Forecasts point to lower future investment returns and, more specifically, to a lower equity premium return for stocks over bonds. If indeed we will experience a lower equity premium, how much lower of a premium will make stock investing unattractive relative to bonds? I’ll develop an example based on a retirement scenario to help us think through this issue.
The example
I’ll compare retirement outcomes based on the historical average equity premium versus a lower equity premium. The issue of future investment returns and safe retirement withdrawals was recently addressed in this Advisor Perspectives article by Bill Bengen, and his article responded to this Morningstar paper on safe withdrawal rates that has been receiving considerable publicity.
Here are the specifics for my example. For investments, I establish a baseline by assuming a 0% real arithmetic average return for bonds. Current yields will have a strong anchoring effect on bond returns for retirements in the near term and those yields are currently negative, so I am assuming some offsetting yield improvement over time. For stocks, I set up two scenarios: a 7% equity premium for the historical scenario (based on Ibbotson for large company stocks) and 5% for my alternative scenario. For both scenarios I assumed standard deviations of 20% for stocks and 7% for bonds based on history. I used a 60/40 stock/bond portfolio with rebalancing to maintain that allocation. Other assumptions included $1 million of initial savings, $30,000 of annual Social Security income, and a fixed 30-year retirement period. I set up annual retirement withdrawals that are based on RMDs and will therefore vary with investment performance. More specifically, I used RMD factors developed from the extended IRS table for couples beginning at age 65, and multiplied those factors by 1.25. The 1.25 generates reasonably level real expected withdrawals over the retirement period. These assumptions feed into a Monte Carlo model that generates 5,000 30-year retirement scenarios.
Comparison
In addition to the two equity premium scenarios, I set up a baseline for comparison using a hypothetical TIPS ladder assuming a zero real yield. This generates annual level real withdrawals of 3.33% of the initial $1 million or $33,333, which will use up the $1 million over the 30-year retirement period. I added $30,000 of annual Social Security income, which sets the annual baseline at $63,333. By comparison, the first year RMD at age 65 is 2.95%, and, multiplied by 1.25, equals 3.69%, so the cash flow from the first-year withdrawal plus Social Security will exceed the baseline 3.33%. If annual returns from the 60/40 portfolios were equal to the assumed averages without volatility, annual cash flow would always exceed the baseline, but introducing return volatility means there can be shortfalls.
Here’s the chart showing the comparison:
For each of the 5,000 simulations, I calculated the average annual cash flow over the 30-year retirements and show the median in the first column. By lowering the equity premium assumption from 7% to 5%, we lose slightly more than half the cash flow margin that stock investing provides over the base case.
The next column shows the 10th percentiles, and we see that both equity premium assumptions leave us with retirement cash flow less than the base case.
Next, I show median bequests remaining at the end of 30 years. The base case is designed to fully deplete savings, so any bequest provides a bonus over the base case.
The next measure is what I call the “shortfall,” which represents the total amount by which annual cash flow falls below the baseline of $63,333. Even though the median annual cash flow is above the base amount, there will be some years when cash flow falls short, and I accumulated the amount of such shortfalls over the 30-year retirement.
Finally, I show a measure of cash flow volatility – the average absolute amount of change (up or down) in cash flow from year to year.
Positives and negatives
With both the historical 7% and reduced 5% equity risk premiums, the expected cash flow exceeds what we would expect from bonds only. Another positive is that both cases produced expected bequests. Those are the positives associated with stock investing, and the reduced equity premium lowers both positives somewhat.
But now for the negatives.
There is the risk, as shown by the 10th percentiles, that cash flow over retirement could come in below the base case. Related to this risk are instances where the annual cash flow falls short of the base case, and these get worse with the reduced equity premium assumption. Finally, with variable withdrawals, there is the prospect of disruption due to cash flow bouncing around from year to year.
There are tradeoffs associated with taking stock market risk, and the above chart shows how the tradeoffs become more negative as we lower the equity premium assumption. Different clients will put different weights on the positives and negatives.
What to do
Reacting to the prospect of a lower-than-historical investment returns and a lower risk premium from stock investing creates a dilemma for clients and advisors – whether to invest more heavily in stocks to increase expected cash flows and bequests or take a more defensive posture and reduce the equity allocation to hold down the negatives.
For practitioners, this analysis would need to focus on customized projections that reflect client specifics. For example, if a client has a solid base of lifetime income from Social Security and perhaps pensions or annuities, such income may be enough to cover basic living expenses and make it easier to tolerate a higher equity allocation for the investment portfolio. But if a client has more limited lifetime income sources, the negatives discussed above will loom larger, and the best approach might be to use a lower stock allocation. It may also be worth exploring delaying Social Security and other options such as an annuity (e.g., a SPIA) or setting up a reverse mortgage to provide more secure lifetime cash flow and reduce potential damage from poor investment performance.
Is 70/30 the new 60/40?
There has been discussion in the past few years that prospects of reduced returns mean that clients need to take more risk – going to higher stock allocations than the traditional 60/40. This Advisor Perspectives article by Allison Schrager discussed the tradeoffs in taking more risk. Whether to take on more risk depends on the degree that lower future returns reflect lower returns for both stocks and bonds with no change in the equity premium, versus poorer prospects for the equity premium. If prospects for the equity premium look good, it supports the case for raising the equity allocation. But prospects for a lower equity premium may push things in the other direction if the negatives from stock investing outweigh the positives. In this case the new 60/40 may be 30/70 instead of 70/30.
Joe Tomlinson is an actuary and financial planner, and his work mostly focuses on research related to retirement planning. He previously ran Tomlinson Financial Planning, LLC in Greenville, Maine, but now resides in West Yorkshire, England.
Read more articles by Joe Tomlinson