by Sheila Bair
I also learned from that early experience that trade group lobbyists frequently did not reflect the views of better-managed banks. A number of older, more established banks contacted me in support of what we were doing. Under the statutory scheme set up by Congress, the newer, “free-rider” banks would pay the lion’s share of the initial assessments, as older banks were given credit for premiums they had paid to clean up the S&L mess. That made sense from the standpoint of fairness. (However, the existence of those large credits also impeded our ability to replenish the fund quickly.)
Notwithstanding lobbying pressure, we stuck to our guns. We finalized the rule in November 2006, again on a unanimous vote, and started collecting premiums in the first quarter of 2007. But it was too late to build up the fund sufficiently before the crisis hit. Later, we would be forced to increase assessments and require banks to prepay their premiums to maintain sufficient industry-funded reserves. But our financial condition would have been even worse if we had succumbed to industry pressure and shelved the rule.
Another major issue that divided the board was the question of whether Walmart should be approved for a bank charter and deposit insurance. The general rule—somewhat unique to the United States—is that nonfinancial commercial entities such as Walmart cannot own banks. However, there was an arcane exception to this overarching separation of banking and commerce for banks chartered in Utah. Specialty banks, known as “industrial loan charters” (ILCs), had been used in the past primarily by car manufacturers and other companies that wanted to create banks to make loans for their customers to buy their products. Now Walmart wanted to use it to set up its own bank.
Community banks feared that Walmart would use its bank charter to open up full-service banking branches in its thousands of stores, undercutting small local banks that do not have the same deep pockets and economies of scale. Walmart insisted that it wanted the charter only to perform narrow services such as processing credit card payments. My internal directors were uncomfortable with the Walmart application, as was John Reich, who had deep ties to the community banks. John Dugan, on the other hand, was somewhat sympathetic, not surprising given the fact that he had once worked for Senator Jake Garn, the Utah Republican who had championed the ILC exception.
I wasn’t sure where I came out on the policy issues associated with Walmart having a bank. On the one hand, with Walmart’s huge imprint, I could see that its entry into the banking business could theoretically expand banking services into lower-income communities. On the other hand, the impact on community banks could be severe. I agreed with John Dugan on the legal analysis: the law seemed to say clearly that commercial entities such as Walmart were entitled to own an ILC. But our approval of the Walmart application could dramatically change the face of banking in the United States. Was that really what Congress had intended by approving what was supposed to be a limited number of commercially owned specialty banks?
Like Hamlet, I couldn’t make a decision. So I punted. I asked and got approval from the board to place a moratorium on all ILC applications to give Congress some time and incentive to think about whether it wanted to put some limits on who could have an ILC charter. We had already received several strongly worded letters protesting the Walmart application from influential members of Congress, such as Barney Frank, the chairman of the House Financial Services Committee. I basically threw the hot potato back to them. Here, Congress, you created this ILC exception; we will give you more time to consider whether you really want it to be this broad. I’m not usually one to dodge issues; in fact, I pride myself as the type of person who tackles problems head-on. But I didn’t see any downside to delaying a decision on the ILC issue, particularly given the fact that the controversy surrounding it had been a major distraction for the agency.
The moratorium gave us time to focus on what I considered to be more important matters. According to data analysis presented by our economics staff, the housing market was starting to turn dramatically and a down cycle in housing could pose significant risks to banks insured by the FDIC.
In the second quarter of 2006, there were more than $4 trillion in real estate–related assets sitting on bank balance sheets, representing more than 36 percent of total assets. A precipitous decline in housing prices would create real problems for insured banks. I wanted that looming risk to be our primary focus. Safety and soundness regulation had become too lax. It was imperative that the deregulatory trend that had overtaken Washington be reversed. Our first priority had to be to make sure that banks had enough capital to withstand losses from a housing downturn. Capital was the key to keeping banks solvent as storm clouds gathered on the economic horizon. Yet, instead of moving to increase capital levels, I would find myself in a lonely battle against the other bank regulators—indeed, against the entire global financial regulatory community—to prevent the banks we insured from reducing their capital levels. That fight centered on something called the Basel II advanced approaches, and it was one of the most brutal fights of my public career.
CHAPTER 3
The Fight over Basel II
Managing my diverse board on FDIC-specific issues was hard enough. But some of the most important decisions to be made on bank regulatory policy had to be done on an interagency basis to ensure consistency in how banks were supervised. That meant that we also needed to reach agreement with the Federal Reserve Board. Prior to Ben Bernanke’s arrival in 2006, the Fed had been led by Alan Greenspan for nearly two decades, and during that time, the institution had acquired a strong antipathy to regulation.
Early in my tenure, the other bank regulators were still moving in the direction of less regulation, at least for larger institutions. Adding fuel to their fire was the fact that some of our foreign competitors, particularly in Europe, were taking industry self-regulation to new extremes. In particular, a number of European regulators had embraced “principles-based” regulation, which, in my view, meant articulating high-level standards but then leaving it to the banks themselves to interpret and enforce those standards.
Most problematic was Europe’s implementation of a new framework for setting capital requirements for large banks, known as the “Basel II advanced approaches.” They had been developed by a group called the Basel Committee10 on Banking Supervision. The Basel Committee was established in 1974 to promote international cooperation in bank supervision and in particular, to set global standards for bank capital requirements. The group met four times a year, usually in Basel, Switzerland, and was made up of bank regulators from the major developed nations.
Of all the things that a bank regulator does, setting and enforcing capital requirements are probably the most important. Why? Because banks have certain government benefits that other for-profit commercial entities do not enjoy, which also means that they pose big risks to the government if they fail. For one thing, they have deposit insurance. That allows them to readily obtain funds for their operations from bank depositors, who do not have strong incentives to ask about the safety of the bank because they know the FDIC will protect them from loss if they stay below our insured deposit limits. Banks also have what is called “discount window” access, which is the ability to borrow money from the Federal Reserve System to make sure they always have enough cash on hand to meet their deposit withdrawal and other obligations.
There are good reasons for deposit insurance and Federal Reserve discount window lending. They give the public confidence that the money they have in banks is safe and readily accessible. And by strengthening banks’ ability to attract bank deposits, the banks have more money to lend out to households and businesses to support economic growth. However, because the people who own and run banks don’t have to work very hard to attract deposits and because they know the FDIC will have to cover the losses on insured deposits if the bank gets into trouble, they have incentives to take a lot of risks. That is what is known as “moral hazard.” The moral hazard problem is worse for very large institutions that the market perceives
as being too big to fail. With the very largest financial institutions, the markets assume that the government will protect everyone, not just insured depositors, if they get into trouble. And as we saw with the bailouts of 2008, those assumptions proved to be mostly right. (But more about that later.)
The FDIC has a number of ways it can try to protect itself against banks taking imprudent risks with insured deposits. First, we think it is important to charge banks a premium to cover the costs of bank failures. By requiring the banking system to cover those losses, we give well-managed banks an incentive to look out for the weaker ones. Second, we look to safety and soundness regulation, which is why we think it is so important for examiners to conduct vigorous analysis of a bank’s books and operations. Finally, and most important, we can set capital requirements.
A bank’s capital is, in essence, its “skin in the game.” It is the amount of their own money that the bank’s owners have to stake to support the bank’s lending and other investments. The most basic form of capital—also called “common equity”—is raised by a bank selling stock to shareholders or by retaining its earnings (instead of paying those earnings out in dividends or big employees bonuses). Raising money through common equity is different from raising money through debt issuance. Common-equity owners have no right to have their investments paid back. If the bank is profitable, they share in the profits through dividend payments and appreciation in the value of their shares. If the bank does poorly, however, they have no right to dividends and may suffer a drop in the value of their shares. That is why bank regulators say that common-equity capital is “loss-absorbing.”11 In contrast, when a bank issues debt to fund itself, it is legally obliged to pay the loan back, along with the agreed-upon interest. If it is unable to fulfill that commitment, it is in default—in bank regulatory parlance, it fails.
Insured deposits are a form of bank borrowing. When you put money on deposit at a bank, you are in essence lending the bank your money, which the bank in turn can use to make loans or other investments. The bank is legally obligated to give that money, plus any promised interest, back to you, in accordance with your deposit agreement. If it fails to do so, it is in default, and if the amount of money you have deposited is under the insured deposit limits, the FDIC will step in, take control of the bank, and make you whole.
Left to their own devices, banks will not want to risk much of their own money. Why should they if they can get funding through insured deposits or, if they are a large institution, by issuing debt that bondholders believe is implicitly backed by the government? That is why, even with capital regulation, most banks have much lower capital levels than nonfinancial, commercial entities that do not have access to government-supported funding.
Here are two highly simplified examples that demonstrate why banks cannot be relied upon to set their own capital requirements.
Let’s say we have two banks that have made loans totaling $100 million. Bank A’s owners have funded their loans by putting up $5 million of their own money—capital—and the remaining $95 million they have attracted with insured deposits. Bank B’s owners have put up $20 million of their own capital, and the remaining $80 million they have attracted with insured deposits.
If both banks make a $1 million profit, Bank A’s shareholders’ return on their investment is 20%. However, Bank B’s shareholders’ return is only 5%. As you can see, the rate of return on shareholders’ equity investment goes up quite a bit the less of their own money they invest. That larger return provides more money for dividend payments for them and bonuses for the executives at Bank A. Such “leveraged” returns, that is, investing with borrowed money, can be quite profitable when times are good. In these examples, the borrowing is done through insured deposits.
On the other hand, let’s say those banks made a lot of bad loans and have lost $10 million. Bank A becomes insolvent; that is, it fails. Its shareholders are wiped out to the tune of $5 million, with the FDIC paying out $5 million to fully protect the $95 million in deposits. (The other $90 million would be recouped by the FDIC through selling Bank A’s good loans.) In contrast, Bank B’s shareholders would lose half of their investment. But with $10 million remaining in equity capital, the bank is still solvent. It has not failed. The FDIC suffers no losses, and the bank survives to make further (and hopefully better) loans.
In both scenarios, Bank A’s shareholders come out better. Their return on equity is higher in the first scenario, and their losses are less in the second scenario because they can push half of the losses onto the FDIC.
Now let’s say that Bank A has $2 trillion in loans and other assets, as opposed to $100 million, and that its shareholders therefore think it is too big to fail. Their assumption is that the government will bail the bank out, even if it makes stupid loans and other investments and ends up with losses that exceed its capital. In that case, the shareholders will be completely focused on maximizing returns regardless of risk, because they assume the government will step in and protect their equity investment, if necessary.
These examples illustrate precisely why regulators cannot leave it to banks themselves to set their own capital levels, and that is particularly true of large institutions.
The Basel Committee finalized its first agreement on bank capital standards, Basel I, in 1988. Basel I provided for a fairly simple method of determining bank capital, assigning specific capital requirements to four different categories of bank assets. For instance, for a mortgage (which used to be viewed as low risk), Basel I required the bank to put up capital equity to 4 percent of the loan amount. For other types of loans—for instance, a loan to a business—the requirement was 8 percent. However, as banking activities became more complex, Basel Committee members began work on a new framework that they believed would do a better job of setting capital levels based on risk. Unfortunately, the main idea behind the Basel II effort was to let a bank’s management heavily influence how much capital to hold.
Basel II was controversial from the start. Work on it began in 1998, but the accord was not approved and published until 2004. Studies of how the accord would impact capital consistently showed that it would lead to dramatic declines in the amount of capital held by large U.S. banks. For that reason, the FDIC fought12 and delayed U.S. implementation, and we were even more determined to stop it as we watched capital levels decline among big European banks as they moved forward with Basel II adoption.
It makes sense to consider the riskiness of a bank’s assets as one factor in setting capital levels. Certainly, a prudently underwritten mortgage with a 20 percent down payment is going to be less risky than an unsecured credit card line. Trying to “risk weight” assets for capital purposes is something bank regulators have done for a long time. However, instead of regulators setting clear, enforceable parameters for determining the riskiness of bank assets, Basel II essentially allowed bank managers to use their own judgment. That not only opened the door to lower capital levels, it also inserted a great deal of subjectivity and variation among similarly situated banks in how much capital they would actually hold. That was proving to be the case in Europe. As we discovered during the crisis, the Basel II advanced approaches grossly underestimated the risk of most assets, particularly home loans and derivatives, and also produced wide variations in capital levels among the European banks using it.
Another major concern we had with the Basel II advanced approaches was that their methodology relied heavily on how loans had performed historically. Historically, mortgages had performed well, but that didn’t mean their good performance would continue in the future (as we soon found out). We were also concerned that in good economic times, when loan delinquency and default rates are low, bank managers could say that they didn’t need much capital. But as delinquency and default rates went up in an economic downturn, the Basel II methodology would say that capital needed to be higher, causing banks to try to raise capital during periods of market distress. (As it turned out, that concern was only theoretic
al, but for all of the wrong reasons. Even as Europe later plunged13 into recession and delinquency and default rates spiked, most European banks using Basel II said that their assets were becoming less risky and lowered their capital levels even more!)
Our Basel II staff expert at the FDIC was an intense, soft-spoken career government servant by the name of George French. George was battle-scarred from his years of effort in fighting the Fed and OCC over Basel II. The idea for the advanced approaches had originally come from one of the Fed’s regional banks14, the New York Federal Reserve Bank (NY Fed). It was a source of embarrassment to the Fed that Europe was implementing the framework ahead of the United States. George’s first lieutenant in the war was Jason Cave, an affable but equally ferocious (when needed) advocate of stringent bank capital standards. They had received strong support in this years-old battle from former FDIC Chairman Don Powell, as well as from Vice Chairman Gruenberg when he had served as the acting chairman. But enticed by the prospect of lower capital standards, the biggest banks with all of their lobbying muscle were closing in, and the FDIC was becoming more and more isolated.
Lobbying of the new chairman started early. Both John Dugan and Federal Reserve Board Governor Susan Bies reached out to me during my first few weeks in office, hoping that I would show more flexibility than my predecessors in letting Basel II move forward. As an academic at U.Mass. and member of the FDIC’s Advisory Committee on Banking Policy, I had been supportive of the FDIC’s views on Basel II. It made no sense to me to have a capital framework that let big banks essentially set their own capital requirements while the smaller banks were subject to tougher, more prescriptive standards. Big banks posed much greater risks to the financial system if they failed, so if anything their capital requirements should have been stronger, not weaker. On the other hand, I did not want to appear insensitive to the views of my regulatory colleagues and felt obliged to hear out their arguments.