Eight years after the global financial crisis, Italy’s economy remains weak and the country’s banks have a very high rate of shaky – or non-performing — loans at about 18%. That compares with rates of 5% in France and 1.5% in the United Kingdom. Since Italy is the third-largest economy in the eurozone, a breakout of loan defaults or a run on bank deposits could quickly spread eurozone-wide, where many banks have been struggling, in part because of record-low interest rates cutting profit margins. What’s more, companies in Europe depend on bank loans far more than in the U.S., so struggling banks can mean that even successful companies face a credit squeeze.

This has led to fears of a “doom loop” because the potential failure of Italy’s banking system might require a state rescue at a time when the country is already heavily indebted at around 135% of GDP. What’s more, a financially incestuous relationship between the two already exists — Italian banks are heavily invested in Italian government debt. And it is all further complicated because as of January, new regulations require European banks to bail-in shareholders and bondholders – use their holdings to recapitalize a troubled bank — before any taxpayer-funded bailout can occur. Yet in Italy, many bondholders are actually small investors who were duped into buying bank bonds under the impression that they were as safe as insured deposits, explains Franklin Allen, an emeritus finance professor at Wharton and a finance professor at Imperial College in London. If small investors start to take a hit, it could spark a run on bank deposits and kick off a major crisis. In this Knowledge at Wharton interview, he looks at the big picture regarding risky Italian banks, assesses the odds of significant problems breaking out, and considers how officials might avoid a major new financial crisis.

An edited transcript of the conversation appears below:

Knowledge at Wharton: A new rule passed by the European Central Bank that takes effect next year calls for bank authorities to tap stockholders and bondholders for recapitalization before they would tap taxpayers for any bailout. This so-called “bail-in” is in contrast to what happened after the financial crisis when the U.S. — and many European countries — bailed out their banks using taxpayer money. But many of the bondholders in Italy are small investors. They are similar to small depositors in the U.S. [and the small bondholders in Italy] thought that buying bonds would be a safe investment. These smaller holders may not be aware of the precarious financial position they are in because of these underperforming loans held by the Italian banks. Do you agree with this context? And what are the risks of these bondholders suddenly panicking, if they were to realize that the banks are in a fragile condition, and starting a run on deposits?

Franklin Allen: I agree completely that this is a very important issue and I think it’s off the radar screen of most people who aren’t involved in the financial sector. People in the financial sector realize that this is a potentially existential problem for the European Union. The reason is that Italian banks are very big and if they need to be bailed out by taxpayers, large amounts would be required. But the problem is that the Italian government is already heavily indebted, so there’s what people sometimes call “the doom loop” between banks and the sovereign. That’s the general background.

“These systemic problems can arise out of nowhere — or out of small beginnings — and take over very quickly…. This problem with the Italian banks has that potential….”

You laid out very well that under the new rules — the [European Union] Bank Recovery and Resolution Directive — the banks need to bail in shareholders and bondholders before they can get state funds. This is problematic because it seems as though many small investors were missold. In other words, they were not really told the truth about what they were buying. In many cases, they thought they were buying something that was equivalent to insured deposits. But these subordinated and other kinds of debt are not like that. They have the potential to be bailed in.

I think everybody agrees that it’s fine to bail in the shareholders, but it’s problematic to bail in these small bondholders. Politically, this represents a big problem for Italian Prime Minister Matteo Renzi’s government. There was a case at the end of last year where four small banks were bailed in and the same problem occurred there in that many didn’t realize what they had. There was a tragic case of a retired pensioner who lost his life savings and was so distraught that they committed suicide. This obtained quite a lot of publicity, as you would expect.

But I think you’re right, there are still many people who don’t fully understand what they’re holding. It seems that if there are bail-ins, the Italian government is going to recompense the small savers who have this kind of debt. But we get back to this problem of, can they afford to do that? And who exactly will get compensated? Will they go back and compensate the people in the banks that had the bail-ins last year? How far will it go?

Knowledge at Wharton: If that were to escalate up, if they weren’t able to contain it on a local level, then you start to open up all of the things that people have worried about for the last several years since the last financial crisis, which is this whole idea of systemic risk. This could spread to other banks in Europe that are facing some lean times and would have difficulties dealing with big challenges. One in particular is Deutsche Bank. Could you talk about the risk of systemic contagion and Deutsche Bank, in particular?

Allen: In Italy itself, there are a number of banks that have problems. Banca Monte dei Paschi di Siena, which is the oldest bank in the world dating back to the 15th century, has been bailed out twice already but is still on the ropes. The first question is, what will happen to them? They are the No. 3 bank in Italy, so they’re significant. They’re not huge in a global sense. The bank that is globally important in Italy is UniCredit, and it has operations in many other parts of Europe and the rest of the world. They have had a big drop in their equity price since the beginning of the year. This represents the fact that people are worried about these issues that you’ve talked about, the nonperforming loans and how the government will deal with them if they get into trouble.

I think if there is a meltdown, particularly if it spreads to UniCredit, then other banks in Europe will also face problems. I think Deutsche Bank has potential to have problems. The nature of their business is such that the very low interest rates and the fact that there isn’t much difference between long-term rates and short-term rates is problematic for them because of their business model. So, there is likely to be some contagion.

Once it starts, it’s very difficult to stop because I don’t think people understand what’s going on, and they are going to be quite surprised by these new bail-in rules. That in itself will cause significant problems both for small retail investors and, potentially, institutional investors.

Knowledge at Wharton: The new bail-in rules had a good intention. We don’t want to have bailouts funded by taxpayers the way we did last time. But the ECB [European Central Bank] is holding tight to those rules at a time when they might have said, for example, “We will be the lender of last resort if Italy gets into real trouble. We don’t really want to bail them out. However, we’re not going to allow it to spread beyond Italy’s borders.” But it’s those very bail-ins [the ECB says it will enforce] that could cause the runs on deposits and the systematic risk spilling over the border. Could you explain why they want to hold fast to these rules — yet without some flexibility they could create the very problem they’re trying to avoid?

Allen: I think it’s even more complicated than what you just laid out because the other player in all this are the competition authorities in Brussels, because one of the aspects of providing bailouts using government finance is that this is regarded as unfair state aid to these companies and that’s ruled out in the EU.

In addition to the ECB, there are a number of players here — the national governments, the commission and the competition authorities. It’s a very complex set of issues, and I think many of the players are still very much in the mindset of macroeconomists, that problems occur when there’s a big macroeconomic shock or some big macroeconomic problem.

A number of the governments are saying, “We should stick with it and not solve this problem.” Now we should see it play out because it doesn’t look like it’s a systemic problem. But I think these systemic problems can arise out of nowhere or out of small beginnings and take over very quickly. That’s what we saw in the financial crisis. And I think this problem with the Italian banks has that potential, and that’s why it’s so worrying and so important.

Knowledge at Wharton: There are also problems where economics and finance intersect with politics. In Italy, there is the Five Star Movement, a rising political party that has called for a referendum on staying with the euro. If taxpayers had to bail out Italian banks, that would be bad for Renzi. At the same time, that could mean that Five Star could do well in fall elections and introduce a resolution for Italy to leave the EU, which would be very serious. It would not threaten just the financial system in Europe, but the whole EU system.

Allen: The political situation in Italy is complex. They’re going to have a referendum in October on constitutional reform, which will effectively change the way their political system works quite dramatically. At the moment they have a two-house system. If approved, this referendum will change that significantly and effectively eliminate the upper house. The latest polls that I saw were that Five Star Movement is ahead, and we saw in local government elections the Five Star candidates were successful in Rome and Turin, and in a number of other places.

“I think everybody agrees that it’s fine to bail in the shareholders, but it’s problematic to bail in these small bondholders.”

It’s not at all clear at this stage what will happen in the referendum. Renzi has said he’ll resign if he doesn’t get the result he wants. He’s trying somewhat to backtrack from that at the moment. But there are a whole set of volatile events. If it becomes effectively a referendum where people want to deliver a protest vote, which often seems to happen in these referenda in the EU, then he may well lose and then he may have to call elections. Even if he doesn’t, then when elections do come, the Five Star Movement has a chance of doing very well.

While I don’t think they will try and pull out of the EU, they may hold a referendum on pulling out of the eurozone. I think it’s widely perceived that a lot of Italy’s problems and lack of growth for many, many years now is the result of being a member of the eurozone. This, again, is something that has huge ramifications all around. This is a reason to be worried about what’s going on in Italy at the moment.

Knowledge at Wharton: What would happen if Italy were to exit the eurozone?

Allen: Greece and Portugal are very small countries, and their GDPs are a small proportion of the eurozone total, but Italy obviously is a big country with lots of debt. If it were to do something like leave the eurozone, then it would be a major event. This is potentially much more serious than what happened with Greece over the last few years. Again, this is the reason so many people in the financial sector are worried about what’s happening in Italy at the moment.

Knowledge at Wharton: Let’s go back to possible solutions for the financial problem that we talked about. Right now, the ECB is holding to its position, and the Italian government is asking the ECB to give some sort of guarantee and not enforce this bail-in rule. Is there a compromise there? Or if some financial crisis did break out, is there something they could do to contain it quickly? [The Financial Times reported on July 21 that Mario Draghi “now favors a public bailout of Italy’s troubled banks ‘in exceptional circumstances….’” But so far Renzi and EU officials have not been able to hammer out a “deal on state help for the country’s lenders — notably for Monte dei Paschi di Siena.”]

Allen: I think what’s quite likely to happen, which is being discussed, is that there will be a bail-in. And that will be followed by the small people and maybe the institutional investors being compensated for the losses because they were missold. That’s the justification for the small investors.

For the larger investors, I think it will cause problems because they’ll essentially withdraw from many of the markets for subordinated debt and other kinds of debt to banks, and the intent to bail then will also be problematic. So, I think they will try and do something like that, and we’ll see how it all plays out. But it has got the potential to become big quite quickly. If that does happen, what’s likely is that the Italian government may well just simply ignore what the EU says and bail out the banks that need it.

Knowledge at Wharton: So the government would just issue more debt over some period and use that money to bail the banks out?

Allen: Yes, exactly. Either that, or have some kind of taxes or in one way or another raise the money that they need to bail out the banks and make sure there isn’t a systemic event.

Knowledge at Wharton: Then the markets would immediately start worrying about Portugal and Spain?

Allen: They would, but it’s really Italy that is the big problem. The other big problem in the background there potentially is France, because the terrorist attacks have increased the popularity of [ultra conservative party leader] Marine Le Pen. If there were one or two more large terror attacks before the elections next year, they also have as a policy to pull out of the eurozone and also to have Frexit [French Exit] to pull France out of the EU. So that’s also in the background there. Although, there I think the risks are much more of the terrorist type rather than the economic type that they are in Italy.

Knowledge at Wharton: What do you think the odds are that the situation in Italy could spin into a very major financial crisis?

Allen: My own assessment would be something like 15% to 20%. It’s still low, but it’s significant.

Knowledge at Wharton: What else is it important to understand about this whole issue?

Allen: We touched on it briefly, but it’s this basic thing that Italy has 135% or so of GDP as government debt, and that’s problematic. At the moment, interest rates have gone down, they’re at all-time low levels. The Italian government has a spread above the German bund rate, for example, but there’s always this chance that interest rates will jump up because people start pulling their money out in anticipation that there’s going to be a problem.

“While I don’t think they [Italy] will try and pull out of the EU, they may hold a referendum on pulling out of the eurozone.”

That’s always the underlying issue — that people will start running both on the banks and the government. That’s this doom loop, which is so worrying, particularly for Italy, where the banks hold a lot of Italian government debt.

Knowledge at Wharton: The International Monetary Fund recently reported that in Italy productivity remains weak, debt is still climbing and it’s very difficult for the economy to grow when the banks are in such trouble because they are not in a position to lend money to companies. The IMF also said that the economy would not get back to its pre-crisis level until 2025. Could you comment on that?

Allen: I think Italy has a major growth problem. They are essentially, in terms of their GDP, where they were 20 years ago. Their growth rate is very low, and it doesn’t look like they are going to get back to before the crisis anytime soon. This is very problematic for the population in Italy, but also for the political system. They also have a very high youth unemployment and so on, and these are problems that tend to fester and cause feedback problems that are also difficult. I think that if everything goes fine and Italy survives the next few years without some kind of financial crisis, it still has a lot of long-run problems, which always has the potential to bring back this financial crisis kind of situation that we’ve been discussing.

Knowledge at Wharton: Another doom loop, it sounds like. What advice would you give them? Back when the EU was in real crisis several years ago, there was talk of some countries, including Italy, dropping out of the eurozone for a short time so they could take advantage of devaluation, and then coming back in. Could that work or are we past a time when that would be useful?

Allen: Italy is so big in terms of the amount of debt that it has. It’s a couple of trillion Euros. That those kinds of solutions are on the cards again, because they are too big for the Germans or the Dutch or the strong countries to pay off the problem, that’s the real issue. In many ways, this is much more serious than what we saw before a couple of years ago, if it comes and turns into a full-blown crisis.