Monday, August 31, 2009

Rethinking Deposit Insurance

We have become accustomed to thinking that federal insurance of 100% of deposits in US chartered banks (up to some amount) is an unmitigated benefit of the legacy of New Deal financial reform. In the current crisis, it has been observed that, following the bankruptcy of Lehman Brothers, financial meltdown was triggered by a collapse of the repo market, which is equivalent to a run on the ‘shadow banking system’ of investment banks that lie outside the FDIC insurance umbrella. Thus have many commentators opined that the umbrella ought to be extended to the short term liabilities of those institutions.

Indeed, the existence of deposit insurance has mitigated financial dis-order; there have been no runs on US banks. But the crisis has revealed problems engendered by complete insurance of deposits. A much analyzed one is the perverse incentive on shareholders, debt-holders and mangers to take large leveraged risks when the cost of capital to do so is subsidized via deposit insurance (even in the absence of an implicit promise of government bailout). I wish to shed light on another problem: That deposit insurance acts as a barrier to entry to new banking competition.

Non-price competition and the perpetuation of failure

When deposits are fully insured, a new entrant cannot attract depositors by virtue of being a better capitalized –lower risk –investment than an incumbent bank. From the depositors’ standpoint, choice is based on convenience and deposit interest rate. The incumbent will win on convenience, as it has the branches and the sunk cost of opening new accounts has been incurred. The new bank is thus left to compete on price alone. As it competes in the same loan market as does the incumbent bank, it will have little ability to raise its lending rates above those of the incumbent group, and so it will have little room to raise the rate it pays depositors above incumbent levels. Yet it needs to offer higher rates in order to attract deposits away from incumbents.

That is one reason why the US banking industry has experienced a low rate of entry, even after the elimination of deposit interest ceilings. It is also the primary reason many have called for banks to be regulated like utilities –to limit the potential taxpayer losses incurred by insuring deposits of banks that take reckless risks. It is the source of the dilemma of how to design rules for a safe and innovative financial system.

Clogged pipes

Today we face the problem of re-starting moribund bank lending. The Fed and Treasury have pumped unprecedented liquidity and capital into banks, but the banks have not leant it out. They have not because they cannot. In our debt-deflationary meltdown, banks find themselves weighted down with ever declining asset value and the need to set aside ever increasing amounts of capital to cover losses. In this environment, they must hoard what capital they have. Bank capital is ‘locked’ into non-performing legacy loans, and savings are not funding new, productive investments. This is the situation Japan found itself in during its ‘lost decade’ of the 1990’s.

The banking pipes are clogged. The Treasury had tried for 3 years to unclog the system through a succession of programs to remove ‘toxic’ assets from banks’ balance sheets. None have worked because the realization of even greater losses banks would have to reveal in arms-length transactions have caused banks to balk. Meanwhile, attempts to unclog the pipes by forcing debt for equity swaps or granting the government expanded resolution authority that would allow it to take immediate control of bank holding companies and their derivative books (and so avert panic) have foundered on political gridlock.

Unclogging the pipes

If new capital could flow into banks unencumbered by toxic loans, the pipes could be cleared, since these new banks would not have to divert capital into covering loan losses. But for this to work, new banks would have to be able to attract deposits. That cannot happen (very quickly) under the current deposit insurance system.

If a way were found to enable new banks to effectively compete for deposits on the basis of their superior capitalization, political gridlock could be circumvented. An essentially bankrupt bank like Citicorp would presumably bleed deposits and be forced to liquidate or engage in a debt-for-equity swap recapitalization. This would not take an act of Congress or a decision to intervene by the Fed or Treasury; it would occur by the normal operation of market forces.

Moreover, there would be no net loss of deposit capitalization of the banking system –deposits would merely be shifted into new banks. The ‘money multiplier’ would actually increase, as new banks would have a lower propensity to hoard and lesser requirements for loss provision. A meltdown of Citicorp would not mean the financial system was melting down.

An essential distinction

Our deposit insurance was designed to prevent bank runs triggered by a systemic panic, which would cause a financial meltdown. That remains a valid concern, as evidenced by the damage inflicted by the recent panic induced run on uninsured shadow banks and the implosion of the money multiplier. But there is nothing in this logic to suggest that shifting of deposits between institutions in response to market perceptions of differentials in the idiosyncratic prospects and soundness of individual banks would harm the financial system. On the contrary, a competitive market in deposits would improve bank performance by allowing efficient banks to win market share from inefficient banks.

We thus can make a distinction between runs on individual banks due to the normal operations of a competitive market and those induced by systemic panic. Government only needs to intervene in the event of the latter.

In his recent co-authored paper delivered at the 2009 Jackson Hole Fed Conference, MIT economist Ricardo Caballero described the core problem of financial crisis as “a significant surprise, which suddenly changes, at least temporarily, the perceived rules of the game. All of a sudden it is no longer enough for economic agents and financial institutions to understand their local environment since, as we have seen in the current crisis, systemic events can seep through unexpected and distant linkages. When this happens the fundamental shock is compounded many times by panic”. In this atmosphere, where calculable risk gives way to Knightian uncertainty, agents withdraw en masse into safe assets, which “triggers asset fire sales and activates financial multipliers that cause enormous damage to balance sheets and credit markets”. According to Caballero “the main antidote to fear is government backed insurance against what investors fear”.

The proposal

Caballero has proposed an insurance scheme for bank assets in which the monetary authorities could ‘flip a switch’ and activate coverage when they detect a systemic crisis, but otherwise allow the market to function without interference. This finely balances the rationale for intervention to prevent the systemic financial meltdowns which capitalist credit economies are prone to, with the imperative to refrain from impeding competition and innovation in financial markets at all other times . It is, according to Caballero “directly targeted at offsetting the damaging effect of uncertainty-spikes on balance sheets and credit markets”.

Caballero points out that his scheme could be applied to bank equity and the liability side of banks balance sheets, but he does not address deposit insurance, perhaps because he takes a given that deposits are fully insured at all times.

My proposal is this: Eliminate, or reduce the extent of coverage to some percentage of bank deposits, of FDIC deposit insurance in ‘normal’ times. This will allow market forces to operate more effectively in weeding out badly performing banks and shifting the flow of savings into more productive channels. But empower the monetary authority to ‘flip a switch’ to temporarily fully insure deposits during periods of systemic panic.

Resolving our current malaise

We are past the period of systemic panic, but our banking system is not performing its function of channeling savings into investment and working capital. This may impede the nascent ‘inventory restocking’ recovery everyone seems to be predicting. If firms, particularly those too small to access the bond market, are not able to secure loans and lines from banks, they will be unable to restock inventory.

If the monetary authority were able to ‘turn off the switch’ and withdraw deposit insurance (or even a small portion of deposit insurance) it would probably unleash a wave of new banks, unencumbered by toxic assets, taking market share from incumbents and financing new investment. These market forces would ‘resolve’ our banking problems by forcing incumbents to restructure or liquidate and it would engender a market to finance the acquisition of liquidated bank assets (thereby reducing the losses).

The main problem I detect with this approach is a potential relapse of a Lehman type panic if the monetary authority doesn’t have resolution authority that covers the derivatives books of bank holding companies (for those banks that are central counter-parties).

No comments:

Post a Comment