We’re a bit perplexed by the exuberance in the markets and in Washington DC over the expectation that the government may create a so-called ‘bad bank’ that would buy up troubled assets. First of all, haven’t we seen this movie before? We think it was called TARP I: First Blood.
It didn’t work out. The banks needed more capital too quickly, and the logistics of buying the assets were too difficult. There was a fear that the government would either overpay for the assets or render banks insolvent by purchasing them at actual market levels. There was no way to actually price the assets because all the models for projecting expected returns are junk. Political pressure would interfere with the process. So the Treasury Deparment abandoned the plan.
As far as we can tell, nothing much has changed. Big banks are still desperately scrounging for capital. The government is still faced with the problems of pricing illiquid assets. Politics is still politics, even after Obama has been president for over week.
But we think we know how an aggregator bank might actually be able to pay more for troubled assets than banks could fetch in the market place but still not overpay. The trick to this is that the assets may be worth more in the government’s hands than in the banks.
How could that be?
Well, to get to that happy conclusion you have to agree to some pretty questionable assumptions and one really big assumption. Let’s start with the big assumption, which is that mortgage backed securities could be worth more if the terms of the securities and underlying mortgages could be adjusted. That is, modified mortgages might default less and, as a group, be worth more than unmodified mortgages.
There’s good reason to be sceptical of this assumption. If it were true, we’d expect the MBS market take care of this by modifying the terms of the contracts. That’s not happening, which is evidence that modification won’t increase value.
One the other hand, perhaps the market isn’t operating as smoothly as that quick read suggests. We know that transaction costs can often prevent otherwise efficient outcomes. Holders of MBS face huge collective action costs that could prevent effecient modification. organising all the holders and getting them to agree to modifications is difficult. Those transaction costs could be so high that they outweigh the benefit of modification. Thus, individual MBS might trade at what we would call a ‘transaction cost discount.’
If the government were to aggregate the securities, however, it could make the modifications without facing these transaction costs. This means that aggregated MBS would be worth more than dispersed MBS. That added value from netting out the transaction costs could be passed onto banks, allowing the government to overpay for assets without guaranteeing a loss.
Is that true? Well, like we said. You have to do a lot of assuming to get to that conclusion. And you have to be confident that the government can somehow actually measure the price of the transaction cost discount. And that paying for MBS at market prices net the transaction cost discount would be enough to keep the banks solvent.
But, technically, it’s at least possible.