Click here for a printer-friendly version of this article.
On Friday, the Federal Reserve announced the completion of the second round of stress tests for the 19 largest bank holding companies. Unlike the 2009 stress tests, the Fed released no information about individual banks’ performance, but the results were obvious enough given the immediate announcements by J.P. Morgan, Wells Fargo, US Bancorp, and BB&T that they intend to massively increase cash distributions (dividends). More banks are anticipated to increase dividends now or in the near future. Goldman Sachs has redeemed its preferred stock, effectively replacing mezzanine capital with incremental debt. The stress test has therefore served to allow the largest financial institutions to re-leverage, increasing the probability of financial distress and a repeat of the disastrous autumn of 2008.
Others have taken a different view. They trumpet the Fed’s decision as welcome evidence that the financial crisis is officially behind us. Analysts called the existing capitalization requirement a “mark of Cain,” and point to Fed research showing that Tier 1 common capital increased by $275 billion at the 19 largest institutions between 4Q 2008 and the end of 2010. All together, the Tier 1 common ratio increased to over 9% at the end of 2010, two-thirds above where it stood at the end of 2008. Measured relative to pre-crisis leverage ratios, banks look overcapitalized and able to increase dividends.
But are pre-crisis leverage ratios really the appropriate frame of reference? One would hope banks are significantly more capitalized than they were in 2008, as low levels of equity contributed directly to the crisis in two ways. First, and most obviously, the amount of equity was not sufficient to offset the low probability of insolvency. Equity is the residual claim once expected losses, debt, and other liabilities are deducted from asset values. It therefore represents the amount of losses a financial institution could withstand before it would have to seek bankruptcy protection from creditors (or become resolved by regulators). The larger the equity cushion, the larger and more improbable losses would have to be to exceed it. Judging by the systemic character of the crisis of 2008, banks in general held too little equity because insolvency was viewed as a virtual eventuality for nearly all large financial institutions in October 2008.
The second and less appreciated way low levels of equity contributed to the crisis was the way they encouraged “risk shifting.” This phenomenon, also called “asset substitution,” is the process by which managers transfer wealth from creditors to shareholders by assuming greater risk. Equity is a right to all cash flows generated by a firm after creditors and other claimants have been paid. If the firm cannot meet its promised payments to creditors, the equity becomes worthless. This is true if the firm falls $1 or $1 billion short – since corporate equity enjoys limited liability, all losses beyond the zero bound accrue to creditors. As a result, shareholders disproportionately benefit from riskier investments.
To see why this is so, consider a bank manager seeking to borrow $91 million to finance one of two $100 million investments, A or B (this corresponds to today’s 9% equity ratio). Investment A offers the potential to double the money in good times, but would result in 60% losses in bad times. Alternatively, investment B offers a 50% return in exchange for the risk of a 10% loss. Although both investments have a present value of $120 million (assuming good or bad times are equally likely), most people would be intuitively drawn to investment B. It offers a much better risk-return tradeoff, with 1.67% of expected return for every 1% of risk, compared to only 1.25% for investment A.
But, as shown below, it’s actually investment A that a manager acting in the interests of shareholders would probably pursue. In good times, debt-holders receive only a return of the face value of their investment. In exchange, they receive the assets of the bank in bad times (say, bankruptcy) to help offset losses. Thus, in good times the payoff to debt-holders would be $91 for both investments; in bad times investment A would leave debt-holders with $40 million while investment B would deliver $90 million. This means that the present value debt claim to asset A is worth $25 million less than the claim to asset B even though the two investments have the same market value. Where did the $25 million go?\
It was transferred to equity-holders. In bad times, the equity of both A and B is worthless. But in good times, the equity in investment A is worth $109 million, while investment B provides just $59 million. Investment A’s $50 million of added downside risk is of no consequence to the equity-holder because he’s not the one bearing it.
The only way the equity-holder in this example would prefer asset B is if he was forced to finance $60 million of the investment himself, as shown below. Only at this point would the losses in bad times exert enough of a drag to equal the potential upside gains of leverage. Higher capitalization levels are required in this example simply to get the equity-holder to behave as he would if he were financing the investment entirely with his own money.
This is the key point: the asset substitution introduced by leverage causes rational actors to take on more risk than they would if they were paying for the investment themselves. More lenient capitalization requirements would strengthen this incentive. The result will be not just a greater risk of insolvency because of less equity, but also greater risk-taking generally
This is of great consequence to policymakers when the institutions taking increased risk are the country’s largest banks. The social cost of risk is contextual. As we saw in 2008, the managers of large banks do not shift risk onto creditors, but taxpayers instead. Rather than attacking this problem by maintaining capital surcharges on the largest bank holding companies, the Fed [with it’s latest decision to allow banks to restart dividends] has essentially embraced the old state of affairs. This makes no sense and simply sets the stage for bank managers to assume more risk in the early stages of the new cycle.
The Fed decision follows the missteps of the Dodd-Frank regulatory reform bill, which imposed excessive burdens on much of the financial system without fundamentally addressing the undercapitalized status of the largest banks. Amazingly, the bill creates a $50 billion threshold for classification as a “systemic risk,” as though Huntington Bancshares of Columbus, Ohio or similarly situated banks are deserving of the same degree of attention as $2 trillion institutions like Citigroup. Yet, the costs of these and other provisions of Dodd-Frank are actually borne by thousands of smaller institutions whose failure would do nothing to undermine the payments system.
In a letter to shareholders, M&T Bank President Robert Wilmers expressed concerns that conditions are ripe for another financial crisis for this reason. “To categorize (the largest financial institutions) as of the same species as traditional commercial banks is akin to describing dinosaurs as simple reptiles – it is true but profoundly misleading,” he wrote. Much of the support for senseless regulations on interchange fees, for example, is based on a desire to extract concessions from the “big TARP banks.” But if the target were the largest banks, why were regional banks, hedge funds, and asset managers hit with a complex new regulatory regime? As argued by finance professor Douglas Cumming, it was not “the fact that [these other institutions] caused the crisis, but rather, since there is a crisis it affords an opportunity to regulate more heavily.”
The Fed’s decision is hugely unfortunate and let there be no doubt that the next crisis is nearer as a result. Let us hope that in the wave of rules and regulations to follow that one will at least address the nature of the ongoing undercapitalization (and thus, leverage) problem.