By Betty Meredith, CFA, CFP®, CRC®
Making informed retirement income decisions is much more difficult than making retirement accumulation decisions, and millions of pre-retirement and retired Americans currently do not have access to the qualified professional advice they need.
Professionals choosing to work with these millions of middle mass and middle mass affluent clients in the future will be called on not only to generate retirement income, but also optimize the insufficient resources these clients bring to the table by making sure key risks such as longevity, inflation, health and long-term care are first addressed.
Sounds like a job for software.
2003 and 2009 Retirement Software Studies
In 2003 the International Foundation for Retirement Education (InFRE) co-sponsored a study with the SOA and LIMRA to evaluate retirement planning software programs. The software included the retirement planning components of popular programs used by financial planners, the proprietary software of several financial services firms, as well as key consumer programs available via the internet. The study applied the data of six hypothetical case studies to all 19 programs and assessed the following:
- The capabilities and usability of the software
- Approaches used to address key retirement risks
- Income sources
- Income distribution
- Long term care
The study demonstrated that most programs treated post-retirement planning as an extension of accumulation planning. The ability to model income solutions was far less than what planners needed to optimize a retirement income plan and also address the key post-retirement risks most mid-market pre-retirees and retirees face. Results varied widely from one program to the next. In short, the authors recommended that planners employ multiple programs to essentially flush out a second – and sometimes third – opinion.
The study was repeated in 2009 using 12 programs and more in-depth case studies. Software was categorized as to whether it primarily addressed investments and portfolio management, how much to save for retirement, or how to manage retirement resources and risks. The goal again was to determine whether or not the software tools appropriately address the modeling needs of retirement professionals to explore a variety of alternatives for creating income plans while managing post retirement risks. As before, six cases were used to compare the results. This time, however, they were designed to represent the largest six of the twelve market middle-mass and middle-affluent segments identified in the SOA’s 2009 Segmenting the Middle Market Study and test for the following:
ADD CONTENT HERE
2009 Study Results
The goal of the 2009 study was to identify the general progress made in dealing with post-retirement risks by software programs in general since the 2003 study. Did the more recent programs recognize the key risks of longevity, inflation, health and long term care, and various market risks, and if so, what options were provided to address those risks? In particular, the study evaluated the advice given for:
- When to retire
- When to receive Social Security benefits; coordination with spouse’s
- How much should be saved, and whether the savings need was over or under-estimated
- How much can be consumed during retirement
- Asset drawdown recommendations when savings are insufficient
- Use of income annuities to create a floor when needed
- Which assets should be spent down first
- How an investment portfolio should be changed to manage risks and produce income needed
- How well the program provided opportunities to educate the user
- How well retiree health, life, long-term care, Medicare, annuity and inflation-indexed insurance products we integrated and considered
- Down-sizing a home; paying off the mortgage early versus contributing to a tax-deferred account benefit
- Various scenarios and how the advice differed from other programs for the same scenario.
The 2009 study found much progress was made in professional programs, but that there is still room for improvement in key post-retirement risk areas:
Most of the professional programs still do not allow for stochastic modeling of key post-retirement risks as planners might need:
The researchers also felt the need for best practice guidelines on the following during input and interpretation of results:
1. What defines retirement income adequacy?
2. Should replacement rates be used? If they are used, what should be the target? Should target replacement rates differ across groups of peoples such as single persons, single earner couples, or dual-earner couples?
3. Should target replacement rates differ and be modeled by age of retirement?
4. What should be the planning period for modeling and how does that differ among the groups of people above?
5. What probability of success is desired when using stochastic modeling?
6. What should be the rates of return used in deterministic models?
7. What default rates should programs provide for inflation, investment returns, life expectancy, and income replacement?
8. Why are there such different outcomes from program to program?
In sum, the 2009 authors concluded, as did the 2003 study authors, that any one retirement software “should not be used as the sole input for decision-making for retirees.”
Software companies are not at fault for shortfalls in what planners need from their programs. The fault lies squarely with us – the financial planning industry – for assuming retirement income can be secured for all people through management of assets alone even though it is a well known fact that mid-market baby boomers currently have insufficient assets and inappropriately self-insure multiple retirement risks.
The findings of both of these studies point to the need for the financial and retirement planning industries to take the lead in developing best practices guidelines for the middle mass and middle affluent markets for software companies to build into their models. The 4-5% safe withdrawal rate, primarily an asset management approach to retirement income which works well for higher net worth clients, is clearly not enough for the mid-market. Key retirement risks such as longevity, inflation, health and long term care need to be addressed before asset allocation decisions are made, since the limited resources of the mid-market pre-retirees and retirees are required to do double duty. Optimization of assets for these market segments will require software that can model multiple approaches to help advisors decide which strategies potentially meet a client’s needs.
Once best practices for managing post retirement risks and income are identified, modeling methods and client reporting will quickly improve. Creating an income floor based on essential and discretionary expenses, which also transfers as many longevity, inflation and long-term care risks as much as possible, is best for the mid-market, while the 4-5% rule is most appropriate for higher net worth clients. However, more research on modeling methods is needed. It’s not the responsibility of software developers to identify the standards to include for meeting clients’ retirement income planning needs. It’s ours.
Betty Meredith is Director of Education and Research for the International Foundation for Retirement Education (InFRE). She oversees incorporation of research findings and best practices into InFRE’s Certified Retirement Counselor® certification study and professional continuing education programs to help professionals meet the retirement preparedness and income management needs of clients and employees.
A shorter version of this article was published in the August, 2011 Journal of Financial Planning. (www.FPAnet.org/Journal).
1 Retirement Planning Software, Society of Actuaries, LIMRA and InFRE, Sondergeld et al, 2003.
2 Retirement Planning Software and Post-Retirement Risks, Society of Actuaries and the Actuarial Foundation, Tuner and Witte, 2009
3 Segmenting the Middle Market: Retirement Risks and Solutions–Phase 1 Report, Society of Actuaries, 2009
4 Retirement Planning Software and Post-Retirement Risks, Society of Actuaries and the Actuarial Foundation, Tuner and Witte, 2009