Sequence of returns and why it matters
If you spend any time perusing a financial independence, retire early (FIRE) community, you’ll eventually come across sequence of returns risk, or SORR. Now, if there’s one negative thing I will say about FIRE, or finance in general, it is that much like medicine, there’s no shortage of acronyms!
The context where SORR is usually first encountered is when discussing safe withdrawal rate, or SWR. SWR represents the percentage of your initial portfolio that you can withdrawal every year in retirement, while maintaining confidence that your portfolio will not fail. For example, using a commonly-cited SWR of 4% means that if you need $40,000 per year for living expenses, you can retire when your portfolio hits $1 million for the first time. In subsequent years, you can continue to withdraw $40,000 every year, adjusting upwards slightly for inflation, regardless of how your portfolio performs thereafter (note that 4% SWR does not mean withdrawing 4% of your current portfolio value every year, which might fluctuate wildly due to market conditions).
The beauty (and some would say, illusion) of SWR is that you can ensure a steady and unchanging stream of income in retirement without depleting your portfolio and without needing to adjust for market performance. It also gives you a nice target “FIRE number” to hit based on your annual expenses, which would be 25x your expenses if you used 4% SWR, 33x for 3%, and 50x for 2%. Needless to say, the topic of what is actually safe in safe withdrawal rate is endlessly debated. Obviously, the more conservative your withdrawal rate, the more likely that your portfolio will last forever. But at too conservative of a withdrawal rate, you might be delaying your retirement for too long, or leave too much money unspent when you die (although some people prefer to leave behind a sizable estate).
I have an entire chapter in my book on SWR and there are also academic papers and entire blogs devoted to in-depth review of this topic. The truth is that because the future is unknowable, no withdrawal rate is absolutely safe. However, extensive back-testing of historical U.S. market conditions shows that SWR between 3 to 4% survives pretty much everything, although 4% has failed under exceptional circumstances. Note that I use the word market in this article interchangeably with the S&P 500 at times. The S&P 500 represents about 80% of the total US domestic equity market by market capitalization and its movement correlates very well with the total market, but they are not the same.
Looking at historical market conditions reveals that the S&P 500 has a historical compound annual growth rate (CAGR) of around 9 to 10% with dividends reinvested (read my article about CAGR if you haven’t). Therefore, the obvious question is:
If my investment returns 10% per year in the long-run, why can’t I withdraw 10% per year?
The answer is the subject of this article: sequence of returns risk, or SORR.
Basically, the market doesn’t neatly return 9 to 10% every year, even if historical 30, 50, or 100-year CAGR comes out to between 9 and 10%. The S&P 500 has negative returns about 1 in every 4 years. In fact, in some years (such as 2000 to 2002, or 2008), the market crashes and loses 20, 30, or even 40%, and it can take up to a decade to regain those losses:
S&P 500 Historical Annual Returns
Image source: Macrotrends (click to see an interactive version of chart with downloadable data).
For example, if you retired in 2000 using SWR of 4%, withdrawing $40,000 on a $1 million portfolio, by March 2003 your portfolio value is now barely over $500,000 (varies slightly depending on how and when you make your withdrawals). An annual withdrawal of $40,000 now represents 8% of your portfolio value, which does not sound sustainable. If the market continues to decline for several years before recovery, you would face the terrifying prospect of depleting your portfolio. A subsequent market recovery is of little consolation if you run out of money before the recovery happens. You fall victim to an unfortunate sequence of returns.
Therefore, sequence of returns risk is why the concept of a safe withdrawal rate exists in the first place. The idea is to find a withdrawal rate which can survive all previous market conditions, and (hopefully) weather all future market sequence of returns. There are, of course, other ways to withdraw from your portfolio. The most common variant would be a variable withdrawal, where the retiree cuts expenses and withdraws less during market downturns. One intriguing withdrawal method that has generated increasing discussion is a CAPE-adjusted withdrawal rate, which is beyond the scope of this article.
If sequence of returns risk exists, does the opposite also exist? Absolutely! Just as sequence of returns represents risk to the retiree, it can also provide benefit or opportunity to the investor in the accumulation phase. Consider the following phenomenon:
Three different investments with identical CAGR (i.e. 10% over 10 years) can yield dramatically different results, based on the sequence of returns and how you invest.
I’ll use somewhat unrealistic numbers in these examples, just to illustrate my point. Slow and Steady Company is the first investment. It returns precisely 10% per year, every year on December 31st to shareholders who have held the stock for the year. We make a lump-sum investment of $10,000 on January 1st of year 1:
The 10 year CAGR of Slow and Steady is 10%, and after 10 years, we have $25,937.42.
Late Bloomer Company is our second investment. This is a startup company that does absolutely nothing for 5 years, but launches a revolutionary product in year 6. It returns 0% per year for 5 years and returns 21% per year in years 6 to 10. We make a lump-sum investment of $10,000 on January 1st of year 1:
After 10 years, we also have $25,937.42. The CAGR of Late Bloomer is also 10% over 10 years.
Early Blossom Company is our third investment. This is a company that had incredible potential, but became stagnant from mismanagement. It returns 21% per year for the first 5 years, and returns 0% per year in years 6 to 10. We make the same $10,000 investment on January 1st of year 1:
After 10 years, we also have $25,937. The CAGR of Early blossom is also 10% over 10 years.
So far, so good. All three companies had 10 year CAGR of 10% and we end with the same amount.
But what if we invested our money periodically?
Let’s assume that we didn’t invest $10,000 lump sum in year 1. Instead, we made periodic investments of $1,000 per year on January 1st each year for 10 years, but still totaling $10,000 invested.
After 10 years, we have $17,531.17. This is because only the initial $1,000 in year 1 benefited from 10% annual compounding for the full 10 years. The later investments had less time to compound.
After 10 years, we have $22,151.70. This is because more of our invested money benefited from the returns over this period (by the time 21% growth started in year 6, we already had $6,000 invested).
After 10 years, we only have $14,182.99. A small amount of invested money benefited from early growth, but the money we put in after year 5 didn’t grow at all.
So we have 3 investments with identical CAGR of 10% over 10 years yet they yield dramatically different results when we make periodic investments. This is because their sequence of returns during this period were different, and this sequence interacts with periodic cashflows. As you can see, the overall CAGR does not adequately describe investment performance in this case.
If you were wondering, the Money-Weighted Rate of Return, or MWRR, is the measure of your investment returns when cashflows are taken into account. Each of the 3 example investments above have a different MWRR value. MWRR is unique to each investor and each investment because it captures the timing of the periodic cashflows interacting with the investment’s return. MWRR calculations are much more complex than CAGR, which only looks at beginning value, end value, and time period. The bottom line is this: When you make ongoing contributions or withdrawals, the sequence of returns matters.
Without getting too deep into the weeds, the main point about sequence of returns is that ideally, the retiree withdrawing from their portfolio wants to see a strong bull market at the start of retirement. Even if he encounters a bear market later on, he’s less worried, because the bull market has grown his portfolio so much that by the time the bear market comes, his withdrawals might now only represent 1 to 2% of his new portfolio value.
On the other hand, the investor contributing to their portfolio benefits from a bear market at the beginning of their investing career. This is the phase where they are accumulating their nest egg. They get to purchase more shares at a discount early on. IF a bull market eventually arrives, the more money they already have invested, the better.
So, yes, the S&P 500 historically has a CAGR of around 9 to 10%. But almost nobody makes a lump-sum investment at a single point in time. For everyone who is either investing periodically before retirement, or withdrawing periodically after retirement, the sequence of returns matter a great deal.
Happy investing!
