Beware of Dull Tools in Retirement Planning
By W. Andrew Larson, CPC, Retirement Learning Center
Retirement planning is complex, emotional and can be a time sucker. Deciding on the vision of what retirement will look like requires a commitment to time and honest reflection.
We as an industry have deployed a veritable host of tools to support individuals and couples in the retirement planning process
Have the plethora of tools created misconceptions in minds of consumers? Obviously, we need to encourage practical retirement planning. Maybe we have tried to make it too simple in an effort to get more people engaged in retirement planning. We have encouraged people to use tools. In fact, the entire “robo advisor” industry has sprung up advocating tool use. And the easier the tool the better–right?
Consumers seem to implicitly trust tools and their output. I have been surprised at the level of trust placed in retirement planning tools. There seems an almost magical quality about a friendly and colorful retirement planning widget. It’s so simple. Just enter the right data, select the right retirement age and rate of return and *POOF* we are on our way to a happy and blissful retirement. We are in awe of the fancy print outs, and colored graphs and charts. They look so good, they must be accurate.
I met a couple who, to reward themselves for being “on-track” in their retirement planning (according to their retirement planning widget), purchased an airplane!
Their trust in the tool may have been misplaced.
Maybe we need to take a step back and help consumers understand the limitations of virtually all retirement planning tools. Let’s discuss frankly an aspect, and limitation, of every retirement planning tool ever developed. At the core of every retirement planning tool are one or more life expectancy tables or algorithms. Enter your age and the software tells you how long you have to live—on average. Couples enter their information and we have their joint life expectancy! Let the planning begin.
Why should we be concerned about the accuracy of life expectancy tables? Are the tables really that inaccurate? The short answer is life expectancy tables are accurate and inaccurate; the accuracy is based on the timeframe and number of individuals involved. The use of longer time frames and more people to develop the tables equates to greater accuracy. Fewer people and/or shorter time frames mean less accuracy. Actuarial tables were never intended to forecast individual outcomes. Let’s explore this seeming paradox.
Life insurance companies engage actuaries to develop life expectancy tables in order to set premiums for insurance and annuity products. For this purpose, life expectancy tables are well suited. The tables tell you, on average, how long a group of people are expected to live. Obviously, some will live longer than the average and others will not. Insurance companies use the tables to price their products, realizing that, in the long term, the numbers will end up close to the average.
But, the concern is using a tool—the life expectancy table—which is designed to predict average life expectancy of a group to calculate the life span of an individual.
The tables are not accurate in predicting a specific individual’s lifespan. In fact, an individual tool is accurate about five percent of the time. Five percent!
Let’s review again why life expectancy tables are so inaccurate on an individual basis. Let’s say the life expectancy table tells us the group’s average life span for a 62-year-old is 20 years (age 82). An average life expectancy of 20 years at age 62 clearly doesn’t mean the group lives exactly 23 years longer (to age 82). Some will die before that age and some will live beyond age 82. In reality, only about five percent of current 62-year-olds will die on or about age 82; about half of the group will die sooner and about half the group will die later. Remember, the 20 year life expectancy is an average figure and varies due to many factors including current age, gender, race, and geographic location. Any retirement planning software is an approximate tool. The life expectancy calculations are, by their nature, inaccurate on an individual basis.
What are the solutions? In my view, awareness is the first step. Understanding the inherent limitation of life expectancy calculations is necessary in order to prevent excess dependence on a tool with limited accuracy. One solution is to consider the use of alternative tables. For example, David Blanchett, head of retirement research at Morningstar Investment Management, has proposed using annuity purchase tables (which are more conservative) to more accurately project life expectancy and plan accordingly (“How To Better Estimate Life Expectancy,” Investment News, February 12. 2015).
Perhaps we need to slow down the planning process to make sure individuals are aware of the inherent limitations of the tools they may use, and foster the understanding of what a life expectancy table is and how it works. Life expectancy tables tell us on average how long a group of people can be expected to live.
Perhaps we should ask our clients if they understand how the tools work. Maybe we should ask them about the appropriateness of specific tools. Maybe we should discuss the tools’ limitations. Again, I am not against tools, not at all. In fact, I have helped build and develop planning tools. But I believe we have an obligation to help clients understand the proper use and limitations of planning tools.