Commercial Auto: Achieving a Balance
Commercial Auto: Achieving a Balance
by Lane DeCoster
Introduction
The insurance industry reports higher costs—meaning losses—for accidents involving business automobiles. Claims costs of Commercial Automobile insurance are racing to new heights whenever bodily injury is modest or severe. Surprisingly, inflationary insurance costs began surging during a decade of unprecedented improvements in automobile safety, highway maintenance and driver awareness. Additionally, managed care became important in the health care sector at this time. These Improvements resulted in:
¨ fewer auto accidents
¨ fewer highway fatalities
¨ fewer DUI/DWI related accidents
¨ stable medical costs
¨ flat or declining costs of insurance
Even today, governmental and insurance industry statistics suggest that improvements are holding steady. People are less likely to suffer injury or death from an automobile accident.
Although safety improvements are reducing personal injury, claim costs of Commercial Automobile insurance are rising. Paying for severe claims is now at record heights and severity is the culprit.
That is, the cost of a small claim remains nearly unchanged. On the other hand, the cost of severe injury or death is accelerating rapidly. The bottom line is that a small population of expensive claims disproportionately carries the financial burden of the higher insurance costs today.
Actuaries adjust two components of “pure premium” to price for loss costs:
1. Base Rate
2. Increased Limit Factor (ILF)
Basic limits are normally $25,000. Industry information shows a slight increase in the cost of these claims, which are offset by a declining rate of frequency. In total, Base Rates are under little pressure.
Separately, actuaries study the portion of claims greater than $25,000 as the basis of ILFs. Since excess losses are changing with more velocity, we would expect ILFs to respond accordingly. Instead, ILFs from the industry’s leading rating bureau seem to lag the recent loss trends.
ILFs finance higher insurance limits. Increasing total policy premiums (using discretionary modifications) or Base Rates adjusts premiums uniformly and will not price solely for severity. Only ILFs apportion the premiums to basic and excess limits.
An analysis of industry results reveals unquestionably severe loss trends. However, the industry’s method of pricing for severity—namely ILFs—is not keeping pace.
Commercial Automobile Market
Automobile insurance is a staple in the U.S. Property and Casualty industry. Its premium exceeds $140 billion annually, according to OneSource (2001 edition). That represents about 44% of the industry’s premium volume. A small segment of the Automobile market is Commercial Auto insurance with $18.1 billion of premium, as reported in Conning’s Industry Insight: Commercial Automobile Insurance 2001. Commercial Auto is very unhealthy today. One might wonder how big a problem this segment could create when it is barely a “ten percenter.”
A few statistics highlight the dilemma for insurers of Commercial Auto. Conning & Company reports the estimated financial results of Commercial Auto in 2000, are:
| Loss Ratio | 88.2% |
| Expense Ratio | 33.4% |
| Combined Ratio | 119.7% |
| Return on Equity—GAAP | (1.4%) |
Today’s experience is beyond disappointing. It is financially destructive.
History shows the insurance industry is unable to sustain stable markets when the combined ratio is near 120%. For example, a hard market occurred in casualty lines in the mid-1980s, later in workers compensation and after certain property catastrophes.
Financial pressure placed upon companies with results similar to or worse than the industry’s Commercial Auto experience suggests the need for improvement. In late 1999, a few companies disclosed to their shareholders a need to increase rates. Since then, many companies report that they are raising rates in Commercial Auto, among other lines, to return to profitability.
The journey to profitability might be challenging, especially if companies measure success by monitoring earnings. During a speech to General Re’s officers in March 2001, V.J. Dowling, Jr. of Dowling & Partners Securities presented a sobering predicament for the P&C industry (not the Commercial Auto segment). He explained the analysis used in his research comes from proven financial equations; the math is irrefutable. Only the industry’s results and assumptions will change the outcome.
Based upon the performance of the P&C industry, Mr. Dowling determined that the combined ratio necessary to achieve a 12% ROE is 93%. Some investors set even higher ROE goals, which would require lower permissible combined ratios. (Please note General Re does not advocate an ROE goal of a particular amount—it is illustrative only.) We assume today’s results may disappoint many investors.
Factors that influence the outcome include premium-to-surplus ratios, investment returns and duration (or tail) of losses. If the variables affecting Commercial Auto insurance are similar to those of the industry, it is possible to calculate the premium shortfall in this line of insurance. (While the two differ, the variance seems small enough that the following example is meaningful.)
Applying Mr. Dowling’s assumptions for the P&C industry to Commercial Auto, in 2000, the segment needed another $5.2 billion of premium to achieve a 12% ROE.
A $5.2 billion shortfall equals all insured losses from natural catastrophes during that same year. More importantly, all Commercial Auto premiums in the four largest states barely top such a deficiency. Performance in this small segment of Automobile may have large implications on insurers.
| State | 1999 Premium |
| California | $1,840,712,000 |
| Florida | $1,150,786,000 |
| New York | $1,436,524,000 |
| Texas | $1,303,417,000 |
| . | . |
| Top 4 | $5,731,439,000 |
| Market Size | 31.7% |
Source: Conning’s Industry Insight: Commercial Automobile Insurance 2000
The medium range forecast remains troubling. Conning & Company expects poor results for the Commercial Auto segment through at least 2002, the last year of their study.
 |
| . |
In spite of rate increases, Conning believes the 2002 combined ratio will be unchanged and the ROE will slip to minus 8.6%. If this is true, closing the gap between actual and desired results requires diligence.
Any insurer willing to accept an underwriting loss in exchange for investment income should analyze their cash flow assumptions in the new claims cost environment. Under-priced exposures often create a faster than expected cash outlay for insurers, at least temporarily. Invested “float” may disappear sooner than planned.
More Bad News
Generally, the industry divides Commercial Automobile business into two broad categories:
1. Commercial Auto Liability (CAL)
2. Commercial Auto Physical Damage (CAPD)
In the early 1990s, the insurance industry posted excellent combined ratios for CAPD. At the time, underwriting losses hurt CAL only. Now both coverages operate at a significant underwriting loss.
CAPD affects CAL indirectly. Historically, CAPD might be perceived to cross-subsidize CAL. In 2000, Conning reports CAL premium grew (at 4.0%) for the first time in eight years. Coincidentally, annual profits from CAPD in the early 1990s had exceeded 4% of CAL’s premium. Reallocating the underwriting profits away from CAPD would give CAL an additional boost. In 2000, CAL might have had an effective premium increase of 8% after the cross-subsidy, if CAPD had remained profitable.
The redundancy in CAPD rates is gone. CAPD no longer cross-subsidizes CAL. The financial dynamics of Commercial Auto—the sum of CAL and CAPD—changed after both coverages became unprofitable.
Seeking Solutions
Armed with the previous information, a logical next step may be to rush to a solution. “Okay, the financial performance is dreadful, let’s fix it.” Any underwriter with a combined ratio of 119.7%, which equals the Commercial Auto industry’s results, might consider a corrective action plan. Underwriters match rates to the exposures at risk. A natural solution by some underwriters is to raise rates whenever they perceive a rate deficiency.
To illustrate an easy solution, consider a goal of a 100% combined ratio, thus no underwriting loss. Straightforward math indicates a 19.7% rate increase is needed in this example. On average, applying this rate increase to all policies would achieve the selected goal—all things being equal.
Across-the-board solutions treat each risk uniformly. After all, every insured is entitled to the same protection as another when a fortuitous loss occurs (subject to terms, conditions and limits of coverage). To some, this balanced approach may seem reasonable. However, insurers typically do not use across-the-board solutions.
Each risk has unique characteristics that affect underwriting judgment. Traditionally, underwriters apply different rates and eligibility criteria to various exposures. These kinds of rules—including guidelines for rate decreases or increases—typically apply to all limits of insurance. Rarely do underwriting rules vary by limit.
Commercial Auto experience reveals different levels of concern at various limits of insurance. An alternative to an across-the-board solution should be apparent after examining Base Rates and ILFs in relation to non-uniform Claim Cost Trends. Investigating bodily injury trends may be the most useful start. CAL accounts for 70% of the Commercial Auto premium and is responsible most of the deteriorating results. Within CAL, Bodily Injury (BI) claims are more costly than Property Damage (PD) losses, which are less troubling.
Worries are not uniform throughout the business; Commercial Auto is out of balance today.
Frequency is declining slightly, the average size of minor claims is almost unchanged, large losses are rapidly getting more expensive and ILFs are virtually the same. We will review how the combination of these factors relates to the total insurance premium for Commercial Auto and CAL BI.
Jury Verdicts & Settlement Trends
Studies published in the past couple of years show a dramatic change in jury verdicts—Americans are paying larger awards. The trend of jury verdicts is often in the news because of today’s spike. A lot of speculation surrounds the reasons for the sudden and dramatic rise. Regardless of the reasons, it is clear higher verdicts affect insurance company settlements.
Jury Verdicts Research (JVR) is a premier provider of jury verdicts and claims settlement information. JVR reports startling settlement statistics, in a way more startling than jury verdicts. Settlement values are important because jurors resolve very few automobile suits; pretrial settlements are more common. In JVR’s annual publication—2000 Current Award Trends in Personal Injury—over a six year period the mean settlement for vehicular liability cases rose over 200%, i.e., it is now three times higher, as follows:
The pace is rapid. In JVR’s database, there is no mistaking that the cost of insurance is detached from economic inflation of 2.5% (as measured by C.P.I. at this time). A settlement inflation rate of 22% annually resembles inflation in lesser-developed nations, not in the U.S. Insurers might have been unprepared for the swift change.
JVR’s data shows that large settlements are common. The effect on the cost of insurance seems obvious for limits greater than $100,000. If similar results are evident in the insurance industry’s databases, then insurers should consider increasing ILFs enough to capture the disproportionate rise in severity relative to the entire loss.
We do not recommend accepting mean settlements without further investigation because a large item outside of the statistical norm may distort the average value. A few extraordinary automobile settlements could skew the mean and it appears they have.
Mean settlement values are rising faster than median settlements, indicating increasing volatility in vehicular liability settlements.
Looking at median data is useful. These statistics from JVR confirm the general trend. The increase in median settlements rose nearly, but below, 200% during this same timeframe, as follows:
We must accept these statistics at face value because JVR does not provide source data in their publication. Although JVR’s data may be of limited value to actuaries when they calculate insurance rates and ILFs, understanding trend is critical. Underwriters and actuaries alike have taken notice of these astounding increases in settlements. Higher settlements add volatility.
ISO’s Loss Trends
Actuaries obtain technical data from the insurance industry to determine rates. Insurance Services Organization (ISO) provides the best analysis of Claims Cost Trends by layer that we have seen for Commercial Auto (see Commercial Automobile Trend Data & Analysis 08/2001).
This trend study is one of ISO’s important reports. ISO measures change—in percentages—of the claim patterns from year to year. With this information insurers can link the past and the future. In other words, trends help us estimate future insurance costs.
Source data comes from ISO’s vast membership. Due to the large cross-section of insurers reporting to ISO, it represents a credible sample of the industry’s experience. In fact, ISO’s Claims Cost Trends measures changes in the country's largest database of primary Commercial Auto claims.
Data is prepared carefully. For example, if claims from three years ago came from a different mix of members, changes observed this year might be skewed by the shifting profile of participating insurers instead of the claims environment. ISO wants clean comparisons. Cleaning the data is arduous, yet critical to mitigate unwanted bias.
In its trend study ISO analyzes primary policies, omitting umbrella and excess policies. This differs from JVR, which we believe includes all types of vehicular liability verdicts and settlements.
The bias of JVR’s data may be more toward excess than ISO’s trend study. Our experience as a reinsurer suggests that many Commercial Auto policies have a $1 million limit, though occasionally higher. (ISO does not disclose this information in the report.) We believe ISO’s results are reliable to at least a $1 million limit.
Using claims data through 2000, ISO estimates that the average loss from a primary policy of CAL BI is rising 7.6% annually for Total Limits (meaning actual limits of primary insurance). Additionally, ISO estimates Claims Cost Trends at specified limits, as follows:
Capped Limits
| $25,000 | +2.6% |
| $100,000 | +3.7% |
| $500,000 | +6.0% |
| $1,000,000 | +7.2% |
| Total Limits | +7.6% |
Source: Commercial Automobile Trend Data & Analysis 08/2001
Each limit in ISO’s study includes losses arising from the first $25,000. ISO adds credibility to its selections by keeping a large population of smaller claims throughout their analysis. Another aspect of ISO’s methodology is the way it dampens the trend factor from limit to limit. Notice the smooth decline for successively lower limits.
Capped losses trend slower than Total Limits losses. Removing volatility causes this.
ISO segments Total Limits data into capped limits and does not analyze excess layers. This information suits most insurers. Remember that ISO’s members generally want information to price primary insurance at various limits. ISO is not geared to excess risk.
Examining Trends Per Layer
General Re converted ISO’s data into excess layers rather than capped limits. It is important to note the data is the same. From this perspective, trend factors for CAL BI rise dramatically in each successively higher layer. Changes are not uniform.
Excess Limits
| $75,000 x $25,000 | + 5.4% |
| $400,000 x $100,000 | +10.8% |
| $500,000 x $500,000 | +19.5% |
| T.L. x $1,000,000 | +26.9% |
Excess trend factors show the “leveraged effect of trend.” Excess layers carry the burden of financing catastrophic claims plus those losses that once were below the cap, but now are not. The “leverage” is due to volatility and inflation.
For CAL BI losses, evidence shows that cost pressures are in the excess layers, but at an attachment point as low as $100,000. After all, CAL BI losses add only 2.6% to Base Rates, yet the average CAL BI claim is rising 7.6% annually. Severity overwhelms any benefits received from frequency reductions and containment of costs at Base Limits. Pure Premiums and ILFs should not ignore the alarming discrepancies between layers.
The following chart illustrates the discrepancy. Reading it may be difficult without a brief explanation, though the concept is not complex. “Capped” percentages match ISO’s trend factors for the various capped values, i.e., $25,000, $100,000, etc., up to Total Limits (TL). “Excess” represents trends for each of the intermediate layers between ISO’s capped limits, i.e., the excess value between $25,000 and $100,0000 is the trend factor for the layer of $75,000 x $25,000, etc., up to TL x $1million.
Source: General Re’s actuarial review of ISO’s trend data
Loss trends support different rate increases for various layers rather than one-size fits all and a single rate increase for the policy.
The surprise is the dramatic differences between the various layers. We expect the trend factor of an excess layer to be greater than that of a lower layer, but not to this extent. The trends are startling and worrisome. Many insurance professionals might find today’s unbalanced situation to be doubtful without the backup data from reliable sources.
Timeline of Changes
Part of the technical analysis is to review incurred losses (reserves and payments) and paid losses (excluding reserves). Data from both are consistent. ISO’s trend study shows “incurred losses” are accelerating upward since mid-1996 and “paid losses” mirror this trend since late-1998. Seeing “paid” data lag “incurred” losses by two-and-a-half years is normal for CAL BI because it is the average time between a loss occurrence and its settlement.
Source: Commercial Automobile Trend Data & Analysis 08/2001
The above graph is for Total Limits (i.e., actual policy limits), based upon ISO’s data. A graph of any excess layer would reflect the same trend, but with a much steeper incline because of the more dramatic trend factors.
If General Re’s excess trend factors are accurate and remain unchanged, then the average loss for CAL BI will double as follows:
# Years To Double Costs
| 1st $25,000 | 27 years |
| $75,000 xs $25,000 | 14 years |
| $400,000 x $100,000 | 7 years |
| $500,000 x $500,000 | 4 years |
| TL x $1,000,000 | 3 years |
Insurance costs have always cycled through peaks and valleys. It is possible today’s trends will not endure four years, let alone more than seven. Yet, some changes have occurred already, so we can measure these.
It is surprising to see how much “damage is done” already.
Based upon ISO’s “paid loss” from its trend study, General Re measured the growth to date for a couple of excess layers:
Source: General Re’s actuarial review of ISO’s trend data
The indications from ISO mirror those in JVR’s settlement statistics. JVR reports settlements are trending +22% per year, or about +48% over two years. The inflationary rate for severe injury is enormous whether we look to JVR or ISO for information.
| . |
|
An insurer’s mix of business could influence the weighted-average trend factor and, thus, the speed in which average claims costs rise.
If ISO’s actuaries are correct, underwriters of low limits should have plenty of time to adjust Base Rates to maintain adequate premiums. In contrast, companies providing high limits of insurance or only excess limits (e.g., umbrella business) might see their Commercial Auto results deteriorate quickly if they react slowly to the excess trends. Understanding the importance of trend during changing times helps insurers set appropriate premiums in a timely way.
Building Trend into ILFs
Pure premiums are expected to match the pace of rising CAL BI Claim Costs. Specifically, ILFs should adjust for the skyrocketing costs in the excess layers and the resulting non-uniform changes in loss by layer. Yet, this is not happening. ISO reports the nationwide weighted-average change in ILFs is notably lower:
Average ILF Changes
| 1999 to 2000 | +0.7% |
| 1999 to 2001 | +4.2% |
Source: ISO’s Commercial Automobile Liability Increased Limits Data and Anaysis, 09/2000 and 9/2001)
We notice a few things from ISO’s ILF data and analysis:
1. +4.2% is a two-year change, meaning the annualized percentage is less.
2. ILFs increased more this year than during the prior year.
3. Increases in ILFs lag current loss trends for CAL BI.
4. ISO’s two-year change in Total Limits’ ILFs is ten times smaller than two-years of CAL BI trend for $400,000 xs $100,000 (see previous graph).
|
| . |
ILFs should adjust in accordance with loss trends. It is unwise to expect ILFs to be at equilibrium unless their changes match loss trends.
Although the statistics come from the same ratemaking organization, ILFs are increasing more slowly than the costs that underlie them. At a glance this may seem peculiar and, in fact, inappropriate. However, a quick comparison between CAL BI loss trends and ILFs is an oversimplification because the statistics for each come from different pools of data.
For example, CAL BI is only one component of an ILF because an ILF prices for “combined single limit” insurance—BI and PD—and loss adjustment expenses. ILFs must reflect trends in all of the components. Examining only CAL BI trends would be incomplete.
Nonetheless, severity trend of CAL BI stands out as the largest change within the ILF calculations according to ISO’s data. However, the other factors are changing more slowly and they temper the overall change. Therefore, the overall ILF increase will be at a slower pace than CAL BI trends.
Unlike JVR, ISO provides detailed information in its ILF study. Statistics confirm a disparity between the loss component of ILFs versus CAL BI severity trends—more than the mitigating factors would suggest. The following compares “loss to loss”:
Source: General Re’s review of ISO’s selected ILFs and ISO’s trend study
(The trend study does not isolate the layer $1M x $1M, so the factor for Total Limits x $1M is the substitute. General Re feels T.L. is a reasonable proxy since few Commercial Auto policies exceed $2M.)
In summary, ILFs appear to lag loss trends. The offsetting benefits derived from the non-BI components of ILFs do not fully explain the mere +4.2% rate increase in ILFs over two years. Commercial Auto ILFs should have increased by a larger percentage if they are to match the empirical loss severity trends reported by ISO.
Lagging ILF Data
A standard practice by actuaries is to use the past to forecast the future. However, using past information “as is” might be inappropriate, so trending adjusts any lagging data to make it meaningful in the future.
The source data underlying ISO’s ILF calculations has some lagging characteristics.
For various reasons, ISO promulgates ILFs from paid loss data. The nature of paid loss is that it lags incurred loss by two years for CAL because of claims settlement practices. Another lag occurs due to the time-consuming process of gathering data from ISO’s member companies. Finally, ISO intentionally selects old years for its ILF calculations since they are subject to fewer adjustments (e.g., late reporting by a member company because of delays in processing the paperwork).
Because of the lagging characteristics, ISO promulgates ILFs from data that is older than those from the recent periods in its trend study. In fact, most of the data supporting the ILFs filed and approved today seem to predate the inflection point of 1998.
ILF data “lags” recent trend experience. Current ILFs are based on losses that seem to omit most of the severity trend “spike”. Lagging data might explain the disconnected changes in loss trends and ILFs.
Specifically, ISO published ILF studies in September 2000 and 2001. It appears that ILFs selected in 2000 include merely one-fifth of the severity spike. The remainder comes from stable times. We were surprised to learn a year later from ISO’s 2001 report that most of the ILFs in use today were published in 1999, which might predate the inflection point entirely. ISO says most of the insurers are not using ILFs from the 2000 study.
ISO uses five years of paid losses to create ILFs. Each year ISO drops an old year from the equation and adds a new one. General Re expects ISO’s ILFs for CAL will increase slowly for several years as less costly years drop from the ILF calculation and are replaced by more expensive years, unless ISO adjusts its formula. ILFs are beginning to rise, albeit slowly, and they should continue to do so.
ISO’s methodology of promulgating ILFs requires patience. Claims costs, jury verdicts and insurance company settlements in the higher layers are increasing faster than ILFs. It might take as many as five years before ISO’s ILFs fully include the effect of today’s Claims Cost trends.
In the meantime, revenues must cover costs to maintain a viable market for Commercial Auto insurance. The gap between loss trends and premium trends (including changes in Base Rates and ILFs) strains rate adequacy. The link between the past and the future is trend. It is possible to select a Total Limits trend factor within the ILF formula that would match recent loss experience in each layer. Such an adjustment seems appropriate today to equalize premium and loss trends.
Limits Pull Affects Severity
Several studies done by General Re analyze ILFs in a newly recognized way. We asked how the insured’s Available Limit affects ultimate losses? [The sum of all limits from all policies is the “Available Limit.”] During the mid- to late-1990s, many businesses bought higher limits of insurance (including umbrella). The jump in severity at that time seems more than coincidental. While we are unable to demonstrate why a correlation exists, we know it does and have now measured its impact on ILFs.
Typically, each limit has one ILF for each type of vehicle (e.g., light trucks). Many years ago, pricing insurance limits was simple. One policy provided all of the limits of insurance and that limit had a corresponding ILF.
Now, many businesses purchase umbrella or excess insurance. The insured may receive at least two policies. Yet, the methodology of calculating the premium of the primary (or underlying) policy is unchanged—one ILF matches the limit for each type of vehicle, regardless of the sum of the limits of all policies.
We believe a more appropriate method is to have multiple ILFs for each limit, depending upon the Available Limits. Currently, ISO has ILF tables for different vehicle types and different state groups. A new dimension arises—subgroups for “Available Limits.” We are unaware of any pricing formulas that work this way now.
For example, each limit might have three ILFs instead of one, as follows:
Sample ILF Table
Without Risk Load
(Specify Veh. Type & State Group)
Limit | Low | Med | High |
$100K | 1.40 | 1.45 | 1.50 |
$500K | 2.15 | 2.25 | 2.35 |
$1M | 2.35 | 2.55 | 2.65 |
Available Limits:
Low = Up to $2M (First Dollar)
Medium = $3M – $20M
High = >$20M
Source: General Re
The phenomenon is “Limits Pull” because the policy limit draws—or pulls like a magnet—losses from lower limits within a policy. Therefore, we identified a different set of ILFs for each Available Limit. Higher Available Limits coincide with “steeper” ILF tables and lower ones have “gentler” slopes. Most importantly, a policy limit for a Commercial Auto policy, e.g., $1M CSL, would have more than one ILF to reflect the possible array of Available Limits because some insureds buy no other insurance, while others buy additional umbrella limits.
The impact of Limits Pull in the Commercial Auto line is significant in many of the excess layers. We believe the issue worthy of serious consideration for all umbrella underwriters. As additional studies are published, we believe momentum will build to create a new matrix of ILFs. Unlike accelerating severity trends, which might be temporary, “Limits Pull” seems to be permanent.
Limits Pull is further evidence that severity trend is not uniform across all policies, limits or types of Commercial Auto business. Limits affect ILFs in a new way.
Limits Pull remains an emerging topic and ISO may publish a new study for its members. Again, we are unaware of any pricing formulas that include Available Limits as a variable in the ILF tables. In fact, it is not a question on the ACORD Application. Any underwriter who uses Available Limits to determine premiums will need its own guidelines because rating bureaus have not adopted this method to date.
Differentiating Portfolios
The characteristics of each book of business are important when reviewing the implications of today’s trends. No doubt, selecting proper trend factors is more challenging when costs are soaring unevenly. Research done by General Re indicates an amplified risk of picking inadequate loss costs and, thus, ILFs for higher limits of insurance.
An oversimplified example should highlight the issues. This example will omit other variables, such as frequency trend. Imagine a Commercial Auto insurer has only two agents selling all of its policies. One specializes in small insureds that buy low limits, e.g., up to $100,000. The other specializes in large insureds that require $1 million limits and keep a $100,000 deductible.
Since the company is familiar with overall trend indications, it adopts a 7.6% rate increase on all Commercial Auto policies this year. Two years later the company is looking at the profitability of 2001 for each agent to calculate profit sharing commissions. The agent of small accounts exceeded expectations and the other one missed its profit sharing targets. Trend may explain the results, assuming the company had adequate rates before adopting a 7.6% increase.
| Small Limits Business: | |
| Basic Limit: | +2.6% |
| $75K x $25K | +5.4% |
| . | . |
| Selected Trend For All : | +7.6% |
| . | . |
| Large Limits Business | |
| $400K x $100K: | +10.8% |
| $500K x $500K: | +19.5% |
In reality, portfolios are not divided as neatly as this example. Nonetheless, insurers are exposed to today’s non-uniform trend in different portfolios:
¨ Small commercial insureds (like BOP business) buy lower limits of insurance.
¨ Larger commercial insureds (like middle market business) buy higher limits.
¨ National accounts underwriting units specialize in deductibles.
¨ Stand-alone umbrella facilities (i.e., unsupported underlying insurance) or companies with a separate umbrella department have no cross-subsidies.
¨ Even those carriers that that have multi-line underwriters handling the primary and umbrella business may be subject to uneven results because virtually all companies track umbrella as a product line.
Additionally, many insurance companies buy excess of loss reinsurance. Reinsurers look for fair, equitable and appropriate terms. Today, reinsurers may be unwilling to share in an “across-the-board” rate increase. We expect reinsurers may evaluate each portfolio carefully, looking for the unique characteristics that affect underwriting decisions.
Summary
Commercial Auto insurance losses have become more expensive recently. Throughout history, insurance costs have ebbed and flooded like the tides. Unquestionably, costs are flooding today.
Actually, claims costs do not ebb like the tide because they do not recede in an equal and opposite way. Instead, claims costs tend to plateau at a new height before surging to the next level. Once trend flattens (and perhaps recedes a little), the industry will have a new and higher threshold of Commercial Auto losses.
During these changing times, not only are claims costs rising, but the burden also is distributed unevenly—across all layers and through all coverage parts of Commercial Auto policies. Data from JVR and ISO seem to agree that the breaking point is around a $100,000 limit. Above this amount, the cost pressure is substantial.
Finding one rate increase to address all of the trends is challenging and nearly impossible. Having a single rate increase today apply to all policies and all layers just magnifies the risk of error.
It is important to understand that all of the trends discussed here relate to losses—JVR’s loss settlements, ISO’s claims costs and loss components of ILFs. They measure loss-to-loss changes and are different from “rate adequacy.” Loss trends are one component of premium adjustments. Obviously, Claims Costs Trends are an integral part of the insurance premium. However, these adjustments must apply to technically adequate rates in order to achieve the desired results. Each company should first determine whether its rates are adequate before adjusting for loss trends. Based upon the industry’s estimated combined ratio of 119.7%, some companies are not profitable, so further work is necessary to remedy underwriting losses other than adjusting for trend.
Sources:
Commercial Automobile Trend Data & Analysis, Insurance Services Office, August 2001.
Commercial Automobile Liability Increased Limits Data & Analysis, Insurance Services Office, Septemeber 2000.
Commercial Automobile Liability Increased Limits Data & Analysis, Insurance Services Office, Septemeber 2001.
Conning’s Industry Insight: Commercail Automobile Insurance 2001, Conning & Company, 2001.
OneSource Market Share, Thomson Financial Insurance Solutions, 2001.
2000 Current Award Trends in Personal Injury, Jury Verdicts Research, LPR Publications, 2000.
Materials presented in March of 2001 by V.J. Dowling of Dowling & Partners Securities, Hartford, CT.
Research of General Re, Stamford, CT.
About the Author
Lane DeCoster is a Vice President and the Business Development Specialist for Automobile for the Program Practice Group at General Re. He is responsible for the marketing strategy, providing underwriting expertise and monitoring emerging issues in automobile insurance and reinsurance.
For More Information
Our Automobile Team, comprised of several underwriters, claims professionals, attorneys and actuaries, specializes in auto insurance matters. Additional in-depth research on commercial auto trends is available. For more information, contact Lane DeCoster at 203-328-5697 or ldecoste@gcr.com; Ralph Ciarmiello at 203-328-5918 or rciarmie@gcr.com; Randy White at 203-328-5775 or rwhite@gcr.com; or contact your local General Re office.
Confidentiality:
These materials contain PROPRIETARY AND CONFIDENTIAL information of General Re concerning issues that may affect the business we write. In consideration of providing you with these materials, you agree to maintain the information in confidence and NOT REPRODUCE OR DISCLOSE the information to any third party, or to any person outside of the group within your company directly responsible for its use.
|