Testing Whether Equity Valuations Predict Long-Run Returns in Excel
Summary
The article uses Excel to investigate whether the cyclically adjusted price-to-earnings ratio (CAPE) predicts subsequent real returns on a broad US equity index. It rebuilds a valuation-versus-forward-return scatterplot from historical data, then questions how much confidence to place in the apparent relationship. Monthly observations of ten-year returns overlap heavily, so the many plotted points do not represent independent evidence. The author compares shorter horizons, annual observations, and non-overlapping decade observations to examine the effect with fewer repeated periods.
The examples show a weak negative association between valuation and future returns, while a rolling standardization of CAPE does not show the same effect under the chosen scaling. The article interprets the evidence as suggestive, not decisive. There are few independent long-horizon observations, early index history is reconstructed, market conditions have changed, and results can depend on measurement choices. It concludes that valuations may inform expected returns, but the analysis is too uncertain to justify a strong investment decision by itself.
Key ideas
- Overlapping forward-return windows make monthly observations look more numerous than the independent evidence warrants.
- Non-overlapping periods reduce repeated observations but leave a small sample for long horizons.
- The article finds a weak negative association between CAPE and later returns in some views of the data.
- Standardizing a drifting valuation series can change the apparent relationship, so scaling choices need scrutiny.
- Historical valuation analysis has limited statistical power and does not by itself support a decisive market-timing call.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.