When advisors are measuring the performance of investment portfolios, they obviously need to make some general assumptions. The challenge, of course, is to make the right ones.

For instance, do most people invest money all at once with a single big lump sum? Or do they usually invest gradually—say, by putting money into the market every month, usually taking it out of their paychecks?

It’s obvious that most people do the latter, and yet the standard assumption behind published performance data for mutual funds, stocks and bonds is that the returns are based on lump sum investments.

Let’s drill down further into the most common assumptions behind the total return figures we see every day:

1. We assume that investors make a lump sum investment on the first day of the investment period in which they are being evaluated.
2. We assume that they put no additional money into the investment after the initial lump sum.
3. We assume they withdraw no money.
4. We don’t account for inflation.
5. We don’t account for taxes.
6. We assume that the investors will reinvest their dividends, capital gains and interest and thus cause a change in price (or net asset value).

How realistic are these assumptions? Not very. The vast majority of investors put money into their 401(k) or 403(b) retirement accounts from monthly payroll deductions. Others invest annually or monthly in their IRAs. Thus, the lump sum assumption is irrelevant to how most people are behaving.

In Table 2, a $20,000 lump sum investment was compared to a $1,000 annual investment made at the start of each year (with the eventual total being $20,000). Because a lump sum investment commits all the money at once, it naturally has the advantage if success is measured by the growth of money—at least, over long time periods.

This all means that the assumption we make about how money is invested can dramatically alter our view of historical performance.

This article shows the gap that arises in performance figures when we use the two different assumptions. It also looks at a scenario involving decumulation during retirement.

Table 1 shows the 20-year average annualized percentage returns for three prominent U.S. equity indexes when we assume that investors made a lump sum investment on January 1, 2002, and compare those to the average annualized percentage return for each index when the investor instead made annual investments at the start of each year for 20 years. On one hand we put $20,000 in our initial lump sum investment, and on the other hand put $1,000 per year in our indexes for the annual assumption.

When one examines the performance of these three U.S. equity indexes, the results are interesting. Over 20 years, the winners of the lump sum investment approach are small-cap and mid-cap indexes. But when we make annual investments every January, the three indexes all turned in very similar 20-year returns—though the large-cap index had a modest advantage.

In Table 2, when we measured the growth in dollars as opposed to percentage returns, the results were similar: A lump sum investment in the small-cap and mid-cap index outperformed the large-cap index by roughly $30,000. The dollar differences in the outcome for a $1,000 annual investment over 20 years were relatively insignificant among the three indexes.

This suggests that periodic investing levels the playing field among investments, whereas some indexes do better with lump sums.

Let’s now turn our attention to a very different assumption—namely, a retirement portfolio in which money is being withdrawn at the end of each year. In Table 3’s retirement portfolio scenario, the starting balance on January 1, 2002, was $250,000. At the end of each year, 4% of the portfolio’s end-of-year account balance was withdrawn. It’s clear that these three indexes produced very different experiences for the retiree. 

Over this particular 20-year period from 2002 to 2021, the small-cap U.S. equity index was the best investment option both in terms of its ending account balance and the total amount of money the retiree was able to withdraw. Mid-cap stocks were close behind, and large caps were a distant third.

Of course, single asset classes such as these three indexes do not represent a well-diversified retirement portfolio. But they do show the differences in outcomes when we consider the sequence of returns experienced by a retiree. In this case, the mid-cap and small-cap indexes had much better performance during the first four years of this particular 20-year period (from 2002 to 2005). That better initial sequence-of-returns is the driver of the outcomes observed in Table 3.

In an upcoming piece I’ll tackle the issue of sensible performance benchmarking of broadly diversified portfolios.

Craig L. Israelsen, Ph.D., is an executive-in-residence in the Financial Planning Program at Utah Valley University (UVU) in Orem, Utah. He holds a Ph.D. in Family Resource Management from Brigham Young University (BYU).