TL;DR
Shiller's S&P composite is the longest reliable equity series, going back to 1871. Actual index funds (VFINX, etc.) only have track records to 1976. For studying long-horizon FIRE risk, we need the older data.
In short
Critics argue Shiller's pre-1957 reconstruction is imperfect — true. But it's by far the best available, and the 1871-1957 period adds critical stress scenarios (1907 panic, WWI, 1920s bull, Depression). Excluding it would make the dataset less honest about tail risk, not more.
We're working on a full deep-dive for this article — including historical data, charts, and worked examples. In the meantime, you can run a free simulation to explore the underlying numbers yourself.
Frequently asked questions
- What about pre-1871?
- Data exists (Schwert, Wilson-Jones) but quality drops sharply and US equity markets were a different animal before the late 1800s. We start at 1871 for that reason.
- Does using fund-level data change the results?
- Slightly. Real funds have expense ratios that the academic series don't. We subtract a default 0.15% from displayed returns to approximate this, and you can adjust the assumption in the simulator.
Stress-test your own FIRE plan
FIRE Wealth OS runs your savings rate and expenses against every historical market starting point since 1871. Free to use, no card required.