How to obtain safe and sound banking?
Traditional "commercial banking" is about taking loans from depositors and making loans. Depositors are promised that their deposits will be returned "at par" with a fixed interest rate. This `assured return' is fraught with problems, because the assets of the bank (i.e. the loans it has given out) inherently face uncertain returns.
India's mutual fund industry has painfully discovered the hazards of making assurances about future portfolio returns. If the assets obtain a lower rate of return as compared with the assurances, the fund manager (mutual fund or bank) faces a problem.
Consider a bank formed of Rs.10 of equity capital, with Rs.90 of borrowings (deposits). The bank has a portfolio of Rs.100 of assets, most of which would be invested in the loans given out. The equity capital serves as a buffer: if the portfolio experiences losses, then the promises made to depositors can be met out of equity capital. If a loss of 10% takes place, then the bank is on the edge of failure. At this point, the liquidation value of the assets is Rs.90, the depositors can be paid off the full sum of Rs.90, the shareholders get nothing, and the bank can be smoothly closed down.
When the liquidation value of the assets of the bank drops below Rs.90, the promise of redeeming their deposits at par is threatened. If depositors suspect that such an outcome is impending, then a run on the bank takes place. Each depositor has an interest in being the first to withdraw his deposits (at par) while the bank is still honouring this commitment. This flight is an efficient strategy for the individual. Suppose a bank has two depositors, who have placed Rs.45 each with the bank. Suppose the assets of a bank are down to Rs.80, so that it is technically insolvent. If the bank were closed down smoothly, each depositor would get Rs.40 each. However, if one depositor manages to pull out his deposit (Rs.45) in advance of closure, then the other only gets Rs.35 upon liquidation.
The Basle Capital Accord is now the dominant set of ideas about how banks should be regulated. A central, and remarkable, insight underlying it was a pro-active approach towards bank closure. The Accord defines a minimum permissible level for the "buffer" of equity capital. Suppose we define the minimum buffer as 10%. Suppose our bank - which started with Rs.10 of equity and Rs.90 of deposits - makes some losses on its portfolio and finds itself with assets of Rs.95. In this event, the Accord requires that regulators proceed as follows:
- The shareholders must first be given an opportunity to bring in Rs.5 of fresh capital, to bring the equity capital back up to 10%.
- If shareholders choose to not bring in fresh equity funds, then regulators should liquidate the bank, raising Rs.95 by selling assets, paying off depositors, and giving the residual Rs.5 to the shareholders.
It is remarkable to observe that the Accord requires closing down banks while they are solvent, as the only way to avoid the pain of closing down banks once they are insolvent. If this policy is well enforced, then the promise to depositors - of redeeming their deposits at par - would always be upheld. In addition, this outcome would be obtained at no cost to taxpayers.
In India, we are some distance away from such a regulatory framework: where banks are pre-emptively closed down while their net worth is still positive. Suppose, instead, the bank is not closed down while it is still solvent. Suppose it goes on to experience further losses. In our example, suppose the bank finds itself with assets worth Rs.75 (a total loss of 25% compared with the starting point of Rs.100). When liquidation takes place, there is a gap of Rs.15 before the depositors can be paid off. At this point, there is a depressing pair of choices which have surfaced in every banking crisis in the world: either depositors have to lose this Rs.15, or taxpayers have to (through a State-financed bail-out).
Suppose we build banking regulation with the orientation and political capability to actually close down banks while they are solvent. The next hurdle encountered is about knowing the true value of assets of a bank. We critically need an accurate assessment of the liquidation value of the assets of a bank. Banks hold a variety of opaque assets: loans, non-traded assets which are accepted as collateral, etc. There is no concept of "a market price" which can be used in marking to market for these assets. In addition, even if a `market price' were known, the market impact cost faced when actually selling the asset is significant and inestimable.
We can now write down the pre-requisites to make all this work. (1) We need a banking regulator which is able to close down banks while they are still solvent, (2) It is somehow possible for the banking regulator to accurately measure the `liquidation value' of the assets of the bank, in real-time, and (3) These assets can actually be speedily sold off for prices which are `close' to these `liquidation values'. Many scholars view commercial banking as inherently unsound, because two to three of these are violated in every country. This is what explains the crisis-ridden history of commercial banking, ever since its inception a few centuries ago.
Is there a way out? There are two paths from which robustness can be rescued. However, each path requires giving up some of the central features of commercial banking today:
- Giving up the promise of redemption at par. If depositors are not promised that their assets will be returned at par, then these problems are greatly simplified. Losses would be passed on to depositors. By avoiding `assured returns', we eliminate the shortfall risk.
- Giving up the concept of investing in opaque assets. Suppose a bank invests purely in highly transparent securities: in assets which are anonymously traded on an exchange, and are highly liquid. In this case, it would be easy to obtain a real-time measure of the portfolio value. Regulators would be able to accurately judge when the bank should be closed down. Closure would be smoothly attained because these highly liquid securities can be sold off on the secondary market.
In summary, to make commercial banking work, we can either eliminate the promise of redemption of deposits at par, or we can eliminate the concept of forming portfolios of illiquid assets. Money market mutual funds are a remarkable institution, for they embrace both these strategies: they do not promise to redeem assets at par, and they only invest in liquid assets. MMMFs are the ideal, robust foundation for a monetary system of the future which is not exposed to the problems of banking.
This vision does not preclude investments in illiquid assets in the economy; it only says that commercial banks forming portfolios of opaque assets is fraught with danger. Other vehicles, e.g. closed-end mutual funds, are available through which investments can go into illiquid assets without either requiring assured returns or involving the payments system.
Back up to Ajay Shah's media page