Equities: The Debate Over Market Sentiment Has Become Sterile
06/05/2017 | 11:02am CEST
Many pundits argue that the U.S. equity market has much further to run due the persistent skepticism concerning this rally and relatively high level of bearish sentiment. To be sure, sentiment surveys have been muted, as the elusive “extreme optimism” reading has yet to be seen.
First, consider the American Association of Individual Investors survey. We took the ratio of Bears to all respondents (Bulls + Bears + neutral responses) and plotted the 4-week moving average of this weekly survey.
The track record of matching low relative Bearish readings with market tops is dubious, at best. Todays Bearish sentiment is right in the middle of the historic range
not much different from readings seen at previous market highs.
The CNN Money Fear & Greed Index, which looks at options activity, market volatility, stock price strength, junk bond demand, breadth, and safe haven demand, is shown below. The index got to extreme greed several times in the past year, but has pulled back to a more muted greed reading. Again, by this measure, Bulls are correct. The market is not at excessively optimistic levels.
Finally, our WMA Market Sentiment Indicator has remained in the optimism mode since the U.S. election. We touched extreme optimism in mid-December, which gave way to an 8-week stall out on the S&P 500. Note that the October 2007 top corresponded to an optimism reading of about 70, not a surged into extreme optimism.
Besides the sketchy historical track record of sentiment extremes coinciding with market tops, we believe that this cycle is different (yes, for once we are uttering the dreaded this time is different phrase). Our argument consists of two points. First, many are dumbfounded by the run that the U.S. equity market has made on a relatively shaky fundamental backdrop. By our logic (wrong or right), stretching the market further, without any attempt to sell-off/take profits, is not going to make Believers out of those questioning the market today. As we mentioned in last weeks Commentary, the S&P 500 has extended its streak without a 5% drawdown to 236 consecutive sessions, not seen since 1996. Participants watching this streak in awe are not going to be throwing money in this market, if not under the constraint to invest. And as for fund managers forced to invest, its hard to imagine many holding back at this stage. In sum, for sentiment to move to extreme optimism, we believe the reset button needs to be hit first to draw more money in and more skeptics into the bullish camp.
Our second reason why sentiment does not matter today is that machines, unlike humans, dont feel emotions. We have been blaming central bank extraordinary policy action for the current market excesses. While central banks set the ball in motion early in this bull cycle, today quantitative algorithms at hedge funds and other institutions, notably the high frequency trading strategies, are perpetuating the seemingly nonsensical, relentless price advance. Algorithmic trading is a system of trading which facilitates transaction decision making in the financial markets using any simple or complex logics, based on user inputs. No emotions enter the equation. Quantitative algos in general, and high-frequency trading specifically, involve computers executing millions of orders a second and scanning dozens of public and private marketplaces simultaneously. They can spot trends before other investors can blink, changing orders and strategies within milliseconds.
Why is this cycle different? Simply because the amount of trading done by machines has never been greater. Quant-based hedge funds gained $4.6 billion of net new assets in Q1 2017 and now hold 30% of the $3.1 trillion hedge fund assets. Trading by quant funds has soared to 27.1% of all stock market trading, up from 13.6% in 2013. And we are just talking about the hedge fund industry. Although precise figures are elusive, stock exchanges say that a handful of high-frequency trading financial institutions now account for more than half of all trades on the U.S. exchanges. We would guess that well over half the trades done on exchanges involve no emotional reasoning.
Clearly algos are programmed with a bias to buy. The run in stocks has left human investors scratching their heads. Since everything goes up, even small dips are jumped on as big buying opportunities by these algos. Who do you believe caught the falling knife on May 17 when the Nasdaq-100 plummeted -2.5%? Not an emotional human. Moreover, machine learning teaches algos that small dips equate to big gains, as this becomes a self-propagating code.
As a result, human trader euphoria would seem to be much less important this cycle than in the past, and the hunt for a sentiment extreme is likely to be a futile endeavor in a machine-dominated market. And unlike investor euphoria, no one can predict when the algos will stop buying equity indexes. Investors should continue to focus on holding a portfolio of companies with strong fundamentals (readers can verify the quality of each of their holdings under the analysis tab on the individual stocks 4-Traders page) and tune out the market sentiment debate between bulls and bears.