• The Health Care Productivity Problem

    I’ve blogged twice already about David Cutler’s recent NBER paper on inefficiencies and costs in the health care system. It’s worth some attention and it is an easy read, so I recommend checking it out  in full. It includes the following figure (original source: Oliner et al., 2007).


    Cutler acknowledges that productivity is hard to measure in health care, and I don’t know how it’s done. Nevertheless, according to the figure, productivity growth in health, education, and social services was negative between 1995 and 2005, whereas the average across industries was 2.4%. This is not a good sign. To have any hope of obtaining a reasonable return on our massive health care (and other social program) spending, we need productivity in the sector to grow, not shrink. If the 1995-2005 trend continues, we risk both the health of our economy and populace.

    Cutler’s paper explains why productivity and productivity growth are low in health care and what can be done about it. He hypothesizes why we see so little innovation in health care and suggests ways to promote it. It’s a familiar set of problems (asymmetric information, inability of plans and providers to capture long-term returns on short-term investments, plan turnover, third-party payment, etc.) and solutions (bundling, provider integration, pay-for-performance, etc.). He concludes,

    To reduce this waste, organizational innovation will be required. To date, however, such innovation has been very rare in health care. This paper argues that lack of information and poor incentives are the key barriers to new organizational models, and accordingly that public action to address these issues is needed. Recent reform legislation has made changes in each of these areas. Whether the legislation addresses these problems sufficiently is something that only time will tell. (© 2010 by David M. Cutler.)

    Actually, there are two levels of innovation required. We need innovation in the health care sector. But we don’t know for sure how best to foster it. So, we first need innovation in health policy. Doing nothing won’t suffice. Try we must.


    Oliner, Stephen D., Daniel E. Sichel, and Kevin J. Stiroh (2007), “Explaining a Productive Decade.” Brookings Papers on Economic Activity, 1: 81–137. [Cutler didn’t list this reference, but I think this is the one for which the figure is the source.]

    Comments closed
    • Austin,

      William Baumol’s (1967) “cost disease of the service sector” (or model of unbalanced macroeconomic growth) certainly plays a role here. He argues that wages in the technologically-stagnant sector (services) are tied to the wages in the technologically-progressive sector (manufacturing). Overtime wages rise in the progressive sector because of productivity gains due to new technologies but wages also rise in the stagnant by assumption. As a result, marginal costs increase (supply declines) in the stagnant sector, thereby pushing price and expenditures upward (given an inelastic demand for services). Consequently, the service sector continues to take a larger and larger share of the economic pie overtime.

      He also had a 1992 Wall Street Journal article specifically applying the cost-disease explanation to health care. Several papers have tested his theory.


    • One of my thoughts reading Cutler’s paper, is that in general service industries have more difficulty increasing productivity. The chart above seems to at least partially support that idea.

      My second thought was lack of integration. Very little in medicine is integrated. A patient having a procedure will likely interact with multiple physicians, nurses and technicians. They will work for multiple organizations, or sometimes, will still be self employed. I suspect that we wont see real productivity gains until you see all of these providers working for the same system in some fashion. Sharing a common information source would also be a huge boon. Every new patient is a mystery that can take a long time to unravel in our current system of fragmented data. We nearly let a patient die last week because the family told us he had metastatic lung cancer. We operated anyway, emergency surgery, and it turned out he did not have any cancer. A reliable, quickly accessed medical information system would have made our guessing not necessary.

      IOW, while you are looking for top-down solutions, I hope that you are looking at the micro side also.