Wednesday, 2 August 2017

Macro and Credit - The Barnum effect

"If you look and act as if you believe in what you are doing, you will be able to sell even a bad reading to most of your subjects" - Ray Hyman, 1977 about the Forer effect.

Looking at the on-going rally in risky assets, making it a "Goldilocks" period for credit in particular thanks to very low rates volatility (when the MOVE index is stuck that is...), we reminded ourselves for our title analogy of the "Barnum effect". The Barnum effect also called the Forer effect is a common psychological phenomenon whereby individuals will give high accuracy ratings to descriptions of their personality that supposedly are tailored to them but that are in fact, vague and general enough to apply to a wide range of people. The Forer effect is also known as the "Barnum effect". This term was coined in 1956 by American psychologist Paul Meehl in his essay "Wanted — A Good Cookbook". He relates the vague personality descriptions used in certain "pseudo-successful" psychological tests to those given by showman P. T. Barnum who is (albeit erroneously) said to have stated that "there's a sucker born every minute". 

This effect can provide a partial explanation for the widespread acceptance of paranormal beliefs and practice such as astrology, fortune telling, graphology, aura reading and of course, central banking. Given it was recently the anniversary of Le Chiffre aka Mario Draghi "Whatever it takes" bumblebee moment, we thought our chosen title and even our above quote where appropriate choosing of ours. Obviously the latest dovish "Barnum statements" from our "Generous gamblers" aka central bankers, have had the desired effect in leading to a continuation of "Goldilocks" particularly in credit where the "trade du jour" remains "carry on". Clearly central bankers have become masters at "Barnum statements" also called the Forer effect in which the interlocutor finds valid the statements they make given the statements are generalization that could apply to almost any kind of economic situation. Such statements are used by fortune tellers, astrologers, and other practitioners of chicanery to convince customers that they, the practitioners, are in fact endowed with a paranormal gift like our wizards in central banking: 
"The moral of the Barnum Demonstration," according to Professor of Psychology at California State University, Fullerton Michael Birnbaum is that "Self-validation is no validation. Do not be fooled by a psychic, quack psychotherapist, or a phony faith healer who uses this trick on you! Be skeptical and ask for proof. Keep your money in your wallet, your wallet in your pocket, and your hand on your wallet."
Same applies with powerful narratives stated in financial markets, be skeptical and contrarian when needed. For instances, in January in our conversation "The Woozle effect", we indicated that we were ready to take a contrarian stance against the "Barnum effect" affecting the long US dollar crowd":
"If indeed the US administration is serious on getting a tough stance on global trade then obviously, this will be bullish gold but the big Woozle effect is that it will be as well negative on the US dollar. It appears that from a "Mack the Knife" perspective (aka King Dollar + positive real US interest rates), it will be rather binary, either we are right and the consensus is wrong thanks to the Woozle effect, or we are wrong and then there is much more acute pain coming for Emerging Markets, should the US dollar continue its stratospheric run. From a contrarian perspective we are willing to play on the outlier." - Macronomics, January 2017
And of course willing to play on the outlier has been the correct path so far from a contrarian perspective, countering therefore the "Barnum effect" played so effectively by the wizards at the Fed. 

In our most recent musing we also hinted that we had become more positive on gold and gold miners and increased our allocation. We do practice magic, but in our own terms...

In this week's conversation, we would like to look at the continuation of the final melt up in risky assets in particular credit thanks to the "Barnum effect".

  • Macro and Credit - European High Yield and credit spreads overall - These go to 11
  • Final charts - Credit Quality ? Lower for longer...

  • Macro and Credit - European High Yield and credit spreads overall - These go to 11
The Barnum statements from our wizards at the helm of central banks have had the desired effect in neutering bond volatility as seen in the repressed bond volatility gauge MOVE index. This has indeed has had the desired effect in prolonging the "Goldilocks" environment for the better crowd. Carry players alike leveraged investors are in love and in particular with one thing, namely low rates volatility. 

As we pointed out in our June conversation "Goldilocks principle", as we indicated recently, it seems to us that in terms of risk-reward, High Yield has moved from expensive or rich to very expensive and in particular Euro High Yield. While our bullet point is a veiled reference to Spinal Tap movie when it comes to central banking "Barnum effect" and their Marshall amplifier, in this movie Spinal Tap lead guitarist, Nigel Tufnel could explain you the importance of going to 11. In our recent musings, we pointed out that in the on-going credit "Goldilocks" scenario, already expensive asset classes would become even more expensive, going to 11 that is.  The continued yield compression à la 2007 is not a surprise given we think we are indeed in the last inning of the current credit cycle. As indicated by Deutsche Bank European High Yield note from the 2nd of August entitled "Justifying tighter spreads", going to 11 is we think akin to the infamously famous "this time it's different" "Barnum effect" mantra:
"Justifying tighter spreads
July provided a return to positive performance for EUR HY credit as the index returned +0.83% including financials and +0.74% for non-financials. YTD nonfinancial returns now sit at +4.4% having been at +3.3% as recently as the second week of July. Looking at Figure 1 we can see that spreads have tightened to fresh post-GFC tights for both BBs and Bs following more than 10bps of tightening for both rating bands during July.
In terms of where current spreads sit relative to their own histories BBs have only been tighter 13% of the time since 2003 while Bs have only been tighter 9% of the time. In fact as we can see in Figure 2 where we show where current spreads trade relative to their own histories across ratings and currencies, 11 of the 21 analysed indices are currently trading in their tightest quartile.
There is no doubting that valuations are looking increasingly stretched and we would certainly argue that the risks are very much weighted to the downside. However with market volatility measures remaining at such low levels we can make an argument that spreads could actually be even tighter. Specifically it's worth noting that the VIX index closed below 10 for 10 consecutive days during July, this is the longest such run for the series. Previously the longest run below  10 lasted 4 days back in 1993. That said we have moved off of these lows in the last few days.
In Figure 3 we update our simple model looking at where rates, equity and FX implied volatility suggest credit spreads should be. The main takeaway here is that despite seeing spreads tighten and a slight shift higher in volatility in the past few days our volatility implied spread series suggests that HY credit spreads could be even tighter.
Looking at the recent history of the difference between the implied spreads and the actual spreads more closely (Figure 4) we can see that if volatility remains at current levels then it could be argued that BB and B spreads could be 20bps and 10bps tighter respectively to be in line with volatility.

Admittedly the volatility implied spread series is a lot more volatile than actual spread levels. As recently as mid-July it was suggesting spreads should be as much as 60-80bps tighter. Therefore we are mindful of over interpreting the results but this analysis does at least provide us with evidence that credit spreads are merely reflecting the very benign environment that we are currently operating in.
In a world where investors are increasingly looking for the drivers of a sell-off the very near-term direction of travel could still be tighter. As September approaches and we get more news flow around potential central bank actions in both Europe and the US we could see higher volatility that may eventually weigh on credit spreads. For now however we continue to believe that the carry trade will prevail." - source Deutsche Bank
With the current "Barnum effect" at play, of course for the "beta investors crowd" it's carry on and the near term direction for European High Yield is indeed 11, Marshall amplifier wise. So while carry trade remains the "trade du jour" thanks to low volatility in rates, we do think that from a "Great Rotation" perspective, credit wise you would be better off increasing credit quality and reducing your beta exposure as we move towards September. 

You might rightly ask yourselves, how tight can we go, in fear of missing out on the last stage of this epic rally, yet you would be wise to take notice that some astute players have already started to "cash in" and taking a few chips from the casino table as indicated by DataGrapple in their post from the 1st of August entitled "When Is Far Far Enough?":
"Credit experienced another positive session today, with all indices trading near the tightest levels of the year – which are effectively the tightest levels of the last 3 years – reached a few sessions ago. The market is benefitting from the light volumes, low volatility environment that investors foresee in August. If such scenario materialises, why would they want to give away some carry and pay for protection. The performance of credit overall was underpinned by the tightening of the miners and basic material sector. BHP Billiton, Mittal, Glencore and Anglo American all saw their 5-year risk premia reach multiyear low after commodity prices continue their surge on optimism the global economy is keeping its momentum. But down at these tighter levels, after what looked like one way traffic over the last 3 months – since early May, Mittal is 120bps tighter at 166bps, Glencore 70bps tighter at 121bps and Anglo 80bps tighter at 119bps - some 2-way flow is developing. While some investors who were holding short risk positions are throwing the towels, others who were riding the tightening wave are taking chips off the table." - source DataGrapple
Of course some "permabears" have thrown in the proverbial towel and as of late we have seen swath of retail investors on the fear of missing out the grand tricks of the sorcerer's apprentices aka our central bankers and their "Barnum effect" have in earnest decided to jump in. But, credit wise, when European Junk Bond yields hit an all-time low of 2.45%, you need to ask yourself if the risk is still worth the reward in terms of forward returns. We don't think it's enticing anymore and are happy rotating even more into Investment Grade, sacrificing style (rating) over substance (carry). It means starting building up "defense" we think:
"Be fearful when others are greedy, and greedy when others are fearful" - Warren Buffett 
It doesn't mean, it's bunker time with food and ammo stacked up, but it means starting to think seriously about reducing your beta exposure in terms of allocation and probably adding some long dated US long bonds to the mix. Yes, everything might be great macro wise in Europe with the latest raft of releases, but it's when everybody is cheerful then one should be skeptical in the tricks being played in this on-going "Barnum effect". Yes, risk assets continued to rally, with the investment grade credit index breaching 100bp for the first time since 2014, but clouds are lining up for September not to mention potential exogenous geopolitical risks that are known unknowns such as North Korea. When it comes to "Barnum" and circuses, US political gridlocks with the upcoming debt ceiling debate is yet another cause for concern we think. 

When it comes to reaching for 11, and the "Barnum effect" you would probably point out as well to the overall tightness in the credit space particularly in investment grade credit. On that subject we read with interest Barclays US Credit Alpha note from the 21st of July entitled "Breaching the Century Mark":
"Tight and Tighter
The Bloomberg Barclays Credit Index dropped below 100bp for the first time in nearly three years. Aside from a brief period in the summer of 2014, the index has not maintained a spread below its current level since the pre-crisis years of 2005-06. Over that period, the index composition has changed dramatically, with notable increases in the market value of lower-rated and longer-dated debt. The index market value weight of AAA/AA credit decreased over 10%, and debt with more than eight years to maturity increased over 6% (Figure 1).

After we adjust for these differences at an index level, we find that spreads are less than 5bp off their average level for 2005-06, although they remain roughly 15bp off the tights (Figure 2).

Dissecting the index into financials and non-financials, the majority of rating deterioration has occurred in the former. While the term structure for financials remains consistent with pre-crisis levels, lower-rated debt nearly doubled its market value weight over the past decade (Figure 3).

In particular, BBB rated debt comprised only 11% of financials’ market value in 2005-06, while today BBBs are close to 40%, even though subordinated bank debt’s share of financial market value has declined by 50% over the period. This rating migration partly explains the cheapness of financials currently relative to pre-crisis levels: from 2005 to 2006, the corporate index traded roughly 20bp wide of financial credit; that difference is about 4bp now.
However, even after adjusting for subordination and compositional differences, financials trade meaningfully wide of pre-crisis levels (Figure 4 compares senior financial spreads across different rating/maturity buckets between 2005-06 and now).

This is driven partly by concerns about TLAC supply and the increase in USD-denominated European bank debt, which generally trades wider than US bank paper (see Banking on Yankees Going into Summer). That said, we see room for further compression in US bank spreads, especially as the TLAC technical abates (see US Banks: 2Q17 Supply Update: Favorable Conditions Keep the Spigot Open).
Industrials have largely seen a more balanced change in composition (Figure 5).

While there has been an increase in BBB rated and 20-30y debt (the sector’s market value share increased roughly 4% for each from 2005 to 2006), we also see added market value share in AA rated credit and bonds in the 1-3y maturity bucket (each increased more than 3% in the same period). Comparing valuations across the ratings/maturity buckets, A rated paper consistently trades through pre-crisis averages, while BBBs appear more in line or slightly wide, particularly for longer-dated paper. One part of the market that still offers substantial pickup versus 2005-06 spreads is BBB rated 20y debt (Figure 6).

Although we do not see an identifiable near-term catalyst for spread widening, we believe that investors should focus on pockets of the index where relative value looks somewhat attractive. Within the industrials sector, 7-20y BBBs stand out as attractive, trading materially wider than pre-crisis levels, unlike the rest of the industrials sector." - source Barclays
While CDS wise with the CDX IG is within 2 bps of its post-crisis tight of 56 bps in 2014, we are still far from the all-time tight of 28 bps in February 2007. It doesn't mean we can't get to 11, even in Investment Grade, but, we still ponder how long this "Goldilocks" period can last before the return of the three bears. Some may argue that this time it's different, when it comes to credit relative to 2007 and as highlighted by Barclays in their note they would be right:

"In 2007, the biggest factor driving spreads to very tight levels was the presence of synthetic CDOs, which were sizable sellers of protection. During that period, there was significant leverage available, which made it attractive to sell protection even at low absolute spread levels. At the time the index reached its all-time tight, more than half of index constituents were trading inside of 20bp (Figure 8).

In contrast, today there is less leverage available for these trades, and structured credit is not a significant part of the market. Currently, only 10% of IG28 constituents trade inside of 20bp, and unless structured products (or similarly prolific sellers of protection) are meaningfully reintroduced to the market, we do not expect a large number of index constituents to trade at very tight levels. Another difference between the index today and the index in 2007 is quality. IG28 is of lower average quality than IG7 was back in 2007: Baa1 versus Baa1/A3." - source Barclays
Sure the CDO printing press has been fairly muted, yet synthetic activity continues as well to be helping spreads grinding tighter in similar fashion to 2007 albeit not at the same rapid pace we witnessed firsthand back in the days. We do agree though with Barclays premises from their note that a low-beta strategy can be a sound long-term approach to capital preservation, just saying, no matter how mesmerizing the "Barnum effect". Some pundits do think that August will be quiet until we reach September or at least after Jackson Hole in late August, but, rising volumes in CDS options suggest that at least some players are actively hedging their risks. As we indicated recently, gamma credit wise is still relatively cheap to own compared to other asset classes rather than going the negative carry way through CDS index purchasing outright.

Nonetheless, lower rates volatility has always been good for credit as pointed out by Bank of America Merrill Lynch in their Euro Excess Returns note from the 1st of August entitled "Lower rates vol = good news for credit":
"Lower rates vol is good news for credit. In July the European credit market posted another month of positive excess returns. Note that IG euro-denominated corporate bonds recorded 62bp of excess returns performance in July, the strongest performance
over the past year.
  • Duration outperforms. In July the back-end of the curve has outperformed significantly the front-end. 1-3yr credit posted 18bp of excess returns last month, while the 10yr+ bucket posted a gain of 128bp.
  • Beta underperforms. Euro-denominated high-yield credit posted not much more than the excess return of its high-grade counterpart (83bp vs 62bp, respectively). Same story within the IG space with AAAs doing better than BBBs, while double-As beat single-As.
  • Sectors: The vast majority of sectors posted positive excess returns last month. Only Autos were slightly on negative territory. Insurers, telcos and tech names feature at the top of the performance table. On the other side autos and transport were at the bottom of the performance table.
  • Issuers: a number of French credits like Credit Agricole, BNP and EDF were among the best performers. Seven French names are featuring in the top 15 spots of the performance table." - source Bank of America Merrill Lynch
Of course during the Barnum effect, low volatility means outperformance of the beta space including High Yield and its CCC rating bucket.  While the party is still going strong during this summer lull, we do remind ourselves that not only there has been an extension in both credit and duration risk during this long in the tooth credit cycle, but, quality wise, there has been a decay. With interest rate buffer close to zero, as we keep repeating this, in the current low yield environment, both duration and convexity are higher, therefore the price movement lower can be larger. Also in the next downturn we indicated previously we anticipated lower recovery rates this time around. We are not there yet, but the credit cycle continues to gently turn. For our final charts below, while rates might stay lower for longer than expected, for credit, quality wise, it has been trending lower for longer as well.

  • Final charts - Credit Quality ? Lower for longer...
Of course the Barnum effect from our central banks' wizards has had the desired effect in shutting down rates volatility since the "Taper Tantrum", pushing in effect the continuation of a "Goldilocks" scenario for credit markets, carry players and all leveraged investors alike. Yet, something which has been prevailing is that not only in the current cycle investors have reached out for more duration and more credit risk, overall, credit quality has been eroding as per our final charts from Morgan Stanley's Corporate Credit Research note from the 28th of July entitled "Living the HY-Life:
"To size up downgrade potential in the next cycle, we think it's important to remember two points. First, credit markets have grown considerably since the crisis, a phenomenon we have discussed frequently. Low rates and easy liquidity drove a need for yield, which led to significant demand for credit. Spreads compressed, and all-time low yields incentivized companies to issue debt. Issuers responded accordingly. As we show in Exhibit 5, US credit markets (measured by index debt) have more than doubled in size since 2008.
Second, credit quality has deteriorated in a number of ways. Most simply, the ratings quality of the IG index has weakened considerably. For example, as we have mentioned in the past, BBBs comprise 49% of the IG index today vs. 28% in 1997. The decline in credit quality, in fact, accelerated this cycle, with A-rated bonds now representing 37% of the IG market, down from 45% at the cycle tights in mid-2014. Note that 49% of the index is currently BBB, and that is after accounting for $133bn of IG to HY downgrades in 2016 (net of the upgrades into IG).

This ratings decline has been relatively broad-based, affecting over half of the subsectors for which we have data going back to 1997 (see Not Your Parents' Market, 6.2.2017). Looked at another way, the lower-quality part of the IG market (rated BBB) combined with the HY and loan indices makes up 64% of total corporate debt today , compared to 35% in 1991, 48% in 1999, 61% in 2007.
Thus, not only has the absolute volume of credit markets grown substantially, but so has the proportion of debt theoretically at risk of a downgrade into high yield. As an example, the 1999-2003 credit cycle was severe, with a 29% 5Y cumulative default rate over the full period . However, in 1999, the IG index was $1.1tn in size, with 32% or $379bn in BBB debt. Today the market is $4.8tn in size with 49% or $2.4tn BBBs in total.
This is not to suggest that all BBBs will fall into HY by any means. The default and/or downgrade rate could be lower in this cycle than in the past. Our point is to illustrate that actual downgrade (and default) volumes will still be magnitudes bigger. 
Also remember that in every cycle some factor causes valuations to deviate from fundamentals on the way down. In the last cycle, as mentioned above, it was the severe deleveraging of the financial system. In our view, the next time around, meaningful default/downgrade potential given the massive growth in credit markets, combined with post-crisis liquidity dynamics in periods of stress, will contribute to the overshoot as markets are heading lower." - source Morgan Stanley
As we recently said, what goes up often goes down to paraphrase Mark Yusko from Morgan Creek's previous quarterly gravity parabola. We would add that what goes too tight often goes too wide in relation to credit spreads. Next downturn, it won't be any different when it comes to overshooting. The only difference is that we think recovery values will be lower, blame it as well on the increase of Non-GAAP measures embedded within the Barnum effect but we ramble yet again...

"There are two kinds of magicians, the honorable kind and the unethical kind. The ethical magician admits that he or she uses tricks to create illusions. The unethical magicians use the same devices to claim to have magic powers. Magicians do not reveal how tricks performed by ethical magicians are done, in order to preserve the mystery. However magicians make an exception when unethical magicians use their methods to defraud and deceive. Harry Houdini, Amazing Randi, Penn and Teller, are examples of real magicians who reveal secrets to expose phonies, quacks and frauds who claim to have psychic powers, mind over matter, or to communicate with the dead." - Professor of Psychology, Michael Birnbaum.

Stay tuned!

No comments:

Post a Comment

View My Stats