Why Obama’s Penalty for Not Having Health Insurance Is Not a Good Pigouvian Tax

The main defense being put forth in favor of the proposed penalty on those without health insurance is that these people should pay for the expected damage that they impose on others in society by not being insured. In other words, there is a Pigouvian justification for taxing the uninsured.

The problem with using this defense for the specific proposals that have been put forth to penalize the uninsured is that the proposed penalty is given as follows:

Penalty for Uninsured = max[((AGI less filing threshold) * (“tax” rate) * (1 if uninsured, 0 otherwise)), maximum penalty]

I don’t think we can consider this even close to a good Pigouvian tax because an uninsured person’s external cost imposed on society is not a positive function of the individual’s income. In fact, if anything, the person’s income being higher would likely decrease his/her external costs imposed on others because there is a greater likelihood that the person could actually end up paying the medical bills. Having a tax bill that is a function of the reciprocal of income would probably be closer to a true Pigouvian tax. A head tax on the uninsured would almost assuredly be more Pigouvian than relying on income as the base.

There is the adverse selection justification put forth, which says that those who select to be uninsured are typically healthier. Unless health is highly correlated with income in this income range and thereby income is a predictor of adverse selection (i.e. healthy people choosing not to be insured), I don’t see much of a justification for using income as a base as opposed to a flat amount per person.