Insights

Revisiting the Relationship Between Value and Sentiment in Equities

  • Where have all the expensive, high-sentiment stocks from July of 2020 gone?
  • Recent months have seen a realignment of the value and sentiment themes. More companies that appear to be a good value – that is, those that look attractively priced on fundamentals – are also showing positive investor sentiment.
CIO, Active Quantitative Equity

In the third quarter of last year, we discussed the deep disagreement that was taking shape between two important themes in equities: value and sentiment. These themes each represent complementary characteristics that we seek in Active Quantitative Equity (AQE) as we build equity portfolios. When there’s a lot of disagreement between value and sentiment, the pool of companies that we would deem attractive in aggregate becomes smaller. What we observed instead were extreme levels of concentration in the high-sentiment, expensive pocket of the equity market.

A lot has happened since we last published on this topic. In this piece, we’ll revisit the relationship between value and sentiment. What does the relationship between value and sentiment look like now? And what has happened with those high-sentiment, expensive names?

Changing Correlation

The relationship between value and sentiment can be expressed in terms of the cross-sectional correlation – that is, the degree of alignment – between the two themes. Two variables are negatively correlated if one decreases when the other increases; they are positively correlated when they tend to move up and down together.

Over the course of about 10 months, the correlation between value and sentiment has moved from a historically low negative correlation to a historically high positive correlation (see Figure 1). When correlations reached those historical lows, there was an extraordinary amount of overlap between companies that were both expensive and that showed strong positive sentiment.

Figure 1. Cross-Sectional Correlation of AQE’s Proprietary Value and Sentiment Signals

This change in correlation was driven more by changes in sentiment than by changes in value. Of the 600 companies we considered to be the cheapest last July, we still consider about two-thirds of that cohort to be cheap. Of the 600 companies we viewed as benefiting from the strongest sentiment last July, only about one-third of that cohort still have the strongest sentiment today.

Changing Concentration

The last time we looked at market concentration for value and sentiment in this commentary, we observed not just a very large number of companies in the expensive, high-sentiment group, but also record-high market capitalization of that cohort. However, the index weight of expensive, high-sentiment stocks has changed dramatically since July 2020 (see Figure 2).

Figure 2. Index Weight by Value/Sentiment Quadrant

Where have all those expensive, high-sentiment stocks gone now? For the most part, they remain expensive, but their sentiment has dropped dramatically.

The following grid shows the current value and sentiment positioning of the most expensive, high-sentiment names back in July 2020.

  Still Strong Sentiment Neutral Poor Sentiment
Still Expensive

Apple

Facebook

Nvidia

PayPal

ASML

Costco

Microsoft

Amazon

Visa

Mastercard

Adobe

Thermo Fisher

Tesla

Netflix

Salesforce

Neutral -- -- SAP
Now Cheap --  -- 

Procter & Gamble

AstraZeneca 

Source: State Street Global Advisors.

As we see here, most of these names (around 70% of them) remain expensive; however, only 29% of the names (including Apple and Facebook) remain in the high-sentiment cohort. Other expensive stocks, such as Microsoft and Amazon, are neutral; still other expensive stocks are currently suffering from poor sentiment, including Tesla and Netflix.

The Bottom Line

Recent months have seen a realignment of the value and sentiment themes. More companies that appear to be a good value – that is, those that look attractively priced on fundamentals – are also showing positive investor sentiment. For investors concerned with both value and sentiment attributes, this leads to several consequences, including:

  • Lower turnover in portfolios; lower trading costs
  • A larger pool of companies with attractive attributes from which to pick stocks
  • Greater ability in long-only portfolios to align overweight and underweight portfolio positions with model preferences
Share