Data used in Retirement Index are 'flawed'

Tuesday, Feb. 23
Posted 2 p.m. EST

Readers continue to comment on the blog I wrote about the Retirement Readiness Index designed by retirement consulting firm Fiduciary Benchmarks. CEO Tom Kmak contends his index shows that Americans on average are on track to meet 92 percent of their retirement income needs -- as long as they wait until full retirement age (66 or 67) to retire.

Last week, author Jane White questioned the conclusions of the study. This week, Certified Financial Planner Greg Kasten -- who is also a physician and president and CEO of Unified Trust Company -- says the calculations used in the index are based on incorrect simplistic assessments. Specifically, he wrote me about seven mistakes. In his words:

1. Account balance (asset averaging) errors.

I believe this is the worst mistake (out of many). The system takes the amount of plan assets and divides by the number of plan participants for an "average balance." Most plans are the "80:20 Rule," or more likely the "90:10" rule, where 10 percent of participants own 90 percent of the assets.

Clearly, the averages will be skewed by a few employees with very large balances.

2. Longevity mistakes.

For longevity, RRI uses the life expectancy of a female employee. This means ... half of employees, at least female employees, would still be alive at the time they assume the participant will have died. They do not use the actual demographics of a particular plan.

3. Inaccurate participant age.

They apparently do not even look at the average age of an employee in the particular plan. Instead, they are looking at the average age of an employee in the same industry.

4. Compensation mistakes.

RRI does not consider even the average wages of the plan, but they are looking at wages (defined as the 75th percentile) based on the industry classification average. They do not take into account the geographic location of employer, nor anything specific to the employer's workforce.

5. Contribution errors.

Likewise, RRI is doing the same overly simplistic average for both employer and employee contributions by again looking at an average tabulation based upon total plan contributions and the number of employees.

6. Investment return errors.

RRI does not look at the plan's investments, let alone the actual allocation of each individual employee. Instead, RRI uses a 50 basis point return over riskfree Treasuries.

7. Investment risk errors.

RRI's use of risk‐free Treasuries means they are applying virtually no volatility to the portfolio. This would skew results to the upside by not taking into account the actual losses a portfolio might sustain.

Bottom line: Every participant is different. You cannot use simple average, even at the plan level of data. The RRI is not very helpful.

Tom Kmak, CEO of Fiduciary Benchmarks, responds as follows:

I agree with many of Dr. Kasten's comments. The major one we take issue with is his contention that the RRI is not useful. In fact, there is nothing in Dr. Kasten's comments that we did not already know or debate at length.

We did the plan level RRI, however, because we wanted to start the conversation in the right place and have something that is directional versus nothing at all. Yes, we made some assumptions for the average participant in a plan, but a company can update all of that data for no charge starting very soon. In essence, we believe there is great potential benefit in setting forth an RRI number for 41,000 companies covering 55 million participants and the impact of meaningful conversations like this one.

While we believe the Retirement Readiness Index will increase focus on the ultimate question that is important to the retirement industry -- "Can I retire how I want, when I want?" -- it should be noted that based on 25 years of experience in the industry, Fiduciary Benchmarks is convinced that helping one individual at a time is by far the most effective method for improving retirement outcomes. This is why the RRI is also available at the participant level.

Thus, the participant-level RRI version, which we will release this year, will account for what Dr. Kasten has outlined. And if a plan computes an RRI for every single participant, a composite of those numbers would replace our "average" number.

My biggest retort to Dr. Kasten would be: Let us not sacrifice the benefit of starting the right conversation for perfection we can never achieve (i.e., a projection is by definition just that, a projection). In any case, we applaud Dr. Kasten and Unified Trust for their work on retirement readiness.

Questions? Comments? E-mail

Read more Boomer Bucks blogs.


Show Bankrate's community sharing policy
          Connect with us

Learn the latest trends that will help grow your portfolio, plus tips on investing strategies. Delivered weekly.


Jennie Phipps

3 ways to outsmart Uncle Sam

Retirement tax planning isn't just for millionaires.  ... Read more

Partner Center

Connect with us