Sunday 7 July 2013

Credit - The Dunning-Kruger effect

"The truest characters of ignorance are vanity and pride and arrogance." - Samuel Butler, British poet

Watching with interest the impressive volatility in the bond space which has yet to normalize, we thought this week we would use a reference to human psychology in our reference title, given so far our "Central Bankers" mind tricks (call them jedi skills if you want) seems to differ widely, between the Fed and Bank of Japan, between the Bank of England and the Reserve Bank of Australia, and the ECB of course.

For those who have been following us, you know that like any good cognitive behavioral therapist, we tend to watch the process rather than focus solely on the content. 

So why our chosen title you might rightly ask?

"The Dunning–Kruger effect is a cognitive bias in which unskilled individuals suffer from illusory superiority, mistakenly rating their ability much higher than average. This bias is attributed to a metacognitive inability of the unskilled to recognize their mistakes" - source Wikipedia

When one look at how central bankers have been previously apt in preventing formation of asset bubbles or identifying asset bubbles, one can easily be drawn to the Dunning-Kruger effect given that ignorance of standards of performance is behind a great deal of incompetence.

Of course we are not surprised to see the Dunning-Kruger effect at play, given it is a continuation of the "Omnipotence Paradox" of our central bankers or deities:
"In similar fashion, the financial crisis and the consequent burst of the housing bubble which had taken aback the beliefs of some forefront central bankers such as Alan Greenspan; have clearly shown that Central Banks are not omniscient either (omniscient being the capacity to know everything that there is to know).
"Those of us who have looked to the self-interest of lending institutions to protect shareholder's equity (myself especially) are in a state of shocked disbelief." - Alan Greenspan -  October 2008." - Macronomics, 18th of November 2012.

In the Dunning-Kruger effect, for a given skill, incompetent people will:
"1.tend to overestimate their own level of skill;
2.fail to recognize genuine skill in others;
3.fail to recognize the extremity of their inadequacy;
4.recognize and acknowledge their own previous lack of skill, if they are exposed to training for that skill." source Wikipedia

In continuation to our "Omnipotence Paradox" conversation,  we believe this time around that the Dunning-Kruger effect can explain the failures of some economic school of thoughts, namely the Keynesian school of thought and the Monetarist School of thought. We have argued in our conversation "Zemblanity", when looking at the evolution of M2 and the US labor participation rate that both were indicative of the failure of both theories:
"Both theories failed in essence because central banks have not kept an eye on asset bubbles and the growth of credit and do not seem to fully grasp the core concept of "stocks" versus "flows"."

"Credit growth is a stock variable and domestic demand is a flow variable" as indicated by Michael Biggs and Thomas Mayer in voxeu.org entitled - How central banks contributed to the financial crisis.

Obviously one can posit that not only do our central bankers suffer from the Dunning-Kruger effect but they are no doubt victim of the well documented "optimism bias" which we discussed in our "Bayesian Thoughts" conversation:
"Humans, however, exhibit a pervasive and surprising bias: when it comes to predicting what will happen to us tomorrow, next week, or fifty years from now, we overestimate the likelihood of positive events, and underestimate the likelihood of negative events. For example, we underrate our chances of getting divorced, being in a car accident, or suffering from cancer. We also expect to live longer than objective measures would warrant, overestimate our success in the job market, and believe that our children will be especially talented. This phenomenon is known as the optimism bias, and it is one of the most consistent, prevalent, and robust biases documented in psychology and behavioral economics."
Tali Sharot - The optimism bias - Current Biology, Volume 21, issues 23, R941-R945, 6th of December 2011.

We have on numerous occasions discussed shipping as being not only a leading credit indicator (with the collapse in European structured finance) but as well a leading economic growth indicator (on that subject please refer to "The link between consumer spending, housing, credit and shipping"), in our "Bear Case", excess capacity and a weak global economy with a China slowdown will drive rates down even with price increases, pressuring margins - graph source Bloomberg:
"The Drewry Hong Kong-Los Angeles 40-foot container rate benchmark rose 21.8% to $2,236 in the week ended July 3, as a $400 rate increase went into effect. Rates are 11.2% lower yoy, as slack capacity pressures pricing. With four increases in 2013, rates are up 1% ytd. Carriers are expected to implement a $400 peak season surcharge ahead of back-to-school and holiday shopping on containers from Asia to all U.S. destinations, effective Aug. 1." - source Bloomberg.

Not only slack capacity are pressurizing pricing in the shipping space but in general, we have long argued that overcapacity has been plaguing various economic segments such as the car industry with European car sales back at 1993 sales levels (on that subject see our 21st of April "European Clunker" conversation), but with the incoming threat of a China slowdown or even hard landing, metal prices such as Aluminum prices are indicative as well of the great deflationary forces at play and the overcapacity fuelled by "cheap credit" (the Baltic Dry reached 11,783 on May 20, 2008 and is now at 1103) - graph source Bloomberg:
"Aluminum prices, which have fallen for three straight quarters, may be poised for further declines as new production in China and the Middle East increases global output even as Alcoa Inc. trims capacity.
The CHART OF THE DAY shows production has gained 5.1 percent since the end of 2011, helping drive prices down 9.3 percent, according to data from the International Aluminium Institute. Output will reach a record near 50 million metric tons this year, up from 45 million in 2012, Harbor Intelligence forecasts.
The market is still looking at over-capacity, over-production and an unprecedented overhang of metal,” said Jorge Vazquez, a managing director at Austin, Texas-based Harbor. “There’s a lack of credibility for the producers, and even if these cuts take place, investors expect nothing to change.” Alcoa, Aluminum Corp. of China Ltd. and United Co. Rusal, the world’s largest producer, are among companies trimming capacity amid ample supplies. The price outlook remains “depressed” as some investors are concerned that producers won’t follow through on planned cuts and amid the prospect of new and expanded plants and restarts at some older sites, Vazquez said.
Aluminum for delivery in three months on the London Metal Exchange dropped 12 percent this year to settle at $1,832.50 a ton yesterday. The metal may fall to $1,675 in the next few weeks said Vazquez, who expects supply to exceed demand by 350,000 tons in 2013 for a seventh straight year of surplus." - source Bloomberg.

Of course our "omnipotent" central bankers "fail to recognize the extremity of their inadequacy" in true Dunning-Kruger effect as far as "cheap credit" and "bubbles" implications are concerned. They have even come up with a new marketing campaign as of late "forward guidance" in Europe.

Forward Guidance:
"Forward guidance arms central banks with fresh ammunition even when they have lowered short-term interest rates close to zero. It allows them to influence not just current rates but those stretching into the future through pledges to keep them low. The forward guidance can be for a period of time or it can be linked to specific indicators, such as an unemployment-rate threshold (which is not, however, a trigger) in the case of the Fed." - source The Economist

Fresh ammunition? Yet another demonstration of the Dunning-Kruger effect at play we think. Thank god, our "central bankers" are not in the guide dog training business to lead blind and visually impaired people around obstacles...
"Forward Guidance" might be as effective as using a "Yorkshire Terrier" as a guide dog, instead of the usual Labrador retriever. The "Yorkshire Terrier" could be trained to do the job, but would they really be effective? We wonder and ramble again.

Unemployment-rate threshold? As we have argued in "Goodhart's law":
"Conducing monetary policy based on an unemployment target is, no doubt, an application of the aforementioned Goodhart law. Therefore, when unemployment becomes a target for the Fed, we could argue that it ceases to be a good measure." - Macronomics, 2nd of June 2011

In this week's conversation, we would like to focus our attention to the "Bail-in" effect and the recent clarifications made surrounding financial subordinated debt instruments which have long been a pet subject of ours ("Subordinated debt-Love me tender?") given the "Bail-in" conversations which took place on the 26th of June (BRRD) which we touched last week, will have as well  "ripple" effects in the subordinated credit space but first our market overview.

The US dollar still rising against one of the most impacted asset commodity classes since the beginning of the year namely gold - graph source Bloomberg:
The greenback is still benefiting from the surge in bond volatility which has yet to recede.

The "Daisy Cutter" effect as displayed by the evolution of the Merrill Lynch MOVE  index, which is still showing sign of high volatility in the fixed income space as witness this week and CVIX indices closely followed by the recent rise in the VIX index - graph source Bloomberg:

No wonder interest rate sensitive asset classes such as Investment Grade Credit and High Yield are as well suffering from the increased turbulences as displayed by the price evolution of the most liquid and active ETFs in the credit space namely the LQD (Investment Grade) and HYG (High Yield) ETFs, graph source Bloomberg:
In the HY Fixed Income space HYG (iShares $ High Yield Corporate Bond, Expense Ratio 0.50%) has lost $1.78 billion year to date in terms of net redemption flow.

In terms of weekly allocation trends, bond market outflows have jumped in the week of the 27th of June until the 3rd of July as reported recently by Nomura's Fundflow insight published on the 5th of July:
"Asset allocation trends: Bond market outflows jumped/money market turned to inflows
- Bond market: -USD28.1bn vs -USD8.0bn in the previous week
- Money market: +USD7.7bn vs -USD25.1bn in the previous week
For the week ending 26 June, the bond market reported outflows for the 4th week in a row — the longest streak since August 2011. Despite bond market outflows easing slightly in the week before, the latest outflows jumped more than 3x to USD28.1bn from USD8.0bn in the previous week. In contrast, the money market turned to mild"- source Nomura, Fundflow insight, 5th of July 2013.

With the recent surge in both US Treasury yields courtesy of a better than expected Nonfarm payroll number coming at 195 K and in oil prices thanks to increased tensions in Middle-East, we wonder how long equities  in general and the S&P in particular will stay immune from the growing nervousness - graph source Bloomberg:

The latest "Forward guidance" European marketing stunt is no doubt meant to prevent a dramatic repricing in the European government space and avoid the trigger of the much "hyped" OMT. The volatility jitters in the bond space, have led to a surge in European Government Bonds yields in the process as indicated in the below graph with German 10 year yields rising towards the 1.70% level and French yields now around 2.30% - source Bloomberg: - graph source Bloomberg:

Of course our "omnipotent" central bankers in Europe had to come up with another playing trick up their sleeves, given as we indicated in last week's conversation, contagion risk is now bigger than ever and the customer/investor security system is now weaker than ever because the LTROs have encouraged banks to increase even more their holdings of government debt to fund fiscal deficit, making them in the process even more "Too-Big To Fail". Yet another demonstration of the Dunning-Kruger effect.

The issue of course is "convexity", given the debt levels for both the private sector and the public sector are high in most developed countries meaning their economies are now even more sensitive to interest rate risk courtesy of global ZIRP! The more sensitive their economies get, the more solvency risk you have, the greater the risk of a sudden spike of defaults you get in a low yield environment with surging yields.

Back in November 2011, we posited the following in our conversation "Complacency":
"In a low yield environment, defaults tend to spike. Deflation is still the name of the game and it should be your concern credit wise (in relation to upcoming defaults), not inflation."
"Low inflation environments, like the one we’ve had for the past 25 years, tend to be ones where defaults can spike." - Morgan Stanley - "Understanding Credit in a Low Yield World.


Moving on to the subject of the "Bail-in" factor and the financial subordinated credit space, given it has gathered much attention in the bank credit analysts sector as of late with very diverging views on the future for legacy subordinated Tier 1 securities, a subject which warrants some attention.

For instance Morgan Stanley on the 25th of June, in their European Banks note entitled "Get Ready: Regulatory Rating Par Calls Soon" argued the following:
"We have long been cautious on high cash price regulatory and rating par calls (see Reg Par Calls:
Closer, October 12, 2012, and The End of RAC Tier 2, April 2, 2013). The finalisation of CRD IV and S&P’s decision on RAC methodology mean possible calls are weeks away.
About 60 Tier 1 bonds have reg par call language (RPC) in our space, which means that if regulations change and bonds lose their Tier 1 regulatory status, they can be called at par. More than half of these RPC bonds are currently trading above par, leaving scope for significant downside from here.
Deutsche could be the first RPC next month… As we believe it will be able to trigger the reg par call of its €9.5% and €8% retail prefs as soon as CRD IV/CRR is published in the Official Journal of the EU, on June 27, which marks the end of the legislative decision-making process. At current levels, the yield to call in, say, a month is -26% for the €9.5% and -37% for the €8%. Bondholders risk losing up to 8 points in a day if the RPC is exercised.
… affecting all RPC bond pricing negatively, in our view. The language of nearly all other RPC bonds is far less clear than Deutsche’s, and such ambiguity could bring legal challenges many of these issuers (if not all) would not want to risk. However, particularly in today’s kind of market, we believe that many holders will simply take fright and it’s hard to say how much above par any bid might be, following a potential par call by Deutsche." - source Morgan Stanley

We were quite baffled by Morgan Stanley's note given we have been watching liability management exercise for a while in the European banking space and we completely disagree with their take on Tier 1 securities trading way above par that could be rapidly called by their issuers. For us, it doesn't make sense as we indicated in our conversation "The Doubt in the Shadow" on the 23rd of March 2013:
"Banks may have an incentive to buy back non-compliant Basel 2.5 hybrids that do not qualify as regulatory capital under Basel 3. To the extent that such "liability management exercises" can result in debt being repurchased at a discount to par (well below a cash price of 100), banks are able to generate common equity Tier 1 (CET1) gains. On that subject see our October 2011 conversation "Subordinated debt-love me tender?"."

Our views have been comforted by Bank of America Merrill Lynch note entitled "Bail-in: the ripples" from the 1st of July:
"Reg par calls – still a no-no?
The new need to have a bail-in buffer if anything supports the idea that the banks need to hold onto their existing subordinated debt and would be ill-advised to rush to redeem it, even if it optically appears to be expensive. The work we have presented on the need for bail-in buffers only underlines that European banks need to retain and rebuild capital, not redeem it, in our view. Why would a regulator permit the calling of an old Tier 1 bond just because it was ‘expensive’ to the bank? We think the emphasis on retaining capital until the banks are more comfortably positioned will remain for the foreseeable future so our base case remains: No reg par calls." - source Bank of America Merrill Lynch.

Morgan Stanley is putting the cart before the horse and we also agree with the below extract from Bank of America Merrill Lynch note:
"European banks’ need to retain and rebuild capital not redeem it. Why would a regulator permit the calling of an old Tier 1 bond just because it was ‘expensive’ to the bank? They would wish to see such a bond being replaced. If we were regulating the banks, we’d also like to see the banks issued the replacement capital prior to our allowing them to call the old." - source Bank of America Merrill Lynch

On top of that it seems to us Morgan Stanley's is not taking into account earnings boosting technique of FAS 159 which allows banks to book profits when the value of their bonds falls from par, meaning for us, that banks will be encouraged to issue more loss absorbing subordinated debt rather than reduce their buffer to avoid having senior unsecured bondholders or unsecured depositors paying the piper like it happened in the Dutch SNS case in the first instance due to lack of deliverables for the CDS trigger, and in the second case like it happened in Cyprus due to lack of sufficient subordinated and senior unsecured bonds buffers.

So what is the "ripple" effect of the latest "Bail-in" discussions for senior debt and legacy subordinated debt?
"But this brings us to the ripple effect: if banks need to focus on building this buffer, we believe it will prima facie entail potentially some new sub or bail-in bond issuance.
What about retaining the existing subordinated stock though where it is evidently cheaper than issuing new stuff? We are thinking particularly of low-back end Tier 1 bonds but there are also fixed-to-float UT2s and arguably even the dated LT2s too. It makes sense to keep these outstanding forever, arguably, or at least until such a time that the spreads on e.g. bail-in bonds are comparable to those low back-ends (which may, equally, be never of course). 
This is a totally separate discussion to whether the bonds in question are included in regulatory capital. Eventually, very little of the old stock of Tier 1s, Upper Tier 2s and Lower Tier 2s should there be any long-dated enough will ‘count’ as regulatory capital. But there is a role now for all these instruments in the liability buffer above own funds that perhaps they didn’t have before last Thursday. What sense for Credit Agricole to call the €4.13% Tier 1 bond with a back-end of +165bp? Or for BNP to call the US$5.186% bond with a back-end of +168bp? How to justify retiring these bonds which are subordinated and – indeed – loss absorbing – to expose senior bondholders to potential losses? A call of these bonds would look like negligence to us, if it increased the risk that senior bondholders would be bailed-in.- source Bank of America Merrill Lynch

We agree with Bank of America Merrill Lynch's take on the subject. From a regulatory perspective and in relation to increasing capital buffers, previous bond tenders have shown that there is a greater call risk with low back-end bonds (convexity issue) and those trading below par:
"In most cases, that CRD 4 is now European Law. It must be domestic law too. It will take several countries some time to put CRD 4 into their own law. So in most cases, there is no immediate threat of a reg par call because there isn’t even the legal basis yet for one." - source Bank of America Merrill Lynch

What is the risk for senior unsecured bondholders and the implication of having low subordinated bond levels buffers for some European banks?
Here is Bank of America Merrill Lynch take on the subject:
"If banks don’t build up their buffers and appear to have no intention to either, then their senior will widen towards their sub and the sub-senior curve (in cash) will flatten at wider levels, mutatis mutandis, we think. Current CDS contracts may not be a reliable indicator of this of course, since they are locked in their own technical until the new contracts come into being in September" - source Bank of America Merrill Lynch

We could not agree more. Of course current CDS contracts are not yet reliable indicators of this risk until the much anticipated need to revamp CDS contracts in September. For more on the importance of this issue please refer to our conversation "The Week That Changed The CDS World" from the 26th of May.

So all in all, Morgan Stanley as put the horse before the cart, and in the case of financial subordinated bonds has even jumped the gun as indicated by a JP Morgan's note from the 5th of July entitled "Tier 1: Upgrade to Overweight":
"The clarification that legacy Tier I instruments will not be eligible as Tier II capital under transitional arrangements will be an undoubted positive for valuations as this will increase the certainty of call being exercised. Whereas previously our assumption was that the Tier I instruments would be eligible as Tier II capital, making the decision to call such instruments dependent on the relative cost of issuing Tier II capital instruments, the fact that the legacy Tier I instruments will not have any regulatory capital value will imply that it will merely be a question of comparing the post-call spread on the instrument versus the cost of senior funding. Under these circumstances, we assume that the issuers will have much lower incentives to maintain these instruments outstanding and as such, the certainty with regard to call has to increase. This would also be applicable for issuers such as DB which in the past have used economic rationale to justify the calling or not of Tier I instruments." - source JP Morgan

On a final note, we have always lacked conviction in the great rotation story from bonds to equities which has been put forward since the beginning of the year. As displayed in the below graph from Bloomberg, the rotation to stocks from bonds has been indeed less than great, so has been QE to the "real economy" courtesy of the Dunning-Kruger effect:
"Anyone who expects U.S. individual investors to push stocks higher by moving away from bonds may end up disappointed, according to Vadim Zlotnikov, Sanford C. Bernstein & Co.’s chief market strategist.
The CHART OF THE DAY illustrates how Zlotnikov drew his conclusion, presented in a report yesterday. He tracked the value of equities, owned directly or through funds, as a percentage of household financial assets. Stocks were 39 percent of assets at the end of March, according to data that the Federal Reserve compiles quarterly. The figure was the highest since 2007 and surpassed an average of 29.2 percent since 1950, as shown in the chart. “U.S. households’ exposure to equities is already above historical levels,” the New York-based strategist wrote. “With rates likely to rise over the next 12 months, the case for a rotation from bonds into equities may become less compelling.Average inflation-adjusted returns on U.S. stocks are negative 2.3 percent a year when bond yields increase at an annual rate of more than 1.3 percentage points, he wrote. The calculation is based on performance from 1871 through April of this year.
Betting against stocks with relatively high dividend yields may pay off as rates increase, Zlotnikov wrote. These shares are 20 percent more expensive than their industry peers on average when judged by ratios of price to book value, or the value of assets after subtracting liabilities, the report said." - source Bloomberg

"Real knowledge is to know the extent of one's ignorance." - Confucius

Stay tuned!

No comments:

Post a Comment

 
View My Stats