The Curious, Counter-Intuitive Relationship Between Medicare Costs and HMO EnrollmentJeff Lemieux
February 8, 2004
In a weird way, Congress set the stage for future savings when it raised payments to Medicare HMOs in last year's Rx drug bill. A healthy rivalry between private health plans and Medicare's traditional fee-for-service program is probably the best way to ensure that Medicare spending never goes back to an unsustainable, business-as-usual trend.
Outline:
Medicare Outlays vs. HMO Enrollment
Political Consequences of the Mid-1990s Surge in HMO Enrollment
Do Medicare HMOs Get Healthier Enrollees?
Conclusion: Saving Medicare Private Plans Might Save Money Too
The biggest Medicare debate is whether or not it was ethical for the White House to suppress the Medicare actuaries' cost estimates of last year's Rx drug bill. Should we have been told that the Administration estimated the bill would cost $540 billion over ten years, not the $400 billion estimated by Congress?
That's an excellent question, but let's be honest. Everyone knew the Medicare bill was going to be very expensive. Ultimately, the cost could range from $200 billion to $800 billion, and much more in the following decade.
Of course, official cost estimates are extremely important, and all members of Congress deserve the full range of opinions and expert discussion.
But cost estimates are not always a good predictor of what will really happen to Medicare. Political forces can overwhelm the apparent impacts of changes in Medicare's payment formulas. And the conventional wisdom about how formula changes will affect Medicare costs is often flimsy.
Case in point: Medicare HMOs. For years, many health policy analysts assumed that expanding Medicare's alternative HMO program would cost the program money, especially if the HMOs attracted healthier-than-average enrollees.
However, studying the Medicare experience over the last decade points toward the opposite conclusion.
Medicare Outlays vs. HMO Enrollment. A simple picture of trends in Medicare spending vs. the program's HMO enrollment tells an unexpected story.
Growth in Medicare spending went down after a big surge in HMO enrollment in the mid-1990s. Likewise, Medicare spending growth accelerated in the late-1990s when HMO enrollment plummeted (see Figure 1).
Figure 1.
Wait a minute! We're not suggesting that increased HMO enrollment caused Medicare spending to fall, and that lower HMO enrollment caused Medicare spending to accelerate, are we?
Actually, yes. But in a very roundabout way.
There are two main points:
1. The surge in Medicare HMO enrollment in the mid-1990s caused a political upheaval that, in turn, dramatically reduced the growth in Medicare spending, at least for a while.
2. The old theory that Medicare HMO enrollment would siphon off the healthiest enrollees from the traditional fee-for-service program, and thereby increase Medicare costs, is almost certainly wrong.
Medicare's experience with alternative, full-service HMO plans has gone through several stages:
The quiet years, before 1995, when HMO enrollment in Medicare grew steadily, but mostly beneath the radar of angry doctors and hospitals (who don't like tough, HMO-negotiated payments), and opportunistic politicians (who love to rant about the restrictions HMOs place on health coverage).
The boom years, 1995-1998, when the growth of HMO enrollment in Medicare surged (and got everybody's attention).
The big bust, 1999-2003, when Congress ratcheted down payments to Medicare HMOs. After a few years, payments to HMOs fell to well below the amounts paid on behalf of Medicare beneficiaries in the traditional fee-for-service system. Of course, HMO enrollment subsequently tumbled, as HMO plans scaled back benefits, raised copayments and premiums, and withdrew from market areas.
The boom in Medicare HMO enrollment in the mid-1990s prompted members of Congress to launch a fundamental re-examination of the Medicare program; the viability of the traditional fee-for-service benefits plan was called into question.
However, when HMO enrollment started to fall in the late-1990s, Congress returned to business-as-usual in the Medicare program. Soon after, the growth in Medicare spending started to rise toward its earlier pace.
Political Consequences of the 1990s Surge in HMO Enrollment. Politicians took notice when Medicare beneficiaries started voting with their feet and enrolling in HMOs in the mid-1990s.
Prior to 1998, Medicare paid HMOs 95 percent of what they would have spent in claims had the HMO enrollees remained in the fee-for-service program.
However, many HMOs were able to take that payment and provide all of Medicare's regular benefits plus considerable "gap" coverage -- usually including drugs and low copayments -- without charging much in extra premiums.
Of course, the HMOs had coverage restrictions. Most required seniors to use a "closed panel" of doctors and hospitals specifically affiliated with the plan. The most successful HMOs were in urban areas, where Medicare fee-for-service payments were already high, and where the HMOs could bargain with a wide variety of hospitals and physician groups for favorable rates.
But the bottom line was straightforward: Unless the HMOs were enrolling much healthier-than-average seniors (which we'll get to below), it was obvious that they were much more efficient than the fee-for-service program during that period.
This caused elation among conservatives in Congress -- who preferred private health plans to the fee-for-service program anyway -- and consternation for liberals, who had touted the fee-for service program as a successful model for national health insurance.
Conservative members of Congress said, in effect, "Look, let's just forget about the outdated fee-for-service program and let all Medicare beneficiaries join HMOs or other private health plans, sort of in the way federal employees choose health coverage."
The first counter-reactions were within the Medicare bureaucracy, which was controlled by liberal Clinton Administration appointees. They started looking for ways to control costs in the fee-for-service program (lest it become uncompetitive with the private health plans).
Medicare administrators make dozens, even hundreds of small decisions each year that affect Medicare payments. When HMO enrollment surged and politicians started talking about how inefficient the fee-for-service program was, Medicare administrators started saving money by saying "no" to requests by health care providers for marginal regulatory changes that would have raised their fees (and Medicare's costs). They also reorganized the Medicare bureaucracy in a crude attempt to keep HMOs from getting sympathetic regulations.
Congress followed up the increasingly strict regulations with tougher laws. In 1996, Congress passed significant anti-fraud provisions designed to prevent health providers from taking advantage of the fee-for-service program's lax rules.
Finally, the Balanced Budget Act of 1997 slashed fees to hospitals, physicians and HMOs alike. This caused a reduction in Medicare spending across the board, but HMO payments were reduced disproportionately. Congress felt that if HMOs were able to offer such attractive benefits that enrollees were signing up in droves, then the payments they were receiving from Medicare must be much too high. By 1999, HMO payments were falling well below 95 percent of average fee-for-service costs.
In sum, the disparity in the efficiency of HMOs and the fee-for-service plan caused the Administration and Congress to clamp down the on fee-for-service program (because it was inefficient and increasingly uncompetitive) and on HMOs (because they were presumably getting too much).
Conversely, when the post-1997 payment cuts took effect and HMO enrollment started to plunge, the pressure to restrain fee-for-service costs was relaxed. A declining HMO program was no longer a threat to the fee-for-service program's viability and political popularity.
To be fair, the mid-1990s were generally a period of fiscal austerity, when federal outlays for most programs were carefully monitored, and often cut.
However, the surge of enrollment in HMOs unquestionably played a large role in the subsequent crackdown on Medicare spending.
Do Medicare HMOs Get Healthier Enrollees? Throughout the 1990s, health policy analysts generally asserted that Medicare HMOs attracted healthier-than-average enrollees. Therefore, they said that Medicare shouldn't pay the HMOs as much as it spent on fee-for-service enrollees (who were presumably sicker).
Figure 1 doesn't disprove the "favorable selection" hypothesis by itself, but is raises some pretty serious questions.
Why did overall Medicare spending fall after the surge of enrollment in HMOs in the mid-1990s? In general, moving supposedly healthier seniors (who would otherwise cost Medicare much less than the average enrollee) into capitated health plans, where the capitation rate is 95 percent of average costs, should increase Medicare spending.
For example, suppose the average claims cost of fee-for-service Medicare beneficiaries is $7,000 a year. Now, assume a group of healthy Medicare beneficiaries whose claims costs were only 50 percent of average ($3,500) moved from the fee-for-service program to an HMO. If Medicare payments to the HMO were 95 percent of average fee-for-service costs ($6,650), Medicare spending would increase by a $3,150 for each beneficiary who switched ($6,650 minus $3,500).
However, Medicare spending fell after lots of seniors joined HMOs in the mid-1990s. This is the opposite of what the "favorable selection" theory would imply.
To be sure, there were many other things happening in Medicare at that time -- anti-fraud efforts, regulatory stiffening, and legislated reductions in Medicare payments. These things certainly were the main reason Medicare spending fell.
But for the favorable selection theory to have been accurate, those payment reductions would have had to be doubly deep, because continued in-migration to HMOs was supposed to be raising Medicare spending during that time.
Likewise, after 1999, when HMO enrollment started to fall, the favorable selection theory said that Medicare spending would decelerate, as healthier HMO enrollees returned to fee-for-service, where Medicare would spend less on their behalf. However, Medicare spending started to rise just as HMO enrollment began to fall.
What probably happened was this. Any favorable selection in Medicare HMOs that may have occurred in the early 1990s (and was the original source of the theory) was gone by the mid- and late-1990s. By the middle of the decade, Medicare enrollees in HMOs were probably no more or less healthy than comparable fee-for-service enrollees.
In fact, by the late-1990s, Medicare's HMO enrollment might have become less healthy than the comparable fee-for-service enrollment, as HMOs pulled back their service areas to inner cities with a disproportionate minority population (and widespread health problems).
Importantly, neither the Congressional Budget Office (CBO) nor the Medicare actuaries assume that private plans get favorable selection.
Conclusion: Saving Medicare Private Plans Might Save Money Too. Medicare payments to HMOs were increased to 100 percent of average fee-for-service costs as part of the 2003 Rx drug bill. This should cause HMOs to re-join the program. The bill also encourages other private health plans like preferred provider organizations (PPOs) -- which have "out-of-network" coverage options in addition to their regular panels -- to offer Medicare coverage.
Indeed, a recent survey trumpeted by the trade association for private health plans predicted significant premium reductions and service expansions following the increase in rates.
The question is, will the re-invigorated private plan options also begin to put downward pressure on Medicare's overall spending?
In the short run, the answer is "no." Higher Medicare payments to private plans will cost the program more.
But over the next 5 years, look for a new round of cost cutting in Medicare's fee-for-service program as private plan enrollment rises.
Hopefully, this next round of Medicare cuts will be more forward-looking than the 1997 cuts. Medicare's fee-for-service program needs to be made more like a private plan in many respects, with greater region-to-region accountability, more flexibility for administrators to made decisions that could save money, and a better ability to coordinate the care of seniors with expensive long-term or chronic illnesses.
A viable private plan alternative in Medicare will force the fee-for-service program to shape up, even if the two sectors are not placed into direct competition.
On the other hand, a stronger, more efficient, fee-for-service program will prompt private plans to continue looking for ways to improve coverage and reduce costs.
Health policy analysts often look at Medicare's rules and payment formulas in a vacuum. Certainly cost estimators are not allowed to make sweeping assumptions about what future Congresses or regulators will do.
But political and regulatory directions or tendencies will be very important to future Medicare spending.
A healthy rivalry between private health plans and Medicare's traditional fee-for-service program is probably the best way to ensure that Medicare spending never goes back to an unsustainable, business-as-usual trend.
Links:
Centrists.Org Explaining Premium Support: How Medicare Reform Could Work
(revised November 6, 2003)
Centrists.Org Testimony: Improving Chronic Care in Medicare (November 4, 2003)
Progressive Policy Institute McDonald's vs. Burger King: A "Nothing Burger" Debate on Medicare Reform (May 6, 2003)
Centrists.Org Issue Summary: Health (Basics)
Centrists.Org Issue Summary: Health Costs, Competition, and Chronic Care
Centrists.Org Issue Summary: Medicare Reform and Prescription Drugs
