in utilization, then co-insurance will not reduce Harvard’s aggregate health benefits expenditure. Yet the uncertain magnitude of the co-insurance requirement puts patients and their families under financial pressure at exactly the wrong time.
If co-insurance does decrease utilization, which seems to be one goal of this policy, I believe that matters will be even worse. By deferring or avoiding medical care or diagnostic tests some employees, or their family members, will later experience serious illnesses or complications. Viewed from any perspective longer than the single year in which the initial decision to forego care was made, medical expenses will be higher, not lower. Thus, if utilization does decrease, we will have both inferior outcomes and higher costs. Co-insurance is a lose-lose proposition.
My colleague, David Cutler, in his masterful book “Your Money or Your Life” has documented that the keys to improving health outcomes in any population are: regular follow ups, adhering to “doctor’s orders”, timely diagnostic tests, early interventions, and active management of chronic conditions. These are precisely the actions that might be postponed or avoided by an employee facing the prospect of co-insurance payments.
Everything that we have learned in recent years, in economics and the other social sciences, tells us that people do not choose wisely, even in very important matters. Start with wishful thinking, add an ounce of procrastination, stir in the anxieties due to illness, and you have a recipe for poor medical decision making by the patient. Add a dash of co-insurance, pour over financial stringency, and this potent cocktail will become dangerous, perhaps lethal.
A few years ago, as everyone will recall, our retirement plans were simplified and investment choices were restricted. The rationale for these changes was beautifully explained on the floor of this meeting by my colleague David Laibson. Citing the same body of academic research I have mentioned above, Professor Laibson confirmed that people – even highly educated and intelligent people -- are not good judges of their own situations. We all are subject to irrationality: over-valuing the present relative to the future, inertia, cognitive biases, and especially over-optimism.
If that is true in the financial realm it is doubly true in medicine. Medical decisions are more complicated, more uncertain, and more emotional. They are frequently made in times of stress, making us even more likely to err.
Harvard acted wisely when it recognized the adverse effects of human psychology on retirement planning. It should now act wisely again. The administration should recognize that co-insurance creates unfair, unnecessary, random transfers of wealth, falling on exactly the wrong subset of our population. It will not reduce the long term cost of medical care, and will result in some avoidable, perhaps tragic, outcomes.
November 4, 2014
FAS Faculty Meeting