We are living longer than ever. This is good news, but it means that we may have to depend longer on a savings portfolio to finance our retirement. The risk of outliving our portfolio is real. Our success depends on choosing the right investment and spending strategy so our savings don't run out after we no longer draw a salary.
One of the best known strategies is the “4% rule”. It states that every year you can take 4% of the initial value of the portfolio, adjusted for inflation, and have a reasonable expectation that it will last. The rule does not say whether 4% of your portfolio is barely enough to heat your house or plenty to pay for leisure travel. It only tells you how much you can take without running your well dry.
This is a problematic rule, not just because it’s potentially inefficient and rarely adequate, but also because of its popularity. Nobel Prize in Economics William F. Sharpe calls it the “the most endorsed, publicized, and parroted piece of advice that a retiree is likely to hear.”
The basic problem with this rule is that it suggests that a fixed annual spending schedule can be supported by a volatile investment policy. Although this is often inescapable, the way we approach this mismatch can make a big difference between failure and success.
Most often this rule comes along with another: a prescription to divide the portfolio investments into a fixed proportion of stocks and bonds. All sorts of historical data showing the returns of both asset classes and how one zigs when the other zags are brought up to prove that the 4% rule, paired with a 60/40 mix of stocks and bonds, will give the average retiree a good chance of success.
But this “proof” is laden with hidden, undisclosed risks. One is the “sequence” risk: historical returns, volatilities and correlations do not account for the huge difference that a few good or bad years at the beginning of retirement can make. A market that doubles and then comes down to its initial value at the end of twenty years is far better than one that goes down in half and then recovers to the same value. If all the while you keep taking money periodically to pay for things, you will soon find out that the sequence of returns matters a lot.
The other problem is more subtle, but not less serious: while an aggressive allocation can boost the expected value of your portfolio after a number of years, in many cases it could increase the chances that you won’t make it.
To see why, imagine that in your way to a movie theater someone offers you to play a game. For $10 you can flip a coin and receive $30 on heads and nothing on tails. The average outcome of this game is $15 (half a chance of $30, plus half a chance of $0). This sounds like a great deal - pay $10 and expect to get an average of $15. But you go from being totally able to pay for the movie ticket to having a 50% chance of not being able to. Your expected wealth went from $10 to $15, but your risk has gone up too.
This simple example also highlights the importance of not losing sight of your goals. In this example all you want to do is to buy a $10 movie ticket. Gambling that certainty to increase your wealth to $30, even though the expected value of this gamble is exceedingly favorable, is an action that has nothing to do with your original goal – namely, watch the movie. Even if you win, you will just buy the $10 ticket and have an extra $20 bill that serves no purpose. Why gamble it?
The equivalent example would be this: imagine that you have saved enough to support your retirement spending needs with a portfolio invested in US Treasury bills. However, your stockbroker convinces you that you can boost your returns with a 60%/40% mix of stocks and bonds. While given historical returns this should give you, on average, a higher value at the end, you are also gambling your retirement. While it is reasonable to expect that in the end your portfolio will be larger, your real goal is to have enough to pay for your lifestyle. Assuming that you have no bequest motive, gambling in order to pad your portfolio can increase your chances of ruin.
It is very easy to make this mistake. Intuitively, it seems to make sense that choosing an investment with a higher expected return than US Treasury bills - like stocks - is a good thing because you would have better chances of building an “extra cushion” just in case. It is also partly consistent with early academic work, which defines people as “rational” insofar as they always prefer more wealth to less. But higher potential wealth without a purpose and subject to risk is not directly comparable with less wealth that satisfies a clearly defined goal with less risk.
The graphs below are from our retirement model (we skipped many details so they are not to be used for investment advice). The one on the left shows the chances that a portfolio invested in a fixed-return asset would run dry after 30 years for different withdrawal rates. As long as less than 3% per year is taken out of the portfolio, success is assured. Take 3% or more, and ruin is guaranteed. But if 75% of the portfolio is invested in the stock market (below, right) the portfolio faces a 19% chance of ruin even if the withdrawal rate is only 2.5%. Although the expected end value of the portfolio has gone up (not shown in this graph), the chances of ending up with less than zero have also increased.
On the other hand, a person who has absolutely no chance of making it with the fixed asset can boost the probability of success by adding stocks. If the required draw was 3%, for example, a 75% allocation to the stock market can reduce the chance of ruin from complete certainty to just 37%. For a 3.5% withdrawal rate, the chance of ruin has gone down from certainty to 56%.
These are better outcomes. However, they can be improved further because a 75% exposure to equities may not be the optimal mix for any given withdrawal rate. For example, for a 3% withdrawal rate, reducing the equity exposure to 70% lowers the probability of ruin from 37% to 34%. For a 3.5% withdrawal rate, increasing the equity exposure to 80% lowers the chance of default from 56% to 51%.
The conclusion is that some retirees have no reason to invest in risky assets like the stock market if their spending needs are lower than a certain breakeven. For others, some exposure to the stock market is necessary but the proportion of stocks and bonds depends on their particular profile.
There are vast implications to all of this. Does it make sense, for example, to measure the success of a retirement portfolio by comparing it to a benchmark like the S&P500, if there is no need to match it or beat it? What is the most realistic fixed-return asset available that can provide decent guaranteed returns? And why would exposure to the volatile market component be US Blue Chips or high-dividend stocks, as is often recommended? Can a diversified portfolio with lower volatility provide a better solution, even at the expense of returns?
We will explore these topics in more depth in an upcoming paper. But the inescapable conclusion is that the 4% rule, especially when paired with an arbitrary mix of stocks and bonds like 60%/40% can not only increase your chances of failure but also inefficiently fund surpluses that are not part of your goals.
This conclusion might not be as catchy as the 4% rule but it is crucial to understand. The key to designing a retirement investment strategy that makes sense lies in a realistic assessment and quantification of your personal goals, of the available investment choices for your portfolio, and of your real money (not percentages) spending needs.
