Perfection is out of the question, better is possible.– Hyman Minsky
Banks are special, but are banking regulators also special? The answer is that the federal banking regulators operate in a cultural mindset different from other financial sector agencies and the larger administrative state.
Administrative law scholar Kenneth Culp Davis has argued the following:
The banking agencies of the federal government have long maintained systems of secret evidence, secret law, and secret policy. The result has been a degree of unchecked and unstructured discretionary power that is far greater than it should be. Sound principle calls for openness, so that discretion may be checked and structured. To some extent the systems the agencies have been following violate existing legal requirements. The banking agencies can and should make procedural changes that will increase both efficiency and fairness.
Surprisingly, this statement was made in 1966, and it remains fresh today.
History explains the separate cultural tradition. Banking supervision developed within the central paradigm of the confidential bank examination and the discretion to control access to bank charters. The tradition of transparency, accountability, and the rule of law is a more recent creature of the New Deal and the reactions to it. These two regulatory traditions have long lived in an uneasy truce, made possible during simpler times when banks engaged almost exclusively in taking deposits and making commercial loans. As market competition made the banking sector more complex, the unexamined truce was, counterintuitively, sustained by the expansion of both traditions. Today, we have both more secrecy and discretion and more transparency.
Some secrecy is justified by the need for financial stability but much of it, by design or happenstance, protects the federal banking regulators from public accountability and the appropriate checks on their discretion that would be provided by more openness. The extensive scope of this shadow regulatory system has become untenable and unstable. The two contradictory traditions need to find a new balance in favor of more transparency, accountability, and the rule of law.
The Central Paradigm of Confidential Supervision
Confidential supervision and discretionary secret lore have a long history in the banking sector. (Note: I prefer the term “secret lore” to “secret law,” even though many banking lawyers, including the author, will in conversation refer to “secret law.” We do well to remind ourselves that, in a democratic country, law in the sense of binding constraints cannot be secret. And, under the Administrative Procedure Act, it is not.) These traditions developed when federal regulation was limited, long before Federal Deposit Insurance, the creation of the Federal Reserve as the lender of last resort, the New Deal, and the Administrative Procedure Act (APA). (Note: The FDIC was created in 1934 but its culture is that of the pre-New Deal banking regulators.)
The bank examination is two centuries old. The job of the traditional bank examiner was to look closely at the loans and liabilities of each bank and confirm that vault cash and reserves really existed. (Simpler times exemplified in It’s a Wonderful Life by Mr. Carter, a diligent bank examiner checking whether the Bailey Building & Loan had cash in its vault.) His mission was to assess whether the bank was safe and sound in an era when rumors could lead to deposit runs and bank panics were frequent. (Note: Panics took hold of the American banking sector in 1819, 1837, 1857, 1873, 1893, 1901, 1907, 1929, and 1933.) Thus developed the tradition of the secret bank examination, the crime of spreading false rumors about a bank, and the view that bank supervision was best done inside a cone of confidentiality to avoid triggering a banking panic. (Note: N.Y. Banking Law § 671 False Statements or Rumors as to Banking Institutions.)
Central to the concept of a confidential bank examination is the need for a free flow of communication in conditions of high trust among bank management, the board of directors, and the supervisory authority. Examiners feel strongly about the need for this candid conversation, which has contributed to the creation of a common-law bank examiner’s privilege that keeps reports out of the public domain and out of the hands of the plaintiffs’ bar. (Note: The need for candid conversations in the supervisory context is hotly defended in the courts by banking regulators. At the same time, banking regulators frequently take the view that the attorney-client privilege should be waived by the banks or limited in supervisory communications. So, sometimes candid conversations are encouraged and sometimes they are not.)
The lack of a solid foundation in federal law for many of the secrecy traditions will surprise those who have accepted them as hallowed texts. There is a federal criminal statute that prohibits bank examiners from disclosing the results of an examination (18 U.S.C. § 1906). The comptroller may, if not satisfied with a national bank’s response, disclose an examination (12 U.S.C. § 48). There is no federal statute that explicitly prohibits nonexaminers from disclosing the bank examination or parts of it, such as CAMELS ratings. (Note: CAMELS is used in this essay for simplicity even though there are other ratings systems with their own acronyms.) Rather, the prohibition arises from the interesting assertion by the federal banking regulators that bank examinations are the property of the banking regulator. This assertion derives not from a statute but a regulation promulgated in the late 1960s after the introduction of the Freedom of Information Act’s exemption for confidential supervisory information. The authority to promulgate it is less solid than one might think, relying on a general statute relating to federal government property (18 U.S.C. § 641). One suspects that the general federal property law was pressed into service by the banking regulators because no other statutory authority was available. The criminal prohibitions on banking organizations revealing bank examinations derive solely from this assertion.
The other central paradigm that infuses the culture of banking supervision is the discretion that banking regulators have over entry into the banking sector. As a result of a Supreme Court decision from the early 1970s (Camp v. Pitts, 411 U.S. 138, 1973), a generation has grown to accept that the granting of bank charters is so up to the discretion of the bank regulators that the regulator need not even give reasons for a denial. It was not always so. An early comptroller of the currency thought he had no discretion to deny a request for a charter if the standards of the statute were met. By the mid-1960s, there was a circuit court split on whether there ought to be hearings on charter denials. One of the leading APA scholars of that time criticized the comptroller for its policy of keeping secret applications for charters and the reasons for denials. (Note: Davis, supra note 3, at 713, 716 (stating “I have asked…the Comptroller’s staff why the evidence is concealed even in the absence of a reason for confidentiality, and I have never heard a satisfactory answer” and noting that not publishing reasoned opinions or findings of fact lends itself toward arbitrariness)). He suggested that the comptroller’s withholding of its guidance for charters was illegal under the APA (Id. at 716, 718). Today, we should question again how astonishing it is that the regulator can deny a bank charter without giving any reasons at all for the denial. (Note: There is no time limit on when a decision has to be made. We owe a debt of gratitude to former Acting Comptroller Keith Noreika, who made clear that the OCC granted charters in the post-financial crisis era upon which the FDIC had failed to act. Transcript of “CFTC Talks: OCC Acting Comptroller Keith Noreika, 12,” Aug. 10, 2017, CFTC Podcasts. From this custom of delay and discretion, a practice has sprung up of submitting applications silently and withdrawing them if a denial is expected.)
The importance of the supervisory judgment exercised in the bank examination and the discretion afforded as to entry into the sector are the bedrock of confidential supervision and secret lore. They are routinely justified by the need for financial stability and to ensure the safety and soundness of individual banking organizations. Many have debated whether confidential supervision and its secret lore are binding upon the banking organization. In a world where the supervisor can punish the banking organization and mold its behavior through these tools, from the perspective of those who receive it, the secret guidance and lore is binding.
The Other Tradition of Transparency
The central paradigm of the other regulatory tradition is that of the disinfectant of disclosure. Created in the New Deal or as an immediate reaction to it, the norms of the securities disclosure laws and the APA are much newer. The APA is a “bill of rights for the new regulatory state” because it demands that regulations be public and subject to notice and comment. It has transparency and accountability as its central core. The APA was the end product of a decade’s worth of political wrangling between New Dealers, who fought to protect the highly discretionary administrative state, and those concerned with the rule of law and transparency when powerful agencies exercised quasi-legislative and quasi-judicial powers. A compromise was finally reached following Harry Truman’s assumption of the presidency in a post-WWII environment more sensitive to the risks of authoritarian tendencies (Id. at 1683).
While the New Deal and the APA may seem like long-ago developments, by the time of their passage, the cultural traditions and institutional path dependency of the banking regulators had already been set. Early versions of the bill that became the APA excluded the federal banking agencies from its scope (Id. at 1618). The banking regulators might be forgiven, in the early years after the APA, for thinking that the APA only lightly applied to them. The APA and the norms of transparency and accountability for which it stands are approaching their 72nd year, and the time for cultural adjustment should be over. (Note: We might have some sympathy for banking regulators. Law schools have also treated administrative law and banking regulation as unrelated topics. Gillian E. Metzger, “Through the Looking Glass to a Shared Reflection: The Evolving Relationship Between Administrative Law & Financial Regulation.” 78 Law & Contemp. Probs. 129, 145 (2015)).
It is safe to say that none of the banking regulators, and certainly not the banking bar, noticed the passage of the Congressional Review Act in 1996. The Government Accountability Office (then the General Accounting Office) ruling that the leveraged lending guidelines are a legislative rule, although completely obvious under the APA, has been a surprise to the traditional cultural mode. From the perspective of the banking regulators, the leveraged lending guidance, originally developed as secret letters to banks, which could not disclose them to their clients, was an advancement in transparency and disclosure. They were, after all, public and had been subject to notice and comment. From the perspective of those who have been thinking deeply about the administrative law and its march toward transparency and accountability, they did not go far enough. The ruling is a cultural shock to the shadow regulatory system.
Expanding Both Traditions
Both of these traditions have co-existed in an uneasy truce for so long for the counterintuitive reason that, as the banking sector has become more complex, both transparency and secrecy have expanded in scope. On the side of transparency, there is more disclosure of banking organization financial information, more notice-and-comment rulemaking, a greater tendency to publish guidance, interpretations and FAQs, and greater disclosure of formal and informal actions against banking organizations. On the side of confidential supervision and secret lore, the supervisors have steeply increased the scope of confidential supervision far beyond the traditional bank examination. They have relied upon this broader view to expand the realm of discretionary secret lore.
Expansion of Disclosure
The scope of financial disclosure by banking organizations and its companion market discipline has been expanding for a long time. Both banks and bank holding companies that are public companies are subject to the securities laws, where there is a strong tilt toward disclosure and the constraints of generally accepted accounting principles apply. Pillar 3 of Basel II, now in full implementation, also requires more disclosure. Even though Pillar 3 seeks disclosure in order to manage risk and improve stability, whereas the Securities and Exchange Commission disclosure philosophy is driven by investor protection and market efficiency goals, the existence of subordinated debt, credit default swaps, and, more recently, total loss-absorbing capacity (TLAC) debt that might be bailed in, all push toward market-signaling functions.
These disclosures have become so embedded in our consciousness that many have forgotten that they were new and shocking not so long ago. Bank stocks were not subject to periodic reporting until 1964, and the constraints of Guide 3 date from 1976 (Id. at 161). Call reports were not made public by the Federal Deposit Insurance Corporation until 1972, and even then it was upon request, with a fee for search costs (37 Fed. Reg. 28607, 28608 (Dec. 28, 1972)). When first created, CAMELS ratings were not even disclosed to bank management. It has been forgotten that the banking regulators once attempted a separate securities disclosure system for banks rather than incorporate by reference SEC regulations (Mathewson, supra note 26, at 153-158). Vestiges of that differing cultural mode can be seen in the long-standing spat between the banking regulators and the SEC regarding the calculation of allowances for loan losses. That particular clash was resolved by a detente in the early 2000s and a recent change in the accounting rules. Even so, unresolved tensions remain between the requirements of the securities laws and the culture of confidential supervision. (Note: Nonetheless, the so-called GAAP/RAAP debates sometimes show up in the footnotes to call reports. It remains to be seen how the implementation of CECL will impact this dynamic.)
The formal and informal punitive actions of the banking regulators against banking organizations have become increasingly more public. Banking regulators were not given formal enforcement powers until 1966. Before that, the banking regulator’s main powers were behind-the-scenes moral suasion and “jawboning,” backed by the nuclear threat – not used – to revoke a charter or terminate deposit insurance. Even after the banking supervisors were given the power to remove directors and officers, impose civil money penalties, and enter into informal written memoranda of understanding or formal consent orders or written agreements, they favored informal – that is, nonpublic – board resolutions and memoranda of understanding. The long litany of very public post-financial crisis consent orders and written agreements has changed that custom. Moreover, there is an increasing tendency to disclose informal and private memoranda of understanding in securities disclosure documents when their contents are material to investors with the express consent of the banking regulators. (Note: Requirements to raise capital and restrictions on dividends are the core examples. Other memoranda of understanding remain undisclosed.)
Many written interpretive positions were kept secret well into the 1990s. It was long a given that the only way to find out the Federal Reserve’s interpretive letters was to file an FOIA request and hope for the best. (Note: As a young lawyer in the early 1990s, the author was frequently tasked with crafting an FOIA request to capture an interpretive letter that was known by the bank regulatory bar to exist but which was not public. Letters received under FOIA were carefully tended in paper files and shared among banking lawyers.) The development of the internet, which intensifies the culture of transparency, has meant that many, but not all, supervisory letters, guidance, FAQs, and other interpretive positions now find their way onto the banking regulators’ websites. There is more in the public domain than ever before through notice-and-comment rulemaking. This trend started even before the financial crisis and the Dodd-Frank Act required 390 new rulemakings by the banking agencies. (Note: Davis Polk & Wardwell LLP. Dodd-Frank Progress Report. July 19, 2016)
Expansion of the Secret Realm
At the same time, the scope of what is considered confidential supervision information has expanded far beyond the traditional bank examination. All of the federal banking regulators used the passage of FOIA, a statute meant to expand the scope of information available to the public, to expand their zone of confidentiality beyond the scope of the traditional bank examination. The precise words of the FOIA statute’s exemption, which were written by the banking regulators, encompass a range broader than an examination report and include matters “contained in or related to examination, operating, or condition reports prepared by, on behalf of, or for the use of an agency responsible for the regulation or supervision of financial institutions” from disclosure (5 U.S.C. § 552(b)(8)). The result has been the creation of a nonpublic, shadow regulatory system that is neither transparent nor subject to accountability. It is only possible, due to the constraints of confidential supervisory information, to speak about those examples that have randomly become public, but many more exist. Those who are at banks or the regulators can fill in their own examples.
Widening the scope of the bank examination & positions taken within it. The traditional bank examination has morphed into something much wider in scope. Banking organizations are subject to multiple examinations, and the number of matters requiring attention or immediate attention have expanded. Operation Chokepoint, which purported to decide which sectors were morally appropriate to receive banking services, started in the bank examination. The leveraged lending guidelines, which purport to guard against the next asset bubble, began in the bank examination context. There have been attempts to take legal interpretations in the context of bank examinations. (Note: In a democratic society, the concept of a secret legal interpretation of an economic regulation cannot be possible.)
Federal Reserve power to limit the activities of a financial holding company. To qualify as a financial holding company (FHC), the holding company and all of its insured depository institution subsidiaries must be both “well managed” and “well capitalized.” The Federal Reserve may impose limitations on the conduct and activities of an FHC that fails to satisfy either condition, and typically the FHC is required to enter into a nonpublic agreement to comply with those limitations.(Note: These agreement are imposed under 12 U.S.C. 1843(m) and are typically referred to as 4(m) agreements.) Because the Federal Reserve treats the failure to be well managed as confidential supervisory information, the existence and scope of 4(m) limitations are confidential. One study that examined the securities disclosures of 60 FHCs between the years 2005 and 2017 noted that nearly all FHCs disclose that they are well capitalized but many do not disclose if they are well managed. The public does not know the conditions on business activities or investments that exist through this mechanism and how long they might last.
Federal Reserve power to limit investment in the banking sector. The Federal Reserve’s legal staff has influence over who can invest in the banking sector through fluid standards on controlling influence and by creating secret lore surrounding the so-called teardown “rules,” which take views on when an investor has broken controlling influence. These standards have not been subject to notice and comment, and most of them are not public. Vice Chair Randal Quarles recently described these standards as “ornate” and noted that “in some cases [they] cannot be discovered except through supplication to someone who has spent a long apprenticeship in the art of Fed interpretation,” characterized in ad lib comments as akin to the relationship between “shaman and novice.”
Limits on expansion and acquisitions. Another expansion of secret lore has come about due to the penalty box rules of thumb that the banking supervisors apply to banking organizations as a result of CAMELS ratings, especially as to management ratings, Bank Secrecy Act and anti-money laundering compliance reviews, and consumer compliance reviews. As noted by Greg Baer, tacit principles in the evaluation of management include the fact that any compliance problem resulting in an enforcement action will result in a downgrade of the management rating and that an entity cannot have a composite rating better than 3 if it has a management rating of 3. Bank expansion is not possible so long as a consent order is pending, meaning that banks of all sizes devote board and management time as well as technology resources toward even the most immaterial compliance concerns to ensure that regulators are fully satisfied. Appeals against adverse ratings are rare because appeals must be made to the same agency that issued the rating, part of the evaluation is the readiness with which management responds to regulator criticism, and banks are warned that “examiners have long memories.”
Substitutes for notice-and-comment rulemaking. The lack of transparency and tensions with the APA in the requirements of Comprehensive Capital Analysis and Review is now being addressed by the Federal Reserve. It is, however, just one example of many where guidance is used as a substitute for notice-and-comment rulemaking. The living wills guidance is also instructive. Some of the guidance is public and some remains confidential. All of it has been created without public notice and comment, including the sudden imposition of a new secret binding liquidity constraint on some global systemically important banks.
Uses and abuses of horizontal reviews. The banking regulators hide behind confidential supervisory information to perform secretive, unreviewable horizontal reviews of compliance with supervisory expectations, often moving the goal posts retroactively instead of prospectively. At the same time, bank organizations are largely prohibited from performing their own horizontal reviews of compliance with supervisory expectations lest they inadvertently disclose confidential supervisory information and risk criminal sanctions for doing so. How these horizontal reviews, especially when industry wide and not tailored to the institution, align with the APA is an open question.
Negotiations in applications. The long tradition of regulation by negotiation in the applications process is also a type of shadow regulatory system. Sometimes regulators strategically use delays to encourage nonpublic withdrawals of applications or impose conditions in the applications process that are unrelated to the application itself. As a result, the staff conducting negotiations during the application process wield policymaking power.
Examples are not complete. The actual list is much longer and the examples could be more specific. But the constraints of confidentiality constrain the public discussion. The key point is that the expansion of confidential supervisory information has created a realm where economic and social policy choices are being made in the shadows.
Uneasy Truce
The uneasy truce has become untenable and unstable. One canary in the coal mine is the increase in leaks of confidential supervisory information. The melody that canary is singing is changed societal mores about transparency in the digital age. But more important is the fact that the increase in confidential supervision makes it too difficult to hold the banking regulators accountable to the public for policy choices made in the shadow regulatory system.
Since the financial crisis, there has been a strong increase in leaks of confidential supervisory information, often by the bank regulators themselves. So far, each of the leaks below has been treated as a one-off situation. It is time to consider, however, whether they are a signal of the pressures felt by humans living in a digital society where there is a strong tilt toward transparency. Here is a list of recent examples that have made their way into the press: an examiner released 47.5 hours of tapes of private discussions about Goldman Sachs; a Federal Reserve of New York employee leaked the results of a management rating for a community bank to Goldman Sachs; someone leaked Deutsche Bank’s examination results to The Wall Street Journal; someone leaked Deutsche Bank’s state of play in negotiations, including a possible fine amount, with the Department of Justice; the FDIC rebuked personnel for downloading living wills files just before leaving for the private sector; someone leaked the results of 2015 living wills before they were sent to filers; someone leaked Wells Fargo’s CAMELS ratings and state of play on enforcement actions; and, during the financial crisis, a member of Congress posted Wachovia’s confidential exam report to a public website.
There is also an increase in the officially sanctioned publication of confidential supervisory information by the banking regulators themselves. The New York State Department of Financial Services recently used its power to release information in the public interest to make public its otherwise confidential ratings of the Bank of Tokyo-Mitsubishi UFJ. The decision about what is in the public interest and its timing is entirely in the hands of the regulators. (Note: A court has recently published CAMELS ratings. [Builders Bank v. Federal Dep. Ins. Corp., N.D. Ill., 15-cv-06044]). At the same time, banking organizations are silenced in the public arena when Congress or the media make statements that might otherwise be corrected except for the relevant information being considered confidential supervisory information, even when that information has been leaked by regulatory staff.
The lack of transparency and accountability makes it difficult to decipher when banking regulators are using the threat of confidential supervisory action as a sword against banking organizations or individuals and when they are using the confidentiality of supervisory information as a shield to preserve financial stability. There are a few telling examples that can be stated publicly. What is one to make, for example, of Henry Paulson admitting that he privately threatened to remove the management and board of Bank of America if it did not complete a merger with Merrill Lynch? He has since stated “[b]y referring to the Federal Reserve’s supervisory powers, I intended to deliver a strong message.” This message was not disclosed at the time. Operation Chokepoint, the penalty box, and many nonpublic examples involve similar threats of confidential supervisory actions. The increasing number of banks requesting to strengthen their ability to appeal examination results reflects the sense that confidential supervision can look like a weapon when it is shrouded in secrecy.
The shadow regulatory system leads to another problem. Many have criticized the banking supervisors for not being hard enough on the banking sector after the financial crisis. This criticism may not be valid. How can Congress, scholars, and others assess whether there is cognitive capture (a well-established term in the field of economics to describe when the regulator begins to think like the regulated industry) or whether banking supervisors are being too tough or too easy on the banking entities they regulate and supervise if what is known publicly is only the tip of the iceberg? The public debate and scholarship are deformed when they know only part of the story.
Thankfully, there is a growing concern, including at the Federal Reserve, that the balance is askew. Vice Chair Quarles has suggested that increasing transparency and rethinking supervision are high on his agenda, characterizing transparency as a “necessary precondition to the core democratic ideal of government accountability – the governed have a right to know the rules imposed on them by the government.” (Note: Quarles, supra note 39, at 3; see also Metzger, supra note 24, at 130). This thoughtful relook will also align more with constitutional norms and the APA to “produce better regulation.” (Note: Baer Testimony, supra note 40; see also Quarles, supra note 39, at 5).
What Policy Purpose Is Being Served?
We can improve transparency, accountability, and the rule of law by narrowing the scope of the shadow regulatory system to the core minimum necessary to achieve its policy goals. For important policy choices affecting economic and social conditions, the tilt should be toward transparency. Going forward, three key questions ought to be asked about confidential supervision and discretionary secret lore. Each one of these questions would involve a new way of thinking.
- Why is this topic being treated confidentially? Banking regulators and banking organizations should begin to ask themselves why a certain topic is being treated as confidential. There should be a tight link to financial stability and the need for candid conversations. There should be a serious re-examination, from first principles, of how the obligations of the securities laws and confidential supervision interact. One example is to ask why banks shouldn’t have the option to make their CAMELS and other supervisory ratings public. After all, the results of Community Reinvestment Act examinations have been public since 1990.
- Who or what is being protected by the confidentiality? Sometimes confidentiality protects the supervisor and shields its actions from public scrutiny. Why is part of the living wills guidance public and part of it private? How is it that the confidential penalty box constraints on banking organizations can exist for years? (Note: Quarles Transcript, supra note 41 (statement by Greg Baer)). Do CAMELS ratings really judge individual institutions or do they follow the trends of the business cycle? Some of these examples are areas where the supervisor knows that its actions would be controversial to the public and so a confidential route is chosen without much forethought. If confidentiality is chosen to protect the supervisor or a policy choice from public scrutiny, it is not appropriate.
- Why is this policy choice or regulatory interpretation being made under the rule of discretion rather than the rule of law? Access to the banking sector via grants of bank charters is at the complete discretion of banking agencies, including the timing of responses and the ability to encourage often silent withdrawals. Why is that discretion not cabined? Why did Operation Chokepoint begin as a confidential element of examinations? Why did the leveraged lending guidance start as confidential letters? In an era of increased transparency and accountability, policy choices that have an impact on access to deposit and payment services, credit allocation, and investment in the banking sector – that is, on jobs and growth – should be open, not secret. We are far away from examining the quality of a bank’s loans or the amount of cash it has in the vault to protect against the risk of bank runs as in the traditional bank examination context.
We should not jettison confidential supervision, but we ought to reform it. Following in the footsteps of economist Hyman Minsky, we should aim for better, not perfect. The public discourse will change and for the better.
