"The Geithner plan: criticisms are off the mark
April 7, 2009 5:21pm
By Michael Spence
Depending on who you ask, the pubilc private investment programme announced by Tim Geithner is either part of a solution to today’s banking crisis or an aggravator of the problems. This debate will likely widen as the US government moves from the design stage to implementation.
What are the key features of the Geithner plan? Government and private investors put in equal amounts of equity; this is supported by the provision of cheap leveraging through government debt (up to six times leverage); and the government provides a non-recourse feature equivalent to a put option in case that the ultimate value of the package of securities turns out to be less than a pre-specified amount. The purpose of this feature is to take away the risk of a large loss of the investor. It is insurance against the left tail outcomes. In return for the put the government takes warrants which add to its equity fraction in case the final value is greater than the debt. Sellers auction packages to the buyers. The private investors in the buying entity (with the government) set the bid prices.
Markets reacted well to Mr Geithner’s efforts to line up capital, financing and management expertise in order to liquefy the market for legacy assets residing on banks’ balance sheets, thereby facilitating their lending activities.
At the same time, in widely-publicised remarks, several prominent economists criticised the plan, arguing that it provides government/taxpayer “cash for trash,” constitutes a huge giveaway and robbery of the US taxpayer. This is a mistaken view. It is based on an assumption that debt will be overused by the government and the participants in cases where the left-tail or downside risk is very high.
The plan is designed to restart frozen markets in securitised assets, and inject capital whose value is easier to understand. These are important steps in restarting the private credit system. But not the whole package. It is not intended to solve the potential insolvency problem. That will take direct injections of capital and expanding government ownership, once the insolvencies become clear.
There are hurdles. One is that there are no assurances that the increase in demand for legacy assets will be matched by increased supply. Many banks may find it difficult to live with the consequences of explicitly valuing the securitised debt on their balance sheets, some of which is toxic. Pressure in the form of the stress tests (forensic audits) may be needed.
Third, implementation is crucial. Investors and the government have to do enough due diligence to know what they are buying. Otherwise there is a potentially serious adverse selection problem in which only the lowest quality packages get offered for sale.
Now we come to the use of non-recourse debt. Packages of securities put up for sale will undoubtedly have widely varying levels of quality as measured by the both the value (in relation to book) and the risk or degree of uncertainty. The programme was designed to accommodate this considerable variety in quality. Recall that markets for securitised assets are broadly frozen and not just for the portion that is characterised as toxic or trash.
Let’s say the goal of the non-recourse provision is to take away the bottom 10 per cent of the outcomes in terms of the value of a package. This will make it attractive to the private investors by lowering the risk premium they need. The government which is participating in all the deals can recover the cost of the insurance they are providing with the warrants. With the ten per cent target, some packages with relatively low variance will support leverage of 3 to 4 times as one often sees in private equity deals. Other packages with very high variance and risk (the low quality ones) will support almost no debt. The intuitive reason is that if you have significant leverage and high risk, the chances of the value falling below the debt will be above 10 per cent.
The crucial point is that the rate at which the value of the embedded put option rises with leverage depends on the variance of the distribution: high variance - rapid increase, low variance - moderate increase. Perhaps the easiest way to see this is in the extreme case where the risk is negligible. In that case one can use a lot of leverage and the insurance or non-recourse feature is worth nothing because the value once realised will never be below the debt level.
Government therefore needs to confine the use and match the level of leverage to cases in which the value may be impaired but the uncertainty is low to moderate. If it permits the use of high leverage in high risk cases, it will end up funneling assets primarily into banks. The investors have to bid on the packages. With excessive downside insurance they would end up paying more than the packages are worth.
The criticisms seem to assume that all the packages are trash (in slightly more technical terms, low value relative to book and high variance or fat tailed) in which case the use of non-recourse debt automatically involves a significant taxpayer transfer to some combination of the buyers and sellers, mostly sellers.
If the critics are right that all the packages will turn out to be high risk (often measured by the variance) and if the government implements properly then there won’t be much leverage in the observed outcomes.
The reason this is important and not just technical, is that stabilising the financial system is going to require a complex set of government and central bank initiatives undertaken with imperfect knowledge of consequences.
That challenge is going to be much harder with a competent and well-intentioned government (which we have) if our fellow citizens who are understandably confused and very angry, think the government is trying to bail out the financial sector and doing it in a surreptitious way.
Michael Spence received the Nobel Prize in economics in 2001 and is chairman of the Commission on Grown and Development"



































No comments:
Post a Comment