S&P 500 Dividend Yield: Long-Term Series Since 1871 (Shiller Data)

The S&P 500 dividend yield measures annual dividend distributions per dollar of S&P 500 price, expressed as a percentage. This dataset compiles the monthly series from Robert Shiller’s Yale dataset (ie_data.xls), spanning January 1871 to the most recent observation — more than 150 years of continuous monthly data. The S&P 500 dividend yield declined structurally from approximately 5%–6% in the late nineteenth century to below 2% in the modern era, primarily reflecting the post-1980 shift from dividend distributions toward share buybacks as the dominant form of cash return to shareholders.

Dataset: S&P 500 Dividend Yield (1871–2026) · Updated 2026-07-01

Source: Robert Shiller, Yale University — ie_data.xls, open academic data on long-term US equity valuation and dividend history

Macro Takeaway

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 dataset is available in CSV and Excel formats.

Direct CSV Access — Eco3min Structured Dataset

https://eco3min.fr/dataset/sp500-dividend-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/sp500-dividend-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/sp500-dividend-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 S&P 500 dividend yield is computed as trailing twelve-month aggregate dividends paid by S&P 500 constituents, divided by the S&P 500 price level at the end of the reporting month, expressed as a percentage. The numerator captures actual cash dividend distributions over the prior year; the denominator is the contemporaneous index level. The ratio therefore represents a backward-looking dividend yield rather than a forward estimate.

The source data is maintained by Robert Shiller (Yale University) in the file ie_data.xls, published as open academic data. Pre-1957 dividend data is reconstructed from the Cowles Commission historical series; post-1957 data is sourced from S&P/Standard Statistics. Shiller updates the file approximately monthly, typically within the first two weeks of each calendar month.

Eco3min mirrors the Shiller dataset via an automated pipeline that reformats the source spreadsheet into a normalized CSV/XLSX schema. The dataset is refreshed when a new release is detected at the Shiller source.

Data Quality & Provider Notes

The S&P 500 dividend yield series is sourced from Robert Shiller’s Yale University publication, the standard academic reference for long-term US equity valuation. The series is not available on FRED; users must work from Shiller’s source file or a mirror such as Eco3min, multpl.com, or commercial data providers.

• Release latency. Shiller updates ie_data.xls on an approximately monthly cadence, generally during the first or second week of each calendar month. There is no fixed publication date, so the latest available observation may lag the current month by two to six weeks.
• Revisions policy. Revisions are uncommon but occur when (i) S&P revises historical reported dividends, or (ii) Shiller corrects historical reconstruction errors flagged by academic readers. Recent observations carry slightly higher revision risk than older data. Pre-1926 dividend figures rely on Cowles Commission reconstructions and should be interpreted with awareness of nineteenth-century accounting conventions.
• Alternative sources. S&P Dow Jones Indices publishes the official S&P 500 dividend point series via its proprietary data platform. multpl.com mirrors the Shiller methodology. Bloomberg (field DVD_YLD) computes dividend yield from current trailing distributions and can differ slightly from Shiller in real-time observations. Aggregate dividend-payment data is also available from S&P Indices for total-payout analyses.
• Known gaps. The series is continuous monthly from January 1871. There are no missing observations.

Before running historical analyses, confirm the last observation date against Shiller’s source page directly. For research requiring strict vintage reproducibility, archive the source ie_data.xls at the date of analysis.

Common Pitfalls When Using the S&P 500 Dividend Yield

The S&P 500 dividend yield is the oldest continuous US equity valuation metric, but its interpretation has shifted materially since the 1980s due to changes in corporate payout policy. Several recurring errors distort its signal.

1. Confusing price yield with total return. The S&P 500 dividend yield reports cash distributions per dollar of price — it does not capture the compounding effect of dividend reinvestment, which is the dominant driver of long-run S&P 500 total return. Multi-decade analyses of equity performance must use a total-return series (with dividends reinvested) rather than a price-only series adjusted by current yield.
2. Ignoring the post-2000 buyback substitution. Since the early 2000s, S&P 500 issuers have increasingly returned cash to shareholders via share buybacks rather than dividends. By the mid-2020s, aggregate buyback authorizations consistently exceeded total dividend payments. The S&P 500 dividend yield therefore understates total shareholder cash return by approximately one-half in the modern era. The aggregate “shareholder yield” (dividends plus net buybacks) is the corresponding total-payout metric, not the dividend yield alone.
3. Comparing pre-1986 and post-1986 yields without tax-regime context. The 1986 Tax Reform Act reduced the differential between capital-gains and ordinary-income taxation. Combined with subsequent tax changes, this shifted corporate-finance preferences toward retained earnings and buybacks. A 4% S&P 500 dividend yield in 1975 reflected a different tax-adjusted economic value than a 4% yield would today — the two are not directly comparable.
4. Confusing trailing and forward yields. The Shiller series reports trailing twelve-month dividends divided by the current price. Real-time market quotes from data terminals often report annualized forward yields based on the most recent quarterly distributions. The two metrics can diverge by 10–20 basis points and should not be substituted in time-series analysis.
5. Source confusion. Shiller’s S&P 500 dividend yield uses aggregate S&P 500 trailing dividends from ie_data.xls. Other providers may publish yield series using different dividend aggregation methods (cap-weighted versus equal-weighted, gross versus net of withholding), different observation frequencies (daily versus monthly), or different lookback windows. Values quoted from Bloomberg, multpl.com, or commercial vendors may differ by 5–20 basis points from this dataset and should not be substituted without verifying methodology.

Historical Regimes

1871–1928 — High-payout era. The S&P 500 dividend yield typically ranged between 4% and 6%, reflecting an industrial economy where dividends were the dominant form of shareholder return and corporate retention was modest by modern standards. The series provides one of the few continuous market-valuation references for the pre-Federal Reserve period.

May 1932 — Depression spike. The S&P 500 dividend yield reached approximately 13.84% — the all-time high in the 155-year series. The spike reflected the catastrophic collapse in equity prices (denominator effect) far more than any increase in cash distributions, as many issuers cut or suspended dividends during the same period.

1946–1980 — Stable mid-range. Following postwar normalization, the S&P 500 dividend yield oscillated between roughly 3% and 5%, with periodic excursions above 6% during the 1970s stagflation period. The 1981 trough near the start of the disinflation rally registered around 5.7%.

1981–March 2000 — Disinflation compression. The S&P 500 dividend yield declined from approximately 5.7% to a record low near 1.1% in early 2000 — the lowest reading in the entire series. The compression reflected both rising equity prices and the early stages of the buyback substitution: by the late 1990s, total payout was already shifting away from dividend distributions.

2000–2008 — Modest recovery. The post-dot-com price correction lifted the S&P 500 dividend yield back to approximately 2% by 2003. The metric remained range-bound between 1.8% and 2.4% through the credit cycle, with a brief spike to roughly 3.6% in early 2009 as prices collapsed during the global financial crisis.

2009–2019 — Buyback-era persistent lows. Despite ongoing economic expansion, the S&P 500 dividend yield remained anchored between 1.8% and 2.2% throughout the cycle. The structural compression reflected the accelerating buyback trend: by the late 2010s, S&P 500 issuers were returning more cash via buybacks than via dividends.

2020–2026 — Structural sub-2% regime. Pandemic-era stimulus and the AI-driven equity rally pushed the S&P 500 dividend yield below 1.5% for extended periods in 2021 and 2024. The current regime is characterized by low absolute yields combined with high aggregate buyback authorizations — a configuration that materially changes how the metric should be interpreted relative to its pre-1980 history. Read in tandem with the S&P 500 price index and CAPE ratio provides the cleanest read of valuation regime through the modern cycle.

Related Macroeconomic Datasets

The S&P 500 dividend yield is most informative when interpreted alongside complementary measures of equity return, valuation, and broader market context. The datasets below provide the principal cross-references.

Macroeconomic Dataset Hub

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

Explore the Eco3min Dataset Hub

Sources

• Robert Shiller, Yale University — ie_data.xls, open academic data on long-term US equity valuation and dividend history
• S&P Dow Jones Indices — official S&P 500 dividend point methodology and current data

Dataset Reference

Eco3min Research. (2026). S&P 500 Dividend Yield — Long-Term Monthly Series (1871–2026).<br> https://eco3min.fr/en/sp500-dividend-yield-dataset/