Table of Contents | Features | News & Trends | Departments | Experts | Classifieds

The myth of the litigation crisis
July 2006 | Volume 42, Issue 7

Insuring against the next insurance crisis

If states change their insurance laws and adopt objective standards for rate increases, they can keep the next insurance crisis—and the next campaign to limit liability—at bay.

Jay Angoff

The recent insurance crisis is now over: Property/casualty insurance profitability reached a new all-time high of $43 billion in 2005, even after accounting for losses caused by Hurricane Katrina.¹

The recent shift in the insurance cycle was predictable. Over the past three decades, periodic campaigns to limit liability have been based on wild swings in rates. The cycle also turned following the insurance crisis of the mid-1980s and that of the mid-1970s, but the turn in the cycle today is even more dramatic.

Now, perhaps legislators will question the wisdom of enacting liability-limiting legislation designed to further increase the profitability of an industry already enjoying record profits. In the long run, however, the only way to prevent campaigns to limit liability is to eliminate the conditions that permit insurance rates to rise dramatically and suddenly, since those are the conditions that enable campaigns to limit liablity to take hold.

Little can be done to change some of these conditions—such as fluctuating investment returns and reinsurance rates—but others can be addressed. These include actuaries’ essentially unlimited discretion in calculating rates; insurers’ ability to accumulate unlimited surplus; and the inability of both state insurance departments and private parties to obtain a meaningful remedy when insurers charge excessive rates or engage in unfair practices.

This article both explains current law in those areas and proposes changes in the law that would minimize the likelihood that insurance rates will rise irrationally again—and thus spark another campaign to limit liability—sometime around 2015.

Actuaries’ discretion

Virtually all states’ insurance codes provide that rates may not be excessive, inadequate, or unfairly discriminatory. However, the extent to which states regulate medical malpractice rates—that is, ensure that they meet the statutory standard—varies significantly.

Some states require insurers to obtain the insurance commissioner’s approval before they can implement rate increases; some require insurers to file their proposed rates with the commissioner and then wait a certain number of days before implementing them; and some allow insurers to implement rate increases at will but permit the commissioner to disapprove them after they take effect if the commissioner finds them excessive.

Finally, some states both permit insurers to implement rate increases at will and define rates in a competitive market as per se nonexcessive. Because under the statutory standard the medical malpractice market has never been held to be noncompetitive, such statutes—as a practical matter—preclude the commissioner from disapproving rate increases even after they take effect.

Nevertheless, in almost all states, including those in which the commissioner has no practical authority over rates, insurers submit to the commissioner’s office so-called rate filings that purport to justify the rate increase they want to implement. Such filings contain data on the insurer’s past and projected claims payments. They also contain the assumptions the company’s actuary makes and applies to such data to estimate the insurer’s ultimate liability for claims covered by policies written in the coming year, and thus to estimate the price the insurer should charge for those policies.

Unfortunately, both insurers and their actuaries convey the impression that the rate-making process is a lot more scientific and exact—and a lot less dependent on assumptions, estimates, and subjective judgment—than it really is. Actuaries have broad discretion in such areas as loss development, trend factor, target rate of return, and investment yield. The different assumptions they can make in these areas affect the level of the rate increase the insurer proposes. Reasonably limiting their discretion should both reduce rates over the long run and eliminate wild swings in rates.

Loss development. The ultimate liability for a given year’s policies is estimated through the process of loss development. The actuary estimates this liability by looking at the claims payments covered by prior years’ policies that were made in different periods. He or she then assumes that the claims covered by the new year’s policies will follow the same payout pattern.

No rules govern either the number of prior years’ data that the actuary must consider in estimating the loss-development pattern for the current year or the manner in which he or she must weight each previous year’s data. For example, the actuary could consider the loss-development pattern of the previous three, five, or more years; he or she could also calculate a weighted average or a straight average, use the mean or the median, and throw out the high and the low number in calculating any average. (For example, in calculating a five-year average, he or she could throw out either the highest or lowest number among the five.) The actuary’s discretion thus enables him or her to project higher developed losses than is reasonable.

States could establish an actuarial standard to address this problem. For example, the legislature or the commissioner could require actuaries to weight the three most recent years of data more heavily than data from earlier years in determining loss-development factors if older data would produce higher losses. Actuaries also could be required to use a weighted average rather than a straight average if the latter would give more weight to higher-loss years.

Trend factor. The trend factor is the assumed rate at which the average claim amount (or “severity”) and the number of claims payments (or “frequency”) will increase or decrease in the future. Actuaries determine the trend factor by looking at the rate at which frequency and severity have been increasing or decreasing. The actuary can produce an unreasonably high trend factor—and thus an unreasonably high rate—by heavily weighting years during which claims payments increased substantially, while giving little weight to years in which claims payments decreased.

In medical malpractice, for example, claims payments generally have been increasing over the last 15 years, but they have been declining in real terms— increasing by less than the rate of inflation—during the last 5 years.² An actuary wishing to justify a high rate will use a trend factor based solely or primarily on the trend in losses over the last 15 years and will disregard or give little weight to more recent data indicating that losses have been declining.

As with loss development, the number of years the actuary considers in calculating a trend factor, and the weight he or she gives to different years, can have a dramatic effect on the size of the trend factor, and therefore on the rate for the current year. In fact, as the Rhode Island Department of Business Regulation has noted, one point of trend can make a difference of as much as 10 points in the rate indication.³ (An increase in the trend factor from 1 percent to 2 percent could raise the “indicated” rate increase from 10 percent to 20 percent.)

For example, assume that losses have been increasing by 8 percent a year on average over the last 15 years, by 6 percent a year on average over the last 10 years, and by 1 percent a year on average (less than the rate of inflation) over the last 5 years. If the actuary used the 8 percent, 15-year trend in calculating the necessary rate increase for the coming year, that increase could be as much as 70 points higher than it would be if he or she used the 1 percent trend that the most recent 5 years of data showed.

Because of the virtually unlimited discretion actuaries currently have in selecting trend factors, rates can go up even though claims are going down. To address this issue, a standard could require that the actuary give at least as much weight to the trend indicated by the most recent five-year period as to that indicated by longer periods, if the more recent data produced a lower trend and thus a lower rate.

Target rate of return. Insurers are constitutionally entitled to earn a fair rate of return.⁴ But state insurance departments generally do not specify a single return level or a range of returns as “fair,” and insurers have used rates of return in rate filings that are as low as 4 percent and as high as 20 percent.

As long as the insurer provides some support for the rate of return it is using—such as an opinion from an economist that its target rate of return is reasonable—the insurance department generally accepts it. So, as a practical matter, insurers have substantial discretion to say what a fair rate of return is—and thus discretion to charge a rate high enough to produce that return.

States could develop a standard under which a rate of return above a certain level would be deemed or presumed to be excessive. Because high rates are needed to produce high rates of return, a standard governing rates of return would effectively reduce rates.

Investment yield. Insurers must set their rates at a level that enables them to earn a fair rate of return after accounting for investment income. Like the trend factor, the investment yield the insurer assumes in its rate filing depends on the number of years the actuary considers in calculating it and the weight he or she gives to each year. Also, as is the case with trend factor, the average investment yield during the last 5 years has been substantially lower than the average yield over the last 10 years.⁵

Although using more recent trend data produces a lower rate, using an insurer’s recent investment yield produces a higher rate. To maximize its rate, therefore, an insurer has to rely on old data on loss trend, but recent data on investment yield. Such an approach would appear to be driven not by any principled rationale but rather by a desire to justify higher rates. No standards currently prohibit this approach; states should develop them.

If state insurance departments adopted objective standards that actuaries were required to follow in calculating rates, rates would decline over the long run, since actuaries tend to be overly conservative in their projections. Moreover, periodic dramatic increases in rates would be eliminated, since the only way to justify such increases is to use unrealistic, ultraconservative assumptions in several areas in the same rate filing, as insurers have been doing in malpractice rate filings in recent years. Promulgating objective standards would provide much-needed guidance to insurers and would make the current rate-making process much less arbitrary.

Maximum surplus standards

An insurer’s surplus is the extra cushion it holds above the amount it has set aside in reserves to make its estimated future claims payments. A company increases its surplus when it earns a profit and declines to distribute that profit to its shareholders (in a stock company) or policyholders (in a mutual company).

Regulators have the dual responsibility of making sure that insurance companies remain solvent and keeping rates at reasonable levels. There is a tension between these two responsibilities.

On one hand, to the extent that the regulator is responsible for ensuring a company’s solvency, there can be no such thing as “excessive” surplus, since the more surplus a carrier has, the more certain it is that the carrier will be able to pay its claims—even under the most unlikely and catastrophic circumstances. On the other hand, because the regulator is responsible for keeping rates at reasonable levels, it is also the regulator’s job to limit rate increases and, as a result, the insurer’s surplus.

To enable the regulator to carry out the former responsibility, the National Association of Insurance Commissioners (NAIC) has promulgated risk-based capital (RBC) standards, which virtually all states have adopted.⁶ (New York is the major exception.) They include a formula that calculates a specific minimum level of surplus each insurer must hold, based on the risk the insurer assumes and the types of assets it holds.

However, the RBC standards do nothing to help the regulator keep rates at reasonable levels, since they do not specify any maximum amount of surplus an insurer may hold. The NAIC should develop such a standard, since there must be some point at which the accumulation of additional surplus provides no meaningful benefit to policyholders.

Each insurer discloses in its annual statement both its actual surplus and the minimum surplus the NAIC requires it to hold. The former divided by the latter is called the RBC ratio. The NAIC collects and compiles the actual and required surpluses of all carriers annually, and it publishes average RBC ratios by size of carrier.

The NAIC could promulgate maximum surplus standards based on these ratios, and on the experience of insurers with very high RBC ratios. State insurance commissioners could then either require that any surplus exceeding those maximums be distributed to policyholders in the form of a dividend, or that an insurer whose surplus exceeds the maximums forgo any rate increase while its surplus remains excessive. Surplus exceeding what is reasonably necessary to protect policyholders would be returned to them in the form of either dividends or lower rates.

Refund authority and private challenges

Consumer protection was not the primary reason that states adopted insurance-rating laws. Rather, the laws were intended to supply the regulation required by the McCarran-Ferguson Act to immunize insurers from an antitrust challenge. McCarran-Ferguson provides that the antitrust laws “shall be applicable to the business of insurance to the extent that such business is not regulated by state law;”⁷ in order to trigger the exemption, therefore, it was necessary for those states to enact some regulation.

Notably, McCarran immunity applies as long as the state enacts some regulation of the industry, even if that regulation is minimal and even if the state doesn’t enforce it. In particular, less regulation is required to immunize anticompetitive activity under the McCarran exemption than to immunize such activity under the traditional antitrust “state action” doctrine. Under that doctrine, anticompetitive activity is immunized if the state has a “clearly articulated and affirmatively expressed” policy of substituting regulation for competition, and the state “actively supervise[s]” that policy.⁸

In contrast, to trigger McCarran immunity, the state need only enact legislation dealing with the regulated industry.⁹ Whether the state intended to substitute regulation for competition, and the extent, if any, to which the state supervises that policy, is irrelevant. State insurance regulation can thus immunize insurers from antitrust prosecution without offering any meaningful consumer protection.

State insurance regulation also limits the liability of insurers in at least three ways. First, in most states it permits insurance commissioners to penalize insurers only prospectively; it does not permit them to order refunds.¹⁰ Second, it contains only nominal penalties. Third, it establishes a cumbersome, multistage process that regulators must go through to obtain even the limited prospective relief authorized by statute.

In addition, insurers have increasingly used state regulation as a shield in litigation. For example, because insurance departments typically have no authority to order monetary relief, insurers that have been sued for alleged insurance code violations routinely defend on the grounds that the dispute belongs before the department. Exhaustion of administrative remedies, exclusive jurisdiction, primary jurisdiction, and the filed-rate doctrine¹¹ are all variations of this defense. Their purpose, of course, is to force the plaintiff to litigate in a forum in which it is impossible for him or her to obtain any meaningful relief.

Insurers’ ability to irrationally increase rates and to retain the benefits of those increases could be reduced, if not eliminated, if the law were changed. First, state insurance commissioners should be given the authority to order refunds to those adversely affected by insurer violations of the insurance code, including violations of rating laws (such as charging excessive rates), violations of unfair insurance practices acts, and violations of specific prohibitions on the use of certain rating factors.

Today, in the majority of states, either the unfair insurance practices act expressly provides that there is no private right of action under that act¹² or the statute is silent and courts have found that there is no such private right.¹³ Many states do permit private plaintiffs to challenge violations of the unfair insurance trade practices act under their general consumer protection act.¹⁴

However, not all unfair insurance practices can be easily characterized as violations of the more general consumer protection act. It is therefore important that states enable private parties to challenge violations of the unfair insurance trade practices act directly.

Finally, insurance rating laws should be amended to permit interested parties to challenge proposed rate increases in adjudicative hearings, and to cross- examine witnesses and submit evidence and argument. By challenging the assumptions insurers rely on in their rate filings and showing that those assumptions are unreasonable, interested parties could demonstrate that proposed rate increases are excessive and prevent them from taking effect.

By establishing objective actuarial standards and maximum surplus standards, states could both reduce rates and eliminate wild rate swings. By empowering regulators to issue refunds and consumers to challenge excessive rates and unfair practices, states could maximize insurer compliance with both their filed rates and the insurance laws.

Such changes in the law will not prevent those seeking to limit their liability from campaigning to do so. Such changes should, however, substantially reduce the likelihood that they will succeed with liability limits.

Jay Angoff, a former insurance commissioner of Missouri, is of counsel at Roger Brown and Associates in Jefferson City, Missouri.

back to top

1. Robert P. Hartwig, Ins. Info. Inst., Commentary on Full-Year 2005 Results 11 (Apr. 18, 2006), available at http://server.iii.org/yy_obj_data/binary/753200_1_0/Year%20End%202005%20Commentary.pdf (last visited May 23, 2005).

2. See Jay Angoff, Falling Claims and Rising Premiums in the Medical Malpractice Insurance Industry, July 2005, available at www.centerjd.org/ANGOFFReport.pdf (last visited May 23, 2006).

3. In re Norcal Mut. Ins. Co., Medical Malpractice Insurance, DBR No. 03-I-0218, Tr. at 227:14-18 (Nov. 10, 2003).

4. Federal Power Comm’n v. Hope Natural Gas Co., 320 U.S. 591 (1944).

5. See, e.g., NAIC Reports on Profitability by Line by State, 1995-2004 (available for purchase at www.naic.org).

6. E.g., MO. REV. STAT. §375.1250 (2006).

7. 15 U.S.C. §1012(b)(2001).

8. California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97, 105 (1980).

9. See, e.g., Federal Trade Comm’n v. National Cas. Co., 357 U.S. 560, 564-65 (1958).

10. A small minority of states do already have refund authority; Colorado is one. See COLO. REV. STAT. 10-4-418(4)(a)(2006).

11. The filed-rate doctrine was originally established to immunize railroad tariffs that had been filed with and approved by the Interstate Commerce Commission from private challenge. (Keogh v. Chicago & Northwestern Ry., 260 U.S. 156 (1922).) In recent years, however, insurers have sought to expand the filed-rate doctrine to apply to insurance rates; they have also sought to expand it to immunize both rates that have been approved and those that have not and to immunize rating factors as well as rates. For a particularly thoughtful explanation of the filed-rate doctrine, see the opinion of Judge Kimba Wood in Wegoland, Ltd. v. NYNEX Corp., 806 F. Supp. 1112, 1114 (S.D.N.Y. 1992), aff’d, 27 F.3d 17 (2d Cir. 1994).

12. E.g., MO. REV. STAT. §375.930 (2006).

13. E.g., Wilder v. Aetna Life & Cas. Ins. Co., 433 A.2d 309 (Vt. 1981); Strack v. Westfield Cos., 515 N.E.2d 1005 (Ohio Ct. App. 1986).

14. See, e.g., Indus. Indem. Co. v. Kallevig, 792 P.2d 520 (Wash. 1990); Miller v. Nationwide Mut. Ins. Co., 435 S.E.2d 537 (N.C. Ct. App. 1993).

Table of Contents | Features | News & Trends | Departments | Experts | Classifieds
Frequently Asked Questions about TRIAL | Past Issues of TRIAL

Send your comments and questions about the online version of TRIAL to us at trial@justice.org