FinUties 101 · History · Beginner
How to read financial history without being fooled
Old charts and famous numbers mislead more often than they inform. Six checks, from inflation and dividends to survivorship and base rates, turn a historical figure into something you can actually use.
In this lesson
- Tell a price-only figure from a total-return figure, and a nominal one from a real one
- Spot survivorship bias in a long-run return or a list of winners
- Use base rates instead of one memorable episode
Financial history is the only laboratory markets have. Nobody can rerun 1929 with different rules. But the record is full of traps, and most famous numbers hide at least one of them. Before you lean on a historical figure, run it through six checks.
1. Price or total return?
Most index charts show price only. Shares also pay dividends, and over long periods dividends have made up a large part of what shareholders earned. A chart that shows an index "going nowhere for 16 years" can hide a positive return once dividends are counted. Always ask which one you are looking at.
2. Nominal or real?
A figure in money of the day is nominal. A figure adjusted for inflation is real. In a decade of high inflation, a market can rise in nominal terms while the buying power of the investment falls. Long-run comparisons only make sense in real terms. Robert Shiller's public data set, used by many researchers, gives both price and dividends with an inflation index back to the 1870s.
3. Which index, which market?
"The market fell 89%" is true for one index, in one country, measured from one day to another. A broader index, another country or a different end date can tell a different story. Check the index, the currency it is measured in, and the exact start and end dates.
4. Who is missing?
Long-run studies of "global returns" mostly measure markets that still exist. Markets that were closed by war or revolution, and companies that failed, drop out of later lists. This is survivorship bias: the record that survives looks better than the average experience. Dimson, Marsh and Staunton built their long-run database partly to correct for it.
5. One story or a base rate?
A famous crash is memorable, which makes it feel typical. A better question is how often something happened out of all the times it could have. "How many 20% falls turned into 50% falls?" is a base-rate question, and it is much more useful than "what happened in 2008?" Reinhart and Rogoff's study of eight centuries of crises is built on exactly this idea.
6. Then, or now?
Rules change after every crisis: deposit insurance, central bank tools, circuit breakers, capital requirements. A pattern from a world without those rules may not repeat in one with them, and new rules create new weak spots. Ask what was different about the system at the time.
Common mistakes
- Comparing a nominal figure from the 1970s with a real figure from today.
- Reading a log-scale chart as if it were linear, or the other way round.
- Treating the recovery time of one crash as the normal recovery time.
- Quoting a number without its source. On FinUties every figure in the history collection carries its source and how it was checked.
In practice
When you read "markets always recover", translate it into checks: which market, price or total return, real or nominal, how long, and how many markets did not. The answer is usually still useful, but narrower and more honest.
Check yourself
An index is at the same level as ten years ago. Did a holder earn nothing?
Not necessarily. The index level is price only. Dividends paid along the way add to the return, and inflation subtracts from it. You need the total-return, inflation-adjusted figure to know.
Why can the long-run return of 'the stock market' overstate what investors got?
Because the markets and companies we measure today are the ones that survived. Markets that closed or companies that failed drop out of the record, so the surviving history looks better than the average experience.
A crash has been followed by a recovery within two years five times. Does that tell you the next one will recover as fast?
No. Five cases is a small sample, and the cases were not identical. History gives a range of outcomes and how often each happened, not a promise about the next one.
Key terms
Sources
- Online data: U.S. stock markets 1871-present and CAPE ratio. Robert J. Shiller, Yale University. Accessed 8 Nov 2026
- This Time Is Different: Eight Centuries of Financial Folly. Princeton University Press (Carmen M. Reinhart, Kenneth S. Rogoff), 2009
- Triumph of the Optimists: 101 Years of Global Investment Returns. Princeton University Press (Elroy Dimson, Paul Marsh, Mike Staunton), 2002