Debt and Uncertainty
An interesting posting by Jonathan Hopkin made me wonder what debt is for, specifically government debt. Of course it allows us to borrow money to meet current needs, especially large unexpected current needs - like nationalising a banking system. But also, crucially, it allows us to fund assets of uncertain, fluctuating value.
Here's what I mean. The problem with ABS is precisely the AB part: the assets back the security. Therefore as the assets fluctuate in value, so does the security. Specifically if the assets are uncertain is value, so are (some or all, depending on the tranching and the credit enhancement) of the issued securities.
But ordinary unsecured debt depends just on the value of the promise to repay by the issuer. If this promise is of good quality, the security is valuable regardless of the assets that are being supported with the liability. In other words, by buying risky assets and issuing sovereign debt, the government does not just take credit risk - it also removes valuation and funding uncertainty from the financial system.
Consider this. (I'm not saying this is a good idea: it is just a thought experiment.) Suppose the government sets up The Office of Security Support. TOSS undertakes to do two things. It will repo risky assets to term at fixed haircuts. And it will provide firm prices on the bid and offer for risky securities in size. Each use of TOSS facilities requires paying a certain equity stake as well as the cash demanded, and the users must covenant to use a certain fraction of the funds raised to conduct new lending. Now undoubtedly running TOSS would be a difficult business and it would have some losing trades. But if the equity stakes required were large enough and the bid/offer spreads were wide enough, then it should be possible to run it without enormous cost to the taxpayer. Would this alone be sufficient to rescue the banking system?
Here's what I mean. The problem with ABS is precisely the AB part: the assets back the security. Therefore as the assets fluctuate in value, so does the security. Specifically if the assets are uncertain is value, so are (some or all, depending on the tranching and the credit enhancement) of the issued securities.
But ordinary unsecured debt depends just on the value of the promise to repay by the issuer. If this promise is of good quality, the security is valuable regardless of the assets that are being supported with the liability. In other words, by buying risky assets and issuing sovereign debt, the government does not just take credit risk - it also removes valuation and funding uncertainty from the financial system.
Consider this. (I'm not saying this is a good idea: it is just a thought experiment.) Suppose the government sets up The Office of Security Support. TOSS undertakes to do two things. It will repo risky assets to term at fixed haircuts. And it will provide firm prices on the bid and offer for risky securities in size. Each use of TOSS facilities requires paying a certain equity stake as well as the cash demanded, and the users must covenant to use a certain fraction of the funds raised to conduct new lending. Now undoubtedly running TOSS would be a difficult business and it would have some losing trades. But if the equity stakes required were large enough and the bid/offer spreads were wide enough, then it should be possible to run it without enormous cost to the taxpayer. Would this alone be sufficient to rescue the banking system?
Labels: MOAB
0 Comments:
Post a Comment
<< Home