Major financial crises seem to rear their ugly head about every decade or so somewhere in the world, each different from the preceding one. 

While there has been a lot of research on the causes of the latest global financial crisis that began in 2008 — and how to prevent it happening again — many experts argue that any new crisis will be different. Like the old saw noting that “generals always fight the last war,” there have been questions raised about whether or not researchers have focused enough on what might cause a future financial crisis, particularly with regard to central bank behavior.

In an effort to avoid planning for the past when it comes to global financial crises, Wharton finance professor Franklin Allen has collaborated on a research paper titled “Financial Connections and Systemic Risk” with colleagues Ana Babus of the University of Cambridge and Elena Carletti of the European University Institute.

Allen spoke with Knowledge at Wharton about his paper and what regulators need to be concerned about in order to avoid future crises. 

An edited transcript of the conversation follows.  

On what the research is about:

The research I’m going to talk about is part of a long agenda. It has to do with the way that central banks and governments intervene in the economy.

For the last 20 or 30 years, central banks have, by and large, focused on fighting inflation. After we had the big shocks in the 1970s, that was the major problem, and that’s what they’ve spent their main efforts doing. Some of the central banks, like the Federal Reserve, have a dual mandate. In addition to worrying about inflation, they also have to worry about unemployment.

The way that this has been implemented in most countries, either explicitly or implicitly, is that the central bank has focused on inflation, and it’s usually granted formal independence from the government. The idea there is to stop it from lowering interest rates just before an election and making the economy boom, but then having inflation going up and so on. That was an idea that has been widely accepted for some time.

Financial stability was in the mix, but it was usually regarded as something that was secondary — so that was dealt with either by the central bank or in many countries, such as the U.K. or Japan, by a separate financial services authority or FSA. They would deal with problems to make sure that there were no difficulties in transmitting monetary policy because banks were having problems. The way they did that was to stop banks taking risks, one by one. They would look at each bank and make sure that they weren’t doing risky things. The idea was that would stop any problems in the financial system.

Fiscal policy was done separately by the treasury or the finance ministry, depending on the country. We could break up all these different parts of the way the government and central banks intervened, and they could all do their job separately. Now the problem is — what the crisis has shown — is that system didn’t work properly. What we need to do, I argue in this research, is to think carefully about how we should proceed going forward.

It may well be that we need to change the architecture of the structure of policy-making and have a much more holistic view, because all of these things are intertwined. Financial stability isn’t secondary. It’s a primary target in the same way that inflation fighting is. We also have to worry about the fiscal side, because that also is an important part of the mix. 

The key takeaways for regulators:

The most important is that central banks need to worry about the effects of their policies on asset prices. We used to think of asset prices as being independent of what they do to a large degree. But what the housing bubble illustrated was that they can have a big effect. And if things go wrong, in the sense that they rise quickly and get out of line and then collapse, that can trigger a whole set of very unfortunate events. That’s what we saw during the crisis.

[Second,] when they lower interest rates it has an effect on house prices, but it potentially has many other effects. One of the problems that they now face is that interest rates are going to go up at some point. And when they go up, it’s important they go up slowly enough that we don’t have a big fall in prices of long-term bonds, mortgages and so on, which will get us back into a crisis in just the same way that house prices led to a fall in the price of securitized mortgages and so on. So, that’s the second big takeaway. They need to worry about that.

The third is that they have to start thinking holistically, because when they run up debt in large amounts and interest rates go up again, there may be a fiscal problem and that also can be significant.

Japan as an illustration:

The country that is a very good illustration of some of these issues is Japan. Japan had a massive property and stock bubble back in the late 1980s and early 1990s. Since then, they’ve had 25 years of low interest rates and very slow growth. Now they’re trying to get out of that by having these three arrows of “Abenomics” (Prime Minister Shinzo Abe’s mix of quantitative easing to raise inflation levels; increased government spending; and economic, labor and regulatory reform).

“Financial stability isn’t secondary. It’s a primary target in the same way that inflation fighting is.”

The first arrow is monetary policy. They are trying to create inflation, and the way they’re doing that is by having large amounts of quantitative easing. Their target is to get inflation up to 2% — it’s currently around zero or negative.

Now, most of our theory suggests that if inflation goes up by 2%, nominal interest rates should go up by about 2%. That creates a number of problems.

First, is a potential financial stability problem, because the banks hold many long-term assets like mortgages, government bonds, loans and so forth. If interest rates go up, their value falls.

In April 2013, the Bank of Japan in its financial stability review looked at the effect of a hundred basis point rise (1%) in interest rates on the assets held by banks. The large banks are fine. They don’t hold too much, and they’re reasonably well hedged. The regional banks and the small banks have a bigger problem. A hundred basis-point rise has the potential to knock their asset values by an amount that would wipe out 30% to 40%, on average, of their tier one capital. A two hundred basis point rise would do twice that — so, 60% to 80%. 

That’s clearly a problem. Their solution was [to say in effect], “We’re going to extend the current regulations,” which allow them to not mark their assets to market. Of course, that was one of the big problems in the subprime crisis. People didn’t know where the subprime securitized mortgages were, and so everybody came under suspicion. I think that this is potentially a problem in Japan. So, that’s the first issue.

The second issue is that if interest rates do go up by 2%, then they’re going to have a fiscal problem because they have over 200% of GDP in gross debt, and about 135%-140% in net debt.

In recent years, people have focused more on the net debt. But an important issue in this case is: What’s the maturity of the gross debt versus the net debt, and in particular, the assets that offset the gross debt? Much of that is the money held by the central bank — the foreign exchange reserves — and that’s invested, by and large, short term. I don’t think we have a good idea of precisely what the maturities of all these debts are. But … over the long term, if interest rates go up 2%, then on the gross side, their funding costs are going to go up by 4%-5% of GDP. On the net debt, they’re going to go up by roughly 2.5%-3%. 

Those are both very large numbers. Even the 3% is a large amount in terms of actually raising revenue to pay those amounts. That’s going to create a fiscal problem, which they already have, and which the tax rise that came into effect on April 1st is designed to offset. [Japan increased its sales tax, also called a consumption tax, from 5% to 8%.] But this is a similar order of magnitude kind of problem to their deficit problem that they’ve been running for many years now. So, that’s the second issue.

It may be that interest rates don’t rise and in fact there isn’t much evidence so far of that happening. The current yield on 10-year JGBs — Japanese Government Bonds — is about 60 basis points. So, they haven’t ticked up. Last year, just after the start of these policies, they did go up a little bit to 90 to 100 basis points, but they’re now back.

That seems to suggest that the market is not so sure these inflation policies are going to work. If they don’t work, there’s another problem, which is, if inflation is at 2% and long-term interest rates are at 60 basis points or less, then there’s a negative yield on these very large quantities of assets. That raises the issue: Will people take their money out of the country and put it in higher-yielding assets so they avoid the negative yield?

If the interest rates and the inflation rates both go up by 2%, then there is an effect in the sense that the value of the debt will be eroded in real terms, and in that sense it will be easier. But it’s still the case that the cost of raising the taxes is large. And at the beginning, that’s a large burden. So, there’s again a complicated trade off which needs to be taken into account in designing these policies. 

“We need to have policymakers worry about the effects of their policies on asset prices…. [That] was at the heart of the start of the financial crisis….”

I think these things illustrate that there’s a holistic problem — that you can’t focus on one particular target and forget about everything else. The whole set of policies has to be consistent. And the current architecture of the policy entities is not well-designed to deal with those kinds of problems. 

The most important lesson:

The key takeaway is that we need to worry a lot more about financial stability, and worrying about financial stability is very different from worrying about inflation. The reason is, we can’t just separate it out in the way that we’ve done in the past. We have to have a holistic view of policies by central banks and governments and treasuries. They’re all closely inter-related when it comes to financial stability, and the reason is that systemic risk is a very complex and involved phenomenon — it pops its head up in many ways.

The practical implications of the research:

Let me talk a little bit about systemic risk, because I think it’s very important in this whole issue. Systemic risk has at least five components. 

The first component is panics. Traditionally, I think that’s the way people thought about financial crises, that there were good equilibria and bad equilibria. Good equilibria — people keep their money in the bank, everything’s fine. But if you think other people are going to take their money out of the bank, then you’d better get your money out, too, because they only have a limited amount of liquidity. And if there are going to be problems in the bank, you want to make sure that you get your money out first.

Panics are a big problem but, as we’ve seen in the current crisis, they’re not the only problem. What we saw here was falls in asset prices and, in particular, falls in real estate prices. That lies at the heart of many financial crises if you go back historically. And that was the case, particularly here in the U.S., but also in Spain and in Ireland. That was really what drove them.

But there are many other reasons for asset price falls. Rises in interest rates are one. The flash crash is another example. Sovereign default is another example – so many. We need to understand all those different aspects. But that doesn’t exhaust the list of forms of systemic risk. Contagion is another one. If we have one financial institution go down, then others with claims on it may also be affected and we may get a whole domino effect.

A second form of contagion is, if you see a bank like Lehman default, then you may think that other banks like Goldman Sachs and Morgan Stanley, which at some level do similar kinds of things, may also be threatened. And there may be a problem for those banks. That’s what we saw during the crisis. If they hadn’t been granted access to the Fed’s discount window, many people believe that they may also have gone down.

The third type of contagion is a very important one, which is that many countries which weren’t that exposed to the subprime crisis had massive falls in GDP after Lehman defaulted. Japan is one example. It had a 10% fall in GDP in the subsequent year even though their banks weren’t involved with subprime, and in fact stayed fairly strong throughout the whole episode. Another example is Finland, which also had a GDP drop of about the same magnitude. Trade played a part in that but, again, it’s interesting that these effects are so large. We don’t really understand that very well.

The fourth kind of systemic risk is problems in the financial architecture, what many people refer to as the “plumbing.” Many markets, like the interbank markets, stop working very well. There were repo runs and so on, and problems in derivatives markets and things like that.

The fifth kind of systemic risk is problems in the foreign exchange markets or foreign exchange liabilities that banks and firms have. That wasn’t a problem in this crisis, because the central banks did a very good job having foreign exchange swaps. But it was a big problem, for example, in the Asian crisis [1997s]. So, we still have to keep worrying about it.

Which conclusions were surprising?

I think the relationship between fighting systemic risk and the kinds of policies that are needed to do that — this interrelationship — is not quite as obvious, particularly when we started on this research…. 

Central bank independence is at the heart of the whole way that monetary and fiscal policy is run at the moment, although there have been calls that it may be too much because much of what central banks currently do is fiscal in nature. I think it still isn’t quite appreciated how the whole set of policies needs to be coordinated in the way that I think will be done going forward.

“As we saw in the financial crisis, most governments, most central banks, just missed it. One of the big worries that I think many people have is whether they’ll miss the next one too….”

Policy makers need to coordinate much more and they can’t do things separately in the way that the architecture is designed to implement. They need to worry about, for example, the effects of interest rates on government deficits. Now, in many countries we had debt-to-GDP [ratios] over 100%. Every 1% rise in interest rates potentially means that there’s going to be, in the long run at least, a 1% increase in the deficit. So, that’s an important issue.

We need to have policy makers worry about the effects of their policies on asset prices.This is not something that they currently do that much, but that was at the heart of the start of the financial crisis, because we had this big run up in prices and then the collapse which triggered the subprime mortgage problems and the other mortgage problems. And then that, in turn, triggered a whole set of problems in the financial system.

On interest rates in China:

One of the major systemic risks at the moment I would say is a rise in interest rates. The question is, why should that happen? Interest rates are low in most parts of the world, and so it seems the central banks can control them and gradually raise them.

That’s true, but that’s changing in China. In China, many entrepreneurs and firms are willing to pay much higher interest rates than we see in the U.S. and Europe or Japan. They’re paying 15%, 20%. Now, part of that is a risk premium, a default premium, but the real interest rates on average — accounting for default problems and inflation in China — seem much higher.

At the moment, that is not affecting the rest of the world, because China has very stringent capital controls. But they’re going through a set of policies to ease those capital controls. The start of that is an experimental phase which is going to be conducted to a large extent in the Shanghai Free Trade Zone, which is now been going for a few months. What they’re doing there is experimenting with policies to allow financial firms — banks and other financial institutions — to go overseas and start having products based on overseas [markets]. But also they’re going to allow foreign institutions to come in to make the Chinese financial system much more competitive.

When they do relax capital controls, which presumably will be done quite slowly, then there’ll be an inflow of capital into China and an outflow. One of the big questions is, what’s that going to do to the exchange rate? One of the interesting things is that they’ve been allowing the renminbi to fall, which hasn’t happened for a long time. I think this is part of this preparation for opening up the financial system.

One of the interesting aspects of China’s economy today is that trade isn’t nearly as important as it used to be. People quote the gross figure, which is usually that about a third of GDP is exports. But that includes a lot of things that were imported and embedded in the products — for example, energy, raw materials. A few years ago the net trade was about 10%, so [that is] still significant. Now it’s down to about 2% or thereabouts. So, trade is not that important.

The traditional justification for the vast foreign exchange reserves that they have in China, of $3.8 trillion now, was that this was to make sure their exporters were competitive. But they don’t need that anymore because it’s such a small part of the economy, because the rest of the economy has grown so much. What China is doing is investing lots of money in low-yielding assets, while at the same time, the entrepreneurs are paying very high rates of 15%-20%.

As we unwind these capital controls, what we’re going to see is that there’s going to be an effect on global interest rates in my view, as China is now such a large part of the global economy. This is the part that the central banks are going to find difficult to control — long-term rates going up globally, driven by the real rates in China — this is potentially a problem.

The question as always in finance is, how much will people anticipate these changes? If they don’t anticipate these changes very much, then the period where this will be relevant could be stretched out over years. But if they anticipate that this is going to happen, then it could be shorter than that. And I think that that’s where one of the risks lies.

On common misperceptions:

One of the interesting issues is quite the extent to which the central bank and government have everything under control. As we saw in the financial crisis, most governments, most central banks, just missed it. One of the big worries that I think many people have is whether they’ll miss the next one, too, or whether now everything is under control.

I think one of the things that our research is trying to do is to get governments and central banks and policy makers to think ahead and not just try to react to what happened last time — put in place measures that will control that particular aspect of the phenomenon. What they need rather to do is worry about what is the general nature of the problem — in this case, systemic risk — and how it may it rear its ugly head going forward.

We are trying to look at the inter-relationships between various policies — between banking regulation, monetary policy, fiscal policy — the whole mix of all these policies. Typically, what we do — and there are good reasons for doing it — is look at each one separately. When you try to do it as a whole, it necessarily becomes complex.

On what’s next:

We need to understand systemic risk much better. And so, a lot of it will be looking at the individual components of systemic risk, trying to understand any additional components that are out there, drilling down into each area separately, and then going back and trying to look at the whole, holistic picture again.