Risk Is Best Managed From the Bottom Up

From Five Ways of Looking at Risk.

Federal Reserve Chair Ben Bernanke stated the obvious during his March 10 speech at the Council on Foreign Relations in Washington: The global financial system contains too much risk and too little regulation. As the central figure in a multitrillion-dollar bailout exercise that has done little to contain, let alone reverse, the current economic crisis, Bernanke then strained to make some risk-fighting suggestions for future stabilization.

First, he recommended regulating "the financial system as a whole … not just its individual components." Second, he warned that classifying any firm as "too big to fail" only "encourages excessive risk-taking by the firm." Third, he stressed the need to help credit flow by adopting "new policies to limit the incidence and impact of systemic risk." All of these are efforts to contain risk from the top and to spot big bubbles before they get out of control. The problem is that they overlook all the layers of risk created from the bottom, before they are visible in the financial system as a whole. We don't just need to prevent big problems at the top. We need regulations to address risk in every layer of the system, from the loan or bond, to the bank, to the very structure of the global financial industry.

At the bottom of the risk pyramid lie the mortgage loans. Financial models select and package them into new securities according to criteria such as diversity of geography, size of loans, and length of mortgages in years. Standard securitization models make numerous assumptions before repackaging a bunch of loans into an asset-backed security. These include the likelihood that a certain number of loans will default or be delinquent with payments. The results the models spit out are only as good as the information put into them. In the case of the mortgage crisis, the input relied too heavily on recent history, of which there was little for sub-prime loans. This introduced a risk that wasn't taken into account.

Meanwhile, under pressure from investment banks and their investors, lenders pushed loans onto borrowers. Even when loans deteriorated and delinquencies rose, demand persisted, which caused lenders to make even riskier loans. From pension funds to university endowments to Icelandic municipalities, institutions everywhere relied on high-grade securities manufactured from the base of low-quality U.S. mortgages.

Even if the models had detected possible losses better, things still would have gotten ugly because the market obsessively reused, or leveraged, these loans in multiple ways. One was the layering within the securitization market. You could have one asset or mortgage-backed security dependent on a certain bunch of sub-prime loans. But parts of that security could be repackaged into a whole new security. Then two securities were related to the same underlying loans.

These asset-backed securities, or collateralized debt obligations (packages of asset-backed securities known as CDOs), could be used as collateral to purchase and package new bunches of sub-prime loans or to simply borrow money for other endeavors. There is no market limit on the number of times loans or asset-backed securities or CDOs can be repackaged. And each time a repackaging occurs, it introduces more risk and leverage (since a widening swath of securities or subsequent borrowing is dependent on the same loans) into the system. This borrowing, or leverage, acts like a magnifying glass for risk.

To make matters systemically worse, different institutions are allowed to take on different amounts of leverage. The Securities and Exchange Commission had set a net capital rule in 1975 that required broker-dealers (or investment banks which ran broker-dealers) to limit their debt-to-net capital ratio to 12 to 1. In other words, they couldn't borrow more than $12 worth of debt for every dollar of real capital, or equity, they held. In 2000, former Treasury Secretary Henry Paulson, in his capacity as chief executive officer of Goldman Sachs, began the fight to change that rule. Four years later, he testified against it before the Senate Banking Committee. Finally, on April 28, 2004, investment banks like Lehman Brothers, Bear Stearns, and Goldman Sachs got the SEC to increase their 12-to-1 leverage to a whopping 30 to 1. Final approval to lift the rule was unanimous and took 55 minutes. At one time, Merrill Lynch's assets were leveraged 40 to 1. Hedge and private-equity funds didn't even have official leverage rules. To this day, there is no official compilation of their global or domestic leverage.

Leverage increases the likelihood of a bank entering the "too big to fail" category, since it expands a bank's borrowing capacity and hence, the amount of debt it holds. Plus, as banks become bigger, it's more difficult to detect all the pockets in which risk can gather. Case in point: Bank of America was allowed to acquire Merrill Lynch. The acquisition rendered the institution bigger, more leveraged, less transparent, and riskier. The size—and the convoluted balance sheet that comes with it—makes the possibility of seeing risk, let alone containing it, that much harder. Smaller, simpler banks, are easier to monitor and cheaper to save.

In his March 10 speech, Bernanke suggested establishing better regulatory oversight for institutions deemed too big to fail. What he should have demanded is that they not be allowed to get too big to fail, which, along with restrictions on risky practices for commercial banks, was the intent of the Glass-Steagall Act of 1933, which was slaughtered in 1999. Instead, mergers were encouraged at every step—with dire consequences for the nation's biggest banks. Citigroup is in critical condition. The Bank of America and Merrill Lynch merger was a train wreck. The merger of Wachovia and Wells Fargo bled losses. The bargain merger of JP Morgan Chase, Washington Mutual, and Bear Stearns only did a bit better because of government backing.

Multilayered risk has reverberated throughout the financial system with catastrophic results. Whether he explicitly said it or not, Bernanke knew things had gone massively wrong and that the banking system took on more debt than it is able repay. And so the risk-holders of last resort became the federal government, the taxpayers, and international governments.

There are remedies to the problem. But for those remedies to be fully effective, risk must be mitigated at every point from the loan level to the leverage amount to the size and complexity of each institution.

Loans must be better regulated. Lenders should be subject to greater restrictions and transparency on hidden costs and loan complexity and not push loans that borrowers are ill-equipped to understand or repay. Multiple loans should be capped. Lenders should not be allowed to keep preying on the same borrower to take out multiple loans on their only collateral, their home.

The leverage within each security should be restricted by increasing the real assets and limiting the number of repackagings—no more CDOs of CDOs. By spreading the loans so thin, any small failure has a greater impact on a higher pyramid. Plus, multiple repackaging renders each new security less transparent, which makes it harder for regulators to keep tabs on it. The net capital rule should be 8 to 1 for every financial firm that operates in the global market—with no exceptions and no loopholes. The system shouldn't be allowed to leverage itself beyond its capacity to absorb the risk, requiring the government to step in and be the safety net. Financial firms should be required to build their own better safety nets. And all transactions should be on the books. More leverage should not be created in such a way as to obscure capital rules and hide the true nature of institutions' debt.

Last, the riskier components of banking should be separated from the consumer oriented, deposit-loan components. The government has a responsibility to back consumer deposits, because those consumers didn't think their deposits were going to be used to create the other, massively risky transactions that banks took on. These riskier elements should be allowed to fail; otherwise, "super regulators" will continue to have a tough time keeping up with Wall Street and its risk-creating prowess.

Alas, Treasury Secretary Timothy Geithner seems to be maintaining Bernanke's top-down approach to fixing the financial system. When Geithner unveiled his plan for economic stability in late March, he echoed the need for a stronger regulator rather than a banking system that is easier to regulate. This was supposedly a remedy to the idea that "the bigger they are, the harder they fall." But in fact, more careful logic holds that if financial institutions don't get so big, they won't fall so hard. In other words, Geithner failed to grasp that we can contain risk by reducing the possibility of its growth, not by watching it more carefully afterward.

In 1933, Franklin D. Roosevelt understood that the creation of the regulatory agencies, the Securities and Exchange Commission and the Federal Deposit Insurance Corporation, worked because the Glass-Steagall Act rendered the banking arena into smaller, clearer components that were easier to regulate. The current banking crisis screams for a similar approach, one that is focused not on strengthening big regulators at the top but on addressing every level of the system to make banks easier to regulate.

Next: The Rich and Powerful Can Avoid Risk