Are broad measures of public sociopolitical sentiment relevant to investors? Do they predict stock returns as indicators of exuberance and fear? To investigate, we relate S&P 500 Index return and 12-month trailing S&P 500 price-operating earnings ratio (P/E) to the percentage of respondents saying “yes” to the recurring Gallup polling question: “In general, are you satisfied or dissatisfied with the way things are going in the United States at this time?” Since individual polls span several days, we use S&P 500 Index levels for about the middle of the polling interval. To calculate market P/E, we use current S&P 500 Index level and most recent quarterly aggregate operating earnings. Using Gallup polling results, S&P 500 Index levels and 12-month trailing S&P 500 operating earnings as available during July 1990 (when polling frequency becomes about monthly) through June 2019, *we find that:*

The following chart tracks the S&P 500 Index (on a logarithmic scale) and the level of public sociopolitical satisfaction (% Satisfied) over the full sample period. Visual inspection reveals no obvious consistent relationship between the two series.

Because the S&P 500 Index is not stationary, we relate S&P 500 Index return to % Satisfied.

The following scatter plot relates S&P 500 Index poll-to-poll return to % Satisfied for the later poll. The Pearson correlation between the two series is 0.08 and the R-squared statistic is 0.007, indicating that % Satisfied explains less than 1% of concurrent variation in index return. Nor does it appear that the relationship has interesting tail effects.

In case there is some lag between stock market behavior and sociopolitical satisfaction, we perform a lead-lag analysis.

The next chart summarizes correlations between S&P 500 Index poll-to-poll return and either % Satisfied or (in case % Satisfied is not stationary) change in % Satisfied across scenarios ranging from return leads % Satisfied measurements by 12 polls (-12) to % Satisfied measurements lead return by 12 polls (12). 12 polls is about a year. Notable points are:

- All correlations are small.
- There are persistently positive correlations indicating that S&P 500 Index return leads % Satisfied, suggesting that a series of positive (negative) returns predicts growing (falling) sociopolitical satisfaction.
- Correlations between S&P 500 Index return and change in % Satisfied are noisy, with scalloping perhaps due to combining monthly index levels with quarterly changes in earnings. The strongest indication is coincident. There is perhaps weak indication that change in % Satisfied leads stocks positively by one or two polls.

Might % Satisfied relate to S&P 500 P/E?

The next chart tracks S&P 500 P/E and % Satisfied over the full sample period. There appears to be a relationship between these two series in terms of both broad trends and finer features.

For perspective, we look at a scatter plot.

The following scatter plot relates S&P 500 Index P/E to % Satisfied over the full sample period. At the polling frequency, the Pearson correlation between the two series is 0.60 and the R-squared statistic is 0.36, indicating that variation in % Satisfied explains 36% of concurrent variation in P/E.

In other words, P/E is a rough proxy for level of sociopolitical satisfaction, perhaps reasonably interpreted as level of willingness to take risk.

In summary, *evidence suggests that public sociopolitical satisfaction and stock market valuation (market P/E) move somewhat in step, but with no reliable way to exploit the satisfaction measure for stock market timing.*

Cautions regarding findings include:

- Sample size is modest with respect to the numbers of independent annual earnings measurements and very small with respect to number of four-year political cycles and number of economic/market cycles.
- Precise alignment of polling results, S&P 500 Index returns and (especially) S&P 500 P/E is problematic. Recently, Gallup has widened the polling interval, exacerbating this caution.
- The above analyses are in-sample (retrospective). An investor operating in real time during the sample period would not know all the data used in the charts above and may draw different conclusions at different times based on strictly historical information.