Increase Bank Dividends? Not So Fast

A debate is developing over the current level of capital at many banks.

Although it was unthinkable a few years ago, some bankers and pundits believe banks have excess capital that should be returned to shareholders through dividends and stock repurchases. Banks have rebuilt capital positions from their depressed crisis levels largely through improved earnings from reduced credit costs. Lackluster returns on equity, limited investment opportunities and crisis-related curtailed payouts have raised the call for the return of perceived excess capital to improve ROE.

Many banks have filed capital plans with their regulators seeking increased shareholder payouts. Once approved, an avalanche of dividend increases is expected. This bears a similarity to early 2007, when the same requests were made by then-overcapitalized banks like Citibank, Bank of America, National City and Wachovia just before the financial crisis. A framework is needed to determine the appropriate level of capital to distribute to avoid repeating the erroneous 2007 dividend increases.

Dividends distribute, but do not create value. Hence, share prices fall once a stock goes ex-dividend. Funds that cannot be profitably reinvested within the institution should be returned to shareholders. This assumes, of course, the institution's capital structure is consistent with the existing and expected operating environment. Many analysts, however, measure the level of capital based on a simplistic static Basel III core tier 1 equity capital ratio to risk-weighted assets. Levels above 10% are deemed excessive, as the 2019 requirement is 7%. This approach is wrong in two respects.

First, the RWA metric, like the discredited value-at-risk measure, is a managed-model-driven number. Risk weights can be manipulated to keep a substantial portion of assets, up to 40% at some domestic institutions, off the regulatory balance sheet. This overstates the level of capital strength. Next, the 10% standard used to determine excess capital is questionable even if measured correctly.

Banking is undergoing deep structural changes. These changes have increased the risk banks face. Consequently, greater capital levels than previously deemed adequate are required. The great moderation is over. Business cycles are shorter and earnings swings are more violent. The higher capital levels seem wasteful only if nothing unusual happens. The same can be said of having a wasteful fire insurance policy just because your house did not burn down. As Nassim N. Taleb notes in his book "Antifragile," something unusual such as, the Great Recession or the euro crisis, usually happens.

Additionally, the opaque nature of bank financial statements makes it nearly impossible to assess the reliability of capital to offset potential asset quality problems. Even managers have a difficult time assessing this risk, as was demonstrated by the 2012 London Whale losses at JPMorgan Chase.

In this unstable new normal environment, equity levels of 15% to 20% of total assets, not simply RWA, are needed. During the crisis, bank asset losses exceeded 8%, according to the International Monetary Fund. Another 8% or more is needed to keep banks operating and extending credit during such an event to avoid emergency governmental support. Rating agencies also indicate 15% capital levels are needed to be deemed strongly capitalized.

Some argue such capital levels unduly depress ROE. True, but value is not created by high ROE. Rather, it is created from the spread between ROE and an institution's cost of equity. Higher capital levels reduce the cost of equity. Others allege this exceeds regulatory requirements. Keep in mind, regulators set a floor, not a ceiling, for capital requirements. Each institution needs to establish its own capital structure requirements consistent with its strategy and expected future market conditions – not just the current market.