Sent by anonymous, 4 June 2007:
As an investment IFA I can sympathise with the question dated 23.05.07 – in relation to asset allocation and the resultant bond exposure so many stochastic modelling tools tell us to have. We have been underweight in Fixed Interest in general as an asset class for about 1 year now and going overweight in UK commercial property funds. This I am now taking back down to neutral (as I think UK property has had its day) – but I am still nervous about increasing the fixed interest exposure. I have looked at funds that use the UCITS III powers ie "strategic" bond funds and have to say that they look exciting. On the flip side however, I read a report that suggests some managers do not have sufficient knowledge to use the derivatives effectively and this could cause a major problem.
I would be interested to hear your views on this topic – particularly in relation to CDS, CDOs and the impact the US sub prime market problems could have on these.
The CDS (Credit Default Swap) market has developed rapidly, and in just a decade is now more liquid than the conventional corporate bond markets. There are undoubted benefits to CDS, namely that derivatives can provide investors with pure exposure to a company’s credit risk (so it doesn’t matter whether interest rates go up or down), and investors can make money from correctly forecasting that a company’s credit worthiness will deteriorate. The development of CDS has also been integral in spreading and diversifying risk within the economy as a whole, although the sheer size of the market has led to concerns.
The biggest worry is whether derivatives can withstand the impact of a large unforeseen shock. By definition, it’s not possible to predict a large unforeseen shock, but all we can say for certain is that there will be one. Last week, Terrence Checki, a vice president at the New York Federal Reserve, warned that "abundant global liquidity has been a powerful wind in the back of economic and policy progress and has bought substantial benefits [but] as we all know, liquidity is ephemeral: it disappears at the most inconvenient times". This is precisely what happened after the Russian default of August 1998, which was the last quake to shake the bond markets.
The Russian default caused financial markets around the world to seize up. Global liquidity evaporated, even for US Treasuries. The sudden and sharp spread widening wreaked havoc on what were supposedly low risk investments. Higher risk investments were inevitably hit hard – between July 20 1998 and October 5 1998, the FTSE 100 fell from 6179 to 4648. Perhaps the highest profile collapse was that of the Long Term Capital Management (LTCM), a hedge fund that lost almost $5bn in a matter of months and had to be bailed out by the Federal Reserve in order to prevent a widespread global meltdown. Now that the derivatives market is many times bigger than in the 1990s, what will the impact be of another Russia?
The problem is that the CDS market hasn’t been properly tested yet, because defaults in the global bond market have been incredibly low for an almost unprecedented period of time. This is not to say there have been no defaults at all though – Delphi, the world’s biggest auto parts manufacturer, defaulted in 2005. Delphi had around $2bn worth of bonds outstanding, but the value of the CDS riding on Delphi was over $20bn. Traditionally in the CDS market, contracts were settled by delivering the actual bonds, so investors that had bought protection needed to go into the market, buy the defaulted bonds, and then sell the bonds at par (100% value) to investors who had written protection. With Delphi there were concerns that the huge number of derivatives would create settlement problems, however a large proportion of the derivatives were either netted off against each other or settled for cash, and settlement in the end proved fairly straightforward. Cash settlement is becoming the norm in the CDS market, and the risk of any major settlement issues cropping up in the future is falling as the market becomes more transparent and efficient.
Moving onto CDOs (Collaterised Debt Obligations), the issues facing CDO investors (specifically synthetic CDO investors) are similar to those facing CDS, because synthetic CDOs are invested in CDS. With a CDO, an investor can choose precisely how much risk they want to take. Someone who invests in an equity tranche faces potentially large rewards but knows that if the portfolio is hit by defaults then they will be the first taking a hit on their capital. An investor in an AAA rated tranche of a CDO knows that he or she won’t lose any money until all the investors ranked beneath them are wiped out. The risk facing CDO investors, though, is knowing how much risk they are actually taking. There’s been a fair amount of criticism directed towards ratings agencies (see here for a good piece from the FT) who, it is argued, aren’t up to the job of rating sophisticated structured products. Does a AAA rated synthetic CDO tranche really carry the same default risk as a US Treasury? We’ll only know for sure when there’s a liquidity crunch.
The woes of the US mortgage market have hit a number of CMOs quite hard (CMOs are like CDOs, except the underlying portfolio is invested in mortgages rather than bonds). We’ve covered the US sub prime mortgage market in a fair bit of depth on this blog (see here, for instance). The yield on a number of CMO tranches that were previously rated investment grade suddenly shot out to 15% once the size of the collapse became apparent, but at the moment, it doesn’t look like the US sub prime crisis is the next Russian default. After pausing for breath, equity markets have more than recovered their losses and jumped to record highs in the US.
Finally, the concern over fund managers’ experience in dealing with products and instruments such as CDOs, CLOs and CDS is certainly a legitimate one. Some asset management firms are able to rely on their experience in alternative assets to build up the necessary operational infrastructure. M&G are lucky to have a market-leading CDO team who have been using CDS for over five years. The use of Value at Risk (VAR) analysis is also critical to control portfolio risk. But it’s less likely to be an operational issue that hurts investors rather than bad judgement from a fund manager. Investors need to be aware that while buying protection is not a risky strategy, since the downside is capped, writing protection has the potential to result in a fairly large loss. Given our views on credit risk in the corporate bond market, Richard Woolnough has mostly used CDS to buy protection in the M&G Optimal Income Fund, since he thinks there are a greater number of unattractively priced corporate bonds out there than attractive ones. When the next Russia comes along, it will likely be those funds taking excessive risk by writing protection on companies whose bonds default that will feel the pain. In short, if you make bad investment decisions using CDS and CDOs you will lose money, just as you would in traditional instruments, but there is nothing intrinsically bad about the derivatives and structured credit markets, and indeed they are often used to reduce and diversify risk rather than increase it.