One of the many advantages of working at a firm with a rich, long history is occasionally stumbling upon interesting things in our archives. We were reminded of this recently while updating the assumptions for our cash flow modeling analysis, which we use most frequently to advise clients on retirement planning. In the archives was a summary document of asset class return assumptions that our founder Roger Gibson created in the 1990s.
This sort of document provides some useful perspective. The scope is truly long-term. Investors over the roughly 25 years since its creation have experienced plenty of economic cycles, geopolitical flare-ups, market downturns, and recoveries. Through it all, we have guided clients toward their most important financial objectives, which for many clients includes a secure and comfortable retirement. Likewise, with this letter, we zoom out from the headlines of today and share some observations from our firm’s three decades of experience in retirement planning.
Some Observations on Retirement Planning
Model vs. Reality
All conversations about retirement planning must begin with a big disclaimer. We do not have a crystal ball and cannot predict the future. We try to make reasonable assumptions that incorporate historical data and our best judgment about the future. We use simulation modeling software to account for the variability in potential investment outcomes. In the end, we have a model that can help us explore the likelihood of a client meeting their financial objectives, given our inputs.
Things change along the way—client objectives shift, we update our return assumptions, etc. That’s why it’s important to refresh our analysis every few years. But another reason to periodically revisit our analysis is that the model does not perfectly represent actual client decisions. One of the most significant differences between model and reality is what tends to happen in a prolonged market downturn. Our models assume no change in spending behavior. In practice, some people may rein in spending during a down market. This difference between assumed and actual spending behavior embeds an element of conservatism in our retirement planning.
Some might assume that a primary role of the financial advisor is to protect clients from spending too much or saving too little. We are more frequently in the role of encouraging clients to spend more on the things that are most important to them. We find that the majority of our clients “underspend” relative to their resources. By this we mean that many clients plan lifestyle expenditures that are on the safe side of what we deem to be an acceptable range.
Financial conservatism is perfectly fine, so long as it doesn’t result in unnecessary sacrifice. If you find yourself stressing about a purchase or some expenses, let us model the long-term impact for you. Our analysis can help you differentiate between the emotional and financial aspects of the decision. Sometimes the right answer is to take the big family vacation, travel more, spend more on the renovation, etc., and you might just need an impartial sounding board with a sharp pencil.
More on Consumption
Research on retirement spending suggests a typical pattern through retirement. Expenses start relatively high early in retirement, then dip in a person’s 70s and 80s, before going higher later in life due to increased health care costs. Researchers refer to a graphical representation of this pattern as the “retirement spending smile”.
We do not have sufficient data to counter this hypothesis, but our anecdotal observations generally do not support it, at least for our client demographic. We find that some clients tend not to adjust their regular distributions that they take from their portfolios, sometimes for years at a time. We generally do not see the dip in spending in one’s 70s. We find that clients tend to enjoy many of the same activities and experiences in their 70s that they did in their 60s. Lastly, we have not observed many instances of high, sustained health expenses far in excess of prior lifestyle expenses.
Rules of Thumb
Some retirement planning rules of thumb can be useful in simplifying a complex topic. An example is the “four percent rule”. The four percent rule comes from research showing that a safe withdrawal rate for a stock-bond portfolio over thirty-year horizons was an initial withdrawal of four percent of one’s portfolio value, followed by increases at the rate of inflation.
Two important caveats apply. The first is that the study supporting this rule of thumb looks at historical time periods and, therefore, historical returns. Our future return assumptions are lower than the historical returns. Consequently, we calculate an initial sustainable withdrawal rate for many investors that is lower than four percent. Second, sustainable withdrawal rates are sensitive to an investor’s time horizon. Relatively long time horizons result in lower sustainable withdrawal rates and vice versa.
While we’re on the topic of time horizon, we will share that we commonly find a mismatch in how our clients think about their investment horizons and how we think about their investment horizons. In short, the relevant investment horizon often is longer than our clients perceive it to be. “Not as long as it once was” is not the same as “not long”. For example, a couple, both of whom are 75, may instinctively feel that their investment horizon is “not long”. Yet, the life expectancy of the survivor of the two is 17 years–close to a two-decade investment and planning horizon. For most purposes, two decades is a long time horizon.
Living With Volatility
A surprising conclusion from our modeling analysis is that some clients in retirement have a similar probability of meeting their objectives if they decided to reduce their allocation to equity investments such as stocks. This conclusion can be a surprise because we rightfully think of stocks as the primary return driver in one’s portfolio. However, stocks also are the primary source of downside risk. So, reducing one’s stock allocation can reduce the negative impact from poor stock returns, albeit with a cost. The cost is a likely lower portfolio value many years out into the future.
As Chad put it, “asset allocation decisions are not necessarily about how much volatility you can tolerate, but rather about how much volatility you want to tolerate.”
The output to any retirement planning analysis is highly sensitive to the portfolio’s assumed rate of return. In our judgment, it is a mistake to blindly extrapolate historical returns when making forward-looking estimates. In developing our return assumptions, we account for current stock market valuations and interest rate levels to develop our best estimate for future returns given current market conditions. Over time, our return estimates have been reasonably accurate.
More on Return Assumptions
Portfolio values and future expected returns tend to be inversely related. For instance, prices for bonds, stocks, and real estate securities have moved lower this year. Our future expected returns for bonds, stocks, and real estate securities have increased.
The exact opposite happened over the prior two years. In that period, investment asset values rose, and future expected returns came down. Picture this relationship as a seesaw with investment portfolio values on one side and portfolio expected return on the other side. All of the attention is on the price changes, which is understandable because they make us feel relatively more wealthy or less wealthy. But the change in future expected returns is equally important.
The stream of future cash distributions that your portfolio can support is a function not only of today’s portfolio value but also of future expected returns. Therefore, the great silver lining to any market pullback, such as the one that we have experienced this year, is that your portfolio’s wherewithal to meet your objectives might change relatively little.
The Most Important Part
The planning conversation often is more valuable than the results or output of our analysis. Articulating the things that you care about the most, and what concerns you the most, has a way of putting everything else into perspective. For us as advisors, the planning conversation allows us to shift focus away from short-term market fluctuations and toward your primary objectives, as well as provide reassurance that your plan remains on track.
The price declines in major asset classes such as bonds, stocks, and real estate securities that began in the first quarter picked up steam in the second quarter. Market prices are adjusting to new expectations of higher inflation, higher interest rates, and a higher probability of an impending recession. None of these changes in expectations are good news for markets in the short run. However, a little perspective may be helpful. From the beginning of 2019 through June 30 of this year, diversified portfolios have produced compound annual percentage returns in the mid-single digits. Two good years (2020 and 2021) and what so far has been a bad year (2022) have come together to produce a reasonable return for the combined period.
Good and bad years tend to offset one another over longer horizons. Consider the course of events in the two-plus decades since Roger’s late 1990s return assumptions. We had the tech meltdown and bear market in the early 2000s, the 9/11 attacks, the Global Financial Crisis in 2008, a roaring recovery in the 2010s, the COVID crisis in 2020, and now the first major inflation scare in 40 years. Each of these market events was significant and yet none of them was likely to have derailed the retirement plans of an investor who remained disciplined and diversified.
With newly higher bond yields and lower stock prices this year, we are making some modest changes to our long-term planning assumptions. If you would like to see how this year’s market movements impact your plan, we encourage you to reach out to your advisor. We are grateful for the opportunity to serve you, and we look forward to our next conversation.
- Loosely defined, a “safe” withdrawal rate is an amount that an investor can reasonably expect to withdraw annually through retirement without depleting his or her portfolio.
- Bengen, William P. Determining Withdrawal Rates Using Historical Data. Journal of Financial Planning. October 1994.
- A caveat to this statement involves time horizon and cash flows. Investors with long time horizons, who are regularly contributing to their portfolios, will be better in the off in the long run as a result of a price decline. Shorter-time horizon investors and those who are taking distributions from their portfolios may be relatively worse off after a price decline.