In a new policy note, Jan Kregel draws out some of the policy lessons of the Cypriot deposit tax episode for plans to create a system of EU-wide deposit insurance. In addition to the necessity of a strong central bank (the ECB in this case) standing behind the deposit insurance scheme (which does not appear to be part of the current plans), Jan Kregel explains why a certain amount of moral hazard is inescapable.
We can see this by looking at two types of deposits that correspond to the dual functions of banks: deposits of currency and coin, and deposits created when loans are made. If a bank makes bad loans — and as Kregel points out, “it is the failure of the holder of the second type of deposit [loan-created deposits] to redeem its liability that is the major cause of bank failure” — the first type of depositor (of currency and coin) should not bear the brunt of these bad decisions. The role of deposit insurance, one might argue, is to provide such protection.
But since deposit insurance has to be extended to all of a banks’ deposits (up to a certain level), including those created by loans, moral hazard is inevitable. Ideally, deposit insurance would be structured in such a way as to distinguish between deposits based on currency and coin and deposits generated through loans, as well as between deposits created by good and bad loans (loans that default). But if you examine how the banking system functions, says Kregel, this isn’t operationally possible:
Unfortunately, it is impossible in practice to make these distinctions between reserve deposits, defaulted-loan-created deposits, and deposits created by loans that are current. It is for this reason that there are limits on the size of insured deposits based on the presumption that the first type of deposits will be relatively small household deposits created by the transfer of reserves and used as means of payment or store of value. It thus limits coverage of the other types of deposits. However, this is clearly inequitable for the deposits held by borrowers who are still current on their loans.
As he explains, with some help from Minsky, it is the means of payment function that makes these distinctions practically impossible (for more on why this is, read the note here). Kregel concludes that “It would thus seem impossible to design a truly fair deposit insurance scheme that eliminates the inherent moral hazard and the necessity of a contingent guarantee of the central bank.”