Print this article | Return to Article | Return to CFO.com
In his new book, Too Big to Save?, Robert Pozen proposes fixes for the biggest financial crisis of our lifetimes.
Tim Reason, CFO.com | US
November 10, 2009
Often, when ominous noises emanate from this country's complex financial machinery, Robert Pozen is called on for a diagnosis. The chairman of mutual-fund company MFS Investment Management and a lecturer at Harvard Business School, Pozen has a knack for delving into the nuts and bolts of tortuous policy issues and seeking the most practical solutions, even if, as he says, "they're not perfect."
Most recently, he presided over the Securities and Exchange Commission's blue-ribbon Committee to Improve Financial Reporting — a year-long effort that resulted, in August 2008, in a series of recommendations widely called the "Pozen Committee Report." In 2001 and 2002, he also served on President Bush's Commission to Strengthen Social Security.
As the financial crisis unfolded, says Pozen, he was asked by everyone from his Harvard students to members of his own book club to "explain what was happening and how were we going to get out of this mess." He began to pen an analysis of the crisis — and, true to form, propose a series of fixes. The resulting book, Too Big to Save?, was released Monday, with each passage on a proposed policy marked in bold type. Pozen spoke about the book with CFO last week. An edited version of that discussion follows.
What prompted you to write this book?
It was clear that there was this tremendous appetite for people to both understand what had happened [to cause the financial crisis] and also to evaluate what the government was doing. A number of books were coming out, but they focused on a particular aspect of the crisis, like Lehman Brothers or credit default swaps or the Fed. They were also very backward looking. Many of them were quite good at financial journalism, but they were essentially a narrative of what had happened in the past. They didn't really [address] what we should do in the future. And I have a lot of ideas about what we should do in the future.
The first three chapters of your book deal with the mortgage securitization issues at the heart of the crisis. Compared with mortgage or even credit-card securitizations, corporate securitizations are a tiny slice of the pie. Yet you argue that the policy issues you address in the book are highly relevant to corporate readers. Why is that?
Like it or not, CFOs are caught up in the wave of reform and reaction to what some people viewed as the misuse of off-balance-sheet entities and the misuse of financial derivatives. I think we'd all have to agree that there surely were a lot of misuses. So the question is whether in responding to that the regulators and Congress will go so far as to make life very difficult for corporate treasurers and corporate CFOs, even though they were not culpable in the least bit. And I think unfortunately that's what's happened.
The Financial Accounting Standards Board has thrown up its hands and said, effectively, that there isn't anything that can be off balance sheet anymore. If a sponsor has modest indices of control and some of the benefits and risk, the sponsor is almost always going to be considered to have that off-balance-sheet entity on its balance sheet.
That has much larger repercussions. I don't think it was aimed at corporates; it was aimed at the money center banks. But FASB doesn't want anyone to do off-balance-sheet [accounting]. I think that's a mistake. We've gone from one extreme, where some bank like Citigroup would have $40–$50 billion off balance sheet, to one where it is all on balance sheet. That's quite a switch for that amount of money. The reality in most off-balance-sheet financing is that the sponsor lays off some, but not all, of the risk. These are never all-or-nothing situations.
We've got to find a middle ground, because loan securitizations are the key to loan volume. President Obama said last week he was very disappointed that banks weren't lending anymore — as if banks were the key to lending. They aren't. The key to lending is loan securitization. And if we force banks and corporates to put all the securitizations on their balance sheets and to treat them as if none of the risk has been laid off, well, I don't think you're going to have much in the way of securitization.
So your book proposes that off-balance-sheet accounting should remain.
You would have an off-balance-sheet entity. The sponsor, whether it be a corporate or a bank, would disclose its contingent obligations, [such as] credit supports. And in the case of the financial institutions, there would be some capital support for that based on actual allocations of risk.
You say that FASB has essentially given up on securitization accounting. But hasn't FASB repeatedly tried to find a middle ground, only to find that the rules it devised were circumvented?
It's a fair point that FASB, after Enron, tried to come up with [better] rules. But it has always had rules that are all-or-nothing tests. Either you're off balance sheet or you're not. The whole nature of those [type of] rules encourages gaming.
So your proposal says there should be a realistic allocation of risk and capital to each of the various stakeholders.
Yes. That's not amenable to gaming. Maybe you could game it so you'll only have 55% of the risk instead of 60%. But it's not an all-or-nothing situation.
Your proposal actually sounds analogous to what FASB and international accounting standards-setters are considering for leasing, which is also off balance sheet.
Right. [Current] lease accounting has that same characteristic. [Editor's note: A lease can be held off balance sheet if the present value of the lease payments is less than 90% of the leased asset's fair value.] [As a result], people have 89.9% and they do these crazy things just to get under the rule. So of course it's gamed. But, yes, if we had a leasing rule that recognized allocations and didn't see this as all or nothing, then we wouldn't have that sort of gaming. It's an important lesson for FASB to learn.
In the actual mortgage crisis, banks first refused to do any sort of workouts for borrowers because they feared violating the QSPE structure, but then later found a rationalization that twisted the existing accounting to justify rewriting whole swaths of loans without even contacting the borrowers. Do you think securitization can recover in the eyes of investors?
That's a very important question. The answer is first to have real transparency. I know that's a buzzword, but remember, if we have these entities on balance sheet, we won't have very good disclosure about them either. You'll have a footnote somewhere in Citigroup's financial statement that will aggregate them all together. We've got to have really good disclosure.
So in addition to having banks disclose their contingent obligations in off-balance-sheet vehicles, you also propose that off-balance-sheet vehicles themselves should remain registered SEC entities.
Yes. They usually are registered in the initial offerings and usually deregister within a year, claiming that they no longer have 300 investors. That's a mistake, because then there's no way for shareholders of the company, the shareholders of the bank, or the investors to really figure out what's happening once there's a deterioration of the assets. Realistically, there are two groups of investors who care: one is the investors in the off-balance-sheet entity and the other is the investors in the sponsor, whether it be a corporation or a bank. So we ought to say that unless the total number of investors who are interested drops below 300, then we shouldn't allow deregistration.
It would seem that credit-rating agencies, also badly tarnished by the crisis, remain a necessity for the capital markets.
Let's be clear: credit-rating agencies are a necessity for two classes of investors: investors who are legally required to only hold A, AA, or AAA securities, and the many pension funds and other [investing entities] where the directors, the trustees, have [established that requirement]. These tend to be small and middle-sized investors. The anomaly that makes the problem [of credit-rating agencies] very difficult to solve is that the large bond investors, like MFS or the IBM pension fund, actually don't think that highly of the credit ratings. They look down on all of them and say, "We're not going to pay for that because we don't think it's worth much."
In theory it's clear what the answer should be: investors should pay for the credit-rating agencies because they're the ones who are supposed to be benefiting from it. Unfortunately, the big players in the bond business don't want to pay. And I don't think we could tolerate a solution where we had all the small and middle-sized investors paying for something that would inevitably be a public good. It would get out, and the big investors would have the slight advantage of knowing the rating, but they wouldn't pay for it. So having investors pay for ratings is a theoretically interesting solution that has no practical chance of being adopted.
The way things are structured now, there's not a lot of credibility at the rating agencies. Congress's first reaction was to create more approved agencies because they thought it was an issue of not enough competition. In my view, they've got it exactly backwards. The problem with rating agencies is a forum shopping issue. The issuer goes to one agency, and if the agency won't give the securities a high-enough rating, then it goes to the next one. So if you go from 3 agencies to 9 or 10, you just increase the possibility for forum shopping.
In your book, then, you propose that the SEC appoint a consultant to select a credit-rating agency on behalf of each issuer.
Yes. It's not a perfect solution, but it focuses on the key problem of forum shopping. For one day, the consultant represents investors, and the consultant's job is to do an RFP and pick the credit-rating agency that's going to have the most accurate rating and at a reasonable cost. Once picked, the credit-rating agency will be paid by the issuer. But the issuer won't pick the agency. It's this shopping around that leads to this terrible result.
And you think this has a practical chance of being adopted?
This is a case in which you have to think through carefully what's the real problem, and what are alternative solutions that will at least get you most of the way [toward solving it], although they're not perfect. I've had talks with some credit-rating agencies and they were receptive to the idea. They don't like the position they're in now; they want their credibility back, too. And for the sake of the small and middle-sized investors, we need their credibility. Those investors don't have the resources of an MFS to look at all these bond offerings. They have to accept the credit rating.
Let's discuss another adoption issue: International Financial Reporting Standards. There's a lot of uncertainty about how soon, or whether, companies in the United States would convert from U.S. GAAP to IFRS. But in your book, you say that such a change makes sense only for the largest two or three hundred companies in the United States, and that the rest should remain on GAAP. Why is that?
We have to look at this as essentially a cost-benefit issue. Companies that have substantial subsidiaries overseas — this would be true of IBM, Hewlett-Packard, General Electric — are already doing financial statements in IFRS in other countries. They have a real interest in being on one uniform standard, they're probably spending all or most of the money [for conversion] already, and they get a substantial benefit out of it.
But if you look at the number of companies that really have significant operating subsidiaries — I'm not talking about sales, I'm talking about operating subsidiaries — if you go beyond the top 300 companies, you really have a very small number of companies. What's the market cap of the last 100 in the S&P 500? It's actually in the $2–$3 billion range. Those aren't really big companies.
You have six or seven thousand publicly listed companies. Of those 6,000 companies, how many of them are already doing IFRS? Almost none. How many of them are likely to do IFRS? None. It's only if we force them to. What will be the cost of their doing it? A lot. What will be the benefit of their doing it? Minimal.
So why would we want to do this? We should concentrate on that tier of companies that are global companies where it makes sense. Now you can argue that will lead to a two-tiered system, but my view is that we have a two-tiered system now.
If you took a survey of the companies that were between say, [number] 300 and 500 on the S&P 500, I bet you none of those companies would want to go to IFRS. I'm serious.
How about the reverse? Would you prohibit smaller companies from choosing IFRS?
I don't like to give companies optionality. I think we need a hard-and-fast rule. Because if you give companies optionality, accounting firms will spend a lot of time and money telling them where they can increase their earnings. There are a lot of substantive differences between GAAP and IFRS, especially in the high-tech area.
True, but doesn't that raise a possible objection to your proposal? Take revenue recognition for software. Under IFRS, Microsoft might have an advantage over a smaller up-and-coming challenger using GAAP.
You could always choose to have the up-and-comers present it both ways. You can always have additional disclosure.
Suppose that your proposal were adopted. Would you still argue that substantive reconciliation should continue with the ultimate goal of eliminating all differences between GAAP and IFRS — however long that would take?
Yes, I would. I think it's a worthwhile goal to try to have a global standard. I think we all gain from that. The closer these standards are, the less difference there would be between IFRS and GAAP [results], though I still maintain that interpretation and audit will make a big difference.
One of the issues that seems to really concern you is the independence of the Fed. In fact, you've said that Treasury is treating the Fed like a hedge fund.
The Fed has done a good job with short-term liquidity, but it has taken on roughly $500 billion of guarantees of truly troubled assets where it could lose several hundred billion.
The Fed has let the Treasury use it as a hedge fund, leveraging it up. The Treasury has only $700 billion [in congressionally approved bailout funds], so they lend the Fed $100 billion and the Fed lends out $1 trillion for these toxic assets programs.
When the government guaranteed $300 billion of Citigroup's worst assets, the Fed took $220 billion of that guarantee. That's a mistake. The Fed ought not to be holding those sorts of assets. Traditionally, treasuries have been more than 90% of the Fed's balance sheet. That's why it used to be very easy for the Fed to drain money for the system — it sold treasuries. Now, treasuries are less than 30% of the Fed's balance sheet. Is it going to sell toxic assets now?
Ultimately, the most important thing is that the Fed needs to be independent so it has the will and the ability to raise interest rates and drain the cash from the system when we need it.
We've been through a pretty frightening financial trauma, but we at least seem to have stabilized things for the time being. Which issues covered in your book do you consider the greatest cause for ongoing concern?
One is that we haven't brought back loan securitization. That's key. Second, we've basically set up, either implicitly or explicitly, an incredible safety net among an incredible number of financial institutions, which is much too broad. We need to bring the bondholders back into play. We need to stop creating the impression that we're bailing out everybody. That's a really big problem and we need to deal with that.
A third big problem is that we really need much more effective boards [at financial institutions]. We don't want to have the regulators micromanaging. And the fourth problem is that we need to make sure that the Fed is independent.