The stock market is driven by the psychology of those operating and participating in it. A very useful approach to spotting changing environments (usually measured in time-frames of many months at least) is looking at metrics that summarize this psychology and the consensus opinion that market participants form about the market.
That is for two reasons.
One, if everybody (or most) has the same opinion, everybody (or most) have already acted on it. That means few to no capital is left to act in that market direction, often marking a change in trend or environment.
Two, the consensus market participant is easy to sell to or buy from.
The stock market works in future time and is a discounting mechanism. Institutional money will accumulate stocks when sentiment is low, everyone is of a ‘bearish’ opinion, and start marking stocks up. When trends become more pronounced and obvious, they will start selling to the public, the amateurs, that come late into the market when they perceive it to be ‘safer’.
But what’s obvious, ‘safe’, or consensus, is almost always wrong or at least late in the market – when the amateurs, including many financial advisors and money managers, start becoming very happy about the market, it is often selling time for the institutional money who appreciates a liquid market bustling with hopeful buyers.
The media machine is complicit in this, either paid by the smart money to encourage small traders and private individuals to buy into a given market environment, or too dumb to know better. But the smart institutional money, the group of actors which demand/supply makes or breaks markets, is not buying anymore – they’re taking profits. Without their firepower, markets can’t advance much further in smooth trends and become stale, stagnant & volatile. All the amateurs and late-comers can do is talk and hype each other up, they sure don’t have the cumulative buying power to keep a whole market afloat.
When the consensus machine of media, people, and their sentiment switches to a happy-go-lucky state, or worse, exuberance, of course you shouldn’t dump all your stock right away – but it pays to be aware of such a development when gauging how far a market top could be off.
The reverse is true for market bottoms – extremely bad sentiment encourages the likelihood of a market bottoming.
Here we will be looking at a couple of sentiment indicators that experience has shown are among the few useful ones. We won’t be looking at economic or consumer gauges (e.g. the SBET SBOI or the University of Michigan CSI) – we’re focusing on investor sentiment.
Sentiment indicators
As for breadth measures, there are a ton of metrics that people have developed over time to look at sentiment. This can quickly become confusing at best and full of contradictions at worst.
The best approach is to focus exclusively at the simplest and most impactful metrics, and always contextualize them with other market health measures.
Two points to always remember:
- for sentiment, only extremes matter, the rest is noise
- sentiment indicators are much more useful for spotting market bottoms than tops (over-bullish sentiment consensus can stay out of whack for much longer than short bursts of genuine cold-sweat panic)
- sentiment indicators are helpful only in the context of a larger market picture; they are most certainly not timing signals to be used in isolation
Bull/bear ratio of financial advisors
Measures of bullish and bearish opinion are widespread, but most contradict each other too much and contain redundant information. Don’t pay attention to most, including the AAII.
What we need to look at are those bull/bear opinions that have potential to impact a large number of ‘dumb money’ decision-making – thus we need to look at financial advisors and newsletter writers.
The finance service provider Investors Intelligence publishes an analysis of how many financial newsletter writers have a bullish or a bearish opinion of the stock market at a given time. By taking the ‘bull/bear ratio’ of the two, we can see where consensus lies – as is so often the case, consensus is often wrong and point to approaching tops or bottoms.
For the interpretation, as for any sentiment metrics, it’s important to realize that only extremes matter – the broad range of ‘middle values’ are noise and don’t mean anything:
- Values below 1 often signal a strong washout of bullish sentiment, a good breeding ground for market bottoms.
- Values above 3 confirm a strong rally but often indicate “overbullish” consensus reflecting an approaching top (this can however stay like this for a long time, see caveat at the bottom).
Again, remember that this ratio is not a timing signal, but rather sets a condition that makes a turn of the market from up- to down-trend, or vice versa, more likely.
The volatility index (VIX), a.k.a. the ‘fear gauge’
Options traded on the S&P500 reflect the forward expectations of market participants – including amateurs and professionals – on impending market moves and their intensity, i.e. the “implied volatility” of the S&P500. The VIX is thus a measure of expected volatility, not of volatility per se – which makes it a sentiment indicator.
The VIX measures the cost of holding a bullish expectation vs. the cost of holding a bearish expectation over the next 30 days, via comparing options premiums paid for 30-day S&P500 index Call & Put options.
This index is often called the ‘fear gauge’, though this is not necessarily what is important about it – by and large, it is a measure of not only implied volatility of equities but also of the implied volatility and thus risk inherent to portfolios holding equities. The risk of losing unrealized gains, experiencing paper- & realized losses and rising margin requirements will go up for any fund running long-only or long/short portfolios when the VIX rises. Risk in the portfolio will rise without the fund managers doing anything at all, and thus the VIX is an indication for the likelihood of de-risking behaviour of institutional money.
Since de-risking (typically involving reduction of exposure and/or hedging via derivatives) for portfolios consisting of hundreds of millions or billions of dollar gross notional exposure can take quite some time and at lot of market operations, transitions in the VIX can suggest a deleveraging and an impending change in the general supply/demand dynamic. There is no rulebook here; it all depends on context of where the market is, where it has been, how the internals are positioned, etc. All in all, the “fear” of (implied) volatility can become the reason for realized volatility.
Of course, the VIX can also be read traditionally as a gauge for the directional opinion of market participants, and that opinion can become victim to a certain consensus bias as well.
Low or dropping values indicate that most market participants have a bullish expectation, i.e. there is low implied volatility, high leverage and few hedging in equities. That is because if the market drifts up consistently, and has been doing so for a while, more and more people believe that it will continue doing so. Fewer and fewer people will buy Put options (i.e. downside bets-/hedges/-‘insurance’), and their cost will drop. Of course, as everybody moves towards one side of the boat, at some point it can topple over – every few months or years, your can see visible VIX spikes in the chart below.
High or rising values in the VIX indicate imminent de-risking, expectation of forward volatility, hedging, and of course some “fear”.
Again, the extremes are what really matters:
- Values, often sharp spikes, over 40 suggest an onset of de-risking/fear/panic consensus. The implications of this are highly contextual, but the main message is often that of a tectonic shift in market dynamics of some kind at least in the short-term. For example, extreme spikes often coincide with late-stage bear market “panic-selling dumps” before a major bottom is experienced. They are however by no means an accurate timing signal, often happening months ahead of a turn.
- Values under 20 suggest complacency and “over-bullishness”. Again, they are not timing signals, but rather conditions that need to be read in context. Markets can stay in this sentiment state for a very long time, especially since the advent of central bank quantitative easing, and the expectations of the infamous “Fed Put”.
The same caveat as for breadth metrics applies here – never use sentiment indicators for timing.
Reading sentiment is what’s called “contrarian analysis“, and contrarians are often right eventually but as good as never on time. In fact, they’re almost always too early.
Of course, sentiment indicators do have their place – to generate supportive information in a larger analysis of the wider market health. Using sentiment as timing mechanism can make you be right but still lose all your money, because the market is driven by psychology – and the latter can sometimes get more out of whack than anyone would believe. As J. M. Keynes is quoted, “Markets can stay irrational longer than you can stay solvent“.
