Health Affairs, 23, no. 6 (2004): 91-102 doi: 10.1377/hlthaff.23.6.91 © 2004 by Project HOPE	New Online   House Health Reform Bill   Paying for Reform   Vetting AHIP's Report   HIV/AIDS Costs   Brief: Insurance Reform   HA Blog Top 10

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Grossman, J. M. Articles by Ginsburg, P. B. Search for Related Content PubMed PubMed Citation Articles by Grossman, J. M. Articles by Ginsburg, P. B. Related Collections Health Reform Legal/Regulatory Issues Business Of Health Health Spending Insurance Market

Future Of Insurance

As The Health Insurance Underwriting Cycle Turns: What Next?

PROLOGUE: Business executives, financiers, and the markets abhor uncertainty as much as nature abhors a vacuum. The boom-and-bust pattern in the health insurance markets known as the "underwriting cycle" has been studied sufficiently enough to understand its basic characteristics. Historically, the underwriting cycle has consisted of three years of profitability followed by three years of losses. Higher profits attract new entrants and drive more intense price competition, which in turn leads to losses as premiums fall below costs. As plans exit the market, the remaining plans raise premiums above costs, driving a return to higher profits and a repeat of the cycle.

In the 1990s, though, something else happened, which sparked uncertainty about future industry profitability. Managed care’s ascendance suppressed prices in unexpected ways, but the consumer backlash and the market’s reactions made for unpredictable price increases later in the decade. The underwriting cycle appeared to break, with "longer and more uneven periods of gains and losses, and less extreme fluctuation in profitability," as Joy Grossman and Paul Ginsburg tell us.

Based on interviews with numerous industry experts, Grossman and Ginsburg conclude, however, that the underwriting cycle hasn’t gone away. But in the future, we will see one with swings that are even more muted than in the 1990s, as continuing consolidation makes it more likely that the industry will forgo heated price competition, and better forecasting tools allow it to set premium trends closer to cost trends. Premiums may be higher, but they won’t be as volatile.

Grossman (jgrossman{at}hschange.org) is a senior health researcher and Ginsburg is president of the Center for Studying Health System Change (HSC) in Washington, D.C. Grossman has been with the center since its founding in 1995. She holds a doctorate in economics from the University of California, Berkeley. Ginsburg, a frequent contributor to Health Affairs, is the founding president of HSC. He holds a doctorate in economics from Harvard.

In an accompanying Perspective, Alice Rosenblatt of WellPoint Health Networks offers six reasons why the underwriting cycle may soon be "a thing of the past" for most insurance companies.

Abstract

 
After twenty-five years of a consistent health insurance underwriting cycle, the pattern of insurer profitability changed greatly in the 1990s, raising speculation about the future. We conclude from interviews with industry experts that health plan competition and limits on plans’ ability to predict costs will continue to drive a cycle, albeit one even more muted than it was in the 1990s because of changes in industry structure and forecasting improvements. Plans will price closer to cost trends and forgo the more heated price competition that drove major losses in the past, reducing premium volatility but possibly leading to higher average premiums.

View larger version (53K): [in this window] [in a new window]   EXHIBIT 1 Health Insurance Industry Profitability: Underwriting And After-Tax Margins Of The Blue Cross And Blue Shield (BCBS) System And Major Public Managed Care Firms, 1965–2003

After a period of sharp increases in health insurance premium trends that began in the late 1990s, underlying health care cost trends have begun to decelerate, and industry and policy experts have been speculating about when premium trends might begin to decline and to what degree.¹ Some have suggested that the cycle, as we know it, is "becoming a myth," while others suggest that it is alive and well, albeit in a muted form, with differing predictions for how fast and how far premium trends will fall in relation to cost trends.²

Although Jon Gabel and colleagues documented and examined reasons for the cycle through the 1980s for both commercial health insurers and the Blues, we are aware of few studies that explore why the cycle changed in the 1990s and what the implications are for health insurance pricing in the future.³ Given the current speculation about where premiums are headed in the next few years, it is timely to revisit this issue.

Study Design

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
We interviewed twenty-one experts, including ten health plan executives representing actuarial, marketing, and senior management perspectives; three independent actuaries; four Wall Street securities analysts who monitor trends in industry profitability; two former state insurance regulators; and two economists who study health insurance pricing. To help understand variations in the underwriting cycle across local markets and to capture the perspectives of executives in different types of plans, we selected three local/regional nonprofit BCBS plans and three multistate for-profit plans (including one Blues plan).

Half- to one-hour telephone interviews were conducted during March and April 2004. Topics included factors driving the underwriting cycle historically, why the cycle changed during the 1990s, and views about the future. We also assessed available health plan financial data.

Drivers Of The Historical Underwriting Cycle

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
A large body of research shows that the underwriting cycle is present, with some variation, across many lines of insurance.⁴ Drawing on the many theories that have been developed to explain this phenomenon, our framework for analyzing the underwriting cycle involves distinguishing between the two steps that plans go through in determining changes in premiums: forecasting cost trends and specifying a premium change that will be either greater than or less than the projected change in costs, reflecting the opposing strategies of sacrificing market share to boost profitability or gaining share at the expense of profitability. Given the traditionally thin profit margins in the health insurance industry, these decisions can lead to substantial short-term fluctuations in profitability.

Consistent with previous studies, there was broad agreement among almost all of the experts we interviewed that the health insurance underwriting cycle before the 1990s was driven by two major factors that interacted to drive periodic gains and losses: the lag in adjusting premiums to reflect unanticipated fluctuations in cost trends; and the competitive dynamics of the industry, with competition being the stronger driver of the cycle.

Lags in correcting forecast error. Setting premiums requires predicting future cost trends. Before the 1990s, this task was made difficult by the cyclical fluctuation in underlying health care cost trends (Exhibit 2). These fluctuations were often associated with shocks to the health care system or the economy, often from federal policy changes.

View larger version (28K): [in this window] [in a new window]   EXHIBIT 2 Annual Percentage Change In Employment-Based Health Insurance Premium Underlying Health Care Spending, 1976–2003

Extrapolating from recent trends with dated information made it particularly difficult for plans to predict turns in trend. Adding to this, when plans were overly optimistic or pessimistic about cost trends, their bias persisted for some time; as a result, profitability was higher or lower than planned for during consecutive years. The effects of these forecasting errors on profitability were compounded by the fact that renewal dates were clustered in January, so that any errors identified too late for bids for January renewals carried over into another year.

Competition. When industry profitability exceeded the norm, this attracted new plans and led to efforts by incumbent plans to expand market share. As these plans priced their products below those of competitors to gain enrollment, this triggered price wars, which depressed profitability and eventually led to industrywide losses. If plans priced aggressively while cost trends—unbeknownst to them—were turning up, then the losses were magnified. The losses, in turn, caused plans to exit the market. With reduced competition, premium trends increased, leading to increased profitability and starting the cycle anew.

The industry’s historical low barriers to market entry facilitated these cyclical swings in profitability. In particular, commercial carriers, which typically sold other lines of insurance as well as health, could easily enter and exit the health market. These insurers, particularly property and casualty companies, make most of their profits from investment earnings rather than from underwriting gains because of the infrequent but large nature of the losses. For those also in the health insurance business, the fluctuations in the level of cash flow in their other businesses often led to expansion or contraction of their health insurance activities.

The pricing behavior of Blues plans also helped drive the underwriting cycle. The Blues, at the time, were all nonprofit or mutual insurers. These plans’ residual underwriting gains became part of accumulated reserves rather than being distributed to shareholders, as in the case of for-profit plans. After building up capital in excess of required reserves during periods of profitability, Blues plans would spend it down by lowering premiums or granting rebates on past premiums, sometimes in response to pressures from state insurance regulators. When capital reserves became depleted as a result of underwriting losses, plans would then raise premiums to build up reserves.

Other factors. Although state insurance rate regulation and fluctuations in investment earnings were cited as other factors influencing the underwriting cycle, there was no firm agreement among the experts we interviewed about how important they were.

Impact Of Managed Care On The Cycle

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
Elongated pattern. Respondents uniformly agreed that unique events surrounding the rapid growth of managed care mitigated the underwriting cycle during the 1990s. As Exhibit 1 indicates for the Blues, rather than predictable three-year phases in the cycle, the period of underwriting profits spanned six years, from 1989 to 1994, before moving to a four-year period of losses from 1995 to 1998. The underwriting cycle for public companies followed a similar elongated pattern.⁵

Exhibit 1 also shows that the swings in the cycle were not as pronounced during the 1990s as they were from the mid-1960s through the 1980s. The largest annual positive underwriting margin in the 1990s was 2.2 percent in 1993, compared with the maximum of 3.5 percent reported in 1967. Losses were even more muted, with the largest negative margin of –1.2 percent in 1997, compared with a maximum of –5.9 percent in 1987.

Cost trends. Underlying the elongated underwriting cycle in the 1990s was an elongated cost cycle that differed from the regular, more frequent swings in cost trend prior to the 1990s (Exhibit 2). During the first half of the 1990s, the rapid shift to restrictive managed care products, especially health maintenance organizations (HMOs), sharply reduced cost trends. This reflected reduced service use from more utilization management and increased use of capitation, as well as larger provider discounts. In the mid-1990s, however, the cost trend first stabilized and then turned upward in response to a number of factors, including the loosening of managed care, increased provider consolidation and pushback on payment rates and capitation, and rapid increases in prescription drug use.

Ability to predict trends. During the first half of the 1990s, because plans only slowly adjusted their actuarial predictions to reflect decelerating cost trends, premium trends lagged cost trends, and the industry remained profitable. However, few insurers accurately predicted the acceleration in cost trends in the mid-1990s, which, in combination with increasing price competition, resulted in a shift to the loss phase of the underwriting cycle. Plans had limited experience with predicting costs for managed care products, particularly the many new plans that entered the market in the mid-1990s. In addition, plans mistakenly believed that it was their plan’s utilization management and other cost control activities, rather than broader system and economic forces, that were driving the cost deceleration. The fact that some plans passed on capitated risk to providers during that period probably masked some of the forecasting error.

Toward the end of the decade, with the combination of additional managed care experience and the exit of most of the weaker plans from the market, the remaining plans had stronger management teams in place, setting the stage for a return to profitability. The plans also benefited from gradual improvements in the ability to collect and analyze data during the decade. Increased electronic processing of provider claims cut the three-month data lag in half or better. Some aspects of managed care contracting improved the available information. For example, the introduction of provider contracts with fixed prices (rather than percentages of charges), typically one to three years long, helped plans estimate costs better.

Competition. There was a rush of new entry that started in the early 1990s and peaked around 1995, as plans responded to opportunities for increasing their managed care enrollment and continued profitability.⁶ New entrants included provider-sponsored plans and the expansion of existing plans into new geographic markets—especially the national for-profit plans, which often entered new areas through acquisition.⁷

Consistent with classic underwriting cycle behavior, stiff competition ensued as plans lowered their prices to gain market share. Exhibit 2 shows a steep decline in premium increases through 1996. Several factors drove this push for market share: (1) the need to increase enrollment to extract larger provider discounts; (2) the advantages to early entrants in developing exclusive provider networks, which were gaining popularity at the time; (3) the building perception that the "top three" plans in terms of local market share were best positioned for profitability; and (4) in the case of for-profits, shareholders’ focus on enrollment growth as an indicator of long-term profitability.

A number of respondents concurred with findings from Center for Studying Health System Change (HSC) site visits at the time that some plans intentionally underpriced to gain market share.⁸ Plans believed that they could price lower than their competitors without harming their bottom line because of expectations that larger market share, along with utilization management, would reduce costs. With the industry in its growth phase, publicly offered plans were valued based on a multiple of members and so had additional incentives to price to increase enrollment. The heavy capitalization of these plans provided a cushion for losses arising from such pricing.

Because most plans did not correctly forecast the turn in trend as cost trends first stabilized and then accelerated in the mid-1990s, they ended up unintentionally setting premium increases too low. Plans that had underpriced to gain market share had even larger declines in profitability. Nevertheless, there were both publicly offered companies and nonprofit Blues plans that remained consistently profitable throughout the 1990s.

Consolidation. Starting around 1995, while new entry continued, the industry began consolidating as national plans merged to increase their concentration in local markets and take advantage of the ability to spread fixed costs over a larger number of covered lives. Some notable mergers during the period included United HealthCare with MetraHealth (1995); PacifiCare with FHP (1996); and Aetna Life and Casualty with U.S. Healthcare (1996), NYLCare (1998), and Prudential (1998). As part of industry restructuring, major multiline insurers separated their health insurance units from their other lines of business.

In the last few years of the 1990s, continuing losses on top of the declining interest in HMOs associated with the backlash against managed care drove a large number of plans from the market, which, in combination with the large-scale mergers of national plans, led to a more concentrated industry. Plans that exited included both smaller and provider-sponsored plans and local units of national plans that did not have sufficient market share to be profitable.

Many of the remaining plans shifted their strategies from gaining market share to improving current profitability, setting in motion a new round of premium increases that eventually moved the industry into a period of profitability by 1999 (Exhibits 1 and 2). Plans that had once been part of multiline companies with more diversified portfolios now had to focus on increasing health insurance profitability. More generally, public companies, including an increasing number of Blues plans, had to respond to investors’ priorities that shifted from enrollment growth to current profitability.

Blues’ responses. The Blues, as dominant players in many local markets, also participated in these two phases of the cycle during the 1990s. Reacting to financial instability in some Blues plans, the Blue Cross and Blue Shield Association (BCBSA) imposed stricter financial standards on member plans in the early 1990s, including higher reserve requirements. These standards helped limit how far plans could spend down surpluses in response to competitive pressures. In combination with the association’s decision to allow Blues plans to become for-profit, some experts suggested that the Blues, as a whole, began to focus on maintaining stable operating margins earlier in the decade. As the incumbent plans in local markets and having been slow to respond to the growth in managed care, however, the Blues plans found themselves responding defensively to new local and national entrants.

A New Phase For The New Millenium

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
The most recent five-year period, 1999–2003, was one of rising profitability (Exhibit 1) that took place in an environment of first increasing and then decreasing cost trends (Exhibit 2). Most plans missed the turn in the cost trend as it first stabilized in 2002 and then decelerated in 2003—or were unwilling to price to it without more data or a similar move by a competitor—resulting in unusually high margins in 2003 (Exhibit 1).

Improved ability to predict trends. Under pressure to improve financial performance, all of the plans in the study noted that by the early 2000s, they had instituted processes to gather more in-depth and timely information from the field on renewal pricing and provider contracts. Plans used these data, in combination with improved claims information, to monitor premium and cost trends in a much more detailed way and more frequently—quarterly or monthly, rather than yearly as in the past. As plans reduced the proportion of renewals done in January, they were then able to put pricing adjustments based on this information into effect sooner for some portion of their contracts.

Less aggressive competition. In contrast to previous periods of industry profitability, there was little entry between 1999 and 2003 because of a number of factors: (1) Existing plans were focused on improving profitability rather than growing market share; (2) barriers to entry were higher with managed care, including the need to develop broad provider networks while having sufficient market share to extract provider discounts and the need for capital investments in information technology; and (3) the exit of the commercial companies that sold multiple lines from the health insurance business removed a ready source of capital.

With few new entrants to stimulate aggressive price competition, executives in both the nonprofit Blues plans and the for-profit plans in this study reported that they and their competitors became more focused on pricing to improve profitability. As markets became more consolidated, plans probably were more likely to factor in expected responses by competitors in making pricing decisions.

A large proportion of the public companies had additional incentives to price conservatively. This was a turnaround period for plans that had suffered financially during the second half of the 1990s for reasons beyond pricing decisions, including reduced profitability from Medicare+Choice and acquisitions made on unfavorable terms. For example, Aetna culled unprofitable business by raising premiums sharply, which greatly reduced its membership but improved its bottom line.

How Well Does The Recent Past Predict The Future?

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
Near term. The analysts, along with plans that had larger-than-expected margins over the past few years, expected premium trends to decline more rapidly than cost trends in 2004 and 2005 and plan margins to contract somewhat from the 2003 high. They viewed this as an adjustment in pricing to bring margins back down to a "normal" rate of return in response to what they expected to be continuing moderate deceleration in cost trends.

Some local plans that had not yet experienced declining cost trends stated that they would continue to price to predicted trend to maintain their current margins and expected their competitors to do the same. Although these plans were expecting flat to slightly increasing cost trends in 2004 and 2005, in the event of a decline in cost trends in the future, they expected premium trends to moderate slowly as they "fine-tuned" pricing in response.

Four of the six plans did report a small amount of "aggressive" pricing behavior by competitors that were the first to lower premium trends. In most cases, the price cutting was not to the degree seen in the mid-1990s that "left you scratching your head." Nonetheless, two of the plans noted that they were carefully monitoring a plan that was actively attempting to expand into their market.

In the press, the most consistently noted example of a potential for a sharp turn in the underwriting cycle has been the reports that nonprofit Blues plans in at least eleven states have been reducing their pricing trends in 2004 or giving rebates on 2003 contracts as a result of very high margins in 2003.⁹ With both unanticipated cost-trend deceleration and increases in investment earnings as the stock market rallied, following on several years of increasing profitability, many Blues plans found themselves with large excess reserves in 2003.

Most experts did not see the Blues’ "give-backs" as a signal that nonprofit Blues as a whole are likely to forgo the goal of maintaining target levels of profitability. As nonprofits or mutuals, returning excess surplus to policyholders is consistent with the plans’ mission—not "irrational" behavior—particularly in light of the reduction in cost trend. In some states, the Blues are subject to either specific reserve regulations (for example, caps on reserves in Minnesota) or regulatory pressures to reduce reserves (for example, refusals to grant individual insurance rate increases in Pennsylvania). Discounts on 2004 premiums were viewed as more disruptive to the market than rebates on 2003 contracts because discounting future premiums requires plans to raise rates even more sharply at the next renewal period.

Two analysts warned that these "give-backs" could set in play even stronger price competition in some markets where nonprofit Blues plans have large market share, reducing industry profitability and driving a return to a more pronounced underwriting cycle in those markets.¹⁰ However, in general, most experts did not view any of the recent signs of renewed price competition by the Blues or other plans as an indication that a sharp turn in the underwriting cycle for the industry as a whole was imminent.

The future. Most study respondents were optimistic that pronounced underwriting cycles (particularly the sharp losses) were a thing of the past. They believed that the radical transformation of the industry structure that took place during the 1990s would make it less likely that plans would cut prices aggressively during times of high profitability and that plans’ improved ability to monitor premium and cost trends would allow them to price closer to trend. Many did temper this optimism by noting that "all business has cycles," and, given the difficulty in predicting the future, good intentions to price to trend may not be enough to prevent a cycle of gains and losses.

All agreed that probably the greatest risk to this scenario is external cost shocks, which unexpectedly reduce profits and introduce uncertainty about restoring them. When asked to identify potential shocks, the experts suggested scenarios such as a robust recovery from the recent recession or the introduction of new blockbuster medical technologies.

The introduction of a new product also could trigger a turn in the cycle, just as it did in the mid-1990s, when managed care reached a critical mass. Some respondents raised concerns that if consumer-directed products grow markedly, errors in pricing them or intentional underpricing by new niche players to gain share could set off new price wars. However, others believed that niche players are not much of a threat to insurers and that major insurers are being conservative in pricing these products.

Taking these perspectives into consideration, we expect the underwriting cycle to continue but to be even more muted than in the 1990s. We do not expect it to have the regularity or large losses of the historical cycle or the exact shape of the cycle through the 1990s, which was driven in large part by the growth and retreat of tightly managed care. We anticipate that the swings in gains and particularly losses will be smaller than in the 1990s because plans have more incentives to price to trend to maintain a stable level of profitability. With more reliable and timely data, they also might be able to take corrective action more quickly when they miss a turn in cost trends. These factors could also shorten the length of the cycle. With competitive dynamics playing less of a role in the underwriting cycle, external shocks to the health care system could dominate the overall shape of the cycle more so than in the past.

The period we are entering now will be a good test of our thesis. The period 1999–2003 was a time of adjustment, when the newly restructured industry faced greater pressures to restore profitability. Now that we have a mature industry with stagnant aggregate enrollment growth, it remains to be seen whether investors and plans will be satisfied with earnings growth from additional consolidation and from new technologies that reduce administrative costs and support care management techniques. Indeed, with most turnarounds completed, most public companies are seeking enrollment growth this year, which will be difficult to achieve in the aggregate. In many cases, public companies may have to decide whether to accept lower growth than forecasted or to price more aggressively to achieve their enrollment targets at the expense of profitability.

Policy Implications

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
The reduced volatility in pricing discussed above is generally a positive development for the major stakeholders in health care. More stability in pricing may lead to more stability in health insurance offering and take-up and to less tension between health plans and providers. However, a more muted underwriting cycle may be a negative development if it reflects an overall reduction in price competition among health plans because of consolidation and other structural changes in the industry. In that case, purchasers and consumers would pay higher premiums at all times. Beyond leading to higher profitability for insurers, a less competitive insurance market could reduce pressures on insurers to innovate. Another potential downside of a more muted cycle is that it could reduce the pressure on purchasers to focus on their long-term goal of reducing cost growth. Sharper cycles may have had the benefit of periodically galvanizing corporate executives to address these issues.

It will remain important to monitor and identify the various stages of the underwriting cycle. Such information can help sort cyclical from secular factors in premium trends and plan consolidation at the market level. This can provide guidance to policymakers in assessing a range of policy issues, beyond premium and cost trends themselves, including antitrust scrutiny of the effects of increasing plan consolidation on premium levels. Similarly, states’ regulatory reviews of Blues plans’ conversion and merger proposals often hinge, in part, on an assessment of the plans’ long-term financial viability. To make this assessment and to more accurately predict the effects of completed conversions and mergers on premiums, a deep understanding of the underwriting cycle is essential.

Editor's Notes

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 
The authors gratefully acknowledge the Robert Wood Johnson Foundation for its financial support. They also thank the following people for participating in the study: Brian Atchinson, James Bailey, John Bertko, Howard J. Bolnick, Charles Boorady, Matthew Borsch, Deborah Chollet, Gary Claxton, David C. Colby, John P. Cookson, Heyward Donigan, Jon Gabel, Roberta Goodman, Audrey Halvorson, Daniel Johnson, Alice Rosenblatt, M. Edward Sellers, William R. Shrader, Dwayne Ullsperger, Scott Warrior, and William C. Weller. Claxton, Goodman, the editors, and anonymous reviewers gave valuable input on earlier drafts of the paper.

NOTES

Top Study Design Drivers Of The Historical... Impact Of Managed Care... A New Phase For... How Well Does The... Policy Implications Editor's Notes NOTES

 

1. See, for example, B.C. Strunk and P.B. Ginsburg, "Tracking Health Care Costs: Trends Turn Downward in 2003," Health Affairs, 9 June 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.354 (23 June 2004); and S. Heffler et al., "Health Spending Projections through 2013," Health Affairs, 11 February 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.79 (23 June 2003).
2. For these contrasting views, compare M. Smith and D. Colby, "Anthem/Wellpoint Investor Presentation," November 2003, www.wellpoint.com/pdf/112003WLPATHInvesPresRev-114XPreVersFinal.pdf (23 June 2004), with M. Borsch and K.P. Huynh, Heading toward an Industry Downturn (New York: Goldman Sachs, 2004); and C. Boorady, S. Mong, and S. Soorya, Stage Set for Strong Late-Cycle Performance (New York: Citigroup Smith Barney, 2004).
3. See J. Gabel et al., "Tracing the Cycle of Health Insurance," Health Affairs 10, no. 4 (1991): 48–61; [Medline]J.R. Gabel and G.A. Jensen, "Can a Universal Coverage System Temper the Underwriting Cycle?" Inquiry 29, no. 2 (1992): 249–262; [Medline]J.P. Cookson, "Underwriting Gain and Loss Cycles," in Group Insurance, ed. W.F. Bluhm (Winsted, Conn.: ACTEX Publications, 2003), 849–862; and R. Kipp, J.P. Cookson, and L.L. Mattie, Health Insurance Underwriting Cycle Effect on Health Plan Premiums and Profitablity (Washington: Milliman USA, 2003).
4. See, for example, K.E. Thorpe, "The Medical Malpractice ‘Crisis’: Recent Trends and the Impact of State Tort Reforms," Health Affairs, 21 January 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.20 (23 June 2004); and R. Klein, "The Underwriting Cycle," in Encyclopedia of Actuarial Science, ed. J. Teugels and B. Sundt (Chichester, U.K.: Wiley, forthcoming).
5. Consistent financial time-series data on public companies are limited. While the study focuses on underwriting margins, only after-tax margins for 1992–2003 are available for public companies. Comparable data are provided for the Blues.
6. J. Gabel, "Competition among Health Plans," presented at Joint Federal Trade Commission/Department of Justice Hearings on Health Care and Competition Law and Policy, 23 April 2003, www.ftc.gov/ogc/healthcarehearings/doc/030423jongabel.pdf (18 August 2004).
7. See R.D. Feldman, D.R. Wholey, and J.B. Christianson, "HMO Consolidations: How National Mergers Affect Local Markets," Health Affairs 18, no. 4 (1999): 96–104, for a discussion of national and local plan entry and consolidation from 1994 to 1997.[Abstract]
8. J.M. Grossman, "Health Plan Competition in Local Markets," Health Services Research 35, no. 1, Part 1 (2000): 17–35.[Web of Science][Medline]
9. Atlantic Information Services, "Blue Cross and Blue Shield: Not-for-Profit Blues’ Strong 2003 Financial Results Could Invite Regulatory Scrutiny," March 2004, www.aishealth.com/ManagedCare/BluesNews/BLUNPResultsBringScrutiny.html (18 August 2004).
10. Borsch and Huynh, Heading toward an Industry Downturn; and Boorady et al., Stage Set for Strong Late-Cycle Performance.

                      What's this?

This article has been cited by other articles:

				M. Hartman, A. Martin, P. McDonnell, A. Catlin, and the National Health Expenditure Accounts Team National Health Spending In 2007: Slower Drug Spending Contributes To Lowest Rate Of Overall Growth Since 1998 Health Aff., January 1, 2009; 28(1): 246 - 261. [Abstract] [Full Text] [PDF]

				S. S. Wallack, C. P. Thomas, T. C. Martin, and A. Ryan Differences in Prescription Drug Use in HMO and Self-insured Health Plans Med Care Res Rev, February 1, 2007; 64(1): 98 - 116. [Abstract] [PDF]

				W. Encinosa and M. Hagan Introduction: AHRQ Research on Health Care Markets Med Care Res Rev, December 1, 2006; 63(6_suppl): 3S - 8S. [PDF]

				J. C. Robinson The Commercial Health Insurance Industry In An Era Of Eroding Employer Coverage Health Aff., November 1, 2006; 25(6): 1475 - 1486. [Abstract] [Full Text] [PDF]

				P. B. Ginsburg, B. C. Strunk, M. I. Banker, and J. P. Cookson Tracking Health Care Costs: Continued Stability But At High Rates In 2005 Health Aff., November 1, 2006; 25(6): w486 - w495. [Abstract] [Full Text] [PDF]

				P. B. Ginsburg Competition In Health Care: Its Evolution Over The Past Decade Health Aff., November 1, 2005; 24(6): 1512 - 1522. [Abstract] [Full Text] [PDF]