Product problems often become apparent only after earnings get hit or cash loss ratios soar. My premise is that it does not have to be that way, that an experienced insurance executive, applying what he already knows, plus common sense, can gain a pretty good idea whether a product is going to be profitable or not. The executive who says "the actuary says the product is all right so I don't have to worry about it" doesn't understand the product design process. The actuarial work is important, but is partially based upon a set of assumptions furnished to the actuary by the marketing department. Those assumptions tend, naturally, to be on the optimistic side. Almost any product can be profitable or unprofitable. It depends on where and how it is sold. In retrospect, the assumptions which underlie every disastrous product prove to have been incredible given the known sales situation in the field. The trick is for the executive to measure those assumptions against his own knowledge and common sense up front.

Occasionally a product and its selling situation may appear to fail the common sense test, but a knowledgeable executive will conclude it can work in a particular situation. For example, a guaranteed issue life product sold by funeral directors doesn't on the surface sound like a business you would want to be in. And yet there are companies that are in it and appear to be surviving. Then the question becomes "What am I hoping for?". What is it about my particular situation that I am counting on to make this profitable? When you isolate that, you know why you are in that business, and you know what to check on down the road to make sure it is really happening. And then there are those situations you know are bad, but you are desperate for activity and agents, hoping to gain some, and are planning later to shift production from a bad product to a good one. Risky, but it has been done, although it was usually by accident rather than design.

The products discussed here are viewed from the standpoint of the company, not the agent or the consumer. Also, it is assumed for the most part that the reader understands the coverages, all of which are widely discussed elsewhere on the web. Finally, the purpose is not to provide a comprehensive discussion of any product, but rather to offer observations and cautions on specific points.

Some products are so heavily lapse supported that it becomes the outstanding feature, and are discussed in the section on lapse support rather than on this page. Those products are so volatile that they can be extremely dangerous to the company, like long term care, or extremely profitable (or dangerous, depending upon what you do with the expected lapse yield), like return of premium riders. Both of those products are discussed on the lapse support page.

Notes

Discussion

To assess the soundness of the company's products, a good first step is to list all of them by sales volume so you do the important ones first, and then determine whether the products seem to make sense to you. Then see what else you have that says the products are profitable, and with what degree of certainty.

Obviously if you have credible experience on a product, that is the best evidence. But most of the time you don't, and have to start with the actuary and/or actuarial studies. If there is no formal study, you need to explore the assumptions behind the actuary's opinion. The company may be reporting profit on the line, but the emergence of profit is also based upon assumptions. So we are back to the credibility of the assumptions. Remember that the assumptions in the product design are not necessarily the same as the statutory and the GAAP assumptions used for the emergence of earnings.

Read the actuary's individual product files, with particular attention to memos written to management, and correspondence with the insurance department. The regulators occasionally ask some good questions during the filing process.They tend to be sensitive to reserving questions which can be disastrous if left to pop up later.

The essence of a product is in the assumptions, not the design. All products come with a workup that shows a profit, or they would not be introduced. If the actuary has doubts about the assumptions, he will cover himself with a memo to the file stating that the assumption was dictated by others, usually the marketing department. An outside actuary will state the assumptions in the design report in the section that everyone skips to get to the profit part. If those assumptions do not reflect reality, the rest of the opinion is useless. The comments on the filing from the insurance department and the response of the company can also provide a tip that further investigation is required.

On top of everything else, talk to the company actuary about the products being sold, and ones that have been discontinued but still have substantial blocks in force. You can tell immediately whether you are getting anywhere. I have never had an actuary intentionally mislead me, but I have talked to plenty that only answered a question if it was asked. If you have looked at the files first, and the assumptions, you should be able to elicit whatever concerns the actuary has about the product.

This is particularly important in the due diligence phase of an acquisition. Don't rely on the consulting actuary you brought along. I have never had a product warning from one. Partly this is because the consulting actuary is expected to opine on a lot of matters in just several days. I also suspect that the habit of relying on detailed analysis (to profit test a single product can take several weeks and $20-$30 thousand, and they still want you to give them all the assumptions) somehow dulls their gut instinct. A surprising number of companies have sold those "whole life at half the cost" products that only an actuary (and the marketing department, of course) could believe in. Along those lines, when you talk to the marketing folks, try to get detail on who is selling the product for them, and how. Don't waste you time discussing the product. Marketing people still believe in the tooth fairy, and have forgotten the part about giving up a tooth.

The most widely applied profit test on a product is the "all the competition is doing it" test. That is dangerous, even if you can determine exactly what the competition is doing, which you usually can't. There is a tendency to assume the "other companies" are smarter than you are. Since there are about 1800 life insurance companies, it seems like it would be more logical to assume that mistakes, ignorance, and desperation would turn up fairly often. Copying someone else is no safe harbor. To the marketing department, anyone they can find that is selling the product that the agents are asking for, automatically becomes "all the competition".

No other company's situation is identical to yours. Market, persistency, and channel may all be different. Any competitor can be wrong, or may have copied some other company simply because "all the competition is doing it". And the competitor may be losing money and not know it. Example: a company exclusively in the payroll market may be able to sell cancer insurance at one premium for all ages and obtain an acceptable average age. However, a company in the brokerage business that tries that will end up with predominately older ages. Since the incidence of cancer increases with age, the later will have inferior margins.

Perhaps less obvious is the fact that the market you are in will influence the premium you have to charge. It is common for a company to move into a new market, having very different characteristics, and assume that its existing products that are working in the existing market will work in the new one. The problem in the new market is usually lower persistency, but it can be higher mortality. One company was advised by its testing vendor that it was getting positives for drug and alcohol abuse at 2 1/2 times the frequency of other companies. Another company moving into that market would not know that the old pricing and underwriting is not going to work.

What profit margin has the actuary been instructed to design into your new products? You should have defined profit percentages in several profit measures, since no one measure gives a full picture. In setting the percentages, the most important factor is your competitive position, and that is determined by the nature of your distribution system and your niche.

Successful companies don't worry about the "competition", because is the life insurance business there really isn't any. Almost every customer needs more insurance than they have. As Pogo said, "We have met the enemy and he is us." If you have developed a genuine market niche, it means that your agents (or other distribution methods) are only writing that niche business with you. If that is the case, you do not need to minimize the profit margins. To be in a position to realize a full profit on business is the purpose of a niche. Since there is no such thing as a free niche, you are also incurring additional costs to support the niche you have. Those costs have to be included in the premium.

The foregoing doesn't ring true with companies that are trying to do business through brokers and have no competitive advantage. Fortunately nearly everyone has some niche or advantage that can be turned into one. So the secret is to concentrate on it, and not someone else's "market". And, obviously, not to try to be in every "market".

What do you do if your company does not have a niche? If you have a big enough block of business in force, you can get along pretty well selling to your existing policyholders and agents. If you focus on service and continuing contact so you are there when they buy again. You can do this with or without agents, depending upon which "owns" the customer, that is, whether the primary contact is directly with you, or through an agent.

A big in force block will also throw off good statutory profits even if the policies were not particularly profitable when they were originally sold. The acquisition cost and any initial poor persistency is all sunk cost. What this does to you GAAP earnings of course depends upon how the deferred acquisition cost was handled.

What do you do if you don't have a niche and don't have a big in force block to either sell around or run off? Did you ever notice that the founders of successful life companies, if living, are always very old? If you are very very good, and still in your 30s, go ahead and start building a life company from scratch. But forget the internet. Anything that works quick and easy will be gobbled up by the existing companies. Unless you find a new niche. The internet is not a niche. No barriers to entry. Neither is a product. Same reason.

Return on investment as a measure of the future profitability of business can be deceptive, since the result is strongly influenced by the timing of reserves. Other measures, such as profit as a percent of premium, need to be examined. A spread sheet of expected cash flows can be more revealing. Clearly it is better to receive cash earlier rather than later, but present value concepts really don't apply to accounting entries.

At issue a reserve is zero, and it is zero after the policy terminates. In between these points the reserve rises, and then falls, all with no effect on cash flow. However, the ROI calculation includes as a cost the present value of the reserve just as if it was a cash expenditure. If the reserve is put up early in the policy duration, ROI will be lower than if the reserve is put up later. The timing of the reserve creation will affect the incidence of statutory earnings, and thus the ability to pay dividends, but will have no effect on the timing of cash flows, which really ought to be the primary concern when thinking about "profit". Put another way, if you defer a reserve, can you invest it for a return?

Specialty A&H products, such as cancer and heart/stroke coverage have cash loss ratios that increase with age of the policyholder, and thus the duration of the business. The mix of duration, i.e. the ratio of new business to in force, adjusted for persistency, determines the acceptable current cash loss ratio.

The actuarial studies should give an approximation of the ultimate loss ratio, but the concrete evidence is the current cash loss ratio. New business has a low ratio compared to the ultimate expected, since the policyholder is as young as he is ever going to be when the policy is first issued. If most of the block is in the early durations and the loss ratio is more than 20 or 30%, the ultimate could be 3 times that. On a closed block, the loss ratio will increase over time, but only by the shape of the morbidity curve. Rate increases can probably be made later, but can have the effect of losing more of the younger policyholders than the old.

The cancer morbidity curve increases with age almost as rapidly as the mortality curve. And yet if cancer rates are age graded, it is usually only to a few intervals, and not year by year as is life insurance. The reasons usually given are agent and payroll clerk convenience in the payroll selling situation where most cancer insurance is sold. This probably needs to be rethought, since the same agent sells life in payroll settings without that "convenience", and payroll clerks now use computers.

There are a number of advantages in grading cancer premiums yearly. There is enough difference in morbidity to show an annual increase, and it makes you cheaper in the early portions of the competitors larger intervals. You then have at least a tendency to get the younger ages, and it makes you harder to replace later.

The morbidity curve should also make you wonder what will happen to earnings if you stop selling cancer coverage, or even if your rate of growth slows down. You know your cash loss ratio will steadily increase, but your loss ratio for earning purposes includes an active life reserve which ideally would hold the loss ratio steady as the block ages. That is the theory anyway, and so that is a good thing to discuss with your actuary. That reserve is not a bad place to be conservative, particularly if your cancer policies start getting replaced with critical illness policies, which some carriers have under priced so severely that the premium is about the same as for a cancer only policy, notwithstanding that CI will have twice the benefit payout of the same size cancer plan (given that 50% of the benefits under CI are paid on cancer claims).

There was a period of popularity for limited pay cancer plans, mostly 20 pay, and some are still being sold. An interesting question: how do you raise the premiums on paid up policies if you get into trouble? You are not going to get much lapse on those paid up policies either.

The limited pay cancer plan actually is a worse problem with regard to rate increases than the obvious paid up problem. Suppose it takes you 10 years to discover a rate problem. The extra premium of a limited pay plan, the part that goes to the reserve to cover claims after the policy is paid up, is spread over the premium paying period, originally 20 years. But now you only have 10 years to go on the stuff sold 10 years ago. You are not going to get the increases needed to make up for that approved. Your current cash loss ratio doesn't look any worse (to the regulator who you need to improve your increase) than the ratio on continuous pay plans. Apparently one company came up with a solution: the product is 20 pay all right, but also coverage terminates after 20 years. It is listed under "Other Advantages". Unless I am misreading that, that could cause a problem down the road.

Critical Illness

Critical Illness -- See separate page

Lapse Support

Lapse Support -- See separate page for Return of Premium, Long term care.

Universal Life

Universal Life -- See separate page

It can be very helpful to calculate the effective level commission rate on an A&H plan. If you know this you can add it to your other costs and see if you are over 100% of the premium. Most commission structures are not level, having a first year commission that is substantially higher than the renewals. What level commission this would be equivalent to depends upon the persistency of the business. The effective level commission is one that has the same present value as the stream you are actually paying.

The common 70% first year and 15% renewal on cancer plans can produce an effective level commission of 35% at fairly common persistency. If the required loss ratio is 55%, you are already at 90% of the premium and you haven't covered premium taxes, expenses and profit(?) yet.

Many A&H products have a state required loss ratio between 50% and 65%, and higher for "group" products. It is necessary to convince the regulators that the ultimate loss ratio, the average loss ratio over the life of the product, will meet the requirement. Some states are requiring premium refunds when experience is below that required. That makes the percentage very real. If management is projecting ratios to the regulators that are higher than the ratios management is depending on to make a profit, there is a disconnect that will become a problem.

The lack of Non-Smoker rates can lead to selection situation for the company. If you have a life product (often the lowest face class) that has a blended rate, the same for smokers as nonsmokers, and the next band that has separate smoker and nonsmoker rates, either the face or the premium is going to overlap. You will get a higher portion of smokers than you premiumed for in the blended rate product. The smoker rates are so much higher that most smokers will take the combined rate, and the non smokers the non smoker rate the next band up.

Even if you have no product with non-smoker rates, or have isolated the ones you have so there can be no overlap, you can still have adverse selection created by the availability of non-smoker rates from other companies. Since smoker mortality is about 2 1/2 time non-smoker, the selection goes beyond the time of issue. The non-smoker is a target for replacement at a lower premium for a number of years. As in other respects, you also face selection if your underwriting requirements are weaker than those of other companies. The rate penalty for smokers is so severe, it is not uncommon for companies to require a nicotine test from applicants that say they are non-smokers. That in turn creates pressure from the agents for a blended rate in order to avoid asking the prospect to take the test. Some combination of the foregoing probably explains the continued availability of blended rates even if selection against the company makes them undesirable.

If you have a life product where the commission plus the surrender value of a product exceeds the premium by a large enough amount in the first year (or cumulative in any duration) you are almost certain to attract some large "producers" of fraudulent business. If after paying the first annual premium there is enough left to make it interesting, the insurance becomes "free" and pretty easy to sell.

How much is enough to attract this bad business? It is common for first year life commissions to exceed 100% by 10 or more points. My guess is that the danger point is around 140% plus (commission and surrender value), where the agent can pay the first year premium with his commission and have enough left over to cover taxes and transaction costs and still be worth while. Someone who specializes in this approach probably has to provide something for his cohort of free insurance takers. Usually the favorite product is whole life, since most universal life products have a the surrender charge for a number of years. However, the specialists may have a UL product they would like you to issue that compromises the surrender charge. And, of course, they like to have it issued "nets".

Modified premium life insurance generally has lower premiums in the early years, increasing later when the insured may better afford it. However, there are product designs that do the reverse, where the premium is higher in the first year and reduced thereafter. There the purpose is to increase the effective commission, as the first year commission is paid on the higher amount rather than the lower level premium in subsequent years. What is important is that management recognize the effect of this design on the value of its product and the distribution cost.

Since the purpose of this modification in the premium pattern is to increase agent compensation, you would assume the impact on sales cost would be obvious, but apparently it is not. I have heard it argued that there is no extra commission, and in any event the extra commission is "paid" by the insured, not the company. For example, take a $1000 first year premium that reduces to $600 in subsequent years. Usually the insured pays an annual $1000, all of which goes to insurance the first year. After that $600 goes to insurance and $400 to an accumulation vehicle. Since the first year commission is at least 100%, from the company's viewpoint what it receives is a level premium of $600 beginning the second year, exactly what it would had received if the first year insurance premium has been $600 instead of the $1000. Considering the income stream to the company, the effective first year commission is between 175% and 200%, depending on the renewal commission rate. This has the same effect on product value to the consumer, and on the company margin, as would paying that effective percentage commission on any non modified product.

Looked at from a different angle, if you are going to pay a first year commission of 200%, why conceal it with a modified premium? You might as well get the bang for the buck with the agent. My hunch is that most companies that do this really don't understand the arithmetic.

For pure term insurance to be profitable, it is a good bet that there has to be something like a protected niche or a controlled field force that protects it from the full blast of competition. Most term sold through brokers or over the internet has premiums which are lower than pure mortality cost, so the companies are "hoping for" something. So it comes down to figuring out what that is, and the likely hood of it happening. .

The problem with term insurance is that it is easy to understand and impossible to differentiate from the product offered by everybody else. That explains why it is the only product sold in significant volume over the internet. It is hard to see how these sales can be profitable, so the firms involved must be hoping for some other benefit. Internet sales are discussed in the eBusiness section. The internet is the economist's dream and the businessman's nightmare, the perfect market. The annual premiums for 20 year term are listed with the cheapest at the top. So who, exactly, is going to choose the second one on the list, and pay more for the exact same product? And what is the company at the top of the list hoping for? We don't know what it is, but we do know what it isn't.

Term insurance offered over the internet or through highly competitive markets now is quoted at standard, preferred and super preferred rates. Super preferred rates are about half standard rates. It is common for web sites to state the qualifications for preferred and super preferred rates. It appears that competitive pressure has been expanding the super preferred category over the few years. For example, the untreated cholesterol standard has moved up at least 20 points, and blood pressure to 140/90.

If the super preferred requirements are getting looser, a higher percentage of the population will qualify, and the rates will have to increase or issuer profits go down. It will not be long before we see super super preferred rates, since the super category will be broad enough for someone to subdivide again and offer a lower rate than can be justified for the category as a whole. The pressures will almost certainly lead to unprofitable business.

Subdividing risk categories can trap an unwary company. If you have a standard rate and decide to divide it into categories, such as standard and preferred, you will have to increase your standard rate or your margins will be less. Removing the best risks to a preferred category leaves the standard category with higher average mortality.

Generally companies estimate that 30% of the applicants will qualify for super select term rates. This is a broad range that makes sense when dealing with a field force, but it creates an opportunity to subdivide the category to offer lower rates on internet sales. This may already be happening without being visible. If a company is selling only over the internet, it is easy to make the qualifications stricter, and the premium lower, without ever saying you are creating another class.

When you offer the same product through agents that you do over the internet, you have to set your super select qualifications so that the agent does not feel the rates are illusory, that "no one can qualify". Internet sales are different. If you have the lowest rate and only one out of a thousand can qualify, you will still get plenty of business because the universe is so big. So it is logical to expect more super super qualifications, e.g. cholesterol at 160 (instead of 220) and BP at 110/70 (instead of 140/90). This should fly with the agents as long as they are told up front these are the "one in a thousand" rates and not to present them up front to their clients. The rare agent client that qualifies will get the lower rate upon issue. It may be that super select rates are already so low as to be unprofitable, but that is not likely to prevent further subdivision and even lower rates.

Juvenile life insurance. In late 1999 the Washington Insurance Department conducted an informative survey of life companies writing in the state, and included in the report a detailed analysis of the policy questions involved.