Is the Veralytic Report a Useful Tool for Advisors in Evaluating Life Insurance?

Posted at on May 30, 2012

Veralytic is the trademark of, Inc., a Florida-based firm founded in 1999 by Barry D. Flagg. The trademark application describes the firm as "providing information in the area of pricing, and performance, and suitability data on life insurance policies and products through a global computer network."

Veralytic's main product is the Veralytic Report, described on their website as "the only patented, objective and transparent evaluation of suitability for life insurance policies." It is protected by two U.S. patents: #6,456,979 (Method of Evaluating a Permanent Life Insurance Policy, issued 9/24/2002) and #7,698,158 (Life Insurance Policy Evaluation Method, issued 4/13/2010).

Michael Kitces recently wrote a favorable review ("Will Veralytic Reform the Life Insurance Industry?") on his blog at

I have occasionally ordered Veralytic Reports (formerly Confidential Policy Evaluator Research Reports) during the past few years. I encourage advisors to become familiar with this tool by ordering a report for each of the five main types of cash value life insurance: traditional whole life, traditional (non-indexed) universal life, guaranteed universal life, indexed universal life and variable universal life. You can take advantage of discounts through Michael Kitces's blog and Veralytic's e-mail list to reduce the cost.

Here are a few questions and comments to help advisors form their own opinions about the report's usefulness.

1. Should you take the Veralytic Appropriateness Rating seriously?

Veralytic approaches life insurance policy analysis as a multi-criteria decision-making problem. Its five selected criteria are Financial Strength & Claims-Paying Ability, Cost Competitiveness, Pricing Stability, Relative Policy Value and Historical Performance.

Its weighting method is simple. Each criterion is quantitatively measured, and the result is assigned a value of zero, one-half or one star. The five results are summed, leading to a possible Appropriateness Rating of zero to five stars.

For example, financial strength is measured by the Comdex percentile ranking, a ratings composite calculated by Ebix, Inc. (My critique of the Comdex is at A Comdex of at least 91 gets one star; 76 to 90 gets one-half star; and less than 76 gets no star.

Pity the company whose 90 Comdex causes it to lose half a star and whose 3.5 Appropriateness Rating is then no match for another company's 4.0.

In addition to the Comdex, Veralytic shows the rating from each rating agency, as well as the watch list status, and that is a useful feature of the report.

Multi-criteria decision analysis (MCDA) is a serious field, with many competing methodologies. For a survey, see Constantin Zopounidis and Michael Doumpos, "Multi-criteria Decision Aid in Financial Decision Making: Methodologies and Literature Review," Journal of Multi-Criteria Decision Analysis, July-October 2002.

Compare Veralytic's approach with the methodology presented by Robert Puelz in "A Process for Selecting a Life Insurance Contract," Journal of Risk and Insurance, March 1991. Puelz's selected criteria — for exposition only — were net payment index, insurer's financial strength, contractual features and 20-year cash value.

Instead of simple additive weighting, Puelz used the analytic hierarchy process, an MCDA method that is based on linear algebra. He also did not try to impose his preferences on everyone: "The complexity of the decision requires a decision model which considers an individual's unique circumstances, objectives, and constraints."

To be fair, Veralytic does not make extravagant claims; each report contains this disclaimer on the first page: "Furthermore, while Veralytic Star Ratings are produced using generally accepted mathematical algorithms and a consistent and objective rules set developed by Veralytic, this rating system, like all ratings systems, relies on certain judgmental techniques, which are fully described in the attached User Guide, and with which certain insurance professionals may disagree."

At bottom, Veralytic is just saying that it has an opinion about life insurance policies, and anyone can have an opinion.

Veralytic shows the components of the Appropriateness Rating, so advisors can easily produce their own opinions by changing the weights. For example, you could ignore the low Relative Policy Value rating of a guaranteed universal life policy with no cash values if there will be other policies with high cash values in the portfolio, or you could apply a different MCDA technique to the components.

2. What is the relationship between the Veralytic Appropriateness Ratings and future policy performance or future policyholder satisfaction or future anything?

To my knowledge, there is no published research on the predictive value of the Veralytic methodology. This would be a good topic for independent researchers in the academic and actuarial communities. Ideally, the Veralytic Appropriateness Rating should be benchmarked against a cheaper or easier alternative, such as flipping a coin.

3. Are advisors well served by a dumbed-down formula for life insurance pricing?

The formula that Veralytic promotes is: Premiums/Performance = Cost-of-Insurance Charges + Policy Expenses – Policy Interest/Earnings.

This is okay as far as it goes, but it doesn't go very far. Veralytic's User Guide says that "insurance pricing is simple when reduced to its three fundamental components." It would be more accurate to say that insurance pricing is simple when the complexities are ignored.

Mr. Flagg's patent #6,456,979 gets it right: "The permanent life insurance product is the most complex financial instrument that is purchased or owned by the general market of consumers, combining the benefits and costs otherwise found in insurance and investments into a single financial instrument."

The price of nonguaranteed life insurance policies is determined by the decisions of the insurance company's current and future management and board of directors. Those decisions are informed by the pricing work done by the company's actuaries, and that work is documented in profit tests, model office projections and actuarial memoranda.

Profit tests are the crown jewels of life insurance due diligence. They take account of all of the significant pricing factors: distribution, underwriting, administrative and product development costs; state premium tax and Federal DAC tax; mortality rates and reinsurance costs; lapse rates; investment earnings; statutory, GAAP and tax reserves; and target surplus. For each pricing cell (face amount, issue age, underwriting class, premium schedule), they demonstrate that the product will satisfy one or more profit objectives, such as profit margin, return on investment, present value of after-tax distributable earnings or embedded value.

If you want a comprehensive list of the risk factors that can affect future policy performance, understanding profit tests is the place to start. This should be part of financial planner education, but unfortunately it isn't.

As far as I know, Veralytic has no access to the internal pricing documents of life insurance companies. The users of Veralytic Reports are betting, explicitly or implicitly, that performance risk factors can be adequately assessed without intimate knowledge of the product pricing process.

Knowing what I know, I would not make that bet. So there are only two possibilities: people who rely on Veralytic Reports know more than I do, or they know less.

4. Can one set of analytical tools apply to all cash value policies?

Some measures, such as financial strength and liquidity, can reasonably be applied to all types of policies, although the criterion weight might vary. However, Veralytic's simple vision of life insurance pricing misses important features.

For guaranteed universal life with a shadow account design, you can't understand the policy without understanding the calculation of the shadow account value; this information is contained only in the contract, which Veralytic does not review.

For indexed universal life, you need to understand the hedging budget that is available to produce the credited interest rate. Veralytic does not examine that.

Combination products (life insurance with long-term care benefits) require an analysis of both the life insurance and long-term care insurance pieces of the policy.

Veralytic seems to want the world to conform with its simple vision, rather than adapting its methods to the world.

5. Is Veralytic's Pricing Style Box useful?

Veralytic has created a two-dimensional matrix that imitates Morningstar's style boxes. A policy is classified by target market (Retail, Institutional, Experience-Rated) and "optimal funding strategy" (Minimum Premium/Defined-Death-Benefit, Maximum Accumulation/Defined-Contribution, Mixed Funding Strategy).

The target market dimension may leave advisors puzzled. I recently ordered a Veralytic Report for an indexed universal life policy that most people would consider retail, and it was classified as Institutional because of a $1 million face amount. And since Institutional and Experience-Rated are described in the User Guide as having better features than Retail, your retail clients may wonder if you have really done your job when you propose a Retail policy.

Life insurance is complicated enough without creating arbitrary distinctions.

The funding strategy dimension attempts to offer guidance on the best use of the policy given the load structure. It falls far short of guidance that could be called "optimal," however, because it does not highlight year-to-year changes in policy loads that can create a strong incentive to pay non-level premiums.

6. Are Veralytic's benchmarks useful?

The Veralytic Report compares cost-of-insurance charges, policy expenses, premium loads, premiums and projected cash values for the evaluated policy versus two sets of benchmarks: target market and all policies. The benchmarks are based on actuarially determined factors or Veralytic's growing database of evaluated policies, or both.

In the 1980s, the Tillinghast Universal Life Analytic Study (TULAS) published the percentile distributions (10th, 25th, 50th, 75th, 90th) of policy loads, cost-of-insurance rates, interest rates and cash values at several durations. Veralytic publishes only one value for each benchmarked item, so advisors cannot see the range.

For dial-a-commission products (such as whole life with term and paid-up additions riders), Veralytic does not use a lower-commission version of the policy as an additional benchmark against the evaluated policy, although this seems relevant to appropriateness.

Veralytic's benchmarks affect the Cost Competitiveness, Pricing Stability and Relative Policy Value ratings, so the construction of the benchmarks is important. There are at least two issues here.

First, are the benchmarks sufficiently calibrated to the policy under evaluation? In "The Dynamic Threshold" (Product Development News, July 1992), William H. Drinkwater and Jerrold M. Norman showed that product profitability varies greatly across pricing cells and that you need a complicated load structure to solve this. The actuarially determined factors and the relationship between Veralytic's model and asset-share models also merit investigation. Do Veralytic's benchmarks produce false positives or false negatives because they are too crude?

Second, can Veralytic's Appropriateness Rating be gamed by designing a policy that scores better versus the benchmarks without actually providing better consumer value? For an analysis of manipulability, see Ralph A. Winter, "On the Choice of an Index for Disclosure in the Life Insurance Market: An Axiomatic Approach," Journal of Risk and Insurance, December 1982.

7. What is the relationship between companywide statutory accounting data and product-specific pricing factors?

Veralytic apparently believes that there is a close relationship. The Historical Performance rating for general-account policies (whole life, universal life) relies on net portfolio yields taken from the statutory statements of the rated company and all other insurers.

I'm not aware of published research that supports the premise of a close relationship, but I'm aware of research that disputes it. This is from Joseph J. Buff, "Measuring Insurers' Investment Performance," Contingencies (published by the American Academy of Actuaries), January/February 1995: "We spent considerable time perusing publicly available data, trying to develop comparable, meaningful performance information. Annual statements, annual reports, rating agency reports, and research reports were all reviewed. But they weren't of much use... we had to conclude that statutory data really aren't useful for measuring investment performance."

8. What is the effect of a Veralytic Report on litigation?

I suspect that most users of Veralytic Reports never read the full report; they just glance at the Executive Summary, especially the Veralytic Appropriateness Rating, and put the report in a file, thinking that they have purchased cheap insurance against lawsuits. To my knowledge, Veralytic Reports have never been subjected to the scrutiny of a lawsuit, so their prophylactic value remains untested. Does the presence of a Veralytic Report deter litigation? Given litigation, does a Veralytic Report improve the defendant's position?

Even water-soluble raingear looks effective on a sunny day. I don't see why a smart plaintiff's attorney should be deterred by a Veralytic Report. In fact, reliance on a Veralytic Report might be a fruitful line of inquiry in some cases.

9. If not the Veralytic Report, then what?

The third-party tools that I use most often in my practice are Compulife (term insurance), Full Disclosure (cash value life insurance), Morningstar (variable universal life) and LifeSpecs (indexed universal life). These services give me time-saving information about important product features.

The opacity of nonguaranteed life insurance policies is not unique to the U.S. The British have struggled with this problem for years, as shown by the title of a paper published in the British Actuarial Journal in 2001: "Transparent With-Profits — Freedom With Publicity."

State insurance regulators could improve the life insurance marketplace in several ways:

  • Require companies to disclose all elements of the calculation of illustrated values for each filed product (including guaranteed universal life). For universal and variable universal life, that would include premium, per-thousand and per-policy charges, cost-of-insurance rates, interest rates, and the formulas that produce the illustrated values. None of this information is proprietary; it can all be determined, much less efficiently, by reading the contract and the annual statement and reverse-engineering the illustrations. Traditional whole life is harder to deal with, although Joseph M. Belth, editor of The Insurance Forum, has proposed disclosure of dividend formulas.
  • Require companies to disclose all repricing actions for all existing policies. This should be viewed as a price of admission to the marketplace. Why should consumers and their advisors have to try to piece together a track record of policyholder treatment from limited data?
  • Greatly expand the supplemental interrogatories to Exhibit 5 of the statutory annual statement. The U.K.'s Principles and Practices of Financial Management (PPFM) might serve as a model (for more information, go to the Financial Services Authority's website at A 2009 study found that even the consumer-friendly version of PPFMs was generally not used by policyholders or financial advisors, so there is still room for improvement.

Final thoughts

There is a real need for credible, comprehensive information about cash value life insurance policies. Veralytic's visibility has been built on relentless marketing, but at some point marketing needs to be accompanied by validation.

I don't know if the Veralytic Report is a step forward or a step backward. It is a step forward if it stimulates research on how to use publicly available information to predict future policy performance or puts pressure on insurance companies and state insurance regulators to make more information public. It is a step backward if it makes people think that the problem of comparing cash value life insurance policies has been solved when it certainly has not.

September 22, 2012 update

Regarding the relationship between statutory accounting information and policy performance, here’s a research paper that I overlooked: James M. Carson and Randy E. Dumm, “Insurance Company-Level Determinants of Life Insurance Product Performance,” Journal of Insurance Regulation, Vol. 18, No. 2 (Winter 1999). A related article also appeared in the September 2000 issue of the Journal of Financial Service Professionals.

The authors looked at data compiled by A.M. Best for 73 universal life policies issued in 1985 to 45-year-old male nonsmokers. The face amount was $100,000, and the annual premium was $1,500. They examined the relationship between actual policy performance, measured by 10-year cash surrender value, and selected company-level information, mostly based on insurers’ statutory annual statements. Their regression analysis showed that three company-level data items were significantly related to policy performance: lapse rate (statistically significant at the 0.01 level) and general expenses and investment yield (significant at the 0.10 level). Variables that showed no significant relationship included company size, organizational form (mutual vs. stock), A.M. Best’s financial strength rating, net gain as a percentage of total income, and change in product mix.

It makes sense that lapse rates should have the strongest relationship; that data item is for life insurance only, whereas other items are aggregates for all lines of business.