• A privatized, exchange-based Medicaid will not be cheaper than public Medicaid

    I’ve heard that there are folks who think that Arkansas’s plans to move the Medicaid expansion population onto the state’s ACA exchange will save money. Now, it is not completely out of the question that it could, on net, save the state money, though that claim itself would require careful calculation and explanation. But if some are claiming that it will save money overall, even to the federal government, which is on the hook for nearly all the cost, they’re wrong. It won’t happen. Here’s why:

    First of all, what motivates privatizing coverage for the expansion population? There are several possibilities. Let’s put aside the obviously political benefit of pumping more money to private enterprise. Another idea is that private plans will innovate in ways that the public Medicaid program cannot or does not. I have no argument with that. Yet another is that the exchange will offer choice, which, to a point, is good for consumers. Sure! And then there’s the idea that private plans offer better quality.

    Let’s pause here and dissect this. Do private plans cause better health outcomes than public Medicaid? You’ll find a lot of studies that show an association on this score, but not causation. Private plans enroll a healthier pool of patients (even controlling for observable diagnoses). Private plans serve a less vulnerable population in terms of socioeconomic status and community resources (even controlling for observable income, education, and the like). What you will not find is any study that shows outcomes are better for private plans than public Medicaid using a research design that any health economist would buy as valid for causal inference. I’ve asked the experts on this, and they agree. (Sorry, no name dropping. But think of the leading health economists who definitely know this literature and you’ve got the idea.)

    So, the notion that private plans cause better outcomes than public Medicaid is empirically unfounded. But, it is theoretically plausible. Why? Because private plans have more expansive networks than Medicaid, offering policyholders easier access to (presumably) higher quality services. And how, pray tell, do they conjure up these more expansive networks. THEY PAY PROVIDERS MORE!!! This is the very limitation that critics of Medicaid (myself included) raise. It pays providers too little, so too few participate. On the margin, perhaps those that do are of lower quality, but that is itself debatable. Still, less participation is less access, and that can harm outcomes.

    And that brings me to the final motivation for privatizing Medicaid, that it will lower overall costs due to competition. This is magical thinking, but not for the standard (and sometimes flawed) reasons people attack the concept of markets. Markets can and do reduce costs and improve quality, if they’re sufficiently competitive and in the right ways. But the ACA exchanges will not be competitive in the right way and certainly not for the Medicaid expansion population.

    The vast majority, about 80-85%, of insurers’ spending is on medical care. The customary ways to substantially reduce costs is to lower payment rates to providers or reduce the volume of care insured.* Either one of these reduces policyholders’ access to care. This, in large part, is why Medicaid is so cheap. Of course, policyholders won’t like this. Nobody likes to be told, “No.” The way they are compensated for less in the market is through lower prices. People are willing to get less if they pay less. Sometimes they prefer it, since they have other uses for their money.

    But wait, who’s paying the bills here? It’s not individuals, mostly. It’s the government. Most exchange enrollees and all of those in the Medicaid expansion population will buy highly subsidized coverage.** We know that this reduces sensitivity to price. It’s the very reason employer-sponsored plans are so generous and expensive. The government is paying a big subsidy through the tax exclusion of such plans.

    The same reasoning applies to exchange coverage. There is very little motivation by consumers to willingly trade less generous coverage for lower cost. Under such conditions, competition can’t drive costs down very far. Consumers won’t stand for it when they’re paying with someone else’s money. A plan that tries to go too far will likely lose market share and experience reduced profit. This is not a set-up in which I would expect private plans in competition to beat Medicaid on price. (It doesn’t happen in Medicare Part D either. As beneficial as competition has been in that market, it has not brought prices down to the level paid by Medicaid or the VA.)

    Of course, consumers don’t necessarily like Medicaid either. But Medicaid is not in competition with anyone. It can reduce generosity and have a sparse network without concern for market share or profit. That’s a perverse incentive too, but it is the very reason it is so cheap, and why it will remain cheaper than private, exchange-based plans. Count on it.

    * There is a third way: deliver care more efficiently. It is possible private plans could do this, but it would require big changes in the delivery system to close the gap between Medicaid and private costs. I’m deeply skeptical they could pull this off. It certainly can’t happen quickly. There is no precedent for it. But that doesn’t mean it’s impossible, eventually for private plans to induce tremendous efficiency. And, for the record, I’m not opposed to letting private plans try. But it will require an investment of more spending in the short run, and, if it doesn’t work (which is my bet), forever.

    ** Crucially, the complete premium and cost sharing subsidy scheme does not expose enrollees to the full marginal cost of care. One could design one that does, but this isn’t it.


    • Nice analysis. Medicaid pays providers too little, limits access, causes delays in diagnosis. Sometimes you get what you pay for, but not always. And cost is not always the issue in every case. For routine care, sure, cost is an issue. For complex care, it only costs a LOT more to go first class. Some docs are better than others. Some health care systems are better in quality and cost. than others. The exchanges will not have enough flexibility in coverage or cost, kind of like the way the VA system only has 2 annual reimbursement rates for veterans in each VISN.

      • “kind of like the way the VA system only has 2 annual reimbursement rates for veterans in each VISN.”

        I should know what this means, but I am not sure I do. Can you elaborate or provide a link?

    • I suspect some plans use a sparse network as a way to hold costs down in and of itself, by reducing utilization.

      I know of a doctor who tried to join our State’s high-risk-pool (very sparse) network, accepting the low reimbursement rates, and was refused (without explanation IIRC).

      I don’t have any sense of how common this might be or whether my theory of the motivation is correct.

    • Good analysis. I am going to comment on this phrase:

      “Thus, the only way to substantially reduce costs is to lower payment rates to providers or reduce the volume of care insured.”

      I don’t dispute this. However, we should note that some medically complex individuals see a variety of medical and sometimes non-medical providers (e.g. long-term care providers). The way our health system is set up, these folks don’t communicate well with each other. Therefore, for the most complex patients, it is possible that better coordination of care could reduce costs if done well. And a lot of the ACA focuses on better care coordination in Medicare and Medicaid, which will help.

      However, in my opinion, a lot of the public discourse on the Democrats’ side has focused on producing savings through changing the way care is delivered. Well, if you’re solely focusing on coordination of care, then the magnitude of the savings is not that great when considered in light of how much we spend in total. Intensive care coordination efforts would, in some ways, be a wasted expenditure on less complex patients (e.g. working adults with only minor medical conditions).

      Ultimately, if we want to reduce overall medical costs to society, it is indeed about either reducing the volume of care delivered, or paying less. Our medical providers make quite a bit more than they do in other OECD countries. Maybe they should not make as much as they do now and we should admit it.

    • You write “The vast majority, about 80-85%, of insurers’ spending is on medical care. Thus, the only way to substantially reduce costs is to lower payment rates to providers or reduce the volume of care insured. ”

      That is absolutely not so unless what you are trying to say means something completely different. One can reduce costs by using cost efficient technology, a bit more cognitive thinking before ordering testing, making sure patients don’t end up in the ER or hospitalized etc.

      None of the above has to involve lowered payments to providers or reducing the volume of care that is insured.

      • That’s true. But in that case, doesn’t it end up putting less money in the pockets of providers?

        • Not necessarily, at least not in the context of your comment. Even if one tries to stretch the meaning of your words to the breaking point it is again not necessarily so. There is so much overhead in healthcare that if one maintained total provider compensation at the same level healthcare could still be provided for less money if the incentives were appropriately aligned.

          That is one of the reasons I disagree with your over all conclusions despite the fact that I would list many of the same variables you listed though sometimes I might have a different spin.

          [Take note that lowering “payment rates” which is part of your quote is quite different from lowered total payments.]

          • So, I’m in 100% agreement that there is a way to reduce health care spending without cutting payment rates or reducing the quality of care. I still remain puzzled how it isn’t taking money out of someone’s pocket. Witness the recent pushback by Pioneer ACOs. It’s not about payment rates! (I will amend the post.)

            • But of course. If total outlays are reduced it has to come from someone’s pocket and that is the trick. Incentives have to be created to remove money from the right pockets and create an appropriate division of labor. That is something all of us do every day.

              The problem we face is that every system permits some to unjustly reward themselves. Thus we have become reactionary constantly trying to figure out who is getting too much and frequently our emphasis has been on the wrong things. That has caused costs to rise.

              Doctors make too much is a frequent mantra. I don’t know if they make too much or too little and that isn’t my primary concern. Let us say they make the right amount, but if they earned more total costs could fall without a change in quality or quantity. That would be a good thing. The opposite would also be true.

              If we look at things that way we open up a whole new bunch of possibilities for controlling healthcare costs. Presently almost all the energy is spent aimlessly cutting and that seems to work in the wrong direction. That is probably why Medicare costs keep rising even though the intention was that costs should fall. Perhaps the ones that are doing the cutting are doing so from too far a distance meeting political needs rather than the needs of our health system.

              With reference to your recent change on the blog (with much appreciation) and the * and **. There is a fourth way and perhaps many more. It seems we have restricted our methodologies focusing too much on the insurer and third party type payment.

    • so are you expecting managed medicaid patients will have more access, higher quality healthcare?

      seems to me that private health insurance plans have extended their market to adjoining government space …first with medicare through cutely names “medicare advantage” plans and now with medicaid programs….two enormous pools of covered lives beyond the employer market….

    • Y’see, we don’t want ACTUAL health “CARE,” we want an inscrutable myriad of health care PLANS.

    • “The health plans that compete for Arizona’s combined Medicare-Medicaid patients receive a fixed monthly fee. The companies are then expected to cover all of a patient’s needs.

      Critics have long argued the incentives for companies to keep the cash and withhold care are too great — the potential profits too tempting. And they point to the scandal-filled 1990s when some HMOs kept costs down by denying treatment.

      Today in Arizona, though, patients like Joseph Ford and advocacy groups say the health plan case managers are less like Grim Reapers and more like guardian angels. “Dave is cool,” Joseph tells me. “He’s in my cellphone.”

      “I appreciate somebody on the other side not only looking out for my well-being, and not just looking at me as a number on a paper,” he says.

      Ford is a number on a paper, of course. A big number. Nine million people nationwide are eligible for both Medicare and Medicaid, and their care costs some $300 billion a year.

      Arizona though, unlike other states, has long paid private health plans to manage care for public beneficiaries, and that has saved a lot of money. Arizonans in nursing homes cost the state $5,400 a month for custodial care. If kept in their homes, the cost is $1,400.”