The Financial & Risk business of Thomson Reuters is now Refinitiv

All names and marks owned by Thomson Reuters, including "Thomson", "Reuters" and the Kinesis logo are used under license from Thomson Reuters and its affiliated companies.

March 2, 2017

Riding The Wave 

by Amareos.

Hokusai’s The Great Wave Off Kanagawa woodblock print is a fine work of art; one we have always found aesthetically pleasing because like many people we can happily sit for hours in front of the sea watching the waves rolling in. Perhaps this is one of the reasons why we chose to work in finance. After all, waves are nothing more than oscillations in a medium, and this medium can take the form of water as in Hokusai’s print, air in the case of music, or asset prices in a financial context.

While the latter may have less artistic merit, as every investor knows riding an asset price wave – “calling the market right” to continue the analogy – is one of the greatest feelings; it is deeply satisfying both financially and emotionally to correctly anticipate how the future will evolve.

The downside is that such waves, especially in finance, are ephemeral. Every investor encounters unexpected, often sudden, movements in financial markets and experiences a “hero” to “zero” feeling. Such transitions are not pleasant, either financially or emotionally. Yet, they are nevertheless invaluable, which is a good thing because they are unavoidable. (If you have never experienced such transitions and have a 100% investment hit-rate then we, along with every other asset manager in the world, would love to hear from you!)

Their value, while perhaps not apparent in the moment, lies in the fact that they provide us with an opportunity to learn. If the future unfolds as anticipated our information set is unchanged, it is only when something unexpected happens, challenging our mental model, that something new is revealed us. That said, to gain as much as possible from such episodes one has to go back and analyse why the outcome differed from what was originally anticipated.

As our regular readers will know, a few weeks back we published a Market Insight warning that the stock market rally witnessed since the US election was at risk of being corrupted because of a change in crowd sentiment, specifically the rise in the Fear sentiment [1]. As we noted at the time,

“a marked rise in crowd fear towards US equities has, historically, been a reliable harbinger of future equity market weakness. Conversely, when the Fear indicator is at historically elevated readings and begins to subside, such as was the case just after the presidential election, it is typically a harbinger of future equity market strength. Right now, given this sentiment’s current trajectory, it should be a concern to equity market bulls.”

In short, the sentiment stars appeared to be aligning in such a way as to flag a potential stock market correction. However, over the past fortnight, Wall Street has continued to hit fresh cycle highs and in a manner that is rather unusual (more on this later).

Subsequent to the publication of this research note, and incorporating other crowd-sourced sentiment elements that we have explored in earlier Market Insight [2], we decided to do a bit more digging to explore how the combination of crowd-sourced Sentiment and Fear previously impacted future US equity market returns. In doing so we developed a checklist approach incorporating crowd-sourced sentiment characteristics based on the following questions:

Crowd-sourced Sentiment Checklist – S&P500

  • Is Sentiment Positive or Negative?
  • Is Sentiment Extreme?
  • Is Sentiment momentum Positive or Negative?
  • Is Fear Rising or Falling?
  • Is Fear Extreme?

The five questions that go into the checklist generate 32 possible sentiment scenarios. For each of these scenarios we then calculated the historic future price returns for the 11-year period for which we have sentiment data. We will refrain from sharing the full set results at this stage as this is still work-in-progress but for the S&P500 crowd-sourced sentiment combination observed in late January (answers: Negative, No, Negative, Rising, No – Scenario 14) the average future returns are shown in the exhibit below together with the equivalent price returns for the S&P 500 over the same period for comparison purposes.

Exhibit 1. S&P500 Forward Price Returns

Exhibit 1


In the short-run (the 30 to 60-day windows) we found little difference between the full sample future S&P500 price returns and those under sentiment scenario 14 [3]. However, beyond the 60-day horizon a significant underperformance is observed under sentiment scenario 14 relative to the full sample future S&P500 price returns. This suggests that perhaps we were just a little pre-emptive in terms of anticipating a downward correction, especially as crowd sentiment towards US equities has continued to weaken and is now marginally negative, while crowd Fear also remains elevated.

That said, as mentioned above, one aspect of the US equity markets march upwards into unchartered territory that has attracted considerable investor attention [4], and has certainly pricked our interest, is the extremely low level of volatility. The VIX index – commonly known as the Fear gauge [5] –  remains at very low levels historically, something that is very much at odds with the rising Fear sentiment observed in the crowd-sourced sentiment data.

One explanation cited for low volatility at the index level is the marked decline in correlation between the equity prices of individual companies within the index, a change widely assumed to be a reflection of investors anticipating divergent effects of the macroeconomic policies to be adopted by the incoming administration. This argument has merit. However, there may be something else at play that accounts for the seemingly benign nature of the US equity market rally.

In a recent tweet [6] we noted that there was a high degree of divergence in the level of sentiment towards US equities dependent upon media-type. This is highly unusual as one would naturally expect the views of the two media types (mainstream more reflective of a professional audience and social media more retail) to be fairly closely correlated. After all, Wall Street luminaries prognosticating on financial TV channels have long been able to move the crowd – and hence markets – for reasons we outlined in the Market Insight referenced in footnote 2 above. Yet, as the chart clearly illustrates, the sentiment divergence between the two media types, which started around September last year, has become even more pronounced.

Exhibit 2. Crowd-sourced US Equity Fear – By Media Type

Exhibit 2


In contrast to the significant negativity being expressed towards US stocks by the mainstream media sentiment, social media is much more positive. This divergence has also been reflected in a high degree of polarity in Fear attitudes. In mainstream media it is at extremely elevated levels while in social media it is at extremely low levels. Recent price action combined with low levels of implied and realized volatility, suggests it is the retail element of the investor base that presently is the key market driving force.

One possible interpretation of this divergence is that it is only transitory. And, in keeping with the old market adage that retail investors are always the last to arrive at the party, this constitutes yet another warning sign that the risks for US equity markets are skewed to the downside.

There is, however, an alternative interpretation, one that could point to a more prolonged divergence.

No one really knows the extent of the fake news problem – that would require accurate ways to determine what is fake news and what isn’t, negating the issue in its entirety – but one impact from the furore it has created has been the eroding of the public’s trust in all media. Moreover, for mainstream in particular, the increasingly subjective nature of reporting during the US election campaign as we documented in an earlier Market Insight [7] was without doubt an exacerbating factor (it fits with the timing of the sentiment divergence).

Even though the subjectivity of mainstream media reporting has returned to more normal levels, the damage to the public’s trust appears to have been done, leading to a weakening in the linkages between the two media types (joe public mentally switching off from the talking heads on the TV).

If this latter interpretation is correct, not only would it account for the decorrelation in the sentiment tones between the two media types, it suggests the unusual divergence could prove to be persistent (trust is a highly asymmetric emotion – “hard won, easy lost”). In turn, this could make Wall Street more resilient than one would expect given the aggregate sentiment indicators as retail investors, tuned-out to mainstream media and hence less influenced by its negative overtones, continue to ride the equity bull market wave.

Sentiment Analytics are based on MarketPsych indices


[1] See:

[2] See:

[3] We have observed this sentiment combination 4.2% of the time over the past 11 years.

[4] Even the Fed appears to be surprised and concerned by the low levels of implied equity market volatility which “appeared inconsistent with the considerable uncertainty attending the outlook for such policy initiatives” according to the minutes from the last FOMC meeting published this week.

[5] Incorrectly so, as we noted in previous Market Insights.

[6]  Our Twitter handle is @Amareos_info. The tweet we are referring to was published on February 13th.

[7] See:

Article Topics

Get In Touch


We have updated our Privacy Statement. Before you continue, please read our new Privacy Statement and familiarize yourself with the terms.×