• Where’s the private sector cost control?

    This is a TIE-U post associated with Jonathan Kolstad’s The Economics of Health Care and Policy (Penn’s HCMG 903-001, Spring 2012). For other posts in this series, see the course intro.

    Mark Pauly’s Taxation, Health Insurance, and Market Failure in the Medical Economy (ungaged PDF) [1], published in 1986, is a review of much of the health economics literature and thought as of the mid 1980s. It’s interesting to see what has changed since then. For instance, we are now aware of several examples of adverse selection death spirals, and we now better understand the nature and consequences of the employer-sponsored health insurance (ESI) tax subsidy. The latter is a principal focus of Pauly’s paper.

    Since it covers a great deal of territory, one could use the paper as a springboard for just about anything in health care. I’ll use it to return to a question I posed on Twitter and by email to colleagues last week. Pauly asks,

    The fundamental question that has yet to be answered is […] why did not market forces induce individual insurance firms to supply “cost control” of some type, and so obviate the need for government regulation? […]

    One answer [A] is a conspiracy or pressure theory. The notion is that providers of hospital and physician services would prefer fewer to more restrictions. Physicians in particular have been alleged to move collectively in overt and covert ways to resist insurer cost controls that exist and to discourage their introduction. […]

    A second argument [B] relates to the nature of the industry. […] An insurer that imposes a cost-control program on its insureds, at some explicit or implicit cost to them, may be unable to recoup all of the benefit from lower-service intensity in the form of lower premiums, because service intensity falls for all users of health care, not just for the firm’s insureds. […]

    Partly it is due to the inability of firms, either doctors or hospitals, to offer different quality levels at the same time to those with different types of insurance. […]

    A third [C] explanation finds the cause of undersupply of cost-control activity to be the tax subsidy directly. […]

    I find the first two, labeled A and B, especially plausible. The third, C, that the ESI tax subsidy deserves some blame, is also plausible, but becoming less so as health care costs are exerting an ever-greater strain on businesses and workers. There is a documented outcry for lower-cost insurance. Market participants on the demand side want to lower costs (spending). My question is, why has there not been a private sector revolution in evidence-based medicine to squeeze out the waste?

    Perhaps the best complement to Pauly’s explanation  is that the private sector has concluded it is not profitable to invest in the type of research that is needed. As I’ve written before, comparative effectiveness research is a public good. Also, at least as of the mid-1990s or so, insurers were highly constrained by the law and the courts as to what they could get physicians to do (and not do).

    Co-blogger Kevin suggests things have changed, however. Plans are getting smarter about how they define benefits and coverage. Moreover, some insurers are moving toward global payment (or capitated) models with quality- and cost-based bonuses, much like Medicare’s ACOs. If this trend holds, it should be providers, not just insurers and self-insured employers, that demand information on how to provide care of equivalent (or better) quality at lower (or the same) cost.

    In other words, cost control will not merely be a demand-side concern. The supply-side will care too. By itself, that does not mean the cost cutting will be efficient. A lot rides on how providers are monitored, what incentives they face. I don’t expect that our current plans are the solution. There’s no good reason to believe that we have somehow figured it all out.

    Pauly’s question is a good one, and I think it will remain relevant for years to come.


    [1] Pauly, Mark, “Taxation, Health Insurance, and Market Failure in Medical Care,” Journal of Economic Literature, 24(2), June 1986, 629-675.


    • I agree with you that this is one of the key questions in health care. Insurers and consumers stand to gain substantially from cost control, and one would think insurers are in a good position to make progress on the issue.

      Something like [B] must be going on, and I think [B] includes the point on research that you made. Successful cost control, and the many fixed costs that go into it, probably do result in public goods. The research, the management techniques, the compensation arrangements, etc. are costly, and the gains from them cannot be restricted to the innovative insurers.

      Also, seeing the ACO regs being created illustrates just how complex it is to institute a new compensation scheme. Imagine if every insurer had to invent these processes for themselves! Even with better appropriability, I think there would be substantial obstacles to progress because of high returns to scale in cost control investments.

    • Isn’t it true that Medicare Advantage plans spend less money in direct patient care compared to the traditional Medicare? Yet there is no proof that their medical outcome is any worse than traditional Medicare. So why can’t we all do what Medicare Advantage plans did and save money? Of course we have to take away those huge profits now enjoyed by Medicare Advantage plan owners.

    • -Seems like if you really want to make an accurate evaluation of cost control over time you’d have to control for changes in the quality, volume, composition, and scope of medical care delivered over time and hold all of these factors constant. Do that and you can meaningfully determine the extent to which consumers are being charged more for the same care over time. Fail to do that and you are simply measuring spending growth, not cost growth. Since spending growth and cost growth are two different things, this distinction matters quite a bit.

    • Austin, you’re mixing two desirable cost control methods together, and they’d be easier to think through if you separated them:

      1. Cost control by insurers (i.e., risk management)
      2. Cost control by providers (i.e., finding ways to deliver care more efficiently and then lowering price)

      My theory for number one is that insurance search frictions have made it okay for insurers not to try risky innovations in cost-saving prevention, plus short time horizons makes fewer such efforts a good investment if they can’t recoup the upfront investment before the insured moves on. In anticipation of insurance exchanges that will remove many search frictions, insurers are starting to try out more risky cost-saving prevention ideas (e.g., WellPoint paying doctors more for making medical homes).