Global Banks Are Undercapitalized. Is that So Wrong?

After the 2008 financial crisis, I became obsessed with bank capital.

So much, in fact, that I have spent a considerable amount of time lately conducting empirical analysis on exactly how much capital regulators demanded banks hold and how much capital they eventually funded themselves with. Everyone was talking and theorizing about the topic, so I just took a look at the actual data, principally for large American and European banks.

And for those who oppose high leverage, it's not a pretty picture.

The analysis shows that the Basel guidebook yielded a "zero capital" regime, with big banks being allowed to trot along essentially noncapitalized for over two decades and with many firms only too eager to oblige. Basel minimum capital charges have always been low (5% to 6% of on-balance-sheet assets, typical of U.S. banks) or very low (2% to 3% of on-balance-sheet assets, typical of European banks), increasingly so through the years. This is explained by low, or very low, Risk Weighted Assets numbers as a proportion of total real assets, increasingly so through the years. While banks have always maintained very high regulatory ratios (12% to 20% total capital to RWA), they were in truth very or extremely leveraged (the amount of actual capital held remained small or very small throughout – 2.5% to 4.5% average equity to assets in Europe, 5.5% to 7.5% average Tier 1 to assets for U.S. commercial banks, 3% to 4% average equity to assets for U.S. investment banks). Regulatory and actual capital ratios went up substantially after 2008, but leverage remained high or very high for the most part.

Given the evidence, it may seem difficult to understand why the Basel system was allowed to reign for so long. The capital requirements it churned out were so obviously low and so obviously declining (and even lower if we included off-balance-sheet positions in the assets number). How could policymakers miss the unavoidable signals? Even though banks did in the end hold more capital than the minimum, that number was for the most part still quite small. Even as new formal rules have been introduced to increase capital requirements, the calculation machinery continues to produce RWA densities that are lower than precrisis levels. Have regulators been neglectful, prone to inaction, or they just did not care?

How about banks? Why did they happily trot along undercapitalized? After all, the less quality capital, the closer to insolvency you are. There are two possible reasons for banks' acceptance of high leverage:

• They wanted it: Lots of leverage can lead to very high returns on equity, a key success and compensation metric; raising new equity can lead to dilution of existing shareholders with potentially serious consequences for bank management; debt funding has the huge advantage in most places of being tax-deductible.
• They had no other choice: Maybe it was or is impossible for banks to raise more equity than they have, perhaps due to a lack of investor demand much beyond preexisting levels; equity is assumed to be much more expensive than debt and this would be particularly true for banks (debt enjoys deposit insurance and implicit government guarantees). These two points should be the focus: could banks really raise lots more equity even if they truly wanted to, and would the cost of that equity really not be prohibitively expensive?

What if there is a limit to the amount of capital banks can raise? What if it´s simply not possible to attract yet another sovereign wealth fund or institutional asset manager to a large new issue? What if retaining extra earnings is insufficient or infeasible? In that case, higher regulatory capital requirements like those imposed by Basel III would have to be satisfied in two ways: "RWA optimization," where the quant team tries to come up with yet lower risk estimates for the bank´s assets (lower RWA numbers lead to a higher regulatory ratio even if capital remains the same), or balance sheet shrinkage. The latter route sounds worrisome in principle, as it would be limiting the bank´s activities, including potentially big drops in lending and market-making.
By looking at balance sheet equity, one could argue that banks already raise quite a lot of capital, nominally at least. A list of the 50 U.S. and European banks and nonfinancial companies with the most balance sheet equity in nominal terms at the end of 2012 included around 25 banks. Four of the top five were banks: Bank of America (BAC) at No. 1 with $237 billion in equity; JPMorgan Chase (JPM) in third place with $204 billion, and then Citigroup (NYSE: C) and HSBC (HBC) with $191 billion and $175 billion respectively. Gazprom, the Russian gas company was the only nonbank to make the top five, with $224 billion of equity.

The top 10 included five banks – the four mentioned above plus Wells Fargo (WFC). The top 20 had 10 banks (add BNP Paribas, Royal Bank of Scotland, Santander, Barclays, and Unicredit). The top 30 U.S. and European banks by balance sheet equity held roughly as much equity as their 30 nonfinancial equivalents. These rankings are bound to be less than perfectly exact, given currency rate fluctuations and lack of data in a few cases, but the overall picture would not change.

If we consider balance sheet equity to be a good indication of the ability of a firm to raise equity, then we must conclude that the big banks are at least as capable of raising core capital as the largest, most successful, most profitable corporations in the world. The fact is that big banks hold as much, if not more, equity as the Exxon-Mobils, the Gazproms, the Apples, Volkswagens, and Chevrons of this world. No other single industry, in essence, holds as much balance sheet equity as international banks. If we assume that there is a fixed supply of equity capital available worldwide (only so many investors, only so much profits), then banks would already capture more of it than any other sector.

Of course, given that banks have also proven successful and willing to issue lots of debt, those amounts of equity are a much smaller portion of total assets than in the case of the nonbanks (where equity-to-assets ratios of 40% to 50% are quite common). Banks may have tons of core capital, but they are infinitely more exposed.

Yet the point remains that when it comes to equity, banks hold just as much as the healthiest of corporations. Could they raise even more? That is the vexing issue. And given their current, comparatively bloated, equity bases, it may be hard to enhance them much further. Perhaps the overall global supply of capital available is already at full use, or close to it.

If banks, for whatever reasons, can´t raise lots of extra capital, the only way they can maintain large balance sheets is via large debt funding, which is only possible if regulators allow and condone lots of leverage. The Basel system has sanctioned balance sheets on the verge of insolvency, but at the same time it may have permitted much more lending and much more trading than otherwise might have occurred.

If we assume that those two things can be a huge positive, and if the only way that they can happen on the required scale is by allowing 30-1, 40-1 or 50-to-1 bank leverage, and if the positives somehow compensate for the monster negatives of a likely toxic leverage-induced crisis, then the zero capital regime fully in place since 1996 may, rather paradoxically, have been unavoidable, natural, and even desirable.

Pablo Triana teaches at ESADE Business School in Barcelona.