Insurers and the Consumer
What is Ratemaking?
All states have specific criteria that insurers are expected to follow when calculating rates. While the criteria may have some variation, all insurers follow some basic ratemaking patterns.
As we know, underwriting is a vital function of successful insurance marketing. Closely allied to the function of underwriting is that of rate making. Ratemaking is very technical in most lines of insurance. In general terms, it involves first the selection of classes of exposure units on which to collect statistics regarding the probability of loss. In life insurance this task is relatively uncomplicated since the major task is to estimate mortality rates according to groups (age, gender, occupation). In other fields, like fire and workers’ compensation, very elaborate classifications are necessary. Workers’ compensation, for example, has hundreds of classes of industries and each has its own individual rate. Actuaries supervise ratemaking.
Even after distinct classes have been established, reliable statistics are necessary for each class collected over a sufficient period of time. Next, the statistics (data) must be converted into a useful form of some type for the purpose of developing a final premium. The data will be used to determine premium, but also estimates of the cost of doing business. As with any business, insurers have overhead. They must pay for their premises, utilities, direct employees, underwriting expenses, commission to field staff, and so forth. In addition, insurers must estimate the amount of claims that will be submitted for policy benefits. There may be additional costs associated with writing business on a risk for which no data whatsoever has been accumulated.
The insurance rate is the amount charged per single unit of exposure. The premium is the product of the insurance rate multiplied by the number of units of exposure that is purchased. For example, there will be a rate for one unit of life insurance. Let’s say that a single unit represents $1,000 of life insurance coverage. If an individual wants $10,000 worth of life insurance, then he or she will buy ten units ($1,000 X 10 = $10,000). When we sell insurance, we do not speak in terms of the insurance rate or rate per unit. Instead, we are more likely to speak in terms of money: a $50,000 policy or a $100,000 policy. Even though we do not discuss the possibilities in terms of units, however, that is how we are representing it.
Each unit of insurance has a specified cost. For example, one unit may sell for $25. Therefore, if the applicant wished to purchase 10 units the agent would multiply $25 times 10 to determine the total premium cost.
Premium is designed to cover two major costs: the expected loss and the cost of doing business. The expected loss is called the pure premium and the cost of doing business is called the loading. The pure premium is determined by dividing the total expected loss by the number of exposures. The loading is made up of such items as commissions, general company expenses, taxes, and so on. The sum of the pure premium and loading is termed the gross premium. The loading is typically expressed as a percentage of the expected gross premium.
In any form of ratemaking, it is the actuary that develops the charge. When multiple insureds are combined, it produces an overall adequate sum that provides sufficient funding for losses, the expenses of operating, reasonable profits, and accumulation of a reserve for catastrophes, where applicable. Obviously, a primary function of ratemaking is to provide an adequate flow of premium to meet the needs of the insurer and, in some cases, the stockholders. In addition the rates must reflect the varying loss-producing characteristics and exposures of individual insureds.
Insurance is one of the few industries that can legally group people by class. In fact, when it relates to insurance, class groupings are essential to rate making procedures.
Classifications are used to measure risk. Initially, this must be based on the judgments of the actuaries, but these judgments come from consideration of common loss-producing characteristics even when there is not yet statistical information available. Once statistical information on that particular risk becomes available (through statistical accumulation, knowledge, and acquired understanding) the risk classifications will be further defined and, as time goes by, redefined again. Each time a risk classification is further defined, the rate could change.
Classifications are established as a function of the balance between the need to minimize the number of classifications and the need to have a sufficient number of classifications to achieve reasonable equity and avoid competitive disadvantages.
Classifications must produce a reasonable volume of experience sufficient to ascertain the underlying pattern of loss development. Actuaries use past experience as the basis for estimating future costs. Therefore, the volume must be sufficient to establish patterns of past experience from which to base their conclusions. Insurers use the law of large numbers when developing the pattern of loss.
It is important for the classifications to be as exact as possible. If the classification takes in too broad of a definition, it may not be accurate. That is why class groupings are so important to the insurer. The actuary must use clear terminology to avoid accidental misclassifications. Even when classifications or groups of risk seem to be exact, there will be endless reclassification as greater knowledge is gathered or statistical information becomes more extensive. In addition, risk conditions tend to change continuously, requiring constant reclassification of those risks. Experience may also uncover previous class weaknesses, which then means the actuary will redefine the risk to offset the problem.
Grouping risks by territory is an important step in attaining equity among a company’s insureds. For instance, in the automobile lines the differences in hazard relating to location is one of the most significant factors affecting loss costs. This is not surprising. Consider the amount of traffic in a large city compared to a small rural town. It is more likely that a driver in the thick of city traffic would be involved in a fender bender than would a driver in a less congested rural town. Therefore, the risk factor for the city would be greater than for the small town. Theft is also more likely to be a factor in a city where there are more people. Even such things as climate, attitude of judges and juries, traffic regulations, conditions of roads, and local laws will affect loss and cause differences in insurer expenses. All of these factors will be considered by actuaries when setting policy rates.
How territories are established can vary depending upon many factors. Territories may be separated by zip codes, counties, or even block by block. It will depend upon all the factors relating to loss needed to achieve a reasonable equity and avoid competitive problems due to risk groupings.
Territory groups will depend upon their type of insurance it relates to. Some types of insurance will make an entire state as one territory (Workers’ Compensation Insurance is based on state law so an entire state is one territory) while other types may be based on an entire country. Boiler and Machinery insurance is a type that is based on an entire country. Certain inland marine lines require countrywide schedules because of the low premium volume and the scarcity of statistical experience. Boiler and Machinery Insurance is so complex in its rate structure and the significance of uniform inspections is so great that it needs countrywide schedules.
There are two special aspects of manual ratemaking that should be mentioned. One is the provision for catastrophes and the other is the minimum premium.
In liability insurance, the basic coverage allows a specific limit per claim, per accident, or per occurrence for bodily injury liability. There may also be a specific limit per accident for property damage liability. The rates in the manual are for these limits and excess limits rates usually are calculated by adding percentage loadings to the basic limit premiums. The excess limits tables of percentages are based on countrywide experience.
Not all risks have enough exposure units to produce sufficient data. For risks of this nature, a combination of a very low rate and/or a small number of exposure units require the insurer to meet its minimum expenses in handling individual contracts by establishing minimum premiums. These are based on careful expense studies. The actuaries and insurers certainly do not want to experience claims that exceed premium income.
Ratemaking represents the average expected losses costs for each classification. Classification groups are based broadly on the basis of their loss-producing characteristics. If all risks falling within a particular class were exactly the same, this would be easy to estimate. In reality, risks of the same classification have the potential of wide variance. How is this possible? Consider two companies in the same business. One company is ran by management that is very frugal. They are aware of even the smallest expenditures. Their awareness of financial issues also means they are aware of safety, since small losses would affect their resources. This results in an awareness of loss prevention measures.
The second business, while exactly like the first in every way except management, has an entirely different approach. Management does not have the same devotion to spending. They give little thought to small expenditures and do not give any attention to safety and loss prevention. As a result, their claims tend to be higher.
When loss experience is lower than average, there is often recognition of this by the insurers. The recognition may affect the premium the company is charged. Those with lower-than-average risk may allow a “merit rating.” Merit rating attempts to measure the extent to which a particular risk deviates from the average of others in the same class.
Are the insurers just being good guys giving a safety conscience company a premium reduction? Not at all. Merit rating makes loss-prevention activities commercially attractive. The hope is that other companies will want the same lower rating and, therefore, become more alert to safety and loss-prevention issues. It may also minimize the possibility of larger companies turning to self-insurance programs.
There are two broad types of merit rating: experience rating, which is based on the insured’s experience with regard to the risk, and schedule rating, which is based on physical characteristics not measured by the experience rating. Under each of these two types, there may be multiple subcategories.
Since schedule ratings are based on physical characteristics, it is not hard to comprehend. For example, in schedule rating for fire insurance, other than dwellings and a few classes of homogeneous risks, it is necessary to inspect the insured property in order to establish the relative hazards of the individual’s risks.
In lines other than fire insurance, schedule rating has been used for a different reason. Schedule rating is often used to attain equity. This may be the case when low frequency of loss and great variation in physical characteristics exists. In lines that experience high loss, such as casualty lines, greater reliance is placed on the individual risks’ loss record to have a greater degree of reliability.
Some areas of insurance have especially been made aware of loss prevention through the use of schedule rating. Workers’ Compensation Insurance is one of those. There was a significant impact on the development of accident prevention, safety education, and improved appreciation for loss prevention as it relates to insurance cost.
Experience rating is based on statistics. It relies on the record of insured experience produced by the individual risk. Experience rating is based on results in contrast to expected causation, so it reflects all of the loss-producing characteristics of the risk. Of course results are useful only to the extent of data reliability. Because of the limitations of reliable data, experience rating is used most effectively in lines such as liability and compensation insurance where loss frequency is relatively high and the number of units is large. Eligibility rules restrict the use of experience rating to certain lines of insurance and to those risks that exceed a minimum size.[1]
There are subcategories under both schedule rating and experience rating. One such subcategory is Prospective Experience Rating. This method utilizes the experience of the risk over a past period, usually three years, to calculate the rate for the period of time that it would apply to.
While there may be variations in the approach, typically a risk experience loss ratio is determined, the degree of deviation from the expected loss ratio for the class is measured, and the reliability of the deviation is assessed. Although the credibility formula under experience rating is not the same as that used in manual-rate level determinations, the principle is the same.
Retrospective Experience Rating makes a modification of the rate after the expiration of the policy period that is applicable. It is based on the experience of the individual risk developed within the policy period.
Retrospective experience rating is generally superimposed on prospective experience rating. While the manual rate is modified by experience rating to produce what is termed a standard rate, even the standard rate is further modified after the expiration of the policy to reflect the actual experience that became evident during the policy term. Adjustments that follow actual policy experience may be referred to as retrospective adjustments. These are subject to minimum and maximum premium limitations, which would have been specified in advance in the policy terms. Since losses determine premium adjustments, this means that the insured actually has the ability to determine their own future rates by minimizing the possibility of loss. How does one minimize losses? By following safety rules and being aware of how losses happen.
Generally, retrospective rating is tied to a program of grading expenses according to the size of the risk. Obviously, the insurer will charge more for higher risk and less for lower risk. A basic premium is established based on the information the insurer has. It will include a provision for expenses other than loss adjustment and an insurance charge to provide for losses that have not been considered by the minimum and maximum premium limitations. The losses that happen during the time the stated policy period are increased by a loss-conversion factor to include loss adjustment expenses. This figure is added to the basic premium. In addition, a tax multiplier is applied to include the provision for premium taxes, which gives the final retrospective premium. This would, as we said, be subject to the minimum and maximum premium limitations. The actual equation would look like this:
[Basic premium + (Incurred losses X Loss conversion factor)]
X Tax multiplier =
Retrospective premium subject to the minimum and maximum premiums selected in advance.
It is not surprising that some expenses tied to insurance do not vary according to loss experience. For example, insurer general overhead expenses and payroll audit expense is not proportionate with the size of policy premium. Therefore, the retrospective rating must recognize this. It should also be recognized that, as the size of the risk increases, the individual experience loss ratios tend to deviate less from the expected loss ratio of the class. In most types of insurance, the more insureds there are, the easier it is to recognize the amount of loss that will occur. Additionally, as more insureds step forward to share the total risk, it becomes less expensive for the insurer to offer the coverage.
While ratemaking can be viewed as following standard procedures, there do tend to be some variance based on the type of insurance being considered. The lack of an ideal unit of exposure, along with the low frequency of loss, makes fire insurance ratemaking slightly different.
The exposure unit used in fire insurance ratemaking is $100 of coverage for one year. Since the unit of exposure is defined in terms of amount of coverage, for many years loss experience was not particularly useful in determining classification validity. In order for loss experience to be useful, it must be able to reflect loss-producing characteristics. Loss experience data is used to redefine classifications of risk. When a house fire happens losses are often high, but the frequency of house fires is low. As a result, it is difficult to gather the type of statistics needed to influence ratemaking.
There are several factors that must be taken into account in fire insurance ratemaking. While there may be subcategories, the factors most often considered include:
It is easy to understand how a structure’s construction would affect fire losses. Some types of structures would pose a higher risk than other types. Underwriters have known for years that there are factors that affect loss between masonry and frame structures. They have reflected this knowledge in the rates that are charged for fire protection.
Over the years, the two basic structure types (masonry and frame) have been modified to include:
There may be subcategories to the previous list.
Other factors may also be considered, such as structure height, flooring, construction details relating to walls, floors, joists, roof, partitions, and other features. All structure details will be considered from the standpoint of fire loss.
Occupancy
Many fire insurers consider occupancy to be the single most important factor influencing the risk of fire. In any particular type of structure, there are hundreds (perhaps thousands) of possible hazards that will reflect the people occupying or visiting the building. For example, consider two identical structures, with the only variable being the people who use them. One building is used for office space while the other one is used for manufacturing. Obviously, the type of manufacturing will play a role, but regardless of the type it is likely to have a higher danger of fire than would the office building.
The actual causes of fire are so varied and numerous that it would be impossible to cover every one here, but substances used in the building, types of labor utilized, types of manufacture and commerce, and the attention paid to safety all affect the risk of loss. Underwriters consider every aspect in their ratemaking process. Insurers must change their rates to meet changing conditions, since the conditions so strongly affect the chance of loss.
The area surrounding the structure will also affect the chance of loss. Some areas are especially vulnerable. This would include a building located in woodland where the surrounding area would be vulnerable to fire from lightening and other natural causes or from people who have campfires in the area. Mountain lodges would be in this category, for example. It would also include those located where people not associated with the building pose a risk. This would include buildings located in a congested area where structures located besides the building pose a risk from fire.
There are two kinds of protection: public and private. Public protection relates to the availability of help (from city firemen, for example) if a fire breaks out. Obviously, not every location is in close proximity to help from paid firefighters. Even when paid firefighters exist, if the facilities from which they work are poor, it will have an affect of the effectiveness of the protection. Areas that have excellent fire departments and an ample water supply are the most desirable.
When a good fire department exists and an ample water supply is available, underwriters feel that the loss will be reduced even if a fire occurs. On the other hand, if no protection exists or if there is a poor water supply, then a loss would be greater through no fault of the insured. In fact, towns and cities have been rated according to their fire departments, available equipment, water supply, and police departments by underwriters. It is important to make such distinctions since loss can be greatly affected by such conditions.
Private protection relates to devices installed by the owner. This would include water storage tanks (for a water supply), sprinkler systems, fire extinguishers, alarms, or hiring private watchmen. Insurance definitely reflects conditions that reduce the extent of the hazard being insured against. Construction, occupancy, exposure and types of protection will determine the total fire hazard, which will be applied to insurance rates.
The location of the structure being insured will affect premium rates. Besides individual characters of the risk, underwriters know that the geographical area where the structure is located will have a bearing on loss if a fire happens. Losses vary from state to state so consistently that underwriters have programmed geographical location into the premium equation.
Time is not actually a rating factor, but underwriters do know that it must be considered. It is important not to base rates solely on one year’s loss ratio since these can vary from year to year depending on factors that may have affected a particular year. When it comes to fire insurance, five years is generally a minimum time to obtain reliable figures. Recent years will be the most important since they are the most likely to reflect current conditions. Since fire insurance is a low frequency line, good and bad years must be averaged for ratemaking purposes. Years of light losses will enable insurers to have extra funds available for years of heavy losses. If an unusual event occurs there may be extraordinary losses in a particular year, which previous low loss years will help to offset.
Although fire insurance is considered a low-frequency loss insurance line that does not mean that it is not complicated to underwrite. Rates must be instituted in a way that will bring enough premiums in to pay losses and operation expenses. In addition, as every businessman knows, the desire is to also have enough extra to earn a profit. Obviously, a company cannot continue to operate without earning a profit. At the same time, rates must be reasonable. Otherwise, no one will purchase the policy. Besides, most states require rates to be reasonable according to some type of predetermined standard of “reasonable.”
In an effort to simplify the rating procedure, companies try to group risks together based on their similarities, referred to as risk classifications. There must be enough in the group to provide reliable statistics. Groups that are not large enough may produce inaccurate statistics, which could result in a pricing error. The law of large numbers applies to these group ratings. When the number of properties having the same basic loss-producing characteristics is large enough, a major class can be established. As variations develop within this large group, it may become evident that it should be broke into smaller classifications to further define the existing loss producing characteristics. As a result, there is likely to be a greater number of small and intermediate sized classifications rather than large major classes.
When there is such a degree of similarity that it would not be prudent to differentiate among individual risks within the group, an average rate based on the loss experience of the class is applied to each defined risk. There are basically two broad groups of class-rated risks: residential and small commercial buildings. Residential usually applies to one or two family dwellings. Because the numbers are large, occupancy consistent, and premium per risk is small, there is no need to further divide into smaller groups.
With small commercial buildings, the same general conditions exist, although there is more variation in occupancy, construction, and loss exposure than that experienced in the residential group. There are three categories of commercial properties class rated for building and contents:
Each of the three groups may have subcategories that combine specific characteristics related to loss. These subcategories are called “sub-classifications.” Flat charges may be made to account for important nonstandard features that increase the risk of loss. This might include such things as an inadequate water supply or an unusual number of people occupying the structure. These rates for sub-classifications are established and printed in rate manuals.
No matter how hard an insurer tries to avoid it, minor discrepancies will always exist. Even though class rated risks make up the majority of all risks, they still represent less than half of the total fire premiums written.
Some types of risk may seem similar to a particular classification, but their loss results are outside of what would be normal (higher or lower than the average). For example, when more people occupy a structure than is normal, there is more likely to be loss from fire or other damage to the structure. Therefore, even though it appears to fit the parameters of a particular classification, the loss results will be higher than normal. Such distinctions are handled through the use of schedule-rated risks. The underwriter begins with the average for the class, which is how insurance functions, but the average rate is then modified to fit the specific characteristics of the total risk.
There was a time when there were many rating schedules but that has been replaced with the Commercial Fire National Schedule. Developed by using the Commercial Fire Rating Schedule, it is a massive volume detailing specific charges and credits used to determine rates. This text will not go into the Commercial Fire Rating schedule since it would take far more time than this course is designed for. In simplistic terms, there are four major areas involved:
Ultimately, it is the underwriter that brings all elements together to formulate a fair and reliable rating for each type of risk.
Thank you,
United Insurance Educators, Inc.
2014
[1] Property and Liability Insurance by S. S. Huebner, Kenneth Black, Jr., and Bernard L. Webb, Page 631