Saturday, August 18, 2012

Setting Voucher Levels or Setting Supplier Reimbursement Rates?

One issue in the Medicare debate that I have not seen anyone address concerns the political economy of setting voucher levels versus setting supplier reimbursement rates.

The Ryan plan would set a level of "premium support"-- think of it as a voucher -- which seniors would use to buy their health plan.  While the Ryan plan might have in mind a path for the voucher amount each year in the future, the actual level is of course going to be up to the Congress at the time.

Current Medicare sets thousands of individual prices at which hospitals and docs are reimbursed -- the notorious fee-for-service regime.  Each year, at least ostensibly, the Federal government, through the Centers for Medicaid and Medicare Services, determines all these prices.  Under ACA, it is true, there is some incentive to move away from fee for service to bundled payments -- payment for treating a disease condition over a period of time -- or even to capitation, whereby an entity such as an Accountable Care Organization will be paid for maintaining the health of a whole population.  Even in these cases, there will still be a lot of individual prices being determined.  Fee for service is not going to entirely disappear.

In order to say which regime will be less generous to Medicare beneficiaries, it is necessary to address the political economy of setting the different prices.  Will the political process really be able to hold the voucher level below the average cost of a senior buying a reasonable health plan on the open market? How does the political process deal with the setting of individual doctors' reimbursement rates?

I won't pretend to have done a full analysis of this.  But I think the visibility of the voucher and its sufficiency will be a key issue, and those who say that the voucher will be set at too-low levels need to think twice.  Seniors are a powerful political lobby.  Meanwhile on the other side, there is the invisibility of all the individual suppliers' prices, and the power of the American Medical Association.  The so-called "doc fix" where a previous cut in doctors' reimbursement rates has been put off year after year suggests the nature of the problems in setting individual payment rates.

The issue reminds me of  Stigler's paper "The Theory of Oligopoly."  Is it more conducive to collusion to having many small buyers or a few large buyers?  Stigler argued many small buyers is more conducive to collusion, as a seller will not risk defecting from a collusive agreement for just a small increase in sales:  "It follows that oligopolistic collusion will often be effective against small buyers even when it is ineffective against large buyers."



Some Honesty in the Medicare Debate, Finally

In the Washington Post, Ezra Klein has an interview of Rep. Chris van Hollen, who sits with Rep. Paul Ryan on the House Budget Committee.  The interview is here.  Klein does an admirable job of keeping the discussion on an even rational basis.

As one example, many opponents of the Ryan "premium support plan" claim that it cuts Medicare benefits, while the Affordable Care Act did not cut benefits.  See for instance Eugene Robinson in the WaPo who says this:

"The Affordable Care Act, otherwise known as Obamacare, slows the rate of growth of payments to Medicare service providers by more than $700 billion over a decade. But no impact is felt by seniors themselves, whose benefits and costs remain the same."

Sarah Kliff, in an otherwise very informative post, makes a statement that is very similar:  

"It’s worth noting that there’s one area these cuts don’t touch: Medicare benefits. The Affordable Care Act rolls back payment rates for hospitals and insurers. It does not, however, change the basket of benefits that patients have access to."


Now current Medicare does not pay anything directly to beneficiaries; the beneficiaries visit hospitals and doctors, who are paid directly by Medicare.  ACA does cut payments to doctors, hospitals, and some private insurers, and Kliff's otherwise fine post details quite well what those cuts are (they are very complex).  To say that ACA's cuts to doctors, hospitals, and private insurance companies (who provide supplemental Medicare insurance) are not cuts to benefits is really semantics.   Taken to an extreme, ACA could have cut doctors' reimbursement rates to Medicaid rates (extremely low, that is) and the ACA sympathizers would still be saying "but we didn't cut any benefits."  

Getting back to the Klein interview of Rep. van Hollen, there is this exchange.   I like Klein's question more than van Hollen's answer, which doesn't really address the question  (though he does have a valid point on what ACA attempts to do).

EK: Until now, I think that insofar as folks knew anything about the Medicare debate, they probably thought that Republicans had a plan to cut Medicare spending, but it was a bit cruel, and Democrats were completely unwilling to touch Medicare spending at all. I think we’re starting to get closer to the truth here, which is that both parties have very different visions for how to cut Medicare spending. But one thing the Democrats like to say is that they’re just cutting providers, not beneficiaries. But providers often pass their costs along to beneficiaries, either by making them pay more or giving them worse service. So how real is that distinction?

CVH: Obviously, if you were just to do across-the-board, arbitrary cuts, that would be the case, but the whole idea behind Obamacare is to change the incentive structure behind Medicare so the payments to providers focus on the value of care rather than the volume of care.
So, for example, before the Affordable Care Act was passed, hospitals would get reimbursed every time a patient was readmitted to a hospital even if they were readmitted continuously for the same underlying condition. Hospitals had no financial incentive to coordinate the care of the condition once the beneficiary left the hospital. We’re now changing the model so hospitals don’t get reimbursed every time the patient gets readmitted. Now they’ll get readmitted for managing that underlying condition. There’s also a major initiative underway to better coordinate care for dual-eligibles, people both on Medicare and Medicaid, who are a small portion of the population but a very high percentage of the costs. There are lots of misaligned incentives between the Medicare and the Medicaid program, and we’re working on them.

Sunday, August 12, 2012

Employment Effects of Medicaid Expansion

In thinking about the effects of ACA on employment, I have neglected the effect of the Medicaid expansion.

Overall there are a multitude of effects of ACA on the supply and demand for labor (that framework of course being my model of choice for the analysis).  On the demand side, we should expect that the employer mandate in the ACA -- provide acceptable insurance or pay a fee -- should decrease the demand for labor.  Offsetting this negative effect on demand is an increase in the supply of labor, reflecting the value that employees get from the mandated coverage.  As a first approximation, these two effects would perfectly cancel out, causing the money wage to fall and employment overall stay constant, but that is only a first approximation.  It should be expected that the value of the insurance to the employee is less than the cost to the employer, as many employers are observed to not offer insurance.  Also, there is the small problem that with very low wage employees, the money wage cannot fall so with demand being the binding constraint employment will fall.

A confounding effect here is caused by the subsidies offered to employees who buy insurance on state exchanges -- in some cases, the subsidy will be worth more to the employee than the fine that the employer will have to pay if their employees avail themselves of a subsidy (there are tax effects here too that I will ignore for now).

These are the effects that most analysts seem to take into account when looking at the effects of ACA on employer-sponsored insurance, see for  this Urban Institute article or this article by Holz-Eakin and Smith. 

But for the effect of ACA on employment, one more variable is important, and that is the expansion of Medicaid coverage.  Right now, of course, Medicaid coverage in many states is quite meager, with able-bodied adults often if not typically ineligible.  With the new Medicaid coverage, adults will be covered by Medicaid if income is below 133% of the Federal poverty level.  In comparing the decision to be unemployed pre-ACA, the cost of health insurance (or the cost of not having it) would have loomed large.  Post-ACA, the difference between the cost of insurance if unemployed is zero (Medicaid coverage) as it is if employed (subsidized or employer sponsored).  This seems like it would be a rather large impact on the supply of labor -- reducing it, as one of the major costs of being unemployed has fallen dramatically.