Roger Ibbotson and Larry Siegel spent 100 years of stock and bond data building a new book, and found that before 1926 the case for owning stocks over bonds barely existed at all.
Everyone assumes the US market's long record of beating bonds is simply how markets work. Ibbotson and Siegel's own data says otherwise: extend the record back to 1793, and for long stretches the two asset classes paid about the same.
"If you go back before 1926, let's say 1793 to 1926, the stock market did about as well as it did in the 20th century. But the bond market did much better, and so the equity risk premium, stocks minus bonds, was low. And so over some periods it was zero."
Ibbotson is Professor Emeritus of Finance at Yale School of Management and chairman of Zebra Capital Management, and built the original Ibbotson Associates stock and bond indices 50 years ago; Siegel is the Larry P. Brinson Director of Research Emeritus at the CFA Institute Research Foundation and co-author of the book behind this conversation, Exponential Wealth: Centuries of Stock and Bond Returns.
I listened to the full interview so you can skip it. 35 minutes of audio, 12 minutes of reading.
Here are the 10 lessons that matter.
👤 Guests: Roger G. Ibbotson, Professor Emeritus of Finance at Yale School of Management and chairman of Zebra Capital Management, and Laurence B. Siegel, Larry P. Brinson Director of Research Emeritus at the CFA Institute Research Foundation
🎙️ Host: Lotta Moberg, co-founder and Head of Investment Research and Strategy at Wealth Horizons and a trustee of the CFA Institute Research Foundation
📰 Published: 10 September 2026
🔴 YouTube | 🔗 Episode page | ⏱️ 35 min | ✅ Time saved: 23 min
Key Takeaways
Before 1926, US stocks barely beat bonds at all
From 1793 to 1926 the equity risk premium was low, and zero in some stretches, because bond yields fell as the US went from emerging to developed market
The small-cap premium has mostly disappeared over the last 50 years
It worked in the 1970s and 1980s, before index funds repriced small stocks upward
The US market's own market cap is at close to a historic concentration high
A handful of technology-edge companies now carry an outsized share of the cap-weighted index
Market volatility has been falling for decades, and the new Fed chair thinks that is a problem
Kevin Warsh reportedly wants the Fed to stop talking markets into artificially calm conditions
SpaceX's IPO pop said nothing about its long-term prospects, because 95% of its stock was old, not new
Ibbotson called it publicly: a great short-term trade and a terrible long-term one
A forecast built only on US history is probably too optimistic
The book's 2050 forecast blends US risk with lower global returns to correct for that bias
Going back to 1793, Ed McQuarrie found a fuller record than Ibbotson's own team ever had
Ibbotson used McQuarrie's data instead of his own where it was better
Home bias is a form of survivorship bias, seen clearest in Argentina in 1900
It looked as promising as the US did at the time, and it never delivered
Bonds are priced almost entirely off math; stocks are priced off statistics and noise
Every bond is priced relative to another bond; a stock's fate depends on what the company actually does
Emerging markets look like a bad long-run bet only because the losers get relabeled "emerging" after the fact
1. Fifty Years Building It
Ibbotson opened with the history of how this data came to exist at all. He and Rex Sinquefield built the first stock and bond indices 50 years ago, covering 1926 to 1975 with forecasts out to 2000. Ibbotson Associates published those indices as yearbooks for decades before Ibbotson sold the firm to Morningstar in 2006; Morningstar later bought CRSP, the data source behind the book.
Owning historical market data now means owning licenses, not just running the numbers. All the underlying data is now owned by somebody, Ibbotson said, so building the book required working out licensing deals as much as building the indices themselves
The book pulled in outside co-authors specifically to push the data further back than Ibbotson's own team had gone. Co-author Will Goetzmann found stock quotes from a French water mill dating to 1372, and evidence of priced capital markets in ancient Rome
"So this seems to be a part of human nature. It's not just an invention by Americans in the 20th century."
2. Cap-Weighting Beat Price
Moberg asked what a reader new to indices should know. Ibbotson explained why the original Dow Jones Index — a price-weighted index that simply adds up share prices — gave way to cap-weighting almost everywhere else.
Price-weighting was a computing shortcut, not a design choice. "The main reason you would do that was it was an easy calculation. They didn't have calculators. They had to add things up by hand, so they just price-weighted it"
A price-weighted index rewards a high share price rather than a large company. Splitting a million-dollar company into a million shares instead of one share changes nothing about the business, but it changes the Dow's weighting entirely: "Because then one share would be worth a million as opposed to $1"
Cap-weighting fixes that by multiplying price by share count. The result sums to 100% of the market, with large companies carrying a large share of the index and small companies a small one
3. Slicing the Market by Cap
Cap-weighted indices still have to decide where large-cap ends and small-cap begins, and Ibbotson said the book used the simplest common method: a fixed company count rather than a percentage of total market value.
"So the large cap are the first 300 largest caps now for the last 50 years." That count is arbitrary but standard — S&P and Wilshire draw their own cutoffs the same way
The same count does not work going back further, because there were only about 500 public companies total in the 1920s. CRSP instead splits the historical market into deciles for periods where a fixed count of 300 would swallow most of the market
Morningstar and CRSP's own Vanguard-linked indices take a different cut: their large-cap index holds a variable number of stocks but always represents about 80–85% of total market capitalization
4. A Century of Bond Math
Bond returns turned out to be the harder half of the project. Ibbotson said the starting point for a century of bond returns is the Treasury yield curve, and that the actual bonds trading in the market are never priced at the clean par value a yield curve assumes — so the team had to work backward through forward interest rates to build synthetic yields, work Ibbotson has done with Tom Coleman since the 1980s.
"Every bond is priced relative to another bond." Once you know one bond's price, Ibbotson said, the rest of the curve follows mathematically
Stocks and bonds fail in opposite ways. "The thing is that you don't know what a company is gonna do when you buy the stock. It could prosper, it could fail, it could purchase other companies." Bonds instead carry a fixed schedule of cash flows the issuer is contractually supposed to pay, which is what makes them mathematically tractable where stocks are merely statistical
The book measures the US market monthly but had to lean on Elroy Dimson, Paul Marsh and Mike Staunton's 125-year, 20-country dataset for the rest of the world, which is annual only
5. What Survivorship Hides
Moberg pushed on the book's biggest methodological risk: building 100-plus years of returns data necessarily means picking which countries and companies to measure, and the ones that disappeared along the way are the ones a simple US-only history forgets.
The US looks exceptional partly because it is the country that survived. Russia, China and Austria-Hungary dropped out of investable markets outright; Japan disappeared for a period and came back. Ibbotson: "The US has a survival bias because it not only survived all these wars and so forth, but it also prospered and so it turns out to be the winning country over the last 100 years"
The same bias operates at the level of an individual investor's home market, and Argentina in 1900 is Siegel's example. "It had tremendous prospects, people thought, and that one did not turn out at all"
Moberg framed the standard emerging-markets argument as circular. Investors see emerging markets underperform over a century and conclude they are a poor bet — but a country is still labeled "emerging" a century later specifically because it failed to develop, so that comparison already excludes every market that emerged successfully and stopped being counted as emerging at all
Even inside the US alone there are correction problems: a buyout has to be captured at its final price, and a bankruptcy has to carry its negative return forward even after the stock stops trading, in both directions
6. The Forecast to 2050
The book's own forecast, running out to 2050, is built specifically to avoid the trap the rest of the conversation just described.
"If the US was such a big winning country, the forward expectation is probably too high if you take the US, and you're better off taking the results from the international data." The book blends today's US yield curve and US-measured equity risk with the lower, broader returns Dimson, Marsh and Staunton found across the rest of the world
The method starts with simple extrapolation of the past, adjusted for inflation and the shape of the yield curve, then discounts the US-only result down using the global comparison
7. US Data Back to 1793
Ibbotson had already pushed his own US data back before 1925, working with Will Goetzmann. Co-author Ed McQuarrie went further, using internet-era research tools to recover a fuller record back to the New York Stock Exchange's founding in the 1790s — and the book uses McQuarrie's data instead of Ibbotson's own where the newer work is better.
"If you go back before 1926, let's say 1793 to 1926, the stock market did about as well as it did in the 20th century. But the bond market did much better, and so the equity risk premium, stocks minus bonds, was low. And so over some periods it was zero," Siegel said
The reason is the era itself. The US went from an emerging market to a developed one across that century, and bond yields fell as it did — which handed bondholders a large capital gain that shows up nowhere in a stocks-only history
Ibbotson credited the choice to defer to a co-author's better dataset as ordinary academic practice: "As an academic, I gotta say, we're not trying to always use our own work always. If somebody does something better and they have better techniques, you wanna use that"
8. The Free-Float Problem
SpaceX's public listing came up as a live test case for how index construction actually works, because its structure exposed the difference between a company's full market cap and the free float that index funds can actually buy.
"I said on TV that it had really poor long-term prospects because first of all, 95% of the stock was still held by other people who wanted to sell, and only 5% of it was newly issued here," Ibbotson said, describing the stock's run from 135 to around 110
The 5% that traded was oversubscribed; the 95% held by venture and private equity investors still wanted out. "So I definitely recommended that this was a great short-term investment with the oversubscription but a terrible long-term investment"
Index providers use free float rather than full market cap specifically to avoid this distortion. Including SpaceX's entire capitalization would force index funds to buy far more of the stock than the market could actually supply
The book's own historical indices do not use free float at all — reliable free-float data only goes back about 15 years, so the earlier decades of the dataset measure total market capitalization instead
9. The Vanishing Premium
The small-cap premium — the long-observed tendency for smaller companies to outperform larger ones — is one of the book's more surprising findings, because it has mostly stopped showing up.
"In the last 50 years, there really hasn't been much of a small cap premium, actually," Ibbotson said. It ran hardest in the 1970s and early 1980s, after the 1973–74 recession, and did well again this year, but the multi-decade trend has flattened
Siegel's explanation is that the premium got arbitraged away. Small stocks used to be called secondaries and were mostly ignored in favor of blue chips; once index funds and academic research made the small-cap effect well known, they got repriced upward. "But after they became more fairly priced, in other words, more expensive, then the forward-looking returns on them became more like any other stock"
Siegel still thinks the underlying case is sound, even if the data has not cooperated lately. "So there are reasons why there should be a small cap premium, but there hasn't been lately" — smaller companies carry less available information and are harder to buy, which should command a premium in theory
10. Concentration Rising
The conversation closed on two related findings about the shape of today's market: unusually high concentration in a few large companies, and unusually low volatility.
Cap-weighted indices, long considered the diversified default, are less diversified than they used to be. Ibbotson said "we have almost historically high concentrations in the market," driven largely by a handful of technology-edged companies. He compared it to the 1950s, when General Motors, AT&T, Ford and a handful of large chemical companies carried outsized weight before their relative importance faded as other companies grew
Market volatility has fallen steadily since the middle of World War II, with the exception of shocks like the 1987 crash. Ibbotson said that "the crash of '87 was beyond what could reasonably happen from a probabilistic perspective, and yet it didn't have much impact on the long-term returns"
Ibbotson said Fed Chair Kevin Warsh sees that calm as artificial rather than earned, a byproduct of the Fed talking markets through every eventuality in advance. In Ibbotson's telling, Warsh "wants to be able to react to things that happen and not have artificially low volatility" rather than have the Fed pre-commit to smoothing every shock away
Bonus Insights
Bonds are analyzed mathematically because they have to be. "The thing is that you don't know what a company is gonna do when you buy the stock. It could prosper, it could fail, it could purchase other companies." A bond instead carries a fixed schedule of payments the issuer is contractually obligated to make
Moberg pointed listeners toward the show's earlier conversations with Rob Arnott on fundamental indexing, a different methodology from the cap-weighting and equal-weighting discussed here
Siegel closed by framing this episode as the narrow half of the book. Index construction and the case for total-return data were the focus here; later conversations with the other co-authors will cover the other asset classes, centuries and countries the book contains
Ibbotson and Siegel's bottom line is that a hundred years of US outperformance is the exception rather than the rule, and any forecast built only on that record is probably too optimistic.
Products, Companies & Tools Mentioned
Zebra Capital Management (Where Ibbotson is chairman and director of research)
Yale School of Management (Where Ibbotson is Professor Emeritus of Finance)
CFA Institute Research Foundation (Publisher of the book behind this conversation; Siegel's research-director role and Moberg's trustee seat are both there)
Morningstar (Bought Ibbotson Associates in 2006, then later acquired CRSP, the data source behind the book)
CRSP (The Center for Research in Security Prices, source of the US stock and bond data underlying the book)
SpaceX (The free-float case study — Ibbotson called its long-term prospects poor on TV even as the stock was oversubscribed at listing)
Books & Resources Mentioned
Exponential Wealth: Centuries of Stock and Bond Returns with Ibbotson and Siegel (Part 1) (The first half of this conversation, covering how these returns connect to GDP growth)
The Long-Run Drivers of Stock Returns: Total Payouts and the Real Economy (Related research the show's own notes point to)
Stocks, Bonds, Bills, and Inflation (SBBI): 2020 Summary Edition (Ibbotson's earlier flagship publication, which the book updates and extends)
If this was worth your time, send it to someone who has to have a view on this.
Get the latest market chatter as it happens: