Can We Reduce Risk from the Shadow Banking System?

Q: What was the impetus behind your research on congruent financial regulations?

After the global financial crisis, we paid a lot of attention to the traditional banking system. This was funny and somewhat anomalous because while the global financial crisis was happening, it was fairly well understood that it was parts of the shadow banking system, the nontraditional banking system, that had been a very large accelerant, if not the primary cause, of the upheaval. But it’s the traditional banks that got the most attention, because it wasn’t until the traditional banks, the Citigroups and the Bank of Americas, got in trouble that people really noticed the financial crisis.

So the post-crisis regulatory changes were mostly things to beef up the traditional banking system. And those changes were largely successful. We had a very big COVID crisis, an enormous global recession, but did you hear about banks going out of business? The banks had been forced to get much safer over the previous 10 years, and they weathered the storm and have continued to weather the storm.

The bad news is that the things that were vulnerable before the global financial crisis outside of the traditional banking system were still vulnerable in the COVID crisis—and in some ways had become even more vulnerable.

It’s like we had the highway and then we had the service road. The highway is the traditional banking system. The service road is the shadow banking system. When there’s traffic on the highway, you get on the service road. We strengthened the infrastructure on the main road, put in more tolls, made it a little more expensive to drive on. And it held up very well. But that encouraged a lot of drivers off the main roads and onto the service roads.

In early days of COVID, we saw weird types of breakdowns in the financial system that got fixed by massive federal interventions. But sometimes those federal interventions were ultimately helping players who had not paid into the system to begin with. Ultimately when the government came in and spent an enormous amount of resources to fix the government bond system, a lot of the beneficiaries of that were hedge funds and money market mutual funds and other players in the system who hadn’t been paying any insurance beforehand to be kept safe.

So in part what we saw happen over the last 10 years since the GFC has been a continuation of trends that we saw beforehand, which is chunks of the financial system moving out of the traditional regulated space and into the shadows. When those parts of the system start to malfunction, it’s more difficult for the government to stabilize them.

So we did make the traditional banking system safer, but what’s left to do? My co-author Dan Tarullo was a senior official at the Federal Reserve in the post-GFC period, and as much as anybody in the world was responsible for making banks safer. But all through that time, he was concerned about the non-bank financial institutions, the shadow banking system. He and I had been talking about it for years and years. When we saw what the early days of COVID were like, and the parts of the non-bank financial system that were malfunctioning, we thought this might be a good time to talk about how you might fix that.

We gave ourselves the following constraint: we’re not going to get new laws. We don’t expect that post-COVID we’re going to generate the momentum necessary to get substantial new financial regulation through Congress. So what can we do with the authority that we already have under current existing law, to try to have a system that is more balanced, to try to have financial regulation that we call congruent? Which means not writing exactly the same rules for everybody, but rather trying to have rules in place that at least recognize you’ve got a main road and you’ve got a service road. You don’t want to have totally different laws for those things, but you can’t do exactly the same thing because the roads aren’t exactly the same. And so you want to do things that are congruent, even if they’re not identical.

An example of congruence is that banks have capital requirements. That means that if a bank has a certain asset on their balance sheet, the rules say they have to hold a certain amount of loss-absorbing capital, typically—best would be equity, but some forms of debt also count—against those assets as a function of how risky those assets are.

We said, well, there’s a similar animal that exists out there in market-based finance, which is called a haircut on a repo transaction or an initial margin on a futures or derivatives transaction on an exchange. And we should be using those analogs in a congruent way to capital requirements, so that it’s not the case that in one place you don’t have to set aside any capital and if you do the transaction using a different way, you have to set aside a lot.

We try to sketch out in the paper the actual legal authorities that exist that would enable you to force this congruence on the system—what agencies would have to do what types of things, and how might you go about starting a process of putting something in place which mechanically would impose this congruence across the system.

Q: You looked at a couple of case studies. How did you choose those and what did you learn?

We were looking for markets that, first, had changes in their institutional structure over the last 10 years that moved them towards non-bank financial institutions, at least in part driven by regulatory arbitrage—trying to get away from regulation. And second, places that appear to have had a malfunction during the COVID crisis.

The first one that we ended up focusing on was the government securities market. In the end, everything in there is government guaranteed. So why did liquidity go away in this market that everybody should have perceived as being very safe?

And then the other place was non-prime mortgage finance, which had malfunctioned in 2008. It’s remarkable how different it looked in 2020 than 2008, but yet how it was vulnerable for similar reasons, in that a lot of the activity had moved away from banks. The old-fashioned way of a bank writing a mortgage and holding it on its balance sheet, which left us 20 years ago, and before the global financial crisis had led to a whole other set of imbalances—those were totally gone. We didn’t have anything that looked anything like subprime, mortgage-backed securities. We had a completely different looking system. The government was very involved. But the institutions at the core of it were not banks.

Q: How have Treasury securities become part of the shadow banking system?

Historically the way that the government would sell treasury securities is that there are a couple of dozen primary dealers—that’s an official title that they would get from the U.S. Treasury. Most primary dealers are subsidiaries of big banks—say, Goldman Sachs Securities. I’m a subsidiary of Goldman Sachs, which is a bank holding company, and as a primary dealer, I am obligated to bid in all primary auctions of Treasury securities; usually I do that after finding out from my clients what they would like ultimately to own. I’m an intermediary; I’m like a wholesaler of U.S. government securities. And historically, even up until the global financial crisis, that’s essentially how that market worked. The U.S. government needs to sell securities. They sell them to Goldman Sachs in an auction. Goldman Sachs holds them on their balance sheet and gradually sells them off to their clients.

Post-global financial crisis, the capital rules at banks changed to make them safer, and one of the ways in which they changed was to make it so that it cost banks some scarce balance sheet space to perform this activity. In the old days they used to be able to perform this activity and effectively it didn’t really cost them any regulatory capital, but in the post-GFC world, every time they use their balance sheet for anything would bind some of their constraints. So banks said, “We don’t really want to do this as much as we used to do it.”

Think about how much the government is selling. It’s trillions of dollars a year of securities. Even though they’re safe securities, they have to get sold. You have to move them through the system, from a government balance sheet ultimately to being held by someone who’s an investor. And that process takes money along the way.

If the banks say, “We don’t really want to commit as much of our balance sheets to this,” others will come in and fill that void. And in one major case that we described in the paper, that void was filled by a combination of hedge funds, futures markets, and money market mutual funds, all working through different types of subsidiaries of the big banks, but in ways that the big banks didn’t need to actually commit as much of their balance sheet to. If you could figure out a way to commit less of your balance sheet for that activity, but still make some profit off of it, you’re going to try to do that—but someone needs to commit balance sheet. So ultimately where was it happening? Through some nexus of non-bank financial institutions.

In the early days of COVID, when liquidity began to dry up, people really wanted to have cash. Now, cash is an obligation from the government, as is government securities. But it has different uses, just like the dollars in your pocket have different uses than the zeros in your account at Bank of America.

When people said they really wanted to have plain old cash and not government securities, that placed a lot of pressure on the non-bank part of the system that had been financing it—and suddenly they just couldn’t do it. And it made this plumbing part of the system break down.

The way it was fixed, ultimately, was the willingness of the Federal Reserve to come in and say, “Fine, you want to give up your Treasuries? We will give you dollars for your Treasuries.” But if you think about who ultimately was benefiting from that, it was not people who were paying insurance into the system. It was hedge funds and money market mutual funds who had been the ones providing that capital who got rescued by the Federal Reserve.

Q: What is the larger meaning of this trend towards shadow banking?

Well, I think it’s a big deal, in part because of, as we have recognized over the last 20, 25 years, how tightly ingrained the financial system is with the broader real economic system. A malfunction in the financial system can have broad repercussions on the everyday lives of people in the country. Because of that, we have taken much tighter control over what was going on in finance. But if every time you try to fix it by regulating what happens in these public spaces that activity moves elsewhere, you haven’t really fixed any of the problems that we observed growing over the last 25 years. We’ve just swept them under the rug.

We are vulnerable to weird disruptions from esoteric things in finance, affecting your ability to run your corner business—for example, because you were financing your inventory through a credit line at the bank, which the bank has to pull because the bank suddenly lost all this money in total return swaps with hedge funds that were replicating the activity the bank used to be able to do on balance sheet. Now you can’t keep any inventory in stock. The people who shop at your store can’t get it. You have to lay off your employees. And all of that happens, not by any design, not because the people who are overseeing the whole system thinks this is an appropriate response to the trouble we’re in, but because they lost sight of where this activity was happening and they lost control over the levers that could keep that part of the system running.

Q: Do we know how much of the COVID economic downturn was related to this type of activity?

I think that in the absence of the massive response that we saw in the United States from the Federal Reserve and in Europe from the ECB, it would have been an enormous problem. In the end, it was not. The enormous problem was the actual suspension of a lot of economic activity, but there was not a follow-on problem from financial markets—in part, because there was a really muscular response right away, in March and April 2020.

The wakeup call was really more that, look, we just had a non-financial shock. COVID was a shock to the real economy. It was like we all woke up in the morning and decided, I don’t want to consume or produce. So it wasn’t something that grew out of the financial system or that we could blame the financial system for, but even in that case, the financial system almost immediately started to implode. A very rapid response kept that from exacerbating the underlying real economic shock, so I would not say that it caused the problem or made things worse. It did however, signal to us a new type of vulnerability we didn’t know about before. Which, next time, could be the source of the problem instead of just a possible link in the chain making it worse.

Q: Should your average person should feel secure that the government will act?

This was a problem where they were able to act quickly and they were able to fix, but the next one may not be that way.

What would you say if the following happened? A fire started in some tall buildings, and the fire department went and put it out. They succeeded in putting this fire out. But in the course of putting this fire out, they noticed many structural weaknesses. If the fire had been of a slightly different type, or if they hadn’t been lucky in a specific way, the whole building would have come down and lots of people would have died.

Do you take comfort from the fact that the fire department succeeded and people didn’t die this time? Or do you get a little nervous because of what you learned? I would say it’s the latter. We shouldn’t be complacent. We should say, “Wow, we just learned about a lot of vulnerabilities. Let’s fix them.”

Q: Is the regulatory system always a step behind, because you can’t keep up with the innovation of the financial system?

There will always be new problems. The regulatory system will always be a little bit behind. That’s not a reason to be a lot behind. You don’t want the perfect to be the enemy of the good.

It’s often the last refuge of the scoundrel who does not want there to be any regulation at all is to say, “Oh, well, you’re never going to be able to catch up to them.” So what? That means we shouldn’t bother at all? In other words, if you think criminals will always be out ahead of the police, figuring out ways to break the laws, that’s not really a reason to stop enforcing the laws that we have. It’s just a recognition that we’ll never have perfect enforcement.

Q: What should the average person be thinking about our financial system? Should they be taking any different actions than they would otherwise?

The warning that I give to my friends and relatives who ask me for financial advice is, if something is not a bank, it’s not a bank. Don’t think that just because your friend had learned about this particular financial product and thinks it’s really safe, that somehow that means that it is. The only way to make sure that your money is safe is to have it in a government-guaranteed, insured account. That is as true now as it has ever been.

The more important thing for individuals is recognizing, as citizens and voters, that the regulation of the financial sector is an ongoing challenge. Even with the very best intentions, there are going to be mistakes and oversights.

I tell people, support our efforts to regulate and get more of this under control, and let’s not have an attitude of, “They’re all crooks. They’re all evil. They’re all trying to rip us off all the time.” It’s just not particularly productive. We’re going to have a financial system. The people in the financial system are going to be chasing profit. If we’re going to regulate them effectively, we have to recognize that the system is very complicated, and we need to give some scope of action to the people we’ve entrusted with that regulatory responsibility.