Nearly every source I've read or heard has said that for long-term investing, the best option is to invest in an ETF based on the S&P 500 Index. The logic is that, over the course of many decades, there are almost no investment strategies that have proven to yield a higher average total return (subtracting fees and taxes) than the S&P 500.

As far as I'm aware, the S&P 500 index is mathematically quite simple - straight-forward weighting of the top 500 (or so) stocks based on free-float market capitalization. I find it extremely surprising that out of all the complexity of the market, the best performing strategy (in the long run) would be the S&P 500.

For example, by looking at the numbers, it seems that the S&P 600 index, which follows the same weighting formula using the same class of stocks but only using "small-cap" American stocks, has produced a better long-term total return than the S&P 500. The average total annual return over the years 2004-2020 for the S&P 500 was 10.9%, whereas for the S&P 600 it was 12.9%. I understand the S&P 600 is more volatile due to its small-cap focus, but in the long-term it clearly seems to be the better option, right?

It seems to me that a slight shift of focus to small-cap stocks (e.g. S&P 600) generates better long-term returns than focusing only on large-cap stocks (e.g. S&P 500). To me this seems like a very obvious (and naive?) guess - smaller companies have more room to grow. But if this is true, how does it not directly contradict with the common saying that nothing beats the S&P 500 in the long term?

  • 2
    One problem with the S&P 500 index is that its components are all American companies. Depending on your requirements, global indexes may be more appropriate. Examples of such indexes: FTSE All-World index, FTSE Global All Cap index, MSCI World index, MSCI ACWI (All Country World Index), etc.
    – Flux
    Commented Jul 4, 2021 at 6:23
  • 2
    "The average total annual return over the years 2004-2020 ..." is an example of cherry picking to prove a point. You'll need to randomly select multiple sets of beginning and end points to have any validity in your comparison. Commented Jul 4, 2021 at 13:44
  • 1
    @BeginnerBiker Well, I thought that 17 years was at least a moderately decent time-frame to pick up a broad overall trend, smoothing out shorter term volatility. I have read elsewhere that smaller cap stock have better long-term returns, a point which I believe is demonstrated in the exact same way as the average 10% annual returns of the S&P 500 is. Basically, I felt that my point was as cherry-picked as the phrase "the S&P 500 will likely give about 10% annual returns over a sufficiently long time frame, e.g. over 30 years". Commented Jul 4, 2021 at 22:03

3 Answers 3


It is of course not true that nothing beats the S&P 500 in the long term. There's certainly nothing magical about the number 500 and it may be that a slightly different allocation would be better, even over the long term.

In practice, "nothing beats the S&P 500" is used as a short hand for "nothing beats the market average". The S&P 500 is used as a proxy for the market average because it contains roughly 80% of the total US stock market value. But it's not a perfect representation of the market. In fact, some index-fund advocates have begun suggesting it's better to invest in a "total stock market" fund that tries to buy all tradable US stocks, or a "total world stock market" fund that tries to buy all tradable stocks in the whole world. These are just different variations on the same idea, which is to approximate the market average.

Another important reason that the S&P 500 specifically is recommended is that, because it is well-known, index funds that track it are almost universally available. Almost any mutual fund provider will have an S&P 500 fund. In this sense "just buy the S&P 500" is good advice, but is again approximate advice. It means that if you're faced with a huge array of fund options (for instance, when signing up for a 401k plan) you can just look for the S&P 500 index fund and ignore the rest. This won't always be the absolute best choice but it will usually be very close and will be substantially easier than carefully checking all the other options to see if there's something better.

So basically, the answer is that the S&P 500 isn't "the best" conceivable vehicle, but it's the best one that is widely available and better ones are probably not that much better. So "buy the S&P 500" shouldn't be thought of as a precise law of nature, but rather a rule of thumb. For most people that rule of thumb is a good policy, not because it provides the absolute best returns but because it's a way to make a very good choice with essentially zero thought or effort required.


In investing:

Perfect is the enemy of good.

Investing is not really about precision and eking out every last basis point of gain. It is about coming up with a good asset allocation and trying to avoid doing something dumb that will cause you to lose many years of gains in an instant.

People should be diversified in their investments. You will likely have at least some bonds in your portfolio.

Let's assume for the sake of discussion that the S&P 600 is clearly a better investment choice than the S&P 500.

With that assumption, the following two portfolios might have similar risk/reward:

  • 62% S&P 500 and 38% bonds
  • 60% S&P 600 and 40% bonds

In the grand scheme of things, you will get a similar benefit in choosing S&P 600 over the S&P 500 as you would with simply slightly increasing your ratio of stocks to bonds.

There are many ways to create a good investment portfolio and you could use either the S&P 500 or the S&P 600, and it just isn't that important which one you choose.

  • Thanks for the response. You say "it just isn't that important which one you choose", but isn't the difference between 10% and 12% annual return over the course of 35 years (the time frame I'm looking at) quite big? After 35 years, with the 12% annual return you would make nearly double the amount on an initial investment, as compared 10% annual return. I suppose this consideration is a bit idealistic, but to me it seems as idealistic as the common advice about investing in the S&P 500 and expecting approximately 10% annual returns (long-term average). Commented Jul 4, 2021 at 22:15
  • 2
    @Arturo don Juan: "Past performance is not a guarantee of future results." That one index performed better over one past period that you chose doesn't mean that's going to be the case in the future. You might want to consider splitting your investment between several funds.
    – jamesqf
    Commented Jul 5, 2021 at 4:10
  • @ArturodonJuan, yes, but you get the same increased return by adjusting your stock/bond ratio. In the big picture, it doesn't matter which one you do.
    – minou
    Commented Jul 5, 2021 at 13:13
  • @jamesqf the same is true of the S&P 500, as I'm sure many have realized since you posted your comment Commented Nov 23, 2022 at 12:14

I understand the S&P 600 is more volatile due to its small-cap focus, but in the long-term it clearly seems to be the better option, right?

The key word is risk-adjusted returns. I didn't run the numbers but I'm willing to bet that the risk-adjusted performance (e.g. Sharpe ratio) of the S&P 500 beats the S&P 600.

Additionally, the S&P 500 is often-times used as a proxy for the "market" as it's one of the world's most-followed benchmarks.

If the S&P 500 (VOO, IVV, SPY, etc.) isn't broad enough for you, you could look at the Vanguard Total Stock Market Index Fund ETF (VTI) but you'll find that they're closely correlated. Alternatively, you could look at something like the MSCI World Index.

If you care about absolute returns rather than risk-adjusted returns, there are many hedge funds, alternative investments, single stocks, etc. that beat the S&P 500. You just have to get lucky once.


I'm not sure how you get to 10.9% vs 12.9%. Assuming it's correct, make sure it's an apples to apples comparison. For instance, are you talking about price returns or total returns, including dividend reinvestment?

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .