Objective 1 - Individual Long-Duration Health Contracts

Question 1

Types of antiselection

Answer 1

1. External antiselection - occurs as the person is first becoming insured. Those with expensive health conditions will seek insurance.
2. Internal antiselection - occurs while the person is insured
   a. Buy-down effect - upon receiving a rate increase, some policyholders switch to lower cost plans, so the actual premium increase will be less than expected
   b. Premium leakage - unhealthy individuals are less likely to buy down their benefits. So the claim cost reduction is less than the premium reduction and not enough premium is collected.
3. Durational (cumulative) antiselection - occurs as people make decisions about whether to end coverage. Higher cost insureds tend to keep their coverage in force longer.

Question 2

Mechanisms for controlling external antiselection

Answer 2

1. Individual underwriting before issue - includes initial screening of applicants by the agent
   a. In life insurance, medical U/W is a physician medical exam
   b. In health insurance, medical U/W is a medical questionnaire
2. Pre-existing condition limitations
3. Requiring an enrollment mechanism that doesn’t permit antiselection (for example, minimum participation percentages for associations)

Question 3

Tools used in the underwriting process

Answer 3

1. Individual application - includes medical history, financial information (if needed), and a release to obtain information from 3rd parties
2. Attending physician statement - the insurer may choose to request an APS from any physician listed in the application
3. Commercial databases (such as MIB) - used to check info provided in application
4. Internal data - such as prior applications and claim databases
5. Telephone interviews - these can replace the need for requesting third party information, thereby speeding up the U/W process; also higher level of honesty than on written questionnaires
6. Inspection reports - any info obtained through direct contact with the applicant or others related to the applicant
7. Lab testing - may detect tobacco, illegal drugs, or the presence of some medical conditions
8. Medical exams - due to high costs, rarely used in U/W for medical coverages
9. Tax returns - often the best source of financial info
10. Pre-existing condition provisions - used to protect against antiselection. For some coverages (such as hospital indemnity), these provisions replace U/W entirely.

Question 4

Actions available to the underwriter

Answer 4

1. Offer full coverage with no restrictions
2. Decline coverage
3. Offer coverage at a higher premium rate - the added load may be either temporary or permanent, based on the condition
4. Offer a standard policy with an exclusion rider - the rider excludes coverage for a specific condition or body system
5. Offer a different policy than the one applied for - eg, offer coverage in a substandard risk pool or limited benefit plan
6. Offer a different benefit plan than the one applied for - eg, offer a longer elimination period or shorter benefit period on a disability income policy

Question 5

Criteria for selecting claims to investigate

Answer 5

1. Timing - usually do not investigate claims beyond the time limit for rescinding a contract
2. Conditions - certain conditions (eg, accidents) can be ruled out as being a pre-existing condition
3. Size - don’t investigate a claim if the cost of investigation exceeds the cost of the claim
4. Sentinel conditions or procedures - some conditions are related to others that lend themselves to antiselection (eg, certain diseases may be an indicator of the presence of HIV)

Question 6

Situations in which the CAST model does not work well

Answer 6

1. In the first 3-4 durations, when the impact of U/W wear off overwhelms the CAST effects. The solution is to apply additional U/W selection factors.
2. In later durations, where only a fraction of the original population remains. The solution is to choose a higher value of k2, and recalibrate the model.
3. At all durations, when a rate spiral is severe and volatile. The projection formulas may need stronger terms to fit this type of situation, such as:
   ShockLapse = [RateIncrease - Trend] / [(RateIncrease - Trend) + (1 + Trend) / EF], where
   EF = Elasticity Factor, which measures the price elasticity of the population (eg, may be 1.3 for healthy lives and 0.8 for unhealthy lives)

Question 7

Essential characteristics of a good model

Answer 7

1. Reliable accuracy - a model must be good at predicting the future. It must also be robust (can be used to model input variables over range of possible values).
2. Suitability for use - the model should produce the results it was designed for, without adding unnecessary complications
3. Appropriate precision - this relates to how many decimal places should be kept in the values
4. Sensibility - the model should reflect a logical construction of what is being modeled
5. Effectively communicated - this includes communicating everything necessary to understand and use the model’s results

Question 8

Uses of health insurance financial models

Answer 8

1. Pricing - financial and sales models are used to determine premiums
2. Reserve calculations and reserve basis evaluation - some reserves (such as gross premium reserves) are calculated by forecasting models
3. Monitoring of results - to validate assumptions, to warn of deviations from expected values, and for resource planning
4. Solvency testing - may indicate a need for gross premium reserves
5. Financial forecasting - corporations forecast results for various reasons
6. Actuarial appraisals - these are studies of the value of a block of business, typically used when transferring ownership

Question 9

Steps in building a forecast model

Answer 9

1. Choosing the basic structure of the model
   a. Tools used include spreadsheets, database models, and sequential programs
   b. Model types include asset share models and reserve development models
2. Choosing the info to be carried - the info needed will depend on the purpose of the model
3. Choosing assumptions and building a prototype projection
   a. Starting values and assumptions bust be built into the model
   b. A prototype cell is defined, ad then projected to the end of the forecast period
4. Extending the prototype - after the prototype cell is built, the model must be extended to other cells which represent the different subsets of the business being modeled
5. Validating the model (see separate list of validation methods)
6. Documenting the model - this allows the model to be evaluated by other professionals, and it makes it easier to make modifications
7. Designing output and communicating results - the model output can be useless unless it is put into the context of the question being asked

Question 10

Methods for validating forecast models

Answer 10

1. Starting values are compared directly to the actual values for that year
2. Year to year changes in the model are compared to actual past historical results
3. Model results are checked for reasonableness by people familiar with the business
4. Stress testing - analyze how the modeled results behave when some of the underlying assumptions are changed (includes sensitivity testing)

Question 11

Assumptions needed for forecasting

Answer 11

1. Lapse assumptions - lapse rates vary widely by product, duration, company, and member or policy characteristics. They are generally highest in the first year, and then decrease thereafter.
2. Mortality - some models treat mortality as a separate decrement, but most models combine mortality and lapses (because mortality is a minor assumption for health insurance)
3. Claim costs - it is best to use actual experience when possible. Trend assumptions are needed for determining future claim costs.
4. Expense assumptions - expenses are usually expressed on a per unit basis (such as per policy or a % of premium or claims)
5. Profit assumptions - profits can be measured as an ROI, an ROE, or a % of premium
6. Model office assumptions - these assumptions define the proportion of the block of business that is represented by each model cell

Question 12

Bases used as expected amounts for A/E analysis

Answer 12

1. Original pricing assumptions - management likely reviewed these assumptions when the product was being developed, so management expectations may be based on these assumptions
2. Profit targets - this is the bottom line metric that most senior management is interested in
3. Current pricing - may be the most useful measure for inflation sensitive products, since inflation targets are not reliable over the long term
4. Tabular - for DI coverage, a published table is often used for comparison. For DI and LTC, companies with large amounts of data may develop their own internal tables for comparisons

Question 13

Types of disability income claim experience studies

Answer 13

1. A/E morbidity - this is the most preferable method of examining disability income experience, but there is often not enough data for morbidity studies. Morbidity consists of:
   a. The rate of disability - the # of disabled lives / 1000 lives exposed
   b. The rate of recovery -measures the length of disability. The # of disabled lives that will recover at different points in time / 1000 disabled lives
2. Loss ratios - due to the limited amount of data, most studies are based on claims ratios:
   a. Cash ratio - claims $ paid out / earned premiums
   b. Incurred claims ratio (preferred) - (claims + ALR + claim reserve) / earned premium

Question 14

Types of reserves in disability income insurance

Answer 14

1. Active life reserve - exists for policies priced on a level-premium basis. Consists of the excess premiums charged in early years to cover the premium shortfall in later years.
2. Disabled life reserve - established to cover each disability claim and its projected length

Question 15

Factors that stimulate product development for disability income

Answer 15

1. Responding to the competition - because the marketplace is so product sensitive, the disability income insurer must be quick to react to its competitors’ product changes
2. Consumer demands - as the consumer has become more aware of the need for long-term disability protection, the DI product has evolved. The product development process must be responsive to whatever changes emerge in the future.
3. Claims experience - this must be monitored on an ongoing basis. When pricing assumptions prove to be inaccurate, changes in product language, rate structure, or U/W may be needed for future sales.
4. Gov’t influences - for example, the expansion of the Social Security disability program in the 1970s affected DI insurers. Regulatory and tax changes can also impact DI insurance.

Question 16

Areas that particiate in the product development process

Answer 16

1. Sales and marketing - these individuals are the closest to the consumer and the competition, so they are a good source of product ideas and competitive info
2. Other home office disciplines
   a. Actuarial - must determine how the product is to be priced and what safeguards may be necessary to develop competitive rates
   b. U/W - must determine whether new U/W approaches are necessary
   c. Claims - must determine what risks and factors it will face in administering claims
3. Data processing and systems - must be involved early since new products may require substantial modification to existing systems. Timely identification of potential issues may reduce project costs.
4. Legal - some product ideas may present special legislative and regulatory problems, and the attorney may be able to offer alternative approaches or solutions that avoid problems in state approval
5. Investments - the significant impact of investment returns increases the importance of this function

Question 17

Considerations in pricing LTC

Answer 17

1. Morbidity (see separate list)
2. Investment earnings assumption - very significant because of the long term nature of the product
3. Expenses (excluding profit) - non-commission expenses are expected to average 13% to 18%. Commissions are very high in the first year.
4. Voluntary lapses - higher 1st yr and lower later year lapses increase premium
5. Mortality - 1994 GAM is the table specified in NAIC Model. Many actuaries feel the rates in this table are too high.
6. Surplus strain/reserves - reserves have a significant impact on pricing due to the conservatism required and long duration of policies. Significant surplus strain occurs due to reserve and RBC requirements, the cost of capital, and policy issuance expenses.
7. Profit - takes 7-10 yrs to emerge. Use % of premium, statutory IRR, or GAAP ROE.
8. Loss ratio requirements - most states require 60% minimum for individual (higher for group). The NAIC Model Reg removes the LR requirement, in favor of a certification of rate adequacy under moderately adverse experience.

Question 18

Parameters to consider in disability income claims or persistency study

Answer 18

1. Occupation class - there are significant differences in morbidity and U/W approaches from one occupation class to another
2. Occupation - each class is made up of many occupations, and each occupation may perform somewhat differently based on socioeconomic trends
3. Policy form - a study by policy form is needed to determine whether pricing assumptions were correct for new forms
4. Extra benefits - some optional benefits (such as cost of living) require significant reserves
5. Age - changes in medical treatment and technology will affect age experience
6. Duration - due to the wear off of U/W selection, loss ratios will be higher on older blocks of business and extremely low on new blocks
7. Elimination period - changes in experience may occur at one elimination period and not at another
8. Benefit period - to-age-65 and lifetime benefits may affect the election of early retirement
9. Indemnity (benefit amount) - some studies have shown that the larger the indemnity, the poorer the experience
10. Income - studies have shown that higher replacement ratios (benefit amt / income) lead to higher morbidity
11. Geography - densely populated areas may have higher morbidity than less populated areas
12. Agent and agency - data by agent can provide info on the ability of the agent to select good risks
13. Sex - higher morbidity for women has been demonstrated at least up to the mid-50 age grouping
14. Mode of premium payment - annual premium payment mode generates more favorable experience, while the quarterly mode is least favorable
15. Smoking status - nonsmokers have lower disability costs than smokers
16. Combinations of the above parameters - to determine interactions

Question 19

Considerations in establishing morbidity assumptions for LTC

Answer 19

1. Data sources - insured data (such as intercompany study) is best but more population-based data exists. Population data must be adjusted for data bias and other issues.
2. Integration of coverages - if there is only one type of benefit, then utilization is increased by inappropriate placement
3. Reinstatements - should be handled consistently in pricing and administration
4. Transfers - reflect consistently in frequency and continuance curves
5. Coordination with other coverage - tax-qualified plans must coordinate with Medicare
6. Pre-existing requirement - not expected to produce dramatic savings due to the heavy level of U/W
7. Level of care/charge levels - daily costs of care vary by type of care
8. Area - utilization and charges vary significantly by area for NH and HH services
9. Policy options and benefit triggers - richer plan designs have greater potential for adverse selection
10. Age/gender - costs increase greatly with age. Females have higher costs.
11. Marital status - costs for married may be 1/2 the costs for singles
12. Morbidity improvement - claim costs for a given age may be decreasing over time (debatable)
13. U/W - strongly impacts experience in early durations (and potentially even later durations)
14. Marketing - broker and career agents
15. Claim administration - may include various levels of care management
16. Reinsurance - arrangements include coinsurance, modified coinsurance, yearly renewable term
17. Regulatory considerations - the actuary must certify that rates are sustainable under “moderately adverse experience”

Question 20

Rating approaches for Medicare Supplement

Answer 20

1. Attained age - rates are based on the individual’s current age
2. Issue age - rates are based on the individual’s age when the policy was issued
3. Community rates - all participants pay the same rate. Some modified community rating approaches differentiate based on age, sex, duration, or other parameters.

Question 21

Medicare Supplement pricing assumptions

Answer 21

1. Morbidity - past claim costs need to be trended forward to the rating period
2. Mortality - this is not a significant assumption for Med Supp, and is frequently combined with the persistency assumption
3. Persistency - this should be based on the company’s experience for similar products
4. Investment earnings - these will be credited to the various types of reserves that are held
5. Selection factors/ U/W - for underwritten policies, selection factors may be used for the first 1-3 yrs
6. Age and sex distribution - most policies are sold to individuals turning 65
7. Smoker vs. non-smoker - if rates vary by smoker status, then the distribution of smoker status must be estimated
8. Area factors - claim costs by area may come from rating manuals or gov’t statistics
9. Expenses and taxes
10. Other considerations - modal factors and policy fees are sometimes used

Question 22

Steps for developing critical illness incidence rates

Answer 22

1. Start with general population age-specific incidence rates from gov’t sources and research organizations for the various illnesses covered
2. Adjust these rates to fit the condition definitions in the policy
3. Apply any applicable trends (such as a decrease in heart attack rates)
4. Use ratios of insured lives to population mortality to adjust rates from the general population to an insured population
5. Use ratios of nonsmoker to smoker mortality to segment rates into nonsmoker and smoker rates
6. Use ratios of select to ultimate insured mortality to create select and ultimate rates
7. Compare the rates to any available insurance experience and adjust as deemed necessary
8. Sum the rates for each of the major conditions covered, and then add small amounts (about 1%) for each additional covered condition

Question 23

Types of critical illness policies

Answer 23

1. Standalone - offers coverage only for CI
   a. Basic - covers only cancer, heart attack, stroke, and sometimes coronary artery bypass graft
   b. Enhanced - includes 15-20 additional conditions and costs about 30% more
2. Acceleration - combines coverage for both CI and death. Pays the face amount on the earlier to occur of CI or death.
   a. An alternative is partial acceleration. Some percentage (25-50%) of the face amount is paid for CI, after which the remaining face amount remains in force as death protection only.

Question 24

Definition of critical illness insurance

Answer 24

Critical illness is an insurance product that pays the face amount when:

1. The insured is diagnosed with a condition covered by the policy. The diagnosis must be made by a doctor and must be supported by objective evidence.
2. The condition meets the definition in the policy and is not excluded by any other policy provision (see separate list of covered conditions)
3. The insured survives for a specified period (usually 30 days) following diagnosis

Question 25

Conditions covered by critical illness insurance

Answer 25

Typically covered in basic policies:

1. Life-threatening cancer
2. Heart attack - may exclude mild heart attacks that occur within a couple of days following angioplasty
3. Stroke - the definition requires a measurable neurological deficit that persists for 30 consecutive days
4. Coronary artery bypass graft - similar procedures which do not involve grafts are always excluded

Covered in enhanced policies:

1. Multiple sclerosis
2. Kidney failure requiring dialysis
3. Major organ transplants
4. Cardiovascular: heart valve replacement and aortic surgery
5. Degenerative: motor neuron disease, Parkinson disease, and Alzheimer disease
6. Brain: coma and benign brain tumor
7. Head: blindness, deafness, and loss of speech
8. Body: loss of limbs, paralysis, major burns, and occupational HIV
9. Loss of independence (covered by only some companies)

Question 26

Optional product features on critical illness policies

Answer 26

1. Return of premiums on death
2. Return of premiums on expiry - returns premiums on the policy’s expiration date if the policy is still in force
3. Return of premiums on surrender - returns a defined % of the premiums upon surrender prior to the expiry date. The % may increase to 100% over time.
4. Face amount increasing (to keep up with inflation) or decreasing (to match the declining principal remaining on a mortgage loan)
5. Partial benefits (10-25% of the face amount) payable for some non-life threatening conditions which have been excluded in the policy
6. Assistance benefit - provides medical consulting advice for the diagnosed CI
7. Guarantee that the premiums will not change

Question 27

CumulatIive Antiselective Theory (CAST) Model

Answer 27

1. An underlying involuntary lapse rate is assumed for all insureds (healthy and unhealthy). Above this rate, lapsation varies between healthy and unhealthy (lower lapse rates) lives.
2. Cumulative antiselection is measured by looking at experience by duration.
3. Healthy insureds can move to the unhealthy group over time.
4. Partition models rank members from highest to lowest cost and split them into healthy and unhealthy subsets. Projections for these subsets are modeled separately.

Question 28

Pricing challenges for disability income insurance

Answer 28

1. Pricing must react qucly to social and economic changes
2. DI pricing actuaries must understand the cyclical nature of the business
3. Scarcity of experienced actuaries
4. Lack of mature experience data
5. Competitive pressures
6. Steep claim cost curve for DI

Question 29

Issues in LTC claim reserving

Answer 29

1. Relatively final data is used in developing continuance tables, but tables may be applied to ongoing claims with poorly defined service dates
   a. For example, if payment on services is pending for some reason, the claim appears to be inactive for longer than is really the case, resulting in application of larger termination probability in the reserve calculation.
2. Bias may be created due to claim coding practices.
   a. For example, if claimant has recovered or died, administrator may close the claim, but if there are still some payments for services outstanding, a reserve may not be set up and removes the shortest remaining claims from data used to set reserves.
   b. Approach to account for bias is to keep all claims open, but this makes the reserve more conservative.

Question 30

Advantages of LTC Combination Products

Answer 30

Product in the article is an LTC accelerated death benefit rider on a universal life policy.

1. Lower premiums compared to standalone LTC coverage
2. Avoids concerns with steep claim cost curve for LTC

Question 31

Considerations in pricing LTCI accelerated death benefit riders

Answer 31

1. Benefit structure - are LTCI payments based on the original death benefit or the amount at time of acceleration? How do other riders impact the benefit (i.e. COLI)?
2. Mortality - must be identified separately for those triggering LTCI benefits (will have higher mortality)
3. Guaranteed minimum benefits - if minimum death benefit is guaranteed even for claimants receiving the entire death benefit in LTC payments, more detailed modeling is needed
4. Additional reserves - multiple reserve bases may be needed
5. Claim reserves - DLR should be held for insureds receiving benefits and the death benefit reserves should be reduced
6. Taxation - does not attract a premium tax or the federal DAC tax