Texas Private Passenger Automobile Insurance Profitability 1990-1998 Conclusion

Rate-Regulated versus Non-Rate Regulated

Some have suggested that Texas automobile insurance consumers would pay lower premiums if the benchmark with flexibility band rating system were eliminated and insurers were no longer subject to any rate oversight by the Texas Department of Insurance. A Progressive Insurance official has claimed that if automobile insurance rates were deregulated and if all insurers did what Progressive does, Texas consumers would save $500 million annually.

Table 7 shows that Progressive County Mutual – a non-rate regulated county mutual – has paid out much less of the premium dollar than rate-regulated insurance companies. Clearly, if every insurer did as Progressive does, then Texas automobile insurance consumers would have paid hundreds of millions more in insurance premiums. In fact, while the Texas Commissioner of Insurance has determined that a rate of return of around 11.5% to 12% is reasonable for Texas private passenger automobile insurance, Progressive has earned annual returns on equity from 17% to 19.3% countrywide from 1996 to 1998.

Comments on the Bickerstaff & Whatley / State Farm Profit Study

In June 1998, the firm of Bickerstaff and Whatley (B&W) issued a report, commissioned by State Farm Insurance, responding to the Center for Economic Justice’s March 1998 report on Texas private passenger automobile insurance profitability. B&W argue the following:

1. Longer time periods should be used to measure profitability;
2. Comparisons of paid-to-written and incurred-to-earned loss ratios are inappropriate;
3. Current rate level adjustments (e.g., January 1, 1998 benchmark rate changes) were not adequately considered; and
4. The CEJ report included two mathematical errors.

B&W argue that a ten-year period is necessary to evaluate auto insurance profitability and that, when viewed over a ten-year period, Texas auto insurance has not been excessively profitable. The B&W arguments fail on several grounds. First, in the real world, profitability is not measured over a ten-year period for any purpose. Investors look to recent profitability. Ratemaking is based upon the most recent two or three years of experience. Insurers’ decisions to enter or exit markets are based on less than ten years experience. Regulators would not review ten years of loss experience to evaluate an insurer’s solvency. The Texas Legislature, in establishing mandatory tort reform rate reductions did not reference direct the Commissioner to look at ten-year historical profitability in determining the necessary rate reductions.

Second, B&W are incorrect to claim that insurers have earned "below reasonable" returns on Texas automobile insurance from 1988 to 1997. B&W fail to consider the excessive reserves included in the incurred results over that period, as described above and shown in Table 4 of this report. B&W’s effort to demonstrate that paid-to-written and incurred-to-earned loss ratios cannot be compared over time is also in error, as demonstrated by their own Exhibit 3 hypothetical model. B&W’s model shows lower paid-to-written loss ratios than incurred-to-earned in the early years and then lower incurred-to-earned loss ratios than paid-to-written in the later years. But the B&W model shows, as hypothesized in the CEJ reports, that over time the excesses the incurred-to-earned loss ratio should be matched by excesses of paid-to-written loss ratios in later years. The actual Texas experience shows incurred-to-earned loss ratios exceeding paid-to-written loss ratios for many years and by significant amounts. There is no actual matching period of higher paid-to-written loss ratios. It is only in the past few years that paid-to-written loss ratios have exceeded the incurred-to-earned loss ratios and then by only a few percentage points.

The 1998 experience has shown that B&W were demonstrably wrong to argue that the CEJ analysis did not fully consider the modest rate reductions filed by insurers following the January 20, 1998 benchmark rate change. Contrary to the B&W arguments, Table 1 and 2 of this report show that Texas automobile insurance premiums and profitability were substantially excessive in 1998. Table 1 shows that the overall 1998 loss ratio was only 61.5% and Table 2 shows that premiums were excessive by $900 million compared to premiums levels based upon reasonable loss ratios. CEJ’s analysis in early 1998 – that the modest rate decreases filed by insurers were insufficient and that deeper rate reductions were necessary – was accurate. It should be further noted that the Texas Department of Insurance challenged many insurers’ early 1998 rate filings, resulting in many insurers filing additional rate reductions later in 1998. The excessive premiums shown for 1998 in Table 2 reflect even those additional rate reductions. Given the modest rate decreases filed by top insurers in early 1999, current rate levels continue to be excessive.

Finally, B&W did identify one mathematical error – the handling of service fees – that had a minor impact on the original analysis and is corrected in this report. B&W’s allegation of a second error is incorrect – both allocated and unallocated loss adjustment expenses were, and are, correctly considered in the calculation of excessive premiums.