Richard Berner’s distinguished career is highlighted by service in business, government and academia. He is currently Executive-in-Residence and Adjunct Professor at the Center for Global Economy and Business at the NYU Stern School of Business. From 2013 until 2017, Berner served as the first Director of the Office of Financial Research (OFR), an agency established by the Dodd-Frank Act to support the Financial Stability Oversight Council, the Council’s member organizations, and the public. He was Counselor to Treasury Secretary Geithner from April 2011 to 2013. Berner was a Managing Director and Chief U.S. Economist at Morgan Stanley from 1999 to 2011; he became Co-Head of Global Economics in 2007. He began his career in 1972 at the Board of Governors of the Federal Reserve System.
Berner’s speech at the 2018 Macro Financial Modeling (MFM) group’s winter meeting, “Challenges for Macro Financial Modeling,” provoked many questions from conference participants. In that spirit of inquiry, we offer the following discussion on his views relating to modeling, data, regulation, and other issues facing the financial sector, regulators, and policymakers.
In your remarks, you recount the many ways that economists, policymakers and market players were surprised by the breadth and depth of the financial crisis. Much has been learned since, but how confident are you that we are prepared for the next financial challenge?
The answer to that question has several dimensions. First, of course, we do not know what the next financial challenge is going to be. To assess their potential impact on the economy, we have to separate the potential sources of those challenges and the way they can affect various parts of the financial system. And then we have to think about their consequences, whether positive or adverse for the economy.
I’ll give you an example: We ought to expect that innovation within our financial system and within our economy will be constant. Recently, financial innovation has accelerated, and not just because of new financial products, but also because of new technologies. And those new technologies convey lots of benefits for households, businesses and potentially even governments. However, they may also create risks we don’t see. We do know about so-called operational risks, including failures of that technology. Cyber security incidents are a subset of those; they can have devastating effects if they are severe and widespread.
So, we know something about those risks, but we need to know a lot more about them, and a lot more about how to make our financial infrastructure more resilient. I think that kind of challenge is at the top of the list, because while cyber shocks can be accidental, there are also bad actors out there trying to subvert our financial systems. This is an emerging risk that has recently come more sharply into focus.
You note that “systemic risk,” which is the motivator for so much post-crisis regulation, is an imprecise term. Could you describe more what you mean by such imprecision, and how could our lack of understanding inform our policymaking?
One of the co-sponsors of the MFM conference, Lars Peter Hansen, wrote an eloquent paper on the subject [“Challenges in Identifying and Measuring Systemic Risk”], which I cited in my talk. The term systemic risk in an abstract sense correctly conveys the idea that you are looking at risk across the financial system. However, the financial system is complicated and constantly changing. So, the connotation from the term “systemic risk,” that there may be one metric summarizing financial-system risk, cannot reveal nuance; it hides important, complex, and dynamic information. And you can’t operationalize the idea of systemic risk to arrive at any actionable conclusion. Hansen and I are in complete agreement on these problems.
Instead, I much prefer using the notion of threats to financial stability. Describing threats helps direct you to unpack where the different problems in the system might arise. Doing so avoids painting the entire financial system with the same brush, which is not possible.
For example, take the size of financial institutions. Some people believe that if an institution is big it is necessarily systemically risky. That’s only one factor. Size and risk may be correlated, but there is not necessarily a causal relationship. If you’ve got a financial institution that’s large but its business model is plain vanilla—they accept deposits and make loans—it may still be risky on some levels, but it may well be less risky than a smaller institution with a more complex business model. So, the Basel Committee on Banking Supervision developed a series of factors to identify so-called Systemically Important Financial Institutions, or SIFIs, based on their risk, rather than size alone.
Relatedly, you describe the importance of modeling in your talk, but also acknowledge that current models must improve, and you list a number of challenges and questions facing researchers today. How do we reconcile the uncertainty that comes with economic models, no matter how good they are, with reality (especially the realities faced by policymakers)? Put another way, how far do we trust our models?
Models are very useful tools to inform our judgement, to structure, discipline, and organize our thinking, and to help us assess the effects of changed circumstances on the economy and the financial system. But they are no substitute for judgement. Nor will they ever completely capture reality. We can trust our models if we constantly validate and verify them, and constantly ask whether the results stand up to review.
As you note in your speech, with so much data available these days, it is sometimes hard to imagine that we may not have enough. What are the data gaps that exist when it comes to analyzing financial turbulence in real time?
It would be hard to say what the single most important data gap might when monitoring financial stability. Partly that’s because we don’t know what all the vulnerabilities and risks are in the financial system. There’s a feedback loop here, in the sense that when we discover things that we didn’t know before, we realize that we didn’t know about them because we didn’t have the data.
One thing we do know is that what we call securities financing transactions in markets such as repurchase agreements, or repo, still carry risks. For example, if a broker-dealer were to default, and if its counterparties abruptly liquidate the collateral that the broker-dealer repoed to them, that could trigger a so-called fire sale that could possibly destabilize financial markets.
To understand how those markets work, and what makes them stronger, we need more information. That’s one of the reasons that, when I was at OFR, I focused on the data gaps in repo that still exist. Also, while data gaps are important, data quality is also key. If data across institutions or sectors aren’t standardized, then they can’t be compared or aggregated in a sensible way to assess risks across the financial system.
As a follow-up, do we get too beholden to data, too caught up in getting more without thinking about why and how it might be used?
You’re spot on with that question. I tried to emphasize during my career in government that we should never collect data for collection’s sake. I tried to initiate a review of the extent of the reporting burden from regulatory data collections. If regulators collect data that they really didn’t use, or use well, then that wastes everybody’s effort. We need to review whether data we collect are still useful and whether they duplicate other efforts. [Berner gave a talk on this subject while serving as Director of OFR, “Reducing the Regulatory Reporting Burden.”]
Your speech refers to OFR’s approach to financial stability monitoring, which includes “macroeconomic, market, credit, funding and liquidity, solvency and leverage, and contagion risks—wherever they arise.” No small task. How would you assess OFR’s success in this regard? Are we getting there?
I think we’re getting there. Monitors like the OFR’s are tools. Tool building is not an end; it is a means to an end; an ingredient that helps you arrive at a judgment of where financial stability is overall. We must recognize that financial stability means that all parts of the financial system, across various institutions and across markets, are functioning even under stress, and that requires monitoring all those parts and their interaction.
The key insight in that regard over the past decade has been to analyze the financial system as a system. Shocks in one part of the system can spill over into other parts. That can create disruptions in more than one part of the system—all at the same time. And that’s why “wherever they arise” is so important, because people are engaged in certain activities that are riskier than others and can have broader impacts.
Another important aspect of this is regulatory arbitrage. That’s when you regulate one part of the system and the activity may migrate to less regulated parts of the system. That happens often, it’s endemic. It’s particularly problematic here in the United States because we have a fragmented regulatory system. Each regulatory agency has its own mission, mandate, and authorities, and they must respect them. Of course, regulators, particularly since the formation of the Financial Stability Oversight Council, pay a lot more attention to financial stability issues than they did before the crisis. But we haven’t sufficiently institutionalized financial stability as an important part of their mandate.
The OFR successfully took several important steps in the right direction in pursuing financial stability monitoring. I think what’s important to recognize is that OFR’s job was to inform the other financial regulators along these lines. So, this wasn’t something that the OFR could or did do by itself. The more regulators can work together, the better.
What does your experience at Morgan Stanley tell you about the resiliency of organizations and the value of good risk management? How much more resilient are our financial institutions today?
Financial firms today are more resilient because of the steps that the firms have taken, and the steps that policymakers have taken to make the system stronger. Risk management practices are much better now than in the past. The steps that policymakers have taken, which you might call the pillars or tent-poles of the new financial regulatory structure, have really made the system more resilient: capital regulation, liquidity regulation, making derivatives markets more transparent, using stress tests to determine weaknesses within institutions.
In addition, the creation of a resolution authority that allows for the orderly wind down of a large, complex and troubled financial institution is very important. Now, this process is still untested—we haven’t done this yet—but to have such a process in place that is credible and can shield the rest of the financial system, will be key to addressing possible future crises.
In my talk I describe shock absorbers and guard rails, and the latter term involves incentives; it is really important to get the incentives right so that firms will take the necessary steps to make themselves more resilient.
Many of our readers have certainly heard of the Dodd-Frank Act of 2010, but may know little about it. What is your broad assessment of the Dodd-Frank Act of 2010?
I think Dodd-Frank, which arose out of the crisis, did many important things to make the financial system stronger. Like any substantial piece of legislation, particularly one that touched virtually all parts of the financial system, it has some things that could be improved, and might even have things that are missing.
However, it’s important to remember that the financial system has evolved since Dodd-Frank was passed, and in ways that may require change. I think, broadly speaking, Dodd-Frank was legislation that enabled regulators to take the action that they needed to take, and gave them the authority to do so.
That’s a very broad assessment, but let’s take a look at one small piece of Dodd-Frank—the Volcker rule. I think all of us can stand up and salute the idea that, in principle, institutions shouldn’t use depositors’ money to take on very risky debts. However, when you get down to the details it is very hard to distinguish between proprietary trading and market making. It’s these details, once they’re put into place, that may require review.
So, it’s right to look at Dodd-Frank; we should make it better and reduce the burden where it’s overly burdensome. That’s going to involve lots of judgment, and it’s not going to be easy to do.