Credit - The Week That Changed The CDS World
"One must change one's tactics every ten years if one wishes to maintain one's superiority." - Napoleon Bonaparte
Looking at the epic compression in recent weeks of the Itraxx CDS
financial subordinated index versus the Itraxx Senior Financial CDS 5
year index, tied up to the recent ISDA proposals to include Bail-In
Credit Event, we decided our reference this week ought to be a shorthand
for describing surprising and uncharacteristic actions in similar
fashion to Kissinger's 1971 secret trip to China. This secret trip laid
the groundwork for the historic visit of Nixon to China that followed in
1972.
Given the upcoming clean up of ISDA's 2003 Credit Derivatives Credit
Event definitions which were in dire need of a brush up following the
recent Dutch banks SNS subordinated debt saga, as per our Napoleon
Bonaparte quote goes, arguably, one indeed must change tactics every ten
years if ones wishes to maintains one's superiority. It could not be
more truer than for the viability of the CDS market. What happened this
week in the CDS world, with the proposed introduction of a new credit
event for financial CDS in the case of a bail-in triggered by a
government agency and the change in deliverability rules, made this week
an important one from a credit perspective.
We already discussed the implications of the SNS case in our previous conversation "House of pain and House of cards":
"The SNS case this week has had some major significant risks to the "House of pain" in the European banking sector that warrants additional close attention for the remaining subordinated bondholders.
If the recovery rate for SNS LT2 subordinated bonds is zero, the significance for the European subordinated CDS market is not neutral given the assumed recovery rate factored in to calculate the value of the CDS spread is assumed to be 20% for single name subordinated CDS and 40% for senior financial CDS. On top of that, a nationalization, such as SNS case, is not by itself a credit event trigger. Appointing an insolvency official is.
As far as delivery of LT2 underlying subordinated bonds referenced in any CDS contract referencing SNS, you would have to ask the Dutch state for delivery (if the subordinated bonds are not simply cancelled or converted into equity...).
So what's the value of your subordinated single name CDS on SNS? Could it mean single name subordinated CDS are a "House of cards"? We wonder. Oh well..."
"The SNS case this week has had some major significant risks to the "House of pain" in the European banking sector that warrants additional close attention for the remaining subordinated bondholders.
If the recovery rate for SNS LT2 subordinated bonds is zero, the significance for the European subordinated CDS market is not neutral given the assumed recovery rate factored in to calculate the value of the CDS spread is assumed to be 20% for single name subordinated CDS and 40% for senior financial CDS. On top of that, a nationalization, such as SNS case, is not by itself a credit event trigger. Appointing an insolvency official is.
As far as delivery of LT2 underlying subordinated bonds referenced in any CDS contract referencing SNS, you would have to ask the Dutch state for delivery (if the subordinated bonds are not simply cancelled or converted into equity...).
So what's the value of your subordinated single name CDS on SNS? Could it mean single name subordinated CDS are a "House of cards"? We wonder. Oh well..."
So in this week's conversation we will look at the wider implications
for the financial CDS market on the proposed ISDA Credit Events revamp
on the 10 year anniversary of the ISDA 2003 Credit event definitions
because the validity of the CDS market as a hedge had been put in
jeopardy quite significantly by the SNS case. We will also look at debt
disturbances and price-level disturbances, revisiting the wisdom of
Irving Fisher in the process.
The CDS compression story in one chart - Itraxx Financial Senior index
versus Itraxx Financial Subordinated 5 year index - source Bloomberg:
From the above chart one can see the severity and rapidity of the move in the subordinated financial CDS space.
So is the move justified?
Here is BNP Paribas take on the move from their 23rd of May entitled "ISDA Proposes Bail-in Credit Event:
"Is the Sub CDS move since last Friday justified?
Sub CDS has collapsed by more than 40bp
since 16 May and the Sub/Sen ratio is now just below 1.4x, after having
been at a mean of around 1.7x for a long time. The timing
surprised us, as the new definitions are not finalised nor implemented
yet. In addition, the market could have already reacted more than it did
after the SNS news. Therefore, while we were proponents of the general
Sub/Sen compression theme, we were surprised both in terms of timing and
severity of the market move.
How do we explain the move then? The
rapid compression of Sub vs. Sen over just a few trading sessions was
probably due to the realisation that existing financial CDS contracts
will over time be replaced by the new ones, making the old contracts
less valuable from a long protection perspective. Thin market conditions due to European holidays may have exacerbated the move.
Other possible explanations of the significant move are the general
bull market and search for yield, the gradual acceptance and pricing in
of senior bail-in and the possibility of depositor preference over
senior. Finally, investors may expect that, with the arrival of a new
CDS contract, authorities would be less careful about legacy CDS (i.e.
about making sure that there are deliverables, as the Irish authorities
had ensured). That said, the change of definitions had been mentioned
for a while and the implications clear, i.e. the new contracts should trade at a wider level.
The old contracts can still be useful, especially as we believe that
bail-in will trigger a restructuring event (as was the case with SNS),
but the existence of sub deliverables is uncertain and therefore they
are less valuable to a protection buyer than the new one. This
information was previously available but the market reacted last Friday.
Can the magnitude of the move be justified by relative recovery
expectations between senior and sub contracts? Chart 1 shows the implied
Sub/Sen ratio (for the existing contracts) as a function of the senior
expected recovery rate for different sub recovery rate assumptions. In
most cases the sub recovery rate has been in the 0-20% range. At current market pricing, this corresponds to a senior recovery rate of 30-40%.
This does not strike us as too low, especially when we consider that
(i) the Moody’s average historical senior corporate recovery rate is
around 38%; (ii) further developments towards resolution regimes and
senior bail-in should increase the senior credit event probability
relative to the sub probability; (iii) depositor preference, if
forthcoming, would reduce the senior expected recovery rate; and (iv)
the recent SNS event highlights a growing likelihood of events with a
significantly higher sub recovery rate (for the existing contracts) than
0-20%." - source BNP Paribas.
We disagree with BNP Paribas on the implied recovery rate of 30-40%. It is not too low, it is not low enough at least on the "old contracts" because it relies on Moody's average historical senior recovery so this analysis is backward looking. The senior expected recovery rate due to the evolution of resolution regimes and senior bail-in implies lower recovery rates and wider spread levels in the new contract.
We disagree with BNP Paribas on the implied recovery rate of 30-40%. It is not too low, it is not low enough at least on the "old contracts" because it relies on Moody's average historical senior recovery so this analysis is backward looking. The senior expected recovery rate due to the evolution of resolution regimes and senior bail-in implies lower recovery rates and wider spread levels in the new contract.
Why the new contracts should trade at a wider level you might rightly ask?
We have touched on that subject previously in February 2013 in our conversation "Promissory Hope":
From EDHEC-Risk Institute in their January 2012 note entitled "The Link between Eurozone Sovereign Debt and CDS Prices" provides us with some insight on the aforementioned impact:
"To examine the difference between these spread measures, we priced a 5-year bond with a 5% coupon in an environment where the default-free yield curve is assumed flat at 3% and the Libor risk-free curve is also assumed to be flat at 3.5%. We considered two cases - first an expected recovery rate of 40% and second an expected recovery of 0%. We then varied the 5-year survival probability assuming a flat term structure of default rates11 and calculated the implied bond price and spread measures. In all cases we assumed k = 1.
Figure 2 Comparison of the model-implied CDS, bond yield-spread and par asset swap spread measures as a function of the full price of a 5-year bond with a 6% coupon. We show this for an expected recovery of 40% (above) and 0% (below).":
"To examine the difference between these spread measures, we priced a 5-year bond with a 5% coupon in an environment where the default-free yield curve is assumed flat at 3% and the Libor risk-free curve is also assumed to be flat at 3.5%. We considered two cases - first an expected recovery rate of 40% and second an expected recovery of 0%. We then varied the 5-year survival probability assuming a flat term structure of default rates11 and calculated the implied bond price and spread measures. In all cases we assumed k = 1.
Figure 2 Comparison of the model-implied CDS, bond yield-spread and par asset swap spread measures as a function of the full price of a 5-year bond with a 6% coupon. We show this for an expected recovery of 40% (above) and 0% (below).":
"The results are presented in Figure 2. When the expected recovery rate is 40% we find that as the
bond price falls (and it cannot fall below 40), the CDS spread grows and asymptotically tends to infinity while the yield-spread and asset swap spread tend to different large but finite numbers. However, if we set the expected recovery rate to zero then the yield-spread also tends to infinity and is very close in value to the CDS spread as the bond price falls to zero." - source EDHEC-Risk
As we repeatedly pointed out, the importance of liquidity is paramount,
particularly in the credit space, given the low level of inventories on
dealers' book that can accommodate large selling movements. The lack of
liquidity in the financial CDS space without a revamp of the 2003 ISDA
Credit Events definitions would exacerbate potentially the movements in
financial bonds.
The liquidity in credit is already impacted by the poor liquidity in the
secondary space, in this low yield, and yield hunting environment as
described in a recent presentation made by Wells Fargo Credit Strategy
team:
"Anecdotal evidence from our trading desk suggests the performance of
secondary bonds is lagging that of new-issues. In addition, trading
flows point toward more investors buying new-issues “on switch” rather
than outright from cash. This is particularly true at the long end of
the curve and for frequent borrowers." - source Wells Fargo
Since January the price action has been more volatile in the Investment
Grade than in the US High Yield ETF space, which has mirrored much more
the price action of equities, namely the S&P 500. Investment Grade
is therefore a more volatility sensitive asset, whereas High Yield is a
more default sensitive asset, as indicated in by the price movement of
the the iShares iBoxx $ Investment Grade Corporate Bond Fund ETF (AMEX:
LQD) versus the US High Yield ETF HYG and the S&P 500- source
Bloomberg:
Looking at the latest minutes from the FOMC and the discussions
surrounding the Fed's QE stance and the possibility of a reduction in
bond purchases in coming quarters provided an improvement in the
employment data, it does make Investment Grade corporate bonds highly
more volatile to interest movements in this "Japonification" of the
credit markets courtesy of global ZIRP.
As per a recent Wells Fargo credit presentation, we agree with them in relation to the key risks for Q2 2013 given that:
"Lower coupons and longer maturities increase a portfolio’s duration/interest-rate sensitivity. The duration of the entire HG corporate bond market has extended 1.0 year to 7.0 years over the past five years. In maturities of greater than 10 years, duration has extended 2.0 years to 13.6. With
the Fed signaling a potential shift in policy this year, long-duration
corporate bond prices could be at risk of falling sharply and quickly." - source Wells Fargo Credit Strategy.
We are not surprised that the price of "stability" courtesy of massive
liquidity injections from global central banks has come at a cost of
increased "instability" as per Hyman Minsky's definition:
"A Minsky moment is the point in a credit cycle or business cycle
when investors have cash flow problems due to spiraling debt they have
incurred in order to finance speculative investments.
At this point, a major selloff begins due to the fact that no
counterparty can be found to bid at the high asking prices previously
quoted, leading to a sudden and precipitous collapse in market clearing
asset prices and a sharp drop in market liquidity." - source Wikipedia.
You might want to read again, Irving Fisher's book "The Debt-Deflation
Theory of the Great Depression" published in 1933. Because in this book
Irving Fisher dealt extensively with "business cycle theory".
For Irving Fisher, the two big bad actors in great booms and depressions
are debt disturbances and price-level disturbances. We have both...but
that's us ranting:
"Debt Starters
Easy money is the great cause of over-borrowing. When
an investor thinks he can make over 100 per cent per annum by borrowing
at 6 per cent, he will be tempted to borrow, and to invest or speculate
with borrowed money. This was the prime cause leading to the over indebtedness of 1929." - Irving Fisher
Just a fact:
Investors borrowed $384.4 billion in April, a 1.3% gain from the
previous month which was at $379.5 billion and conveniently the second
highest in the history of the NYSE going back to 1959. The April surge
was a 29% rise from the same month last year. The highest level was
$381.4 billion recorded in July 2007. We have an all-time record for
margin debt and it exceeds the previous high mark.
And what else did Irving Fisher wrote in his 1933 book?
"When the starter consists of new
opportunities to make unusually profitable investments, the bubble of
debt tends to be blown bigger and faster than when the starter is great
misfortune causing merely non-productive debts."
So no "speculation" going on, it is all going well...
As far as credit is concerned, the rapidity in the tightening movements
in credit spreads is reminiscent of the warnings given in 1933 by the
wise Irving Fisher.
A recent note from Societe Generale on the 22nd of May is clearly
indicative of the rapidity of how this time around the credit bubble is
being blown by our "omnipotent" central bankers:
"When the music stops, we pause and then just add another chair. It's
usually the slightest of breathers and the market isn't waiting too
long before that relentless grind and lifting of paper resumes. Clips,
blocks of paper and new issues are managing to get taken down without
much fuss, and we're not seeing any contagion from the volatile stocks
impacting the cash market. We're well poised here. Of course there's
plenty of apprehension and it is understandable that we all need a
little convincing that still adding risk at these levels is the right
thing to do. It is. In the meantime, the low/high beta compression
continues and was clearly evident from yesterday's deal from Plastic
Omnium. The unrated - but implied non-IG - French borrower managed to
raise €500m for seven years, paying just 3% for the privilege. Add in a
premium for being unrated and one has to concede that the level is a
funding coup for the borrower! For the broader market, the iBoxx cash
index closed below B+133bp yesterday and will be lower again today after
the tightening seen in today's session for corporate spreads. Even
taking into account the massive February/March wobble (when the index
widened 22bp), credit is tightening faster than even we - the most
bullish of observers - would have expected. Hitting our original 2013
target of B+120bp is now a case of when not if, and we can only expect
investor nervousness to rise even more as a result. As long
as money continues to come into the asset class and supply remains at
these (low) levels given the size of the demand, then the current
tightening/compression dynamics will stay in place. Position for it." - source Societe Generale.
Moving back to the subject of the evolution ISDA Credit Events and the impact of expected changes recovery rates on financials, the impact of the new CDS contracts would make the CDS market in the financial space more relevant as per a note from Societe Generale from the 24th of May on the subject:
"Event:
CDS protection may be more valuable should reported proposals for amending credit derivative definitions be accepted. The proposal is to amend credit derivative definitions for banks and comprises three main points. First, it adds a credit event to capture government enforced bail-in. Second, it expands the list of deliverables in the new credit event and keeps current deliverables available for old events, despite their potential loss-absorption ability in the future. And third, it improves successor provisions to keep CDS protection attached to the debt. Taken together, these may help to avoid a repeat of the CDS insurance failure of SNS while capturing the increased tool-kit available for governments to restructure banks out of bankruptcy. We do not expect these provisions to be retrospectively applied to existing contracts; however the amendments suggest much less value in outstanding sub contracts.
Assessment:
ISDA’s proposed changes would make CDS protection more robust and, therefore, valuable. First, a new ‘hard’ credit event would capture government-enforced bail-in. This would be broadly defined as an action taken by government authorities that alters creditor rights under bank restructuring and resolution laws. By our understanding, this does not include the institution triggering Tier 2 contingent capital (CoCo) clauses, as this is an action undertaken by the entity itself, but it would capture Bankia-type events. If the new credit event trigger is a write-down, the event would not occur until the write-down is permanent or there is nonpayment under prior contract terms. A ‘hard’ event eliminates the maturity buckets of restructuring events. Second, deliverables in the new credit event auction may include the written-down or conversion/exchange proceeds. Again, this is Bankia-event type protection. In the event of complete write-off, à la SNS, par payment would be received by the protection buyer. In addition, deliverables under a current or new credit event could include Tier 2 or more senior securities with write-down provisions as mandated by legislation, provided they are not yet written down. This would enable most Lower Tier 2 debt to be deliverable even if it is loss absorbing Basel III-compliant via legislation. CoCos could be delivered in the new credit event provided they have not yet triggered. This captures many bail-in eventualities. Third, successor provisions would track the debt, enabling subordinated and senior CDS to succeed to different entities. This would keep CDS viable in a good bank/bad bank situation. Also, importantly, the new credit event could occur on subordinated CDS without triggering senior CDS. This may have implications for sub/snr trading levels once the new amendments are in place." - source Societe Generale
On a final note, the US equities market is increasingly being boosted by buybacks, yet another artificial jab in the on-going liquidity induced rally as indicated by Bloomberg's chart, great for CEOs and stock options and their shareholders but probably less so for the health of the balance sheet:
As the CHART OF THE DAY shows, the Nasdaq Buyback Achievers Index has more than tripled in the current bull market and has left the Standard & Poor’s 500 Index behind. The Nasdaq gauge consists of companies that repurchased at least 5 percent of their shares in the previous 12 months.
“Corporations have been aggressively buying back shares,” Jeffrey Kleintop, chief market strategist at LPL Financial Holdings Inc., wrote two days ago in a report. He added that the repurchases are largely designed “to boost earnings per share as revenue growth slows.” - source Bloomberg
"The public psychology of going into debt for gain passes through
several more or less distinc phases: (a) the lure of big prospective
dividends or gains in income in the remote future; (b) the hope of
reckless promotions, taking advantage of the habituation of the public
to great expectations; (d) the development of downright fraud, imposing
on a public which had grown credulous and gullible". - Irving Fisher.
At some point the Fed will have to normalize, probably not now, but the
more they delay the adjustment, the more painful it is going to end up.
Stay tuned!