Measuring Investor Sentiment More Confidently

Investor sentiment can help guide stock pickers when recession is a risk and earnings prospects are uncertain. Yet do traditional measures of sentiment capture the full picture?

The Active Quantitative Equity (AQE) team uses a wide range of metrics to form a multi-dimensional view of stocks, based on our main themes of Value, Quality, and Sentiment. Within the Sentiment theme, directly captured data (price action) as well as indirectly captured data (throughout the supply chain) give us a highly granular way in which to assess overall investor sentiment. These measures complement the other ways in which we view Sentiment – i.e., earnings sentiment as well as sentiment captured through dialogue with company management¹ or through hedge fund positioning. We have higher confidence in our stock selections when these sentiment measures align.

Segment Implications

The AQE team believes that we can build a richer picture of investor sentiment by taking advantage of the time it takes for sentiment to trickle through the supply chain. Therefore, we use supply chain linkages to map the network of suppliers to customers and, in so doing, build metrics that look deeper into investor sentiment. We believe that sentiment is best assessed when examined from as many angles as possible.

For that reason, we look at indirect measures that use the supply chain to gauge the sentiment toward a company’s customers (which we call “customer sentiment”), in addition to direct measures of sentiment based on stock-price movements (here referred to as “share price momentum”). Customer sentiment presents an opportunity to take advantage of the time it takes for investors to process information about the flow from customers to suppliers before investors at-large process the complexities of the supply chain.

In some segments the AQE team is seeing a disconnect between direct measures of investor sentiment using price movements of company stock prices (“share price momentum”) and indirect measures that use the supply chain to gauge the sentiment toward a company’s customers (“customer sentiment”). The fact that this disconnect can happen from time to time is exactly the reason why it is important to evaluate from multiple dimensions. No one angle represents a “better truth,” so by incorporating multiple angles, scoring will be diluted when the disagreement exists, and scoring will be amplified when the measures are aligned, thereby giving us more confidence in the overall assessment of the stock.

The segments where we see customer sentiment meaningfully weaker than share price momentum are Telecommunications Services, Semiconductors, Materials, Autos, Transport, and Consumer Durables (see Figure 1).

In Consumer Durables and Telecommunications Services, while there remains a difference in the two measures, share price momentum and customer sentiment have started to converge in the last few months, a pattern which builds confidence that our multi-dimensional sentiment measure is helping us forecast stock prices more successfully. In the case of Consumer Durables, our overall measures of investor sentiment are weak, while in Telecommunications Services they are strong and improving relative to other segments.

In the Materials, Semiconductors, and Transport segments, the differences between customer sentiment and share price momentum have widened as customer sentiment has further deteriorated.

Within the Materials segment, we see the disconnect across gold, paper, chemicals, metals, and other subsectors across the world. Within Transport, shipping and other cargo companies are seeing the greatest disconnect and weakening of customer sentiment. Within Semiconductors, Japanese companies are experiencing more disconnect in share price momentum versus customer sentiment.

There are many possible explanations for disconnects between price momentum and customer sentiment. A closer look at a specific sector example sheds further light on this topic. For the Semiconductor segment, for instance, supply chain shortages, a recurring theme, are likely driving the disconnect between more positive sentiment for semiconductor manufacturers and more negative sentiment for their customers. The demand for semiconductor chips is notably higher than supply right now. Goods like cars or data centers cannot be finished without the chips. Many years ago such a supply shortage would only have impacted the production of personal computers. In contrast, in today’s economy, the usage of semiconductor chips is far more widespread across sectors. While customer companies are currently challenged and at times unable to complete production of their goods, suppliers are mostly at over capacity. Many of the chipmakers’ open orders are many times higher than production capacity levels. This implies the impact of a recession on customer demand would have to be very deep and long to put a dent in the expected revenue observed in some of these semiconductor companies.

Relying on metrics, in this case a dataset that can map these supply chain complexities, is beneficial and enables us to calculate a measure of sentiment objectively and to make a determination by means of algorithms as opposed to more subjective means.

Sentiment measures shift more rapidly than Value and Quality measures, and over the last few months, while the different measures of Sentiment have changed, the AQE team’s overall preferences for the segments discussed here have not changed dramatically, primarily because when Sentiment improves, stocks become more expensive, and vice versa. The changes can offset each other. Figure 2 summarizes our current view on the segments discussed, how they have changed, and the underlying themes shaping that view.

The Bottom Line

In today’s market of high uncertainty about recession risk and inflation pressure on corporate earnings, investor sentiment provides an additional input to assess potential equity performance. By examining both direct and indirect measures of investor sentiment, the AQE team can develop a multi-dimensional outlook on companies and a more comprehensive way in which to analyze overall sentiment. This approach to measuring investor sentiment serves as an informational edge which helps us to project potential winners.