• Basic Membership Access to Journal, Shadow Stock portfolio & monthly updates for stock screens
  • A+ Investor Full access to AAII.com, Journal, and a power data suite for investment ideas
  • AAII Platinum Full access to all of AAII's model portfolios and premium services at a significant discount
  • Dividend Investing Newsletter and model portfolio that focuses on dividend growth
  • AAII Growth Investing Find secular growth stocks poised for outperformance regardless of macroeconomic conditions.
  • Retirement Investing Discover trusted strategies for every stage of retirement. Get actionable insights and resources for smart retirement planning.
  • Sentiment Investing A research-driven newsletter with 10+ signals decoding sentiment across retail and institutional behavior.
  • Stock Superstars Report Get stock ideas inspired by the market's leading professionals
  • VMQ Stocks For the value investor looking to harness academic research in an actively managed portfolio
  • Bits and Bips Newsletter delivering insights on linking traditional finance, macro trends and cryptocurrency opportunities to help investors build stronger, more diversified portfolios.
All premium products come with free basic membership.
Explore All Membership Options
Join Now
Asset Allocation Portfolio Strategies

12 Hidden Lessons From Investment History

Investors who are willing to learn from nearly a century of data can discover some important takeaways.

by Paul Merriman | October 2024

Paul Merriman leads a class in AAII's new Essential Investing Video Course. Go to https://www.aaii.com/ves for more information and to subscribe.

  • Analysis of 96 years of U.S. stock market data
  • Insights into the benefits of diversification across asset classes
  • Importance of sticking to a long-term investing strategy

Once upon a time, investing for the future was pretty much a gamble, without much meaningful data to give you any clue as to what might be coming down the pike. Now we are blessed with detailed and reasonably accurate data going back 96 years to 1928.

It’s unfortunately (and inconveniently) true that in the next year, next five years or next decade, anything can happen. Making short- and intermediate-term predictions—in spite of Wall Street’s awesome computing power, hard work and sharp minds—is still something of a crapshoot.

But investors who are willing to learn from nearly a century of data can discover some important lessons that have been mostly hidden.

In this article, I show you the data and share 10 lessons based on it. And for readers who make it to the end, I throw in a couple of bonus lessons at no extra charge.

Before we dive into the data (although I have described it as “hidden,” it’s actually pretty colorful), let me tell you what it won’t help you with.

If your goal is to reliably get rich quickly, this article won’t help. If anything, this data will show you how unlikely that is. But if your goal is for your investments to do well enough that you can stay the course during whatever life span you have left, this will point you to a path forward.

Relative Performance of Four U.S. Equity Asset Classes

Figure 1 on the next page shows four U.S. equity asset classes (as well as an equally weighted combination of all four) and their relative return rankings each year from 1928 through 2023. Chances are you have never seen anything quite like this.

Figure 1 U.S. Asset Class Indexes and Four-Fund Combo: Relative Return Ranking (1928–2023)

I’d like to start by telling you where it came from.

A few years ago, I asked Daryl Bahls, the director of analytics at The Merriman Financial Education Foundation, to come up with a visual tool that would let investors quickly and easily see what’s important about nearly a century of investing data. The colorful quilt chart is the result.

The chart tracks year-by-year returns from 1928 through 2023 for the equity asset classes that are owned in one form or another by the great majority of investors. As you probably know, those asset classes are large-cap blend stocks (represented by the S&P 500 index), large-cap value stocks (labeled LCV on the chart), small-cap blend stocks (SCB) and small-cap value stocks (SCV).

Bahls went one step further and delighted me by cleverly adding the combination of those four asset classes. As we shall see, this grouping, which I sometimes refer to as the Four-Fund Strategy, might be the most important lesson of all to emerge from this data.

I call this a strategy because it can be a master plan for investing in U.S. equities. It’s also a portfolio in the sense that, when implemented, it becomes something real: a collection of four funds. In this article, I refer to it as a combination, or combo, which can satisfy either of those meanings.

Each of the 480 colored boxes in Figure 1 shows the one-year return for one of these investments. For each year, they are presented in order of return, from best to worst.

The colors make it easy to follow. For example, you can track the performance of the S&P 500 by looking at the red boxes or that of small-cap blends by looking at the green ones.

Lessons From the Historical Returns of Stocks

The lessons that follow are designed to reinforce a fundamental truth you’ve heard before: Sticking with your plan through thick and thin isn’t necessarily easy, but it is essential. One key is to manage your expectations. That’s much easier if those expectations are based on a good understanding of market history.

At first glance, the quilt chart looks like a mishmash of random colors and numbers. That is one of the key points I hope you remember.

Lesson #1: There Isn’t a Consistent Winner

No single asset class was ever “best” most of the time.

As you can easily see, the popular S&P 500, representing large-cap blend stocks, was all over the map. In 27 years, it had the best return; in 39 years, the worst. Only seven times was its performance squarely in the middle of the pack.

Yet, the S&P 500 has had a few “winning streaks” that undoubtedly convinced many investors it was all they needed. The four years of 2017 through 2020 was one such period. On the other hand, the S&P 500 was the worst performer for seven years in a row between 2000 and 2006 and for six straight years in the 1940s.

Bahls produced a second chart, shown in Figure 2. This chart shows how much of the time each asset class (and the Four-Fund Combo) was at the bottom of the pack (rank 5), at the top of the pack (rank 1) and in the broad middle (ranks 2, 3 and 4).

Here’s a hint: For your peace of mind and your ability to stay the course, the broad middle is a much better place to be than switching back and forth between best and worst asset classes. The broad middle is also where the Four-Fund Combo really shines.

Figure 2 U.S. Asset Classes and Four-Fund Combo Return Rank Frequency (1928–2023) Total number of times, and the percentage, that individual returns were in each quintile rank

Lesson #2: Small-Cap Value Stocks Have Been Volatile

Like the S&P 500, small-cap value stocks spent quite a few years either on the top or bottom of the return rankings. This asset class ranked in the best quintile 38% of the time and in the worst quintile 25% of the time. This gave investors plenty of opportunities to love it (in 36 individual years) or hate it (in 24 years).

The good years for small-cap value weren’t always sprinkled at random through the decades. They came in streaks that gave investors opportunities to be staunch believers in this asset class (see 2000 through 2004 in Figure 1).

Yet, that belief and enthusiasm were shattered nine separate times when small-cap value abruptly fell from the best performer to the worst. The years 1950 and 1951 and the years 1968 and 1969 are stunning examples of these reversals. (The same thing happened to the S&P 500 six times, though nobody was complaining in 1958 when the index gained “only” 43.4% after losing 10.8% in 1957.)

Despite their superior long-term returns, small-cap value stocks are not the place to invest if you’re seeking the comfort of low volatility.

Lesson #3: Relative Performance Has Been Random

In any particular year or decade, the best and worst performers were essentially random. That stark reality leads directly to the next lesson.

Lesson #4: The Data Is Noisy

You may find some meaningful patterns in these returns, but they won’t be much help in predicting anything.

In 24 individual years, large-cap stocks were clearly in favor, as the S&P 500 and large-cap value took the top two spots. Yet numerous times, they simply ranked favorably because they lost less than small-cap stocks.

In contrast, small-cap stocks (blue and green boxes) held the top two spots in 32 individual years. In only two of those years (2002 and 2008) were small-cap stocks there because of lower losses than large-cap stocks.

In years when large- and small-cap blend funds outperformed value stocks, it was because growth was in favor with investors. The opposite was true in years when value stocks outshone blend funds. So, which is better:

  • Large-cap or small-cap?
  • Growth or value?

There’s no right answer to either question. If you trace through the chart in Figure 1 following the color of the boxes, you can see that these winners and losers changed places much too often to conclude that either one was better or worse.

Lesson #5: Returns Have Varied Greatly Annually

In any single year, the difference between the top and bottom performer can be dramatic.

The spread between the highest and lowest returns was 70.7 percentage points in 1933. In 1943, the spread was 52.7 percentage points; in 1967, it was 55.1 percentage points.

Small-cap value outperformed the S&P 500 by 28.8 percentage points in 2000. In 2001, small-cap value led by 40.3 percentage points. It outperformed by 38.4 percentage points in 2003.

Lesson #6: Stocks, as a Group, Move Up and Down Together

If you really want to find a pattern that is predictable, here’s one: In most years (81 of the 96 in this quilt chart), the best and worst performers moved in the same direction, either up or down.

So, a good year is likely to be good across the board. A bad year is likely to be bad across the board.

Lesson #7: Stocks Gained 71% of the Time

This is another bit of good news from 96 years of data: In the majority of cases, the trend was up, not down. In 68 years (about 71% of the time), at least three of the four major U.S. asset classes had positive returns.

12 Hidden Lessons From Investment History Video

We think you’d like this related webinar! Which Is the Best 1-, 2-, 3- and 4-Fund Strategy?

Lesson #8: The Four-Fund Combo Was Mostly in the Middle

I mentioned earlier that Bahls surprised me by showing our four-fund combination of large-cap blend, large-cap value, small-cap blend and small-cap value asset classes (computed assuming annual rebalancing). You’ll see this in the yellow boxes throughout Figure 1.

Looking at the quilt chart, I immediately noticed how dependable and unexciting—two traits that should be enticing to long-term investors—this combination was. Each individual asset class moved up and down freely, but the Four-Fund Combo spent most of its time in the middle. In fact, it was in the middle 78% of the time.

Lesson #9: The S&P 500 Hasn’t Lived Up to Its Reputation

Although many investors tend to regard the S&P 500 as reliable and comfortable, Figures 1 and 2 show that’s just not true over the long haul. Compared with the Four-Fund Combo, the S&P 500 has been overly dramatic. Plus, among the four major U.S. asset classes, the index representing large-cap blend stocks has given investors the lowest long-term performance.

The numbers tell a powerful story. The S&P 500’s very long-term compound annual growth rate (CAGR) is 10.0%. For the Four-Fund Combo, the number is 11.8%.

That is a huge deal, and here’s why: I have said for years (See “Winning the Battle Against Investment Fees and Biases,” in the May 2024 AAII Journal) that an increase of 0.5 percentage points of return can be worth $1 million over a lifetime. Here, we have a portfolio that provided more than three times that much additional return: 1.8 percentage points.

Lesson #10: Diversification Keeps You Exposed to the Top Performer

Here’s something else that’s also appealing about the four-fund combination: Every year, it guarantees that you’ll have 25% of your equity portfolio in whatever happens to be the highest-performing asset class.

With this combination, you don’t have to give up those bragging rights. And no matter what’s happening, you will never be in last place.

Bonus Lesson A: Even Without Rebalancing, There’s a Big Advantage

This doesn’t come from the tables, but it is important to note. If you hold the four-fund combination inside an individual retirement account (IRA) or a 401(k) plan, the annual rebalancing does not produce any unpleasant side effects. But in a taxable account, the annual sales and purchases will generate taxable events that could affect your annual tax bill.

Without any rebalancing, a lump-sum investment split equally among the four major U.S. asset classes had a slightly lower long-term return advantage over the S&P 500: 1.5 percentage points instead of 1.8. However, 1.5 percentage points is still a huge advantage.

Bonus Lesson B: Diversification Improves a Portfolio of Equities

Here’s why all this is so important. The academic research is unanimous on two points. First, diversification is the best way to improve a portfolio of equities. Second, the most effective way to diversify is by adding asset classes (not just more stocks with similar characteristics).

Figure 2 makes it clear that, over the decades, any one of the other major U.S. asset classes (large-cap value, small-cap blend, small-cap value) would have been a worthwhile addition to the S&P 500. The Four-Fund Combo, as I’ve said, provided a higher overall return along with lower interim losses. To my mind, that’s a serious win-win result.

Another point I want to address is the relationship between risk and return.

The charts clearly show that the S&P 500 has the lowest long-term return among the four major asset classes under review here. If it’s true that lower risks go together with lower returns, then you would expect the index to be less risky than the others. And yet, the beloved S&P 500 had the worst return in 39 years.

What of the Four-Fund Combo and its significantly higher returns? It was at the bottom of the pack exactly never. If there’s any bottom-line “magic” that results from putting the four funds together, it’s the combination of higher long-term returns and lower year-by-year volatility.

Making Use of These Lessons

Finally, there is the question of what (if anything) you should do about all this. While there’s no guarantee about the outcome of any choice you make, these lessons certainly give us some guidance.

Lesson #1 underscores what you already knew: Diversification in equity investments is a good deal, providing reliable benefits at little extra cost in time or money.

Lessons #6 and #7 teach us three encouraging things regarding these four major asset classes. In most years, the stock market moves up, not down. There’s usually nowhere to hide in a bad market year. Finally, in a good market year, any choice is likely to be profitable.

So, if you limit your equity investments to these asset classes and faithfully stay the course over many years, you’ll probably do at least okay—as long as you can avoid the temptation to chase recent performance by jumping from one asset class to another.

Lessons #8 and #9 point to a four-part combination strategy that’s relatively boring, making it easier to stay the course and delivering lots more money in the long run. For any serious investor, that outcome should certainly be worth studying a bit of colorful history. 

Discussion

JAMES L from TN posted almost 2 years ago:

A rich compendium of historical data that is very well displayed, illustrated and discussed for the chosen categories, but which does not include several other of the multi -factor categories - my thanks to the author for the fine work presented to us. Take the data and try to improve on it if you can, the exercise alone will be worth the effort, while the data presented gives a strong foundation in major subsets used as a foundation. My kind of article (among many)!


BOBBY M from NJ posted almost 2 years ago:

Excellent analysis. If possible, can mid cap blend and mid cap value be added in the next go around? The conclusions will certainly be the same but I like allocating across these 6 asset classes instead of 4. Thanks.


BARRY J from TX posted almost 2 years ago:

#1 All AAIIers should be thankful that we have access to the creative mind of Paul Merriman and his Four Funds portfolio, which is well known, widely used, and emulated in various forms. #2 AAII’s own Charles Rotblut’s retirement portfolio is a similar model that he uses to demonstrate the power of rebalancing. #3 Vanguard Bogleheads advocate the use of a similar assortment of low-cost ETFs and constantly debate which configuration of low-cost ETFs to use for each asset class. #4 Everyone tries to improve on this portfolio in one way or another – +/- relative allocations, +/- diversifications through FI or Ex-US tilts, + “boxes,” use all 13 market segments, etc. I have seen lists similar to this 4FundPF that ranging from 1 fund (a US Total Stock Market ETF) to over 20 funds, but no one provides the data Merriman does to prove their relative superiority. #5 Liz Ann Sonders of SCHW uses a similar “quilt” graphic to show the relative performance of the 11 GCIS sectors over time which she rolls out to relate movements to economic and monetary factors to explain similar relative stock sector rotations. #6 “Combo meal” approaches all work for the same reason – (as this article repeated reminds us) the power of asset class diversification offsets market swings (covariance or “beta”) in asset classes. All brokers report market “beta” for ETFs/funds. This was the 1962 principle that earned the late Harry Markowitz Nobel recognition in 1990. #7 4FundPF long-term returns are tilted to favor large caps; 2 of the 4 funds are large caps assets – Lg-Cap Blend and Lg-Cap Value and two are Small Caps. #8 Merriman does not state why his 4Fund Big Boy “combo meal" ignores the power of mid-caps growth stocks. #9 One caution: The stacked quilt analogy may innocuously convey a false sense of cozy, warm, “feel good” comfort, but stacked colorful quilts do not convey the magnitudes of the very wide range of returns for each box over the years (due to market beta again). As the article says, asset class variances are very wide. Merriman simplifies and reinforces the main market mantras that the market dynamics pictured here MAY create optical illusions that your returns are temporarily “at risk” of decreasing in total value. #10 I would like to see the MVO data for each year and the ranges, means, medians and SDs for each of the 4 asset classes. #11 The closest data I have seen is from Wiliam Berstein, one of Merriman’s “Truth Tellers.” That data would provide a rich opportunity for statistical analysis that would show how widely SPX varies as the article says. The spreads (variances) would be frightening. #12 One small quibble. The statement, “Every year, it guarantees that you’ll have 25% of your equity PF in whatever happens to be the highest-performing asset class” is not a completely accurate statement because the 4FundPF does not include any FI asset classes (ST, MID, LT, bonds). #13 This is a great article to help each of us THINK (as Watson Sr. demanded of all this employees) about WHY we invest and HOW we invest. Thank you, Paul for access to your thinking. Good on ya.


ROBERT A from NC posted almost 2 years ago:

Another outstanding analysis by Paul Merriman. Kids, bonds were left out of this analysis apparently for a very important reason: they're a HUGE drag on long-term asset growth. Start early and stick with equities.


JIM L from MI posted almost 2 years ago:

Thanks to Paul Merriman for your many thoughtful, useful articles backed by deep research and analysis of data beyond the reach of individual investors. Like many of us, Paul's views have evolved with time and experience in the market: he's come a long way from when I first encountered him with his book 40 years ago on moving averages! The market humbles us all (well most of us!) and offers instruction for those with a mind willing to learn.


EDWIN W from PA posted almost 2 years ago:

well, it's not particularly surprising that a combination of four things could never be better than the best of the four or worse than the worst of the four. the only rankings mathmatically possible would be 2nd, 3rd or 4th of the group of 5.


SCOTT B from NC posted almost 2 years ago:

It's hard to take any of these 'lessons' to heart when there's no discussion of how these four equity classes are defined. Accordingly, it's actually impossible to create an investment program from this information - assuming you had the means to buy 2000 stocks over four classes (or maybe 1500 stocks if the value classes are only 250 stocks each - see question below). #1 Are all four classes made up of 500 stocks like the S&P500? #2 This is all small cap and large cap - what happened to mid-cap or micro-cap? #3 On what basis is 'value' defined and is it the same for large and small caps? Does the value metric exclude all stocks that have some kind of negative revenue, income or cash flow period as many growth stocks in particular might? Further, is value determined relative to historic norms for each stock, a stock's industry/sector, or some multi-sector average? #4 How discrete are the non-value stocks from the value stocks in the blend or can they overlap? #5 Is a blend exactly half value and half non-value stocks? Since the S&P500 LCB class is 500 stocks, this would imply the value classes are only 250 stocks. So, are all of the value stocks in a blend class part of the value class? Alternatively, if the value classes are 500 stocks, which half of the value stocks are included in the 500-stock blend classes? Does the S&P500 consist of half value and half non-value stocks? #6 Are all four classes market cap weighted like the S&P500? If so, what are the rules then for constructing these weightings? Or is the S&P500 the only cap weighted class in the study? #7 Is there any screening done to minimize the percent of 'dogs' in the non-value portion of a blend class? How about exclusions based on stock price or liquidity (average number of shares or dollar trading volume)? #8 There were no funds available for the early years of this 'study' but what collection of mutual funds and/or ETFs can be shown to closely mimic the annual class rankings reported since the advent of such funds? A portfolio of four funds would actually be practical for a retail investor to implement.


Hugh P from WA posted almost 2 years ago:

Edwin, Agree that Relative Returns Risk is foregone conclusion. In combination with CAGR, attraction of 4-Fund is returns are not far off from the best, while minimizing withdrawal timing risk. A side benefit could be bolstering your resolve to not bail in down years, since your relative return would not be as bad as the worst choice return.


MICHAEL B from WI posted almost 2 years ago:

Great article, but... Scott B's comments and questions are right on. When I was in my short post-retirement career as a financial advisor, I had a chart from Fidelity which looked at the performance of some asset classes from I believe 1926 to 1998. (Since I am currently on trip, don't have it with me.) What I do remember was that bonds were at the bottom, followed by large cap stocks, followed by small cap stocks, and topped by mid-cap stocks. Since value outperformed growth over that same time period, I have been a little partial to mid-cap value as a significant part of my portfolio.


RICHARD M from AZ posted almost 2 years ago:

Great article. Even though value outperforms growth over the long term, I would really like to see small growth and large growth added to your matrix. Richard M


Peter N from TN posted almost 2 years ago:

As mentioned by others, glaring omissions include large-cap growth and mid-cap growth, bland, and value. All the usual suspects should be tested. Then if the returns of the top four or six portfolio or asset classes are substantially better than the others then say so and focus on those. Failing to acknowledge where other portfolio or asset classes have performed relative to the ones mentioned gives the impression that others were omitted based on subjective preferences. As table 2 in the Israelson article "Striking the Right Balance Between Growth and Value Stocks" shows pretty clearly regarding the dead horse of small cap value since 2004: "He's dead Jim!" regardless of occasional dead cat bounce years. Data since 2000 should be given more weight than earlier data. Data prior to 1970 is probably irrelevant noise.


WARREN L from GA posted almost 2 years ago:

The article does not identify what Funds or ETFs were used to represent the Asset Class Indexes. Without that information an investor has little hope of replicating the performances shown in the quilt chart. Please provide that information.


LISANNE P from OR posted almost 2 years ago:

Agree with Warren L. Please name the specific indexes that were used for SCB, SCV and LCV. It would also be great to know, if we prefer to use an all-market ETF like VTI, how much SCB, SCV and LCV would need to be added to get to the equivalent of the 4 Funds used in the article? (Theoretically, I could calculate this algebraically from the Morningstar style box for VTI, but I’m pretty rusty!)


MARTIN S from OH posted almost 2 years ago:

Re: 12 hidden lessons from investment history, October 2024. I really enjoyed the 12 hidden lessons from investment history. I would like to find three ETFs to build the US 4 fund. I think I and other readers would like you to publish ETF options for each of the four categories, SCV, SCB, LCV. And of course, S&P 500, which is well known.” thank you very much, I’m Marty Schirmer. Long time, reader and lifetime member.


MARTIN S from OH posted almost 2 years ago:

Re: 12 hidden lessons from investment history, October 2024. I really enjoyed the 12 hidden lessons from investment history. I would like to find three ETFs to build the US 4 fund. I think I and other readers would like you to publish ETF options for each of the four categories, SCV, SCB, LCV. And of course, S&P 500, which is well known.” thank you very much, I’m Marty Schirmer. Long time, reader and lifetime member.


MIODRAG D from IA posted almost 2 years ago:

Miodrag D from IA: Dear Paul: This is an excellent paper with a brilliant way to compare U.S Asset Classes Performance and Relative Return Ranking going back 96 years. However, I am afraid that there is redundancy in your selection of asset classes. While selecting Small-cap Value stocks is consistent with your previous research and does make overall sense, selection of Large-Cap value stocks and Small-cap Blend stocks is redundant and does not contribute to the quality of the paper. I would kindly suggest that instead of these two categories you include into comparison International Stocks, so that the 3 categories to compare would be S7P 500, U.S Small-cap Value stocks and International Stocks. John Boggle and Burton Malkiel had very nteresting discussion if International Stocks should be included into portfolio construction to satisfy diversification criteria of modern index based portfolio. Your way of analysis would defintely provide this crucially important answer. One other suggestion: When you construct the three-fun Combo you may consider S&P=55% International=35% and US Small-Cap value=10% in addition to equal weighted Combo of 33% each of these 3 categories. Obviously the issue is that there would not be 96 years of data for International stocks, but even smaller time window would be very much indicative. Again, my congratulations for brilliant paper and my gratitude in advance for the paper that would include these 3 suggested asset classes.


MICHAEL C from MI posted over 1 year ago:

"Once upon a time, investing for the future was pretty much a gamble, without much meaningful data to give you any clue as to what might be coming down the pike. Now we are blessed with detailed and reasonably accurate data going back 96 years to 1928." Assuming the future is entirely predicted by the past, especially with so many things fundamentally changing about the world is hopeful.


You need to log in as a registered AAII user before commenting.
Create an account

Log In

Get your free copy of our special report analyzing the tech stocks most likely to outperform the market.
Download the FREE Report Here: