Most books about the nation's financial crisis tell us what happened. In his new book, HBS senior lecturer Robert Pozen tells us how to fix the system. A financial industry veteran and chairman of MFS Investment Management, a Boston firm that oversees more than $170 billion in pension and mutual funds, Pozen writes with authority and unusual clarity about complex issues in Too Big to Save? How to Fix the U.S. Financial System (John Wiley & Sons).
Roger Thompson: How does the government figure out which financial institutions are too big to fail?
Robert Pozen: There are two valid reasons for bailing out a financial institution. First is to protect the system for processing payments, like checks, because that system is critical to the operation of the U.S. economy. Second is to avoid a situation where the failure of one large, interconnected financial institution is likely to lead to the failure of many other large institutions.
Most of the 600 institutions recapitalized by the federal government over the last year do not satisfy either criterion. A lot of bailout decisions were made ad hoc without a clear rationale. For instance, the government bailed out Bear Stearns, but why not Lehman Brothers? Ironically, Congress in 1991 passed a statute establishing specific procedures (including stating a rationale) to be followed before a bank could be rescued and mandating an after-the-fact audit by the Comptroller General. But because Bear and AIG weren't banks, no one had to explain what they were doing under the 1991 statute. To hold senior government officials accountable for all bailouts, Congress should extend the 1991 statute to any type of financial institution.
Q: How has Congress tried to restore confidence in credit rating agencies?
A: The first thing Congress did in 2006 was to boost competition. We now have nine approved rating agencies instead of three. However, if you are a bond issuer and you don't like what one rating agency says, you can choose another. So now, you have tripled the number of choices. That doesn't solve the problem of forum shopping. In fact, it makes it worse.
Q: What if rating agencies were paid by investors rather than by bond issuers? Wouldn't that stop forum shopping?
A: In theory, yes. The people who are being served by the rating agencies, the investors, should pay for the service. But that's not going to happen because the largest investors in bonds—banks, insurance companies, and mutual funds—aren't willing to pay because they think they do a much better job than the rating agencies.
What I propose is a neutral third-party approach to ratings. The SEC would designate a knowledgeable person, independent of both issuers and rating agencies, to select a rating agency for the bond issuer and negotiate a rating fee. This would eliminate the two worst abuses: the issuer shopping for a higher rating, and the issuer paying inflated fees to get a higher rating. But the issuer would still pay for the fees of the rating agency after it was selected by the third party.
Q: You note in your book that loan securitization collapsed at the end of 2008. Has it revived yet, and why is it important for a healthy economy?
A: The monthly volume of securitization in 2007 was over $100 billion. Now, it's $1 or $2 billion a month. So we've got a long way to go. And we need to get securitization going because that's what drives loan volume.
Q: Why has loan securitization been slow to recover?
A: We had a terrible system of securitization where everything was off balance sheet, and we made believe that the sponsors, the biggest money center banks, had zero risk of loss. They did not fully disclose what was happening, and they did not put up enough capital to cover potential risks. Now, the FASB [Financial Accounting Standards Board] has overacted by adopting rules that effectively force all securitizations on the balance sheet. Since the new rules treat banks as if they have 100 percent of the risk of loss, they must put up capital as if that were true.
So we've gone from one extreme to another, with the reality lying somewhere in between. If we are going to resuscitate securitization, we should utilize off-balance-sheet entities but with a transparent process and capital charges that are based on actual risks. We want banks to disclose their potential liabilities to these entities and to put capital behind these risks. Unless we have that sort of transparent process backed by capital, we're not going to revive securitization.
Q: Corporate boards have been criticized for being asleep at the wheel leading up to last year's financial meltdown. Are boards at fault?
A: After Enron and WorldCom, Congress hastily passed Sarbanes-Oxley, basically a very elaborate set of procedures for boards to follow. But it's very clear that the boards of megabanks—the nineteen banks with over $100 billion in assets—complied with Sarbanes-Oxley and somehow didn't realize what huge risks they were taking.
Q: If boards don't need more procedures, what do they need?
A: Some of the most effective boards are those at companies that are owned by private equity. They are composed of the CEO and six directors, all of whom have relevant industry expertise. The directors make the time commitment, spending several days each month at the company. And they all have significant stock incentives.
The question is, what can we learn from the private equity model? When it comes to megabanks, I'm in favor of a smaller board with deep financial expertise, substantial time commitments, and a different pay structure. We can try to be clever and add more procedures. But unless we rethink the board model in a very fundamental way, I believe we're kidding ourselves. Is it likely that somebody who isn't a financial expert can show up six times a year and really understand Citigroup?
Q: In an attempt to head off the next financial crisis, should Congress designate a systemic risk regulator? If so, who should that be?
A: The Treasury has proposed that systemic risks be monitored by a newly formed Financial Services Oversight Council, with the Fed becoming the primary regulator of all institutions posing such risks. I disagree with making the Fed the primary risk regulator.
First of all, the notion that the Fed could be the primary regulator of every systemically risky institution is just not practical. That means it would need to be an expert on money market funds, hedge funds, and life insurance companies as well as banks. Second, if you identify an institution as systemically risky, you're creating moral hazard [implicit federal guarantees] by that very process. Third, you're assuming that we can know in advance every institution that's systemically risky, but I don't think that's possible. Finally, why do we assume that it's just institutions? Sometimes rapidly growing new financial products are systemically risky, like credit default swaps.
Q: What would you do?
A: My proposal is just the opposite of the Treasury's. I would put the Fed in charge of risk monitoring because that's where it has the most expertise. When the Fed comes upon a systemic risk, it should turn to the relevant regulatory agency to resolve the problem. So you get the best of both worlds. The Fed does what it's best at, macro risk monitoring. And the agencies with the deepest expertise in the relevant financial area would be in charge of problem resolution. Moreover, if the Fed assumed the role as primary regulator of all systemically risky institutions, it would jeopardize its political independence. And that would be a big mistake.
Excerpt: the New Structure Of U.s. Financial Regulation
by Robert Pozen, from Too Big To Save
To provide taxpayers with an equitable stake in a mega bank that needs exceptional assistance, the Treasury should contribute capital by purchasing common shares. As a result, the Treasury is likely to hold a majority ownership of the bank's common shares, while the ownership interests of other shareholders will be reduced. This is not permanent nationalization in the socialism sense; this is temporary majority ownership by the government until it can dispose of the mega bank. By owning a majority of the troubled mega bank's shares, the Treasury can enjoy most of the bank's upside gains as well as absorbing most of its downside losses.
With majority ownership of seriously troubled banks by the federal government, it can divide them into good banks and bad banks. The good banks would return to the normal business of taking deposits and making loans; the bad banks would work out and sell toxic assets over several years. If the U.S. Treasury and other existing securities holders were given equal ownership interest in both banks, taxpayers would participate in the potential upside as well as shoulder losses. Moreover, splitting a trouble institution in this manner would avoid the intractable problem of setting a fair price now for the sale of its toxic assets.
The splitting of a troubled institution into two banks is a much better approach than the creation of heavily subsidized public-private partnerships to try to buy toxic assets. These partnerships are another example of one-way capitalism: Private investors receive 50 percent of the upside but little of the downside on toxic assets that are actually purchased. Moreover, the partnerships are likely not to set a market price on many toxic assets, because the government will not provide generous subsidies to buy them on a regular basis.
The government's focus on recapitalizing banks and buying their toxic assets seems to be based on the assumption that banks are the primary originator of new loans. In fact, banks accounted for only 22 percent of the credit extended in the United States. The main cause of reduced lending has been the collapse of the loan securitization process, which allows banks and nonbanks to sell loans and re-lend the cash proceeds multiple times. The volume of new issues of securitized loans has fallen off a cliff, from $100 billion a month in 2006 to almost zero at the end of 2008.
To revive the process of securitizing loans, the United States needs to establish proper incentives at each stage of the process. We need to ensure that mortgages are appropriate for the resources of borrowers, and that mortgage brokers have skin in the game when they sell loans. We need to control the conflicts of interest of credit-rating agencies and reformulate the capital requirements for bank sponsors of special purpose entities (SPEs) that issue asset-backed securities. But all asset securitization should not be forced back on the balance sheets of banks. Instead, bank sponsors should publicly disclose and back with capital their continuing or contingent obligations to any SPE they sponsor.
In response to the financial crisis, the federal government has substantially increased its intervention into the financial markets. Although such intervention is justified in certain cases, federal guarantees of debt offering are too extensive. To avoid moral hazard, the FDIC should guarantee 90 percent, rather than 100 percent, of debt offerings by banks and thrifts. Further, Congress should not extend beyond 2013 the higher limits on FDIC deposit insurance. The previous limits covered 98 percent of all depositors. The United States should move away from a financial sector with broad-based government guarantees to one with market discipline exerted by sophisticated and at-risk investors in bank debt.
The federal government should not be encouraging mergers among large institutions in the financial sector, which is now dominated by a handful of mega banks. The Justice Department should reject mergers that are likely to create more mega banks that are too big to fail. However, we should not attempt to increase competition in the financial sector by reinstating the barriers of the Glass-Steagall Act to the securities activities of banks. Freestanding investment banks present systemic risks because they have limited sources of short-term liquidity: commercial paper and repurchase agreements. Banks with securities powers can also obtain short-term financing through Fed loans and retail deposits.
Given the decline in investor discipline and market competitions, the monitoring of financial institutions has been left mainly to federal regulators. But there are limits to the effectiveness of any federal regulator in light of the fast pace of financial innovation and complexity of financial transactions. On a regular basis, the outside directors of a mega bank should be responsible for monitoring its activities. However, most outside directors of mega banks are not financial experts, do not spend enough time on board matters, and do not have a large equity stake in these institutions. If the United States wants effective board oversight of complex financial institutions, we should move to the private equity model for their boards. Under that model, a small group of super-directors with extensive financial expertise would spend several days every month at the bank; they would also have substantial holdings of the bank's stock.
A board of super-directors would be well-placed to monitor the financial condition of a mega bank and set the compensation of its senior executives in order to attain its dual goals of maximizing long-term profits for its shareholders without taking risks that would materially jeopardize the bank's solvency. Thus, super-directors are critical to fixing the U.S. financial system and moving it to accountable capitalism.