Legislators in Western countries are in the process of rewriting financial sector regulations. Most commentators encourage the politicians in this mission. Yet few politicians or commentators pause to ask themselves why such regulations are required in the first place – that is, why market mechanisms alone do not properly constrain banks’ risk taking.
The answer is that government guarantees of bank deposits greatly reduce the “risk premium” that banks must pay on their debt capital. Retail deposits (i.e. deposits under some threshold) enjoy the explicit guarantee of the government regardless of the bank’s credit worthiness. Only bank creditors who are not guaranteed by the government (e.g. large commercial depositors and other “wholesale” liability holders) will demand higher interest rates when banks take more risk. And even they will do so only to the extent that they doubt the government will cover their losses should their debtor bank fail.
The history of bank bailouts – especially for the “too big to fail” institutions – vindicates the view that debt holders will be made whole, and therefore justifies the low levels of “bailout doubt” observable in the prices wholesale bank creditors charge banks. Since 1988, 28 of the world’s largest 100 financial institutions (as measured by assets) have failed. This equates to a 1.3% annual probability of default, which should equate to a BB credit rating. In fact, the top 100 banks have enjoyed an average credit rating of A+, which equates to a 0.05% annual probability of default. This apparent anomaly is easily explained by the fact that in only two of these 28 cases of bank failure (Lehman Brothers and Washington Mutual) did the national government allow creditors to suffer losses.
These explicit and implicit government guarantees of bank creditors effectively subsidize bank risk-taking. They remove the normal market discipline on risk-taking: namely, the premium that creditors charge for risk. Given this subsidy, banks can extend their risk taking, and the revenues that come with it, well beyond the point where an unsubsidized risk premium would have stopped them.
Excessive bank risk taking is simply another example of the general principle that subsidies create excess supply. And, as usual, to avoid a wasteful misallocation of resources, governments must then try to rein in the overproduction created by their subsidies. In agriculture, this is typically attempted by imposing production quotas or “caps” on the crops for which the government guarantees above-market prices. In banking, the standard cap used is on leverage. Banks are legally required to fund no less than a specified percentage of their assets with equity capital rather than debt. Since the advent of the Basel Accord on Banking Supervision in 1988, this minimum capital requirement has been expressed as a percentage of “risk weighted” assets.
This approach, of trying to regulate away the excess risk taking encouraged by debt-holder guarantees, failed catastrophically in the crisis of 2007-08. The standard response of policy makers and advisers has been that the approach is right in principle and that the catastrophe was caused by nothing but faulty execution. The regulations need only be corrected. Unlike the Basel 1 and Basel 2 accords, whose measures of bank risk taking were hopelessly inadequate, the post-crisis Basel 3 regulations will get things right.
This is difficult to believe. Bankers will seek out activities whose risks are overlooked or mispriced by the new regulatory regime. In other words, they will continue to engage in the “regulatory arbitrage” that occurred on a massive scale under the Basel 1 and Basel 2 regimes. It is mere wishful thinking to believe that it will not also occur under Basel 3.
“More active supervision” is the mantra that encourages many to indulge in this childish optimism. No matter how “active”, supervisors cannot be expected to win this game of cat and mouse. For the mice have too many advantages. They are more numerous, mostly smarter, better informed and, given the complexity of modern financial products and the global scope of the financial markets, they have a near limitless supply of nooks and crannies in which to hide their risks. What’s more, bankers are far better incentivised to win the game than government employees are.
This inevitable regulatory arbitrage is disastrous, not simply because it prevents regulations from having their intended effect of reducing aggregate bank sector risk-taking, but because it distorts the allocation of capital in the economy. Debt capital is allocated not to those uses that deliver the best return on risk; it goes to the uses that deliver the best return on the bank equity capital required by the government’s rules. In short, bank capital rules undermine the market mechanism that ensures resources go to their most valuable use.
Nor will the politicization of lending be restricted to such unintended effects. The government’s role in deciding the cost of lending to various classes of borrowers (through the risk-weightings and other rules) provides a strong incentive to lobby government officials for preferential treatment. The borrowers who seek discounts, the banks who serve them and the politicians who wish to be seen as the champions of these groups all have an incentive to distort the framework and thereby add to the misallocation of capital. Such shenanigans have already been incorporated into the 2010 Dodd-Frank Bill, which requires banks to hold 5% of loan securitizations on their own books, except when the securitized loans are “qualified mortgages” (which are yet to be specified).
The perverse result of tighter risk regulation will be more risk, as bankers seek ever more ingenious ways of evading the rules, and lower economic growth, as the rules distort the allocation of capital in the economy.
Government guarantees of bank debt-holders and risk regulations are a package deal; the former demands the latter. But the package is calamitous. The regulations cannot do their intended job of eliminating the “moral hazard” created by the guarantees, and the failed attempt merely perverts capital allocation. The guarantees are the problem. The solution is to eliminate them and the attendant regulations that prevent market participants from pricing bank risks properly.