Sunday, January 24, 2010

More On Health Savings Accounts

I'm still infatuated with the idea of establishing a Health Savings Account for every person or family in the the US. Here's a little more meat on the bones:

First, an HSA is a way of setting aside funds exclusively to be used for purchasing medical care or insurance. Every independent person in the country would get one, but spouses could share one. Children and other dependents share one with a parent or guardian, unless they're wards of the state, which would set up and manage one for each ward.

Money in an HSA is sticky: It can accumulate over the lifetime of the account holder. It can be invested, and any investment gains go back into the HSA tax-free. (Limitations on the kinds of investments seem like a reasonable restriction.) But you can never get the money out for anything but health care and, if you have no dependents or a surviving spouse, any remaining money goes back to the government when you die. (No doubt there are all sorts of fancy things that need to be done to divide the proceeds between the feds and the states--left as an exercise for the reader.) By sending the excess back, we avoid the temptation to use the HSA as a way to park tax-free assets for one's heirs.

A big problem with such an HSA is the possibility of fraudulently using the account to pay for non-medical stuff. To avoid this, HSAs would be managed by an HSA clearinghouse. The clearinghouse is responsible for ensuring that all payments go to qualified medical providers or insurers. I suspect that this qualification process would involve applying to the government for certification that your goods and services were qualified, and the clearinghouse validating some kind of a qualification number.

The way that employer-based insurance is administered changes, but in a way guaranteed to minimize the impact to employees that would just as soon leave well enough alone. Rather than paying tax-deductible premiums on behalf of the employee, the employer dumps the (still tax-deductible) money into the employee's HSA. If the employee wants to use the employer's group plan, it goes right back out to pay the premiums. However, employees can also opt to buy some kind of third-party insurance. (Enabling legislation to create non-employer "super-group" plans would be handy here, but isn't essential.)

Here's a picture of the relationship between sources of HSA funds and legitimate forms of transfer or payments:

Note a couple of things: Parents or guardians can transfer a part of their HSA to their children to set them up as they become independent. Also note that clearinghouses get paid some regulated sum for their services managing and disbursing funds.

So why is this scheme useful? On the face of it, it does almost nothing. In fact, in arguably introduces a certain amount of inefficiency in the system. Instead of individuals paying out-of-pocket expenses directly, they have to get the clearinghouse to do it. This inefficiency can be mitigated through appropriate automation. I'd think that a credit-card like system, similar to those used by some FSA plans, would work nicely.

But in the long run, such an HSA forces every individual to be completely responsible for all health care expenditures. The clearinghouse can tell you at any point exactly what you're getting in contributions and what you're paying to insurance and/or out-of-pocket. The system is completely transparent. Only when transparency is achieved will individuals be responsible for their payments. Indeed, you can even put Medicaid and other forms of public assistance into the HSA, so even low-income people understand the costs associated with their care.

Note that this is not incredibly far off of the Wyden-Bennett plan, although the clearinghouse, only with the consent of the consumer, becomes the payer, rather than the government.

Transparency begets responsibility, and responsibility begets cost control. This is a big procedural change to how health care is administered in the US, but the actual cost to implement it is relatively modest.

No comments: