• Antitrust and Health Reform

    This post is co-authored by Austin Frakt and Ian Crosby.

    In the wake of AHIP’s promotion of the report sham study it commissioned from PricewaterouseCoopers (PWC), Senate Democrats are pushing to repeal a 1945 statute that partially exempts insurers from national antitrust law. While it may be wise to regulate insurers at the national as opposed to the state level, legislating in anger is not. At this stage, however, this isn’t anywhere near serious legislation. It is a signal of Democrats’ intent and a warning shot across the bow of the insurance industry.

    AHIP’s PWC provocation was not only poor politics, it was poor timing. Only days after its release Olympia Snowe traded a “yes” committee vote for a seat at the Senate negotiating table. It now seems less likely health reform legislation will emerge with a public option. That would be a victory for the industry and would place the burden of negotiating better treatment at lower cost on the private sector, with government regulators looking over the industry’s shoulder. It will take market muscle to shoulder that load.

    But not too much market muscle. Taxpayers will be best served by insurers with sufficient market power to bargain down provider rates, but with not quite enough power to keep the savings (“rents”) for themselves. That is, we want low provider rates to translate into low premiums. Though liberals may be skeptical that this balance is achievable, it is not at odds with their objectives in principle. After all, one of the arguments for the public option is that it would be a large insurer with commensurately large negotiating power but would use that power on the behalf of consumers.

    How to balance the power of insurers and providers is far from simple. Many have pointed to the alleged dominant market position of insurers as a substantial source of high health care costs. However, the health economics literature supports the notion that recent increased market power of insurers does not lead toward monopolistic pricing, but rather it provides a counter-balance to the power held by hospitals and provider groups.

    Moreover, insurance companies are partially exempt from federal antitrust law for an important reason: so they can share rate-making data. This function actually benefits small insurers who would not otherwise have sufficient data to properly adjust premiums. Paradoxically, removing the legal cover for data sharing would harm small insurers more than large ones.

    All this suggests that repealing the federal antitrust exemption for insurers may be misguided. Though the insurance antitrust exemption is a popular whipping boy for Democratic politicians, it is by no means clear that repealing it is practical or beneficial to consumers. Instead, antitrust law might better aid the cause of health care reform by first focusing on providers. While a few proposed hospital mergers have been blocked recently, it follows a long period of hospital consolidation.

    Yet even here, antitrust exemptions may be inevitable to allow doctors and hospitals to negotiate bundled payments contingent on performance under the guise of accountable care organizations (ACOs). It is not unreasonable to worry that providers might try to use the ACO structure to lobby Congress and negotiate more favorable Medicare payments and regulation (in fact, such activities are constitutionally protected under the the Noerr-Pennington doctrine). So we have to encourage provider integration without unduly increasing the strength of providers with respect to insurers or their regulators. This will be hard indeed and could be harder still with a weakened insurance industry.

    All this suggests a confused focus in Washington, though one consistent with populist sentiment. Insurers, rightly or wrongly, are the scapegoat and providers (with the exception of the drug industry) are viewed more favorably. There are clearly insurance reforms worth implementing, but weakening insurers’ market power while strengthening that of providers may not be one of them.

    Beating up on the insurers feels good but may not be the right medicine.

    • I am an antitrust lawyer. Your portrayal of the McCarran Act is absolutely unbelievably incorrect. The McCarran act does not, repeat, does not permit insurance carriers to agree among themselves as to what they can pay providers. This activity has been held by the US Supreme Court not to be in “the business of insurance” as required by the McCarran Ferguson Act.

    • @Bill – Neither of us meant to suggest that McCarran-Ferguson allows carriers to agree on what to pay providers. But it does exempt them from rule of reason review of insurance practices that may help them to build and defend their market share, and arguably limits the role of federal antitrust authorities in reviewing some health insurer mergers. The point is, without a public option, it is worth considering whether measures that may limit concentration on the insurer side may have unintended consequences given the increasing concentration on the provider side.

    • Wrong again. (By the way, I am an adjunct prof in antitrust) and am the former vice chair of the ABA Antitrust Section McCarran-Ferguson Committee)

      The United States Supreme Court in Pireno made it absolutely clear that agreements relating to the procurement of goods or services was not the “business of insurance”, and hence does not fall within the McCarran-Ferguson Act. See, Pireno at http://supreme.justia.com/us/458/119/case.html

      Furthermore, McCarran has nothing to do with “of reason review of insurance practices”–it is a total exemption, not an exemption that confers rule of reason status on anything.

      I recommend you confer with a faculty member at your local law school and retract the errors of this post.

    • As to McCarran being necessary for carriers to exchange historical data on losses, that’s not the case either. California opted to apply antitrust to the business of insurance. The State AG opined that the carriers could aggregate historical loss data without violating the state antitrust laws.

      And, nothing compels a carrier to disclose loss data to a rival, certainly not McCarran, although the state insurance commissioner could compel such disclosure, McCarran notwithstanding.

      You might be surprised that I share your views that a carrier may have to have sufficient presence to make it attractive for hospitals or doctors to offer it a discount or to participate in managed care protocols. But, McCarran is irrelevant to that.

      Now, I haven’t seen or studied the health bill, but to the extent that it permits interstate issuance of policies (and I don’t know if it does), this does raise the question of whether McCarran applies at all, because the carrier’s policy must be “regulated by state law”–if there is federal, and not state, pre-emptive regulation, the antitrust laws would apply (since state regulation does not apply), absent a new exemption in the legislation.

    • @Bill 1 – I certainly do not pretend to your depth of knowledge in the area of McCarran-Ferguson, and am happy to benefit from it. But I believe you misunderstand me again. By “insurance practices that may help them to build and defend their market share” I am referring to their insurance relations with their customers and perhaps their interactions with each other in relation to the risk-spreading function (such as sharing rating information) that I understand to be exempt from federal antitrust scrutiny. I also understand that the exemption is total with respect to the business of insurance (other than boycotts, coercion, and intimidation) that is regulated by a state, and I didn’t mean to suggest that the examples I gave were exhaustive. Rather, they were given as examples of potentially exempt conduct that may facilitate concentration on the insurer side.

    • @ Bill 2 – As for exchanging loss data, I don’t believe the claim is that McCarran is necessary for such exchanges to occur, but rather that it may facilitate them, inasmuch as state antitrust regulators may be less aggressive in concluding that a particular arrangement operates as a vehicle for fixing prices. For the record, I don’t believe that McCarran is an ideal vehicle for facilitating such exchanges. Leaving aside other issues, the ABA proposal for repealing McCarran while adopting safe harbor provisions for beneficial collaboration seems promising to me on this front. But I’m not sure that outright repeal without more, subjecting such exchanges to ex post review under the rule of reason, is the way to encourage them. If I were a large insurer considering the potential consequences of error in such review, I might just take my loss data and go home.

    • @ Bill 3 – As an expert in McCarran-Ferguson, do you believe it has no impact in facilitating concentration in health insurance markets?

    • Ian,

      I do think McCarran can facilitate concentration in the industry, but I do not believe that once you give people power that they will exercise it in a way that improves public welfare–they operate for their own welfare.

    • Not to be a pest, I looked at one of the articles on increasing monopsony power by healthcare plans and consumer effects that formed the basis for the initial premise that McCarran somehow increased buyer power (it only affects competition on the sell side between carriers, not the buy side) and this would be beneficial to consumers.

      My initial reaction would have been that, unless the monopsonist was a coop non-profit that redistributed gains back to consumer owners, that monopsony would not necessarily benefit consumers. And, lo and behold, that’s what the article said: monopsony has to be associated with non-profit ownership; having a private carrier with monopsony power does not solve the problem.

      So, going back to this post, McCarrans facilitation of concentration on the sell side does not benefit consumers unless there is a monopsony of a non-profit consumer HMO. I do not see how McCarrans facilitation of market power on the sell side translates into power on the buy side (if two carriers agree on rates, but can’t agree on what to pay providers, sell side price goes up but nothing should change on the buy side since that can’t be coordinated). Furthermore, you would have to posit that both carriers were nonprofit monopsonists who distributed gains back to their customers.

    • @Bill – I believe the premise is that effects on competition among insurers for customers on the sell side can lead to increased concentration, which can result in increased market power on the part of individual insurers on the buy side. Austin can no doubt speak better to the economic consequences of this, but I don’t expect you would find either of us saying that increased concentration in the insurance market is an unqualified good. Rather, the claim is that in the absence of a public option, and in the face of indications that reformers intend to go easy on providers to get them on board, to the point of perhaps facilitating their opportunities for collusion, focusing on competition among insurers may be barking up the wrong tree, and could lead to perverse consequences. I understand the research Austin cites to show that increased concentration in the insurance market can lead to provider monopoly busting without necessarily producing insurer monopsony. And certainly, some insurers who benefit from relaxed antitrust regulation are nonprofits and cooperatives who may be expected to pass savings on to their subscribers. For-profit insurers may be less inclined to do so, but they face the prospect of increased regulation as a result of reform, and increased political risk if their profits go through the roof in wake of it. In addition, it is not clear to me that concentration sufficient to confer an advantage on the buy side necessarily confers a completely offsetting advantage to set prices on the sell side. In sum, the point is a modest one: don’t let ire at the insurance industry result in legislation that alters the balance of market power between insurers and providers without careful thought for the consequences.

    • I want to challenge the premise that McCarran leads to increased concentration on the sell side, separating from that statement that it certainly does permit agreement on competitive terms among individual actors.

      One of the premises of McCarran was that it would enable small carriers to enter a market, usually participating in a rating bureau which established rates for the group. And, indeed, that is what happens if you look at products such as workers comp insurance. You may have 30 or so carriers, each offering the same rate for the same policy coverage. The experience has been that when states have deregulated, permitting rate deviation, rates and policy premiums declined.

      I think we can agree that having a large number of carriers operating under a price-fix umbrella of a rating bureau mistates number of competitors for competition. But, it also demonstrates that McCarran, rather than reducing the number of ostensible “competitors”, increases the number of fringe players who, according to the monopsony hypothesis of the post, would be unable to bargain for a better price.

      I think the issue of concentration and collusion on the sell side via McCarran is different than concentration on the buy side.

      Now, I do agree with you that a public option may be necessary, but not necessarily for monopsony purchasing reasons, but for yardstick measures and for the threat of new entry. Also, a public option (or even a national coop of states) also creates an opportunity for someone with a book of business to go to the 50,000 person community with one hospital that has resisted managed care (often in cahoots with the local Blue Plan) and offer a discount to get the new business, or offer to a would be outpatient surgery center some new business if the hospital doesn’t discount or change managed care protocols. This may just be dreaming: it could be the public option would be just a dumping ground for bad risks of the health insurance industry; it could be that the public option would be too timid to adopt managed care protocols and limit doctor participation to get discounts and managed care changes from doctors. To say one is a public option is not to be a public option. Little devils are in devilish details.

    • @ Bill – I’m a little confused, because two posts back, you said “I do think McCarran can facilitate concentration in the industry.” Are you backing off that statement? Do you disagree with Christine Varney’s testimony on Wednesday that market allocation can fall within the scope of the McCarran exemption? What about actions in relation to their own insureds that, for example, increase switching costs for consumers and thereby barriers to entry for competitors? I’m not saying that any of these are good things, only that going after the carriers while giving providers a pass may just work to strengthen provider market power rather than benefit consumers.

    • @Ian What I mean when I said–and if you read the entire sentence I think it is clear– “McCarran does not facilitate concentration” in the industry is that it does not do so in a structural sense, ie., create fewer firms. Concentration is a term that I used in industrial organization economics sense. I made that clear by continuing in the sentence the following: “separating from that statement that it certainly does permit agreement on competitive terms among individual actors.” That is, it does not structural concentration–lead to fewer firms–but it does permit agreement. And, since insurers can’t agree on procurement–as it is not in”in the business of insurance” –the fact that McCarran supports more firms operating under a cartel umbrella does not support the view that it leads to monopsony procurement concentration as was the underlying thesis of the post. To the contrary, if you look at workers comps and other lines, there are many, many firms–and they all charge the rating bureau rates as the state minimum rate. As to Varney, Yes, market allocation could fall within McCarran, as does every violation of the antitrust law except boycott, coercion or intimidation, but it’s been a while and would like to check if any recent decisions treat market division as to customersas boycott. I don’t understand the remainder of the post other than if it says give insurers market power so they can fight with providers who use illegal contract terms to inhibit competition. My response would be to go after both: the insurer and the provider and let the antitrust laws apply to both the insurer and provider. That’s the consumer protection and that’s the way to help the market.