Regulators need more tools to keep system safe: Boston Fed chief

Register now

BOSTON — Eric Rosengren, who heads the Federal Reserve Bank of Boston, has gone on record as praising the advantages of central banks having a clear view inside financial services firms.

"That can only occur when you actually have supervisors from the central bank that are actively engaged in the largest banks, which is the case in the United States,” Rosengren said in a recent interview on the sidelines of a stress test conference hosted by the Boston Fed.

But Rosengren, who took the reins of the bank in 2007 just before the financial crisis, says U.S. regulators' knowledge of firms' inner workings could run deeper.

While the United States has some regulatory advantages over other countries, Rosengren says the Fed’s macroprudential toolkit lacks some of the authorities that other countries entrust to their regulators — particularly the extension of prudential authorities beyond the banking sector.

“We don't have all the macroprudential tools that we might need in the future,” Rosengren said. “Ideally, financial stability is worried about risks wherever they are in the financial system, not just at the largest banks. So the coverage in the United States is quite limited.”

What follows is a transcript of the interview, which has been edited for length.

How would you assess the Fed’s stress testing program compared to stress testing programs elsewhere?

ERIC ROSENGREN: The stress test in the United States, I think, is the gold standard relative to most international stress tests. We had an advantage that, during the financial crisis, we took the stress test quite seriously. We also had the advantage, even though it was politically unpopular, to have the [Troubled Asset Relief Program] funding that would help recapitalize banks. The result is that our banks became recapitalized much more quickly than banks in other parts of the world.

That had the benefit that, rather than decreasing lending, it had an impact on the banks [of] preserving capital as quickly as they could, get the government money out, but didn't do it by shrinking their institutions and cutting back on lending. That started in the financial crisis, but it also set a standard for reporting publicly on the results of the stress tests, and I think the public at large has a high degree of confidence in what the stress test is designed to do. That's not as true in some other parts of the world where there have been instances where stress tests occurred, and shortly thereafter banks were undercapitalized and need to recapitalize themselves.

We've yet to have it completely tested by another recession. So I think it's a little premature to know how successful it is until we've weathered a recession, and it's not that there can't be improvements to a stress test — this conference is going to highlight some of the improvements that at least the paper givers and discussions think are important.

Do you think the evolution of the U.S. stress testing program is going in the right direction?

I think there has been a general improvement. Our largest banks are very well-capitalized. I think that's a very positive aspect of the stress test. I think [the Fed has] a good opportunity to reassess some elements of the stress test. There have been some rule changes, [such as] some large banks only have to do a stress test every other year. My own personal view would be that ... I'm not sure the cost-benefit [trade-off] is such that I wouldn't have made a slightly different determination. But I think by and large we're in the right place for the stress test.

You’ve written about the need for the Fed to utilize both monetary and macroprudential tools to respond to future shocks. Do the Fed’s regulatory and monetary policy functions talk to each other, or are they considered separately?

During the financial crisis I think we got a lot of bank supervision information, and I think it was very useful. I was on both the research department and the supervision side during my career. I think understanding both what large banks do — which large banks are involved and what kinds of activities and which large banks were having particularly acute problems — was very useful insight during the crisis. So I do think there is good communication.

The [Fed's] Board of Governors is responsible for setting the regulations. The governors are responsible for the supervision and the regulation; they also all are voting members at every single meeting. So it's not as if there's a wall between what they learn when they get a briefing from a group of bank supervisors and when they get a briefing from a group of economists about the economy — they can put the two together. There's a lot of information that you can get from bank balance sheets about what's growing quickly what's not growing quickly, what the terms of credit are, how easy lending standards are. So I think there is a lot of advantages of a central bank for having a clear vision of what's happening inside the banks, and that can only occur when you actually have supervisors from the central bank that are actively engaged in the largest banks, which is the case in the United States.

Former Fed Chair Janet Yellen said earlier this year that the U.S. lacks important macroprudential tools that other countries have. Do you agree?

I agree with Janet that we don't have all the macroprudential tools that we might need in the future. The Bank of England is a good example of a situation where they have far broader powers to think about financial stability. So when you're thinking about financial stability in the United States, the primary financial stability tool is the countercyclical capital buffer.

In other countries there are a lot of other tools. In particular, right now corporate leverage has gotten quite high, but we don't have the authority to broadly change how leveraged firms can be. If individuals take on a lot of debt, that debt could be in the banking sector [or] it might not be in the banking sector. Same with the corporate debt — it may go through banks, it may not go through banks.

Ideally, financial stability is worried about risks wherever they are in the financial system, not just at the largest banks. So the coverage in the United States is quite limited. A good example is, for example, commercial real estate. We do stress commercial real estate in the stress test. The largest banks don't have ... particularly large exposure to commercial real estate relative to their capital as a result of that. Some of that's moved to smaller banks that are not subject to the stress tests. That's one undesirable feature, if it ends up in a lot of small banks and they all have trouble in the next recession.

But it's also migrated to other kinds of financial intermediaries who are doing the commercial real estate lending. And certainly with financial services evolving as quickly as it is, there are a lot of fintech players that are getting involved in commercial real estate, mortgages to individuals — a wide variety of activities that traditionally had been done by banks. To the extent that we're focused only on the banks, you're going to expect over time that some of those risks are going to migrate to the rest of the economy.

The Fed board decided earlier this year not to raise the countercyclical capital buffer. Did you agree with that decision?

Fed presidents don't get to actually set the countercyclical capital buffer. ... So this opinion is from the stands, not involved in the actual decision-making. My own personal view and I've given plenty of speeches on this is that I think that it would be advantageous to have a countercyclical capital buffer during normal times that's not zero.

Many other countries have countercyclical capital buffers that have the feature that they get built up in good times. You don't have to be at frothy times to increase the countercyclical capital buffer. So I don't think the way we've structured it is the way that we ideally would, and a lot of people focus on the countercyclical capital buffer in terms of raising capital during the good times, but what's critical is that you can release the capital during the bad times so that the financial institution doesn't react by shrinking its assets.

If you look around the world, particularly in northern Europe, they have a somewhat different standard to determining the countercyclical capital buffer. It's country by country and there are a lot of differences across countries, but most of them now have set the countercyclical capital buffer at non-zero rates. I think, given the stage of the cycle we're at looking at, where financial pricing is in a wide variety of financial assets, that it would be advisable at this stage if we actually had a positive buffer rather than a zero buffer.

Is part of the problem with the CCyB that it only applies to the largest banks? Should it be broadened to include smaller institutions?

Well, the issue is that those large banks actually control a very high percentage of bank assets and bank lending. So while it is true that, for example, in commercial real estate they've reduced their exposure, there [are] plenty of other areas where they have fairly significant concentrations. Because they have roughly three-quarters of the assets and loans of the nation ... if they cut back it really does make a big difference for the economy.

I think you could easily have the view that the countercyclical capital buffer should extend beyond those institutions. But I think for those institutions ... we want them to have the flexibility to have the capital released if we have an economic downturn.

What do you see as the biggest sources of risk to the economy?

There are a number of areas that I would focus on. One, corporate leverage has been going up so I'm worried about the number of firms that are highly levered. I don't think it's going to be a triggering event for an economic downturn, but I do think it'll increase the amplitude of the downturn. That is it will make the recession worse than it otherwise would be. So a levered firm is going to have to cut back more dramatically than a firm that has a much bigger capital buffer, in the same way that a bank that has a lot of capital buffer isn't going to have to cut back on lending nearly as much. So the same thing applies to individual firms — how is a leverage firm going to react? It's probably going to do more layoffs. It may have to substantially change its business model as a result with the amount of leverage that it's taken on. So I think that's certainly one significant risk.

A second risk that I would highlight as commercial real estate something I've been worried about for some time — including, increasingly, the movement of commercial real estate to shared spaces with different kinds of institutional arrangements than we've seen in previous recessions.

Some banks have complained about "matters requiring attention" acting as quasi-enforcement actions. Some Fed board members have agreed to review supervisory practices. Is an adjustment needed, or is the concern overblown?

The MRA is a supervisory tool — I think they are an important supervisory tool. I think they should be used judiciously. I think there was a period where the list of MRAs became so daunting that institutions had a difficult time prioritizing what was important and what was not important. I think that we should be focused on those issues that [constitute a] substantial concern about the safety and soundness of the institution. But I think it's a perfectly legitimate supervisory tool that we should use judiciously, but that it's an important tool that we actually need to have.

The MRAs is are still based on the regulations that the banks are subject to, and so they're not willy-nilly just imposing MRAs. They're looking at the rules and regulations related to safety and soundness of the institution and saying, "If this concern isn't addressed, we're worried that the safety and soundness of the institution over time could be impaired." ... The question is whether we've set the threshold appropriately. And I think that's a legitimate argument about where we should be setting the threshold, and how many MRAs should we be expecting. ... We should focus on those that we think are the greatest danger to the financial health of the institution.

Last month Facebook announced its intention to create its own cryptocurrency, Libra. How far away are we from central banks having to compete with digital currencies?

There are central banks, including our own central bank, that are actively researching distributed ledger technology. And not necessarily just cryptocurrency, but looking at financial innovation and how that should affect what we're doing. There are some central banks, for example the central bank in Sweden, has had a program where they've tried to make very few cash transactions. So almost everything's electronic in Sweden now. And the next step, my guess is to thinking about a situation where they could have a distributed ledger that you don't even have to have the same kind of standard electronic transactions that have normally occurred. So there's innovation occurring in central banks around the world.

I don't think it's going to impede our ability to do monetary policy. I think it may change the nature of some of the aspects of what we do. In the end, any transaction has to clear through the central bank, because it's reserves that ends up moving from borrower to lender from the two sides of the transaction. So you have to have a finality of transactions that can only occur in the central bank's books. So I'm not worried about whether this kind of replaces central banks.

I do think that it's a technology that we need to continue to research and see if it's appropriate to use. I think we're a long way away from worrying about it completely disrupting, for example, currency holdings. But in other countries, they're actively discouraging currency holdings. Currency has actually been growing in the United States, it's not shrinking right now. So I think it's certainly something to closely monitor. I'm not sure it's much of a threat to the conduct of monetary policy, but it may change the way many consumers and businesses do transactions.

Might Libra or some other cryptocurrency get more traction in another country with a weaker central bank currency?

One of the worries you have to have is cybersecurity. If I'm holding a dollar bill, I have no cybersecurity concern. If I hold whatever that other currency is, there have been lots of cyber intrusions that have wiped out accounts for individual account holders and for clearinghouses. So those cyber concerns already make a distinction between just holding actual cash and having some of these other alternatives. I do think they have to get around that issue, and because they have an alternative right now which is to hold for example U.S. dollars, and the reason U.S. dollars have been growing is because they still seem to be one of the major ways that you hedge against an unstable currency. ... Cryptocurrencies over time may be an alternative, but I think we're a pretty long way away from them supplanting U.S. dollars.

For reprint and licensing requests for this article, click here.
Stress tests Libra Minimum capital requirements Monetary policy Cryptocurrencies Nonbank Leveraged loans Eric Rosengren Federal Reserve Bank of Boston Federal Reserve
MORE FROM AMERICAN BANKER