Felix Salmon's article, 'The Silver Lining to Synthetic CDOs' is interesting. I do not disagree with him. In fact, I agree with one of the implications of his article that a CDS market makes it easier to hedge. It is true that the prevalence of sovereign CDS has helped Greece deal with its debt rollovers, as investors buying Greek bonds know that they can hedge these using long sovereign CDS positions.
However, one of the most important aspects in the merits of cash CDOs over synthetics is in the role the former plays by virtue of the way it is constructed. It may be true that cash CDOs helped create a demand for asset backed securities although one may well argue that asset backed securities created a demand for cash CDOs. But more important is the fact that cash CDOs play a pivotal role in allowing an excess supply of asset backed securities, be it credit card debt, auto loans, prime and subprime mortgages, loans - virtually any type of asset backed by a projected cash flow structure - to be absorbed into an investment vehicle structure and be able to be sold to end investors.
Synthetics, by definition, do not involve the actual creation of such asset backs. In fact, synthetics only allow an investor to build exposure to a particular long or short position in the asset backing the synthetic CDO tranche. This is one reason why the notional amounts on some types of synthetics can, and have, outgrown the actual cash securities underlying the synthetics. Synthetics play a different role, important but entirely different. Unlike cash CDOs, synthetics do not help lubricate the channel from the point of creation of actual asset-backs to the point of offloading these asset-backs to end investors.
Effectively, I am arguing that synthetics are ideally structured to help create a marketplace for speculators and hedgers to interact. This, in itself, is an extremely important function as it provides a marketplace for off-loading and on-loading exposures to an eclectic mix of assets. This allows market participants to manage risk or take prop exposures. In this way, synthetics help in the discovery of the market price of risk for exposures. Regulators are belatedly grappling with the issues that CDS markets are opaque and lack liquidity, and are affected by technicals or are diminutive compared to the underlying cash markets, yet CDS still help in the price discovery process. Cash CDOs, on the other hand, perform a completely different economic role.
While tranches of cash CDOs can help in price discovery when it is traded or changes hand in a market place, since the cash flow underpinning the tranche in question depends on the specific structure of the rules underlying the distribution of cash flows amongst the different tranches (the waterfall), price discovery is limited. Another way to put this is to say that since synthetic tranches are standardized contracts unlike tranches on cash CDOs, the former play a better role in the price discovery process and are, therefore, better benchmarks for price discovery. Standardization helps prices be more of an-apples-to-apples comparison over time and across asset-types.
However, unlike synthetics, cash CDOs are animals that chew the cud in that they actually part take in the circular flows of goods and services in an economy - recall the circular flow of goods and services diagram in any basic macro 101 text book - by allowing loans, mortgages, credit card debt to be taken from one part of the economy represented by borrowers to another part of the economy, the savers. Thus, they help grease the circular flow of goods and services in an economy. This is a very important economic function that cash CDOs perform – in many ways the same as the monetary and resource transmission mechanism a bank performs. A synthetic does not perform this role.
Another way to put this would be to say that a cash CDO results in real effects in an economy as in production of real goods and services. It helps lubricate the factors of production by allocating resources from the savers to the borrowers. If, as it happened during the housing boom, such production is tilted towards overproduction of a particular product leading to a supply glut and a fall in prices, the fault is not necessarily in the use of cash CDOs acting as a distribution channel to help match supply and demand for funds between borrowers to lenders. The fault is in regulation and the inappropriate use of that channel. The argument is no different than genetic research that may be used for common good of mankind or to play havoc by manipulating genes to create Frankensteins.
In a way, this is like the concept of classical dichotomy in economics that says, in the long-run, real variables affect real variables and nominal variables affect nominal variables. Cash CDO is a real variable. It is an important factor of production that results in helping join savers and borrowers. A synthetic CDO is like a nominal variable, more of a 'management' tool since it warrants no 'ramp up' and is ready to work as an exposure building or reducing mechanism almost from the outset.
As the Shleifer paper on financial innovation says, "Optimism about the profitability of the new claim at t = 0 encourages the intermediary to over-invest in an unproductive activity, eventually triggering a loss… Investment in A occurs only if new securities can be engineered, so financial innovation bears sole responsibility for unproductive investment. It can be argued that the expansion in the supply of housing in the last decade was an example of such inefficient investment needed to meet the growing demand for securitization of mortgages".
The damage caused by synthetic CDOs may be less than that caused by a cash CDO not because synthetics are less (or for that matter more) harmful but because synthetics are neutral in terms of real output. Consequently, if directed towards a socially harmful investment they are less harmful than cash CDOs as their nominal nature has less of a real effect. As a caveat, it is necessary to point out that in the same vein, just as classical dichotomy breaks up in the short-run, synthetics may have limited real effect at times as CDS spreads do affect cost of funding of corporations and, thus appear, in the weighted average cost of capital of a firm. I am implying that synthetics do not have much of a non-nominal, direct real effect in the long-run in terms of output.
The best way to defend the cash CDO is to imagine the following.
India Inc. sets up cash CDO - called IndiCash – in India that takes SLABS (student loan asset backed securities) and distributes it to end investors in the form of tranches. So a large number of loans given to young people to study IT in Bangalore are packed up into tranches and sold to investors. Tranches have different prepayment, extension and default risk and different investors buy the tranches as per their investment preferences. As a result of the cash CDO structure, borrowers (the students) are allowed to interact with savers (investors) through financial intermediaries (banks like Citibank (C), Bank of America (BAC), JP Morgan (JPM)). The cost of loans is lower than what it would otherwise be since the financial intermediary is keeping only a smaller portion of the loan on its books and the rest is offloaded to end investors. At the same time, US Inc., situated in the United States, sets up a synthetic CDO - called AmeriSyn - that has, by virtue of a pen and a paper contract, set up tranches that pay off exactly the same returns as that of IndiCash.
Graduates using SLABS find work in the IT sector in Bangalore thus creating quality software most of which is exported to the United States. In return, India earns foreign exchange and increases its overall level of literacy and knowledge, especially in the IT fields. The Indians churn out a steady supply of educated IT graduates who build set up services and technologies that over time are thick enough to stand in competition with Silicon Valley. Additionally, there are positive externalities as technologies and wealth is created and real goods and services prop up in Bangalore, neighborhoods are gentrified, property prices increase, schools are set up etcetera.
US Inc. on the other hand, with its synthetic CDO reaps the same cash returns from the tranches - as students graduate and find jobs in Bangalore, the investors owning the tranches of the synthetic AmeriSyn get returns on their investments. However, for US Inc. the externalities effect is absent. If US investors reinvest the investment proceeds, there would be trickle-down externalities but those accrue from an entirely different investment decision.
Which society is better off down the road? Both, since investment returns have been earned in both countries, but Indians, with the positive externalities, are markedly much better off.
If on the hand, imagine for a minute that IndiCash comprised of tranches made up of shark loans given to the Mumbai underground gangs to fund their illegal activities. Assume further that the underground gangs are quite 'successful' in doing what they do, then the end investor, whether the one investing in the cash CDO in India Inc. or the synthetic in US Inc. will get the tranche’s indicated return. While the return will not be in line with an ethical mandate, it will still be there. Yet, in such a case, US will, undoubtedly, be better off using the synthetic CDO rather than India with its cash CDO and the concomitant negative externalities - murders, shootouts, encounters, drugs etcetera.
The bottom line is this - Had the loans been directed towards a socially beneficial investment like education, or a reasonable allocation of resources towards building up a housing stock, they would have played an overall beneficial role, as cash CDOs are conduits to direct production of real output, unlike synthetics that are more of a management tool.
Ultimately, the fault, dear Brutus, is not in cash CDOs that they ended up as Pangloss investments and an avenue to 'over-invest in an unproductive activity'. The fault is in regulation, or lack thereof.