March, 2009 Newsletter
Provided by Leimberg Information Services
See other issues.
Longevity Curve Report - Life Expectancy Analysis - What Every Estate Planner Absolutely Positively
Better and more complete information leads to better decisions - – in some cases much better decisions! By knowing more, you can help more. A new tool exists which may help. Certainly, planners need to know about what some call a life expectancy analysis and what I call a longevity curve report (LCR).
This will be Part I of a two part commentary. (A version of this appeared as "Life Expectancy Analysis: Planning Tool of the Future is Here Now," Estate Planning, September 2008, Vol. 35, No. 9. My thanks to Al Barnes for his insight and assistance.)
PART 1 – WHAT IS A LONGEVITY CURVE REPORT AND WHY DO ESTATE PLANNERS NEED TO KNOW ABOUT IT?
THE BUCK STARTS HERE:
A client's longevity, that is, life expectancy, is the starting place for almost every key assumption that must be made in shaping a financial plan. But until recently, that number, the linchpin in every major element of a financial plan, has been little more than a guesstimate with respect to a specific client. "One hundred and five and still alive" has been the blind motto of too many planners, a "one size fits all" approach necessitated by the absence of a more precise or realistic method of arriving at a reasonable estimation of a number reflective of a specific person's likely life span.
Consider the potential misinformation mischief (and therefore serious planning mistakes) possible from even the best of financial planning software packages if the key assumption, i.e. when the client will die (how long is the client likely to live) is significantly incorrect?
Shockingly, there is no good reason to believe that the actual mortalities of our clients will in any way approximate the mortality assumptions we've used since all too often! When it comes to an individual client, the number has generally been based on a government actuarial table or worse yet, a wrong-handed rule of thumb that might not even remotely reflect the client's actual life expectancy.
· If you could determine, within a relatively narrow range of years, how long your client would live, would this not significantly improve the accuracy and therefore the credibility of your planning?
· Would you not be better able to choose more appropriate tools and techniques – and use them more effectively - if you could better approximate the range of how long your client would likely live?
· Suppose there was a way to obtain a customized longevity curve report that cost effectively evaluated the information for a specific client – say Mr. X – age 64.7 years – and could assess a median life expectancy[i] at age 78.4, i.e. a life expectancy of 13.8 years. Could you not create a much better plan if you knew that your client's life expectancy was – say 31% shorter than the median life expectancy for an "average individual" of your client's age and gender?
· Wouldn't the ability to better ascertain a specific client's expected longevity enhance your ability to demonstrate (and if necessary defend) the needs analysis you have performed and the suitability of your recommendations?
· Would third party expert input as to specific probable longevity help fulfill your duty of care and add objectivity, credibility and assurance to your recommendations?
· If your client knew his own life expectancy, would he better understand the urgency and significance of action and preparation if he or she knew his or her specific longevity curve?
· If a client could see and understand his/her longevity odds, could they make better decisions with regard to the choice of financial tools and techniques?
· What if the report also highlighted the medical factors that could have a positive or negative impact on his/her longevity curve, i.e., what positive change in behavior[ii] could help to extend his or her life expectancy (and perhaps enhance life quality); would that information help the client make better life choices?
· Would positive answers to these questions mean a planner could significantly strengthen the relationship with his/her clients?
All these things are not only currently possible; they are now available to professionals and their clients on a cost efficient and time sensitive basis. The short-hand term being used for the process is Longevity Curve Report (LCR) or more commonly Life Expectancy Analysis (L.E.).[iii]
WHAT A LONGEVITY CURVE REPORT IS:
An LCR or L.E. is a scientific assessment of expected life span based on the medical and family history as well as personal lifestyle and social habits of a specific client. It is a combination of (1) medical information and (2) actuarial and (3) statistical analysis designed to assess the relevance and significance of risk associated with various impairments and translate that into meaningful longevity numbers. Life expectancies[iv], typically done on clients aged 60 or older[v] are generated by assessing these three major factors and correlating the findings with actuarially-generated mortality tables.
The life expectancy analysis process involves a sensitive weighing of dozens (and in more sophisticated analysis hundreds) of data points. These data points are translated into debits and credits which then are used to ascertain how a specific client's life expectancy is likely to play out in terms of a median (e.g. years and months) and how it will differ – if at all – from other similar age and gender lives.
Typically performed by professionals with experience in life and heath underwriting, the L.E. has been used in the past mainly for the life settlement community to help investors represented by life settlement companies determine potential payback/profit likelihood scenarios. But today, LCRs are poised to become a unique and invaluable tool for Estate Planners in a multiplicity of situations.
The LCR report, while varying in name and format from company to company,[vi] typically provides information[vii] that depends in level of detail upon the price paid for the report and the mortality based transactional purpose for which it is to be used. For financial planning purposes, it will typically be sufficient to obtain a mortality rating which compares the individual's expected longevity to a cohort[viii] of similar persons and provides an expectancy estimate and gives a percentage (e.g. 85%) curve and spread of probability of dying sooner or living longer than that given number of years.
HOW DOES AN LCR WORK?
The LCR report, while varying in format from company to company in general will work like this:
· As a preliminary, information about a client's functional status and medical, family, and personal history must be gathered.[ix] That information can be gathered by the life expectancy company itself[x] or by a comprehensive phone interview with the client, and/or by obtaining what are called APSs (Attending Physician Statements or APQs (Attending Physician Questionnaires) from doctors, a clinic, or hospital.[xi] (The more work in gathering data the LCR company must do, the more expensive the report typically is.)
· Data is immediately verified with respect to social security number and date of birth.
· A medical "abstractor"[xii] pre-screens, reviews and highlights the medical record using proprietary underwriting manual and forwards findings and the records on to an underwriter. The abstractor confirms both the currency of the data and that the records are for the correct person. Pertinent records, diagnostics test results, and clinical records are highlighted based on underwriting manual guidelines.
· The abstractor supplies the underwriter with a worksheet containing the "debits and credits" culled from the records as well as comments.
· The underwriter will then focus on those problems most likely to affect survival. Among the facts typically examined are the individual's gender and build (i.e. height and weight), family medical history (father, mother, siblings, and children), "social habits" (tobacco, alcohol, drugs, and level of exercise and social activity and travel), cardio, cerebro, and peripheral vascular issues, pulmonary problems, renal and genitourinary, gastrointestinal, hematological, cancer, neurological/psychiatric, orthopedic/rheumatologic, autoimmune, disorders of thinking and emotion, and others. Emphasis is placed on what is likely to cause the individual to live a shorter life than a longer one.
· What occurs next will vary from company to company. At 21st Services, for instance, ( there are 5 major providers: 21st, Fasano, AVS, EMSI and ISC Services – see below)[xiii], unlike some others in the field, the underwriting process involves a proprietary "rules-based"[xiv] underwriting system. The underwriter fills out a "lead sheet", a nine page work paper based on his or her findings in the medical records. After the review in the next step, the items noted on that lead sheet are entered into a computer-based debit/credit underwriting model.
· The records and worksheet are re-reviewed by a team leader or other senior underwriter (or if the work was done by a senior level underwriter it is signed off by that person).[xv] Regular and random audits are also performed at all levels.
· Debits and credits are at this point entered into the software program which will generate a final result.
· Ratings are assigned. Generally 100% mortality is a standard rating. "Standard" means that the individual has no significant problems and that he or she will likely survive to the age which most persons of the same gender, age and smoking status will attain. A lower percent mortality means the person is likely to survive a longer period. A higher percent mortality means the person is likely to survive a shorter time.[xvi]
· One L.E. company, to ensure accuracy, objectivity and consistency, uses a proprietary diagnostic system capable of analyzing and quantifying more than 200 data points. The debits and credits are applied against the mortality tables to produce a mortality multiplier and mortality curve. The interplay between medical conditions[xvii] and how various debits and credits drive the mortality multiplier may be built into system logic along with features such as the erosion of debits and credits[xviii] over time ("age-based debiting"), the impact of time elapsed since onset of a condition ( "elapsed time debiting"), mortality credits, co-morbidity logic, and debits and credits impacted by functionality.
· A longevity curve report is then issued which will show mortality as a median life expectancy and in some cases also note the percentile of probability of those dying before and after that median. The report will also verify the subject of the LCR, list the key factors that comprised the basis for the report, state the median life expectancy, and in some cases provide a longevity curve reaching out until all the persons in the client's cohort are presumed to have died.
· The LCR company will then scan and electronically archive the complete file for permanent storage.
Behind the scenes of this carefully managed and cross-checked process are in at least one company is a Medical Advisory Board, comprised of nationally and internationally-renowned physicians who are board certified in the disciplines of geriatrics, oncology, nephrology, cardiology, neurology, endocrinology, and infectious diseases. Some of the members of this board also have expertise in public health and epidemiology within their fields. This enables the company to draw on their knowledge of, and experience with, the cutting-edge research and treatments in their respective fields which can be essential in calculating expected longevity. These authorities are used to help the company better understand and incorporate – from over 200 possible impairments or gradations of impairments – the latest predictive relevance of each. Longevity Curve Report companies also have close working associations with actuarial consultants with a strong working knowledge of mortality statistics.
HOW AN LCR DIFFERS FROM MAKING ASSUMPTIONS FROM A LIFE EXPECTANCY TABLE:
With a standard table, the average life expectancy for the general male or female population is obtained. This, of course, is not customized to the individual client. Nor does the standard table provide the planner with the client's maximum life span. As a Estate Planner, it is essential to have both of these numbers to provide a solid financial plan. The LCR (Insert Chart A) provides this insight and more.
The Table below is based on four government mortality tables[xix] and on an actual longevity curve report performed by one of the leading report providing companies. It compares key estimations based on a general population with – in the column on the far right – an analysis of the life expectancy of a specific individual aged 64.7. (Note that the input for the four government life expectancy tables was age 65).
90 CM (1990)
Sec. 1.72 (1983)
Life Expectancy (Yrs)
Age at Expectancy
The table above illustrates first that the expectancy for a specific individual can (and often will) vary considerably from any estimation from a general population pool.[xxi] The government's annuity table, for instance, shows a person aged 65 can be expected to live another 21 years – to age 86. But based on the actual person's L.E. , he could only expect to live another 13.7 years – to age 78.4. Secondly, planners should note that for retirement planning, social security purposes, and many other needs, use of any of the government's tables, although convenient and readily accessible, would have yielded what is likely to be a very wrong place from which to start planning for this particular client.
What is especially valuable about the LCR is that, in addition to providing the median life expectancy, customized for the individual, it also contains that person's entire longevity curve. The client's longevity curve begins with the current date and projects out the client's maximum life expectancy. The curve also illustrates all points between these two extremes, indicating the probability of living to various ages.
For example, one government mortality table suggests a median life expectancy for a 65 year old male to be age 86. Alternatively, a customized LCR for this same individual gives you much more appropriate and comprehensive insight into his longevity picture. The LCR tells you that this person's median life expectancy is actually 78.4 and that his maximum projected life span is age 90. It also shows that he has a 60% probability of living to age 76, a 40% chance of surviving to age 80 and a 20% chance of reaching age 83. (Insert Chart B)
LONGEVITY CURVE ANALYSIS NOW PRACTICAL!
The state of the art[xxii] is just that, art. It is not certain. But it is customized, based on the latest and previously unavailable scientific data and on advanced statistical analysis. In short, it's a much better starting point to begin the assumptions underlying any financial planning product, tool, or technique based on or impacted by a client's longevity. LCR (L.E.) companies are now providing such reports in considerable detail to planners for under $100.[xxiii]
WHY NOT GET IT FOR FREE?
It is true that there are several "longevity calculators" that can be found online and quickly completed with rapidity. These calculators are easy to use, and in fact entertaining. But from a professional advisor's viewpoint, the essential question must be, "How accurate, reliable, and credible are these calculators in predicting a client's life expectancy?"
Try them and spreadsheet the results. You will find that the answers will be far from even approaching consistency of results. Life expectancies may vary by a decade or more! The reasons for such variability are obvious:[xxiv]
· the calculators do not use the same parameters for making the estimates.
· although they all include some lifestyle modalities, they don't all use the same ones.[xxv]
· it is not clear how various factors are weighted – but it's likely that they are not all weighted the same.
· Most importantly, universally lacking is an input for detailed medical history. The medical questions asked are superficial.
Furthermore, there is no "trapping" factor that filters out a self-serving or wish fulfilling answer. For instance, if a person subconsciously wanted to obtain a report which generates a longer life expectancy, it would be relatively easy to fool the calculator and enter responses that will generate a longer lifespan.[xxvi] (One calculator even generates "applause" when a "healthy" answer is given.) In fact, it's all too easy to do a "do-over", and keep doing-over until one gets what is felt to be a "desirable" longevity prediction. It's also easy to respond to a different calculator and go with the one that gives the most desirable result.
So there is a lack of both professional detailed analysis and rigorous objectivity. The result is worse than meaningless as a guide and certainly not defensible as a financial planning tool. Absent scientifically reviewed information about a specific calculator's accuracy, it can only be regarded by a professional as entertainment rather than as the basis for taking financial planning actions and implementing tools and techniques related to a person's specific longevity curve.
Lifestyle,[xxvii] diet, safety measures, wealth and income, and mental and physical activity level are all very important in predicting longevity.[xxviii] But without a thorough and professional evaluation and analysis of documented medical facts,[xxix] the single most important basis for a longevity determination is absent.[xxx]
HOPE THIS HELPS YOU HELP OTHERS MAKE A POSITIVE DIFFERENCE!
LISI Estate Planning Newsletter # 1426 (February 26, 2009) at http://www.leimbergservices.com Copyright 2009 Leimberg Information Services, Inc. (LISI). Reproduction in Any Form or Forwarding to ANY Person Prohibited – Without Express Permission.
[i] Median life expectancy is the point at which half of a group of individuals with a given client's same health profile would still be living. It is an average rather than a prediction. It is based on a pool of people of an actual client's age and gender.
[ii] The leading causes of death in the U.S. in descending order of occurrence are heart disease, cancer, stroke, accidents, and diabetes Mellitus. To some extent, changes in behavior, life style, and proper medication can impact on all of these. The quality of health care, level of physical and metal activity, nutrition, and exercise can in some instances vastly expand life expectancy. For instance, a person with hypertension or high blood pressure can increase life expectancy by increasing physical activity and making dietary changes such as eating more fruits, vegetables, low-fat diary products and by decreasing fats, red meats, sweets, and sodium – as well as by increasing the frequency of medical intervention and maintaining the proper medical treatment. People who have a family history of longevity, lead a vigorous lifestyle, and demonstrate excellent exercise tolerance generally tend to live longer than their peers. Some studies show that people of higher net worth in certain geographic locations who have better access to both health care and health care education are likely to live longer. The credibility and weight given to those studies is likely to vary from company to company (and perhaps even from underwriter to underwriter in some companies). Note that, to date, most L.E. companies have been doing analysis on mainly high net worth individuals. This of course, will now be "fine-tuned" as the demand for L.E. information increases beyond the traditional ages.
[iii] Some also call this life expectancy evaluation.
[iv] According to Paul A. Seigert, President and CEO, Insurance Studies Institute and co-author of the Tools and Techniques of Life Settlement Planning (National Underwriter Company), "Life expectancy"(LE) is an actuarial calculation and is best represented in formulaic form. An LE is a determination of the average future lifetime of someone currently at age x, and is typically denoted by the symbol ex. In formula form, life expectancy is:
where tpx is the probability of living from age x to age x+t, and includes calculations through the end of the assumed mortality table (age x=oo), which is some age greater than 100 for all recent tables. Another way of viewing the life expectancy is this: if 1,000 people were alive at age x, then roughly half of them would still be alive at their life expectancy, or age x + ex, or roughly half would also have died. (While this is a reasonable analogy — a 50/50 chance to live to one's life expectancy, the theoretical calculation does differ from this by a few months due to the actual shape of the mortality curve so that slightly more than 50 percent of a population will typically die before their life expectancy is reached. This is particularly noticeable for older ages and/or mortality assessed with impairment ratings. See "Mortality Considerations and Their Affect on Portfolio Valuations", March, 2008 by Ed Mohoric, FSA, MAAA and Robert O. Kinney, M.D., FLMI
[v] Some companies are going as low as age 40 in certain cases but most currently concentrate on the senior ages. It is likely, however, that L.E. companies will be going lower in age as planning professionals learn more about their services and demand for their products increases.
[vi] It is quite possible for different L.E. companies to report significantly different (months or even years) expectancies (longer or shorter) due to (a) missed or absent data, (b) differing interpretation of medical examination results, (c) differing determination of the severity of a person's illness, and (d) differences in the presumptions of the actuarial impacts of various medical problems, i.e. in the conversion of rated debits to months of life expectancy.
[vii] Although most (there are three major companies and a few smaller ones). L.E. companies focus on older (65 and up) individuals, a few will now do evaluations on individuals younger than 60.
[viii] As used here, a cohort is a group of subjects of the same age studied from a statistical/demographic characteristic viewpoint, i.e. their health and mortality.
[ix] Records will generally will preferably be current within one year from the date of review. Five years of medical records are generally obtained. The client's "initial intake" and history and physical performed when a person initially visits a new doctor is studied in detail as are social habits such as alcohol and tobacco use. The L.E. company will want more comprehensive and older information on "events" such as a heart attack or diagnosis of cancer.
[x] The process can start, depending on the life expectancy company, by fax, overnight carrier, mail, or over the internet. Note that not all life settlement companies will offer individual services to non institutional sources.
[xi] Medical Records Information Form and Authorizations for Disclosure of Protected Health Information (HIPPA) must be obtained.
[xii] An abstractor is a person with insurance underwriting experience or a person with medical expertise such as a fourth year medical student, nurse, physician's assistant, or in some cases medical doctors.
[xiii] There are others listed on the chart at the conclusion of this article.
[xiv] Rules based means that the underwriting system is designed to grade every medical disease in the same manner in order to achieve a high degree of objectivity and repeatability, i.e. the system will evaluate and rate everyone with a given impairment profile in a consistent manner. This should be compared with a "subjective overlay" approach where the underwriter's individual judgment will much more greatly affect the bottom line (but may differ widely from underwriter to underwriter and possibly from case to case). (I am not suggesting or making a value judgment on either approach.)
[xv] Some underwriters will have clearance to sign off on their own work in limited circumstances based on the size and complexity of the records. Otherwise, without clearance, underwriters will pass their work onto a Senior Underwriter, the Chief Underwriting Officer, Medical Director, or Underwriting Manager for final review.
[xvi] Dr. Barry Reed, co-author of Tools and Techniques of Life Settlement Planning (published recently by The National Underwriter Company).
[xvii] There is a concept known as "co-morbid" conditions (e.g. a person has a history of both coronary and diabetic conditions) which together have an impact on longevity that is likely to be greater than the sum of the two individually, i.e., the death rates are higher when there is a combination of these two things. In other words the person's risk profile "accumulates."
[xviii] According to Dr. Barry Reed in Tools and Techniques of Life Settlement Planning (National Underwriter Company 2008), "The final sum of the credits and debits is linked to the percent mortality which allows for calculating survival of different ages, genders and smoking status based on all medical problems identified. Persons who are sicker are assigned more debits. Credits are assigned for problems which have been minimized. Examples are well controlled uncomplicated diabetics or a person who has coronary artery disease but who has had coronary artery bypass surgery or an angioplasty." Some individuals, due to their typical relative affluence, better nutrition, high quality health care, and other positive characteristics, often live longer than previously actuarially projected.
[xx] The author wishes to acknowledge 21st Services (www.21stpartners.com) who generously provided me with a customized comprehensive evaluation and walked me through the entire process of the life expectancy process. 21st is one of several companies that provide life expectancy evaluations to investment institutions. Life expectancy estimates have been provided mainly by five major underwriter firms (The other well-known L.E. companies are AVS, Fasano, and EMSI, and ISC) but newer firms have recently entered the market.
[xxi] According to reports from the New York Times, The National Center for Health Statistics puts the average American life expectancy at 77.8 years. Note, however, that gender, race, health, genetics, lifestyle choices, and even socioeconomic status will have a great impact on the reality in a specific situation.
[xxii] A life expectancy estimate is just that, an estimate and can not and does not represent that a given individual will die on or near a projected date. It should be taken as more guidance than gospel. It merely establishes a client-specific date and range of probability of death. For instance, the report might show, based on the population of a group with a medical profile identical to the client, the likely life expectancy curve of the client as compared to the standard curve and show that the expectancy for 85% of the people with a profile identical to the specific client is between the ages of 69 and 87 and that the median life expectancy for the client is 78.4.
[xxiii] The most comprehensive analysis performed by 21st Services, for example, is currently $400. Its least expensive is in the under $100 range. Most companies' L.E.s run from $300 to $750.
[xxiv] Dr. Charlotte Lee, Medical Director for 21st Services.
[xxv] A few take into account one's engagement in activities such as teeth flossing while others consider such safety issues as using seat belts.
[xxvi] It is important that no documented medical history be ignored, regardless of whether it will lengthen or shorten a person's LE. The goal should not be either "long" or "short" but rather to be as accurate as possible given the relatively young state of the life expectancy art.
[xxvii] "Yes, I know there are sites where one can enter such information as how many beers you drink, whether you smoke, how fat you are, that you drive your car without the seat belt buckled…But I have a variety of medical conditions and can't seem to find any expectancy information." Reader comment quoted by columnist Scott Burns, Boston Globe, December 1, 2007
[xxviii] It's also important to ask how lifestyle factors impact on the ultimate life expectancy prediction or how valid they are in drawing conclusions about the long-term impact of these behaviors.
[xxix] " One on-line calculator never asked if the individual taking the "test" had been diagnosed with cancer or some other such serious life shortening factor. It's unlikely that any of these on-line models are based on any sound evidence or any model of multiple variable analysis that puts all the factors together with the proper weight given to each (and to combinations of issues). Nor is it probable that any are based on any substantial real data.
[xxx] However, online lifespan calculators do have some benefit for the client personally, that of creating more awareness of the importance of health and lifestyle. For example, someone who have never had a cholesterol check before might be prompted (scared into?) a medical check-up and analysis. Certainly, on-line calculators drive home the undisputed (except perhaps by tobacco companies) fact that smoking is a major factor in shortening life expectancy – if in no other way than by asking the question, "Do you smoke?" Likewise, on-line calculators that ask about seatbelt usage remind users of the importance of buckling up.
All NAEPC-affiliated estate planning councils are eligible to receive a discounted subscription rate to the Leimberg LISI service. Please see more information about the offering. You may also contact your local council office / board member to find out whether they are offering the service as a member benefit.