A Check On Market Sentiment: Has The Pandemic Changed The Signal?
It is a truism that measures of market “sentiment” are usually contrarian. They point “backwards.” They suggest an obvious interpretation — but actually signify the opposite. If, for example, investor opinion surveys show optimism, enthusiasm, and confidence about the future… it is taken as a sure sign that the market is headed for a downturn.
With that in mind, what are sentiment metrics signaling now?
The answer is clear. Everything points to an imminent market correction.
But can we believe it? Has the psychology of the pandemic changed the way we should read the “sentiment” tea-leaves?
What is Sentiment?
“Sentiment” — a technical term in Finance – refers to the distillation of the opinions of individuals or groups concerning a company, the market, or the economy as a whole into signals or metrics that can reliably forecast the market direction in the not-too-distant future. Sentiment mixes fact and rumor, experience and expectation, “feelings, attitudes, emotions, and opinions.” Sentiment takes in everyone from the fabled shoe-shine boy…
- “In 1929 Joseph Patrick “Joe” Kennedy, Sr. JFK’s father, claimed that he knew it was time to get out of the stock market when he got investment tips from a shoeshine boy. On that moment Joe Kennedy had the intuition that we were at the end of the bull market and subsequently he decided to short the market… Ever since, the shoeshine boy has been the metaphor for “time to get out”; for the end of the mania phase in which everyone, even the shoeshine boy, wants in.”
MORE FOR YOU
… to the top analysts on Wall Street, experienced business journalists, or the economists of the Federal Reserve. They all have opinions/feelings/emotions about the market, which can be probed and mined for predictive value.
Academics will often characterize “sentiment” as inherently – definitionally – misguided.
- “Investor sentiment is a belief about future cash flows and investment risks that is not justified by the facts at hand.”
But this is going too far. Sentiment is what it is – a melange of viewpoints, some of which may be more accurate, more “justified,” than others – embedded in an emotional matrix that often has a certain definable “grain” – positive or negative – which can be read with confidence as a market indicator.
Making Use of Sentiment
Exploiting “sentiment” to help make investment decisions involves a two-stage process.
First, you have to get access to people’s opinions, which is not always easy. Just asking them, in polls or surveys or focus groups, is one way to do it, but the answers may not be reliable (as the recent failures of political polls have reminded us). You can listen in on their Twitter traffic, or read what they write on social media posts, in blogs or online articles. More reliable are measures of behavior, such as the traffic in financial options (puts and calls), or patterns of funds flow (money going into or out of certain asset categories). In the digital era, creative sentiment harvesters track Google searches (e.g., searches for the phrase “gold price” – a very simple metric which mimics the much more complex volatility index, known as the VIX), or mine machine-readable text sources for words and phrases that may carry information about what people are thinking.
The second step is interpretation. Retail investors are simply wrong most of the time – so interpretation is easy. Do the opposite. It is often a fruitful strategy. Analysts’ formal opinions – Buys, Holds, Sells – are also best read “backwards.” Sell recommendations usually outperform Buy recommendations.
Consumer surveys, measures of business confidence, positive and negative headlines in the business press, investor newsletters, word counts – there are many sources of sentiment. And many interpretive models. It is a very active field of research, with dozens of proprietary recipes on offer from academics and the major financial firms. Most of these metrics are read as contrarian. If they scream Buy, it really means Sell — and vice versa. This principle is widely accepted.
What is Sentiment Telling Us Today?
The common sentiment metrics right now are all in alignment… pointing Up, and therefore signaling Down. Let’s consider some examples that are behavioral in nature — they reflect opinions as expressed in concrete actions taken by investors which put money at risk.
Investors who think a stock will go higher in the near future may buy Call options, giving them the right to purchase shares at a defined “strike price” which they think will be surpassed by the rising market price of the company’s stock. If you think a $50 stock will be at $80 in two months, and someone is willing to take the other side of that bet and sell you an option to purchase it at, say, $55 – then, if you are right, you could make a nice profit on that contract. Successful calls are often much more lucrative than simply owning the stock, because of the inherent leverage. The buyer makes money if the share price rises higher than the strike price. So, an investor’s purchase of a Call option reflects positive sentiment.
A Put option is the opposite. It is a right to sell the shares at a defined price. It is a bet that the market price of the shares will fall even further than the strike price. Puts are expressions of negative sentiment.
The ratio of Puts to Calls (negative to positive bets) in the market as a whole is a classic sentiment metric. If the volume of Puts is much greater than the volume of Calls, overall Sentiment is negative – and it is often a signal that the market will rise soon! If Calls are more prevalent, indicating a surplus of optimism, there will probably be a market downturn.
Here then are the Put/Call Trends over the past 12 months, plotted against the S&P 500 Index (from Barron’s, with permission).
Here is how one would read this as a sentiment metric.
The contrarian character of this metric is obvious. The accuracy of the signal is striking. There seems to be a threshold at about the 70% level – a de facto “normal” ratio – a modest 30% surplus of Calls relative to Puts, which reflects a general bias towards optimism among investors. When the ratio crosses below that line — meaning that optimism grows even stronger than normal – it is followed by a market downturn about 2 months later. This happened twice in 2020. When Puts begin to surge and fill the market with negative sentiment, the threshold crossing was followed in about 1 month by a market upturn.
At the moment, the Put/Call ratio is crossing the threshold again heading downwards – a fast-growing surplus of Calls over Puts, optimists over pessimists — which classically signals an imminent market downturn.
The Short Interest
Short sellers often bet on a decline in the value of a particular stock. This is barrel-strength negative sentiment. Selling a stock short is not just expressing a negative opinion. It has the aura of a self-fulfilling prophesy – the very act of shorting a stock drives down its price. Short sellers piling onto a stock are both an effect, and a cause, of weakness in a company’s situation.
If you reasoned thus – you’d be wrong. Contrary to what one might expect, an increase in the short interest – higher negative sentiment – often precedes a rise in the share price. And vice-versa.
Here is a chart of the short interest in Chipotle, from mid 2011 to mid 2016. The company went through two bouts of intense short selling in late 2012 (14% of the total shares shorted) and again in the first half of 2016 (16% shorted), driven by operational and food safety concerns. In between, the short interest fell to a low of about 2% at the beginning of 2015.
These turned out to be contrarian signals. The spikes in the short interest were followed by strong multi-year upward trends in the share price. The 2015 trough was followed by a 3 year decline.
Today, the total short interest for NASDAQ stocks is falling.
- “In fact over the past five years our short interest ratio, short interest as a proportion of shares outstanding, has fallen from 4.25% to around 3.75%.”
The decline accelerated in the last year. The short interest is now lower than it has been since early 2011. Meanwhile, the NYSE short interest has declined 23% just since March.
As a contrarian signal, this distinct weakening of negative sentiment would normally be interpreted as a forecast of an imminent market downturn.
Insider Selling/Buying Ratio
Insiders are executives and board members of public companies who are required to promptly report stock transactions to the SEC. Normally, they report more selling than buying. Much more. There are many reasons an insider might sell (taxes, tuition, real estate purchases, diversification …) but only one reason to buy. The weekly report in Barron’s – shown here – calculates the “neutral” range at about a ratio of somewhere between 12:1 and 20:1.
Interestingly, the Insider Sell/Buy Ratio is one of the few classic sentiment metrics that is not contrarian. A rise in selling is an overt expression of negative sentiment, which in fact often signals an impending decline in the company’s shares. In the aggregate, insider transactions also predict the market movements in a non-contrarian fashion.
Today the sell/buy ratio is quite high. This would also normally signal a market decline in the near future.
Overall, general retail sentiment is “exuberant.” Market multiples are high – another contrarian signal, as a rule. Proprietary metrics show high levels of positive sentiment. For example, Citigroup’s proprietary panic/euphoria index locates the current market mindset firmly in the “euphoria” zone.
Read the fine print: “euphoria” projects that the market has an “80% probability of being lower one year later” — which sounds almost-for-sure ominous. The “panic” reading in March indeed proved to be a contrarian signal, followed by a strong market upswing.
Market sentiment today is pushing towards extreme optimism. Following the contrarian logic, shouldn’t we expect a negative outcome? A Barron’s headline this week invoked memories of the Dotcom surge 20 years ago: “From Airbnb to Tesla, It’s Starting to Feel Like 1999 All Over Again. It May End the Same Way.”
Has the Pandemic Changed the Psychology of “Sentiment”?
The interesting unknown here is whether these metrics have been “distorted” by the extraordinary shock, and the unprecedented character of the pandemic. It may take some time to figure this out.
The market has (mis)behaved in strange ways. As has been widely commented, there is a “disconnect” between the news of the day, which has been mostly dire since March (until the Pfizer vaccine announcement) and the stock market’s surge back to record-breaking levels. Investors have apparently shaken off the catastrophe of hundreds of thousands of Covid deaths and the devastation of significant segments of the economy. Go figure.
The economists at the Federal Reserve Bank of San Francisco recently launched a new sentiment metric called the Daily News Sentiment Index (DNSI) – defined as “a high frequency measure of economic sentiment based on lexical analysis of economics-related news articles from 16 major US newspapers compiled by the news aggregator service LexisNexis.” It has been calculated back to 1980 on a daily basis. In general, the DNSI has been found
- “to move downward with key historical events that have a significant impact on economic outcomes and financial markets, such as the start of the first Gulf War in August 1990, the Russian financial crisis in August 1999, the terrorist attacks of September 11, 2001, the Lehman Brothers bankruptcy in September 2008, and federal government shutdown in October 2013.”
The peak decline in the DNSI due to the Covid shock is smaller than in the previous crises cited in the passage above. Puzzling. Did the Covid crisis really impact U.S. market sentiment less than, say, the first Gulf War? Or…whatever it was that happened in August 2011 (which is a bit of a mystery)?
Consumer sentiment (as measured by the Michigan Consumer Sentiment Index) is more positive. It also fell in March, but remained at about its average for the last 20 years, and well above levels of negativity reached in previous recessions.
The real question is: What does the Covid sentiment signal mean? What does it tell us about the direction of the market in the coming months? The data is the same, but has the interpretation changed? Is “sentiment” telling it true, as per normal – “backwards,” contrarian — or is it giving us “True North” true, what-you-see-is-what-you-get true? Does the current high level of positive sentiment retain its stereotypical significance – i.e., does it project a market decline? Or is there an idiosyncratic component of sheer existential relief at the positive vaccine news, overriding the “irrational exuberance” allegations (which support contrarianism), and inverting the normal interpretation?
We’ll soon see. Perhaps a little giddiness is in order as the vaccines start rolling out this week. Relief from catastrophe can be a powerful tonic. The economic impact of the pandemic has been compared to the impact of World War II, in terms of the massive government stimulus it called forth, and also to some degree the public mindset. Perhaps the endgame will also be similar, in terms of sentiment and the market. As the war approached its conclusion, most observers foresaw an economic relapse, perhaps coupled with a burst of inflation. Instead, the American economy launched on a multi-decade expansion. After languishing for ten years following the 1929 crash, the Dow Index more than doubled between 1942 and 1952. The pent-up demand today may be similar, a coiled spring, ready to explode upwards when the crisis is (suddenly) over. Sentiment metrics may be pointing true. For once. For now. It happened on VE Day and VJ Day. Maybe VV Day (virus victory day) will be similar. We could see sailors kissing nurses in Time Square again soon.