Key Takeaway: A myriad of quantitative signals suggests investors would do well to be positioned rather bearishly (within the context of their respective mandates) as we progress throughout 2H15.
Late last week, it was reported by EPFR that global equity fund outflows totaled $25.9B, which was the largest sum on record (data going back to 2002). U.S. equities experienced a $12.3B outflow – the most in 16 weeks. Per Jonathan Casteleyn of our Financials Team:
“Domestic equity funds continue their streak of outflows, now at 25 consecutive weeks with another -$5.2 billion reined in by investors in the most recent 5 days. The table below shows all domestic equity outflow sequences greater than 4 consecutive weeks in data going back to 2008. We considered a streak of outflows broken by 4 weeks of consecutive inflows. Within these parameters, there have been 8 total outflow sequences with the mean lasting 40 weeks with $105 billion lost on average. The current sequence, as of August 19th, is only the 8th longest in weekly duration, but at -$101.0 billion in total outflows lost, it is the 4th largest in magnitude. With an average outflow of -$4.0 billion per week, the running 2015 outflow is fastest pace on record in our data back to '08”.
So everyone is bearish, right? Right.
While it’s easy to naval gaze at flows data or feel warm and cuddly about the spate of bullish reports from bulge-bracket U.S. equity strategists that hit the tape last week, more thoughtful analysis of trends across global macro markets suggests investors would do well to remain on the defensive. The current setup is not unlike the setup we identified in our 7/20 note titled, “Dangerous New Highs for the Market?” where we flagged the pervasive deterioration of market breadth as a bearish indicator, in addition to other factors.
I: Credit Concerns
Much ado has been made over the trending backup in domestic credit spreads. While certainly not a new factor, we found it worthwhile to analyze what such a backup might imply for U.S. stocks after coming across the following anecdote from Bloomberg:
“Yield premiums on investment-grade debt have widened by 32 basis points over the past three months, according to Bank of America Merrill Lynch index data. Since 1996, there have been five occasions when credit spreads showed similar expansions. Two of them preceded recessions in 2001 and 2007 when stocks went on to drop 50 percent or more. In the other three instances, the S&P 500 fell at least 16 percent while the economy continued to grow. The latest was in August 2011, when the S&P 500 was mired in a 19 percent retreat that almost ended the bull market. Assuming the U.S. economy will be able to avoid a recession this year, as predicted by economists surveyed by Bloomberg, and stocks fall by the average magnitude to reflect similar credit stress in the past, the S&P 500 would hit 1,742, a 18 percent decline from its all-time high reached in May.”
Given the seemingly arbitrary nature of the aforementioned look-back period, we thought we’d perform a similar analysis that holds both the historical OAS observation period and prospective SPX return period constant – at 6 and 12 months, respectively:
Based on this analysis, which uses weekly closing price data going back as far as we can get it (2001), the S&P 500 has declined an average of -11.1% over the following twelve months whenever HY OAS have widened as much as they have over the most recent trailing 6-month period (+116bps).
Obviously the Frequentist nature of this study leaves a lot to chance, but if there’s anything to the old adage “credit leads equities”, there may still be a fair amount of downside ahead for the U.S. equity market.
II: Consolidated Positioning
Heading into the drawdown, hedge fund investors were net short of the S&P 500 to the tune of 127,130 futures and options contracts. More importantly, this lean was -1.5 and -1.7 standard deviations below its TTM and trailing 3Y averages, respectively, which implied that said bearishness was becoming rather crowded.
Looking at that same data set, we see that said positioning had tightened by 86,285 contracts WoW heading into last week’s bounce (data though 8/25). While we won’t know for sure until late Friday afternoon, we’re willing to bet that last week’s dead-cat bounce on waning volume was perpetuated by additional short-covering. Even without that information, we can be sure to conclude that the short position has consolidated with a z-score of 0.4 (TTM) and -0.8 (3Y).
In summary, the data currently suggests that investors aren’t positioned nearly as bearishly as market chatter would imply. Incremental consolidation would imply that investors aren’t at all positioned appropriately for the next leg down.
III: TACRM Signaling Very Bearish
Our Tactical Asset Class Rotation Model (TACRM) continues to be dynamite at prospectively signaling meaningful inflections in performance at the primary asset class level. It’s latest two signals (SELL U.S. equities and SELL international equities) continue to appear appropriate in spite of last week’s short squeeze across global equity markets. The signals prior to that – specifically SELL foreign exchange, SELL commodities and SELL emerging market equities – appear equally as prescient. Refer to slides 6-13 of the following presentation for more details:
With TACRM signaling to reduce exposure to all six primary asset classes, the only thing for investors to do is raise cash – although the “SELL” signal on domestic fixed income, credit & equity yield plays could be sidestepped by deft factor exposure selection (i.e. long long-duration treasuries vs. short credit – particularly junk bonds – as the Fed risks perpetuating deflation by tightening into a global growth scare and a classic #LateCycleSlowdown in the domestic economy).
Other not-so-bullish factors worth highlighting:
- TACRM is not generating a BUY signal for any of the 117 factor exposures the model tracks across the U.S. (47), Developed International (33) and Emerging Market (37) equity markets. In fact, only one (Irish equities, ticker: EIRL) is registering positive volatility-adjusted VWAP momentum across multiple durations. Refer to slides 17-19 of the aforementioned presentation for more details.
- Looking beyond domestic and global equity markets, TACRM is generating a BUY signal for only one factor exposure in the entire global macro universe – the ever-defensive Japanese yen (FXY). Refer to slide 16 of the aforementioned presentation for more details.
- The trending breakout in cross-asset volatility continues unabated, having remained intact since the first week of October. Recall that the model accordions its [independent] historical observation periods according to the trend in global macro volatility, shortening the periods when volatility is trending higher and elongating the periods when volatility is trending lower. The thought behind this is that A) trending breakouts in volatility are usually associated with material changes in the fundamental narrative underpinning any asset class and B) VaR models force investors to de-risk their portfolios and shorten their holding periods on existing and prospective investments. In the context of cross-asset volatility trending higher since 2H14, it is reasonable to conclude that the recent breakout in U.S. equity volatility is unlikely to be a one-off event.
In short, neither domestic nor global macro markets are signaling that it’s appropriate for investors to do anything other than be positioned as bearishly as they can be within the context of their respective mandates.
IV: Broken Market
Keith’s PRICE/VOLUME/VOLATILITY model remains core to our quantitative process. And across all three durations (TRADE: 3 weeks or less, TREND: 3 months or more and TAIL: 3 years or less), the S&P 500 remains broken.
The key takeaway for investors here was best summarized by Keith on today’s RTA Live broadcast: “I’d be a seller of pretty much anything U.S. equities until we get the right flush to the downside.”
All told, the aforementioned three signals suggest to us that domestic and global capital markets are likely to remain under pressure throughout 2H15. We expect a series of lower-highs for almost any factor exposure that isn’t tightly correlated to the U.S. Treasury market.
Best of luck out there,