Does the S&P 500 always go up?

philipkd1 pts0 comments

Does the S&P 500 Always Go Up? - by Philip Dhingra

Philosophistry

SubscribeSign in

Does the S&P 500 Always Go Up?<br>A look at the index’s “Lost Decades”

Philip Dhingra<br>Aug 19, 2026

Share

Every so often I make it a point of reminding people that the S&P Composite Index (herein: S&P¹) can be flat for a decade or more. Everyone seems surprised when I say this, because the conventional wisdom is that index funds are a no-brainer, long-term investments. But is this true? The conventional counter-wisdom is actually that the S&P has had more than a handful of “lost decades.” But everything I’ve read on the subject is low-quality. So as part of my attempt at regular fact-checking, I want to settle this once and for all.<br>The decades where the S&P was flat for a decade or more

There have been six periods since 1871 when the S&P was flat for a decade or more, adjusted for inflation, and accounting for dividend re-investment. One of these periods overlaps with this century! So we shouldn’t take for granted that the S&P is a safe investment, one where you can “set and forget” your money.<br>Thanks for reading Philosophistry! Subscribe for free to receive new posts and support my work.

Subscribe

The dataset I used for this analysis comes from Robert J. Shiller, the Nobel Prize-winning economist and author of Irrational Exuberance. As an accompaniment to his book, he provides a continuously updated spreadsheet with historical S&P prices, as well as dividend, inflation, and other numbers such as total real returns. With some creative spreadsheeting magic, I identified six periods where if you had bought the S&P, you would have had to wait more than 10 years before getting a return on investment (ROI):<br>1901–1903 (20-year wait)

1904–1913 (17-year wait)

1928–1931 (20-year wait)

1936–1937 (13-year wait)

1961–1973 (19-year wait)

1996–2001 (14-year wait)

It’s easy to make excuses for various periods. The 1920s and 1930s were the Great Depression. And the 1960s and 1970s spanned the Vietnam wars, a tumultuous time for America. But consider the dot-com crash. If you had bought the S&P in August of 2001 at $1179, which was already far off from the dot-com peak of $1553 on March 24, 2000, you would have had to wait 10 years to get your money back. 10 years would have been August of 2011, well after the Great Recession. Perhaps the timing of these economic setbacks represents an unusual streak of bad luck, but COVID came just 12 years later.<br>For the purposes of this analysis, I’m defining ROI as how long you’d have to wait before never having to see the same price again. There are some periods where if you had bought, the index would go up for a few months or years, then come back down—but stay down—usually for a long time.² As an extreme example, if you had bought at the bottom of the dot-com crash and sold five years later just before the crash of 2008, you’d have seen an increase of around 90%! But if you had held for a little longer, you’d be back below your purchase price.<br>But you can slice and dice the data to tell whatever story you want. My focus is on the “set and forget” strategy. An even better analysis would be to take median household income and track whether regularly investing a fraction of it would have resulted in an annualized return over one’s lifetime. Such an analysis is beyond the scope of this post, but as a sanity check, you can play with the raw price graph at Macrotrends. With just a few clicks, you can easily eyeball that the S&P has had all sorts of flat, multi-year periods, and not just a few years, but ten or more. So we can’t really take for granted that the S&P “always goes up.”<br>Thanks for reading Philosophistry! Subscribe for free to receive new posts and support my work.

Subscribe

[1]: The S&P Composite Index expanded to the “S&P 500“ in 1957, up from 233 stocks in 1926.<br>[2]: I’ve gone back and forth using em dashes in my writing, especially since I was previously accused of writing my Irish Travellers post with AI. That accusation initially got my post removed from Reddit, but the mods quickly reversed their ruling.

Share

Discussion about this post<br>CommentsRestacks

TopLatestDiscussions

No posts

Ready for more?

Subscribe

© 2026 Philip Dhingra · Privacy ∙ Terms ∙ Collection notice<br>Start your SubstackGet the app<br>Substack is the home for great culture

This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts

wait years year periods index subscribe

Related Articles