Excess CAPE Yield: Inverse CAPE Minus Real 10-Year Treasury Yield Monthly Since 1881

This is an Eco3min composite built from three Shiller series — the CAPE ratio, the 10-year nominal Treasury yield (GS10), and trailing 12-month CPI — combined into a single monthly series since 1881. No equivalent native series exists on FRED or Bloomberg.

Dataset: Excess CAPE Yield (1881–2026) · Updated 2026-07-01

Source: Robert Shiller, Yale University, ie_data.xls — Open academic data (CAPE, GS10, CPI-U series)

Macro Takeaway

Construction & Components

The Excess CAPE Yield is the cyclically adjusted equivalent of an equity risk premium computed against real bonds. It removes two distortions present in the simple earnings yield minus nominal Treasury yield: short-term earnings volatility, and the inflation premium embedded in nominal bond yields. Data reference: Our ERP dataset.

Formula:

Excess CAPE Yield = (1 / CAPE) × 100 − Real 10Y Yield
Real 10Y Yield   = GS10 − Trailing 12-month CPI inflation

Components:

• CAPE (Cyclically Adjusted Price-to-Earnings) — Shiller dataset (ie_data.xls), monthly. S&P 500 real price divided by the 10-year trailing average of real earnings. Provides the inverse used as the smoothed earnings yield.
• GS10 (10-Year Treasury Constant Maturity Rate) — Shiller dataset, monthly. Long-term nominal US Treasury yield, available continuously since 1871.
• CPI inflation (trailing 12 months) — Shiller dataset (CPI-U for post-1913, pre-1913 reconstructions from Warren and Pearson). Subtracted from GS10 to produce the real bond yield leg.

Frequency reconciliation: All three components are monthly within the Shiller dataset. The composite is therefore monthly with no interpolation required. End-of-month observations are used throughout.

Coverage: 1881–present (limited by the start of the CAPE 10-year smoothing window, which itself requires earnings data back to 1871). Real bond yield reconstruction relies on Shiller’s CPI series and is most reliable from 1913 onward; pre-1913 readings carry larger measurement uncertainty.

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns.

Column names match the CSV headers exactly.

Download the Complete Dataset

The full Excess CAPE Yield dataset is available in CSV and Excel formats.

Direct CSV Access — Eco3min Structured Dataset

https://eco3min.fr/dataset/excess-cape-yield.csv

This URL returns the complete dataset in CSV format. It can be used directly in pandas, R, curl, or any data tool.

Using the Dataset in Python

import pandas as pd

url = "https://eco3min.fr/dataset/excess-cape-yield.csv"
df = pd.read_csv(url, parse_dates=["date"])

print(df.head())
print(df.describe())

Using the Dataset in R

library(readr)

url <- "https://eco3min.fr/dataset/excess-cape-yield.csv"
df <- read_csv(url)

head(df)
summary(df)

Both examples load the dataset directly from the URL — no download or API key required.

Methodology

The Excess CAPE Yield series is recomputed monthly by an Eco3min pipeline that pulls Robert Shiller’s ie_data.xls spreadsheet from the Yale University site. The three required inputs (CAPE, GS10 nominal yield, CPI-U) are read directly from the Shiller workbook, aligned on end-of-month dates, and combined according to the Excess CAPE Yield formula. The pipeline runs on the second business day of each month, after Shiller typically refreshes the workbook for the preceding month.

Eco3min uses the same trailing 12-month CPI window for the real bond yield leg that Shiller himself uses in his 2018 paper introducing the indicator. No additional smoothing or filtering is applied beyond the 10-year CAPE smoothing already embedded in the CAPE input.

Data Quality & Provider Notes

Latency is dictated by Shiller’s monthly update cadence: the workbook is typically refreshed during the first 5–10 business days of each calendar month for the preceding month. Eco3min cannot publish fresher data than the upstream Shiller release.

Revisions propagate: when Shiller updates historical CAPE earnings (which occasionally happens for S&P trailing earnings restatements) or BLS revises CPI, the entire Excess CAPE Yield series upstream of the revision changes. Eco3min replaces the full historical CSV on each release rather than appending only the latest observation, so users always retrieve the most recent vintage.

No equivalent native series exists on FRED, Bloomberg, or other major data providers. The closest commercial alternative is the GMO 7-Year Real Return Forecast, which uses a different methodology and is paywalled. Eco3min’s composite is constructed transparently from the public Shiller workbook and reproduces Shiller’s published methodology.

What This Index Captures (And What It Doesn’t)

The Excess CAPE Yield is a structural valuation indicator, not a market-timing tool. Understanding its scope avoids the most common misreadings.

What it captures:

• The cyclically smoothed earnings advantage of US equities over real bonds, comparable across decades of differing inflation regimes
• The combined effect of valuation pressure (high CAPE) and discount-rate pressure (low real rates) on the equity-bond risk gap
• A long-horizon signal: in Shiller’s 2018 paper, the indicator showed meaningful correlation with subsequent 10-year real equity returns, weaker correlation with horizons under 3 years

What it does NOT capture (common misinterpretations):

• Short-term market direction. The indicator’s predictive power is concentrated at 10-year horizons. Readings can persist far from their historical median for years before mean-reverting — periods of “expensive” or “cheap” Excess CAPE Yield have lasted entire decades.
• Sector- or factor-level valuation. CAPE is computed on the aggregate S&P 500. The Excess CAPE Yield averages out wide sector valuation dispersion that can exist beneath the index level.
• Risk-free rate dynamics that aren’t captured by trailing CPI. Inflation expectations (as measured by TIPS breakevens) can diverge from trailing 12-month CPI, especially during inflation regime shifts. In those periods, the “real 10Y” leg of Excess CAPE Yield lags forward-looking real rates by months.
• Buyback effects on earnings yield. Aggregate earnings used in CAPE include the effect of share buybacks, which mechanically lift earnings per share. The indicator does not separate organic earnings growth from buyback-driven EPS lift.

The Excess CAPE Yield is most useful as a regime-classification tool — placing the current environment in its full historical distribution — rather than as a tactical asset-allocation trigger.

Historical Regimes

The Excess CAPE Yield has spanned a wide distribution since 1881, with several distinct regimes:

• 1929 peak (negative reading). The September 1929 Excess CAPE Yield turned negative as CAPE reached ~32 while real rates were positive. This was followed by the largest equity drawdown in US history.
• 1949–1955 (very elevated). Post-WWII equity valuations remained depressed despite low real rates, producing an Excess CAPE Yield consistently above 5%. The subsequent 1950s decade delivered some of the strongest 10-year equity returns on record.
• 1981–1982 (highest readings of the modern era). The combination of very high real rates (~5–7%) and depressed equity valuations produced unusual dynamics in the spread. Equity returns over the following 20 years were exceptional.
• 1999–2000 dot-com peak (compressed). CAPE climbed to ~44 while real rates were near 4%, driving the Excess CAPE Yield close to zero for the first time since 1929. The subsequent decade produced negative real S&P 500 returns.
• 2009–2020 (zero real-rate era). Sustained negative real 10Y yields kept the Excess CAPE Yield positive despite CAPE rising to 30+. Shiller’s 2018 paper specifically highlighted this regime as a structural argument for elevated CAPE.
• 2022–2024 (regime shift). The rise in real rates from −1% to +2% materially compressed the Excess CAPE Yield while CAPE held near 30+, returning the indicator to readings comparable to the 1999–2000 episode.

For analytical context on how real rates and CAPE jointly determine equity valuation regimes, see the Eco3min study US Real Interest Rates vs CAPE Ratio — The Tent-Shaped Relationship. The CAPE ratio dataset isolates the valuation input, and the real interest rates dataset isolates the bond yield input.

Related Macroeconomic Datasets

Related Research

Macroeconomic Dataset Hub

This dataset is part of the Eco3min macro-financial data repository.

Sources

• Robert Shiller, Yale University, ie_data.xls — Open academic data (CAPE, GS10, CPI-U series)
• Shiller, R. J., Black, L. J., & Jivraj, F. (2020). CAPE and the COVID-19 Pandemic Effect. Yale ICF Working Paper — Methodology reference for Excess CAPE Yield

Dataset Reference