Do Sentiment Metrics Matter to the Markets?
Consumer spending accounts for almost 70% of nominal US GDP. As such, consumer sentiment ought to have some correlation with market performance.
Financial journalists certainly act as though it does. Whenever new sentiment or confidence numbers — consumer or otherwise — are released, pundits spring into action, speculating on what the data’s implications are for the markets and the overall economy. But how much do these measures actually matter to market performance?
To answer this question, we explored the correlations between consumer and business sentiment metrics and market returns. Specifically, we examined monthly data from the University of Michigan Consumer Sentiment Index, the Conference Board’s US Consumer Confidence Index (CCI), and the Business Confidence Index (BCI) and compared their relationship to the performance of nine different MSCI stock and bond indices going back to the 1970s, focusing on US high-yield bonds, US long-term bonds, US short-term bonds, US aggregate fixed income, US growth equity, US value equity, US small cap, US large cap, and international equity.
In aggregate, we did not find any significant or sustained correlation between market returns and the three sentiment measures over the entire 50-plus year sample period. The highest correlation, between the University of Michigan Consumer Sentiment Survey and US small-cap stocks, maxed out at a weak 0.21.
Correlations between Changes in Consumer Confidence Indices and Investment Returns, 1970s to 2020s
Michigan Consumer Sentiment Index | Consumer Confidence Index (CCI) | Business Confidence Index (BCI) | |
US High-Yield Bond | 0.18 | 0.17 | –0.01 |
US Long-Term Bond | –0.01 | 0.04 | –0.10 |
US Short-Term Bond | –0.01 | 0.03 | –0.11 |
US Fixed Income | –0.01 | 0.08 | –0.13 |
US Growth | 0.14 | 0.12 | 0.07 |
US Value | 0.17 | 0.15 | 0.07 |
US Small Cap | 0.21 | 0.14 | 0.11 |
US Large Cap | 0.15 | 0.15 | 0.06 |
International | 0.15 | 0.18 | 0.12 |
Yet over time, the correlations exhibit some illuminating trends.
The University of Michigan Consumer Sentiment Index’s correlation with equity returns has diminished. Indeed, since 2010, it has fallen precipitously and been statistically indistinguishable from zero.
University of Michigan Consumer Sentiment Index: Historical Market Correlations
1970s | 1980s | 1990s | 2000s | 2010s | 2020s | |
US High-Yield Bond | 0.24 | –0.05 | 0.34 | 0.35 | –0.09 | 0.20 |
US Long-Term Bond | 0.24 | –0.19 | 0.01 | 0.17 | –0.13 | –0.07 |
US Short-Term Bond | 0.23 | –0.09 | –0.09 | 0.05 | –0.16 | 0.14 |
US Fixed Income | 0.22 | –0.15 | –0.01 | 0.13 | –0.18 | 0.09 |
US Growth | 0.09 | 0.29 | 0.12 | 0.24 | –0.04 | –0.05 |
US Value | 0.13 | 0.27 | 0.11 | 0.31 | –0.07 | 0.01 |
US Small Cap | 0.08 | 0.33 | 0.18 | 0.36 | 0.00 | 0.04 |
International | 0.08 | 0.31 | 0.10 | 0.28 | –0.12 | 0.06 |
US Large Cap | 0.11 | 0.25 | 0.13 | 0.28 | –0.03 | –0.02 |
International | 0.08 | 0.31 | 0.10 | 0.28 | -0.12 | 0.06 |
The CCI, however, has displayed the greatest positive correlation to equity returns since the 2000s. And since 2020, equity correlations and bond correlations have averaged a rather significant 0.30.
Consumer Confidence Index (CCI): Historical Market Correlations
1970s | 1980s | 1990s | 2000s | 2010s | 2020s | |
US High-Yield Bond | 0.25 | 0.014 | 0.16 | 0.15 | 0.20 | 0.35 |
US Long-Term Bond | 0.09 | 0.01 | –0.04 | –0.02 | –0.09 | 0.26 |
US Short-Term Bond | 0.04 | –0.04 | –0.09 | –0.09 | 0.10 | 0.34 |
US Fixed Income | 0.16 | 0.03 | –0.07 | –0.04 | 0.05 | 0.36 |
US Growth | 0.00 | 0.01 | 0.03 | 0.25 | 0.18 | 0.22 |
US Value | 0.04 | –0.01 | 0.04 | 0.30 | 0.19 | 0.27 |
US Small Cap | 0.08 | 0.01 | 0.06 | 0.22 | 0.17 | 0.32 |
US Large Cap | –0.02 | 0.01 | 0.04 | 0.29 | 0.18 | 0.24 |
International | 0.03 | 0.01 | 0.10 | 0.28 | 0.22 | 0.41 |
The BCI shows a similar trend. The BCI has charted its highest positive correlations with the equity return measures, with the upswing beginning in the 2010s.
The Business Confidence Index (BCI): Historical Market Correlation
1970s | 1980s | 1990s | 2000s | 2010s | 2020s | |
US High-Yield Bond | –0.29 | –0.15 | 0.03 | 0.13 | 0.19 | 0.22 |
US Long-Term Bond | –0.35 | –0.21 | –0.11 | 0.05 | –0.06 | 0.09 |
US Short-Term Bond | –0.12 | –0.17 | –0.22 | 0.04 | 0.06 | 0.06 |
US Fixed Income | –0.39 | –0.18 | –0.16 | 0.08 | 0.06 | 0.14 |
US Growth | 0.14 | –0.04 | 0.07 | 0.09 | 0.20 | 0.11 |
US Value | 0.05 | –0.09 | 0.05 | 0.10 | 0.23 | 0.23 |
US Small Cap | 0.13 | –0.02 | 0.10 | 0.15 | 0.23 | 0.23 |
US Large Cap | 0.06 | –0.09 | 0.07 | 0.09 | 0.21 | 0.17 |
International | 0.11 | 0.01 | 0.15 | 0.16 | 0.17 | 0.28 |
That markets correlate more with the CCI and BCI than the University of Michigan Consumer Sentiment Index has several potential implications. Perhaps the CCI and BCI have grown in prestige over time relative to the Michigan index and now the market pays more attention to them. Or maybe their methodologies better reflect an evolving market and economy.
Of course, whatever the roots of these phenomena, the larger takeaway given the relative weakness of these correlations is that financial journalists and commentators may derive more meaning from these metrics than they warrant.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©Getty Images / Natee Meepian
Professional Learning for CFA Institute Members
CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.
Why was a correlation used rather than a regression using different lagged periods? The sentiment index this month and market returns 1 month later, 3 months later and maybe one year later.
“Sentiment metrics” are examples of information cited by the media that are right (correlate) some of the time so they keep citing them. It’s similar to the suggestion that there are only “sellers” on down days and “buyers” on up days. Your bottom line conclusion is right on.