This site is 100% ad supported. Please add an exception to adblock for this site.

Group Ins.- Experience Rating & Special Funding


undefined, object
copy deck
Experience Rating and Special Funding Methods
This is the most important chapter in the syllabus. 10 out of the 120 points on the exam will come from this chapter. You will be asked to:
ï‚Ÿ Compute the renewal premium rate for a group, using both stop-loss and credibility pooling;
ï‚Ÿ Compute the retrospective experience refund for a group.
ï‚Ÿ Describe considerations in establishing a credibility factor for a group.
Experience Rating vs. Pooling vs. Community Rating
Experience Rating
Experience rating means that an employer group’s premium is based, at least in part, on its prior claims experience. Employer groups are charged low rates if their past claim costs have been low, and high rates if their past claim costs have been high.
Pooled premium rates are rates based on the demographics and underwriting factors of the employer group, but not on its actual experience. A rate manual specifies rates to be charged to an employer group policyholder based on the age, sex, health, family status, industry, and location of the employees, and on the benefit plan type and the group size.
Any two groups with the same demographics will be charged the same premiums, even if one of them has never had a claim and the other has had many claims before. Pooling is also called manual rating.
Community Rating
Community Rating is a regulation, not a methodology in and of itself. Community Rating regulations limit the extent to which age, sex, health, industry, location, and the rest of the above can be used in differentiating rates to be charged to different policyholders. Community Rating also prevents or severely limits the use of experience rating.
In pure community rating, each policyholder with the same benefit plan must be charged the same premium rate, regardless of its age, sex, demographics, or prior experience.
In modified community rating, policyholders with the same benefit plan can be charged different rates based on age, sex, industry, and family status, but those are the only factors that can be considered.
Blended Rates
Only the largest employer groups have claims experience that is so statistically valid that it can be used alone to determine a premium rate. For medium-sized groups, a blend of experience rating and pooling (manual rating) is used.
Essentially, when setting a premium rate for a medium-sized group, two premium rates are computed: one based purely on the group’s past experience and one based on the manual rate. The group is then charged an average of those two premium rates.
For small groups, the claims experience is too volatile to be statistically credible at all; thus manual rating is used exclusively.

Pooling can also be applied to just a portion of expected claims; say, very high claims. For example, a group might be experience-rated, but include stop-loss pooling; that is, the group’s potential for a very high claim is combined with the potentials of other groups for very high claims when establishing its premium rate.
Prospective vs. Retrospective rating
Prospective rating
Prospective rating means that the premium rate is set in advance of the coverage period. Regardless of the claim costs that occur during the coverage period, there is no refund or additional charge.
Retrospective rating
Retrospective rating means that the premium rate is somehow adjusted after the year of experience is over. The policyholder receives a refund if claim cost has turned out to be lower than the premium. Conversely, the policyholder must pay an additional amount if claim costs have exceeded the premium that was charged.

Manual Rating is always prospective, since the premium is set in advance (based on the demographic factors of the policyholder) and no adjustment is ever made after the policy year is over.
Experience Rating can be prospective or retrospective. Usually it is prospective. Retrospective adjustments are administratively expensive, and thus are only offered to the largest employer groups.
Self-Insuring vs. Partial Insuring vs. Full Insuring
Self-insuring means that an employer pays for all of its employees’ medical costs itself. A self-insuring employer usually hires an insurance company just to handle the administrative work. The type of policy it buys is called an ASO policy (Administrative Services Only).

Partial Insuring means the employer pays for all of the insureds’ costs up to a specified dollar amount. A partially-insuring employer purchases stop-loss insurance from a commercial insurance company to cover larger claims.

Full insuring means an employer purchases a commercial insurance policy or HMO plan to cover all of the medical claims of its insureds.
Reasons to Use Experience Rating
ï‚Ÿ Competitivity: Groups that have very few claims prefer experience rating, because the rates are lower. Under Pooled Rating, they have to subsidize unhealthy groups.
ï‚Ÿ To prevent antiselective lapses: If the healthiest groups are unsatisfied with their premiums, they will lapse, leaving the unhealthiest (highest-claim groups) remaining.
 When a group’s experience is credible.
 Large employer groups have the strongest credibility.
 High-frequency, low-severity coverages, such as medical insurance, have the best credibility
 Claim types that are not independent from one year to the next (e.g. medical claims) also give a group’s experience high credibility.
Reasons to use Pooling
 When group’s experience is not credible:
 for small groups
 for low-frequency, high-severity coverages, such as LTD
 for claim types that are independent from one year to the next (e.g. accidental death)
ï‚Ÿ b/c Pooling is easier and cheaper than experience rating.
ï‚Ÿ When Community Rating regulations prohibit experience rating.
Other Considerations in Deciding between Experience Rating and Pooling
 The ins cpy’s administrative ability to do experience rating
 Management’s Philosophy regarding experience rating
Difficulties With Experience Rating
(“Complexities in developing an experience rating system”)
The shaded items below are from SN 601 Ch. 6. They are relevant here, but please note that these are on the extension syllabus. I believe that they will not be found in the solution to a morning question.
 It’s difficult to combine experience rating and pooling or community rating
ï‚Ÿ Retrospective experience rating is difficult
 Experience rating can’t be used for new groups with no prior experience
ï‚Ÿ An MCO may have to offer a medical group a different capitation rate per patient, based on which employer group each patient belongs to. This is difficult to administer.
ï‚Ÿ A group could have high experience costs for hospital, but low experience costs for physician services. (The MCO must analyze all the components of health care use independently)
ï‚Ÿ Converting from a pooling method (or community rating) to experience rating can cause undesirable results (like sudden required rate hikes)
 use a formula to cap the impact on existing rates
 phase in experience rating gradually.

ï‚Ÿ State approval is required for the experience rating method

The steps in prospective experience rating for a group policyholder are as follows:
1. Analyze the Group’s Past Claims Experience
2. Determine the Credibility to be assigned to the group’s past experience
 The main determinants are the size of the group and the type of insurance coverage.
3. Apply Pooling to all, some, or none of the group’s assumed claims distribution
 Pooling removes a portion of the group’s liability for its actual claims, and replaces this liability with a flat pooling charge.
 The more credible a group is, the less it is pooled.
4. Apply a Trend/Inflation factor
 Claim costs must be projected forwards from the midpoint of the prior experience period to the midpoint of the upcoming coverage period.

The result is the Net Premium (Pure Premium).
5. Add loadings (margins)
6. Final Adjustments
e.g. special negotiation with the policyholder.

The result is the Gross Premium.

Each of these steps is discussed below.
(“Development of Claims Experience”)
 The goal is to compute the group’s incurred claims.
ï‚Ÿ Incurred Claims include all claims, regardless of whether paid yet or not, which became liabilities during the experience period.

Incurred Claimsexperience period
= Paid Claimsexperience period
+ Increase in Claim Reservesexperience period
Paid Claims include:
ï‚Ÿ Claim amounts paid
 Loss Adjustment Expenses — the expenses of processing the claims
 Liabilities due to conversions of participants’ policies from group to individual
Companies are often required by law to offer conversions. But converted policies are not self supporting, since they are usually elected by people who know that they need ongoing medical treatment.
Claim Reserves Include:
ï‚Ÿ IBNR reserve
ï‚Ÿ ICOS liability
ï‚Ÿ Due and Unpaid liability
ï‚Ÿ miscellaneous recurring liabilities
ï‚Ÿ Loss Adjustment Expenses expected to be associated with future claims.

(See SN 305 for more information about incurred claims and claim reserves.)
Considerations in recognizing the experience of small groups
Small groups’ claims are especially volatile. However, even though small groups should technically be fully pooled, they often request to be experience-rated, and, in an effort to remain competitive, insurance companies often comply with this request. This section describes two ways in which an insurance company can take into account the past claims experience of a small group, despite its inherent volatility.
Formula-Based Rating (aka Band Rating or Tier Rating)
 Each small group is initially assigned an expected claims cost, μ, based on its demographics.

ï‚Ÿ Eleven rate bands are created:
40-49% of μ
50-59% of μ
140-149% of μ
 After each policy year, the group is assigned to a new rate band, based on some formula for how its claims experience has deviated from μ.
ï‚Ÿ Regulations, Competitivity, and Market Forces are also considered when reassigning the group to a new rate band.
 There are limitations; for example, a group can’t be shifted by more than three bands in any one year.
Case-By-Case Reunderwriting
ï‚Ÿ Examines the causes of each small group's experience in order to determine a the premium rates.
ï‚Ÿ The Ins Cpy tries to determine whether a large claim was a one-time anomaly, or is likely to reoccur.
 Obviously, the more small groups the ins cpy insures, the less possible it is to do this. (Time-consuming / expensive)
A Combination
ï‚Ÿ A combination of formula rating and reunderwriting can be used.
ï‚Ÿ For example, larger groups have more significance to the insurer.
 So, a tiny group with 2-5 employees might be reunderwritten only if total claims > 150% μ. But a slightly larger group of 50 employees might be reunderwritten if total claims are > 120% μ.
Credibility is a number between 0 and 1.
C=0 means the group's actual claims experience is not to be used at all in determining the premium rate to be charged.
C=1 means the group's experience is fully credible, and that the group will not be pooled at all.

Considerations in evaluating the credibility level:
ï‚Ÿ The larger the group, the more credible its experience
ï‚Ÿ Coverages with high frequency and low severity (e.g. medical insurance) are the most credible.
 Stated another way, the lower frequency and higher severity the coverage, the greater the exposure base required in order to achieve full credibility.
ï‚Ÿ The higher the variability (variance) of possible claim costs, the lower the credibility.
 The insurer’s desired confidence interval
ï‚Ÿ The measure of credibility
(in other words, the measure chosen as one exposure unit.)
 Usually the unit is each life. But if claims are stochastically independent over time, then life-years can be used as an independent unit. There are more life-years than lives, and so the insurance company can increase credibility by using this unit instead.
ï‚Ÿ Effect of a change in credibility philosophy on existing business.
ï‚Ÿ Competitive pressures
 Insurer’s administrative ability to analyze the claim distribution
Pooling is an algebraic method of combining a group's experience with the experience of other groups in the same demographic. Pooling is usually not part of the contractual relationship between the insurance company and the policyholder. It is merely a way for the insurance company to decide what premium to charge each policyholder.
Pooling removes a portion of the group’s liability for its actual claims, and replaces these so-called pooled claims with a flat pooling charge.
Several types of pooling can be used together.

Pooling is poorly described in this chapter. For more information, I recommend reading the following materials:
 11/2002 Core Case Study: Tables MM-2a, MM-2b, and MM-3a.
 The 2001 Exam – Question #3 and its solution

These materials provide the numerical examples that you will need to handle the exam questions from this chapter.
Regular Pooling (also called Credibility Weighting)
 The group’s incurred claims experience per employee per month is computed.
ï‚Ÿ A manual claims rate per employee per month is also computed. (see note below)
ï‚Ÿ The premium charged to the group is:
Z * Experienced Claim Rate + (1–Z) * Manual Rate
where Z is the credibility of the group.

ï‚Ÿ Another way of writing this same formula is:
Z * Incurred Claims + (1–Z) * Expected Incurred Claims.

Note: Claims Experience and Manual Rates in the Case Study are provided on a per employee per month (PEPM) basis, rather than on a per member per month (PMPM) basis. PEPM rates are also called composite rates.
Computation of the Manual Rate:
Manual Rate PEPM =
Base premium rate PEPM (from the rate manual)
x age/sex factor
x regional factor
x benefit plan factor
(e.g. plans with higher deductibles have lower premiums.)
Catastrophic Claims Pooling (Stop-Loss Pooling)
 Individual Stop-Loss pooling removes the portion of each claim that is above $50,000 from the group’s experience.
 In exchange for this “forgiveness”, a pooling charge is added to the group’s premium rate.
ï‚Ÿ Aggregate Stop-Loss pooling reduces aggregate claims at year end to, say, $1,000,000, if aggregate claims are greater than $1,000,000. There is a pooling charge for this type of pooling also.

On the Exam:
ï‚Ÿ The Case Study only makes use of individual stop-loss pooling (for claim amounts in excess of $50,000).
 This stop-loss pooling applies to every group’s experience, even 100% credible groups. Pay close attention to this. It is completely separate from the regular “manual-rate vs. experience rate” type of pooling.

(See example on next page)
Multi-Year Averaging
This type of pooling allows you to pool one group with itself!
 The claims experience used in setting a group’s premium is actually a weighted average of the group’s experiences over the last few years.
ï‚Ÿ Like stop-loss pooling, Multi-Year pooling is independent of credibility pooling, and is applied first.
ï‚Ÿ It is unclear how multi-year pooling interacts with stop-loss pooling. The solution to Question #3(d) on the 2001 Sample Exam shows multi-year pooling being used instead of stop-loss pooling.

ï‚Ÿ Multi-year averaging is not useful when there is
 high turnover in the group
 any time older experience is not useful.
 for medical insurance
Note: At another point in this chapter, the author says that medical claims are dependent from one year to the next; therefore, this would seem to imply that multi-year averaging would be useful for medical claims.
Loss Ratio pooling
ï‚Ÿ Loss ratio pooling is equivalent to aggregate stop-loss pooling.
All of the above examples have been inaccurate, because we have not accounted for inflation.
 The group’s experienced claims have occurred during a prior coverage period.
 The rate manual’s assumed claim costs were only valid on, say, January 1, 2003.
 When choosing next year’s premium rate, all prior claim costs must be projected forward at an assumed inflation rate.
Applying Trend Factors
ï‚Ÿ Experienced claim costs are trended from the midpoint of the experience period to the midpoint of the upcoming coverage period.
ï‚Ÿ Manual claim costs are trended from the beginning of their effective date to the beginning of the coverage period.
Numbers that must be trended:
Experienced Claims
ï‚Ÿ Incurred claims during the experience period
ï‚Ÿ Catastrophic claims (> $50,000) during the experience period
ï‚Ÿ Prior coverage period claims used in multi-year averaging

Manual Claims
ï‚Ÿ Base manual claims rate
ï‚Ÿ The stop-loss pooling charge in the manual table

At this point, I recommend that you read Question #3 from the 2001 Sample Exam, and its solution. This is the format that you should imitate when you write your exam.
Note At one point, the student appears to trend the manual rate backwards instead of forwards. I believe that this is either a typo on the part of the SoA in transcribing the solution, or a mental lapse by the student.

Note You can easily score full credit on these questions without getting everything right. For example, just by demonstrating that you’re aware that certain things have to be trended, you will probably score most of the “trending” points.
Claim Cost Trend includes the effects of:
ï‚Ÿ inflation
ï‚Ÿ legislation
ï‚Ÿ economic factors
ï‚Ÿ utilization of medical care
ï‚Ÿ provider reimbursement trends
ï‚Ÿ plan benefit changes
ï‚Ÿ mortality/morbidity trends
ï‚Ÿ demographic changes
See “General–Trends” for a consolidation of this list with similar lists.

 some trends are group-specific; some trends apply to all groups
 some trends are one-time (e.g. benefit changes); some are continuous (inflation)

Types of Retention Items (loadings) (margins)
ï‚Ÿ Expense Loading
 the largest component.
ï‚Ÿ Commissions
ï‚Ÿ Deficit Recovery Charge
 Added to premiums if the policyholder still owes money from past losses.
ï‚Ÿ Termination Risk Charge (aka Risk Charge)
 This loading protects the ins cpy in case a policyholder that is in debt terminates its contract.
 The size of the TRC should reflect
 the ph's ability to pay,
 any existing deficits
 the inherent risk of the group and its funding arrangement
ï‚Ÿ Profits
ï‚Ÿ Investment Income
 An Ins Cpy sometimes credits a policyholder with interest on the reserves that it is holding for this policyholder. (Investment Income is usually a negative loading)
ï‚Ÿ Explicit Margin
 a comfort factor.

ï‚Ÿ Misestimation Risk Charge and Trend Risk Charge
 these loadings apply when premium rates are guaranteed over periods of longer than a year.
 They reflect the risk of underestimating the claims, expenses, risks, and trend over the guaranteed period (during which the ins cpy will not be able to raise rates to compensate).
The Insurance Company should also consider the following factors in setting premiums:
ï‚Ÿ Politically sensitive policyholders
ï‚Ÿ Exposure changes
 if a group’s exposure grows rapidly during the year, then its experience should be trended not from midpoint to midpoint, but from, say, 2/3 to 2/3, since the average exposure date is later in the year.
ï‚Ÿ Manual rates and factors must be updated frequently
ï‚Ÿ Multi-Option Considerations
 An insurance company providing Indemnity coverage to an employer must find out whether the employer is also offering HMO’s and PPO plans (which are perhaps being provided by other carrier). If so, the ins cpy will be hit with unexpected antiselection, and so should charge higher rates.
Employers usually offer their employees an HMO, a PPO, and an Indemnity health plan. It is well-known that healthy people tend to choose the HMO and unhealthy people tend to choose the Indemnity plan.
Retrospective Rating is Most Appropriate⬦
ï‚Ÿ For large groups
 retrospective adjustment is expensive, so not worth it for small groups
ï‚Ÿ For financially stable groups
 so that the group can pay the additional premium at end of year, if necessary
ï‚Ÿ For groups that want a competitive rate
ï‚Ÿ When the ins cpy wants to be able to recoup any losses that occur
ï‚Ÿ When the Ins Cpy is financially stable (has a large surplus)
 Ins Cpy is subject to default risk (the group might terminate while in a deficit position)
ï‚Ÿ depending on company policy
ï‚Ÿ depending on the Funding Method used
see “Special Funding Arrangements”, below

ï‚Ÿ Retro. Rating is difficult under Managed Care health coverages:
 Managed care providers are offered contractual bonuses/incentives if claim costs are low.
 If claim costs are low, who should get the refund: the providers or the employer group?

Claims Reserves are larger for Retrospectively-rated groups, because:
 any overconservatism is eventually refunded to the policyholder, so policyholder isn’t damaged by the high margins
ï‚Ÿ There is added risk that the ph will terminate while in a deficit position
ï‚Ÿ more conversions
retrospectively-rated groups are large, and have more employees who may convert to individual policies.

Recall: Claims Reserves are part of the group’s incurred claims experience.
There is a Rate Stabilization Reserve
 This is a buildup of the policyholder’s surplus, held by the ins cpy.
ï‚Ÿ It protects the ins cpy from default risk.
ï‚Ÿ The lower-frequency, higher-severity the coverage (like LTD), the more moneys should be held.
Investment Income is credited
ï‚Ÿ It is important for the ins cpy to credit interest on the surplus it is holding (since retrospectively-rated groups are large and want competitive policies)
ï‚Ÿ Investment Income = r% * mean fund
 where “mean fund” is a really rude, angry fund.
 mean fund = average of starting and ending fund amounts.

Note: There is no investment income mentioned in the Case Study on the exam. Just assume it is 0. Mentioning it might be worth a point, though.
How to compute the Experience Refund:
New account balance =
prior account balance carried forward
+ premiums paid in
+ investment earnings credited
– claims charged (see below)
– Loadings, commissions, and expenses.
Experience Refund = Account balance in excess of rate stabilization reserve.

Note: On the exam (in the case study), the Rate Stabilization Reserve is $50,000.

As in the prospective methods:
Claims Charged = Incurred claims – pooled claims + pooling charge
[+ conversion charges]
(unclear whether conversion chgs s/be
added here or as part of incurred claims)

All amounts in the above formula must be computed on an aggregate dollar amount basis, not on a PEPM or PMPM basis. After all, the bottom line is the dollar amount of the refund.
Funding Arrangements refer to:
 how and when cash will flow,
 whether and how risk is transferred between the ins cpy and the policyholder.
Prospective and Retrospective rating are two types of funding arrangements.
Self-Insuring and Fully Insuring are also two types of funding arrangements
Reserveless Plans
Typical agreement:
ï‚Ÿ The ph is exempt from premiums up to a specified percentage of the claims reserve. Periodic premiums also have a grace period.
 This effectively means that the policyholder is holding his own reserves, instead of the insurance company holding them.
 This allows the policyholder to invest the money as it desires.
 A reserveless plan creates a one-time¬ reduction in premiums for the policyholder.
ï‚Ÿ The ph must pay all outstanding liabilities if the contract terminates (a terminal premium).

Risks to Insurer:
ï‚Ÿ Insurer must set aside some offsetting, existing, asset to take the place of the ordinary claims reserve.
ï‚Ÿ The main risk is default risk.
 Therefore, ongoing “financial underwriting" of the group is necessary to ensure the group’s ability to pay.
Self-Insured Plans
ï‚Ÿ The employer is responsible for all claim payments and expenses.
 No insurance company is a guarantor.
ï‚Ÿ A trust is used as the funding vehicle. All cashflows flow into and out of it.
ï‚Ÿ An insurance company or Third Party Administrator is used for ASO services.
Minimum Premium Contracts
ï‚Ÿ An old (1960's) method, now obsolete.
ï‚Ÿ Hybrid of self-insured and fully-insured.
ï‚Ÿ Similar to stop-loss coverage.
Partial Self-Insuring
ï‚Ÿ Employer purchases stop-loss insurance (specific or aggregate) to protect himself against unusally high claims.
ï‚Ÿ Employer may also pay an insurer to cover the conversion policies for ees who terminate.
Stop-Loss Insurance
Considerations in Stop-Loss:
For the employer group
ï‚Ÿ Individual or Aggregate
ï‚Ÿ Large groups buy stop-loss policies with higher attachment points
(since they can withstand a few large claims.)

For the Ins Cpy
ï‚Ÿ Large groups have more predictable claims distributions, so the premium can be less.
ï‚Ÿ Misestimation risk
 Expected Claims must be very carefully estimated (using demographics, prior experience, etc.)
 extra conservatism is required
 especially for high-severity coverages like LTD or AD&D
 Setting expected claims too low causes the assumed probability of a claim exceeding the attachment point to also be inaccurate.
since the Attachment Point is expressed as “150% of expected claims”
ï‚Ÿ The stop-loss deductible causes leveraging.
See “General–Leveraging” for more info.

Retrospective premium rider (on a standard insurance policy)
ï‚Ÿ The ph takes on some of the aggregate claims risk, in return for a lower premium rate.


Deck Info