Banks entered the financial crisis undercapitalized. Consequently, after heavy losses, they needed to raise large sums of capital. Banking as an industry failed. It was kept afloat only through extraordinary and unpopular government support.
Regulators responded by strengthening the quantity and quality of required capital to cushion against future mistakes and accidents. Accidents happen — and more frequently than formerly thought. Consequently, more capital is needed than previously thought. For example, we have experienced two unexpected 40% stock market collapses in the last 10 years alone. It seems difficult to argue against higher capital levels, given what happened during the crisis.
Nonetheless, some banks claim that higher capital requirements would depress economic growth by limiting their ability to extend credit. They assert that every dollar increase in capital is one less dollar working in the economy. They cite current weak loan growth as evidence for their claim. Weak loan growth, however, reflects tepid loan demand — not constrained lending capacity, which is at near record levels. The current recovery in lending based on improved borrower confidence supports this view.
The confusion is due to a mischaracterization of capital. Capital (equity) is the share of a bank's asset value held by shareholders, usually as common stock. It is not cash set aside for a contingency that could otherwise be used for more productive purposes. Many rapidly growing companies are entirely equity financed, illustrating that high equity levels do not inhibit growth. In fact, better capitalized companies, including banks, can grow faster. They also are less likely to be disrupted by adverse events.
The belief that higher capital requirements cause balance sheet contraction is a myth. Banks can increase capital through retained earnings and maintain and expand their balance sheets without issuing new equity. Capital is the cost of risk and not a constraint on viable institutions. Current research shows that higher capital levels, properly phased in, would not compress bank balance sheets or produce any material economic harm. In fact, no historical relationship exists between bank capital and economic activity. Furthermore, operating at low capital levels is a relatively recent development.
Also, the impact of high capital levels on bank costs and profitability is small for two reasons. First, equity is only a small portion of a bank's capital structure and funding costs. Second, higher capital levels reduce the institution's risk. Thus, debt and equity costs decline. Consequently, the effect of additions on the weighted average cost of bank capital is largely dissipated or absorbed. Each 1% capital increase affects bank funding costs only by a few basis points.
Bank objections to higher capital levels are understandable. First, government guarantees, both implicit and explicit, provide a subsidy that gives banks access to inexpensive funds. Second, debt can be less expensive than equity, given the tax deductibility of interest. This is partially offset, however, by the tax advantage of equity at the investor level. Higher capital requirements reduce the value of the subsidies. Thus, it is in bankers' best interest to keep capital levels low to support shareholder returns, management compensation and current business models.
Furthermore, any perceived private costs of higher capital requirements must be compared against the social benefits brought by the larger capital cushions. The reduction of social costs from bank failures and government support achieved from a decline in leverage will probably exceed private costs to banks. Capital merely shifts risk costs from the taxpayer to the capital provider.
Banks need a sustainable, through-the-cycle capital cushion. We now know that 7% is insufficient. There is no perfect solution to avoid banking crises. For example, improved liquidity through match funding and high liquid asset levels should also be considered.
Higher capital levels would at least reduce the likelihood and cost of a crisis. Equally important, they are unlikely to produce material economic harm. Bank capital is clearly not expensive in a social sense. Consequently, the question is not why banks should hold more capital. Rather, it is why they do not.