“...only 7% of investors can stand to have more than 75% of their total investments in stock, and only 1% can handle more than 87%” Finametrica
Target Date Mutual Funds (TDFs) are big business, attracting assets at a blistering pace. Assets in TDFs now top $500 billion according to a
recent Morningstar report, a 21% increase over the previous year.
Nearly $100 billion now flow into TDF products annually from defined contribution plans around the country. A recent study by The Vanguard Group predicts that 55% of all defined contribution (DC) participants and 80% of new entrants will have assets in TDFs by 2017. The Target Date concept as a product, has been a huge success for the industry, whether this success transfers to the participant is another question.
I believe there is evidence participants are being exposed to more risk than they are comfortable taking. TDF providers in general appear to be building their products disconnected from the risk tolerance of the participant.
As a financial advisor I’ve witnessed the rise of TDF’s with both optimism and fear. My optimism is based on the simplicity that TDFs offer DC programs and participants. My fear is based in watching how the TDF has become a vehicle to promote the concept of “Stocks for the Long Run” without regard to a participant’s risk tolerance.
In short, I believe TDF’s underlying equity allocations are disconnected from participant’s comfort level with those allocations.
I believe this disconnection will ultimately lead to significant problems for plan sponsors and a retirement that is less dignified for participants.
A common assumption underlying most TDF’s and many professionally managed portfolios is the younger you are the more loss you can handle (because you have a long time period to make up for it).
The conclusion being that young people need higher allocations to equities. Another common assumption is that inflation is a major threat to a retiree’s portfolio and the only way to combat that inflation is high equity allocations.
Both assumptions have underlying truth’s to them, but represent only one facet of a person’s risk profile.
Risk Profile is a combination of the following (for an in-depth analysis of Risk Profile, please see the September 2008 edition of The Kitces Report which you can access by becoming a subscriber at
www.kitces.com):
Risk Need The minimum rate of return required to meet a person’s goals.
Risk Capacity Bob Veres (another must read industry personality,
www.bobveres.com defines risk capacity as “a (participant’s) ability to sustain a market decline without suffering an unacceptable loss of lifestyle or quality (of) life now or in the future.” In other words, how much loss can a participant “afford” to absorb.
Risk Attitude How much a person is comfortable with fluctuation in their portfolio, this can be interchanged with the term Risk Tolerance.
Risk Perception “...Evaluates the extent to which the client understands the financial markets, and thus heavily influences how “risky” the client perceives the market to be.” Michael Kitces September 2008
Under the above Risk Profile paradigm the various components of a person’s risk profile are split out and analyzed separately by a financial professional and this information is used to make a recommendation. TDFs do not have access to Risk Need, Risk Attitude (Tolerance) or Risk Perception - only, to some extent Risk Capacity.
Yet, TDFs are built as a one-stop portfolio for all. TDFs assume that if someone is 25 years of age, they can handle a 50% drop in their portfolio. A retiree may (or may not) need an engine for growth in their portfolio to combat inflation, but what if they can’t handle the fluctuation that comes with that engine? TDFs ignore this reality.
Fund companies claim to have sophisticated methods for developing glidepaths, but the primary data point used to determine which portfolio a defined contribution participant is defaulted into is their date of birth (DOB). While an informative data point, DOB does not convey general or specific information about what the proper asset allocation of a participant should be, it’s a partial component of their overall Risk Profile. DOB does not communicate the need, perception or tolerance for risk.
None of this is a secret, everyone already knows that TDFs are geared toward the general; the problem is that from my perspective, they get the “general” very wrong.
In my opinion Risk tolerance should be the main driver in the portfolio development process. There are many other factors that one should consider (required return, wealth, human capital, etc), but accurately determining how much fluctuation a participant can handle is foremost.
A participant who bails on an asset allocation at the worst possible time (buying high and selling low) will never achieve their goals and will likely become discouraged, afraid to invest in stocks ever again after experiencing the inevitable large drawdown. In this regard a lower allocation to equities may produce more wealth for many participants (or at a minimum, less nervousness).
Measuring risk tolerance is not simple. Many recordkeepers, mutual funds and broker-dealers routinely put prospective clients through a short 5 or 10 question test to place them into a portfolio, yet these tests are typically inadequate and do not actually measure risk tolerance in a scientific fashion.
FinaMetrica, a company out of Australia has created a risk profiling system that claims to accurately measure a person's ability to assume risk. FinaMetrica describes their system as follows:
“The system provides a scientific assessment of an individual's personal financial risk tolerance in plain English. The system uses psychometrics to ensure validity and reliability. The 25-question risk tolerance questionnaire can be completed in 15-20 minutes and the comprehensive risk profile report is available immediately.”
“FinaMetrica, an Australian company that has developed a respected risk questionnaire used more than 250,000 (nearly 500,000 as of 2013) times by financial planners, has found that only 7% of investors can stand to have more than 75% of their total investments in stock, and only 1% can handle more than 87%. "The investment industry tends to encourage people to take on more risk than they're emotionally equipped to handle," says FinaMetrica co-founder Geoff Davey.”
Only 1% of investors can handle equity allocations above 87%, only 7% above 75%. If true, it seems that TDFs are negligent in their management of participant assets (or Plan Sponsors in choosing the TDF).
If the FinaMetrica system can be relied on (and all evidence points to this being the case), the implications for glidepath construction are significant.
The following chart links different age groups and their corresponding target date retirement year with their mean FinaMetrica score (derived from the above mentioned study) and then compares the Big Three TDF provider’s (T. Rowe Price, Fidelity and Vanguard) equity allocation with what the FinaMetrica score would suggest for an allocation to equity assets.
***Age 65 assumed retirement
****This represents the range of growth assets that FinaMetrica defines as “Comfort,” an allocation to growth assets above this range is “significantly greater than (a person) would normally choose to take on.”
Glidepath Equity Allocation Comparison of The Big Three
A FinaMetrica score of 62, which is the mean score for the age cohort of 20 - 29 equates to a portfolio that is 54 - 73% in equities. Anything higher than 73% in equities and these participants (on average) begin feeling what FinaMetrica terms “Marginal Discomfort.”
If the average twentysomething begins feeling discomfort when their equity allocation reaches above 73% and the average equity allocation of the “big three” for this age group is 90%, doesn’t this suggest a disconnect?
A 60 year old would most typically be in the 2020 target year and would have an allocation between 60 - 72% in equities with the “Big Three.” The average 60 year old has a FinaMetrica score of 51 suggesting the allocation to equities should be between 38 - 57%. Even the lowest equity allocation (Fidelity) exceeds the highest comfort level.
If we look at the whole universe of TDFs there is one 2020 fund that is 80% in equities. There appears to be a major difference between what the fund companies believe a participant’s risk tolerance is and what current research shows it to be.
Even if it were the case that participants were highly risk tolerant near their retirement date, the
data shows that the vast majority of participants remove their money within three years of retiring.
Could it be that the fund companies have confused Risk Tolerance with other aspects of a person’s Risk Profile? It appears that the major fund companies have placed greater importance on participant’s Risk Capacity and Risk Need than on their Risk Tolerance.
This begs the question, which pieces of the Risk Profile are more important?
“...I have always been a strong advocate that a client's risk tolerance must be the driving factor in determining someone's maximum exposure to investment risk. Long time horizons may be an appropriate factor to encourage someone to invest up to their risk tolerance, but clients should never go beyond that comfort level. I don't care if you're 22 years old and you have a 70 year investment horizon until the end of your retirement; owning investments that are riskier than your comfort level is simply a recipe for disaster (emphasis mine), even if you "have time to recover", as the inevitable downturn eventually comes and the dramatic losses destroy your willingness to save and invest. Who wants to save more when all you can do is look at your existing savings, the losses it has experienced, and think about how you could have enjoyed the money more by spending it instead of losing it investing!
The bottom line is that as planners, we need to bear in mind that clients do have an emotional experience to investing, and that even if risky portfolios experience declines that turn out to be temporary, the adverse impact on client savings behaviors can be far longer lasting.”
I’ve long shared the same opinion, which is why I’ve been a critic of TDFs almost as long as they’ve been around.
I want to be clear that I’m not saying Risk Capacity is a useless metric, simply that a person shouldn’t invest above their Risk Tolerance, even if their Risk Capacity is a higher number. Tolerance should act as a ceiling (not a floor) on Risk Capacity when setting an allocation.
The FinaMetrica data gives a voice to the participant where they had none before.
Another area that the FinaMetrica data shows a disconnect is with gender.
Target Date Funds treat equity allocations among genders the same. In “TDFLand,” if you’re a female born in 1950, your portfolio should be the same as a male born in 1950. FinaMetrica data on gender risk tolerance reveal that treating genders the same may not be appropriate.
The average 60 year old female begins to feel ‘discomfort’ when equities reach 53% of her portfolio, yet males the same age don't feel the same level of discomfort until equities reach 68%. TDFs treat the genders the same even though the data appears to show a measurable difference in risk tolerance at all levels.
I realize it might not be sensible to create male and female target date funds, my larger point is that TDFs appear to be taking more risk than a significant portion of their population is ready to assume and they ignore ALL demographics of the population except date of birth.
If an Investment Advisor based their recommendations solely on date of birth, they’d be breaching their fiduciary duty. Even a Registered Representative would likely find themselves in hot water.
No competent financial advisor would assign a client a portfolio solely based on their birthday.
Yet, the vast majority of new participants entering retirement plans are defaulted into portfolios which they may not be comfortable with. Even more are holding higher levels of equities than the data would indicate they are comfortable with.
The FinaMetrica questionnaire may not be perfect, but it does have a reputation for accurately portraying a participant’s risk tolerance.
I'm not saying we should base all TDFs’ equity allocations on FinaMetrica data alone, but I think the data indicates a disconnect between the risk taken by the fund companies and the comfort of the participant with that risk. This is a significant concern - an entire population is being defaulted into portfolios that appear to be to risky for them.
What Can A Plan Sponsor Do?
First, I think additional research is warranted into the demographics of Risk Tolerance. I believe better defaults can be built by including additional demographics. It’s impossible to determine a participant’s correct Risk Profile purely from demographic data, but building better defaults is a step in the right direction. Surely a default based on additional data, such as gender, income and job title would be more precise than the current one size fits all method.
I’m going to sidestep the question for now in deference to more thoughtful dialogue that I hope this blog post creates.
Conclusion
We will never be able to default a participant into the perfect portfolio, but if a spectrum of portfolios exist from “Terribly Wrong” to “Perfect” and we are currently closer to the “Terribly Wrong” end...should we not strive for perfection, even if we fall short? Perhaps improving our shortfall will help participants improve theirs.
PostScript
My Thoughts on The Big Three
While it looks like I’m attacking Price, Vanguard and Fidelity I want to make it clear that I do like these companies in other areas. I think Vanguard is one of the most amazing companies in the history of money management, I believe T. Rowe Price is one of the best active money managers in the world. Fidelity has some amazing products and programs (full disclosure, I custody asset at Fidelity). I simply have a fundamental disagreement on their glidepaths. I am happy to see that T. Rowe Price is set to release a less aggressive version of their TDF suite and am looking forward to analyzing it.
Scott Dauenhauer, CFP MSFP, AIF
Disclosure and Thank You: Any opinions in this piece belong to me as do any errors. I do want to give a special thank you to Michael Kitces for helping me flesh out my concerns, but this does not mean he does or does not endorse my conclusions. The same goes for FinaMetrica, I used their data, but my conclusions are my own.