Posts Tagged ‘credit’

floating shorts

January 26, 2010

The markets have pounded Greece heavily last week, both in the equity and the debt front. There is widespread concern on the ability and comittment of the new government to service the country’s debt. Social unrest is evident. Greece also stands acused that, for years, statistics produced by her agencies were flawed at least.

Unlike statistics that are subject to interpretation, creative public accounting or outright reporting fraud, I take a look at the debt Greece issued over the last few years. These amounts are unambiguous, and any patterns that change over the years can give useful insights. To this end, I collected all information on Greek bonds and bills that Bloomberg records from 1993. As debt before 2001 was issued in old dracmae as well as an assortment of different currencies (ranging from pessetas to yen), I focus on the data after 2001. I expect Bloomberg to keep a representative, if not complete, sample.

Hopefully the data can shed some light on any policy changes that relate to the structure of debt. Also, the historical issues can serve as a yardstick to assess the magnitude of new bond or bill issues. One of the key years is 2004: this is the year of the overly expensive Olympic games, and also the year when the “socialist-technocrat” government of Costas Simitis was ousted, with a “conservative-liberal” government of Konstantinos Karamanlis taking their place in power.

In the attached tables you can find 105 short term Bills issues and 88 longer term bond issues. Number 23 in this list is the infamous structured bond that was in the center of the scandal that rocked the country, involving hedge funds, Greek state-run pension funds and JP Morgan.

The table below gives my summary, year-by-year:

2009 14,560 60,589 7.25 18.30%
2008 1,788 35,736 6.66 15.67%
2007 1,364 46,527 18.50 0.60%
2006 1,804 24,562 7.11 11.11%
2005 2,072 40,416 13.40 14.56%
2004 2,273 32,526 7.81 13.37%
2003 1,702 33,004 9.94 1.00%
2002 1,471 31,713 10.36 2.21%
2001 1,178 10,041 8.21 4.86%

Treasury Bills are issued many times in a year, offer no coupons, and have maturities of 13 weeks (~3 months), 26 weeks (~6 months) or 52 weeks (~1 year). Bonds offer coupons payable every six months that can be fixed or floating, and have maturities that range from 2 to 50 years.

One immediate observation is the spike of T-Bills issued in 2009, which is an order of magnitude larger than the typical amount of the previous decade. Presumably this is an attempt to accommodate the appetite of the lenders (or lack of it) to long-term commitments. It also indicates severe cashflow or liquidity problems on the part of the borrower (in the same way one goes to loan sharks for a few weeks until the benefits’ check comes through). And this trend looks likely to continue, if January auctions are an indication (EUR1.6bn and EUR1.2bn borrowed this January, compared to a total of EUR2.55bn last year).

Longer term borrowing has also increased substantially, standing now at about six times what is was 10 years ago, and two times the 2004 levels. But there are some details here worth mentioning: up to 2004, nearly all debt had maturity up to ten years, with the exception of only EUR18bn issued for longer maturities. After 2004, there appears to be an attempt to spread the debt across maturities, a policy that ended abruptly in late 2008. From this point on, most of the debt is very short, with the typical maturity being 5 years. And the new bonds issued this month will also have a 5 year maturity. Investors don’t seem to like putting their money on Greece for the long run.

Another interesting policy shift that happened in 2004 was an increase of floating debt. Up to this point practically all bonds offered fixed coupons, while at the moment about 20% pay interest which is linked to an external index, like the Euribor or some other combination of interest rates. A notable exception is 2007, which follows the scandal of the structured bond (yes, I know that correlation does not mean causation…) The last issue of 2009 pays 2.5% above Euribor; the new EUR8bn 5-year January issue will pay a premium of an extra 3.5% (for comparison, in January 2009 Greece borrowed EUR12.5bn, at 5.50% fixed coupon rate). This means that any interest rate rises in the next years will be increasingly painful for the Greek government, since interest payments for these new bonds will also increase.

These observations indicate investors that are only prepared to lend short-term, demand substantial premiums to do so, and do not want to bear any risk of future interest rate moves. These inverstors are nervous, and nervous investors can pull their money from the table if the going gets tough. Had Greece been able to borrow long-term and at fixed rates, she would not have to worry about investors getting itchy and pulling their money.

And all this matters. The Greek government keeps repeating that it is “they” and not “the markets” that determine their policy, but reality is different. As the figure above shows (borrowed from this FT article), Greece is heavily dependent on foreign investors, as they are holding about 30% of issued debt. As Greece takes her place as a central node in a potential systemic crisis, the probability of foreign banks that try to get out first increases. And this is an event that sets the domino off. The Greek government is desperately trying to front-run this eventuality, trying to expand the debtors’ base by promising bonds in USD and JPY.

The finger will remain on the trigger until investors decide to move away from short term government debt into long term fixed-rate issues. Only then will Greece have the space needed to produce a meaningful strategy that will put her house in order. It is chicken and egg once again…


the small print

January 11, 2010

It appears that the IMF will send a team to Greece in order to offer ‘technical assistance’. I am as clueless as you are on what ‘technical assistance’ actually means, but everyone in Greece is forcefully repeating that it does not mean ‘money’.

Greek bond description

I noticed the small print in the screenshot above, which describes a bond issued by Greece which is denominated in Yen (and is held mostly by a Japanese syndicate)

If issuer loses IMF membership or use of IMF funds, majority of hldrs (sic) may request issuer to call

I was not sure how to interpret that statement. Does it mean that using IMF funds constitutes an event of default, or that losing the potential use of IMF funds is actually the default event? If the former is the case, the Greek government is right to steer clear of any note with the IMF stamp on it. But I suspect it means the former, which is rather unlikely as even Zimbabwe remains an IMF member to date.

something’s got to give

December 18, 2009

For all terms and purposes Greece is a country under a fixed rate regime. Yes, it is part of the Euro which is a freely floating currency, but in reality (i) Greece’s state or actions do not affect the Euro rate, apart from creating worries that other more systemically important countries might find themselves in choppy waters, and (ii) the largest part of Greece’s trade is with Eurozone partners, and this takes place on a fixed trivial Euro-to-Euro basis (55% of imports and 65% of exports are with EU partners, of which the bulk is with Eurozone members).

This implicit peg has served well as an anchor for inflation, but might not be optimal in the current circumstances. If one were to give an IMF-style advice to Greece, then one would probably recommend a currency devaluation or resetting the peg (as this IMF surveillance document points out, where advice is given this is towards more flexibility). It would not inflate the debt away, as this is denominated in Euro, but it will (i) improve the balance of payments in the short run, (ii) inflate away wages that are currently sticky, and (iii) reduce the size of the housing bubble which is also sticky due to the unwillingness of sellers to accept reductions. Perhaps this would have been already optimal in a beggar-thy-neighbor fashion, as recession hit the country.

Paul Krugman suggests, in the case of Spain, that the only option is for nominal wages to drop. This is the route that Ireland has taken, but Greece seems to be unwilling to follow due to pre-election commitments. Does that mean that for Greece the only remaining alternatives are (i) exit the Eurozone, get back to the old Drachma and devalue, or (ii) wait for the IMF to enter the frame and force the wage cuts that the government will not?

The case for fragile regulation

April 7, 2009

Arnold Kling has a very interesting piece on regulation. Crises are endogenous, as economic agents and market participants learn how to game the system. He argues, that we are facing two alternatives: a system that does not break easily, but when it does it is costly and difficult to fix, or a system that breaks more frequently but is easier and cheaper to repair.

Kling argues that measures taken following crises are pivotal in creating the next one. He draws his examples from the mortgage markets from the Great Depression of the 30s, to the Savings and Loans crisis of the 80s, to today’s securitization dead end.

After the Depression it was acknowledged that mortgages with balloon payments are too sensitive to credit shortages. Balloon mortgages have a short term, say 5 years, but the monthly payments are similar to a long term mortgage, say 30 years. This means that in the end of the 5 year period there will be a substantial part of the capital still in place (the balloon), which has to be refinanced. If there is a credit crisis, refinancing might not be an option, leading to foreclosures. To this end long term fixed rate mortgages and adjustable rate mortgages (where refinancing is essentially mandatory) were subsequently encouraged. Such mortgages were given by Savings and Loan associations in the US, or Building Societies in the UK. Fannie Mae was established to promote and guarantee mortgages to families with lower income.

But this shifts the maturity mismatch from the mortgagor (borrower) to the mortgagee (lender). They would borrow short (from their depositors) and lend long in the form of mortgages. In the high inflation and interest rate period of the 70s and 80s, they would receive low cashflows from past mortgages, while at the same time they had to pay higher interest rates to their depositors in order to cling on their deposits. This eventually led to the crisis, where it was made clear that such maturity mismatches were also unsustainable. What emerged from the crisis was securitization, as a means of offloading the long component of their assets. In an ideal world, institutions that had long term liabilities (like insurance companies) would be the buyers of these long term assets.

Unfortunately, it didn’t work as intended with the current crisis being a direct result. Rather than insurance companies, it was investment banks and hedge funds that were left with mortgage backed assets. Also, the originating mortgage providers lost the incentive to monitor the quality of the mortgages they were granting.

The bottom line is that regulation, albeit well meaning, cannot be all encompassing and accommodating. Market participants will eventually work their way around, and there will be unforeseen consequences. It would be better then to have a fast moving framework that keeps the spirit rather than putting rigid constraints.

This is al good in paper of course, but how would such a flexible regulatory framework look like? What capacities should regulatory bodies have to make sure that the system does not descend into chaos? In a world run by politicians that want to cover their bases, the easy option is to talk big and construct a rigid international financial system. Which will probably take us back to the 1930s.

They told us that they told them to lend

April 1, 2009

James Hamilton in a brilliant post discusses the recent expansion of the monetary base.

As it stands, the Fed has bought an array of securities from the banks, of a varying degree of toxicity. They did not go to the bank headquarters with a truckload of money, they, sort of, wrote a check or an IOU.

Every bank is required to keep some reserves with the central bank. In that way they force banks to keep some of their assets in cash form, thereby reducing the risk of banks over-leveraging themselves (or at least that was the principle). In order to buy these dodgy assets, the Fed did not actually pay for them, they just extended the reserves of the corresponding banks, without the banks putting any money themselves.

To make things more concrete, say that Pitibank is required to keep USD50 with the Fed as reserve. Also, say that the Fed decided to buy from them a bunch of mortgages (of face value USD300), but which they agree to buy at USD200. These are “toxic” because if the bank sold them in the market, then at the moment they cannot find a buyer that will pay more than USD30.

The Fed gets these mortgages from Pitibank, and instead of paying in cash, they just increase Pitibank’s reserve to USD250. It is essentially the same thing, since Pitibank can draw from these reserves if they want to (down to the requirement of USD50). The Fed will then essentially “print” money to give Pitibank when asked.

What politicians are telling us, is that they want banks to start lending again. This essentially means that they want Pitibank to ask for these reserves and lend the money to individuals, small businesses and each other.

But banks don’t appear to be doing that. They prefer to let the money in their reserves with the Fed. Why do they do that? Are they keeping good money idle? Well not really, because the Fed has decided to actually pay interest on these reserves. This increases the opportunity cost of lending the money to you and me, and acts as a dis-incentive for Pitibank.

Does that make sense? Well, no matter what the politicians are saying, central banks are having nightmares of all that money reaching the consumers and sparking inflation. As J Hamilton points out, doubling the monetary base will eventually translate into 100% inflation: all prices will double.

It seems to me that with that interest rate on the reserves, the Fed is trying to control how fast this huge amount of money reaches the real economy. And they don’t seem to be very keen for this to happen.

Already below the adverse scenarios

March 31, 2009

The Federal Deposit Insurance Corporation (FDIC) is an agency “created by the Congress that maintains the stability and public confidence in the nation’s financial system by insuring deposits, examining and supervising financial institutions, and managing receiverships.” They were heavily involved in the bailout of Citi and have been active in the so called “Legacy Loans Program”, which is the vehicle via which the US taxpayer is buying all those dodgy “toxic” mortgages.

A few months ago, the FDIC released this FAQ, outlining how they would determine if financial institutions “have sufficient capital buffers”. To do that, they devised two scenarios: a “baseline” and an “adverse” one. The baseline scenario is some sort of expert expectation, while the adverse is a stress that is “unlikely”, but “cannot be ruled out”, as their document explains.

House price scenarios used by the FDIC

House price scenarios used by the FDIC

The index they use is the Case-Shiller index of house prices in 10 US cities. This index is reported monthly with a two-month lag. The average index value over the 3rd quarter of 2008 is 165.92 which is the base of the figure above. Today we got the index value for January 2009, which was 158.04 or a drop of 4.75%. This is the first month of the scenarios that the FDIC constructed, and it seems to me that we are already covered the baseline scenario for the whole quarter. We are also en route to go through the adverse scenario pretty soon.

Well it is only one month, but certainly not an encouraging one. The US taxpayer must be delighted with the guys that buy loans for her…

barclays’ creative accounting

February 10, 2009

Barclays announced that they made a profit to the market’s delight. It is interesting to see where this money comes from, and which part is actually real.

The announcement document hides a few real gems.

On page 7, fourth bullet point, one reads that Barclays books “gains on own credit £1663m”. And this amount is clarified on page 29, naked in all its glory, as “[…] gains of £1663m (2007: £658m) from the general widening of credit spreads on issued notes by Barclays Capital”

What does that mean? Well Barclays have issued bonds that other market participants bought. These bonds are promises, made by Barclays, to deliver a fixed amount in the future.

Say that the amount is GBP100. This GBP100 are clearly a liability for Barclays, but now they are making the following claim:

Well we might default in the near future, in which case the debt is forgiven and bondholders take zip. The credit spread quotes put a price on this default probability, say that they imply a probability of 5%. Then, according to the current accounting standards, we don’t really have a liability of GBP100 but a liability of GBP95 only. So compared to last year when our default probability was negligible we have made a gain of GBP5.

What is the bottom line? the higher our probability of default => higher probability that our debt is forgiven => lower current value of our debt => higher gain we book. Nice!

This blog is doing a good job in explaining the rationale behind the standards that allow such acrobatics.

There I found the quote: “If your own credit spread widening counts as revenue, and you pay compensation as a percent of revenue, the most profitable and lucrative day in the history of your firm will be… THE DAY YOU GO BANKRUPT!

W Buiter, on the other hand, goes berserk about that. He calls it “mark-to-market gone mad”.

A similar construct appears in counterparty risk management a’la Basel II, where capital can be released as one becomes a worse counterparty.

my wife wants a cooker

January 26, 2009

My wife wants a new gas cooker, and as a decent husband I took her down the shops to choose. We went around the usual stores, including Currys and Comet, the largest white goods retailers in the country.

After doing the rounds, a very interesting pattern started to emerge: pretty much all cookers that we asked for were out of stock. This was true even for items that were on discount, which we found a bit bizarre. The sales guys seemed to have no clue as to when they would come back on, with most of them guessing a four to six week wait. They were all keen to point out that we should nevertheless put our order forward, in order to secure our enamel dream.

Now as an expert in all things financial, I thought it would be interesting to check the stock and CDS prices for DSG International (owner of Currys) and Kesa (owner of Comet).

The CDS price is the insurance premium one would pay to protect themselves against a company defaulted. Typically it is used by bondholders as insurance (but, as you might have noticed, all sorts of unscrupulous financiers have been speculating with that stuff recently). Nevertheless, for DSGI the CDS quote was a whopping 18%. This means that if I want to buy insurance for DSGI bonds that promise to pay me EUR100, the premium would be EUR18 per year. Now that’s quite a lot of money, which means that someone believes that these guys might actually default. In fact, the implied probability of default would be at least 18% per year, depending on what proportion of the EUR100 I will be able to recover if DSGI default. With the standard 40% recovery assumed in the market, the default probability is an amazing 30%. This is roughly 2.5% per month.

To recap, some guy wants to sell me a cooker he doesn’t actually have yet. Actually, as far as I could tell, he doesn’t seem to have many cookers at all. But he nevertheless wants me to pay more than 500 quid in advance for it, in full. He promises that he will go and get it for me in the next couple of months, but there is a decent chance that he will make a runner.

Is this how we are going to get ourselves out of this mess? No way mate.

(Unfortunately –or fortunately– Kesa don’t have CDS contracts, and for that reason I could not replicate the analysis. The same is true for Home Retail Group that own Argos.)

bailing out greece

January 14, 2009

Standard and Poor’s today downgraded Greece to A-, which was largely expected. Although Greece is a member of the Eurozone, S&P see the increasing deficit as a problem. And arguably, a bigger problem is that although Greece has experienced some remarkable growth in the last few years, the deficit issue was not adequately addressed by the government. In fact Greece has one of the highest public debt to GDP ratio in the world, and topping that with one of the worst current account deficits does not really help.

The folks at the have a very interesting recent presentation which touches on the matter. On p.11 there is a discussion on who would bail out Greece in case of troubles (either deserved or undeserved). As I understand it, their point is that no institution will be interested or capable of rescuing the Greek economy. They also see some cracks appearing in the <<weakest link>> of the Euro area.

And Greece will need to go to the global credit markets for EUR50bn+. Only yesterday they raised a mere EUR2.5bn, amidst higher spreads, which are now about 250bps (2.5%) above the German yield. Now they will have to raise 20 times that amount.

Should we expect spreads shooting up and Greece defaulting on the existing debt as a result of its inability to roll over? A speculative attack on Greek bonds? A test for the euro?

An IMF document discusses the Greek economy. Overall Greece is exposed to south-eastern Europe and the Balkans, which is not surprising. What surprised me is what a crap job Greek banks do within Greece, with such a high percentage of non-performing loans recorded in a boom period. I can only guess the quality of the loans that they have given in Albania, Romania and Serbia…