April 9, 2007
Testimony of Lawton R. Burns re. the Highmark/Independence Blue Cross Merger
Good morning. My name is Lawton Robert Burns. I am the James Joo-Jin Kim Professor, Professor of Health Care Systems and Management, and Director of the Wharton Center for Health Management and Economics - - all at the Wharton School at the University of Pennsylvania. I appreciate the opportunity to present testimony about the Highmark/Independence Blue Cross Merger.
My remarks are drawn from research I have conducted on the history of the Pennsylvania insurer and hospital markets since the 1980s, and national research on the relationship between insurer and hospital market structures. They are also based on my understanding of the field of industrial organization economics, which examines (in part) the causes and consequences of mergers. I have taught the graduate-level course in industrial organization as applied to healthcare at The Wharton School for the past ten years.
2. The Highmark/Independence Blue Cross Merger
On March 28, 2007, Highmark and Independence Blue Cross (IBC) announced their merger. The press statement mentions several benefits of their combination:
1. Generate $1 Billion in economic benefits via:
a) $650 Million to expand health insurance access to the uninsured
b) $280 Million savings from better management of prescription drug costs
c) $300 Million savings from holding administrative fees flat for two years
2. Better serve customers and providers
3. Improve health care quality and the health of communities served
4. Generate new business, create jobs, and stimulate business opportunities for Penna firms
5. Meet shifting customer demands for new products
6. Combine the best practices, talents, and resources of the two firms
7. Fund essential technological and infrastructure improvements to deal with external, competing, investor-owned health plans
The aims of the merger are lofty. Unfortunately, there is no detail provided regarding how these benefits are to be achieved. My personal view is that most of these benefits are probably not attainable for several reasons.
3. Why the Merger Won't Produce the Anticipated & Espoused Benefits
First, the merger is labeled by both companies as a "combination". The two firms will maintain their respective headquarters in Pittsburgh and Philadelphia. There thus seems to be little integration or consolidation of the infrastructure of the health plans. As a consequence, it is difficult to envision where any savings and efficiencies will spring from. In fact, there may be duplication due to the use of a combination rather than an integrative merger. There may also be higher costs of operations, simply due to the need to coordinate two giant operations located 300 miles apart. The Allegheny Health Education & Research Foundation (AHERF) discovered this sad fact prior to its bankruptcy nine years ago. Many mergers achieve at least short-term savings by combining administrative functions and reducing administrative headcounts. That does not seem to be the aim here, since one goal of the merger is to create jobs.
Second, the literature on corporate mergers and acquisitions is quite clear in showing that efficiencies and synergies result from defined pre- and post-integration efforts. There is no detail regarding these efforts in this pre-merger phase. Specifically, the economic literature on scale economies outlines the different areas in which efficiencies can be reaped: spreading of fixed costs over larger volume, increased specialization of labor, enhanced ability to raise capital, lower costs of carrying inventory, learning curve effects, marketing economies (e.g., branding, advertising), promotion economies (e.g., lower consumer transaction and search costs), and purchasing economies (greater leverage over suppliers). One of the espoused rationales for the merger is better management of drug costs. It is hard to see how the combined firm would have any more leverage over pharmaceutical suppliers than the individual firms already enjoy. Suppliers grant discounts based on local market penetration; Highmark and IBC already have achieved this, and their combination will not increase it. Indeed, to the extent they exist at all, many of the proposed benefits of the merger may already be attainable by the separate organizations. Highmark and IBC should be explicit in outlining where the efficiencies and synergies are to come from, and how they will be achieved. Simply combining two firms without integrating them does not yield operating efficiencies.
Third, even in the presence of such efforts and defined post-integration strategies, scale economies and merger efficiencies are difficult to achieve. The econometric literature shows that scale economies in HMO health plans are reached at roughly 100,000 enrollees. Enrollment levels at the HMO plans operated by Highmark and IBC far exceed the minimum efficient scale. Moreover, the provision of health insurance (e.g., front-office and back-office functions) is a labor-intensive rather than capital-intensive industry. As a result, there are minimal economies to reap as scale increases. Moreover, there is little econometric evidence for scale economies in multi-plant firms - - e.g., firms that have multiple sites or plants of operation scattered geographically. This seems to be the case with this merger. Finally, there is little econometric evidence for economies of scope in these health plans - - e.g., serving both the commercial and Medicare populations. Serving these different patient populations requires different types of infrastructure. Hence, few efficiencies may be reaped from serving large and diverse client populations. Indeed, really large firms may suffer from diseconomies of scale.
Fourth, the recent historical experience with mergers of managed care plans and other types of enterprises does not reveal any long-term efficiencies. Indeed, a recent Wall Street analysis shows that the mergers of investor-owned health insurers under-perform the market two years after the merger. More broadly, the strategy literature shows that the majority of corporate mergers (60-70%) fail. What explains the low success rate? A major problem is the failure to deliver on the sources of value, which is extraordinarily difficult to do. Mergers are complex situations; mergers of large companies are even more complex. In fact, the literature shows that mergers of two evenly-sized firms are the most complex and difficult to extract value from, given the convoluted politics of integration between firms that consider themselves equal.
4. So Why Do Mergers Continue to Occur?
If all of the above is true, when then do mergers (and mergers of health plans) continue to occur? One reason is "managerial hubris": the feeling of firm executives that they can do what others have not done to extract value from existing operations. Another reason is "empire building": executives of larger firms derive monetary and psychic rewards from administering a bigger enterprise. A third reason, and one the Senate should consider, is that mergers serve to reduce the number of competitors in a market by at least one.
What is so important about the sheer number of competitors? Econometric evidence shows that in the managed care field, an increase in the number of competitors is associated with lower health plan costs and premiums; conversely, a decrease in the number of competitors is associated with increases in plan costs and premiums. The evidence also shows that the sheer number of competitors exerts a stronger influence on these outcomes than does the penetration level achieved by plans in the market. Perhaps the most significant effect of the Highmark/IBC merger is the removal of one competitor from the Pennsylvania health plan landscape.
One might then wonder what this landscape looks like statewide? The Commonwealth has four Blue Cross plans and one statewide Blue Shield plan. The four Blue Cross plans are: Highmark, Capital Blue Cross (CBC), Blue Cross of Northeastern Pennsylvania (BC-NEPA), and Independence Blue Cross (IBC). Highmark operates the one Blue Shield plan. The Blue Cross plans operate in various regions in the state. For purposes of discussing the Pennsylvania market today, I have identified eight regions as defined by the Hospital and Healthsystem Association of Pennsylvania (HAP). In HAP's report on HMO managed care, these regions and their dominant HMO health plans include:
Northwest Penna Highmark - Keystone Health Plan West
Southwest Penna Highmark - Keystone Health Plan West
Altoona/Johnstown Highmark - Keystone Health Plan West
North Central Penna Geisinger Health Plan
South Central Penna HealthAmerica
Northeast Penna Blue Cross of Northeast Penna - First Priority
Lehigh Valley Capital Blue Cross - Keystone Health Plan Central
Southeast Penna Independence Blue Cross - Keystone Health Plan East
This initial list suggests that the four Blues plans dominate the Western and Eastern portions of the state, with the Central region controlled by two non-Blues plans. The situation is a bit more concentrated than this, however. Highmark has 40 percent ownership of the Blues plan operations in Northeast Pennsylvania, and has joint operating agreements with BC-NEPA to market its traditional, comprehensive, senior, and PPO products. In effect, Highmark controls not only the Western portion of the state but also a solid piece of the Northeast. With the pending merger with IBC, Highmark would control not only the Western portion but most of the Eastern portion as well. One might surmise from this that Highmark's strategy, beginning with its formation in 1996 with the merger between Western Blue Cross and Pennsylvania Blue Shield, has been and continues to be its desire to be the only Blue Cross plan in the Commonwealth.
This would not necessarily lead to any further concentration in any of these eight regions. This is because the Blues plans have typically operated in their own regions and not poached on the territories of other Blues plans. One exception has been in South Central Pennsylvania, where Highmark ended its joint operating agreement with CBC around 2001 and has since competed with them for market share. Another reason why there would probably not be more concentration is because the various markets are already concentrated. Data published by the American Medical Association on both HMO and PPO enrollments in Pennsylvania's metropolitan statistical areas (MSAs) reveals that the vast majority of these markets are already highly concentrated with respect to HMO products, and most are concentrated with respect to their PPO markets as well. That is, there is relatively little competition within these markets. Philadelphia and Pittsburgh, in particular, are two of the most concentrated markets in the US.
The net effect of the Highmark-IBC merger might then be a nearly-statewide confederation of Blue Cross plans controlled by Highmark with strong domination in each region. What has changed is not so much the local market-level concentration but rather the common ownership and control of the plans that enjoy this concentrated market power. Is this a cause for concern? One might surmise that a powerful Highmark, with control over the Eastern and Western portions of the Commonwealth, might then set its sights on seeking to enter or combine operations with the Blues and other plans operating in the Central regions of the Commonwealth. This would have the effect of reducing what little competition already exists between rival Blues plans in South Central Pennsylvania and the LeHigh Valley. Indeed, given that Highmark and IBC are the two most powerful Blue Cross plans in the Commonwealth, one wonders whether the proposed merger eliminates any possible future competition between them as they eye one another's regions for market entry and expansion.
Are there other possible rationales for the Highmark/IBC merger? I think that one of the rationales for the merger espoused by Highmark may in fact be legitimate: the desire to confront the growing competition from out-of-state, investor-owned health plans such as UnitedHealthcare and Aetna. United has declared a major commitment to expand into Pennsylvania, which would serve to link up its more extensive operations to the East (Maryland, Virginia, New Jersey) and to the West (Ohio). United's strategy has been to contract with national employers with whom it does business elsewhere and which have operations here in the Commonwealth (e.g., Boeing).
The Blues plans in Pennsylvania are worried about the entry of these national plans into their marketplace for several reasons. First, the investor-owned plans have lower medical loss ratios and administrative costs. Second, they are less restricted in medical underwriting practices than the Blue Cross plans. Third, they have begun to take market share away from the Blue Cross plans in certain portions of the state (e.g., Southeast Pennsylvania). The Blues plans are worried that large national firms have the financial ability to under-price the market and sustain losses over several years in order to gain market share. A merger of Highmark and IBC might enable the combined firm to pool their reserves and stave off the threat of market entry and growth by these firms. Blue Cross plans commonly use their reserves to generate investment income that helps to moderate premium increases.
Alternatively, the national firms might competitively price their premiums but use their financial resources to pay providers in the Commonwealth more than the Blues plans currently pay. This strategy would enable them to develop contracts with hospitals more readily than in the past, and would surmount the historical tendency among providers to retard new market entry (and thereby shoot themselves in the foot) by asking for higher levels of reimbursement from the new insurer on the block. Most insurers could not afford to do this for long, and quickly exited the market - - leaving the market less competitive and more concentrated.
At present, there is little econometric evidence for the merger of large health plans like Highmark and IBC. To date, the two firms have failed to provide a convincing rationale and game plan for extracting the value and efficiencies from their proposed combination. There is some legitimate concern that the proposed merger has potential anti-competitive effects on existing Blue Cross plans in other regions of the Commonwealth as well as market entry and expansion by national investor-owned firms.
1. For purposes of definition, I define "concentrated market structure" in terms of the number of competitors and their relative share of the market. These two components are often summarized as the Herfindahl-Hirschman Index (HHI). This index measures how much market share is concentrated in one or a few large health plans. The HHI is measured as the sum of the squared shares of each firm in the market. Thus, a market with three firms whose shares are 25%, 25%, and 50% would be equal to: 252 + 252 + 502 = 3,850. The higher the HHI, the more concentrated the market, and the more powerful are one or a few plans. According to the Department of Justice's Horizontal Merger Guidelines, markets with HHI greater than 1,800 are highly concentrated.