Antitrust debates have largely ignored questions about the relationship between market power and productivity, and scholars have provided little guidance on the issue due to data limitations. However, data is plentiful on the hospital industry for both market power and operating costs and productivity, and researchers need to take advantage, writes David Ennis.
In a paper published in 1935, the British economist Sir John Hicks made the following observation: “The best of all monopoly profits is a quiet life.” He went on to say that there were “subjective costs” to the monopolist, such that he would “not bother” to get to the position of maximizing profits. So, the monopolist wouldn’t necessarily bother to put in the effort to maximize productivity. In the field of economics, this issue comes under the province of Industrial Organization (IO).
There has been considerable IO research outside of healthcare on this issue. But University of Chicago economist Chad Syverson notes that research in this area is complicated by difficulties in measuring market power, the classic definition of which is a firm’s ability to mark up prices above marginal cost of a competitive firm. Often, good price or cost information is not available and, consequently, market concentration has been used as a proxy for market power. Syverson, however, notes that this is not a reliable measure of market power, with the literature showing both positive and negative correlations in some cases. A market can become more concentrated, for example, when an innovative, high productivity producer drives the low productivity firms out of the market.
Alternatively,, there are a number of published papers that show market power and productivity to have a clear negative correlation, with representative studies including the following: A Federal Reserve Bank of Minneapolis study by James Schmitz found that monopolies not only increase prices but also reduce productivity, consistent with Hicks’ “quiet life” observation. In a 2016 NBER paper, Bruce Blonigen and Justin Pierce found that mergers and acquisitions in manufacturing increased market power (evidenced through higher markups) but had no impact on operating efficiency, which should have been enhanced through economies of scale. This same finding has also been shown in healthcare consolidations (see Carnegie Mellon University professor Martin Gaynor’s testimony before the United States House Judiciary Committee discussed below). This is clearly not productivity enhancing behavior. Looking at this from the reverse direction, Ralf Martin of the London School of Economics studied the removal of trade barriers in Chile in the early 1980s and found that after the market power of domestic manufacturers was reduced, productivity increased across all affected industries. This one area, where market power emanates from trade barriers, shows consistency in the research for improved productivity when market power is reduced. And this would appear to be causal rather than just correlational.
It is curious that, in all the discussion about antitrust in recent years, the issue of productivity and market power seems to get relatively little attention. If operating costs are higher than you would find in a competitive market, then there is economic welfare loss because resources are not being employed efficiently. Furthermore, reduced productivity may not appear in the short run but could develop in the longer run as management becomes accustomed to entrenched market power. This would imply a need to involve more IO research in antitrust considerations and the need to identify the conditions under which market power leads to lower productivity in the long run.
There has been minimal study of market power and productivity in the hospital industry where I have observed, anecdotally, that hospitals with high market power are operationally less efficient. Concentration and market power have been increasing for decades in this industry and costs continue to increase well above general price level inflation. The increase in market power in the hospital industry and the availability of extensive hospital operating cost information would make this a good setting to explore the behavior of executives in managing operating costs under various levels of market power. There is already an extensive literature on higher prices for hospitals that have market power (as shown by Zack Cooper, et al., and Gaynor discussed below) but very limited research on whether market power results in lower productivity, as Hicks implied.
From a management perspective, non-profit hospitals (which constitute 76% of community hospitals in the U.S.) are in an unusual situation: Unlike a for-profit organization, they do not have shareholders demanding higher dividends or more efficient operations. Rather, there are community boards, typically consisting of local business leaders, medical professionals, and community leaders, who expect management to achieve reasonable profitability. To achieve excessive profit would not be in the community’s interest, either for the patients in the community or the local employers who pay the bulk of their employees’ health care bills. Obviously, the definition of “excessive” will vary among organizations, but it is not necessarily “maximum.” In any event, non-profit hospitals appear to be a good industry to explore the “Hicks Effect.”
The question for management of a hospital that has high market power is to what degree you rely on price increases versus operational cost reductions to achieve target profitability. Significant operating cost efficiencies are hard to achieve and often risky for management in a financially healthy organization. In addition to the hard work involved, there is the risk of angering the medical staff and/or the hospital workforce. No matter how justified the effort to reduce costs, the executive’s job could be at risk if significant organizational anger develops. So the rational thing for management of a hospital with market power to do is achieve some comfortable operating cost improvements and then rely on price increases from the private payers to achieve target margins. The higher the market power, the more management can rely on increased prices from the private payers. A hospital with low market power has limited bargaining power with insurers and must rely much more on operational efficiencies to achieve target margins. The Hicks’ Effect, then, is not behavioral economics…it is rational behavior in the context of the organizational and industry structure.
This situation begs the research question of whether high market power is associated with lower productivity (and, whether for-profit hospitals at the same market power level have higher productivity than non-profits). Unfortunately, I could find only one study of the hospital industry directly addressing this market power-productivity question. In a 2010 Health Affairs article, Jeffrey Stensland, Zachary Gaumer, and Mark Miller, all of whom were working for the Medicare Payment Advisory Commission, found that hospitals with high market power have higher cost per unit of service, often resulting in operating losses in their Medicare business, losses which are facilitated by the high margins in their private pay business. Alternatively, hospitals with low market power constrain costs to make a margin on Medicare. However, the sample for this study was relatively small and the authors did not have access to strong measures of market power.
Indirect measures of the correlation between market power and operational efficiencies can be found in the literature on the cost and price effects of hospital consolidations. Gaynor discussed price and cost impacts of hospital consolidations in his testimony before the House Judiciary Committee in 2019. He noted that consolidation frequently results in significant price increases and that there is little evidence that consolidated entities are more efficient. So, in merger situations, the increased market power allows management to increase prices to private insurance payers to achieve target operating margins rather than undertaking the hard and unpopular work of achieving all of the potential operating efficiencies that a merger might afford.
A recent NBER working paper by Benjamin Handel and Kate Ho reviewed a large amount of IO research in health care. These studies covered pricing, price negotiation, impact of M&A on price and quality, impact on price of insurance and provider market design, and consumer choice, among other things. Nowhere in their overview of the IO literature can be found research on the relationship between market power and productivity. And, unfortunately, this area of study was not even suggested in their recommendations for future IO healthcare research. But there is a clear need for a more analysis of this issue.
Measuring market power is quite difficult, as noted above. In a recent paper, Maximilian Pany, Michael, Chernew, and Leemore Dafny demonstrated that efforts to regulate hospital prices in concentrated markets would miss many high-priced hospitals in nominally competitive markets when geographic concentration of providers in a market is used as the measure of competition (and market power). For example, an academic medical center in an urban market that has many hospital competitors will command much higher payments for identical services due to the market power that emanates from its perceived higher quality, even where the actual quality difference cannot be demonstrated.
So, market concentration does not always work well as a measure of market power in the hospital industry, either, so would it be better to measure the size of the gap between price and marginal cost? But marginal costs, as noted above, are difficult to measure in many industries, including hospitals. Furthermore, marginal costs in a low-productivity hospital will be higher than necessary and understate the degree of market power. Fortunately, an excellent alternative for hospitals is to measure the gap in price between private insurance payers and Medicare for identical services. Hospitals and health insurance companies negotiate prices and the higher the market power of the hospital, the higher the prices it can negotiate. With respect to Medicare, all hospitals, regardless of market power, must accept government-set prices. So, with Medicare prices fixed, the ratio of private insurance price to Medicare price should be a direct measure of a hospital’s market power. This measurement of private insurance payments versus Medicare has been done by Cooper et. al. in their groundbreaking study of hospital prices. Their data allow for computation of the Private-pay/Medicare ratio for individual hospitals. This provides an excellent measure of market power for any hospital that has a significant volume of privately-insured patients.
This ratio measure can be combined with cost data from the Medicare Cost Report system in the manner done by Stensland et. al., cited above, to determine the correlation between market power and operating cost. It may well be that many of our hospitals with higher prices have the additional welfare cost of inefficient use of resources.
Access to reliable and comparative industry-wide cost and price data is difficult in the private sector and makes broad-based studies a challenge to undertake. However, the public availability of the Medicare Cost Report data for nearly every hospital in the country, together with information on actual prices, creates a unique opportunity to explore the market power-productivity question that Hicks raised over 85 years ago.
Author Disclosure: David Ennis previously worked in hospital management and strategic consulting but is now retired and has no financial interest in the publication of this article.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
