Reconstructing the Individual Mandate as an Escrow Account

Gregg D. Polsky


The recent health care reform law's most controversial provision is the individual mandate, which imposes a fine on individuals who fail to obtain a minimum level of health insurance coverage. Many object to this policy, arguing that the government shouldn't force individuals to purchase health insurance. Others believe that the mandate is a necessary component to health care reform. What has been missed in the discussion is that Congress could restructure the individual mandate to avoid the requirement that individuals purchase health insurance while still fulfilling its principal function.

The principal purpose of the mandate is not to require individuals to purchase health insurance. Rather it is to prevent individuals from gaming a system that prevents insurers from excluding people with preexisting conditions and from fully differentiating prices for those with different levels of health. Proponents of the health care law do not care very much whether someone chooses to insure against future health costs or to self-insure against those future costs. But they do worry that individuals who decide to self-insure will ultimately renege on this decision when they require significant medical care. Self-insurers can renege either by not paying for medical services they receive or by buying into the insurance system after they have become sick.

I. The Proposal

Once it becomes clear that the primary purpose of the mandate is not to require individuals to purchase insurance but rather to protect against the failure of self-insurers to fully perform, a solution emerges. In many contexts when parties make important promises on which others rely, they are required to fund an escrow account or post a surety bond to ensure that they perform. If those parties fail to perform, these security devices compensate those who are harmed by the failure. Likewise, Congress could restructure the individual mandate as an escrow account. If an individual decided to self-insure against her future medical costs, then this proposed law would require her to make annual escrow contributions in the same amount as the penalty under the current mandate.

Once the individual fully performed, the federal government would return the contributed amounts, with interest at the market rate for lending money to the government. Ordinarily, full performance would be assured once the individual reached Medicare age. At that point, if the individual had remained out of the insurance pool and owed no outstanding medical bills, then the government would return her money. If an individual died before reaching Medicare age, then her estate would get the money back, so long as she had fully performed.

Under this proposal, the government would use the money contributed to the escrow account in the same manner that the individual mandate revenue is to be used. The right of the account holder to redemption upon full performance would therefore constitute an unfunded contingent obligation of the government to repay the contributions plus interest.

By reconstructing the mandate as an escrow account, there would no longer be a financial penalty for self-insuring future medical costs. Because a performing self-insurer would receive the market rate of interest for lending to the government, she would be no worse off. While it is true that a self-insurer would effectively be forced to lend money to the government, the self-insurer could easily adjust a balanced investment portfolio to account for this by buying fewer U.S. debt securities. Some might object that those who lack sufficient liquidity would be unable to make the required escrow contributions. But those lacking sufficient liquidity to make the contributions would surely be unable to effectively self-insure, because self-insuring requires significant liquidity to pay for any medical emergencies that may arise. Furthermore, the health care reform law provides people with significant liquidity issues (i.e., poor and lower-middle class people) free insurance coverage or substantial subsidies to purchase insurance, thereby reducing the likelihood that they would elect to self-insure in the first place.

Restructuring the mandate as an escrow account should be palatable to legislators who object to the government forcing people to buy health insurance, but who favor opening up the insurance system to those with preexisting conditions. The escrow structure would allow people to choose to self-fund without penalty, while providing protection against those who try to game the system. Furthermore, reconstructing the mandate as an escrow account would be relatively simple to implement. Everything in the current health care law could remain the same, except that fully performing self-insurers would now be entitled to a refund of their penalties with interest.

II. Possible Objections

A. Too Many Healthy People Would Opt Out of Health Insurance

By reconstructing the penalty as an escrow account, this proposal would appear to substantially reduce the cost of self-insuring. If it did, larger numbers of healthy people might choose to opt out of the insurance system, which would diminish the average health level of the insurance pool, requiring an increase in premiums to provide the same level of coverage.

On first glance, it might appear that the proposal would substantially reduce the cost of choosing to self-insure. While a penalty is a sunk cost, the cost of funding the escrow account would be negligible. The only apparent cost would be that a self-insurer might be required to over-invest (relative to her preferences) in government debt; but as mentioned earlier, even this cost would be lower than it might appear because a self-insurer could adjust her portfolio to account for this forced investment. If this were in fact the only cost, then self-insuring would become significantly more attractive than it is  under the current penalty regime.

But the cost of choosing to self-insure under the escrow proposal is actually significantly larger than it might appear. Recall that to receive the escrow payout, the self-insurer must remain out of the insurance pool until the relevant date (i.e., either upon reaching Medicare age or upon dying, whichever happens first). Thus, only people who have a very high degree of confidence that they would want to remain out of the insurance pool until the relevant date might be influenced by the escrow proposal. Younger people would have difficulty predicting both their future medical needs and the evolution of the health care market over several decades, so they would typically not be motivated by the proposal to opt out of the insurance market. Older people, who face a much shorter "prediction horizon," would typically not choose to opt out because buying insurance (in the reformed insurance market) would generally be a good deal for them after the new law is implemented. For example, the new health care law will prohibit insurers from price discrimination based on preexisting conditions, which older people are more likely to have. Additionally, the law prohibits insurance companies from charging older people premiums that are higher than three times the amount charged to younger people; this aspect of the law does not fully reflect the additional health care costs of older people. In short, it is likely that relatively few people would be motivated to opt out of the insurance market as a result of reconstructing the individual mandate as an escrow account.

B. It's Too Complex

Turning the mandate into an escrow account would add some administrative complexity because the government would have to issue refunds when self-insurers fully performed. But this marginal administrative burden would be minimal because, as discussed above, the number of self-insurers who actually performed through the relevant date would likely be very low. Where self-insurers failed to perform fully, the system would operate just like the current penalty regime, because the government would not be obligated to repay the funds it had collected. Only in those cases where self-insurers fully performed would refunds need to be processed.

In addition, refund processing in these very few cases would not be especially onerous. When a self-insurer reached Medicare age, or if she died before that date, in order to receive a refund, she (or her estate) would need to show that (i) she owed no outstanding medical bills, (ii) she had paid in full all of her previous medical bills, and (iii) she had never obtained insurance coverage. The self-insurer would attest under penalties of perjury that she satisfied these requirements. The government could then substantiate these facts by making inquiries to credit reporting agencies and insurers. If the facts were verified, the government would issue a refund with interest, just like it does when taxpayers file claims for refunds for overpaid federal taxes.

C. It Does Nearly Nothing

The responses to the first two objections have essentially been that the proposal should operate, as a practical matter, almost exactly like the current individual mandate system. This similarity means that nearly the same number of people would opt into the insurance pool and that the marginal administrative burden of the proposed reform would be low. Given the similarity, one might object to the proposal on the ground that it appears to do nothing or, at most, very little.

Though the practical significance of the proposal would likely be minimal, it still does serve an important purpose. This proposed reform clarifies that the government's objective is not to require the purchase of health insurance but rather to open up the insurance market while preventing strategic behavior that undermines this goal. Under the proposal, the government would not punish those who wish to self-insure, but would prevent individuals from selectively participating in the insurance market only when they are ill. The fact that few would actually choose to self-insure under the proposal does not mean that the proposal would not be worthwhile.

D. It Does Not Solve the Constitutional Problem

Whether restructuring the mandate as an escrow account bolsters the argument that the health care law is constitutional is unclear. Those who feel that the Constitution forbids the federal government from penalizing the nonpurchase of health insurance might conclude that requiring an escrow contribution is likewise prohibited. Other skeptics of the current mandate may be swayed, believing that the absence of a financial punishment (in present value terms) is constitutionally significant.

Regardless, the argument made here is a policy argument, not a constitutional one. It is targeted at policymakers who like the idea of opening up the insurance market but are offended by the notion that the government is requiring the purchase of health insurance. A pure constitutional solution (if one is necessary) is easy to envision, but it would not in any way address the concerns of those who object to the mandate on normative grounds. For example, the mandate could easily be restructured as a tax credit for purchasing insurance. This could be accomplished by increasing the income tax liability of everyone through a 100 percent tax on the first specified amount of taxable income, and then giving a tax credit in the same amount to everyone who purchases insurance. While this sort of restructuring would solve the constitutional problem, it would not ameliorate the concerns of those who object to the mandate on the ground that the government should not require the purchase of a health insurance plan. Reconstructing the mandate as an escrow account, however, is responsive to this concern.


Turning the mandate into an escrow account would remove the financial penalty that would be imposed on self-insurers under the health care law. At the same time, it would protect the insurance system from the gamesmanship that would take place without a mandate or security device. Furthermore, changing the mandate to an escrow account would technically be very easy, as the practical effect of the change would be minimal. The entire health care law would remain the same, except that fully performing self-insurers would be entitled to a refund of their "penalties" with interest.

Gregg D. Polsky is the Willie Person Mangum Professor of Law at the University of North Carolina School of Law. He would like to thank Al Brophy, John Coyle, Brant Hellwig, Joan Krause, Andrew Lund, and Richard Saver for helpful conversations about this idea.

   //  VIEW PDF
Volume Archive
Monthly Archive
Tagged Posts

& Other Current Events

Crawford v. Washington: A Ten Year Retrospective

No one disputes the significance of Crawford v. Washington, 541 U.S. 36 (2004), which fundamentally transformed Confrontation...

Come Back to the Boat, Justice Breyer!

I want to get Justice Breyer back on the right side of Confrontation Clause issues. In 1999, in Lilly...

Crawford v. Washington: The Next Ten Years

Imagine a world . . . in which the Supreme Court got it right the first time. That is,...

The Crawford Debacle

First a toast-to my colleague Jeff Fisher and his Crawford compatriot, Richard Friedman, on the...

Confrontation and the Re-Privatization of Domestic Violence

When the Supreme Court transformed the right of confrontation in Crawford v. Washington, the prosecution...