OFR and UM Event on Activity and Entity-Based Regulation

U.S. Office of Financial Research (OFR) and University of Michigan’s Center on Finance, Law, and Policy

“Function and Firms: Using Activity and Entity-based Regulation to Strengthen the Financial System”

Thursday, November 15, 2018

 Key Topics & Takeaways

• Keynote Address
  • FDIC Chairman Jelena McWilliams questioned if the rise of non-bank companies offering new products and services will drive banking activities out of the traditional banks, and how this might change risks in the financial system. Regarding regulatory coordination, she noted the FDIC is part of multiple international committees and working groups and that there is “a lot to be done” internationally, as there are jurisdictional issues driven by host country regimes.
• Panel 1: How Should Our Existing Regulatory Structure Be Applied to Support an Activities-Based Approach?
  • University of Michigan’s Jeremy Kress argued that moving to an activities-based approach “is a mistake,” as an entity-based model is also needed because non-banks were the first to fail in the crisis. SEC Commissioner Robert Jackson stated that an activities-based approach “makes a lot of sense” but is not a substitute for the entity-based model and that both are needed.
• Panel 2: What is the Role of Firm-Based Regulation?
  • Blackrock’s Joanne Medero discussed the differences between asset managers and banks and other financial institutions. She explained that designating asset managers does not reduce systemic risk, because given the agency model, the risk would just move from one entity to another.
• Panel 3: Do Regulators Have the Data They Need to Promote Financial Stability?
  • Davis Polk’s Annette Nazareth questioned whether regulators have the data they need and how they should take stock of what the industry has “before asking for more stuff.” Nazareth and JPMorgan Chase’s Debra Stone discussed regulatory coordination and standardizing reporting, which would help the industry by not having duplicate efforts domestically and internationally.

Keynote Address

Jelena McWilliams, Chairman, FDIC

In her remarks, McWilliams noted that the current U.S. regulatory model focuses on the legal entity but that sometimes there are similar products and services offered, raising the idea of regulation being focused on activities. She continued that the traditional role of banks is evolving, changing due to legislation and regulation, questioning if the rise of non-bank companies offering new products and services will drive banking activities out of the traditional banks, and how this might change risks in the financial system. McWilliams noted that this shift to non-banks is not new, adding that competition spurs innovation. She explained that through regulatory reform, banks are able to make new products, which levels the playing field but sometimes results in additional risk to the system. She questioned how regulators will be able to ensure customers are treated fairly if transactions, such as those through payment platforms, are outside of the regulator’s jurisdiction.

Consumer Protection

When asked how the FDIC can enforce consumer protection at financial institutions, McWilliams replied that the FDIC has enforcement mechanisms and can go after companies for misconduct, as well as refer them to the Department of Justice for further review and enforcement.

McWilliams was then asked how the FDIC deals with derivative exposures that are difficult to capture and may endanger deposits, to which she explained that the FDIC is not the main regulator for the four largest banks, and that they are referred to the Federal Reserve (Fed) and Office of the Comptroller of the Currency (OCC) on primary supervision. However, she noted that the FDIC would step in if there were risks to the deposit insurance fund and take action if other agencies do not.

Resolution Planning

An audience member noted that the FDIC has been engaged in resolution planning for years and that a number of financial institutions and large insurance conglomerates will no longer be subject to such planning, asking how the FDIC will be able to “clean up the mess.” McWilliams replied that the FDIC specifically looks for exposure to the deposit insurance fund, and that for those large non-banks that are no longer subject to federal supervision, there will be coordination with state regulators to ensure there is an appropriate regulatory framework in place. She continued that financial technology (fintech) are allowing companies to shift activities in new ways but that regulators are “slow to catch up” to the new technology, stressing the need for the FDIC to work together with fintech developments.

Central Counterparties (CCPs) and Non-Bank Resolution

McWilliams explained that the FDIC has limited authority over how CCPs and non-banks are regulated, but that third-party service providers look at servicers, vendors and counterparties. She continued that under Dodd-Frank’s Title II, CCPs must be resolved by collaborating with other regulators and that the FDIC has weekly discussions with the OCC, Fed, state regulators, Securities and Exchange Commission (SEC) and Commodity Futures Trading Commission (CFTC). McWilliams noted that there is a point when a banking entity has too much CCP exposure and that the FDIC would step in at that point to see what the market regulator would do to address the problem, and that if the FDIC is not happy with their response there are other ways to deal with it, such as the Orderly Liquidation Authority (OLA).

When asked about the ability to resolve non-banks that have not previously been subject to the supervision of other regulators and have no living will, McWilliams replied that mechanisms have been put into place to resolve banks and non-banks efficiently. Regarding how well state regulators are prepared to supervise and navigate the resolution mechanism, she pointed out that Dodd-Frank’s preferred methodology is bankruptcy and acknowledged her hope that the courts are able to handle it. She added that if such an entity poses a systemic risk to the overall financial system, she hopes they would have been designated by the Financial Stability Oversight Council (FSOC) prior to such an event.

Fintech Startups

An audience member noted the many ways a fintech startup company could access the regulatory financial system, such as the OCC’s fintech charter if they do not take deposits or through an industrial loan company (ILC) if they do take deposits, asking which option McWilliams would prefer for a fintech startup. She replied that while she has no preferred choice, factors include their business profile and risk model. McWilliams explained that if fintech startups apply for an ILC charter, certain difficult requirements must be met to be approved for the charter, and the threshold is not low. She added that since there has not been an ILC charter recently, she would have to carefully review to see the exposure to the deposit insurance fund and consumers.

Regulatory Coordination

McWilliams noted that the FDIC is part of multiple international committees and working groups and that there is “a lot to be done” internationally, as there are jurisdictional issues driven by host country regimes. She continued that while it is difficult negotiating with international counterparts, they have been working “rather well” together in her five months at the FDIC.  Regarding coordination with other U.S. agencies, McWilliams explained that while they do joint exams with other banking regulators and state authorities, they do not conduct joint exams with market regulators, and that there is no effort to combine the market and banking regulatory exams into a joint exam as they look at different things.

Panel 1: How Should Our Existing Regulatory Structure Be Applied to Support an Activities-Based Approach?

In the first panel, Treasury’s Brent McIntosh explained that FSOC has been making entity-based designations but that it has been criticized for being a “blunt instrument” with no off-ramp and is a “competitive inequity” between similarly situated firms. He explained that in the November 17, 2017 Report to the President, Treasury suggested moving from entity-based designation to activities-based, and the panelists discussed their views on each approach.

University of Michigan’s Jeremy Kress stated that the move to an activities-based approach “is a mistake,” as an entity-based model is also needed because non-banks were the first to fail during the financial crisis. Stanford University’s Anat Admati stressed the need to start with the largest institutions and follow their activities and exposures “in the shadows” and that “too big to fail is still here.” Davis Polk’s Margaret Tahyar stated that the U.S. regulatory structure “is just nuts,” questioning why there are so many regulators and why agencies like the SEC and CFTC are not combined. She discussed the concept of having a single systemic risk regulator, noting that FSOC originally had much more power but was weakened throughout the changes made before Dodd-Frank’s final passage. Regarding fintech, Tahyar questioned why the U.S. is the only country thinking of having state-by-state regulation while other countries are looking at it at the national level, adding that the U.S. system is not fit for digital transformation. SEC Commissioner Robert Jackson stated that an activities-based approach “makes a lot of sense” but is not a substitute for the entity-based model. He explained that the entity-based approach is needed because while activities lead to risk, the government does not have perfect knowledge of them and the market will move faster than the government and regulators can. Jackson continued that the activities-based approach is still an important part of the model but that it is “hard to have by itself.”

Identifying and Monitoring Systemic Risk

Tahyar suggested that OFR “up its game” and help FSOC by having “better data and better studies,” adding that it may require OFR having more resources. She continued that the OFR study on asset managers was “criticized” because it did not rely on primary regulators. Jackson stated that some of the data has not been used as much as it could be by OFR and others to understand where risk may lie. Kress noted that since Treasury released their report on shifting to an activities-based approach, there has been no word on how FSOC will identify systemically risky activities, explaining that it “will be hard.” He continued that in the coming months industry will be looking to FSOC to propose a formal activities-based approach and what the framework will look like.

Role of Data

Jackson explained that he has been “encouraged” by SEC Chairman Jay Clayton by what he has shared about FSOC’s outreach with primary regulators and how they are doing a good job sharing data. He added that FSOC relies on the primary regulators to keep the data safe and ensure it is used in a way that “makes sense.”

In response to a question about FR Y-15, the type of data used to set the global systemically important bank (GSIB) surcharge and whether it could benefit FSOC in designating, Tahyar replied that it is an interesting idea but that currently Stage 1 metrics rely on public data, whereas FR Y-15 is not public.

Activities-Based Approach Execution

Tahyar noted that since the financial crisis, there are now resolution plans and living wills for the largest institutions that have now moved to single point of entry (SPOE), in addition to total loss absorbing capital (TLAC) and increased capital. She noted that since there are currently no non-bank designations, there are no living wills for them.

Panel 2: What is the Role of Firm-Based Regulation?

University of Minnesota’s Dan Schwarcz focused much of his remarks on insurance entities, arguing that insurance-focused financial firms can be systemically significant and state-based insurance regulation is not focused on systemic risk. He explained that he believes the SIFI designation regime effectively supplements the state-based insurance regulatory regime for addressing systemic risk. Schwarcz also discussed the activities-based FSOC approach, arguing that it is limited in capacity to deter and limit systemic risk, particularly in insurance.
Blackrock’s Joanne Medero discussed the differences between asset managers, banks and other financial institutions, reiterating that asset managers have an agency model and do not have access to central bank liquidity. She explained that designating asset managers does not reduce systemic risk, because given the agency model, the risk would just move from one entity to another. Medero noted that while Congress thought a bank’s balance sheet size could be an indicator for systemic risk, since asset managers are structured differently, this does not translate to assets under management indicating systemic risk. According to Medero, this analysis supports the pivot to focusing on products and activities. She stressed that U.S. regulators are collecting “substantial” data on products and activities, and suggested that the following issues are worth evaluating from a systemic risk perspective: the future of the London Inter-Bank Offered Rate (LIBOR), cybersecurity, bondholder rights, pension underfunding, Brexit and CCPs.
Columbia University’s Kathryn Judge discussed the “crisis-era shadow banking system,” and argued that entity designation could make sense because it would give the government a basic understanding of an entity, which would be helpful should the government have to step in and provide support. She agreed with Medero’s point that asset managers should be regulated differently than banks due to their different structure, however she also noted the migration of banking activities to the asset management sector since the financial crisis.

Identifying Activities

In response to a question as to what activities should be identified in a products and activities systemic risk review, Schwarcz replied that it is “almost impossible to answer,” but suggested that regulators could begin by looking for things that create “run-ability.” He continued that activities-based approaches are great “if you can get them right” but since that is difficult to do, added that it should only be one part of systemic risk regulation. Medero discussed the difficulty with predicting the next financial crisis but argued that regulators can use their previous experience to know when an issue deserves further consideration and review.

Panel 3: Do Regulators Have the Data They Need to Promote Financial Stability?

University of Michigan’s HV Jagadish gave an overview of the types of data regulators need: global data for model building that is comprehensive, multi-source, multi-modal, large, diverse and incomplete; and local entity-specific data for regulatory actions that is still large and diverse, but not as much as global data, and is hopefully not incomplete. Regarding privacy, Jagadish noted that companies care about proprietary data and individuals care about privacy, so data may be provided to regulators by a third party with “data fiduciary” responsibilities.

The Federal Reserve’s Andreas Lehnert questioned why information deficiencies have emerged in the sector, as anonymized high frequency data on random sets of households is available without any lag and there are increases in data in other segments of the economy. Lehnert stated that firms have improved their risk management and risk measurement systems, and that investors now have access to information through stress tests and other efforts. Looking ahead, he explained that data, automation, and information technology “is the future” and that high frequency trading “is here to stay,” adding that data and information “will only become more important going forward.”

Davis Polk’s Annette Nazareth questioned whether regulators have the data they need and how they should take stock of what the industry has “before asking for more stuff.” She noted that OFR has a goal to standardize reporting, which would help the industry by not having duplicate efforts domestically and internationally, though she questioned how much progress has been made. Nazareth discussed the Consolidated Audit Trail (CAT), noting that FINRA could have started with the “huge amounts of data integrated across the market” as about 90 percent was already there, and how it is now delayed because it does not have the capacity to process transactions that are required. She stressed the need for FSOC to bring functional experts on certain matters into analyses and noted the competition between FSOC and members, adding that there has to be a consensus among regulators. She concluded that data sharing among agencies is essential but that it is natural for them to ensure others “don’t play too much in their sandbox.”

JPMorgan Chase’s Debra Stone discussed some of the ways the industry is falling short, including issues around the quality and coherence of data; a lack of coordination among regulators; who is obligated to report, what is reportable, the required formats for reporting and what data will be made public; and the cost and lack of efficiency for firms to build multiple infrastructures for the different regulatory reporting formats, stressing the need for regulatory coordination. She discussed some of the gaps identified after the financial crisis and how “critical” it is for everyone to be able to look at the same information and for data to be gathered quickly and efficiently.

Legal Entity Identifier (LEI)

Stone recommended the LEI be globally mandated to adopt and monitor at the global standard level like capital requirements and resolution regimes. OFR’s Acting Director Ken Phelan noted that OFR came out with a proposed rulemaking for repos that included the requirement for LEI and that there will be a final rule later this year or early next year.

Data Standardization

Jagadish discussed how data standardization “moves slow but technology moves fast,” stressing the need to be able to react quickly and share things like best practices. Nazareth and Stone discussed the Freedom of Information Act (FOIA) and questioned whether exemptions may translate when data is shared and how supervisory data would be categorized versus other commercially-sensitive data being classified in different ways.

For additional information on this event, click here.