S&P 500 Concentration and Forward Returns: Does a Top-Heavy Market Predict the Next Decade?
The apparent relationship depends heavily on the 1965 starting point.
Across six exact decade starts, the sample Pearson correlation between S&P 500 top-10 concentration and the following 10-year annualized price return is -0.72. Remove June 1965 and the correlation becomes -0.04. That sensitivity is the main result, not a forecast.
Six starting points, six subsequent decades
The scatter plot uses only exact decade endpoints that can be reconstructed from S&P DJI's published table values. The x-axis is starting top-10 concentration. The y-axis is the annualized S&P 500 price return over the following 10 years.
Starting top-10 concentration vs. subsequent 10-year S&P 500 price return
Six points compare starting S&P 500 top-10 concentration with the following decade's annualized price return. June 1965 is the highest-concentration start at 38.24% and is followed by the lowest return at 1.17%. The other five observations do not preserve a strong monotonic relationship.
| Starting point | Starting top-10 weight | Forward period | Annualized S&P 500 price return |
|---|---|---|---|
| June 30, 1965 | 38.24% | 1965-1975 | 1.17% |
| June 30, 1975 | 29.08% | 1975-1985 | 10.84% |
| June 30, 1985 | 19.47% | 1985-1995 | 12.53% |
| June 30, 1995 | 17.67% | 1995-2005 | 10.00% |
| June 30, 2005 | 20.92% | 2005-2015 | 5.46% |
| June 30, 2015 | 17.81% | 2015-2025 | 11.66% |
The -0.72 correlation is not stable
The full six-point sample produces a correlation of -0.72. But the June 1965 start combines the sample's highest concentration, 38.24%, with its weakest subsequent annualized return, 1.17%. Removing that one point changes the correlation to -0.04.
The historical rows are still informative. The second-most concentrated start, June 1975 at 29.08%, was followed by a much stronger 10.84% annualized decade. The strongest decade in the sample, 12.53% annualized from 1985 to 1995, began with top-10 concentration of 19.47%.
Why this differs from S&P DJI's concentration-change result
S&P DJI describes a mildly positive statistical relationship between changes in concentration and index performance over its richer history. This study does not reproduce or reverse that statistic. It asks a different timing question: whether the starting concentration level at six decade endpoints was associated with the following 10-year return.
What this history does not prove
Returns exclude dividends and are not inflation-adjusted. The study does not control for starting valuation, interest rates, earnings growth, sector composition, changes in index membership, or the mechanism that caused concentration to rise or fall.
The June 2025 top-10 concentration endpoint is intentionally absent from the forward-return sample because its 2025-2035 outcome has not happened yet.
Methodology
- Start with the exact June 1965 top-10 concentration of 38.24%, reconstructed from the 10 company weights in S&P DJI's Exhibit 3.
- Add each decade's published change in top-10 weight from Exhibit 5 to reconstruct the next decade endpoint. Use the first six endpoints, June 1965 through June 2015, as starting observations.
- Pair each start with the annualized S&P 500 price return for the following 10-year period from Exhibit 5: 1.17%, 10.84%, 12.53%, 10.00%, 5.46%, and 11.66%.
- Calculate the ordinary Pearson correlation across the six concentration-return pairs, then repeat it after excluding June 1965 and in six leave-one-out samples.
- Do not digitize the quarterly concentration chart, add overlapping windows, or infer a return for the June 2025 starting point.
Primary source
Download the six forward-return observations
The CSV contains the six decade starts and subsequent 10-year annualized price returns. The JSON adds the sensitivity statistics, source boundaries, and reviewed source URL.
Research data
- Download CSVCSVCSV downloadTabular public study data for spreadsheet analysis, independent checks, and new charts.Data snapshot June 30, 2025 · Reuse with attribution to the canonical study.
- Download JSONJSONJSON downloadStructured public study data for programmatic verification while preserving the published field names and research context.Data snapshot June 30, 2025 · Reuse with attribution to the canonical study.
Citation and reuse
When citing the correlation, describe it as a six-observation descriptive statistic using reconstructed decade starting concentrations and subsequent S&P 500 annualized price returns. Do not attribute the -0.72 value to S&P DJI.