Talking Point

What rhymes with synthetic CDO?




DealVector coo and co-founder Dave Jefferds suggests that synthetic CDOs may fulfil their potential if investors can solve three issues

Now that synthetic CDOs are back in vogue (SCI 15 August), will history repeat itself, or only rhyme? Over the last two years, volumes have grown from US$10bn to US$30bn for the instrument widely blamed for causing the 2008 global financial crisis. What gives, and should we be worried?

To answer this question, it is important to understand what makes synthetic CDOs attractive instruments. Two words: customisation and leverage.

Perfect customisation of risk is something that everyone should agree is a good thing. Synthetic CDOs allow an investor to tailor exactly the basket of credit risk they want, including maturity, diversification and subordination. This precision is very difficult to achieve by investing with cash bonds.

Trading cash bonds is 'lumpy' and slow. For example, I may want to take an exposure in Company A until March 2019 exactly, in an amount of US$5m exactly. But there may exist only a bond that matures in January 2020 and, after hunting around, perhaps the only position available for sale is a block of US$7.3m.

To accomplish my risk preference goal, I would need to buy the US$7.3m, immediately sell US$2.3m and then sell the remaining US$5m in March 2019. Both extra steps incur transaction costs and uncertainty.

Multiply this inefficiency by 100, if the aim is to assemble a diversified portfolio of exposure.

Further, synthetic CDOs allow investors to calibrate perfectly their leverage. They may dial up their risk and return by taking the first loss of a portfolio, increasing their leverage.

Conversely, they may take a more secure and safe senior position in the basket for a lower return. In this case, they are deleveraging.

Some point in the middle would be equivalent to the simple return on the portfolio with neither lending nor borrowing. The point is that these choices allow for single-stop shopping and total customisation of the exact portfolio that is desired. The process provides significant efficiency for portfolio managers.

So what is the catch? The catch is the counterparty risk.

Both parties to a synthetic CDO contract (that is, the investor and the bank that provides the service) must live with each other over the life of the transaction. In the example above, the two parties would depend on each other to perform through March 2019.

By contrast, in any cash bond trade, settlement is in three days and neither party sees each other again. So in a synthetic CDO, as with any swap, guarantee of performance over time is critical.

To ensure performance, investors commit to a variety of contractual terms that are fraught. The bank, for example, determines the value of the contract at any point in time.

The value of the contract, in turn, determines how much cash must be wired between the parties as margin to guarantee continued performance. A global credit support annex (CSA) between the parties determines how much additional cash must be wired when the mark-to-market value on all contracts together (not just one) decreases by specified amounts.

Imagine if the bank could tell you the value of your house on any given day, and the valuation the bank provided then determined how much additional money you had to wire to the bank that same day. Furthermore, imagine the bank had a veto over who could buy your house, based on the bank's assessment of the buyer's creditworthiness. In any drop in housing prices, your options would be very limited and the likely result would be that the bank would own your house for a song.

This is the position that hedge funds found themselves in during 2008. Synthetic CDOs were marked down (across the board).

Hedge fund monthly performance suffered. Simultaneously, banks asked for more collateral to guarantee the trades.

Also simultaneously, hedge fund limited partners wanted their money back because of the poor performance. So hedge funds then had to liquidate these synthetic CDO swaps at prices they thought were ridiculous on a fundamental basis.

In essence, using synthetic CDOs, hedge fund investors on the one hand tailored their investment assets perfectly. But they totally screwed up their hedge fund liabilities.

They left themselves open to needing cash exactly when the cash would not be available to them. This is called asset-liability mismatch.

So, will history repeat itself? The volumes of new synthetic CDOs printed now are trivially small compared to what we saw during 2007 and 2008, when the annual volumes exceeded US$1trn and the aggregate volumes were over US$10trn.

So, there is no systemic risk issue at the moment. The real question is, if these volumes grow, will investors be able to negotiate terms that more effectively balance their asset-liability risk?

To do so, they will need to address at least three important issues. First, they will need to make sure they have terms with LPs that match their investments.

Planning to invest in synthetic CDOs with four-year maturities? Better make sure they have some ability to lock in their cash for that period of time.

Second, can they obtain objective third-party analysis on mark-to-market? During the 2008 crisis, the typical method for ensuring good marks was to get three quotes from brokers.

But in a world where all the brokers were similarly situated and the number of desks that even had the ability to price such instruments could be counted on one hand, this was no protection. It will be important that investors address this risk and it is not an easy issue to solve.

Third, investors will need to negotiate stronger CSAs that will mitigate the risk that they will be 'carried out' by margin calls before the investments can realise their true fundamental worth. This issue is also a tricky one in a world where there are relatively few bank providers of synthetic CDOs and these few are heavily concerned with regulators. The best solution here for investors probably is to confine their risk taking to short durations, where the mark-to-market discretion is less likely to become the primary issue.

If investors can address these issues, history will rhyme but not repeat. Maybe this useful product will fulfil its potential and mature into something more robust that benefits rather than destroys the market.

11/09/2017 14:52:04



Copyright © structuredcreditinvestor.com 2007-2019.