“There’s a bull market somewhere” is an old market saying. Over recent years, driven by a global economy flush with central bank liquidity, it is tempting to substitute the “somewhere” with “everywhere”.
The perception that all of the major asset classes are fundamentally expensive is widespread with many investors struggling to find value in equities (developed and emerging), bonds (government and private) or real estate.
It is not hard to be sympathetic to such views and also recognise, and understand, worries that with central banks turning off – or signalling their intention to turn off – the liquidity tap, the uptrend in global asset prices could be corrupted.
That said, as made clear on many occasions, overvaluation is not a reliable tool for timing when markets are about to undergo a significant correction. This has not stopped investors from establishing some rather sizeable shorts in some assets.
From a highly unscientific scan of the financial media, two stand out, longer-dated US government bonds and Tesla shares (two assets that have very little in common, aside from current investor positioning).
The logic behind these shorts appears compelling but does this mean they will be winning bets?
As discussed in a previous Market Insight, crowd-sourced sentiment indicators suggest public confidence towards the global economic recovery is faltering. Since these results were published a month ago, this decelerating undercurrent has, if anything, strengthened.
Updating the charts showing how economic growth sentiment has evolved over the past month, amongst developed economies, all of those we track are now falling (indicated by the outer most ring being coloured red – see exhibit below). Whereas growth sentiment in developing economies has proved more resilient, the outer most ring is still predominantly red.
Exhibit 1: Crowd-sourced Economic Growth Sentiment
Concomitant with this increased negativity towards global economic growth prospects, inflation outlook sentiment, the sentiment-based proxy of inflation expectations, has also been wilting. The number of countries where inflation sentiment has fallen over the past month may not be as widespread as it is in relation to economic growth, nevertheless, it is a tangible change.
Exhibit 2: Crowd-sourced Inflation Outlook Sentiment
The macroeconomic picture suggested by these crowd-sourced sentiment indicators is, therefore, one of decelerating economic growth and an abatement of rising price pressures. Translating this into a standard business cycle regime framework, this would equate to Regime 4, with the prospect of Regime 1 not too far off in the distance – see exhibit below. (This may seem like a dramatic shift relative to where economic data indicates where US business cycle is, but bear in mind the input data is where the crowd “thinks” the economy is going not where it is presently.)
Exhibit 3: Stylized Business Cycle Regimes
Naturally, one expects asset price return characteristics to be markedly different between these four business cycle regimes and the preliminary work on the sentiment-derived business cycle regime classifier bears out such expectations.
Given these classifiers are using crowd-sourced sentiment indicators to identify the regime, as opposed to more standard macroeconomic variables such as GDP growth, PMI surveys or inflation indicators, there is only the data to identify the business cycle regime since 2008. With this caveat in mind, the exhibit below shows the average annualized return by asset class in each of the four regimes.
Exhibit 4: Asset Class Returns By Business Cycle Regime (2008-to-date)
The least amount of variance between the four regimes was in money market returns, an unsurprising result given the extremely accommodative monetary policy stance adopted via the Fed throughout much of the period in question.
Between the other two asset classes, there was a more pronounced divergence in performance. As expected, stocks did best in the early stages of the economic expansion (Regime 2) whereas government bonds did best during the contraction phases (Regimes 4 and 1).
The only surprise was that equities also fared well in Regime 4, a macroeconomic mix that does not readily appear to be favourable. Aside from potential sample size issues, always a problem when using new alt-data sources, one possible explanation is that although there were several instances when crowd confidence in the post-Great Recession recovery faltered, they each proved short-lived. Most likely this was because for much of the period in question the close proximity of the target Fed funds rate to the zero lower bound made the Fed unusually sensitive to downside risks boosting investor confidence in the equity market friendly “Yellen put”.
As the exhibit below illustrates, crowd-sourced economic growth and future inflation sentiments (aggregated up to form a sentiment-based nominal GDP proxy) typically move in sync with the nominal 10-year US Treasury yield. With US economic growth and inflation outlook sentiments having dropped sharply, the clear implication is that public perceptions about the US macroeconomic situation have moved to a more bond-friendly regime (4, or possibly even 1).
Exhibit 5: Crowd-sourced US Nominal GDP Growth Sentiment Vs. 10 Year Treasury Yields
Lending weight to this assessment, the crowd-sourced sentiment towards US equities has also seen a very significant drop in positivity over recent weakness – see exhibit below. In level terms, sentiment towards the S&P500 is best described as neutral, but the momentum of sentiment change is very negative, indicating that the balance of risk is tilting in favour of a downside break in the S&P500.
Exhibit 6: Crowd-sourced Sentiment – S&P500
In direct contrast to the indications from the crowd-sourced sentiment indicators, CFTC data show speculative shorts in the US 10-year Treasury note future at a record high. This suggests that a fairly significant number of investors are positioned for the much-anticipated bear market in US government bonds (3% yields in the ten-year segment of the curve – a level the market has been flirting with recently – has been a major focal point).
Such positioning could, of course, reflect non-cyclical factors not captured by our sentiment data. These include potential offloading by Chinese official accounts as part of the US trade war (the Market Insight discussed, and dismissed, this threat, referenced in footnote 1 above) and/or worries about possible forced selling of government bonds by risk parity funds to respect their target portfolio parameters in the event of a major downturn in stocks. It could also be driven by perceptions that the Fed’s reaction function has changed under Chairman Powell and US central bankers are less inclined to err on the dovish side.
All are possible.
However, even if these are motivating factors, the drop in US macroeconomic sentiment suggests the bond bears face significant cyclical headwinds. Even if the Fed’s pain-threshold is higher under Powell than Yellen, it still has one and faced with a marked downturn in growth/emerging disinflationary forces and an equity market on the slide, at a minimum one would expect a dialling-down of the monetary policy normalization rhetoric. On balance, therefore, we would caution that now is probably not the best time to have a large short position on in US Treasuries.
Turning to the other “significant short” that attracted our interest recently: Tesla.
Tesla and its stock price were last written about a year ago when the stock was trading around $330, which warned that the near doubling in the electric car company’s share price during the first half of 2017 was no longer being matched by rising crowd positivity. On that basis, there was a consideration that the share price correction had further to run. Although Tesla’s share price is 10% lower today than when that article was written (it was 20% lower in the aftermath of Musk’s recent, being polite, “unusual” earnings call), given the stock managed to hit record highs in the interim it was a painful ride.
As acknowledged in the aforementioned research note, Musk is very much a corporate evangelist and he attracts considerable attention in the media. He is, in other words, a character. The problem though with having an adoring public is that they can be a very fickle beast.
On the back of several accidents, some fatal, and amid delays in ramping up Model X/S production, the financial segment of the crowd has become decidedly bearish on the stock’s prospects. Indeed, after the May 3rd conference call, short interest in Tesla stock hit more than 40 million, the biggest short in the US stock market.
Like the US Treasury short, the rationale behind the trade appears compelling. In addition, to repeated failure to hit production targets, mainstream car producers are set to release a slew of electric cars onto the market over the coming years, companies with stronger financial reserves than Tesla which continues to burn cash at a rate of $ 6,500 per minute according to Bloomberg estimates.
Such concerns about Tesla’s long-term future (Musk has always been about “buy the dream” – see his tweet in the aforementioned research note) many consider its share price to be fundamentally overvalued. Indeed, recently one hedge fund analyst estimated its fair value at zero. (Musk’s April Fools’ joke tweet used in the feature image above did little to defuse such speculation).
Consistent with such bearish perceptions, crowd sentiment towards Tesla has slumped this year and is now extremely negative. In fact, the only time it has been lower was in August 2016. (The near doubling in Tesla’s share price over the following 12 months was a classic illustration of our “crowd fail” concept).
Exhibit 7: Crowd-sourced Sentiment – Tesla
Given the outstanding Tesla stock available to short via inclusion in stock lending programmes is estimated to be around 6.5 million shares compared with a cumulated short position of around 40 million, and given the extreme pessimism being expressed by the crowd towards the company, it is clear which is the weaker side of the market at the moment, i.e. a move higher in Tesla’s share price would constitute a pain trade.
Moreover, with Musk having announced he purchased almost $10m of Tesla stock earlier this week, and using his Twitter account to publicly display a combative mood (see exhibit below), just like Treasuries now is probably not the best time to have a large short position, irrespective of what you think about the company’s underlying fundamentals.
Exhibit 8: Musk In Combative Mood
Amareos sentiment analytics incorporate Thomson Reuters MarketPsych indices.
Photo by Hans Eiskonen eiskonen (https://unsplash.com/photos/wn57cSQ7VzI) [CC0], via Wikimedia Commons
More on Tesla
Electric Cars Will Not Save the World
Electrical vehicles (EVs) are not a recent innovation of only the past years. At the beginning of the 20th century, about...
The Paypal Mafia: How eBay’s Cheque Seeded a Generation of Tech
In January this year, eBay announced it would not be continuing with PayPal as its back-end payments service as of...
Humans Will Still Be in Control of Self-Driving Vehicles
Self-driving cars were born with the objective to eliminate the need to have a human behind the steering wheel. Driving...