Interview with Nobuhiro Kiyotaki

Nobuhiro Kiyotaki (Princeton University)

Paper: “Wholesale Banking and Bank Runs in Macroeconomic Modelling of Financial Crises”

The paper’s main takeaway, according to Kiyotaki:

  • We have to worry about the new type of financial crisis centered on wholesale banks and shadow banks. The development of shadow banks itself is not the bad thing. But the potential risk is that they are vulnerable to the rollover risk, and policymakers should be prepared to act.

The Federal Reserve Bank of St. Louis hosted its 40th Annual Fall Conference on Oct. 15-16, 2015. David Andolfatto, a vice president and economist with the St. Louis Fed’s Research division, sat down with each of the conference presenters and discussed their work in plain English. The content below is from those interviews. All interviews have been edited for clarity and length, so the content below should not be considered a transcript. 

ANDOLFATTO

For people who are not familiar with banking and bank runs, what is wholesale banking? I suppose it differs from retail banking.

KIYOTAKI 

Nobu KiyotakiThe traditional bank run is focused on the retail banks. Household depositors worry about health of the bank, and they run up to get money before the bank runs out.

ANDOLFATTO

I guess this is like an It’s a Wonderful Life scenario, something like a retail level of households running to the bank.

KIYOTAKI 

That’s the traditional bank run. But the recent bank runs in the 1970s, Continental Illinois bank run or perhaps more importantly the recent recession’s bank run is not on the retail banks but actually is centered in the interbank market or the wholesale funding market. Basically, nowadays a lot of financial institutions borrow money from other institutions, financial banks. Investment bankers, for example, raise funds mostly from other financial institutions like pension funds as well as commercial bankers. They don’t raise many funds from the household.

There, the bank run is when big lenders like other commercial bankers or the insurance companies stop rolling over the loans, short overnight loans.

ANDOLFATTO

So it’s much like a demand deposit liability? When people have an account at their bank, they’re effectively rolling over their loan to the bank night after night. And then they might all run to withdraw their cash.

KIYOTAKI 

Yes, but here you stop rolling over and financing overnight. And then this borrower—say, an investment banker—runs into trouble. They start selling the assets. When a lot of investment bankers start selling the assets—like mortgage-backed securities or lots of financial products that they hold—the price starts falling. At that point, even if at the beginning they’re healthy, they run into trouble because of asset prices dropping. They become insolvent.

ANDOLFATTO

What you’re describing to me—and probably this might be familiar to a lot of our audience—sounds like a standard retail-level kind of run, that the banks have to meet the large wave of demand for cash by selling off assets.

KIYOTAKI 

The difference is that the traditional bank run is based on something called the sequential service constraint, which is basically: If you go early, you get the full return. First come, first serve. And if you don’t go early enough, you lose the money.

But the wholesale case, basically every contract is short term. If you stop rolling over, then the borrower runs into trouble, starts selling the assets, then at the end of the day, everybody gets hit because asset prices dropped so much.

The difference between the liquidity shortage, the shortage of their liquid assets versus illiquid, is much more subtle. When you look at the popular writing of, say, Paul Krugman, they always say insolvency and liquidity shortages are different things.

But, here, liquidity shortages quickly transformed into insolvency because the asset prices, like mortgage-backed securities prices, weredropping. Then they are in trouble. So the question is, how do you stop this kind of bank run? Or what is the cost of the bank run?

ANDOLFATTO

Before we get there, I’m aware of different theories as to what actually triggers these events. In your view or in your model, what causes these things to happen?

KIYOTAKI 

Before the crisis, we have a big buildup of the wholesale funding market. Basically, the investment bankers and all these financial intermediaries will rely on the funding from other banks. Other financial intermediaries become very big.

That itself is not bad. If a wholesaler has better lending opportunities, more funds go to the wholesale banker, which is a good thing. But when a negative shock hits—like, say, default of the mortgage markets—it will hit the most vulnerable sectors or, in this case, the wholesaler who relies heavily on the short-market funding.

If nobody stopped rolling over as the price stays high, the wholesale banking sector is still solvent. But if people stop rolling over and if government doesn’t intervene, then the price drops enough, and then entire wholesale banking sector gets hit.

ANDOLFATTO

So there is a psychological component here that’s triggered by a small fundamental shock?

KIYOTAKI 

Yes.

ANDOLFATTO

The traditional kind of retail-level bank runs rely on esoteric things like first come, first serve. But, here, this is different. It doesn’t rely on that. And the institutions operating in this wholesale market, they would be who? JPMorgan? Bear Stearns? Lehman Brothers? The money market mutual funds as suppliers? These are the agents that are operating in this wholesale market?

KIYOTAKI 

Yes.

ANDOLFATTO

So that’s the triggering event. There’s a possibility of a contagion, but if everybody remained calm, it would be fine. But people stop rolling over the debt, and this precipitates a sale of assets. Is there some friction in these? They’re illiquid or something?

KIYOTAKI 

Yes. If these assets are sold in the market for the potential value, like with other retail banks or maybe nonfinancial sectors, they are not good at dealing with these assets, say mortgage-backed securities. They cannot use your household to buy these kind of things.

ANDOLFATTO

They’re pretty esoteric products. They’re idiosyncratic. You don’t know if they’re good or bad.

KIYOTAKI 

Yes, and usually, only professionals buy. But if professional buyers are concentrated in the wholesale banking sectors and start selling the assets, the potential buyer may not want to pay too much. In a height of crisis, they try to sell very high-grade tranches of the mortgage-backed securities, and the price drops to like 22 cents for a dollar, even if the default rate is tiny.

ANDOLFATTO

I think that’s an important point. It’s even these AAA-rated tranches of private label securities, these mortgage-backed securities. People may laugh that the ratings were not appropriate. But in fact, as you pointed out, they continued to service them. They did not default. Arguably, they were rated correctly, because they don’t measure liquidities.

KIYOTAKI 

The liquidity here is that the potential buyer is not as good as the usual buyer. And the usual buyers in this case, like wholesale bankers or investment bankers, are shorting cash.

ANDOLFATTO

Who’s shorting the cash here? Lehman? Why don’t we use an example?

KIYOTAKI 

So Lehman’s lender stopped rolling over the short-term credit, stopped lending to Lehman. They have to sell the assets in the market.

ANDOLFATTO

I heard Lehman was heavily invested in subprime. And this is not exactly the AAA-rated stuff you are talking about.

KIYOTAKI 

No, at the end of the day, their subprime securities default rate was not that bad.

ANDOLFATTO

So when people label these things "toxic assets," what are they talking about?

KIYOTAKI 

If you look at the most senior part of the toxic asset, asset-backed securities, the default rate is not that high.

ANDOLFATTO

But was that what Lehman was investing in?

KIYOTAKI 

Yes. They do have a lot of the asset-backed securities, especially the subprime-backed securities.

ANDOLFATTO

So you’re talking about the senior tranches of the subprime?

KIYOTAKI 

Yes. Subprime itself can default. But if you have an asset backed by the subprime, the first loss is absorbed by the other people, then the senior part is relatively safe.

ANDOLFATTO

So, arguably, Lehman wasn’t really doing something terribly bad, was it?

KIYOTAKI 

But the lenders stopped lending. Not just Lehman, but Bear Stearns and even Goldman Sachs, they had trouble with funding during the crisis. Then they started selling the assets.

When the price is dropping, usually it’s good news for the buyer. But this time, they have a big debt. Then the net worth, the assets minus the debt, moves a lot, and sometimes it becomes negative.

ANDOLFATTO

We’ve got Lehman here. They’re going long on these mortgage-backed securities. They’re going short on Treasuries and cash. We have this event that causes the prices of these mortgage-backed securities to go in the wrong direction. And Treasuries are actually going the opposite. So Lehman has just made a bad bet.

KIYOTAKI 

And their net worth becomes negative. Then they are going to be insolvent.

ANDOLFATTO

Do you think that if the creditors to Lehman had just maintained their cool that Lehman would be in business today? Is this your feeling?

KIYOTAKI 

If the majority of the investment banking is still getting funded, then it would be.

ANDOLFATTO

We have this disruption in the wholesale banking sector. Lehman is getting creamed on its positions. All its bets are going all the wrong way. Why should the rest of America care?

KIYOTAKI 

It affects the spread between the assets, illiquid versus liquid assets. During the crisis, as you said, the interest rate on Treasury securities drops. But the interest rates on risky mortgage-backed securities or commercial papers go up.

So the spread expanded, and the spread expanded particularly severely after the summer of 2007, and then shot up during the Lehman crisis. And when the spread goes up, it will affect the households borrowing money from the bank or the nonfinancial businesses borrowing money from the bank.

ANDOLFATTO

This directly impinges on the credit conditions of households or businesses just wanting to borrow funds to buy homes, to do capital spending, and that’s how it affects the real income. So what should we do about it?

KIYOTAKI 

Generally, expanding the wholesale bank itself is not bad. If the wholesaler has better loan opportunities or better financing technology than retailers, that’s a good thing. But if they become too big relative to their net worth and so-called leverage becomes too big, then they become vulnerable for this type of wholesale bank run.

Basically, before the crashes hit, you might not want to have too much leverage. Or the other way to say it is that this important banker should have more capital relative to the assets.

That’s ex-ante policies. Ex-post policy is that you might want to stop the asset price collapse during the crisis. That’s more like a lender of last resort style intervention during the crisis.

ANDOLFATTO

They call this sector the shadow banking sector. Many, like the Fed, did not have jurisdiction to kind of actually regulate many of these entities. We can kind of see Lehman, but there are many others of these institutions. And they operate in these opaque over-the-counter markets.

It’s one thing to say that we should impose capital controls or some sort of regulation on this sector, but as a practical matter, given that it’s in the shadows, do you think that it’s possible?

KIYOTAKI 

Yes. By definition, like I said, the regulation of banks is very hard. Therefore, ex-ante, you worry where the risk is building up.

ANDOLFATTO

But how can we even see the risk building up? We can’t even see these people’s portfolios.

KIYOTAKI 

There is some balance sheet data. The BIS tried to get some data on shadow banks. And some of the banks are no longer shadow, like many other investment bankers who survived are already under the regulation of the Fed.

ANDOLFATTO

Right. Much of the shadow banking sector probably is outside of the existing regulation. I guess the worry is that these people are clever. They’re going to try to innovate around. What can we do?

KIYOTAKI 

What we can do is the ex-post.

ANDOLFATTO

There’s nothing we can really do, but maybe ex-post, the Fed could serve as a lender of last resort?

KIYOTAKI 

Or try to limit the damage, in case shadow banks can become insolvent. At the same time, you don’t want to do regulations that are too severe. As you said, as soon as you put in very severe regulations, people try to avoid that.

One example recently is the Bank of England, which is saying, “We don’t want to ban the high loan-to-income-type mortgage.” Some people don’t have much income but nonetheless want to buy the house, especially when they are very young. If the banker is lending this, they have extra capital requirements. And this regulation is not too big of a tax on the extra capital requirement.

At the same time, if housing prices are going up, the loan-to-value regulation doesn’t work well because the value is already getting high. But the long-term-income one is going to work when the housing price is booming. During the boom, the requirement gets a little tighter, and during the recession, it’s relaxed. So that’s the kind of policy people are talking about.

ANDOLFATTO

Do you have a view on the Fed’s intervention in 2008? Was it successful?

KIYOTAKI 

I think so.

ANDOLFATTO

Was it a bailout?

KIYOTAKI 

No, because the Fed actually made a huge amount of money. If it’s a bailout, the price is higher than the real value. That is a subsidy to the seller of the assets.

But the Fed actually made a huge amount of money with almost every program they did. How effective? People are debating. But it’s not a bailout or money transfer to the big bankers.