Printer Friendly

The effect of Proposition 103 on insurers: evidence from the capital market.

The Effect of Proposition 103 on Insurers: Evidence from the Capital Market

Introduction

On November 8, 1988, California voters passed an insurance-reform ballot initiative known as Proposition 103. In essence, the initiative was a populist response to rising automobile insurance premiums in a state where property-liability insurance rates were already among the highest in the nation. Proposition 103's most controversial provision mandates a minimum 20 percent reduction in property-liability insurance premiums from their November 1987 levels. An exemption from the rate cut and any future rate increases were to be permitted only if an insurer was substantially threatened by insolvency. On May 4, 1989, the California Supreme Court upheld the constitutionality of the rate rollback but overturned the insolvency provision and ruled that insurers were entitled to an adequate return. Comments in the financial press indicate that Proposition 103 has evoked a considerable debate about its effect on the property-liability insurance industry.

Opponents of Proposition 103 have deemed it to be one of the most expensive political defeats any American industry has faced. This view is based on several features of Proposition 103 (see Gross, 1989; and Schmitt and Steptoe, 1988). First, the 20 percent rollback in premiums is estimated to produce an annual loss of $6 billion for the insurance industry. Second, revenues earned by the property-liability industry are further reduced because Proposition 103 introduces more competition into the insurance market by allowing Califoria banks to act as insurance brokers or agents. Third, because of the proposition, the insurance industry is no longer exempt from the state's antitrust laws. This would effectively curtail the widespread practice of exchanging and pooling actuarial and other data. Insurers contend that exchanging information is useful in setting accurate and fair rates. Moreover, they argue that removal of the exemption threatens the soundness of medium-size companies which depend on pooled actuarial data to make sound estimates of risk.

An alternative view expressed by some groups is that the regulatory changes instituted by Proposition 103 may ultimately benefit the insurance industry. For example, some industry experts argue that by injecting the adequate return standard into Proposition 103, the decision rendered by the state high court is favorable to insurers in California. As most California auto insurers claim to be earning little or no return in the state (see Hill and Celis, 1989), the adequate return standard provides a loophole that may largely negate the 20 percent rate reduction provision. Moreover, some observers believe that under Proposition 103 insurance rates will increase rather than fall if insurers can demonstrate that their current rate structure deprives them of an adequate return. (1) According to Standard and Poor's, the adequate return clause was a positive development for insurance firms operating in California. Indeed, as a result of the court decision, the rating agency removed two large California insurers, Safeco and Allstate, from its credit-watch list (see Schmitt and Wells, 1989).

While the debate on the effects of Proposition 103 on the insurance industry continues, no empirical evidence has been presented on how the California initiative affected the shareholders of property-liability insurers. This study examines share price reactions of property-liability insurers around key dates surrounding Proposition 103. The findings indicate that Proposition 103 is viewed by the capital market as an unfavorable development for insurers.

Data and Research Methods

A preliminary sample of firms most likely to be affected by the consequences of Proposition 103 is identified from the list of "The Two Hundred Largest American Property-Casualty Companies" published in Best's Key Rating Guide (1988) (2) and from articles discussing the California initiative in the Wall Street Journal (WSJ). The sample is restricted to property-liability firms (or their parent companies) listed on the New York Stock Exchange (NYSE) or the American Stock Exchange (ASE) and to firms trading in the Over-The-Counter (OTC) market with closing prices recorded in the "NASDAQ National Market Issues" pages of the WSJ. (3) This restriction limits the sample to securities having sufficient trading volume to ensure the availability and accuracy of daily stock prices over the periods examined. The final sample consists of 42 firms of which 17 are identified by Standard and Poor's, Moody's, or the WSJ as having a large exposure to the California property-liability insurance market. (4) For comparison purposes, a sample of non-insurance firms is constructed by randomly selecting 25 firms from a list of non-financial firms on the 1988 Daily Returns File of the Center for Research in Security Prices.

Event test methodology is used to evaluate the market response to the passage of Proposition 103 in the November 8, 1988 election and to the May 4, 1989 California State Supreme Court decision on its constitutionality. (5) The election and the court decision establish day zero in the event tests. Abnormal returns are calculated for 61 trading days surrounding each event. The abnormal return (AR) for each security i on day t is the deviation of security i's realized return ([R.sub.it]) from an expected return generated by the market model:

[Mathematical expressions] (1)

The coefficients [[alpha].sub.i] and [[beta].sub.i] are the OLS estimates of security i's market model parameters obtained from an estimation period of 110 consecutive trading days beginning 140 trading days prior t the event. The S & P 500 index is used to measure the market return over day t ([R.sub.mt]).

For a sample of N securities, the average abnormal return (AAR) on day t is calculated as:

[Mathematical expression] (2)

Abnormal performance is also evaluated over specified intervals of trading days. The cumulative average abnormal return (CAAR) over an interval beginning on day [T.sub.1] and ending on day [T.sub.2] is given by:

[Mathematical expressions] (3)

To control for heteroskedasticity, test statistics are based on standardized abnormal returns. The standardized abnormal return (SAR) for security i on day t is: (6)

[SAR.sub.it] = [AR.sub.it./S.sub.it] (4)

where

[Mathematical expression] (5)

In the above equation, [V.sub.i.sup.2], is the residual variance of security i's market model regression, ED is the total number of days in the regression's estimation period, [R.sub.m] is the average market return over the estimation period.

To test the hypothesis that the average standardized abnormal return for day t is equal to zero, the following t-statistic is computed:

t = [ASAR.sub.t./SD] (6)

where

[Mathematical expressions] (7)

is the average standardized abnormal return for N securities across day t. To control for possible cross-sectional correlation induced by the clustering of industry returns in calendar time, a standard deviation (SD) based on a time series of average standardized abnormal returns is calculated:

[Mathematical expression] (8)

where [Mathematical expression] (9)

To test the hypothesis that the average cumulative standardized abnormal return over a specified interval of trading days is equal to zero, the following t-statistic is computed:

t = [ACSAR.sub.T1, T2./SD]

where

[Mathematical Expression Omitted]

is the average cumulative standardized abnormal return (ACSAR) from day [T.sub.1] to day [T.sub.2]. The t-statistics in (6) and (10) have 109 degrees of freedom.

Empirical Results

Table 1 presents average abnormal returns (AAR) and cumulative average abnormal returns (CAAR) measured around the November 8, 1988 vote on Proposition 103 in the California election. Abnormal stock price performance is reported for three samples: the base sample of 17 property-liability insurers with a large business exposure to the California market; a comparison sample of 25 other property-liability insurers; and a second comparisons sample of 25 randomly selected non-insurance firms.

As shown in the footnotes of Table 1, favorable pre-election poll results and Proposition 103 precipitated events surround the election date. Thus the totality of the market's response may best be captured by examining abnormal stock price performance across intervals of trading days surrounding November 8. Cumulative average abnormal returns are estimated for a day t = -1 to day t = 1 interval and for a wider day t = -10 to t = 10 interval. Across both intervals, insurers record negative CARRs that are significant at the 0.01 or better level. Over the day t = -10 to day t = 10 interval, the CAAR for 17 insurers with a large California exposure is -3.433 percent (t-statistic = -2.81) and -3.465 percent for the 25 other insurers (t-statistic = -2.58). Over 75 percent insurers are associated with negative abnormal returns. Whereas the difference between abnormal returns of the two insurance samples is not statistically significant, the abnormal return for each insurance sample is significantly different, at the 0.01 level, from the abnormal return for non-insurance firms.

The abnormal stock price behavior around the May 4, 1989 court ruling on Proposition 103 is presented in Table 2. Event tests around May 4 are cleaner than those for the November election because the noise created by the campaign issues of the 1988 election is absent. In addition, the court decision cleared by any uncertainties concerning the provisions of Proposition 103. A large stock price reaction is recorded on May 4 by insurers with a large California exposure. The AAR for these 17 firms is -3.286 percent (t-statistic = -4.43) with 88 percent of the firms posting negative abnormal returns. The AAR for the other insurers on May 4 is negative but not statistically significant. However, significant abnormal stock price performance is captured for these 25 firms over the day t = -1 to day t = 1 trading day interval surrounding May 4. Over this interval, the other insurers record a CAAR of -1.844 percent (t-statistic = -2.05) with 84 percent of these firms reporting negative abnormal returns. [7] The portfolio of non-insurance firms reports no significant abnormal price performance on May 4 or over the trading day intervals surrounding May 4.

The evidence from the event tests show a negative market reaction to the election day approval of Proposition 103 and to the California Supreme Court decision. The systematically negative abnormal returns to the portfolios of insurers can be interpreted as an unfavorable assessment by the market of Proposition 103's impact on the insurance industry. The adequate return criteria resulting from the court decision did not alter the market's negative assessment. The results around the court decision may reflect more the expected impact of the total package of insurance reform in California rather than its main feature, the 20 percent rate cut. Moreover, the events in California had been closely followed in other states where politicians and consumer groups are considering pushing for similar insurance regulation. The significantly negative abnormal returns recorded by the 25 other property-liability insurers supports the expectation of populist insurance reforms similar to Proposition 103 being adopted in other states.

Summary and Conclusions

Considerable debate has arisen about the effect of Proposition 103 on the insurance industry in California. Insurers claim that the large cuts in rates mandated by Proposition 103 would substantially reduce revenues of the property-liability insurance industry. Others point out that regulation of the California insurance market may have positive effects. In particular, the adequate return guarantee inserted into Proposition 103 by the California Supreme Court may ultimately increase rates.

This study examined the effect of Proposition 103 on the common stock values of property-liability insurance firms. The results indicate that shareholder wealth declined in response to both the passage of Proposition 103 in the 1988 election and to the California Supreme Court's decision to uphold its major provisions. These findings suggest that the capital market perceives Proposition 103 to be an unfavorable development for property-liability insurers.

(1) In a representative expression of this view, Allan Fulkerson, Chairman of Century Shares Trust notes, "The court's ruling said that insurers are entitled to make a reasonable rate of return. Now that can be defined in a whole lot of different ways." According to Fulkerson, most insurers would be happy to "earn a reasonable rate of return." He adds "that would probably be better than what they are doing. So, selectively, the ruling could lead to rate increases--not decreases." See Welling (1989).

(2) Moore and Schmit (1989) also use Best's Review to identify property-liability insurance firms affected by the Risk Retention Act of 1986.

(3) Many of the property-liability firms listed by Best's Key Rating Guide are subsidiary companies of parent firms. For example, the 22nd largest property-liability insurer listed by Best's, Maryland Casualty, is a subsidiary of American General. In these cases, the parent firm's stock prices are used in the analysis.

(4] Firms with a large California exposure were identified by Standard and Poor's Creditweek November 14, 1988 and December 5, 1988, by Moody's in the Dow Jones News Wire November 17, 1988 and by the Wall Street Journal November 11, 1988. The percent of total premiums exposed to the California market averaged 30 percent among the 17 firms.

(5) Schwert (1981) discusses the usefulness of financial data in examining the wealth effects of regulation.

(6) The statistical tests chosen in this article are based on simulation results for several event study methodologies explored by Brown and Warner (1985). The procedure used for standardizing daily abnormal returns (equations 4 and 5) has been followed in various studies including Linn and Pinegar (1988).

(7) For the day t = -1 to day t = 1 interval, the abnormal return for each insurance sample is significantly different, at the 0.01 level, from the abnormal from the abnormal return for non-insurance firms.

References

Brown, Stephen J. and Jerold B. Warner, 1985, Using Daily Stock Returns: The Case of Event Studies, Journal of Financial Economics, 14: 3-31.

Gross, Jane, 1989, California Court Upholds Car Insurance Rate Cuts, New York Times, 138, May 5: A1.

Hill, G. Christian and William Celis III, 1989, 'Fair Return' on Each Line of Insurance is Backed by California Commission, Wall Street Journal, 213, May 12: A5.

Linn, Scott C. and J. Michael Pinegar, 1988, The Effect of Issuing Preferred Stock on Common and Preferred Stockholder Wealth, Journal of Financial Economics, 22: 155-84.

Miller, James P., 1988, California's Approval of Proposition 103 Throws Insurance Industry into Turmoil, Wall Street Journal, 212, November 11: B8.

Moore, William T. and Joan T. Schmit, 1989. The Risk Retention Act of 1986: Effects on Insurance Firm Shareholders' Wealth, Journal of Risk and Insurance, 56: 137-45.

Schmitt, Richard B. and Sonja Steptoe, 1988, California Voters Shake up Insurers: Rates Slashed for Property Casualty Lines, Wall Street Journal, 212, November 10: B1.

Schmitt, Richard B. and Ken Wells, 1989. California High Court Upholds Proposition 103: Across the U.S., Insurers Brace for the Effects, Wall Street Journal, 213, May 5: B1.

Schwert, G. William, 1981, Using Financial Data to Measure the Effects of Regulation, Journal of Law and Economics, 24: 121-58.

Welling, Kathryn M., 1989, No Multiple Risk: Insurance Stocks, Claims Allan Fulkerton, are Cheap, Barron's, June 5: 16-17.

Samuel H. Szewcyk is Assistant Professor of Finance of Drexel University. Raj Varma is Assistant Professor of Finance at the University of Delaware.

This study benefited from suggestions by two anonymous referees and an anonymous associate editor of the JRI, from presentations at Drexel and the University of Delaware, from the careful data collection assistance of Peter Dempsey and from the comments and encouragement of Travis Pritchett. The study was funded in part by research grants from the College of Business and Administration at Drexel University and the College of Business and Economics at the University of Delaware.
COPYRIGHT 1990 American Risk and Insurance Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:California
Author:Szewczyk, Samuel H.; Varma, Raj
Publication:Journal of Risk and Insurance
Date:Dec 1, 1990
Words:2560
Previous Article:The market value of the corporate risk management function.
Next Article:Insurance firm market response to California Proposition 103 and the effects of firm size.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters