Imagine you’re retired or approaching retirement and one morning you discover an article outlining Warren Buffett’s portfolio recommendations for his own living trust. As Warren Buffett is deemed to be one of the greatest investors of all time, you are very intrigued by what this portfolio might look like. You flip the page and see his recommended allocation, as was laid out in his 2013 shareholder letter:
Put 10% in short-term government bonds and 90% in a very low-cost S&P 500 index fund.
As your nest egg and retirement portfolio have consistently been top of mind in recent years, you want to give Mr. Buffett’s recommendations some serious thought. Moreover, you would really like to know if this is how your own portfolio should be allocated for the next 30 years of your retirement.
First, you’d like to get a better frame of reference on how the S&P 500 has performed throughout much of history, particularly over 30-year time windows for retirement purposes. Accordingly, we can review Figure 1 below.
Each bar in the graph represents the annualized buy-and-hold return from individual 30-year periods in the S&P 500 index (including returns from dividends) all the way back to 1900. As we can see, the average annualized return across all the 30-year periods is 10.1%, and ever since the Great Depression in 1929, the average returns have been consistently higher.
Figure 1: S&P 500 Annualized Returns by Rolling 30-Year Time Window
At first glance, it seems that a portfolio of 90% S&P 500 / 10% short-term government bonds (i.e. cash & equivalents) would be a great allocation to follow over the next 30 years of retirement. After all, you have always been a long-term investor like Warren, and the consistency of historical 30-year returns nearly speaks for itself.
Given these insights, you call your financial advisor to reallocate your portfolio accordingly, feeling confident that your portfolio will be positioned to grow exceptionally throughout your retirement and there is nothing left to worry about…..Right?
The Risk in Return Sequences, Volatility, and Drawdowns
Given the advice and historical data above, we feel fairly confident in S&P 500’s ability to generate impressive returns over time, particularly given recent history where returns have regularly averaged above 8%. Something we do not know, however, is what path the market will take to ultimately generate these returns over the course of an investor’s 30+ years of retirement. Furthermore, as we have seen throughout much of market history, returns are not evenly distributed over time, despite how much we would like them to be.
So, what if they are muted or even negative for the first 10-20 years, only to be followed by a major bull market (delayed returns), or vice versa (early returns)? To better conceptualize these different return paths, we can review the illustrations in Figure 2 below.
Figure 2: Differentiated Return Sequences with Identical Annualized Returns of 8%
You might be asking yourself, “why does the sequence of returns and portfolio volatility matter?” After all, each of the equity curves above ultimately end up in the same spot.
Well, it doesn’t matter for many younger investors that plan to invest a lump sum and leave it alone (with dividend reinvestment) for the next 30-40 years until retirement. The end result will be the same, no matter the path taken. Unfortunately for most retirees, their investment portfolio is expected to provide a steady income to cover living expenses throughout their retirement years, and it is this need for withdrawals that can cause the return path and volatility of a portfolio to matter quite a bit. (Note: Withdrawal proceeds include income from dividends, interest, and withdrawals of portfolio principal.)
As we will demonstrate, when it comes to return sequence risk, investors ultimately want to avoid prolonged periods of muted or negative returns in the early years of retirement, as these periods can be tremendously detrimental to retirement portfolio sustainability and longevity.
A Real World Example
In order to better think through the impact of path dependency or return sequence risk on a portfolio invested almost solely in the S&P 500, we can walk through a fairly simple example using historical data.
For instance, let’s take Andrew, a successful business man who is planning to retire next year when he turns 65. After a successful career, Andrew has amassed a retirement portfolio of $1 million, and based on his most recent personal budget, he estimates he’ll need $40,000 after taxes and inflation each year to cover his living expenses over an expected 30-year retirement. Lastly for simplicity, we’ll assume he will be subject to a 20% tax rate on withdrawals from the portfolio, whether due to long-term capital gains in a taxable account or ordinary income taxes on a tax-deferred retirement account like a 401(k) or IRA. This will make his annual pre-tax withdrawals equal to $50,000 adjusted for inflation.
Based on this scenario, we can now analyze how Andrew’s investment portfolio would have held up in each full 30-year period since 1900 if he had only invested in the S&P.
As we can see in Figures 3 and 4 below, if Andrew’s 30-year retirement window resembled the early 1900’s, 1930’s, 1960’s, or early 1970’s in the S&P 500, his retirement portfolio would have been fully depleted before reaching the end of 30 years – and in some rare cases before the end of 20 years. More specifically, his retirement portfolio would have failed to reach his 30-year horizon target roughly 25% of the time, which in our view is much too large a risk for any retiree to take on.
Figure 3: Historical Retirement Portfolio Longevity (Years)
Figure 4: % of Initial Portfolio Remaining After 30-Years
Why Does This Happen?
Well, as we have seen throughout much of the market’s recorded history, there have been intermittent periods of tremendous positive returns that tend to be followed by periods of lackluster performance.
If we look back across the past 120 years of the S&P 500 price trends, as depicted in Figure 5 below, we find these periods of growth and stagnation can last for quite some time – with each return regime (or lack thereof) often extending for a decade or more.
As retired investors intermittently traverse these down-to-sideways market regimes, their annual withdrawals outpace the returns provided by the market, systematically reducing the size of their retirement portfolio over multiple years. Moreover, these dynamics are further exacerbated during market downturns, economic recessions, and/or periods of rising inflation.
Figure 5: S&P 500 Historical Price Trends
Looking to the Future
So what about today? Are the prospects for investing the bulk of your assets in the S&P 500 (or U.S. stocks in general) appealing in the current market environment? From our perspective, it depends.
For younger investors with a long career and decades of investing ahead of them, a disciplined buy-and-hold, “set it and forget it” approach in stock index instruments like the S&P is likely still a very good bet. Yet, we still have reservations against too heavy an allocation, even for younger investors.
More specifically, in the 120-year history analyzed, the S&P 500 experienced periods of material volatility and peak-to-valley drawdowns of greater than 50%-80%, which we feel would prove quite challenging to endure for most investors given too large an allocation.
As for investors entering retirement, we feel it to be wise to proceed with extra caution, particularly given the relatively high valuation levels we’re seeing in the S&P 500 today.
To more specifically illustrate, we can point to Figure 6 below, which takes the Historical Retirement Portfolio Longevity chart from above (Figure 3) and overlays it with the Shiller Price-to-Earnings Ratio (“Shiller P/E”) to gauge historical levels of market valuation prior to each 30-year retirement window. Although it’s a not a perfect timing tool, we can see some correlation between elevated Shiller P/E readings and retirement windows that fail to reach their 30-year targets. Given the most recent readings are above the historical average, we reiterate our feelings of caution when it comes to retirement portfolios being overly dependent on U.S. stock returns going forward.
Lastly, it’s also prudent to note that choosing the U.S. equity market for this analysis does have a strong element of hindsight bias embedded, as it has been the most successful equity market when compared to other economies, and by a very large margin. As such, the portfolio longevity scenarios outlined above would have been far worse if other countries’ equity markets were chosen instead – and given the amount of home country bias exhibited in most global investor portfolios (i.e. people tend to materially overweight their portfolio to their home country’s markets), these more dire scenarios would likely have been a reality for many.
Figure 6: Historical Portfolio Longevity with Shiller P/E Overlay
With much due respect for Uncle Warren, we completely agree that the long-term prospects for a disciplined buy-and-hold approach in the S&P 500 for 30+ years are very attractive, particularly if we all had the ability to cover our living expenses from sources outside our retirement portfolio. Unfortunately for many investors, that is not a realistic situation, and due to the impact that return sequences can have on a portfolio, too heavy an allocation to something as volatile as the S&P 500 may prove inappropriately risky – especially for those that unluckily end up retiring in the 25% of retirement windows that fall short.
At RQA, we don’t believe someone’s investment success or financial stability should be subject to luck or lack thereof. That is why we have spent nearly a decade developing and honing our adaptive global portfolio investment framework, which through multi-dimensional diversification and enhanced risk management techniques seeks to provide a more stable, consistent wealth trajectory for investors.
Disclaimer: These materials have been prepared solely for informational purposes and do not constitute a recommendation to make or dispose of any investment or engage in any particular investment strategy. These materials include general information and have not been tailored for any specific recipient or recipients. Information or data shown or used in these materials were obtained from sources believed to be reliable, but accuracy is not guaranteed. Furthermore, past results are not necessarily indicative of future results. The analyses presented are based on simulated or hypothetical performance that has certain inherent limitations. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight.