Consolidation and the Transformation of Competition in Health Insurance

James C. Robinson


Health Affairs. 2004;23(6) 

In This Article

Quantifying Consolidation

Available data on health insurance market shares are distorted because of the differences among products and customer segments in the manner by which insurance is regulated and enrollment is counted. Most HMOs are insured at the state level, and data on firms and enrollment are available from consultants, vendors, and researchers willing to collect the data from state insurance departments. But the majority of Americans covered by commercial health insurance are not enrolled in insured HMOs but in PPOs and in self-insured products not covered or counted by state regulatory agencies.[1] The focus on HMOmarket shares, to the exclusion of insured PPO and self-insured products, sometimes reflects a mistaken view that HMO and PPO product designs, and insured and self-insured funding arrangements, do not compete with one another.

To assess the extent of concentration in the commercial health insurance industry, it is necessary to amalgamate data from multiple sources. Firm-specific data are available for the Blue Cross and Blue Shield plans, both nonprofit and for-profit, and for the investor-owned commercial plans, and state-specific enrollment data are available for insured products (which includes most nonprofit HMOs). Commercial health plan enrollment (employment-based and individually purchased) must be separated from noncommercial lines of insurance offered by the same carriers, including Medicare, Medicaid, and the military TRICARE managed care programs. Commercial coverage does include public employee health benefits, both insured and self-insured, such as the Federal Employees Health Benefits Program (FEHBP) and state public employee programs. The following discussion relies on data compiled by Goldman Sachs Global Equity Research, based on state regulatory filings; investor reports from publicly traded firms; information from nonprofit BCBS plans; the InterStudy HMO Directory (based on state regulatory filings); the InterStudy PPO Performance Report (based on surveys by InterStudy); and direct contacts with individual firms. The figures for the investor-owned plans were last updated as of December 2003; some nonprofit plan data are from 2002.[2]

Data are available at the state level only, even though some states include multiple geographic markets and some geographic markets overlap state lines. Several states have multiple Blue Cross plans that operate in nonoverlapping parts of the state and do not compete directly with one another (except in border regions). Pennsylvania has four region-specific Blues plans; New York has three. Data are available separately for upstate and downstate New York, because of the radically different markets in New York City and the western part of the state (two Blues plans operate in upstate New York). The data for the District of Columbia include northern Virginia (suburban Washington, D.C.).[3] No data are available for Hawaii and Alaska; North Dakota data are not available for health plans other than BCBS. The ensuing discussion refers to fifty "states" but in fact includes forty-seven states, two parts of one state (New York), and one jurisdiction (District of Columbia, including northern Virginia).[4]

Table 1 presents total commercial insurance enrollment (including insured and self-insured funding arrangements) for each state market, plus the percentage held by the largest firm in the market, the percentage held by the three largest firms, the Herfindahl-Hirschman Index (HHI) of market consolidation used by the Federal Trade Commission (FTC) and U.S. Department of Justice (DOJ), and the extent of antitrust concern derived from FTC/ DOJ guidelines (based on the HHI).[5]

The striking feature of the numbers in Table 1 is the large market shares controlled by the leading firmin each state. In thirty-eight states the largest firmcontrols one-third or more of the market; in sixteen states the largest firm controls more than half the market. In all states except California and Nevada the largest insurer is a Blue Cross or Blue Shield plan, or both. The role of leading firms is underestimated in these figures, as some states have distinct geographic markets, with a dominant firm in each. For example, if the four regional Blue Cross plans in Pennsylvania were added together, the resulting firm would control 63 percent of the statewide "market" rather than the 33 percent shown in Table 1 .

The top three firms typically dominate each market, as indicated in the third column of Table 1 . In only three state markets do the largest three plans control less than 50 percent of the total enrollment, and in only fourteen do the largest three plans control less than 65 percent. The fourth column of Table 1 presents the HHI, which is calculated by squaring the market share of each firm in the market and summing across all competitors; the HHI is commonly used to analyze markets and sometimes to block mergers and acquisitions. As evident in Table 1 , the HHI for commercial health insurance at the state level is very high, with only three of the state indexes falling below 1,000 (the FTC/DOJ threshold for low level of antitrust concern), twelve falling between 1,000 and 1,800 (moderate level of antitrust concern), and thirty-four exceeding 1,800 (high level of antitrust concern).

Table 2 provides an alternative perspective on the consolidation of the commercial health insurance industry by highlighting the relative shares of each state market held by the four largest U.S. firms (WellPoint, United, Aetna, and CIGNA) and by the BCBS plans that are not part of WellPoint. The Blues, including Well-Point, hold the largest market share in every state except Nevada (where they tied with Sierra Health Services for first) and California, where Blue Cross of California (WellPoint) is tied for first (with Kaiser Permanente) and Blue Shield of California comes in third. While the state-specific BCBS plans formally are owned by forty independent companies, some covering multiple states and others just one, they cooperate with one another in the market for multistate corporate accounts and in building the Blue brand nationally. If all Blues plans were considered part of one firm, they would control 44 percent of the national market. In comparison, if all nonprofit Blues were considered one firm (thereby excluding for-profit Well-Point and WellChoice), the nonprofit plan would control 31 percent of the national market.[6]

The consolidation of the industry at the hands of the largest health plans is evident in the far right column of Table 2 . The market shares of United, Aetna, and CIGNA are quite modest in most states, with exceptions where they have acquired regional HMOs, but the overall scale of the firms is large because of their wide geographic scope. Together the Blues plans and the three national non-Blues carriers control more than 60 percent of the market in thirty-four states and more than 70 percent of the market in twenty-three states. Regional nonprofit HMOs today remain strongest in the states where they launched three decades ago (California, Minnesota, Massachusetts, and Oregon). Regional for-profit plans such as PacifiCare, HealthNet, Humana, and Coventry have dispersed enrollment and do not dominate any single market.

The figures presented in Table 1 and Table 2 should be interpreted with caution, because of the partially distinct customer segments that make up the commercial insurance industry. Large, multistate employers typically purchase self-insured benefits administration from one or more of the national carriers listed in Table 2 or from a state-specific Blues plan linked to the national BlueCard network. The number and relative size of local health plans may be largely irrelevant.[7] The market for multistate employers in 2003 was served by Aetna (14 percent), CIGNA (9 percent), United (15 percent), WellPoint (12 percent), Anthem (5 percent), other Blue Cross plans (12 percent), First Health (8 percent), and other (25 percent).[8]