Open Bank Resolution: Open to Non-Government Alternatives?

The Reserve Bank of New Zealand is pushing to enhance the bank creditor recapitalisation option for dealing with bank failures under a new term: Open Bank Resolution. To make this option more palatable, and thereby more credible, registered banks will soon be required to pre-position their systems so that they can screw their creditors while remaining open for business (or more correctly, to be able to partly screw their creditors and to be able to re-open the next banking day). So far so good. Well it's fantastic really, and much better than the bank bail-out practices we've seen displayed in Europe and America in the last few years.

However, unlike the covered bonds consultation paper, a full range of options does not appear to have been considered. The bank failure laws and the Open Bank Resolution policy involve a high amount of government/RBNZ involvement such as:

1. RBNZ supervision of registered banks
2. RBNZ powers to give directions to registered banks, with Ministerial consent (including for failure management purposes)
3. Ministerial and government involvement in appointing a statutory manager to a failed or failing registered bank
4. Government guarantees given to failed banks to guarantee the portion of deposits not haircut.
5. RBNZ liquidity facilities provided to the failed bank
   
6. Government involvement in selecting a resolution, including any taxpayer socialisation of losses.

As an example of an alternative regulatory policy, Non-Bank Deposit Takers are supervised by security trustees, who have powers to deal with the NBDT in the event of failure, and could, if the trust deed so allowed, effect a creditor recapitalisation and/or a trading receivership that would continue access to the payments system and other facilities for the institution's creditors and debtors. This approach to bank regulation, supervision and failure management does not appear to have been considered.

This post looks at this alternative in some detail, considering the issues presented by a bank failure, and the options for resolving them without government being directly involved in supervising banks or failure management. I argue that the further the government is from dealing with bank supervision and failure, the less the chance of government funds being used to socialise bank losses.

There appears to be three issues emerging from a bank failure:

1. Loss of principal
2. Loss of liquidity, and
3. Loss of access the payments system.

These points are accepted by the RBNZ document in point 14:

What is required is an option which not only places the cost of a failure in the first instance on shareholders, but also provides flexibility to assign losses to creditors, without causing unnecessary disruptions to the payments system and bank customers’ access to liquidity

However, elsewhere, and in the policy considers as a whole, The Open Bank Resolution policy appears to conflate 2. and 3. :

a solution is sought that solves the urgent liquidity problems associated with a bank failure, but does not force all of the losses to be borne by the government. (point 16)

Loss of access to the payments system can be a significant part of the cost of a bank failure. Consider, for example, that your wages are paid into your bank account, and from your bank account your automatic payments are set up, and your direct debits, and your EFTPOS card. In the event that your bank failed, and you had, say $500 in the account, the loss of, say, $200 out of the $500, would hurt you by $200, and the loss of access to the other $300 for, say, 6-12 months, might hurt you by $10, but the cost of trying to re-set up all your wage payments, automatic payments, direct debits and EFTPOS facilities might be such a pain in the arse as to be worth $100 or $200 to you (especially when hundreds of thousands of other customers are trying to do the same thing at the same time resulting in large queues). In fact being without EFTPOS access even for a day could leave you stranded without fuel for your car or the means to buy food or other urgent small purchases, an experience that could cost you a day's wages ($150) or similar loss.

The RBNZ does not appear to have seriously considered that the cost of a total loss of liquidity might be tolerable so long as the bank accounts in question continue to function for receiving new payments and making payments from these new funds.

There appear to be two ways of keeping a bank 'open' after its failure:

1. Rapidly recapitalising the bank with creditor funds, i.e. a proportional and part conversion of debt to equity. This could be in the form of tranches of preferred shares in the failed bank, or by way of diluting or cancelling existing ordinary shareholders (and subordinated creditors) and converting creditors claims partly to ordinary shares. This could (or should) also involve appointment of new directors by or on behalf of the new shareholders. Under this approach, the bank's insolvency is rapidly remedied, and the bank, as an entity, is saved.
   
2. Leaving creditor claims intact, but selectively subordinating and freezing part of them, while leaving the balance as good. Under this approach the bank remains insolvent but can keep trading because it has enough creditors claims frozen and available to absorb losses. Eventually the bank can be liquidated, and its business either closed or sold.
   

The Open Bank Resolution adopts the second approach -- the 'bank creditor recapitalisation' is not a literal recapitalisation but a selective default that allows ongoing trading:

The haircut represents that portion of the accounts that is effectively frozen, but is not legally cancelled, i.e. they are still obligations of the bank, but the bank has no capacity to repay them. (42 (f))

So, how could the Open Bank Resolution policy work without government supervision, failure management, guarantees and so forth?

1. Both registered banks and Non-Bank Deposit Takers should register a prospectus, execute a trust deed and appoint a security trustee to supervise them, rather than the RBNZ being the supervisor of registered banks but not NBDTs. This would provide a more consistent and more private treatment of financial institution supervision, in line with the Securities Act.
2. Registered banks would incorporate a General Security Agreement into their trust deeds, providing a security interest over all the present and after acquired property of the bank, to be held by the trustee for the benefit of all creditors generally. This means that the trustee, and any receiver appointed by the trustee, would have the power to failure-manage the bank.
3. In the event of a bank failure, the trustee would appoint a receiver. The receiver would be the agent of the bank, pusuant to the trust deed, and have the power to carry on the bank's business, including accepting new deposits and (part) paying creditors (this is actually nothing new or unusual about receiverships).
   
4. New deposits would rank ahead of creditor claims existing at the point of failure (it is standard in finance company trust deeds for receivers to have the power to issue such securities ranking ahead of existing creditors (e.g. see 20.5 (g) of South Canterbury Finance's trust deed). This means that creditors can still have confidence in the failed bank with any new funds, while not having any confidence in its capacity to repay balances at the time of failure (as was the case with the engineered Australian large bank failures of the 1890s, although the process here is intended to take place within 1 business day rather than weeks or months as in that episode).
5. The receiver can make a portion of old funds available to depositors by either (a) providing over-draft facilities secured over balances existing at the time of failure (e.g. the receiver could grant all customers of the bank an overdraft facility equal to, say 50% of their balances as at the time of failure), or (b) converting a portion of the balances as at the time of failure into a security ranking prior to the rest of the balances, and ranking equally with new deposits. Option (a) can encourage existing customers to retain their balances with the bank by way of charging a fairly high interest rate on such overdrafts so that only those who need the liquidity will withdraw the funds.
6. Another alternative to providing liquidity for bank customers is to securitise the funds at the point of failure and list them on a securities exchange. Customers with large balances can get involved in liquidating their holdings on the market, while customers with smaller balances can either rely on the receiver using the above options, or the receiver can sell a modest quantity of the securities on the market during the first week or two of the listing and settle large numbers of customers with smaller balances cost efficiently (the trust deed would need to authorise this).
   
7. The receiver can then consider a wide range of resolution options to the failed bank's situation including (a) a debt for equity swap, to be voted on by relevant investors according to the provisions of the trust deed, (b) selling the bank as a going concern to new owners and paying the proceeds to creditors by rank or (c) winding down the business of the bank and progressively paying out creditors over time.
8. Another approach is to empower the trustee to effect a rapid creditor recapitalisation at the point of failure without going through a receivership. (Receivership would still be possible, but the trustee could select recapitalisation as an alternative resolution.) The trustee would be issued with warrants that would enable it to require issue of large amounts of new ordinary shares, and to appoint new directors in respect of this new super-majority of shareholding in the bank (existing shareholders could be either be diluted or extinguished, depending on the set up).
9. Pre-positioning requirements could be regulated by the Securities Commission and/or the RBNZ to enable or prescribe particular trust deed provisions and/or operational capability to be in place to execute such restructurings/trading receiverships. The same would apply to regulation of payment and settlement systems for the purpose of minimising disruption in the event of bank failures. Likewise for out-sourcing policies and other regulations now in place to facilitate statutory management and Open Bank Resolutions.
   
10. Securities and/or banking laws would need to be reviewed to enable disclosure and other requirements to be met or waived throughout a trading receivership or restructuring of a continuous issuer.