Retirement Planning Using Return Assumptions vs. Return Requirements – A Fiduciary Perspective
Since the onset of the age of Modern Portfolio Theory (“MPT”), practitioners have relied on various forms of “portfolio optimization” to generate retirement projection scenarios in order to analyze various investment outcomes before offering a recommended strategy. These strategies often result in specific asset allocation guidance. It is now clear that misuse and abuse of asset allocation can increase fiduciary liability (see “7 Deadly Sins Every ERISA Fiduciary Must Avoid: The 5th Deadly Sin – Misapplied Asset Allocation,” FiduciaryNews.com, June 9, 2015). Can it be that the very process used to generate this asset allocation advice is predicated on input assumptions that also have tenuous reliability? If that’s the case, does it remain in the best interest of retirement savers to expose themselves to data output that may be framed – whether by accident or not – in a potentially misleading manner?
Most professional retirement projection software as well as popular internet-based “retirement calculators” employ some form of return assumptions, usually by and for various asset classes. As one of the “value added” services they can offer, professionals will regularly interpret program output and present an analysis of it, along with specific recommendations, to their clients. Likewise, individuals will on occasion make use of do-it-yourself internet-based tools (usually “free”) to divine their own conclusion. In both cases, projections are made using generally accepted principles that date back to the MPT era. While any systemic failure on the part of these algorithms represents a mere caveat emptor for individuals, there may be more serious implications for professionals acting in a fiduciary capacity.
How Does the Typical Industry Standard Retirement Projection Software Work?
The idea of using computerized calculations to aid in retirement planning has been a component in the service provider industry for some time. “There are a number of approaches to this,” says Steven Glasgow, Sr. Vice President at Avondale Partners in Nashville, Tennessee. “The most prevalent being some form of backward looking average of historical returns (a naïve approach) or creating assumptions through a form of risk ‘layering’ where a risk-free rate is chosen (there is legitimate argument over whether that should be a T-Bill rate, 10 year treasury rate, or some other derivative of the treasury or inflation curve) and then adding successive layers of ‘risk premium’ based on the assumed and historical risk patterns of a given asset class (e.g. corporate bonds would have a lower risk premium based on lower historical volatility and higher placement in capital structure than corporate equity etc…).”
The most sophisticated software uses advanced statistical techniques to generate output that, at least from a theoretical standpoint, can be considered “statistically valid.” This entire process, however, is predicated on the use of “return assumptions.” This input data generally originates from well-recognized industry sources and is frequently used in academic research as well as for practical applications like retirement projections. “Most software has return assumptions baked in,” says Gage DeYoung, Founder of Prudent Wealthcare LLC in Aurora, Colorado. “It looks in the rearview mirror regarding returns in previous markets of the various asset classes and uses Monte Carlo simulation to project future possible results given a set of parameters including current value, anticipated additions, and type of account (taxable or tax deferred). Morningstar (previously Ibbotson) has acquired and provides a majority of the data utilized.”
The better software not only uses generic market return data, but also permits inputs specific to individual clients. “Projections for participants are a bit like a ‘personal DB Plan,’” says Charles W. Leggette, Chief Actuary at Vantage Benefits Administrators, Inc. in Dallas, Texas. “All the statistical components are available but are seldom addressed as us actuaries would. A stochastic model with a Monte-Carlo element actually works best as it expresses the outcome as a probability (like given the asset mix requested you have a 42% chance of achieving your goal). With the modern web and computers, this computational scenario is no longer out of reach. Assumptions are in two classes, ‘atomic’ and ‘molecular.’ Atomic includes gender, birth, hire, and normal retirement date. Molecular includes current assets, deferral rate employer match rate, presence of profit sharing and pension benefits, marital status, pay, pay increase rate, and asset allocation scheme. Practically speaking, its pay, age, deferral rate, assets, and asset allocation scheme.”
Return Assumptions – A Weak Link that Increases Fiduciary Liability Exposure?
At the heart of this method of calculating retirement projections lie the return assumptions. “They play a huge role,” says Leggette.
The significance of return assumptions cannot be understated. “Retirement projections and Monte-Carlo simulations are completely dependent on these assumptions to derive projected outcomes and probability distributions,” says Glasgow, who adds, “You cannot project anything without them.”
On the face of it, this is not terribly disturbing. After all, when you get right down to it, all you’re talking about is a high school algebra exercise. “In a projection, you are basically trying to solve a math problem,” says J. Dennis Mancias, Financial Advisor at Planto Roe Financial Advisors in San Antonio, Texas. “There are only so many variables you can solve for in a retirement projection: Cost today (savings), rate of return, time (to retirement) and cost in retirement (income). We need to make as many of the variables as static as we can, so return assumptions is one variable we can look to history to provide. We do this not so much as to predict precisely what will happen, but to learn what is reasonable to expect to happen based on the long term historical ups and downs of each asset class.”
Simply plugging in the historic average return yields a straight-forward calculation that might seem obvious. The advanced statistics, however, can sound intimidating. Rest assured, save for a hidden randomizer, all this stochastic wizardry is nothing more than a series of straight-forward calculations. “The resulting projections are straight line but are complemented by Monte Carlo analysis to graphically portray a range of likely outcomes,” says Andrew M. Aran, a partner at Regency Wealth Management in Ramsey, New Jersey.
While the math may be easy, the significance of the assumption itself is not something to misjudge. The bigger the role, the bigger the risk, and the risk associated with using return assumptions – no matter how they are derived – has, unfortunately, long been overlooked. “Relying on a single component such as return is extremely risky,” says Leggette. “The risk is self-fulfilling, however, and is a function of the proximity to retirement date.”
This risk manifests itself in many ways, some of which most people can easily identify. For example, there’s the “garbage-in-garbage-out” danger. “The risk of relying on return assumptions lie in the basis of the assumption,” says Mancias. “The less quality data you use to base your assumptions, the less qualified the assumption will be.”
The bigger problem is that historical return assumptions can give the false impression that returns will be consistent. We know from experience returns can be quite volatile, especially in shorter periods. “The risk is we don’t really know how the asset classes will behave in the future,” says Christopher V Kimball, owner, Christopher V. Kimball Financial Services LLC in Lakewood, Washington. “Over long periods of time there can be somewhat predictable patterns, but over shorter durations making rate-of-return assumptions can become almost meaningless.”
But the granddaddy of them all is the equivalent of Dante’s infamous inscription on the gates of Hell (“Lasciate ogne speranza, voi ch’intrate,” or, as it is most frequently translated, “Abandon hope all ye who enter here”). This, of course, is the SEC’s mandated disclosure with regards to all things related to returns: “Past performance is no guarantee of future results.” “As a matter of fact,” says DeYoung, “many recent articles are projecting much lower returns from portfolios going forward for the long term.”
With this warning, how should someone who’s fiduciary life depends on it go about identifying and assembling the appropriate components to determine a reasonable return assumption? Dorann Cafaro, Founder of Cafaro Greenleaf located in Charleston, South Carolina, bluntly states, “Wrong question: one can’t project returns but can determine what return is assumed to be needed to reach retirement goal if the individual doesn’t save more.”
And this presents an alternative – and perhaps a safer – way for a fiduciary adviser to add value.
How Using the “Goal-Oriented-Target” (“GOT”) Approach May Be in Retirement Savers’ Best Interests
“No brainer, GOT is best,” says Leggette. For an increasingly larger number of practitioners, the shift from using return assumptions to using return requirements (i.e., GOT), has already begun (see “How Does Goal-Oriented Targeting Work?” FiduciaryNews.com, July 15, 2014).
It offers a method that is wholly dependent on the facts and circumstances of the retirement saver and is totally divorced from investment assumptions (and the fiduciary risks inherent in making those assumptions). This means a financial professional can speak to the importance of the savings strategy well before any talk of investments enters into the discussion. Moreover, there are no limits when the discussion does turn to investments because the GOT process remains agnostic when it comes to investment style, discipline, or penchants. “GOT is in the best interest of the client because it is based on the specific circumstances of the client and uses a broader, diversified portfolio mix,” says Mancias.
Furthermore, it supersedes any “risk tolerance” factors. It remain indifferent to the retirement saver’s risk preferences; thus, acknowledging the long neglected truth about the relative lack of importance a retirement saver’s qualitative personal risk assessment has to the quantitative reality of the GOT process. “The GOT should be the primary concern – as a form of liability matching,” says Glasgow says. “Risk tolerance should be looked at, and in the case where a client has accumulated more than enough assets to make the GOT meaningless, then it should replace GOT as the primary determinant of portfolio construction.”
DeYoung describes the formulaic differences between the traditional use of return assumptions and GOT using easy-to-understand examples. He says, “A return assumption is the return you are plugging into a model and again is usually a historic average portfolio return provided by Morningstar.” For example, he says if you have $1,000 and the return assumption for the recommended asset allocation is 7.2%, you will be projected to have $2,000 after ten years. On the other hand, DeYoung says a GOT “is the return required to give a certain result when a current value is given and a future value is desired given a timeframe.” Here, he presents the following example: “If a client wants to save $500,000 for retirement and they currently have $100,000, they can save for 20 years and can only save $10,000 per year they would have a return requirement of 3.77% from their portfolio.” In this example, the GOT would be 3.77% and is independent of any investment recommendation.
Of course, in the end, both processes may yield the same recommendation. Kimball says, “It’s really a different way of reaching the same conclusions. If an appropriate allocation is used with return assumptions and doesn’t allow the money to last long enough, a more aggressive allocation may need to be used. If one is using a goal-oriented target and the money isn’t sufficient, a more aggressive allocation also must be.” On the other hand, the GOT process may more naturally focus attention back on savings habits rather than investment allocation since the entire GOT process is conducted absent of investment assumptions. In other words, rather than looking for riskier asset classes appropriate for obtain a 10% annual return rate, the first choice in the GOT process is to increase saving (or, less likely, decrease expected retirement spending).
It is this difference in framing, however, that offers the best avenue to reduce one’s fiduciary liability. “Using a return assumption could be misinterpreted by a client,” says DeYoung says. “A client could feel they were led to believe that a return assumption utilized by an adviser was prudent and achievable. It is very important for an adviser using a return assumption to point out ‘previous returns are no guarantee of future results.’ By using a [GOT] return requirement, the client would better understand that their portfolio has to perform at a certain average percentage rate to achieve the goal at hand.”
Leggette agrees. He says, “Actually, GOT reduces risk because it expresses the likely outcome as a probability…like, ‘Given your asset allocation you have a 42% chance of reaching your goal.’”
Bear in mind, no matter which process is used, they both represent only a snapshot-in-time of a retirement saver’s life. Competent fiduciaries understand the need to periodically “update the numbers” as circumstances can change. DeYoung says, “Return assumption has historically been the industry standard to project where a client may land given current value, additions and time to save. Return requirement puts the burden of performance on the portfolio. In both cases these are ‘Where-are-You-Headed’ projections. Things can change quickly. People may have increase or decrease in salary, may lose a job, may have an emergency come up. Anything can happen and interfere with the goal of accumulating wealth to achieve financial independence. I have had a few clients achieve their goal of financial independence at retirement age only to find out they have a terminal illness and get robbed of their golden years. It is more important to enjoy life while you are here to do so. Take care of all of your needs and some of your wants. Have adequate protection for your life, health and property. Save methodically for financial independence. This financial life balance is far more important than obsessing over portfolio returns.”
Are you interested in discovering more about issues confronting 401k fiduciaries? If you buy Mr. Carosa’s book 401(k) Fiduciary Solutions, you’ll have at your fingertips a valuable reference covering the wide spectrum of How-To’s every 401k plan sponsor and service provider wants and needs to know.
Mr. Carosa is available for keynote speaking engagements, especially in venues located in the Northeast, MidAtantic and Midwestern regions of the United States and in the Toronto region of Canada. His new book Hey! What’s My Number? – How to Increase the Odds You Will Retire in Comfort is available at your favorite bookstore.