Sun, 28 Oct 2007

Northern Rock and Unified Regulators

In my previous post today, I described what I had learnt about managing liquidity risk from studying Northern Rock; in this post, I shall discuss the implications for the regulatory architecture (separation of bank supervision from the monetary authority).

Many observers appear to think that Northern Rock has revealed the dangers of the UK system of separating bank supervision from the central bank. In my view, it has on the contrary, revealed the strengths of this system. It is clear from the evidence, that the banking supervisor (Financial Services Authority) was keen to solve the problem of Northern Rock by making changes in monetary policy while the monetary authority (Bank of England) was unwilling to do this. If the central bank were in charge of both monetary policy and bank supervision, there is little doubt that monetary policy would have been changed to save Northern Rock. The deficiences if any in banking supervision would then have never come to light at all. This is a form of regulatory moral hazard – regulators will tweak the regulatory system to hide their failures. The UK system where the central bank is not the bank supervisor is less vulnerable to this kind of moral hazard.

That the UK had not had a bank run for 140 years prior to Northern Rock is to me a troubling thing. Banks are highly fragile institutions. If in 140 years including two world wars, a great depression, the loss of a vast empire and over three decades of chronic exchange rate difficulties, the UK did not have any bank runs, then it tells us not that banks were well regulated but that they were saved covertly. In these covert operations, doubtless the reputations of bank regulators (and perhaps bank managers) were also protected at the tax payers expense. That Northern Rock has dented many reputations is not such a bad thing by comparison.

Posted at 19:20 on Sun, 28 Oct 2007     View/Post Comments (0)     permanent link

Liquidity Risk and Northern Rock

I have spent a fair amount of time trying to understand the collapse of Northern Rock in the United Kingdom by poring through the transcripts of the Treasury Committee hearings that took evidence from the Bank of England, the Financial Services Authority and the directors of Northern Rock as well as the financial statements of Northern Rock itself.

In a separate post, I shall discuss the implications for the regulatory architecture (separation of bank supervision from the monetary authority. In this posting, I shall focus on what I have learnt from all this about managing liquidity risk:

• While much of the discussion about Northern Rock assumes that their problem was one of borrowing short and lending long, this does not appear to be correct.
  1. The FSA categorically stated in response to Question 315 that “the figures do not suggest that [Northern Rock] was an outlier in respect of its dependency on very short-term funding ... The actual percentage of its funding which was dependent on three months or under was not a particular outlier, and also just to remind us again, I think it is important to remember that it did not actually fail to fund itself is this period.”
  2. The CEO of Northern Rock stated in response to Questions 517 and 518 that “The average life of a mortgage product is three years and one month and the average life of our funding was three and a half years ... We had 10% of our borrowings which had a maturity of less than one year; we had 10% of our borrowings that were over one year; we had 50% of our borrowings that were three and a half years; and we had 10% of our borrowings with an average life of seven years.”
  3. There does not appear to have been a significant difference in the costs of various maturities of borrowing. The CEO of Northern Rock stated in response to Questions 519 and 520 that “The average rate on securitisation for the stock was about LIBOR plus ten basis points; the average price for covered bonds, which is the seven year, was about LIBOR plus one basis point; the average price of longer term wholesale was about LIBOR plus five; the average rate for shorter, ie less than year, was about LIBOR flat. ... The interest rates on our lending, including fees that are effective interest rate were about LIBOR plus 90 basis points.”
• It appears from the available information that there was no serious problem of asset quality. Northern Rock’s Chairman stated in response to Question 402 that Northern Rock was “below half the industry average of arrears on our mortgage book.” While some of this may be due to the rapid growth of the loan book, it does not appear likely that the credit quality of the loan book was below average, let alone bad. Their exposure to US subprime and CDOs was negligible.
• It appears from all the information that I have read that the standard tools of gap analysis and asset liability mismatches would not have revealed any problems in the liquidity management of Northern Rock. Yet, when I re-read the BIS document on Sound Practices for Managing Liquidity in Banking Organisations, I do not think there is any need to rethink the basic principles of liquidity management. The problem as I see it is that in organizations like Northern Rock, the dynamic aspects of liquidity management (as measured by cash flow projections) are more critical than the static aspect (as measured by mismatches and gap analysis). The problem is that securitization tends to be done in a few large transactions at intervals of several months. Profitability considerations will then dictate that the maturity of the short term funding would be shrunk ahead of an impending securitization. If this is not done, then the securitization would produce a huge cash surplus that cannot be profitabily deployed for several weeks. If a closure of a securitization markets takes place at this point, the liquidity position of the bank can become quite acute. This becomes worse if the non securitization lenders also shrink the maturity at which they are willing to lend in the build up to the crisis. All through the crisis, Northern Rock remained funded but by the end, the only funding that they could access was overnight funding. Dynamic liquidity analysis using what-if scenarios are unavoidable for wholesale oriented banks like Northern Rock. However, it is really hard for a regulator to review and comment on the adequacy of this analysis. Gap analysis is so much simpler for a regulator to understand and comment upon.
• There were more than ten questions to the directors of Northern Rock asking them why they were the only bank that failed (Questions 460, 474, 475, 476, 529, 640, 641, 642, 643, 644 and 649). This was a grim reminder of what Keynes wrote 75 years ago: “A ‘sound banker’, alas, is not one who forsees danger and avoids it, but one who, when he is ruined, is ruined in a conventional and orthodox way along with his fellows, so that no one can really blame him.” (Consequences to the Banks of a Collapse in Money Values, 1931).

Posted at 19:17 on Sun, 28 Oct 2007     View/Post Comments (1)     permanent link

Wed, 24 Oct 2007

Can SEBI be more transparent?

The Securities and Exchange Board of India provided critical clarifications on its policies regarding offshore derivative instruments (participatory notes) issued by Foreign Institutional Investors (FIIs) through a video conference on Monday, but their website still has no video recordings, no transcripts, no press releases on what was revealed during this conference. (I have looked in the “What is New”, “Press Releases”, “From the Chairman” and “News Clarifications” sections of the web site and also searched the site for “video conference”. I hope it is not hidden somewhere else in the site.)

SEBI’s policy announcements on this subject have caused intra-day movements in the market index of several percent in recent days and the potential for insider trading is immense whenever SEBI issues even the smallest clarification on the subject. That a video conference on such an important subject was done without it being webcast live is regrettable; that it was not even followed up with recordings and transcripts is unacceptable.

Posted at 14:57 on Wed, 24 Oct 2007     View/Post Comments (3)     permanent link

Fri, 19 Oct 2007

SEBI Proposal on Participatory Notes

I wrote an article in the Business Standard today on the Discussion Paper put out by the Securities and Exchange Board of India (SEBI) proposing to limit the issuance of Offshore Derivative Instruments (participatory notes) by Foreign Institutional Investors (FIIs) in India.

This discussion paper produced a wide range of commentary in the financial press. An excellent summary of this discussion has been put together by Ajay Shah in his blog. The key points that emerge from this analysis are:

• India allows only registered foreign institutional investors (FIIs) to invest in the Indian market.
• Many foreign investors do not register as FIIs because they are not eligible to register or because they do not wish to pay the registration fees or because they do not want to go through the paper work involved.
• These investors who have not registered as FIIs seek to invest in India through derivative instruments (participatory notes) that reference Indian securities. These participatory notes are issued outside India by registered FIIs who hedge them in the Indian market to which they have access.
• India should move away from the FII framework to a regime of direct access to various classes of foreign investors
• India should also simultaneously develop an onshore over-the-counter (OTC) equity derivative market.
• If we do all this much of the participatory note market would die a natural death.

My Business Standard article did not dwell much on any of these but focused on some of the details of the SEBI proposal.

SEBI’s first proposal is to ban participatory notes that have a derivative as the underlying. This is a very confusing statement. The intention appears to be to ensure that a participatory note is backed by a cash market position and not a derivative position. If this is what SEBI indeed wishes to do, it should explicitly ban the use of derivatives to hedge participatory notes.

SEBI should recognize that the term “underlying” is a technical term with a well defined meaning in the world of finance. The underlying of a participatory note is the instrument from which the participatory note derives its value; it is the instrument which is delivered on settlement of the participatory note or with reference to whose price the participatory note is cash settled. The “underlying” in this technical sense has nothing to do with the portfolio that the FII uses to hedge the participatory note. A participatory note that is cash settled using the Nifty index futures price has the future as the underlying even if the FII hedges it using cash equities. Similarly, if the participatory note is cash settled using the cash price of the Nifty index, its underlying is the cash index and not the index future even if the FII hedges the note using index futures.

A financial regulator should respect the semantic integrity of well defined technical terms and not abuse the term “underlying” to mean what it does not and cannot mean. In this context, the use of the word “against” before the word “underlying” in regulation 15A of the FII regulation is also unfortunate as that word is perhaps the source of this confusion.

Enough of semantics. I now turn to the substance of the proposal. If SEBI bans the use of derivatives to hedge participatory notes, it would have three implications.

1. Since cash equities are less liquid than the futures, the hedging costs would increase. The increase would be even greater if the underlying is an index where hedging using the constituent shares is far more difficult than using the index future. If the participatory note contains some option-like features (non linear payoffs), the hedging risks could also increase as the volatility risk of options cannot be hedged using only the cash market. The FII would therefore have to charge a wider spread to its clients. OTC derivatives tend to be carry large spreads anyway (annualized costs of 4% to 8% of the notional principal are not uncommon). A mere increase in transaction costs would not probably kill the participatory note market.
2. SEBI’s proposal would prevent participatory notes that involve a short position in Indian equities (for example by a long-short hedge fund) since short selling is not feasible in the cash market today. Since short selling is essential for a well functioning market, this is clearly an undesirable consequence of the SEBI proposal.
3. SEBI’s proposal would also prevent issuance of participatory notes that are essentially synthetic rupee money market instruments because these synthetics can be created only by offsetting positions in cash and futures markets. It is doubtful whether any significant amount of participatory notes are of this kind.

The second major proposal of SEBI is to ban participatory notes issued by sub accounts of FIIs. In my view, this is largely an administrative measure which would not have a significant long run impact. An FII which is active enough to issue participatory notes should be willing to register as a full fledged FII.

SEBI’s third proposal is to limit participatory notes issuance by any FII to 40% of the assets under custody of that FII. Today, issuance of participatory notes is concentrated in the hands of a few FIIs. This is a very natural phenomenon. Running a derivative hedge book is a very complex activity and those with superior skills in doing this will get more business because of their lower costs and their ability to offer a wider range of products. There are also significant economies of scale in running a derivatives book because if an FII sells a long position to one offshore client and a short position to another offshore client, it needs to hedge only the net position in the Indian market. When the efficient hedgers have exhausted their 40% limits, buyers of participatory notes would have to buy from less efficient hedgers who have not reached the 40% limit. This would increase the costs and would amount to more “sand in the wheels” whose long term impact would be modest.

I also believe that the 40% limit can be circumvented by an FII buying cash equities and selling stock futures or index futures. This synthetic rupee money market position would not increase the FII’s exposure to the Indian equity market but it would increase assets under custody and allow the FII to issue more participatory notes. In the context of a strong rupee and a positive interest rate differential, this synthetic money market position may also be a profitable low risk investment for the FII. It would indeed be a delicious irony if a proposal designed partly to reduce capital inflows leads to more capital inflows.

Posted at 18:19 on Fri, 19 Oct 2007     View/Post Comments (1)     permanent link

Tue, 09 Oct 2007

Similarities and Differences between Banks and CDOs

In February 2006, I posted a blog entry about how banks and CDOs are very similar in their economic function. I received a couple of comments on that entry and then in August 2007, Francisco Casanova from Madrid began a long email conversation with me on the subject. All this forced to me think carefully about the issues and helped clarify my thoughts on the similarities and differences between banks and CDOs. So I decided to pull the comments and my responses into one long blog entry.

Comment	Response
A CDO is a leveraged play on the underlying, banks forget that the ‘deltas’ of the underlying portfolio can be quite large and mean substantial MtM volatility. I think a good test of the market is around the corner when the credit cycle turns and we see a number of downgrades on existing tranches making them economically unviable. (Mrinal Sharma)	A bank is also a leveraged play on the underlying loan book. But banks do not M2M their loan book and when the credit cycle turns, the impact is gradual. When there is no M2M, downgrades do not matter only defaults do.
‘Correlation risk’ and how it affects pricing is also a term misunderstood by banks and frequently ignored. (Mrinal Sharma)	Most of the risk of a bank’s loan book is also correlation risk though it is more commonly called concentration risk.
Tranching creates a slicing of risk whereby the 0-3% (in 5 years CDOs) is labeled as an equity investor. By receiving this name it would equate to the equity investor in any business, i.e., a bank. Nevertheless the actual risk/return characteristics of a non tranched equity stakeholder (i.e., a bank equity holder) are very different to those of a 0-3% holder in a CDO, aren’t they? (Francisco Casanova)	In a bank also, there are actually many tranches – demand deposits, time deposits, subordinated debt, hybrid capital and equity. If there were no central bank to bail out the bank, this would behave much like a CDO. When things start getting bad, the demand deposits will pull out quickly and will be paid out in full – this is like the AAA tranche. Time deposits might involve some loss if pulled out but the loss might be very small – an A or AA tranche. Some of the subordinated debt and hybrid capital would be like the BB tranche. Equity would be like the equity tranche. The precise attachment points of the tranches in the CDO reflect three things – the quality of the underlying pool, the lack of central bank bail out and the rating errors of the rating agencies. The big difference between banks and CDOs is the lender of last resort (the central bank)
Also, this permanent presence of the idiosyncratic vs systemic signals debate arising from the correlation movements in trading (which I do not know how accurately could be extrapolated to a bank), or the fact that, ceteris paribus (collateral spreads unchanged), a change in correlation reallocate expected losses among tranches in a zero sum game (i.e., an increase in default correlation expectations shifts risk allocation from junior to senior CDO tranche holders)... Or the mere fact that you can trade pure correlation views if you take delta hedged positions... (I guess you could do the same if all stakeholdings in a bank were securities form...). (Francisco Casanova)	It is interesting to note that the Basle II formula for corporate exposures is based on the same one-factor Gaussian copula models that are often used to price CDS index tranches. It is also useful to look at the following paragraphs from the second consultative package (Jan 2001) on Basle II that introduced the formula: 424. Credit risk in a portfolio arises from two sources, systematic and idiosyncratic. Systematic risk represents the effect of unexpected changes in macroeconomic and financial market conditions on the performance of borrowers. Borrowers may differ in their degree of sensitivity to systematic risk, but few firms are completely indifferent to the wider economic conditions in which they operate. Therefore, the systematic component of portfolio risk is unavoidable and undiversifiable. Idiosyncratic risk represents the effects of risks that are peculiar to individual firms, such as uncertain investments in R&D, new marketing strategies, or managerial changes. Decomposition of risk into systematic and idiosyncratic sources is useful because of the large-portfolio properties of idiosyncratic risk. As a portfolio becomes more and more fine-grained, in the sense that the largest individual exposures account for a smaller and smaller share of total portfolio exposure, idiosyncratic risk is diversified away at the portfolio level. In the limit, when a portfolio becomes “infinitely fine-grained,” idiosyncratic risk vanishes at the portfolio level, and only systematic risk remains. 425. The design and calibration of the IRB approach to regulatory capital relies on decomposing risk in this manner. Under an IRB system, the risk weight on an exposure does not depend on the bank portfolio in which the exposure is held. That is, while capital charged on a given loan reflects its own risk characteristics, such as the credit rating of the obligor and the strength of the collateral, it is not permitted to depend on the characteristics of the rest of the bank’s portfolio. To get this property of portfolio-independence, we must calibrate risk weights under the assumption of infinite granularity. Without this assumption, the appropriate capital charge for a facility would depend partly on its contribution to the aggregate idiosyncratic risk in the portfolio, and therefore would depend on what else was in the portfolio. With the assumption of infinite granularity, idiosyncratic risk can be ignored, so the appropriate capital charge for a facility depends only on the systematic component of its credit risk. 426. Of course, no real-world portfolio is infinitely fine-grained. Thus, there is always some idiosyncratic risk that has not been fully diversified away. If this residual risk is ignored, then a bank just satisfying IRB capital requirements will in fact be undercapitalised with respect to the intended regulatory soundness standard. To avoid such under-capitalisation, IRB risk weights have been scale upwards by a constant factor from the infinite granularity standard. The constant factor was chosen to approximate the effect of granularity on economic capital for a typical large bank. In order to capture variation across banks in granularity, we furthermore introduce a portfolio-level “granularity adjustment.” This additive adjustment to risk-weighted assets is negative for banks with relatively fine-grained portfolios, and positive for banks with more coarse-grained portfolios.
If the concepts of CDO and Bank are so fundamentally close, why are CDOs facing this acute lack of consensus in the valuation methodologies which is not present in the valuation of banks by equity and debt analysts? (Francisco Casanova)	Banks are well diversified as compared to many CDOs. The few banks like Northern Rock that are not so diversified have seen huge valuation volatilities similar to CDOs. In the past, banks that have faced deterioration of a major part of their portfolio have seen values go down to zero very quickly. For example, think of what happened to some of the largest East Asian banks during the Asian Crisis. In the late 80s/early 90s, some analysts had a target price of zero for Citi stock before it clawed its way back from the brink. This leads to the interesting question as to why CDO investors did not apply the lessons of centuries of bank management and regulation to CDOs. Why did they accept sector concentrations that would have raised eyebrows in banking? One possible answer is that this was a new field and investors were learning the lessons the hard way. The other answer is that the investors were diversifying across CDOs and so risk concentration in any one CDO did not matter. If this is so, then big value changes in individual CDOs do not matter; what matter is value changes of diversified portfolios of CDOs. These latter changes have been much less dramatic. The attachment points for AAA and AA tranches in many CDOs seem to have been badly off. A well run AA rated bank probably has a capital of 10% and a price to book ratio of 1.5 implying a cushion of 15% on a globally diversified asset pool with very little sector concentration. If we apply this benchmark to a CDO, the AA attachment point was I think badly off the mark. If the rating agencies had thought of CDOs as mini banks, would they have given these ratings? I think the answer is clearly no. Bank valuations have not suffered much because they had huge liquidity support. If the central banks had gone to sleep, what would have been the volatility in the valuations of the big banks? The CDOs have had no liquidity support. It would be interesting to compare the actual default rate in investment grade CDO tranches during this crisis with the default rate of banks during the days before central banking became so widespread. I suspect the default rates have actually been lower so far, but it will take a year or so to find out. If CDOs can survive this crisis without any central bank bailout, they may emerge stronger with better risk management, better valuation models, better rating practices, greater transparency and less moral hazard. In the long run, this crisis might be the best thing that ever happened to CDOs!

Posted at 21:48 on Tue, 09 Oct 2007     View/Post Comments (1)     permanent link