How to prevent Zombie Lending coming to life

This article considers the problems that governments and regulators face in preventing zombie lending, where insolvent banks continue lending to insolvent borrowers in order to hide losses and “gamble for resurrection”

Banks at risk of becoming insolvent have an incentive to keep lending to their insolvent borrowers, as a means of avoiding the realisation of losses.  This is a form of "gambling for resurrection". We saw an example of this in Spain at the beginning of the financial crisis. The non-performing loan ratios reported by banks at this point were staggeringly low - especially considering a real estate bubble had just burst there. At the same time, figures produced by the Spanish central bank showed that banks in aggregate were actually increasing their lending to real estate developers. “Zombie lending” seems to be a plausible explanation. It is likely that such a misallocation of credit towards zombie borrowers rather than healthy firms would have had bad consequences for economic growth. This might explain, at least partially, Spain’s recent episode of low growth and gargantuan unemployment figures.

This research discusses what a regulator should do about forbearance lending? Solutions to problems like zombie lending and the “gambling for resurrection” problem, which arise when entities with limited liability are close to insolvency, have been discussed in the academic literature at least since the 1970s. The proposed solutions often revolve around forms of restoring solvency, and include many implicit or explicit forms of recapitalisation. A typical form that a recapitalisation scheme for banks might take would be an asset buyback, in which the regulator would set up a “bad bank” that buys bad assets from banks at a certain price. Bad assets are removed from the bank, and, if the price exceeds the fundamental value of the bad asset, the bank is recapitalised. (Note that if the price does not exceed fundamental value, a bank would really have no reason to sell the bad asset in question.)

Unfortunately, while a bank may know exactly how deeply it is in trouble, a regulator that is trying to restore bank solvency might not. In recapitalisation schemes, any bank is likely to try exploit this fact, to maximize the value it can extract from any recapitalisation scheme, at a cost to society at large. In the context of an asset buyback, for example, one might want to offer better prices to banks that are in worse shape, because such banks need a bigger infusion of capital. This is of course tricky if one does not know which banks require these bigger infusions — recent  experience suggests that relying on the honesty of banks might not be a good strategy. If two prices are offered, it is likely that some banks will declare whatever the regulator wants to hear in order to get the better price. There is a question, then, as to how to design recapitalisation schemes to minimise the ability of banks to take advantage of the schemes and extract value from taxpayers, while still solving the zombie lending issue.

In this paper, the authors propose that a reasonable solution might be to use a form of price discrimination. In the context of an asset buyback, the idea would be to charge banks “access fees.” Banks would be required to pay higher access fees in order to obtain better prices on bad assets.  Fundamentally, this is equivalent to what a video rental club might do. The regular price of a video rental might be £2, but there might also be the option to pay £10 per month to become a “member of the video club,” where the price of a rental for members would be £1. Why would the video rental club do this? The video rental club would like to offer lower prices to better customers who rent more videos. If it just charges £1 to anyone who declares to be a good customer, then everyone could declare to be a good customer. The membership fee of £10 is a solution: only people who know they are good customers and rent more than 10 videos per month will choose to pay the monthly membership fee, and subsequently pay the reduced rental fee.

Similarly, one could offer to buy bad assets at a low price of, say, 60% of face value, or an “access fee” of 2% of the last reported total assets of the bank in question in order to obtain a higher price of, say, 70% of face value.  Only the banks in real trouble that know that they have many bad assets to get rid of will opt to pay the fee in order to get the higher price.

These kind of “access fees” can reduce the value that banks can extract from a recapitalisation scheme. The surprising finding here is that it is possible to structure the access fees to completely eliminate the ability of banks to extract value, as we show in the context of a model in the paper.

The ideas proposed here could be used to improve on what has already been seen in practice. Take for example the scheme set up under the umbrella of the National Asset Management Agency in Ireland. Under this scheme, banks received a price for each bad loan that they sold to a “bad bank.” The Irish scheme was heavily criticised, precisely because it was thought that it would generate large windfall gains for some banks. Prominent critics included Joseph Stiglitz, who stated that the Irish scheme was something more fitting of a "banana republic". This paper suggests that charging some form of “access fee” related to the price at which bad loans are bought back or to the quantity of loans transferred has the potential to claw back a substantial part of these windfall gains.

The research paper can be downloaded at the link below.


{Preventing Zombie Lending}{}