The Obama administration is reluctant to nationalise banks (or at least some of them). But an increasing number of economists are calling for nationalisation, and even prominent Republicans have now expressed their support for temporary nationalisation.
The argument heard most in favour of nationalisation of banks is financial. Banks’ losses were so important that only the government is in a position to save the financial system by investing heavily in ailing banks. At the end of the Bush administration, the Paulson plan included taking stakes in state banks without voting rights. But the rights of the taxpayers must be protected – the state cannot become the main shareholder of banks without taking control.
There is another argument, implicit or explicit, for the nationalisation of banks – we cannot trust bankers not to leave with the cash, let alone spend any assistance in the general interest. Even if other forms of assistance are feasible, the risk that bankers would find a way to use the contract in their own interest is high. In other words, moral hazard is pervasive. Two studies analysing the experience of recent years show that bankers will not hesitate to enrich themselves at the expense of the public good if they have the opportunity.
The securitisation of loans has been identified as one of the triggers of the crisis. New financial assets group hundreds of individual loans and cut them into shares sold by the piece. The merit of these products is to pool risk. But the danger is that the institution that identifies the borrower no longer owns the assets. If the demand for these new financial products is important (as it has been since their introduction), the temptation for banks to grant loans to borrowers who are increasingly risky is strong. And even stronger is the temptation to make these new borrowers appear less risky than they really are, especially as the buyers of these loans are not traditional banks and therefore have no expertise to evaluate them.
Two Chicago economists (Mian and Sufi 2008) have shown that this occurred by analysing loans outstanding at the district level. They found that the growth of unpaid loans was particularly strong between 2002 and 2007 in areas where the loans were securitised in greater numbers and sold to financial institutions other than banks (investment funds for example). This suggests that banks have take advantage of new buyers’ lack of expertise by selling them fragile loans at a high price. They kept (or sold to other banks) the safest securities. This weakened both the institutions and the entire financial system, causing an underestimation of the systematic risk of non-payment.
Another study confirms this suspicion (Keys et. al 2008). Regulation puts safeguards in the securitisation of the riskiest loans. In particular, each borrower is assigned a score based on verifiable information, and it is recommended not to securitise the loans that have a score under 620 points. This threshold is somewhat arbitrary. Hence, one can compare loans with scores just below and just above this threshold (the latter are twice as likely to be securitised). The authors find that loans just above the threshold are 25% more likely to be still outstanding a few years later (while in principle they are a little better from the start). Banks were much more careful when selecting and monitoring customers for loans for which they held the responsibility than for those loans they thought they could sell easily.
All this suggests that the hesitation of the past few weeks is costly. It allows banks to save as much money as possible for their shareholders, at the expense of taxpayers. Everything indicates that they are able to do so.
Atif Mian and Amir Sufi, 2008, “The Consequences of the Mortage Credit Expansion. Evidence from the U.S. mortgage default Crisis”
Keys, Benjamin J., Mukherjee, Tanmoy K., Seru, Amit and Vig, Vikrant, 2008, “Did Securitization lead to lax screening? Evidence from Subprime Loans”, EFA 2008 Athens Meetings Paper