On Competition

It's been a while since I last used the Speakeasy to muse. But I've got competition on the mind.

One of the axioms I laid out in Single-Pay It Forward was "If well-functioning markets are possible, they are preferable to the alternative." I still believe that, if only because all things in life necessarily involve trade-offs and the most ethical seat of decision-making between those alternatives is with the individual impacted by them. But as always I'm mindful of the realities of bounded rationality and the need for the sorts of Nudge-style decision-making aids I talked about in one of my favorite posts, Order Out of Chaos.

If we could have markets in health care, what would that mean? Broadly speaking, I think there are three possibilities: one centered on active competition between insurers and two more centered on direct competition between health care providers (e.g., doctors and hospitals). The options as I see them are:

Option 1: Trust in the insurers

The world in which we live now is largely one of so-called managed care competition. That is to say, health insurers negotiate contracts with health care providers and through those negotiations they theoretically fight to constrain rising provider prices and encourage cost control on the part of the providers. Meanwhile, on the consumer side they attempt to avoid unnecessary spending by keeping an eye on the services their enrollees use, either by relying on the patient's primary care provider to serve as a sort of gatekeeper to the rest of the health care system or by requiring patients and their providers to receive direct authorization before they undergo any major procedures.

The appeal of the notion that contractual relationships between private actors jockeying for position relative to competitors provided the needed dose of market dynamics in health care is part of the impetus behind the decline of active price-setting for hospitals two and a half decades ago. And it still has an appeal.

The insurer designs a range of products, all with different expectations of the buyer's responsibility at the point of care, different provider networks, perhaps innovative benefit designs and ties it up with a neat bow and a single all-encompassing price (the monthly premium). The trade-offs are there for inspection and the sum total of their impact is represented in the premium of the product. An insurer that does a better job of negotiating competitive prices with providers, designing payment incentives for providers that encourage and enable cost control, crafting high-value networks and smart products that constrain spending growth can offer lower premiums. And the insurer that offers the lowest premium is the most attractive in the open market.

Why it hasn't worked and how it could. What's the catch? Well, other than the very small marketplaces created by the Affordable Care Act for people who don't get insurance through their employer, a competitive consumer market doesn't exist in health insurance. The vast majority of people (~170 million out of the ~190 million privately insured people in the country) get insurance through their job. That means their employer either buys an insurance product for them or, as is the case for most of those people, their employer is their insurer, directly setting aside money to pay medical claims and only buying administrative services, not an insurance product, from health insurers. That severely dampens the incentive for offering the most cost-effective product; as Milton Friedman noted--correctly, I believe!--about health insurance, "nobody spends somebody else’s money as wisely or as frugally as he spends his own."

The experience with ACA marketplaces bears this out. Those marketplaces have proven to be remarkable effective at fostering competition between insurers, in no small part because consumers are incredibly price conscious when spending their own money and making their own decisions. In those circumstances it really is most valuable to be the most cost effective, low premium option in the marketplace. This is true even though the overwhelming majority of shoppers in the marketplaces receive premium subsidies because those subsidies preserve the relative prices/premiums of those competing products.

So I'm coming to believe that if competition between insurers is going to save the whole system we need more people to be in direct consumer markets for insurance. That way insurers really are incentivized to offer the most cost-effective and desirable insurance products available. And I think that means people need to stop getting insurance through their employer. Or at the very least they need to be taking an employer subsidy and applying it to a plan of their choice in an open market, similar to the SHOP exchange concept in the ACA that's regrettably never really found its footing.

Option 2: Sideline the insurers

The next possibility is that we don't ask so much of insurers after all. One of the hallmarks of managed care competition is that there are no real prices among health care providers because every insurer negotiates and pays a different price for the services the provider offers. That's one of the reasons price transparency has been so challenging a concept in health care--if you and I have different insurers, or perhaps even different products from the same insurer, then that X-ray at Hospital A won't necessarily have the same price for us.

Perhaps we shouldn't rely on teams of a handful of (well-meaning!) negotiators from an insurer and a provider system hashing out agreements in conference rooms to shape our prices, choices, and futures.

If instead the rates charged were all-payer--that is to say, the same for all comers--they would act like the prices for services in most industries. And with the veil of proprietary secrecy lifted on the prices of health care services, providers would have to compete directly with each other for customer-patients. Sunlight, here price transparency, would be the best disinfectant. And rather than relying on a privileged few to shape our decisions for us, as in Option 1, the market is shaped by millions of unconstrained consumers voting with their feet.

Why it hasn't worked and how it could. Standard pricers for all comers aren't common these days. But even if they were it wouldn't necessarily matter for reasons I walked through (with visuals!) in The Health Care Market. The only way the price differences matter and impact consumer choices--and through market feedbacks thus impact where providers set their prices--is if consumers feel them. It has to cost you the consumer more when you go to get a $1,500 procedure from Provider X instead of getting it for $800 from Provider Y. Otherwise the price differential doesn't shape behavior and the market is broken.

In this option we're sidelining the insurers: they no longer have to come up with clever reimbursement models that incentivize efficiency or cost savings, they don't need to negotiate prices with providers, they don't need to construct networks of providers (every provider is in! the insurer just pays its part of the bill). But they do need to construct insurance benefit designs that make sure you feel the impact of price differences between health care providers.

Maybe that's through a big deductible, though there are limits to that approach. Maybe that's through reference pricing, though there are limits to that approach. Maybe that's through coinsurance, though there are limits to approach. Or combinations of all these things. Or something else entirely. I don't know. But they need to figure out a way to pass through the price differences--if not the full prices themselves--to you the consumer so that providers are pitted against each other head-to-head.

Option 3: Eat the insurers

Another possibility is to get rid of insurers. Or rather subsume them into integrated health care provider systems. Sure, it wouldn't work in all health care markets (or maybe any). But health care providers themselves would take on insurance risk for their patients, essentially selling memberships of subscriptions to their networks of doctors and hospitals. You would pay a monthly premium right to the provider system--more likely its embedded insurance arm--and they would be at risk for 100% of the health expenses you incurred.

In some sense this is similar to Option 1 in that a single premium dollar amount offered in the marketplace sums up a number of factors bundled beneath the surface. If you want access to the prestigious brand of a higher-cost provider system in your market, you can buy into it but the costs of that choice are readily apparent to you in the premium associated with that system vs. those of its competitor(s). But it's also similar to Option 2, in that the differing cost structures/prices of competing providers are laid bare for the consumers (not obscured from the consumer as in Option 1) and submitted for their judgment in the open market.

Why it hasn't worked and how it could. Not all markets can support multiple competing provider systems. And while provider-sponsored insurance plans have proliferated (i.e., providers have been experimenting with directly offering their own insurance products), the results haven't always been pretty. And those insurance products are generally nowhere close to offering exclusive or near-exclusive access to those provider systems.

But if the push toward accountable care organizations accelerates and intensifies, provider systems will likely continue to tamp down on out-of-system referrals in favor of keeping their managed patients in-system. And as integrated but insular care becomes more common, the financial models (and rewards) accompanying the trend may push provider systems toward pricing and marketing their system to consumers directly. One could argue this is the logical end-state of the ACO craze. Not to mention the only way to credibly marry the need for integrated, patient-care with the imperative for some semblance of market dynamics in health care.

***

So where does this leave us? I'd be remiss if I didn't note that we have to acknowledge the fundamental limitations of applying a market approach to health care. Not least because Kenneth Arrow, author of the definitive early work and unrivaled classic on the topic ("Uncertainty and Welfare Economics of Medical Care"), just died five weeks ago.

And there exist key unavoidable realities that nudge us toward the conception of health care we discussed back in The Heath Care Polis. Most of our national health spending is concentrated on a small segment of the population, many of whom have multiple chronic conditions that send their expenses into the stratosphere. Many expenses, particularly on the inpatient side, will realistically always be beyond the reach of any effort to expose consumers to their price. And even when patients can be exposed to prices of high-cost procedures, such as through reference pricing, savvy providers can lower prices on those procedures to be competitive even as they raise prices on services behind the veil to avoid losing any revenue overall.

That doesn't mean market dynamics where we can have them aren't worthwhile. For instance, under Option 3 a provider system must account for all of its costs in the premium it markets to the consumer for entry to its system. There's no opportunity for sleight of hand like dropping prices on reference priced procedures while raising them on un-reference price-able procedures.

As alluring as the all-payer rates and relative administrative simplicity of Option 2 may be (it's the approach that envisions health care acting the most like other, regular markets), I don't believe it's actually a practical possibility. Direct price exposure is ultimately limited by the same reality that makes health insurance necessary: health care prices dwarf people's financial means. People will hit the inevitable out-of-pocket limits and the incentive will switch off just as we're nearing the  threshold of those services that are truly driving our national spending.

Option 3 may be realistic in some places, but it's hard to imagine it being the answer to the problems of our health care system on a large scale. Patients will always "leak" between systems, either by their own volition or because, oddly enough, it's more attractive for the expensive system to which they're attributed and has financial responsibility for them to send them to lower-priced competitors.

That leaves us with Option 1, which is simply a spin on the world we already inhabit. At the end of the day I think we do need a market organizer, the "dictators exercising great power" (in the words of Bernays) who design the choice architecture we use to make sense of the world and who enable us to make the best possible decisions with what limited time and energy we have. Insurers can design more limited or tiered network products that prod us toward lower-cost providers, they can employ value-based insurance design to nudge us toward the smartest treatment options, they can provide the infrastructure and payment incentives to support the benefits of an ACO-centric provider world without requiring the consolidated mega-provider systems of Option 3, and they can aggressively negotiate on health care prices.

Instead of leaving consumers to respond directly to price signals from providers, insurers can bake their own assessments of relative value into the financial incentives and signals sent to consumers by the structure of their products, all without discouraging those who need chronic care from pursuing needed treatment. And if they're successful in all of the above, they'll best their competitors by putting forth offerings with the most competitive premiums--the single, consolidated price bundling the fruits of all the above-mentioned efforts. They can make the most of behavioral economics and smart payer-provider contracting to market directly to consumers the best-priced insurance products and, in turn, hold health care providers in check. The catch is that consumers have to be in the market for such products: which means no one else, such as an employer, can be choosing their insurance products for them. And insurers will need to get in the business of insuring people, not providing risk-less administrative services. A sharp departure from the current norm that will require some potentially tumultuous policy changes.

But in the end I think I've talked myself into learning to love managed care. Viva la insurer competition!