NJ-PAIP and Voluntary PRIVATE PASSENGER AUTO INSURANCE

IN NEW JERSEY:

A THREE DECADE ADVERT FOR REFORM

Private Passenger Automobile Insurance Regulation

in New Jersey: A Three Decade Advert

for Reform

INTRODUCTION:

New Jersey usually has the highest private passenger automobile insurance rates in the United States. Year after year, state residents have listened to news reports, or read their local newspapers, and found that New Jersey, once again, was able to claim the dubious distinction of being the state with the highest average expenditure for auto insurance. The National Association of Insurance Commissioners (NAIC) annual expenditure calculations are dutifully reported, and State Legislators and Governors tremble. People who need their cars to drive to work, school or church, are not concerned with the intricacies of insurance law or rate making. They simply want to know why they had to pay over $1,100 to insure a car in 2006 while folks in Iowa paid $460.

The answer to that question lies in the nature of the state itself, and in the story of its regulatory history. For most of the 1980's and 90's New Jersey was the most expensive state in terms of average private passenger insurance. 1 Table 1 shows the New Jersey and countrywide average expenditure for 2004 to 2006, the latest five years of data available for this study. New Jersey not only led the nation in average expenditure in each year, but its average expenditure grew faster than the countrywide expenditure. In 2004, New Jersey drivers spent 48 percent more than drivers in the rest of the country. In 2006, New Jerseyeans were spending 62 percent more, with an average New Jersey expenditure of $1,138 for private passenger auto insurance.

(INSERT TABLE 1 ABOUT HERE)

For the insurance economist, New Jersey is a dream. New Jersey's auto insurance system has offered a series of natural experiments which provide examples of incentive response, regulatory lag, market failure, binding price constraints, "excess" profit law, barriers to exit, and cash flow residual market financing. The state is the most densely populated with 1,134 people and 695.66 registered motor vehicles per square mile 2 , has metropolitan areas and cities which consistently rank at or near the top of the auto theft table, has been a haven for notorious fraud rings, and has had a "Cadillac" benefit system. New Jersey is a compulsory insurance state which began the 1970's with a tort system. It introduced a No-Fault choice with a low monetary threshold which was rapidly eroded by medical care price inflation, a recipe for claim build-up and law suits. It switched from an assigned risk plan to a Joint Underwriting Association (JUA) with carrier fees based, in part, on loss payments, hence offering perverse incentives for larger claims payments (I am not making this up). 3 Having seen the JUA run up a three billion dollar deficit, New Jersey switched to a state run "Market Transition Facility" (MTF) which managed to accumulated a billion dollar deficit in two years (Did I mention that I am not making this up?). It is the state which gave the world the "Remand Case" and the subsequent and infamous "Clifford Formula," which is still being given as a rate making rationale in New Jersey rate cases, almost thirty years after its adoption. 4 In 1969 the New Jersey Supreme Court ordered a "Remand" of the 1967 Private Passenger Automobile Insurance rate case (see below). The Remand resulted in a Hearing before Insurance Commissioner Robert L. Clifford, who was himself to become a Justice of the New Jersey Supreme Court. Commissioner Clifford basically ordered that investment income had to be explicitly considered in insurance ratemaking. He offered a formula which allowed a 6 percent return on "necessary" statutory surplus, which he defined as surplus necessary to write business at a two to one premium to surplus ratio. Any surplus in excess of necessary surplus was only entitled to earn a return of one percent. Any investment income earned on invested reserves could be used to reduce the allowed return on surplus. Modern financial economics would reject book and accounting measures of rates of return and call for market based, discounted cash flow analyses. See Cummins (1990) for a thorough discussion of appropriate techniques for the treatment of investment income.

On May 19, 2006, Governor Christie Whitman signed the Automobile Insurance Cost Reduction Act 5 and announced that the state finally had achieved auto insurance reform. Although it is still early in the reform process, rates, like the Berlin Wall, may fall. The Insurance Commissioner has ordered a "15 percent" rate reduction. Some actuaries have priced the impact of the law changes and arrived at estimated cost savings well below 15 percent. 6 Insurers were reporting that expenditures on car insurance finally were decreasing in New Jersey, but this may be because of the mandatory 15 percent rate roll-back which took effect in March of 1999. The true test of system reform is yet to come. It will be a few years before we see if claim costs fall, and the tort system will check the new law as it has all previous reform efforts. As we shall show below (see Table 2), some of the largest carriers in the state have filed for hefty rate increases in late 2000 and early 2001.

We still have strict "prior approval" rate regulation, with all of its attendant potential ills, but hope springs eternal. The New Jersey experience should prove to be an object lesson in how not to regulate rates, and how to write bad insurance law. Surveys of economists and insurance professionals indicate that they have jaundiced views of price regulation. Edward L. Lascher reported on his 1995 survey, with Michael Powers, of members of The American Risk and Insurance Association (ARIA). ARIA is the professional organization which sponsors the top ranked academic insurance research journal, The Journal of Risk and Insurance. Its membership consists of academic and industry based scholars in finance, economics, law, actuarial science and related disciplines with a primary interest in insurance research. Ninety-seven percent of those ARIA members surveyed agreed with the statement, "Regulator imposed ceilings on rates are likely to cause firms to restrict coverage and/or withdraw from the market." Strong agreement was expressed by 70%. In New Jersey, firms have gone to court in attempts to exit the market. At times, half of the insured drivers in the state have found themselves in the residual market. In other states, many of those drivers would have been preferred risks. Lascher reported that ninety-four percent of ARIA members responding believed reductions in claim cost will lead to decreased premiums. Claim costs drive auto insurance cost. Lascher reported that ninety percent of respondents believe attorney involvement increases cost. 7 The Trial Bar in New Jersey has attempted to have "No Fault" repealed, and they have fought attempts to strengthen the Limitation on Lawsuits provision of New Jersey private passenger insurance law. Their view that an injured party should have the right to sue for recovery for pain and suffering may be utility maximizing for some, but it does not come without other costs to society.

I shall begin this essay with a brief overview of recent law changes and the concomitant 15% mandated rate rollback. I shall then set out the framework for rate regulation of the private passenger business in New Jersey.

Marjorie M. Berte gave an overview of the New Jersey private passenger insurance market:


"No discussion of failed political solutions to the auto

Insurance crisis is complete without a look at New Jersey,

The state with the most tangled auto insurance mess in history.

Even the Soviet Union would be hard pressed to match this

Economic disaster." 8

AICRA AND THE 15% RATE ROLL-BACK

The Whitman Administration and the Republican controlled legislature have passed a series of recent reforms to the private passenger insurance law. The first wave came in the Summer of 1997 with the adoption of tier rating to provide pricing flexibility and deal with consumer anger over surcharging; and the formation of Urban Enterprise Zones (UEZ's) to provide increased availability and decrease the number of uninsured motorists in urban areas. The UEZ's, which represent New Jersey's cities, some of the poorest areas in the United States, constitute about 11.5 percent of private passenger automobile policies in force. The 1997 law change also brought repeal of flex rates 9 and adoption of "expedited filings." Although these changes may prove to be significant, the 2006 reform package, Automobile Insurance Cost Reduction Act, has generated the most publicity. Perhaps, this is due in large part to the rate roll-back which was ordered after passage of the law and successful court cases to defend the newly enacted law.

AICRA introduced sweeping changes. The new law tried to strengthen the verbal threshold requirement for lawsuits. The law now requires that a claimant suffer a bodily injury "which results in death; dismemberment; significant disfigurement or significant scarring; displaced fractures; loss of a fetus; or a permanent (emphasis added) injury within a reasonable degree of medical probability, other than scarring or disfigurement. An injury will be considered permanent when a body part or organ, or both, has not healed to function normally and will not heal to function normally with further medical treatment." 10 This provision attempts to eliminate "soft tissue" suits, and to focus on "permanent" injuries. Injuries must be certified as permanent by a physician as defined by the law. This provision of the law will, no doubt, be tested in the courts, and the definitions teased out through the legal system.

AICRA focused on reducing fraud, introducing a Fraud Prosecutor and strengthening staff. It also made fraud a more serious crime. If you live in New Jersey, or have the radio on while driving through the State, you have probably heard the advertisement announcing that you will be caught if you engage in auto insurance fraud, and that you can do jail time for the crime. If a physician certified a fraudulent claim, he could lose his license to practice medicine.

The law called for a revamping of New Jersey's rating territories. A commission has been meeting in 1999-2000 to study the shape of new territories and it has prepared a draft report. The law changed the territorial rate cap system. 11 The provision is designed to insure equity, have more uniform territorial sizes, and focus rating on actuarially sound factors.

The reform introduced a new "Basic Policy" which offers drivers a chance to purchase minimal coverage and avoid driving uninsured. The policy has a $15,000 PIP medical expense limit, and a $5000 property damage liability (PD) limit. It provides no bodily injury liability (BI), but $10,000 of optional coverage may be purchased. The policy carries no uninsured/under insured motorist protection. The policy is clearly designed to give those who have been unable to drive legally in New Jersey's expensive market the opportunity to do so. 12

The reform also introduced a "Named Driver Exclusion." The default under the old law charged high risk drivers to the most expensive car on a policy. Families with teenage, inexperienced drivers, saw their rates skyrocket as their newly licensed 17 year old children were charged to mom or dad's new car. Now, they can be excluded from using a car, and if they do drive it and have an accident, their carrier will not pay a comprehensive or collision claim.

AICRA provided choice in PIP in an attempt to control cost. The mandatory $250,000 coverage remains the default option, but consumers can now select PIP coverage at $15,000. The law change provided for tighter control of the use of medical procedures, and introduced an arbitration system for PIP claims. Obviously, this is designed to limit lawsuits and take cases away from tort and place them in the hands of arbitrators. The law established an Ombudsman to investigate consumer complaints, and it also opened the door to intervention in rate hearings.

As a result of these reforms, on January 4, 1999, the Commissioner ordered a 15 percent reduction in rates for all private passenger automobile insurance policies issued or renewed "processing on or after March 22, 1999." The reductions were ordered to be distributed as follows: Personal Injury Protection, 27.25%; Bodily Injury Liability, Limitation on Lawsuit 24.33%, no Lawsuit Limitation 3%; Property Damage Liability, 3%; Collision, 8.82 percent; and Comprehensive (or other than Collision), 3.0%. 13

Notwithstanding the actuarial studies cited above, the early results appear to indicate that policy expenditures have declined. Twelve member companies of The Insurance Research Council of New Jersey, with a combined market share of over 50 percent, reported that private passenger policies sold in the second quarter of 1999 had an average premium of $954, down from $1,138 in late 1997. 14 This 16.2 percent reduction comes as good news to New Jerseys long suffering policyholders. However, by the end of 1999, The Insurance Council was reporting average expenditure was $988. 15 The true test will be whether economic costs actually decline. New Jersey has had a history of declaring war on costs, announcing that it has won the war, and pretending that deficits do not exist.

PRIVATE PASSENGER AUTO INSURANCE RATE REGULATION IN NEW JERSEY

New Jersey is a strict prior approval state. The Insurance Department's Office of Property and Casualty handles filings for rates, rules and forms. They have a very heavy work load with 3,384 filings in 2006. Of that total, 714 were for personal lines, up from 483 personal lines filings in 1997. Elimination of flex rates and the introduction of expedited filings (see below) by 1997 law change, and the May 2006 passage of the Automobile Insurance Cost Reduction Act (AICRA) were probably major contributors to the increased filing activity. The Department disapproved 16 filings and closed or withdrew another 789 in 2006. They approved 2,392. 16 All requests for rate increases must be filed with the State Department of Banking and Insurance (hereafter, the Insurance Department). The Commissioner makes a determination whether the rates are "unreasonably high, or inadequate, or unfairly discriminatory" and issues an order of approval or disapproval based on the determination. 17

When a private passenger insurer files for a rate change, either the filer or the Insurance Department may request a hearing which may be conducted by the Commissioner, his or her designee, or by the Office of Administrative Law (OAL). 18 The hearing is supposed to start within "30 days of the date of the request or decision that a hearing is to be held" extensions are permitted), the findings and recommendation are to be presented to the Commissioner within 30 days of close of the Contested Hearing, and the Commissioner should render a decision within 60 days of receipt of the findings of fact and recommendations. The filer bears the cost of the hearing. 19 The Commissioner has the power to order temporary partial rates to be used until a hearing and determination can be made. These temporary rates can be subsequently adjusted by refunds, debits or credits. 20 No other rates, dividend plans etc. can be used unless they are on file and approved by the Commissioner. 21 Filers are required to offer premium credits which must be a uniform percentage statewide for deductibles and exclusions under personal injury protection (PIP) and for tort limits under bodily injury liability (BI). Insurers can not vary the commission percentage they pay based on the policy options insured's choose. 22

RECENT RATE CASES

In Table 2, I set out the requests for rate changes filed with the New Jersey Insurance Department. Three determinations handed down by the Department in the Spring of 2000 were all for less than filed. After having one request for an increase denied, New Jersey Re-Insurance, a subsidiary of the third largest private passenger automobile insurer in the state, New Jersey Manufacturers, filed for a 22.7 percent rate increase in June, 1999. The case was heard by an Administrative Law Judge who ruled that New Jersey Re should get an unspecified rate increase. On June 16, 2000, the Commissioner announced that she had "modified the Administrative Law Judge's decision and found a rate need of only 8.9 percent rather than the requested 22.7 percent." 23 It was one year from filing to the issue of the order.

(INSERT TABLE 2 ABOUT HERE)

The Commissioner also ordered a 2.9 percent rate decrease for the Prudential Group that day. Pru Group had filed for a 1.2 percent decrease in January. It took about five months for the decision. The department noted that "Prudential's last prior approval rate increase was 6.9 percent in 1990." (Emphasis added)

An interesting example of price regulation in New Jersey is the GSA private passenger case. Maryland Casualty withdrew from the New Jersey market in 1997 with GSA assuming over 90 percent of Maryland Casualty's private passenger risks. Maryland Casualty had not had a prior approval rate increase since the late 1980's. GSA filed for an 85.5 percent rate increase on September 28, 2006. On October 8, 2006, GSA requested that the filing be sent to the Office of Administrative Law. The Department transferred the case to OAL on December 24, 2006, where it was filed as a Contested Case. On February 10, 1999, the case was assigned to Administrative Law Judge John R. Tassini. After the proper notifications, pre-hearing motions etc., the Plenary Hearing was held on August 11, 13 and 19, 1999. Briefs were filed on November 15, 1999. The record was closed on January 5, 2000. The Initial Decision was rendered on January 6, 2000. The Insurance Department's experts supported a 21.1% increase. On May 5, 2000, over a year and a half after GSA filed 24 , the Acting Commissioner announced that she had "rejected an Administrative Law Judge's recommendation that GSA Insurance Company be granted a 34 percent rate increase for private passenger automobile insurance." She granted a 15 percent rate increase. The Insurance Department's press release announcing her decision noted that GSA had lost more than $13 million in 1999. 25

The three rate requests discussed above are counted by the Department as "approved," even though the Department approved an average of about 79 percent less than the rate request. For completeness, the Department has received 32 auto insurance rate filings from January 1, 2006 to December 31, 2000 with a final determination made.. The Department denied 29 of those request, and the insurers "accepted the denial," or at least they did not request a hearing. By the Department's accounting method, the denial rate was approximately 90 percent. 26 By the Jack Worrall method, all 32 rate requests were denied.

There are currently 11 rate requests pending (see Table 2). These rate requests, filed after the law changes, do not bode well for the early success of the reforms of 2006. The combined market share of three of the insurers seeking rate hikes (State Farm, Liberty Mutual and Selective Group) is 30 percent of the New Jersey market. An Insurance Department spokesperson has claimed the Department will fight every one of these rate hikes, adding that the industry had "not demonstrated a need for these rate increases." 27 State Farm has received permission (April 2001) to stop writing new private passenger automobile insurance policies after sustaining massive losses in the New Jersey market in 1999 and 2000. 28

QUALIFIED PERSONS

Someone designated as a "qualified person" can get copies of any filing and amendments and request that the Commissioner hold a rate hearing on the filing. The designated "qualified person" is required to know New Jersey insurance law, have the ability to evaluate the technical aspects of a rate filing, have access to an actuary, have sufficient resources to undertake an evaluation of the filing and represent the interests of consumers. 29 If the Commissioner finds that the qualified person has made a substantial contribution to his determination, he can order that fees be paid to the qualified person. Although this appears to provide an opportunity for rent seeking, it may also offer a heavily burdened Department the opportunity for expert input and advice. The Office of Property Casualty handled 429 requests from insurers, law firms and the public for records to review filings submitted to the public. As of December 2000, there do not appear to have been any hearings initiated by public intervention.

New Jersey had a Public Advocate who intervened in rate cases, but Governor Christie Whitman eliminated the position during her first year in office (2004). The Office of the Public Advocate was seen by some in the state as an office which simply piled additional costs on the regulatory system, and by others as one that protected the consumer from price gouging by insurers. The issue was hot enough in the 1997 election race for Governor Whitman's popular Democrat opponent, State Senator Jim McGreevey, the Mayor of Woodbridge, to propose re-instituting the Public Advocate's office. 30 An exit poll the day of the election found that almost nine of 10 voters reported that auto insurance was important in their decision on whether to vote for Whitman. 31

EXPEDITED FILING

New Jersey also has an expedited filing rule. A filer may use this option when filing a rate increase of three percent or less or a decrease in its statewide base rate. The technical requirements include certification by a qualified actuary that the filing meets the requirements of the law and regulations set out by the Insurance Department, as well as generally accepted rate making principles. The filing can not result in an increase over five percent for any single coverage, even if it meets the three percent average rate increase requirement. The Commissioner should make his determination within forty five days of receipt of the filing, which he can extend to 60 days. If a negative decision is not rendered, or the rates not modified, within the time limit, the filing is considered approved. 32 This provision provided by P.L. 1997, c.151, s.34, appears to be similar to the "flex rating" which New Jersey had from 1988 to 1997, but the expedited filing comes with no guarantee. Under the flex rating scheme, insurers were entitled to increases of three percent a year; however, the Commissioner could grant larger increases. State Farm took a case to the New Jersey State Supreme Court arguing that the state did not have the right to automatically limit flex increases to 3%. In a 1992 decision the court held (6-1) in State Farm's favor. 33 There were flex increases of 5.2% in 1989, 6% in 1990, 6.5% in 1991, 5.18% in 2004, and 3.83% in 1995. 34 With the minimum 3% approved for 1992 and 1993, flex rates grew 37.6% over the 1989-1995 period. The Medical Care Component of the Consumer Price Index grew 46.9% over the same period.

CAPS

There is subsidization in the New Jersey rates. Under the Automobile Insurance Cost Reduction Act an insured can not be charged more than 2.5 times the territorial rate base, nor more than 1.35 times the statewide average base rate. If the principle operator is 65 or older, the ratio is 1.25 times the statewide average. 35 Previous law had a simple 1.35 cap. These caps obviously fly in the face of actuarially sound pricing and generate poor price signals. Although rates under the New Jersey law should ostensibly not be discriminatory, rates are flattened for social purposes and the cost burden is born by other policyholders. 36 These redistributions might be handled more efficiently through other, more explicit tax policy. The rationales offered for private passenger caps range from avoidance of market disruption to the social insurance aspects of private passenger insurance and minimization of the uninsured motorist problem. There are cap provisions for the Basic Policy, for example. The state has strong interest in urban areas, and, as insurance industry sources have reminded us, caps were introduced in 1983 to subsidize urban areas. 37 Auto insurance costs have figured prominently in New Jersey elections. State politicians have had a delicate balancing act. Some may find their constituency in the cities and urban areas of the state, others in the suburbs. 38

URBAN ENTERPRISE ZONES (UEZ's)

Governor Whitman signed P.L. 1997, c. 151, on July 1. It established Urban Enterprise Zones (UEZ's). 39 The scheme was designed to make insurance more available in New Jersey's urban areas, with 27 New Jersey cities constituting the Urban Enterprise Zones. It has been argued by some that insurers might practice price discrimination based on race. It would seem that the market could be sealed, and different elasticities of demand for coverage could exist. However, a careful econometric analysis of the question has shown that profit rates do not vary across racial groups.

Differences in premiums, where found, were the result of differences in loss costs. 40 Although the Insurance Department characterized the UEZ plan as "based on the expectation that various incentives could be used to encourage insurance companies to write more in the cites," insurers must insure risks in proportion to their writings in the voluntary market statewide. For example if they insure 10 percent of the state's voluntary business, they must insure 10 percent of those who desire coverage and who live in cities designated as Urban Enterprize Zones. Insurance companies can use special agents to acquire the business. If an insurer is not within 95 percent of its assigned goal, the company will be assigned risks under a plan know as the Urban Zone Assigned Risk, which is not to be confused with the state's assigned risk plan described below. 41 In effect, the state has used its statutory and regulatory authorities to alter market shares, and underwriting choices, of private passenger insurers. The Department noted that the UEZ market shares of 36 firms had increased, while those of 10 had decreased. The Department believes the program is having the desired effect. Insured vehicles jumped from 487,618 to 544,067 in the UEZ's. Total written exposures, evaluated on December 31, in the voluntary market increased from 4,509,675 in 2006 to 4,660,450 in 1999. 42 Although the growth of UEZ exposures was 12 percent over the time period, and that of the voluntary market less UEZ's was only 4 percent, both the UEZ provisions of the new law, good economic conditions and other law changes, including the introduction of the new Basic Policy, may all have accounted for some of the growth in policyholders insured in UEZ's.

TIER RATING

On July 1,1997, the state adopted a "tier rating system" with each insurer operating in the state required to notify the Department by March 9, 2006 of its intent "to file a tier rating plan to replace an existing standard/non-standard rating system." Filings were due by April 1, 2006. 43 Under the tier rating plan, New Jersey's system of surcharging drivers for tickets and accidents came to an end. However, the tier rating plan, which required insurers to define a standard tier as drivers with six or fewer points, allowed for preferred and less preferred rating tiers. These preferred and less preferred tiers may be further subdivided. Insurers use different determinants in their risk classification schemes and, hence, price dispersion will be generated in the state and its rating territories. During Governor Jim Florio's administration, a law passed in 1990 which eliminated place of residence, age, sex and marital status as determinants. 44 Florio also proposed and was successful in passing a $2.8 billion dollar tax increase. There was a voter revolt and the Republican party took control of the state legislature in 1991. The legislature passed a bill to repeal the ban on residence, age, sex and marital status as rating factors before they were due to be eliminated effective January 1, 1993. It was conditionally vetoed by Governor Florio, but the legislature over-rode his veto 45 . The traditional rating factors are allowed in New Jersey. 46 Use of these factors is barred in a few states, and there has been pressure to eliminate use of territories, in large part due to political pressure for urban rate suppression. 47 P.L. 2006, c. 22, amended the Automobile Insurance Cost Reduction Act, and called for the establishment of new territorial rating plans. 48 A commission was created to establish the factors for use in the design of the new territories. The state's 27 territories have been in use over 50 years, and the amendatory legislation offers guidelines on the redistricting of territories. It specifically calls for no schemes which "result in unfair inter-territorial subsidization among territories ." Experience teaches that unfair covers a good deal of territory.

The tier rating scheme was designed to be revenue neutral and a number of sources pointed out that the law change could result in increased rates for some risks and decreases for others. 49 The Insurance Department requires quarterly tier exposure reports and will have a data series to check market conditions in each territory. 50 Although a few companies have only one tier, by the end of 2006, the majority had six (34 writers) or more tiers. The Insurance Services Office (ISO) also filed six rating tiers. Six firms have 10 or more tiers. 51 The Insurance Department provides consumers with a booklet explaining the tier rating system and giving a description of each insurer's criterion for tier assignment. 52 They also have made tier rating information, at the company and group level, available at their web site (www.njdobi.org). As information makes markets function efficiently, this type of effort, together with the Department's premium price comparisons within rating territory, should lead to more price search. The Insurance Department facilitates that search by lowering the costs of information. They have established an 800 toll free number and send within-territory price comparisons for a number of typical policy options for drivers with different demographic and driving characteristics. They received 4000 requests for such information in the program's first few months of operation. In the early 1970's, survey research indicated that about half of auto insurance consumers never compared their auto insurance prices with those offered by other insurers. 53 The Insurance Council of New Jersey reported survey results that found that one in four drivers shopped for a new policy in the last six months (during 1999), 55 percent of movers found reduced rates, 38 percent of those saved at least $400. 54 In 2000, the Insurance Council reported that there was a 50 percent increase over the 1997-2006 figures in those consumers who switched their private passenger coverage to another insurer. 55 Since the price dispersion generated under tier rating can be great, Insurance Department efforts to disseminate pricing information should be encouraged.

Since price and quality can both vary across policies, the Insurance Department sends both their Shopping Guide and the Auto Insurance Complaint Ratios for writers insuring 1000 or more risks. This data is also provided at the Department's web site. 56

(INSERT TABLE 3 ABOUT HERE)

In Table 3, I have set out a few illustrative examples of the range and average premiums for coverage in a suburban New Jersey territory in 1999. Notice that with the exception of the "Basic Policy" Example 7) and the single car coverage for a married senior couple (Example 4-A), the differential between the highest and lowest premiums exceeded $1,000 in the territory. A married couple with a newly licensed 17 year old daughter (Example 2-A) could save as much as $3,777 by shopping around. Although the case is not illustrated here, savings can be as great for a family with a newly licensed son.

EXCESS PROFITS LAW

Without belaboring the fundamentals of price theory, those who put their capital at risk should have the expectation that they will earn the opportunity cost of that capital on a risk adjusted basis. In general, expectations and outcomes do not always match. There may be "bad" years and there may be "good" years, but if returns, for example, are (log) normally distributed, one's best expectation would be the mean of the distribution. To the extent that outcomes are not entirely deterministic, a scenario where insurance is not necessary, luck plays a role in profit and loss. Excess profit laws, if binding, eliminate the upper tail of the distribution and reduce the expected return on capital. In general, excess profit laws are bad. In cases where markets function relatively efficiently, and entry is not very costly, the inexorable forces of competition will erode any "excess profits." In states where price ceilings are set and binding, not only will the assigned risk pool blossom, but the excess profit rule could prove superfluous. Rates are set ex ante and in the face of considerable uncertainty about future outcomes. Mandatory rate roll backs, before losses are incurred and developed, may not be efficient. For example, the passage of an insurance law with an estimated savings of 10 percent does not guarantee such savings. If actual losses exceed expected losses, revenues collected will be less than those required to attract and retain capital. If rates prove to be too high, cash flow timing can be changed, dividends paid, etc. to arrive at competitive rates.

The excess profit provision requires that private passenger insurers file annual excess profit reports. Reports are required for each company, including those in a group, and each group for the major liability and property damage coverages. The law is designed to limit the ratio of total "actuarial gain" plus "excess investment income" to earned premiums to 2.5 percent over three calendar-accident years. Abstracting for simplicity, "actuarial gain" is defined as underwriting income minus any allowance for profit and contingency. Underwriting income is earned premium minus developed losses and loss adjustment expense, with other minor adjustments defined in the law. "Excess investment income" is defined as "actual investment income" minus "anticipated investment income." The "actual investment income" is the portion of income generated by investing policyholder supplied funds. "Anticipated investment income" is the percent which represents investment income and used in the insurer's "approved rate filings." 57 Attempts to allocate income on the basis of policyholder supplied funds in the excess profit law imply that insurers are simply holding money for people who sell them a cash flow. Unfortunately for insurers, they stand to lose the policy limits, and the sellers of the cash flow (insured's) do not have to contribute to "insurer supplied funds."

The actual size of "excess profits" has been small relative to the size of the market as measured by written premiums. For example, on April 3, 2006, the Commissioner announced that "Since 2004, New Jersey consumers received a total of $4.15 million in excess profit refunds and 364.9 million in dividends from auto insurance companies." The total of $4.15 million over the 2004-1997 period was less than one one hundredths of one percent of direct written premium. Liberty Mutual Fire Insurance Company refunded $1.5 million on business through 1997. They paid dividends of $26 million on 1997 New Jersey private passenger business. Electric Insurance Company paid $204,625 on the three calendar-accident years through 1997. 58 In 2006 the Property-Casualty Division evaluated 63 excess profit reports. State Farm Indemnity ($38 million), Hartford Group ($2.5 million) and Harleysville-Garden State ($3 million) were those that had to provide refunds under the law. 59 These 2006 excess profits constituted less than one percent of direct written premiums 60 . In 1999, The Hartford Group ($3.2) was again required to pay excess profit refunds commencing by January 1, 2000. 61 Although these "Excess Profits" are large in absolute terms, they are much smaller than dividends and constitute a small percentage of written premiums. In Table 4, I list the dividends paid and the percent of direct written premium they represent. It is instructive to examine the ebb and flow of dividends, as well as the market share of the residual market, as they both can provide

(INSERT TABLE 4 ABOUT HERE)

information about the state of market conditions. Some dividends paid may reflect the "threat effect" of excess profit law, but competition for good risks and differences in marketing and expenses are also likely causes.

Private passenger automobile insurance is sold by independent insurance agents (the independent agency system) who can represent a number of insurance companies or by insurance companies that sell policies directly to consumers (direct writers). Consumers who use the independent agency system will typically expect their agent to recommend the best coverage and price available. Consumers may also expect, and be willing to pay higher expenses, for other services provided by independent agents. Direct writers typically market their product through direct mail, the internet etc. They use their company employees, or agents who work exclusively for them, as well. One of their goals is to capture the expense economies available to their marketing approach, and to pass the savings on to consumers. Independent agency business has never paid a dividend of 1 percent, although it came close in 1987, but direct writers paid between 5 and 8.6 percent of direct written premiums in the form of dividends between 1984 and 1989. They paid their highest dividends before the excess profit law was passed 1988. They paid dividends even through the Joint Underwriting Association (JUA) years, but dividends dropped during the Market Transition Facility (MTF) period of 1990-1993. 62 For the 1997 and 2006 period they averaged a 6 percent dividend.

New Jersey's loss ratio has typically been larger than that of states which have competitive rating laws. There can be a host of reasons why this is the case. Professor Scott E. Harrington uses multivariate analysis to examine the impact of rate regulation on insurance prices and volatility elsewhere in this volume. In Table 5, I list the loss ratio for the New Jersey voluntary market private passenger automobile insurance market, and for groups of states which had either a competitive rating or prior approval law over the

(INSERT TABLE 5 ABOUT HERE)

entire period 1980 to 2006. Notice that for 17 of the 19 years, the New Jersey loss ratio was higher than the average ratio for the 10 states which maintained competitive rating laws over the entire sample period. These competitive rating states are those which provide the most pricing flexibility and the least rate regulation. 63 New Jersey also had a higher loss ratio than the average of the 20 prior approval states in 16 of the 19 years. I excluded two of the other most strictly regulated prior approval states, Massachusetts and South Carolina from the calculations. The loss ratios presented in Table 5 are direct losses incurred divided by direct written premium net of dividends paid. Comparing loss ratios in cross section may give some information about insurer profitability, but loss payouts, expense, dividend and tax flows, premium collection and other cash flows can vary from state to state. Profitability can and does vary over time, as investment yields and cash flows change. The revenues collected in the voluntary market in New Jersey were reduced by the industry's funding of the deficits run up by the Joint Underwriting Association-Market Transition Facility schemes described elsewhere in this paper. These revenue reductions are not reflected in the New Jersey loss ratios presented in Table 5, hence those loss ratios are understated.

Expenditures for policies are down, dividends are up, the residual market share - as we shall see - fell from roughly half of New Jersey's drivers to 2.5 percent between 1988 and 1993. Table 5 shows that loss ratios fell in New Jersey in 2006, a year when New Jersey's ratio was lower than the average for the competitive rating and prior approval set. Before the recent spate of requests for rate increases, the market appeared to be improving. We now will consider what factors got New Jersey into the a position where its private passenger automobile insurance market was used as an example of market failure. 64

SOME HISTORY: THE LAW AND REGULATION THAT PRODUCED PROBLEMS FOR NEW JERSEY

The principle determinant of insurance premiums is claim costs generated by the frequency and severity of claims and the benefit system and its regulation in a state . These costs can be high due to factors beyond the control of the insurance industry - population density; personal income levels of the state's inhabitants, and the age, value, and nature of the vehicles they purchase and insure; the nature of the law and regulation; and the propensity to engage in fraud, claims filing (frequency and severity), etc. - or, due in part to actions of insurers themselves. 65

In New Jersey, a segment of the public has focused its attention on

insurance companies and accused them of "cooking the books," "price gouging," etc. The fact that those same insurance companies were trying to exit the market, and attempting to refuse to write risks may have been lost on consumers. Policyholders wanted to know why they could buy full coverage in 1973 for $225, and have an average expenditure of $957 in 1992. The Trial Bar in New Jersey has blamed the high costs on "No-Fault." 66

NO-FAULT IN NEW JERSEY: THE 1972 LAW

New Jersey is one of three states which offer Elective No-Fault as a choice. Motorists in the state can adopt a Limitation on Lawsuits option or they can retain their right to sue. No-Fault is the default option in New Jersey, and it is "selected" by almost 90 percent of insured's in the state. There seems to be a framing effect as our neighboring state, Pennsylvania, offers choice, with tort as the default, and only half of its drivers adopt No-Fault. New Jersey's first No-Fault law was the New Jersey Automobile Reparation Reform Act (P.L. 1972, c. 70), which became effective January 1, 1973, after much study, debate and rancor. In Table 6, I list some of the major provisions of law changes for the New Jersey private passenger automobile over the last 30 years.

(INSERT TABLE 6 ABOUT HERE)

New Jersey was a financial responsibility and contributory negligence state prior to the New Jersey Automobile Reparation Reform Act. Between 1956 and 1969, the "cost of bodily injury liability coverage (was) up 237.4 percent and property damage liability up 195.5 percent." Insurers, through the National Bureau of Casualty Underwriters, filed for 8 major rate increases over the twelve year period, including three years in a row. In 1967, for the first time, a public rate hearing was called to consider a request for a 19.6 percent hike in bodily injury liability and a 22.7 percent increase in property damage liability. The industry and the Insurance Department brought in expert witnesses, and that rate hearing spawned the famous Remand Case and the Clifford Formula discussed above. 67 The Remand Case and the Clifford Formula mark the onset of an explicit recognition of investment income in insurance rate regulation. Claim frequency was high, and, not surprisingly given the law, so was the propensity to file suits. In 1964, for example, bodily injury liability claims frequency was 2.84 per 100, and property damage liability was 8.44 per 100. Note that one third of property damage liability claims had bodily injury liability claims as well. 68 There was a 15 month delay on liability claims payment. 69 The market appeared to be tightening. Insurers did not want to take new policyholders. Risks assigned to the pool paid 20 percent more than the rate charged in the voluntary market. There were also potential surcharges for bad drivers. Insurers were accused of bullying tactics as "Consent to Rate" applications rose from 16,769 in 1966, to 20,041 in 1967, to 22,326 in the first six months of 1968. 70 These requests to sell policies at above approved rates were accompanied by an increasing residual market share. By 1972, the year the law passed, the New Jersey Automobile Insurance Plan (the residual market) was insuring 407,471 vehicles. It would soon insure over 1 million. In the three years from 1970 to 1972, automobile negligence cases added in Superior and County District courts were, 41,171; 41,695; and 38,967, respectively. 71

The 1972 law introduced a $200 monetary threshold and provide unlimited medical benefits. Ten years after its effective date (Jan. 1, 1973), medical price inflation had eroded the threshold to $77.62 in real terms. It was a simple matter to exceed the lawsuit threshold. The New Jersey Insurance Department sounded the alarm, called for legislative reform of the monetary threshold to reduce litigation and costs and to provide availability. The Department pointed out that from 1973 to 1982 mandated premium costs for PIP, bodily injury liability (BI) and property damage liability (PD), increased 248%, and Best's Insurance Management Reports listed New Jersey as the state with the highest premium per car in the nation. 72 Average claim cost rose 119%, 96%, and 65%, for PIP, BI and PD, respectively, in the first 5 years after passage of the law. 73 The Assistant Commissioner warned the world that the New Jersey private passenger market was in disarray and nearly shouted that the residual market size and growth was a signal of the market's distressed condition.

"Availability of insurance protection mandated by New Jersey's No-Fault

bears an inverse relationship with the assignments made under the New Jersey Automobile Insurance Plan (NJAIP)...A large number of risks placed by assignment would indicate that the automobile insurance market is experiencing serious difficulties. This is the situation in New Jersey today...the No-Fault Automobile Reparations system has not been addressed by the Legislature. Without meaningful tort reform legislation there is very little hope that insurance companies will once again voluntarily seek out automobile risks in New Jersey. To do so under the present circumstances would only increase the insurer's share of automobiles assigned to it by NJAIP, a risk most companies are unwilling to take." 74

(INSERT TABLE 7 ABOUT HERE)

The escalating number of drivers insured through the residual market mechanism was well underway. Table 7 reveals that by 1978, over a million vehicles were insured through the pool, and by the end of the decade, one third of private passenger vehicles were covered in the residual market. Increasing frequency, severity and cost, coupled with escalating premiums for New Jersey drivers raised not only the availability of coverage issue, but also the question of the uninsured motorist. The Insurance Department used the difference between registered vehicles (New Jersey was a mandatory liability state) and insured car years to give a rough indication of the uninsured motorist problem. The differential hovered around a half million vehicles from passage of the new law to the end of the decade. This differential averaged around 12% from 1973 to 1980 (See column 4, Table 7).

Although the law was designed to speed up payment and limit law suits, the low monetary threshold was proving to be no barrier to bodily injury claims. From 1975 to 1981, PIP claims arising increased from 64,696 to 119,239, up 84.3%. Bodily injury claims rose from 45,726 to 86,841, an increase of 89.9%, over the same period. The ratio of bodily injury claims to PIP claims arising was in the 70% range over that time period, but the claims paid bodily injury to PIP claims paid ratio fell from 37.1% to 31.4%, a sure indication of increasing severity and costs. 75

Although there is reluctance for insurers to abandon a market, Geico (1976), Safeco(1977), Nationwide(1981) and Progressive (1983) pulled out of the New Jersey market by the early Eighties. Losing two of the four largest private passenger insurers was a portent of things to come. They got out before the state slammed the door on exit, and the before subsequent spate of suits filed by firms in the early 1990's, during the JUA debacle, seeking to bail out of the New Jersey private passenger insurance business. The pressure was clearly on the Legislature to reform the private passenger law and do something about availability.

NEW JERSEY AUTOMOBILE FULL INSURANCE UNDERWRITING ASSOCIATION (NJAFIUA) - THE 1983 LAW CHANGE 76

The 1983 law change brought a Joint Underwriting Association (JUA) to New Jersey. It proved to be an unmitigated disaster. It would cover residual market drivers for policies effective from January 1, 1984 to September 30, 1992. 77 New Jersey's residual market mechanism had been an assigned risk plan. Drivers unable to secure coverage in the voluntary market had been apportioned according to an insurer's share of the voluntary market. Each insurer serviced its own assigned risks. The insurer was responsible for claims payment on all of its own policies, and since its profits depended in part on its diligence, it had an incentive to economize on claims cost. The Full Automobile Insurance Underwriting Association severely weakened that incentive, because JUA schemes have servicing companies, some of which may not even be insurance companies. Drivers who cannot secure coverage in the voluntary market submit applications for coverage which are processed by the servicing companies. The results for all drivers insured in the JUA are shared by all insurance companies writing private passenger automobile insurance in the state. The insurance company's share of the financial responsibility is apportioned by its share of the voluntary business it writes in the state. If the JUA ran up deficits, an insurance company with 10 percent of the voluntary market would be responsible for making up 10 percent of the pool deficit. Since the insurer was not actually servicing the policy or monitoring the activities of individual risks in the pool, it would have little control over the costs it was assigned to pay. The legislation creating the JUA provided for a Residual Market Equalization Charge (RMEC). The RMEC would have been charged to each policyholder, with the total revenue generated by the charge set so that the JUA would run on a break even basis. If that scheme were enforced, the pool would not have been able to run up the huge deficits that it did.

The JUA had 15 servicing insurers: Aetna, Allstate, Cigna, CNA, Continental, Fireman's Fund, Hanover, Royal, Selective, Keystone, Liberty Mutual, Penn National, Prudential, State Farm and Travelers. The JUA was supposed to have been funded with a Policy Constant and a Residual Market Equalization Charge (RMEC). The Policy Constant was a thinly veiled subsidization scheme. It was not broken out as a separate charge on policyholder bills, hence many policyholders would not realize that they were paying it. Since drivers in both the voluntary and residual market had to pay this charge, poor drivers in the pool were receiving transfers from good drivers. Rates in the pool were artificially low, and nearly half of New Jersey's drivers ended up in the residual market.

In a 1997 New Jersey Senate Commerce Committee hearing, Senate State Senator Raymond J. Lesniak characterized the Joint Underwriting Association (JUA) as "going on a cash flow basis and not reserving." 78 He was referring to the consequences of the decision to ignore unfunded liabilities. In 1985, not long after its formation, the Commissioner did not allow a RMEC because the JUA was generating a positive cash flow. The Insurance Information Institute discussing the incident noted that the JUA began running up huge deficits from its inception, and after the Commissioner's decision, they continued to increase in size. 79

The JUA was running a deficit which ultimately exceeded three billion dollars. To get an idea of the magnitude of the deficit one need only realize that the total private passenger automobile insurance direct written premium for the voluntary market in 1989, the last full year of the JUA, was only 1.7 billion dollars (See Table 5). 80 In order to enable future claims payments, most drivers in the state were surcharged up to $222 a year. This surcharge created political havoc in the state. In addition, according to the Commissioner of Insurance, insurers, who were not supposed to bear the cost burden of any deficits under the original law, paid out almost $800 million in JUA surcharges and taxes between 1990 and 1992. Insurers were to pay about 1.5 billion of the deficit, and the rest was to be paid through temporary charges to drivers and annual fees on professionals - doctors, lawyers and others - working in the system. 81 The drivers' share was to be $1 billion. It was to be paid through a system of higher registration fees, and bad driver surcharges. 82

In 1990, the former manager of the JUA was indicted on charges of "theft by deception." 83 This came on the heels of Insurance Department charges that mismanagement led to $908 million in overcharges and to the pool's huge deficit. Insurers were charged with "ripping off the consumers." The servicing companies agreed to a settlement, negotiated under the auspices of Rutgers University Professor Sanford Jaffe, of $231.3 million, with $54.5 million going to the JUA and $176.9 million going to its successor, the Market Transition Facility. The insurers admitted no wrong-doing. 84 By 1988, the JUA accounted for 43% of insured car years and 50% of auto insurance premiums in the state.

1988 INSURANCE REFORM ACT 85 AND VERBAL THRESHOLD

Claim severity doubled between 1980 and 1988 - bodily injury liability (BI) from $7,592 to $14,484, property damage liability (PD) from $748 to $1,503, and PIP from $1,496 to $3,594; and the ratio of BI to 100 PD claims increased from 17.5 to 21.0. 86 The Joint Underwriting Association (JUA) had half of the private passenger business, and the market was in disarray. Medical care price inflation had eroded the $200 monetary threshold introduced in 1973 to $56.

In an attempt to stem rising claim cost and litigation, the 1988 law change eliminated the $200 monetary threshold and replaced it with a "verbal threshold." The verbal threshold would be sorely tested in the courts of New Jersey, and those tests resulted in the "stricter" verbal threshold in the 2006 Automobile Insurance Cost Reduction Act (AICRA). Incentives to end excessive medical utilization through PIP co-pays were also included in the law, as was a required comprehensive deductible. As noted above, the law change also brought "flex rating" to New Jersey. This law provided some political cover to release some of the general rate suppression in the state, but the law still provided that any rate increase granted could not result in "excessive rates," a point strongly made by State Senator Gerald Cardinale, the Chairman of the Senate Commerce Committee in his summary of the law changes. 87 Insurers could file once a year. Insurers were allowed to drop up to 2 percent of their policyholders in a territory when their insurance policies came up for renewal, and insurers could also drop one policyholder for every two new customers they wrote . The Excess profit provision discussed above was included in the 1988 law change. Insurers had to file reports annually and meet the 2.5 percent of earned premium test set out in the chapter law. 88

Effective July 1, 1989, insurers had to make their own rate filings. Most New Jersey private passenger insurers were members of the Insurance Services Office (ISO). The 1988 law forbids filing of full rates in concert. ISO, which was filing loss costs or pure premiums in other lines and states, announced that it was going to exit the property casualty rate advisory business altogether. Although large writers filed their own rates, most of the smaller insurers used ISO rates, some revising them, some adopting them as approved by the Insurance Department. 89 It was believed that prohibiting ISO rates would foster competition. Whatever the belief, the JUA, its deficit ever growing, was still alive and not well. The race for the Governor's office began as the 1988 reform package took effect in the Summer of 1989, and automobile insurance prices and reform were on the lips and minds of the candidates and the voters. Governor Jim Florio won the race, and auto insurance reform was at the top of his agenda when he took office in January of 1990.

THE FAIR AUTOMOBILE INSURANCE REFORM ACT OF 1990:

THE FAIR ACT 90 AND THE MARKET TRANSITION FACILITY

When Governor Florio took office in 1990 he was confronted with a host of problems, not the least of which was a $600 million deficit which he inherited from his predecessor. But even before he proposed the 2.8 tax increase which contributed to his failure in his 1993 re-election race, his first official act was to introduce his auto reform package, the FAIR Act. 91 The law finally eliminated unlimited PIP medical benefits, reducing them to a mandatory $250,000. It also set up an inspection system for physical damage coverage with cars being photographed before they were insured to help reduce fraud. The law also blocked exit, requiring insurers to give up their license to write other lines. Perhaps, the most important aspect of the law was the elimination of the Joint Underwriting Association (JUA) and establishment of the Market Transition Facility (MTF).

Florio wanted to get the residual market under control and he introduced the Market Transition Facility (MTF) to replace the JUA as the insurer of last resort for the private passenger market. In three years, the state would be out of the private passenger automobile insurance business. The MTF was to issue policies with effective dates for policy inceptions between October 1, 1990 and September 30, 1992. It would then be replaced by the NJ Insurance| Free quotes and help online. (NJPAIP), a standard assigned risk arrangement. The MTF policies were to be serviced by four companies: Computer Sciences Corp., Warner Insurance Services, Policy Management Services Corp. and AMGRO. The complaint ratio for policies insured through the MTF and serviced by these four firms was almost 400 percent higher than those serviced by insurers in the voluntary market. 92 It has been pointed out that there is a general belief that service in the residual market is lower than in the voluntary market, and that rate suppression can lead to the poorer service provision. The complaint ratio in New Jersey lends support to that general belief. 93

The law ended the $222 surcharges mentioned above, but it introduced assessments on insurers. Insurers were required to pay 5% per year surtaxes which were to run from 1990 through 1992, and separate 2.7 percent assessments from 1990 to 1997 for the JUA. Insurers were prohibited from passing the cost on to policyholders, despite rate inadequacy. In fact, they were subsequently denied permission, by a 4-2 vote of the New Jersey State Supreme Court, to even deduct the cost of JUA charges from their Excess Profit reports. 94

Drivers with bad driving records could be surcharged for up to three years under the law change, but from the outset, actuaries warned that rates were too low and that MTF deficits would be massive. Len Freifelder, Special Deputy Commissioner, warned on the second day of the MTF's operations, that he estimated the pool would lose between $200 and $500 million a year. Both Milliman and Robertson, and William M. Mercer actuaries predicted over one billion dollar deficits, and recommended massive rate hikes. The Commissioner claimed he was unaware that the MTF would run a deficit, but the industry took him to court in an attempt to avoid bearing the cost burden imposed by grossly inadequate rates. 95 The industry eventually settled and swallowed a half a billion of the deficit. The state Economic Development Authority issued $750 million of bonds to help pay the rest. The bonds were to be repaid by surcharges on motorists with 6 or more points on their drivers licenses. 96 Despite the warnings of Fellows of the Casualty Actuarial Society, the Market Transition Facility (MTF) Advisory Board was not listening. Ten insurers left the state between 1990 and 1993, and others wanted to do so. 97 (See Table 10)

As the MTF deficit mounted, the pool was gradually being depopulated. Between 1990 and 1993, the number of insured vehicle years in the pool dropped from 1,336,325 to 97,317, or from 33.6% of total insured car years to 2.2%. The principle cause of the pool depopulation was that The FAIR Act had a "Take-all-comers" rule. Insurers could not refuse to take any risk with less than nine motor vehicle points. Insurers were not having much success with attempts to get prior approval rate increases for their "voluntary business" either. Allstate and Aetna sued the Commissioner for failing to respond to rate requests within the time stipulated in the law (30 days), and 11 insurers filed for rate increases within one year of the inception of the FAIR Act. 98 From January 1990 through March of 1993, only three prior approval rate filings were approved. 99 Although the MTF's tenure ended with policies expiring on September 30, 1993, it is still paying claims and defending suits. The New Jersey Insurance Department reported that the MTF paid $70 million to lawyers in 1995, and $40 million to lawyers in the year ending September 1996. 100 In 2006, the Insurance Commissioner announced that 7,013 claims payments, many of which had been delayed for up to eighteen months, were mailed in February. There were still 4500 claims outstanding, but payments were finally being made on time. 101 Governor Whitman had won re-election, and her reform package, described above, was about to be adopted.

NEW JERSEY MARKET

CONCENTRATION

Although one might be tempted to question the need for an analysis of market structure in a state where analysis of the residual market, and firm behavior send strong signals that market power is highly unlikely to have resulted in excess profits, the data set out in Table 8 for the New Jersey private passenger automobile insurance business indicate that concentration is low. There is research which indicates that strict prior approval can lead to fewer insurers and affect the shares by distribution system, 102 but both concentration ratios and Herfindahl indices found in Table 8 are lower than those usually seen in concentrated industries. Although the three Herfindahl Indices are

(INSERT TABLE 8 ABOUT HERE)

low, they have increased over the 19 sample observations for PIP and liability coverages , and fallen slightly for physical damage. Each, however, is still lower than the median state Herfindahls reported by Cummins, 103 and even with the slight increases noted, none are even in the "mildly concentrated" category. It has been argued that market power measures can be misleading in the sense that firms may have greater local market power than revealed in aggregations at the state level. The data available for this study does not allow for territorial desegregations, but the Insurance Department, as noted above, is now collecting such data. Casual empiricism leads me to believe that the argument has less power in a state as small as New Jersey.

It is worth noting that the four firm concentration ratios have increased for each of the three coverages displayed in Table 8. However, the four firm concentration ratios are in the 50 percent range, which are below the state median reported by Cummins in his countrywide study, 104 and have been generated by insurers with shares in the 10 to 15 percent range (Allstate, 16.2%; State Farm, 15.8%; New Jersey Manufacturers, 11.9%; and Prudential, 10.7%), 105 and not by one firm with a share of 47%. Perhaps, under different circumstances, one could argue that firms in a market suffer high costs to discourage entry, but such an argument in the New Jersey market in the face of years of evidence of rate suppression appears farcical on its face.

DIRECT WRITER MARKET SHARE

The data set available for this study does not permit examination of the Stock-Mutual market and expense shares over time, but does provide a time series for direct writers and independent agency business. There were also five years of data available for brokerage and other business, but they constitute less than three percent of the writings in the New Jersey private passenger business. Table 9 reveals that between 1980 and 2006,

(INSERT TABLE 9 ABOUT HERE)

the market share of direct writers, as a percentage of direct and independent agency business, grew from 58.4% to 70.3% for PIP, from 60.1% to 70.0% liability (bodily injury and property damage), and from 61.9 to 68.2% for physical damage. It is quite striking, however, that direct writers have increased their shares relative to independent agency business since the demise of the Joint Underwriting Association-Market Transition Facility regime. From 1980 to 1992, the direct writers' shares increased 2.5 and 0.3 percentage points for PIP and liability, and fell 4.3 percentage points for physical damage. But from 1992 to 2006, they increased: 8.4 percentage points for PIP, 7.7 percentage points for liability(BI and PD), and 10.7 percentage points for physical damage. These translate into 13.6, 12.4, and 18.5 percent increases in PIP, liability and physical damages, respectively - huge increases in just a six year period. Sharon Tennyson studied structure in the private passenger automobile insurance business and found that the share of both the direct writers and the four largest private passenger insurers in the country were lower in New Jersey than in unregulated states, 106 which is not inconsistent with the current data. In 1992, the share of direct writers was lower. Two of the largest insurers do not write in New Jersey, hence one would expect the market share of the four largest firms to be lower than in other states. The reason why they have not written in a state with a market with over 5 million registered private passenger vehicles (See Table 13) is obvious. Rates have been too low, and the regulatory climate too uncertain.

ENTRY AND EXIT

If rates are too low, and quality can not be reduced enough, 107 one should see firms exiting the market. New Jersey is an interesting case both because there has been exit, and it has clustered, in part, around law change; but also because the state has fought exit. Rate compression and suppression can lead to "beggar thy neighbor schemes." 108 Rate regulation and allowed rates of return based on a "whole firm" theory enable cross line subsidy and not just subsidy within a line and across coverages within that line. If each activity is not allowed to stand on its own merits, with its price reflecting its value at the margin, capital will not be allocated to its most appropriate uses. States which continually suppress rates may hope that their neighbors in other states will subsidize them, or that they can pursue social policy within their own state by redistributing income by cross line subsidies. Cummins and Weiss (1991) have pointed out that entry is relatively easy in property-liability markets, and they have drawn attention to states' schemes to require license surrender in all lines and continued participation in pool deficits for a number of years after market withdrawal. 109 The New Jersey private passenger automobile insurance business epitomizes their analysis.

New Jersey not only forbade exit without license surrender, but it also granted exit only after an "agreed" waiting period of up to five years. For example, Atlantic Employers got permission in 1990 to start withdrawing in January 1996. In 2004, it petitioned to leave earlier, after filing for a 104 percent prior approval rate increase. It had a two percent market share, over half of that of the new entrants listed in Table 6 below. 110 Twin City Fire Insurance sued claiming that the requirement to surrender its other lines in order to exit the private passenger automobile insurance business was an "unlawful taking." The State Supreme Court (7-0) unanimously upheld the FAIR Act provision, and a significant barrier to exit was sustained. Between the passage of the Fair Act in March of 1990 and the Supreme Court's decision on July 29, 1992, forty percent (32 firms) had applied to withdraw from the New Jersey market. 111 In Table 10, I list the

(INSERT TABLE 10 ABOUT HERE)

firms which have exited the market, and those that have entered, together with an estimate of the share of in-force policies they had in late 1997.

One of the new entrants has already folded. Another has a Weiss Rating of E (very weak). One, Metropolitan, had a license in New Jersey but it had refused to write any business. It entered with the proviso that it could exit if the market was sour. 112 On August 27, 1997, the Commissioner of Insurance petitioned the State Superior Court seeking an Order of Liquidation for Home State Insurance. At that time they had approximately 40,000 private passenger, and 1,000 commercial, policies in force. 113 New Jersey has a guarantee fund, the Property-Liability Insurance Guaranty Association, which will pay the claims of Home State Insurance. Members of the Association are surcharged one-third of one percent of Direct Net Written Premium, but they can charge their policyholders to recover the assessment. 114

The remaining new entrants had about 3.85% of the policies in force in late 2006. For comparison purposes, the national market shares of SAFECO, Nationwide and Progressive, total about 10 percent. They exited the market during the erosion of the $200 Threshold (see Table 6) and the run-up to the formation of the Joint Underwriting Association (JUA). As seen in Table 10, fourteen firms left the market from 1990 to 2004, a period which roughly coincides with the passage of the FAIR Act (March, 1990) and the expiration of the last Market Transition Facility (MTF) policy (September 30, 1993). Many of those who left are major property-liability and private passenger writers.

(INSERT TABLE 11 ABOUT HERE)

Table 11 shows that the number of firms writing private passenger automobile insurance has declined over the 1980 to 2006 period. The number of firms writing physical damage has dropped from 104 in 1980, to 64 in 2006, a decrease of 38.5%. For direct writers, there was a 50% decrease in the number of firms writing physical damage over the sample period. The number of firms writing private passenger automobile insurance nationally did not decrease significantly over the same period. The change in the distribution of firms in New Jersey can be attributable to both mergers and acquisitions, and entry and exit. The net result is that there are 15 fewer direct writers and 25 fewer independent agency companies doing private passenger physical damage business in New Jersey. It is worth noting that the net loss for each line, PIP, liability (BI and PD) and physical damage, since the end of the Market Transition Facility is only one direct writer; however, it is 13 (PIP), 10 (liability) and 12 (physical damage) for independent agency firms. Direct writers are capturing larger market shares, and the number of independent agency firms has declined much more rapidly than direct writers.

New Jersey has also seen the formation of New Jersey Only wholly owned subsidiaries to shield the assets of the parent from the negative effects of the New Jersey market. Note the sudden increase in New Jersey Only firms which coincides with the Florio administration and the inception of the FAIR Act (1990) and the Market Transition Facility, which started with policy inceptions in October of 1990 (See Table 6). Not only were firms applying to leave the market, some were also transferring massive amounts of their private passenger business into their newly formed subsidiaries. The Insurance Department stated that between 1990 and 1993, 1.3 million of the state's 4.3 million drivers were moved to New Jersey Only, wholly owned subsidiaries. Prudential Property and Casualty Company of New Jersey (487,887) and State Farm Indemnity Co. (287,077) had over three quarters of a million insured's on June 30, 1993, alone. 115

In May of 2006, the New Jersey Department of Insurance fined General Accident Insurance Co. Of America over a half million dollars for shrinking its market share and turning away customers. A Department spokesperson said, "We believe they were trying to withdraw from the state entirely." The Department fined four other insurers for similar activity in the preceding year. 116 Despite General Accident's attempt to protect its Surplus, some insurers may have been more optimistic about the prospects for New Jersey. Progressive and GEICO were reportedly considering re-entering the market, and two insurers, albeit small ones, The Provident Washington Company and State National Insurance Company filed applications in the summer of 1999 to write business in New Jersey. 117 Provident Washington has about 1350 policies in force in the market.

Multivariate analysis is more appropriate to determine the marginal effects of regulation on the distribution system and on the number of firms writing in the market. The New Jersey experience presents a strong prima facie case that the law and regulatory environment have affected both. Firms have departed, more have tried. In the absence of the state's barrier to exit, and its support by the State Supreme Court, even more firms would have abandoned ship. Firms could not recover the opportunity cost of capital because they were saddled with inadequate rates, and the claims costs and pool population which would ensure high loss ratios.

AVAILABILITY AND THE RESIDUAL MARKET

The story of New Jersey's residual market is set out above. From the passage of elective "No-Fault" to the depopulation of the pool with the closing of the Market Transition Facility, New Jersey has relied extensively on the residual market mechanism. There are states and markets - New Jersey and Massachusetts private passenger, Maine's pre-reform Workers' Compensation - where rates have been so inadequate at times, that a large share of the state's risks have found themselves unable to secure coverage in the voluntary market. New Jersey has had binding price constraints, and the market has been in disarray. In Table 12, I set out the direct written premium for liability (bodily injury and property damage)

(INSERT TABLE 12 ABOUT HERE)

and physical damage coverages in the New Jersey and national residual markets. Although New Jersey was always a rate inadequacy "All Star" with over 20 percent of the entire United States residual market private passenger automobile insurance premium, it reached legendary status with the adoption of the Joint Underwriting Association in 1983, when it accounted for over one third of the entire countrywide residual market. The effects of the depopulation of the Market Transition Facility are visible, finally in 1992, and peaked in 1993 when New Jersey's residual market share was 3.7 percent. This is in line with the state's share of the entire countrywide market, voluntary and residual, which was approximately 4.4 percent of countrywide private passenger business in 1993. It is troubling to note that New Jersey's share of the countrywide pool has increase monotonically since its low point in 1993.

(INSERT TABLE 13 ABOUT HERE)

The JUA insured over 40 percent of written car years in the state from the mid Eighties until the passage of the flex rating provision, the $250,000 PIP limit, and the verbal threshold in 1988. Table 13, details the residual market activity from 1985 to 1997. Its share of written car years dropped with the depopulation of the Market Transition Facility, and it reached it low of 2.2 percent of written car years with the last policy expiration of the MTF in 1993.

The uninsured motorist problem has been severe in New Jersey. Affordability is an issue for many low income drivers, 118 and the Basic Policy described above is designed to help to attack it. It is impossible to know exactly the size of the problem, and various proxies, including liability claims in cases involving uninsured motorists have been used to get a sense of its magnitude. The Insurance Research Council estimates that 12 percent of New Jersey drivers are uninsured. In Table 13, Column 5, I set out the difference between registrations and written car years in New Jersey. For the latest year available, there is a 10.4% differential between registered vehicles and exposure units, a percentage close to the IRC research study finding. 119 The gap between registrations and written car years increased from 426,347 in 1985 to 1,210,558 in 1988. The differential remained around a million until 1993, when it dropped to 684,827. This was a drop of 31 percent in one year. Insurance prices were lower that year because the surcharging stopped. The state had a "Take-all-comers" rule in place. Many saw prices drop twenty percent, and some uninsured drivers probably came in from the driving cold. The "Take-all-comers" rule helps in that regard, but it also provides an opportunity for insurers to have to insure cheats.

FRAUD

Most evidence about fraud in New Jersey is anecdotal. Unfortunately, there is a great deal of it. The problem has been considered so serious that combating it was one of the four corner stones of the Automobile Insurance Cost Reduction Act. The new Office of Insurance Fraud has been established in the Division of Criminal Justice. During its first full year of operation, it opened 344 new criminal investigations, prosecuted 87 cases with 134 defendants, got 78 convictions and sent 20 percent of the convicted to jail. 120 Although the office is young, it has prosecuted "staged accidents (arrange for collisions)," "faked accidents, (accidents which have never occurred)" "give-up schemes (arrange for someone to get rid of a car)," and using "runners (to solicit accident victims as patients)." 121 These types of cases are not unique to New Jersey, and insurers have seen them all, but they are not the major source of fraudulent cost. That honor goes to the building up of costs associated with legitimate accidents. Richard Derrig and his colleagues have estimated the extent of the problem.

In 2004, the New Jersey Insurance Commissioner estimated that 10% of all insurance claims were fraudulent, and that the fraud rate was higher in automobile insurance. 122 The Insurance Research Council has estimated that roughly 20 percent of paid losses are generated by the fraud rate and claims build-up. 123 This estimate (1995 data) is identical to the estimate Governor Florio gave for New Jersey for 1991. 124 In a 2004 paper, Weisberg, Derrig and Chen reported that nearly half of 1989 bodily injury liability claims in the state of Massachusetts appeared to involve build-up. 125 The Automobile Insurers Bureau has an active research program to develop fraud detection measures for use in auto insurance. Their research team and colleagues have used linear models, 126 fuzzy set theory, 127 and neural network analysis 128 to improve methods of claims fraud detection. Such measures, coupled with strict enforcement and prosecution, are called for in the New Jersey market.

New Jersey borders two cities, New York and Philadelphia, which actually have auto insurance rates that are higher than some New Jersey territories. 129 Drivers in those cities can save money by fraudulently registering their cars in rural New Jersey areas. Between 1988 and the end of 1992, the Insurance Department received 30,000 rate evasion reports, and in 1990 the Fraud Squad found that approximately 40 percent of the 13,000 cases it investigated were claims submitted by rate evaders. 130 The Insurance Department conducted a Sunday morning raid in May of 1991. They found 1,147 cars with New Jersey tags parked in Philadelphia and New York City. A total of 944 were found to be fraudulent registrations. 131

New Jersey has been a home to fraud and theft rings. Allstate and other insurers have been aggressive in pursuing and prosecuting suspected systems abusers. In 1997, Allstate filed suit against 1200 people in North Jersey, the largest alleged fraud ring know in the country, 132 and in August, 2000, it filed against another 241 person alleged fraud ring in Camden County in South Jersey. 133 Camden Fire Insurance Company filed suit in October of 1996 against an 800 person alleged fraud ring in Passaic County. 134 New Jersey has seen virtually all of the fraudulent schemes seen elsewhere in the country, including: fake auto inspection stickers 135 and car theft rings. 136 Although auto theft is down, New Jersey has had four of the top 100 auto theft areas in the country. The National Insurance Crime Bureau reported that in 2004, one of every 32 cars were stolen in Jersey City, one in every 56 in Newark and one of every 93 in Trenton. This was an improvement on the one in thirty-eight, forty-eight and fifty-nine for those three cities in 1993. 137 Theft and fraud have played a major role in New Jersey's high claims costs.

HIGH CLAIMS COSTS

The law and its changes, and the cost factors set out above lead to our summary cost table for the New Jersey market. In Table 14, I set out the indices of claims severity for New Jersey and the United States using the 2000 Insurance Research Council (IRC) study of auto injury claims. 138 Claim severity has grown faster than medical care inflation

(INSERT TABLE 14 ABOUT HERE)

in bodily injury liability (BI), property damage liability (PD) and PIP in New Jersey, and it has grown faster in New Jersey than in the country. New Jersey started from a higher severity base for each claim type. Bodily injury liability (BI) has grown from $7,592 in 1980 to about $19,600 in 2006 (7,592 x 2.58 = $19,587). Countrywide severity is much lower, growing from a base of $4,955 in 1980 to around $9,600 in 2006, i.e. BI severity in New Jersey is over twice as great as countrywide BI severity. When New Jersey adopted the verbal threshold in 1988, PIP severity, as expected, increased. For the years from 1984 to 1992, the years of the JUA-MTF regimes, the IRC data sample did not include New Jersey residual market claims frequency and severity. Hence, to the extent that pool severity was greater than voluntary market severity, the indices in Table 14 are conservative for the period when New Jersey operated without an assigned risk plan. For its analysis of territorial frequency and severity, the IRC averages data over a three year period, and it uses both ISO and NAIL claims data for a broader exposure base. 139 On a base of 12.4 million earned car years in New Jersey, the IRC found frequency rates of 1.37 per 100 insured cars for bodily injury liability, 3.83 for property damage liability and 2.73 for PIP. The ratio of BI to PD claims was 35.7, but it was 75 or more in 5 of New Jersey's 27 territories. 140 Seventy five percent or more of property damage claims included BI involvement in those territories. The ratio of bodily injury liability to property damage liability provides a good indicator of both potential litigation and the severity of claims costs. Countrywide claim frequencies for 2006 were 1.17 (BI), 4.09 (PD) and 1.87 (PIP), with a BI to PD ratio of 28.7%. 141

CONCLUSION

New Jersey is an expensive place to do business. It is strictly regulated, labor and automobile repair costs are high; there may be upwards of a half million uninsured motorists; it has higher than average frequency for BI and PIP, and its BI to PD ratio is higher than countrywide, as well. State Senator Gerald Cardinale summarized his view of the rise in premiums in the state: "In short, the rise in insurance premiums in New Jersey has been partly caused by the unwillingness of many policyholders and others to accept the concept of a trade-off." 142 Among the others have been the legislators of the state. It is easier to install a benefit system which is rich, and to hide behind price suppression through regulation, than to admit that we live in a world of constraints. New Jersey is a state where it is incredibly difficult to get rate relief. The law has changed in the last two years, but rate regulation has not: 29 of 32 rate filings were rejected, and the three others got partial approval. 143 Other requests for rate increases are pending (See Table 2).

The state law is an exercise in social engineering, affecting firms' market shares, dictating whom a firm can take as a customer and preventing market exit. The state probably has fewer firms writing business and less competition than it would, under different regulatory schemes. The law and its administration have subjected drivers and insurers to unnecessary costs and burdened them with needless administration. It has limited the choices that would enable families with different resource levels to make insurance selections that are in their own best interest.

There have been recent improvements. Re-instituting the assigned risk plan with market driven incentives for cost containment, providing choice in PIP benefit levels, strengthening the fraud provisions of the law, and introducing tier rating are a few. But important reforms have still not come. It is time for New Jersey to join the deregulation bandwagon.

The market, left to its own devices, is a miraculous thing. It is more than capable of allocating capital and arriving at fair prices and returns. There is no need for a regulatory bureaucracy to impede competition, restrict choice and suppress price. Prior approval rate regulation should be dismantled in New Jersey. It is not even clear firms should be required to report prices before they use them; but if reporting were required, it should be for informational purposes only, i.e. use and file..

Willing parties are best suited for determining contract conditions. If the state feels a need to intervene, it should set minimum standards (e.g. 15/30/5), and get out of the way and let the market work. If the state wants to foster choice, it should allow for a true No-Fault choice (no tort period), as well as the verbal threshold and the limitation on lawsuit options. If the people want true No-Fault and they have the ability to pay for it, someone will market it, and people will buy it. If they do not want it, it will take its place with the Beta Format in the dustbin of video market history.

The Insurance Department can be an information provider, as they are attempting to do now with their price comparisons, giving consumers information on cost savings under different choices. All rates, plans, convention blanks, filing information, insurance expense exhibits, paid for by consumers who purchase the insurance product, should be posted to the departmental web page.

The industry has minimal anti-trust exemption. It does not need it, and the free enterprise system will do quite well without it. The state, court rulings notwithstanding, should not be in the business of profit regulation. Excess profit laws are anti-competitive. They not only hurt the lucky, they hurt the efficient. They are barriers to entry, require needless reports and waste taxpayer and policyowner money. Each line of insurance and each coverage sold by a company should meet the market test of competition. Someone who owns a home in suburban New Jersey and does not drive should not subsidize the driving excesses of crazed 18 year olds.

The state should eliminate barriers to exit. Although entry is relatively easy, firms will think three times about entering the market, if they know they can not get out. We need an optimal number of Metropolitans, and they should not be selected by regulators, but rather by the pursuit of economic self interest.

Crime pays. If we want to continue to fight fraud, we should lower the net benefits by increasing the costs of engaging in the crime, and lowering the probability of success. New Jersey has made a leap forward with its Fraud Prosecutor, but the penalties for both auto theft and fraud should be increased. Similarly, we need to know when someone is driving an uninsured vehicle. If we are going to require compulsory liability insurance, we should get serious about a National Center for Insurance Coverage. With computer technology, we should know within 24 hours when any policy lapses. When one does, tags should come off of a vehicle, and fraudulent tag use should bear a heavy fine or imprisonment. There is an optimal amount of fraud, an optimal number of insolvencies, and an optimal number of uninsured motorists. It is not at all clear that state coercion will produce them. It is clear that the provision of greater information and less regulation and bureaucracy will improve efficiency in the New Jersey market.

Illinois has had a free market for private passenger automobile insurance for over twenty years. The sky did not fall when Illinois adopted open competition, and private passenger automobile insurance has been depoliticized in that state. The political will to move to a free market probably does not currently exist in the New Jersey. However, the legislature could make moves that would provide a transition to open competition. Any legislative changes (eliminating the "Take-all-Comers" rule, removing barriers to entry and exit, providing greater flexibility in pricing, repealing excess profit provisions etc.) can not be seen as temporary. If New Jersey wants to capture more of the benefits of competition, insurers who want to leave the state, as well as those we would like to attract to our market, must be convinced that they will have every opportunity to compete for business and normal profits without the fear that they will be held captive by the state if they remain or enter the New Jersey private passenger market.

As this book goes to press, the new rating territories proposed by the commission have yet to be adopted. Not all of the medical cost rules designed to reduce costs have been finalized. Some of the optimism which greeted the passage of the reforms of 2006 and the 15 percent mandatory rate roll back is fading. The set and size of rate increase applications pending (See Table 2), and the rumblings in the state capital remind us that the legislature is up for election in the fall. The governor's race will heat up this summer. Whoever wins will have New Jersey automobile insurance on his mind.



1 Hawaii nosed New Jersey out for the most expensive state in 1992. See Joseph Donohue, "Car Rates Drop but Stay High." Newark Star-Ledger, Feb. 14, 2004.

2 New Jersey has a population of 8,414,350 according to the 2000 Census. The land area of the state is 7.418.84 square miles. Vehicle registrations by year are given below. The vehicles per square mile is for 1997.

3 See Mark J. Browne and Brenda J. Wells, "Claims Adjudication in the Personal Automobile Insurance Residual Market." The Journal of Risk and Insurance, 1999, Vol. 66, No. 2, p. 227.

4 For an example of a recent rate decision, see John R. Tassini, ALJ. In the Matter of GSA Insurance Company's Request for a Private Passenger Automobile Insurance Rate Increase, Initial Decision, OAL DKT. No. BKI 10882-98. Agency DKT. 98-2514. Decided Jan. 6, 2000.

5 N.J.S.A. 39:6A-1.1 et. seq.

6 Victoria Sonshine Pasher reports that Miller, Rapp, Herbers & Terry, Inc. Priced the bill for the American Insurance Association and estimated a 3 percent cost savings. Tillinghast Towers and Perrin did a study for State Farm, and estimated that an earlier version of the bill would save 2 percent. "Studies Seek to Steer N.J. Auto Reform." National Underwriter, Pro/Cas Edition, Vol. 102, No. 19, May 11, 2006.

7 Edward L. Lascher, Jr. The Politics of Automobile Insurance Reform, Georgetown University Press, 2004, p. 47.

8 Marjorie M. Bertie, (1991) Hit Me - I Need the Money. San Francisco: ICS Press, p 66.

9 Flex rating enables insurers to change rates within small percentage bands, three percent in New Jersey, without formal rate hearings, as long as the change would not result in rates that are inadequate, excessive or unfairly discriminatory.

10 N.J.S.A. 39:6A8(a).

11 Territorial caps are discussed below. See Caps.

12 N.J. Dept. Of Banking and Insurance, "Basic Auto Policy Becomes Available." New Jersey Insurance Reporter, Winter 1999. See N.J.S.A. 39:6A-3.1(a) for a full description of the Basic Policy coverage.

13 N.J. Div. Of Banking and Insurance, Order A99-102, "In the Matter of Reduction in Rates to be Made Pursuant to Section 67 of the Automobile Insurance Cost Reduction Act. P.L. 2006, c. 21." January 4, 1999.

14 Joe Donohue, "Reforms Driving Down the Cost of Insuring Those Jersey Wheels." Newark Star-Ledger, Sep. 29, 1999.

15 Joe Donohue, "Prudential Cuts Auto Rates 2.9 percent - However, New Jersey Re-Insurance Receives Permission to Boost Its Rates by 8.9 Percent." Newark Star-Ledger, June 17, 2000.

16 Department of Banking and Insurance, 2006-1999 Annual Report, p. 10.

17 N.J.S.A. 17:29A-14(a).

18 N.J.S.A. 17:29A-14(c).

19 Ibid.

20 N.J.S.A. 17:29A-14(e).

21 N.J.S.A. 17:29A-15.

22 N.J.S.A. 17:29A-15.1

23 N.J. Dept. of Banking and Insurance, Press Release, June 16, 2000.

24 See Sharon Tennyson, "Regulatory Lag in Automobile Insurance," Journal of Risk and Insurance, Vol. 60, No. 1, 1993, for an excellent discussion of intended rate relief delay.

25 N.J. Dept. of Banking and Insurance, Press Release, May 5, 2000. The Department's release did not specify whether the loss wan an underwriting or net income loss.

26 These rate filing statistics were provided by the New Jersey Department of Banking and insurance in private correspondence, January 4, 2001.

27 Sandy McClure, "Legislator: No Car Insurance Rate Hikes." Courier-Post, April 12, 2001.

28 See Eugene Kiely, "N.J. Auto Insurers' Push Could Erase Rate Cuts." Philadelphia Inquirer, April 17, 2001. Mr. Kiely writes that state officials indicated State Farm lost $147 million in 1999 and $128 million in 2000. The article does not state if these are underwriting ot total losses.

29 N.J.S.A. 17:29A-46.8. These potential interveners can not engage in intervention in "Expedited Filings" to be defined below.

30 Joe Donohue and David Wald, "Car Rate Strategy Takes a Detour Political Sidestep by McGreevey Camp." Newark Star-Ledger, Sept. 4, 1997. McGreevey lost to Whitman in a very close and hotly contested race.

31 "New Jersey Race Ends with Undecided Results," The Daily Tar Heel, Nov. 5, 1997.

32 N.J.S.A. 17:29A-46.6.

33 Barrett Carter, "State Auto Insurers Gain a Split in Court," Newark-Star Ledger, July 30, 1992.

34 See Joe Donohue, "Car Insurance 'Flexes' With 5.18% Boost." Newark-Star Ledger, July 2, 2004, and "Auto Insurers Get 3.83 pct. 'Flex-rate Hike." Newark Star-Ledger, June 20, 1995.

35 N.J.S.A. 17:29A-36(c). These caps are exclusive of surcharges and discounts.

36 See J. David Cummins and Mary A. Weiss, (1992) "Regulation and the Automobile Insurance Crisis." Regulation: The Cato Review of Business and Government, Vol. 15, No. 2.; see Dwight K Bartlett, Robert W. Klein and David T. Russell, "Attempts to Socialize Insurance Costs in Voluntary Insurance Markets: The Historical Record." Journal of Insurance Regulation, Summer 1999, Vol. 17, No. 4, 478-511.

37 Victoria Sonshine Pasher, National Underwriter, May 11, 2006.

38 Donohue, "Car Rate Strategy Takes a Detour Political Sidestep by McGreevey Camp."

39 N.J.S.A. 17:33C-2.

40 Scott E. Harrington and Greg Niehaus, "Race, Redlining, and Automobile Prices." Journal of Business, 2006, Vol. 71, No. 3, p. 464.

41 New Jersey Dept. Of Banking and Insurance, "Urban Enterprise Zones Making Auto Insurance More Accessible in Cities." New Jersey Insurance Reporter, Fall 1999. The Urban Zone Assigned Risk is managed by NJPAIP, which will be discussed below. The state has a Take-all-comers rule. Drivers with enough points for accidents, moving violations, drunk driving convictions can be refused coverage in the voluntary market. Such drivers find themselves in New Jersey's Assigned Risk Plan. The Assigned Risk Plan should not be confused with Urban Zone Assigned Risk.

42 N.J. Dept. Of Banking and Insurance, Bulletin 00-13 and Bulletin 99-06.

43 N.J. Dept of Banking and Insurance, Bulletin 98-03.

44 See the FAIR Act below.

45 Joe Donahue, "New Jersey Auto Insurance Market Nearing End of Road to Recovery," Newark Star-Ledger, Feb. 21, 1993.

46 Donohue and Wald, op. cit.

47 For an excellent discussion of rate compression and suppression, see Dwight M. Jaffee and Thomas Russell, "The Causes and Consequences of Rate Regulation in the Auto Insurance Industry," in David F. Bradford, (ed.), The Economics of Property-Casualty Insurance. University of Chicago Press, 2006.

48 N.J.S.A. 17:29A-48.

49 For example, see 2006-1999 Annual Report, p. 11; and Robert Schwaneberg, Newark Star- Ledger, Aug. 2, 2006

50 N.J. Department of Banking and Insurance, Order No.: A99-135. "In the Matter of the Filing of Quarterly Tier Exposure Reports by All Insurers Writing Voluntary Private Passenger Automobile Insurance." May 3, 1999.

51 N.J. Dept. Of Banking and Insurance. Auto Insurers' Tier Filing Status. Updated Nov. 17, 2006. (Http://states.naic.org/nj/tiergraf.htm; accessed July 18, 2000). ISO, briefly discussed below under the 1988 law changes, files only loss costs. Its risk rating method, however, can be adopted by companies who must then use their own expense and profit provisions.

52 N.J. Dept. Of Banking and Insurance. Consumer Handbook on Tier Rating 2000.

53 J.D. Cummins and others, 1974. Consumer Attitudes Toward Auto and Homeowners Insurance. Phila.: University of Pennsylvania. See the discussion in Harris Schlesinger and J-Matthias Graf von der Schulenburg. "Consumer Information and Decisions to Switch Insurers." Journal of Risk and Insurance, 60, No. 4, 591-615.

54 Donohue, Newark Star-Ledger, Sep. 29, 1999.

55 Insurance Council of New Jersey, Mar. 20, 2000.

56 N.J. Dept. Of Banking and Insurance, Automobile Insurance Shopping Guide 2000, August 2000.

57 See N.J.S.A. 17:29A5.6 to 5.13 for more gory details.

58 N.J. Dept. of Banking and Insurance, "Two Auto Insurance Companies Must Return Excess Profit to Consumers," Press Release, April 3, 2006.

59 Annual Report 2006-9, p. 11.

60 Direct written premium is the total amount charged a policyholder less return and additional premium. Return and additional premium could be generated by additional insurance purchased during the year, or by changes to the coverage purchased. Direct written premium does not include reinsurance assumed or ceded.

61 N. J. Dept. of Banking and Insurance, "Hartford Group Must Return $3.2 Million to Consumers." New Jersey Insurance Reporter, Fall 1999.

62 The Market Transition Facility stopped issuing policies in October of 1992. Those policies expire in October 1993. The Joint Underwriting Association and the MTF will be discussed below.

63 See Harrington, "An Econometric Analysis of Insurance Rate Regulation," in this volume for a taxonomy of prior approval and competitive rating laws.

64 J. David Cummins has pointed to the collapse of the voluntary business in New Jersey. See his characterization in Cummins and Weiss (1992) Cato and in the popular press. See, Herb Jaffe, "Still an Assigned Risk Auto Rates Fall...But Face a Crossroad," Newark Star- Ledger, Sept. 20, 1992.

65 See J. David Cummins and Mary A. Weiss, "The Structure, Conduct and Regulation of the Property-Liability Insurance Industry," in Richard W. Kopcke and Richard E. Randall (eds.), The Financial Condition and Regulation of Insurance Companies: Conference Series No. 35. The Federal Reserve Bank of Boston, 1991. They discuss costs, pricing rules and the need for appropriate incentives for cost containment.

66 Joe Donohue, "Lawyer Lobby Targets No-Fault Auto Insurance." Newark Star-Ledger, Dec. 15, 2004.

67 David T. Beale, Leonard Shambon and Anne Hogan, Automobile Insurance in New Jersey. Princeton: Center for Analysis of Public Issues, 1970. See p. 24-27 for rate hearing discussion.

68 Robert E. Keeton and Jeffrey O'Connell, The Need for Legal and Insurance Reform. Homewood, Ill.: Dow Jones-Irwin, Inc., 1967.

69 Beale, Shambon and Hogan, p. 46.

70 Ibid., p. 51.

71 Herman Hansler, An Assessment of the New Jersey No-Fault Automobile Reformation Act. New Jersey Department of Insurance, December 1982.

72 Ibid., p. 21.

73 Ibid. Table VIII, p. 47.

74 Ibid., p. 5.

75 Ibid., Table VI, p. 45.

76 P.L. 1983, c. 362.

77 AIPSO, AIPSO FACTS 1991/92, p. 12.

78 Public Hearing before the Senate Commerce Committee, May 12, 1997.

79 Insurance Information Institute, Auto Insurance Issues, Appendix B: The Case of New Jersey, February 8, 1989.

80 The $1.6 billion in Table 5 represents the independent agency and direct writers' direct written premium only.

81 Joe Donohue, "Auto Insurance Out of the Political Hotseat," Newark Star-Ledger, Oct. 19, 1993

82 Joe Donohue, "Auto Insurers Admit Rates Are Lower But Say The Figures Are an Illusion, Newark Star-Ledger, Mar. 31, 1993.

83 Vincent R. Zarate, "SCI Launches In-depth Probe of Auto Insurance Companies," Newark Star-Ledger, May 20, 1990.

84 Joe Donohue, "Settlement to Reduce Losses From JUA Pool," Newark Star-Ledger, Jan. 7, 2004.

85 P.L. 1988, c 119.

86 Insurance Research Council, Trends In Auto Injury Claims: 2000 Edition, Table A-32, 2000. The 1988 estimates are probably conservative since they are based on voluntary market data.

87 Public Hearing, May 12, 1997.

88 P.L. 1988, c. 119, s.2 and s.3.

89 See Alexander Milch, "July 1 to be More Than a New Month for Insurance Concerns in the State." Newark Star-Ledger, June 21, 1989.

90 P.L. 1992, c. 8.

91 "Unfinished Business," Newark Star Ledger, Dec. 26, 2004.

92 Joe Donohue, "Survey Finds Insurance Pool Firms Have Most Customer Gripes By Far." Newark Star-Ledger, April 1, 1993.

93 See Mark Pauly, Howard Kunreuther and Paul Kleindorfer, "Regulation and Quality Competition in the US Insurance Industry," in Jorg Finsinger and Mark V. Pauly, The Economics of Insurance Regulation, St. Martin's Press, 1986.

94 Kathy Barrett Carter, "Justices Slam Car Door on JUA Cost 'End Run'." Newark Star- Ledger, June 15, 1995.

95 Joe Donohue, "Insurance Firms Charge Fortunato Deliberately Ran Up MTF Deficit," Newark-Star Ledger, May 14, 2004.

96 Joe Donohue, "Whitman Backs Bonds for Auto Pool Deficit." Newark Star-Ledger, July 1, 2004.

97 See Table 6 and exit discussion below.

98 Tom Hester, "Court Faults Fortunato on His Slow Response to Two Rate Hike Requests. Appeals Panel Orders Hearings on Bid by Allstate and Aetna." Newark Star-Ledger, May 15, 1991.

99 Joe Donohue, "Auto Insurers Admit Rates Are Lower But Say Figures Are An Illusion. Industry Sees 'Artificial Reduction' in Current Bills." Newark Star-Ledger, Mar. 31, 1993.

100 Robert Schwanberg, "Legislators Can Represent Insurance Pool Ethics Panel Votes To Let Lawyer-Lawmakers Defend Auto-Policy Cases for Defunct Group." Newark Star-Ledger, Feb. 14, 1997.

101 Robert Schwanberg, "State Insurance Pool Begins On-Time Payments to Victims." Newark Star-Ledger, Feb. 18, 2006.

102 Susan J. Suponic and Sharon Tennyson, "Rate Regulation and the Industrial Organization of Automobile Insurance."in David F. Bradford (ed.), The Economics of Property-Casualty Insurance, University of Chicago Press, 2006, p. 127 & p. 135.

103 C. David Cummins and Sharon Tennyson, "Controlling Automobile Insurance Costs," Journal of Economic Perspectives. Vol. 6, No. 2, Spring 1992, p. 98.

104 Ibid.

105 New Jersey Department of Insurance, UEZ Company Results, Revised Feb. 19, 2006.

106 Sharon Tennyson, "The Impact of Rate Regulation on State Automobile Insurance Markets,' Journal of Insurance Regulation, Vol. 15, No. 4, Summer 1997.

107 See Pauly, Kunreuther and Kleindorfer, 1986.

108 If a company is unable to increase inadequate rates in a strictly regulated state, it may be able to increase rates in other states or lines of insurance to offset losses in the strictly regulated line.

109 Cummins and Weiss, 1991, pp. 132-135.

110 Joe Donohue, "Three Auto Insurers Seek Approval to Drop Policies Due to Big Losses." Newark Star-Ledger, July 20, 2004.

111 Barret Carter, "State Auto Insurers Gain Split in Court." Newark Star-Ledger, July 30, 1992. This measure is calculated as a fraction of firms at the group level.

112 Joe Donohue, "Market Perks Up for Auto Insurers. Jersey Companies See Gains, or at Least, Hope." Newark Star-Ledger, Dec. 7, 1997.

113 Press Release, New Jersey Department of Banking and Insurance, August 27, 1997.

114 Order No. A98-152, Dec. 1, 2006.

115 Joe Donohue, "Car Insurers Maneuver to Limit New Jersey Losses." Newark Star-Ledger, Jan 3, 2004. Since our data set has been aggregated to the group level, some of the large New Jersey writers would not be captured in the Table as Jersey Only writers, since they would show up as pups of parents that write business in other states.

116 Joseph N. DiStefano, "Philadelphia Insurer Fined for Refusing to Write New Jersey Policies." Philadelphia Inquirer, May 7, 2006.

117 Randy Diamond, "Reforms Tempt Some Major Auto Insurers. Firms Taking New Look at Covering New Jersey," Bergen Record, Nov. 29, 1999.

118 See Joan T. Schmit and Yu-Luen Ma, "Factors Affecting the Relative Incidence of Uninsured Motorists Claims." Journal of Risk and Insurance, Vol. 67, No. 2, 2000, for a literature review and brief discussion of the impact of price on the uninsured motorist population.

119 Insurance Research Council, Uninsured Motorists, 1999.

120 Office of the Fraud Prosecutor, Annual Report 1999, Trenton, N.J., March 1, 2000.

121 Ibid.

122 Herb Jaffe, "Insurance Chief Foresees Lower Rates for Drivers." Newark Star-Ledger, August 8, 2004.

123 Insurance Research Council, Fraud and Buildup in Auto Injury Claims: Pushing the Limits of the Auto Insurance System,1996.

124 Angela Stewart, "State Launches Crackdown on Car Rate Evaders," Newark Star-Ledger, Jan. 3, 1992.

125 Richard A. Derrig, H.L. Weisberg and X. Chen, "Behavioral Factors and Lotteries under No- Fault with a Monetary Threshold: A Study of Massachusetts Automobile Claims." Journal of Risk and Insurance, Vol. 61, No. 2, 2004.

126 H. I. Weisberg and R. A. Derrig, "Quantitative Methods for Detecting Fraudulent Automobile Bodily Injury Claims." Automobile Insurers Bureau, 1993.

127 Cummins and Derrig provide an excellent review of earlier econometic work, including three papers by Cummins in, J. David Cummins and Richard A. Derrig, "Fuzzy Trends in Property-Liability Insurance Claims Costs." Journal of Risk and Insurance, Vol. 60, No. 3, 1993.

128 Patrick L. Brockett, Xiaobua Xia and Richard A. Derrig, "Using Kohonen's Self- Organization Map to Uncover Automobile Bodily Injury Claims Fraud." Journal of Risk and Insurance, Vol. 65, No. 2, 2006.

129 See Eric Smith and Randall Wright, "Why is Automobile Insurance in Philadelphia So Damn Expensive." American Economic Review, Vol. 82, No. 4, 1992, for a discussion of high rates in cities, and the related UI problem.

130 Stewart, Jan. 3, 1993.

131 Ibid.

132 "Allstate Asks Court to Halt Litigation of Suspect's Claims." Best's Review Property/Casualty Edition, Vol 98, No. 12, April 2006.

133 Mark E. Ruquet, "Passaic Seen as 'Fraud Central'." National Underwriter, Property Casualty Edit., Vol. 104, No. 37, Sep. 11, 2000.

134 Tim Smart, "Accidentally on Purpose." Business Week, June 30, 1997.

135 Robert Hanley, "Fake Auto Inspection Stickers Lead to Arrest of 2 Merchants." New York Times, Feb. 11, 2000.

136 See Alan Feuer, "Police Break Up Car-Theft-Insurance Ring." New York Times Feb. 3, 2000, for just one example. Passiac County has been referred to as the Car Theft Capital of the World (see Ruquet, Sep. 11, 2000.).

137 National Insurance Crime Bureau, Vehicle Theft Rate Study, Jan. 1996.

138 The IRC is a division of The American Institute for Chartered Property Casualty Underwiters. The Institute is one of the leading educational organizations in the property liability area. It trains, examines and certifies insurance professionals. The Insurance Research Council, a non-profit research division of the Institute, has done a series of research studies on the costs of private passenger insurance. I have used their latest cost study in Table 10.

139 This information was provided by Elizabeth Sprinkel of the Insurance Research Council.

140 Appendix Table B-31, Insurance Research Council, 2000.

141 Table A-1, Insurance Research Council, 2000.

142 Public Hearing, 1997.

143 Although the percentage of the business insured by the assigned risk plan is low, it is worth recalling that New Jersey has a "Take-all-Comers" rule which mandates that insurers take risks. If insurers were free to reject risks when rates were grossly inadequate, the residual market share would probably increase rapidly.

BIBLIOGRAPHY

AIPSO, 1985-1999 (various issues) AIPSO FACTS: A Handbook for the Automobile Insurance Residual Market. Johnston, RI: AIPSO.

Bartlett, Dwight K., Robert W. Klein and David T. Russell, 1999, "Attempts to Socialize Insurance Costs in Voluntary insurance Markets: The Historical Record." Journal of Insurance Regulation 17: 478-511.

Beale, David T., Leonard Shambon and Anne Hogan, 1970, Automobile Insurance in New Jersey. Princeton: Center for Analysis of Public Issues.

Bertie, Marjorie M., 1991, Hit Me - I Need the Money. San Francisco: ICS Press.

Best's Review Prop/Cas Edition, 2006, "Allstate Asks Court to Halt Litigation of Suspect's Claims." April.

Brockett, Patrick L., Xiaobua Xia and Richard A. Derrig, 2006, "Using Kohonen's Self-Organization Map to Uncover Automobile Bodily Injury Claims Fraud." Journal of Risk and Insurance 65: 245-274.

Browne, Mark J. and Brenda J Wells, 1999, "Claims Adjudication in the Personal Automobile Insurance Residual Market." Journal of Risk and Insurance 66: 275-290.

Carter, Kathy Barrett. 1995, "Justices Slam Car Door on JUA Cost 'End Run'." Newark Star-Ledger. June 15.

Carter, Barrett, 1992, "State Auto Insurers Gain a Split in Court." Newark Star-Ledger. July 30.

Cummins, J. David and Richard A. Derrig, 1993, "Fuzzy Trends in Property-Liability Insurance Claims Costs." Journal of Risk and Insurance 60: 429-465.

Cummins, J. David, D. M. McGill, H. E. Winklevoss, and H.A. Zelten, 1974, Consumer Attitudes Toward Auto and Homeowners Insurance. Wharton School, University of Pennsylvania.

Cummins, J. David and Sharon Tennyson, 1992, "Controlling Automobile Insurance Costs." Journal of Economic Perspectives 6: 95-115.

Cummins, J. David and Mary Weiss, 1992, "Regulation and the Automobile Insurance Crisis." Regulation: The Cato Review of Business and Government 15: 48-59.

Cummins, J. David and Mary Weiss, 1991, "The Structure, Conduct and Regulation of the Property-Liability Insurance Industry," in Richard W. Kopcke and Richard E. Randall (eds.), The Financial Condition and Regulation of Insurance Companies: Conference Series No. 35. The Federal Reserve Bank of Boston.

Diamond, Randy, 1999, "Reforms Tempt Some Major Auto Insurers. Firms Taking New Look at Covering New Jersey." Bergen Record. November 29.

Derrig, Richard A., H.L. Weisberg and X. Chen, 2004, "Behavioral Factors and Lotteries under No-Fault with a Monetary Threshold: A Study of Massachusetts Automobile Claims." Journal of Risk and Insurance 61: 245-275.

DiStefano, Joseph N., 2006, "Philadelphia Insurer Fined for Refusing to Write New Jersey Policies." Philadelphia Inquirer. May 7.

Donohue, Joe and David Wald, 1997, "Car Rate Strategy Takes a Detour: Political Sidestep by McGreevey Camp." Newark Star-Ledger. September 4.

Donohue, Joe, 2000, "Prudential Cuts Auto Rates 2.9 Percent - However, New Jersey Re-Insurance Receives Permission to Boost Its Rates by 8.9 Percent." Newark Star-Ledger. June 17.

Donohue, Joe, 1999, "Reforms Driving Down the Cost of Insuring Those Jersey Wheels." Newark Star-Ledger. September 29.

Donohue, Joe, 1997, "Market Perks Up for Auto Insurers. Jersey Companies See Gains, or at Least, Hope." Newark Star-Ledger. December 7.

Donohue, Joe, 1995, "Auto Insurers Get 3.83 pct. 'Flex-rate Hike." Newark Star-Ledger. June 20.

Donohue, Joe, 2004, "Lawyer Lobby Targets No-Fault Auto Insurance." Newark Star-Ledger. December 15.

Donohue, Joe, 2004, "Three Auto Insurers Seek Approval to Drop Policies Due to Big Losses." Newark Star-Ledger. July 20.

Donohue, Joe, 2004, "Car Insurance 'Flexes' With 5.18% Boost." Newark Star-Ledger. July 2.

Donohue, Joe, 2004, "Insurance Firms Charge Fortunato Deliberately Ran Up MTF Deficit." Newark Star-Ledger. May 14.

Donohue, Joe, 2004, "Settlement to Reduce Losses From JUA Pool." Newark Star-Ledger. January 7.

Donohue, Joe, 2004, "Car Insurers Maneuver to Limit New Jersey Losses." Newark Star-Ledger. January 3.

Donohue, Joe, 1993, "Auto Insurance Out of the Political Hotseat." Newark Star-Ledger. October 19.

Donohue, Joe, 1993, "Survey Finds Insurance Pool Firms Have Most Customer Gripes By Far." Newark Star-Ledger. April 1, 1993.

Donohue, Joe, 1993, "Auto Insurers Admit Rates Are Lower But Say the Figures Are an Illusion." Newark Star-Ledger. March 31.

Donohue, Joe, 1993, "New Jersey Auto Insurance Market Nearing End of Road to Recovery." Newark Star-Ledger. February 21.

Feuer, Alan, 2000, "Police Break Up Car-Theft-Insurance-Ring." New York Times. February 3.

Hanley, Robert, 2000, "Fake Auto Inspection Stickers Lead to Arrests of 2 Merchants." New York Times. February 11.

Hansler, Herman, 1982, An Assessment of the New Jersey No-Fault Automobile Insurance Reformation Act. Trenton, NJ: New Jersey Department of Banking and Insurance. December.

Harrington, Scott E., 2001, "An Econometric Analysis of Insurance Rate Regulation." This volume.

Harrington, Scott E. And Greg Niehaus, 2006, "Race, Redlining, and Automobile Prices." Journal of Business 71: 439-469.

Hester, Tom, 1991, "Court Faults Fortunato on His Slow Response to Two Rate Hike Requests. Appeals Panel Orders Hearings on Bid by Allstate and Aetna." Newark Star-Ledger. May 15.

Insurance Information Institute, 1989, Auto Insurance Issues. NY: III.

Insurance Research Council, 2000, Trends in Auto Injury Claims: 2000 Edition. Malvern, PA: IRC.

Insurance Research Council, 1999, Uninsured Motorists. Malvern, PA: IRC.

Insurance Research Council, 1996, Fraud and Buildup in Auto Injury Claims: Pushing the Limits of the Auto Insurance System. Malvern, PA: IRC.

Jaffee, Dwight M. And Thomas Russell, 2006, "The Causes and Consequences of Rate Regulation in the Auto Insurance Industry," in David F. Bradford (ed.), The Economics of Property Casualty Insurance. University of Chicago Press.

Jaffe, Herb, 2004, "Insurance Chief Foresees Lower Rates for Drivers." Newark Star-Ledger. August 8.

Jaffe, Herb, 1992, "Still an Assigned Risk Auto Rates Fall...But Face a Crossroad." Newark Star-Ledger. September 20.

Keeton, Robert E. And Jeffrey O'Connell, 1967, The Need for Legal and Insurance Reform. Homewood, Ill.: Dow Jones-Irwin, Inc.

Kiely, Eugene, 2001, "N.J. Auto Insurers' Push Could Erase Rate Cuts," The Philadelphia Inquirer, April 11.

Lascher, Edward L. Jr., 2004, The Politics of Automobile Insurance Reform. Georgetown University Press.

McClure, Sandy, 2001, "Legislator: No Car Insurance Rate Hikes," Courier Post, April 12.

Milch, Alexander, 1989, "July 1 to be More Than a New Month for Insurance Concerns in the State." Newark Star-Ledger. June 21.

National Association of Insurance Commissioners, 1999, State Average Expenditures & Premiums for Personal Automobile Insurance in 2006. Kansas City, MO: NAIA.

National Insurance Crime Bureau, 1996, Vehicle Theft Rate Study. Palos Hills, Ill.: NICB.

New Jersey Department of Banking and Insurance, 2000, Bulletin 00-13.

New Jersey Department of Banking and Insurance, 2000, 2006-1999 Annual Report.

New Jersey Department of Banking and Insurance, 2000, Consumer Handbook on Tier Rating 2000.

New Jersey Department of Banking and Insurance, 2000, Automobile Insurance Shopping Guide 2000.

New Jersey Department of Banking and Insurance, 2000, 1999 Private Passenger Auto Insurance Premium Comparison.

New Jersey Department of Banking and Insurance, 1999, "Basic Auto Policy Becomes Available." New Jersey Insurance Reporter. Winter.

New Jersey Department of Banking and Insurance, 1999, "Urban Enterprise Zones Making Auto Insurance More Accessible in Cities." New Jersey Insurance Reporter. Fall.

New Jersey Department of Banking and Insurance, 1999, "Hartford Group Must Return $3.2 Million to Consumers." New Jersey Insurance Reporter. Fall.

New Jersey Department of Banking and Insurance, 1999, Bulletin 99-06.

New Jersey Department of Banking and Insurance, 1999, "In the Matter of Reduction in Rates to be Made Pursuant to Section 67 of the Automobile Insurance Cost Reduction Act P.L. 1988, c. 21." Order A99-102. January 4.

New Jersey Department of Banking and Insurance, 1999, "In the Matter of the Filing of Quarterly Tier Exposure Reports by All Insurers Writing Voluntary Private Passenger Automobile Insurance." Order A99-135. May 3.

New Jersey Department of Banking and Insurance, 2006, "In the Matter of the Imposition of a Surcharge for Recoupment of the Property-Liability Insurance Order Guaranty Association Assessment Due November 2, 2006, and Imposed Pursuant to N.J.S.A. 17:30A-8a(3)." Order A98-152. December 1.

New Jersey Department of Banking and Insurance, 2006, Auto Insurers Tier Filing Status. Http://states.naic.org/nj.tiergraf.htm accessed July 18, 2000 .

New Jersey Department of Banking and Insurance, 2006, UEZ Company Results. Revised February 19.

New Jersey Department of Law & Public Safety, 2000, Office of Insurance Fraud Prosecutor, Annual Report 1999. February 29.

New Jersey State Senate Commerce Committee, 1997, Public Hearing. May 12.

Pasher, Victoria Sonshine, 2006, "Studies Seek to Steer N.J. Auto Reform." National Underwriter; Prop/Cas Edition 102. May 11.

Pauly, Mark, Howard Kunreuther and Paul Kleindorfer, "Regulation and Quality Competition in the US Insurance Industry," in Jorg Finsinger and Mark V. Pauly (eds.), The Economics of Insurance Regulation. St. Martins Press.

Ruquet, Mark E, 2000, "Passiac Seen as 'Fraud Central'." National Underwriter, Prop/Cas Edition. September.

Schlesinger, Harris and J-Matthias Graf von der Schulenburg, 1993, "Consumer Information and Decisions to Switch Insurers." Journal of Risk and Insurance 60: 591-615.

Schmit, Joan T. And Yu-Luen Ma, 2000, "Factors Affecting the Relative Incidence of Uninsured Motorists Claims." Journal of Risk and Insurance 67: 281-294.

Schwanberg, Robert, 2006, "State Insurance Pool Begins On-Time Payments to Victims." Newark Star-Ledger. February 18.

Schwanberg, Robert, 1997, "Legislators Can Represent Insurance Pool. Ethics Panel Votes to Let Lawyer-Lawmakers Defend Auto-Policy Cases for Defunct Group." Newark Star-Ledger. February 14.

Smart, Tim, 1997, "Accidentally on Purpose." Business Week. June 30.

Smith, Eric, and Randall Wright, 1992, "Why is Automobile Insurance in Philadelphia So Damn Expensive?" American Economic Review 82: 756-772.

Stewart, Angela, 1992, "State Launches Crackdown on Car Rate Evaders." Newark Star-Ledger. January 3.

Suponic, Susan J. And Sharon Tennyson, 1992, "Rate Regulation and the Industrial Organization of Automobile Insurance," in David F. Bradford (ed.), The Economics of Property-Casualty Insurance. University of Chicago Press.

Tassini, John R., 2000, In the Matter of GSA Insurance Company's Request for a Private Passenger Automobile Insurance Rate Increase. OAL DKY. No. BKI 10882-98. Agency DKT. 98-2514.

Tennyson, Sharon, 1997, "The Impact of Rate Regulation on State Automobile Insurance Markets." Journal of Insurance Regulation 15: 502-523.

Tennyson, Sharon, 1993, "Regulatory Lag in Automobile Insurance," Journal of Risk and Insurance 60: 36-58.

The Daily Tar Heel, 1997, "New Jersey Race Ends with Undecided Results." November 5.

Weisberg, H.I., and R.A. Derrig, 1993, Quantitative Methods for Detecting Fraudulent Automobile Bodily Injury Claims. Boston: Automobile Insurers Bureau.

Zarate, Vincent R., 1990, "SCI Launches In-depth Probe of Auto Insurance Companies." Newark Star-Ledger. May 20.

TABLE 1

AVERAGE EXPENDITURES

PRIVATE PASSENGER AUTO INSURANCE

NEW JERSEY & COUNTRYWIDE

YEAR NJ RANK AVERAGE AVERAGE

(1 IS MOST EXPEND. EXPEND.

EXPENSIVE) NEW JERSEY COUNTRYWIDE

2006 1 $1,138 $704

1997 1 1,126 707

1996 1 1,099 691

1995 1 1,013 668

2004 1 963 651

TABLE 2

DISPOSITION OF RATE FILINGS

NJ PRIVATE PASSENGER AUTO INSURANCE

JANUARY 1, 2006 - FEBRUARY 28, 2001

DECIDED

COMPANY DATE PERCENT DATE PERCENT

FILED REQUEST DECIDED GRANTED

GSA Sep 98 85.5% May 00 15.0%

NJ Re Jun 99 22.7 Jun 00 8.9

Pru Group Jan 99 -1.2 Jun 00 -2.9

PENDING

NJPAIP Sep 99 29.9

Selective Gp Jul 00 18.9

Amer Intrnl Sep 00 20.1

State Frm Ind Oct 00 16.8

Colonial Penn Oct 00 37.6

Founders Ins Oct 00 14.5

Proformance Nov 00 7.5

Lib Mut Fire Nov 00 20.0

Lib Mut Ins Dec 00 25.2

Provid Wash Jan 01 48.1

CGU Ins Cos Feb 01 7.3

TABLE 3

HIGH AND LOW POLICY PREMIUMS

1999 PRIVATE PASSENGER AUTO

TERRITORY 12 (SUBURBAN CAMDEN)

EXAMPLE AVERAGE HIGHEST LOWEST DIFFERENTIAL

PREMIUM PREMIUM PREMIUM

1-A $1,155 $2,537 $ 661 $1,876

2-A $2,062 $5,171 $1,394 $3,777

3-A $1,785 $3,744 $1,167 $2,577

4-A $ 980 $1,324 $ 675 $ 649

5-A $1,129 $1,978 $ 712 $1,266

6-A $ 996 $1,993 $ 837 $1,156

7 $ 262 $ 569 $ 204 $ 365

TABLE 4

AGENCY & DIRECT WRITER DIVIDENDS

NJ PRIVATE PASSENGER AUTO

1980-2006

(1) (2) (3) (4) (5) (6) (7)

AGENCY AGENCY DIRECT DIRECT TOTAL TOTAL YEAR

DIVIDEND DIV AS% WRITERS DIV AS % DIV DIV AS % DIR PREM DIVIDEND DIR PREM PAID DIR PREM

$ 2923619 0.18 $253516681 7.1 $256440300 5.0 98

468842 0.03 175790748 5.1 176259590 3.4 97

709625 0.04 78349932 2.4 79059557 1.6 96

986475 0.06 84403148 2.9 85389623 1.8 95

3482395 0.21 46893987 1.8 50376382 1.2 94

887889 0.05 16351020 0.6 17238909 0.4 93 1972029 0.15 38453784 2.0 40425813 1.2 92

1654420 0.20 42486281 2.9 44140701 1.9 91

1071525 0.15 54575282 4.4 55646807 2.9 90

1782962 0.30 56054309 5.3 57837271 3.5 89

2992365 0.58 57880486 5.7 60872851 4.0 88

3831666 0.80 82085010 8.6 85916676 6.0 87

1028908 0.21 50408990 5.6 51437898 3.7 86

923174 0.19 49743602 5.8 50666776 3.8 85

644135 0.13 44705844 5.5 45349979 3.4 84

376001 0.04 26149609 1.9 26525610 1.2 83

245221 0.03 22538833 1.8 22784054 1.2 82

230590 0.03 19298473 1.9 19529063 1.1 81

246877 0.04 17430008 1.9 17676885 1.2 80

TABLE 5

"LOSS RATIOS"

1980-2006

NJ DIR NJ DIR NJ DIV NJ LOSS COMP PRIOR YR

LOSSES PREM RATIO RATING APPROV

(000,000) (000,000) (000,000) LOSRAT LOSRAT

3178.2 5165.4 256.4 64.7 65.6 65.9 98

3513.5 5114.3 176.3 71.1 63.4 64.4 97

3537.7 4942.0 79.1 72.7 66.2 67.3 96

3291.6 4629.3 85.4 72.4 66.5 68.1 95

3291.5 4253.9 50.4 78.3 66.5 65.5 94

2963.4 4165.5 17.2 71.4 64.6 64.9 93

2113.8 3242.3 40.4 66.0 62.8 66.5 92

1771.3 2289.3 44.1 78.9 68.4 70.9 91

1496.0 1939.2 55.6 79.4 72.8 74.4 90

1288.9 1642.6 57.8 81.3 71.0 75.3 89

1198.3 1533.8 60.9 81.4 66.5 71.6 88

1062.8 1434.0 85.9 78.8 65.2 70.0 87

1026.8 1385.7 51.4 77.0 68.2 72.9 86

879.0 1350.3 50.7 67.6 75.3 74.3 85

1208.0 1320.3 45.3 94.8 76.2 71.4 84

1548.6 2275.3 26.5 68.9 68.1 67.1 83

1459.4 1979.1 22.8 74.6 68.8 68.1 82

1378.6 1701.8 19.5 81.9 64.4 66.6 81

1138.6 1498.3 17.7 76.9 62.0 65.3 80

TABLE 6

TIMETABLE FOR SELECTED

LAW CHANGES

NJ PRIVATE PASSENGER AUTO INSURANCE

1972 New Jersey Automobile Reparation Reform Act

Introduced Elective No-Fault

Unlimited PIP Medical Benefits

$200 Monetary Threshold

1983 NJ Automobile Full Insurance Underwriting Association

The Joint Underwriting Association

Caps on rates

1988 Insurance Reform Act

Flex Rating

Excess Profit Law

Introduce Verbal Threshold

PIP Co-Pays

Mandatory Comprehensive Deductible

Allowed 2% non-renewals of policies, and one non-renewal for every two new policies

No Filing "in concert" (No Full Bureau Rates)

1990 The Fair Automobile Insurance Reform Act of 1990 (FAIR Act)

Market Transition Facility (ended the Joint Underwriting Association)

Re-introduced The Assigned Risk Plan (effective Oct 1, 1992)

Introduced "Take-all-comers" Rule

Eliminated Unlimited PIP Medical Benefit (set $250,000 mandatory benefit)

Required License Surrender (all lines) For Market Exit

Pre-insurance Physical Damage Inspection

1997 Public Law 1997, Chapter 151

Tier Rating

Urban Enterprise Zones (UEZ's)

Introduced Expedited Filings (repealed Flex Rates)

2006 Automobile Insurance Cost Reduction Act

Strengthen Verbal Threshold

Fraud Prosecutor

Revamp Rating Territories

Introduce Basic Policy

Named Driver Exclusion

PIP Choice (eliminated mandatory $250,000 PIP medical)

PIP Arbitration System

Medical Cost Controls

Insurance Ombudsman (investigate consumer complaints)

Qualified Person Rule



i 1. Professor of Economics, Rutgers University, Camden, NJ 08102 jworrall@camden.rutgers.edu

New Venture Commercial Insurance