Claudio Borio, chief economist at the Bank for International Settlements (BIS) in Basel, believes that the question that differentiates what has been happening in continental Europe from what has been happening to some extent in the US, the UK and other economies, including Spain, is the extent to which the banking problems and the asset quality problems have been addressed head-on.
Question: Some of your colleagues consider deflation a “terrible disease”, yet you don’t seem to see it that way. What is your take on deflation?
Answer: The point we have been making for quite some time is that whether deflation is problematic or not depends on the causes of that deflation. “Deflation” is an emotional term, with a lot of baggage: if you talk about “falling prices” instead, it doesn’t sound so bad, does it? What matters is whether the fall in prices is driven by supply side or demand side forces.
In fact, there is an underappreciated, long established empirical literature on this question that makes a distinction between “good” and “bad” deflation. In work which we released in March 2015, we examined this issue further. We found, just as we had in previous work, that the close association between the fall in price levels and weakness in economic activity largely comes from the experience of the Great Depression. Outside the Great Depression, you don’t see it that much. But we also found – and this was new – that when you allow for changes in asset prices, property and equity prices in particular, you see that even during the Great Depression the link between deflation and output weakness disappears, suggesting that the fall in asset prices was more damaging. In addition, our analysis pointed to clear evidence of a costly interaction between debt and property prices, as we saw in the Great Recession recently, but not so much between debt and deflation. Taken together, these results suggest that the causes of deflation do matter. To give you just one recent example out of several: in Switzerland, prices have been falling for some time and yet the economy has been doing quite well.
Q: What about debt? After each economic crisis, the world has become more leveraged (right now debt levels are around 255% of GDP). Is there a limit? Do you think there could come a day when we consider indebtedness equivalent to 255% of GDP as normal, ie not dangerous?
A: The key question is whether companies will be able to generate the future revenue to sustain debt smoothly, whether households will have sufficient income to pay debt back, and whether governments will be able to raise the necessary taxes. There is a clear limit to all this, but it is very difficult to identify it exactly in practice. For example, at the BIS we have done some statistical work to find out what happens when debt levels become relatively high, and we have reached the conclusion that beyond a certain level – 90% or something like that – growth slows down. But how much debt can be sustained in any given economy depends very much on country-specific factors, including the extent to which there has been some kind of natural “financial deepening”, with debt rising in line with sustainable increases in asset prices and healthy intra-financial system links.
More importantly, as an empirical matter, what is critical is whether debt grows very rapidly for a long time in relation to GDP, which is a rough aggregate indicator of income generation capacity. So, this is not about a specific level but about the pace of relative expansion. We have found, and others have confirmed, that significant increases in the credit-to-GDP ratio relative to its long-term trend, itself a rough measure of “financial deepening”, are the most reliable indicator of the build-up of financial risks and of financial distress down the road. To be sure, the indicator is by no means perfect, but it is useful.
From this persepctive, we are concerned with two developments since the financial crisis. First, in a number of economies that didn’t have a financial bust at the time of the crisis, we have seen strong increases in credit, often alongside asset prices, including property prices. These financial booms are not qualitatively different from those that we saw pre-crisis in a number of economies that had problems later on. And in the rest of the world, we have not seen that much deleveraging. Second, all this has been taking place alongside a trend decline in interest rates. Combined, these developments point to the risk of a “debt trap”: at some point it may be difficult to raise interest rates without creating the very problems policymakers have been trying to avoid, because the economy might not be able to withstand higher rates. What may appear to make sense in the short run generates problems in the long run.
Q: To what extent would you think your theory of “financial cycle drag” and Larry Summers’ “secular stagnation” are incompatible? Would there be any meeting point for your respective theories?
A:We should probably first say what the two hypotheses are. The version I was taking issue with was the one that posits a stuctural shortfall in aggregate demand, as a result of deep-seated forces such as demographics, inequality or less need for physical investment, which in turn leads to persistently low growth. The alternative view, more consistent with what I was telling you before, is that what we are seeing is persistent but temporary weakness owing to a financial boom gone wrong – what I called the “financial cycle drag” (FCD) hypothesis. At its heart is the interaction of falling asset prices, especially property prices, and overindebtedness – a debt overhang.
Now, before mentioning the possible similarities, let me stress the differences between the two hypotheses, which are more important. First, they differ in terms of the balance between financial and real factors: the FCD hypothesis is very much a financial story and about how financial factors interact with the real economy; by contrast, the secular stagnation hypothesis is very much one about real factors. Second, they differ in terms of the balance between aggregate demand and aggregate supply: the FCD hypothesis of course is very much concerned with deficiencies in aggregate demand as a result of too much debt, but it also highlights the importance of the supply side of the economy – the fact that you can only generate lasting growth from the supply side as well as the role of resource misallocations linked to the boom and bust. Third, the two hypotheses differ in terms of the interpretation of inflation and interest rates. In the FCD hypothesis, persistently low inflation is more related to the supply side factors that I was telling you about, particularly globalisation and possibly technology, than to a big shortfall in demand. And according to the secular stagnation hypothesis, the fall in interest rates is an unavoidable equilibrium phenomenon whereas according to its FCD counterpart it is largely a disequilibrium one, which can generate problems in the longer run. Finally, they differ in terms of realistic future scenarios – temporary versus permanent weakness – and in terms of policy – the balance between structural and demand management policies and, in particular, the role of monetary policy.
If you look hard enough, you can also find some similarities. Both take a long-term perspective, both question the presumption that the economy is self-equilibrating, and both question more thoroughly the prevailing analytical paradigms used to interpret the workings of the economy. One could also point to some more specific common aspects. For instance, both recognise that demographics affect both supply and demand. Indeed, I think demographics is going to be a big force in the future, reducing growth. And the secular stagnation hypothesis also recognises that the financial cycle bust has created problems.
Let me say that there is another version of the secular stagnation hypothesis, one that says we are going to be in a slow-growth environment for technological reasons and due to low productivity growth. This is purely a supply side story. We are concerned about slow productivity growth too, but I am personally optimistic about technological progress. The challenge is to turn that progress into growth and to manage its social consequences – in Schumpeter’s terminology, its creative destruction.
Q: Most European bankers believe this financial crisis has been tackled according to the rules of the Anglo-Saxon investment banking sector (no defaults, no renegotiation), while a European way would be more in line with burden-sharing measures. What do you think?
A: Let me put it differently. The real question that differentiates what has been happening in continental Europe from what has been happening to some extent in the US, the UK and other economies – and I would also put Spain in this category – is the extent to which the banking problems and the asset quality problems have been adressed head-on, losses have been recognised, balance sheets cleaned up, and institutions recapitalised and put on a stronger footing. I don’t think there is an Anglo-Saxon versus a European way of doing things. It’s just a different policy choice under specific circumstances. It echoes the different responses in the early 1990s crises: in the Nordic countries, which tackled their problems head-on, and in Japan, which chose to wait for better times. That difference helps explain the difference in growth performance after their recession: the Nordic countries rebounded whereas Japan lost a decade.
In Spain, the problems were tackled energetically. That has not quite been the case in every other euro area country which faced banking stress. So I don’t think it’s an Anglo-Saxon versus a European way of dealing with things. It reflects a particular policy choice given the constraints and preferences at the time.
Q: From the BIS Annual Report, we infer that despite the ECB’s efforts the European interbank market is broken. Do you agree? The report provides data underlining that cross-border operations between banks went from 70% of GDP in 2007 to less than 40% of GDP in 2013. Is this decline still happening in 2017? The Spanish Banking Association expressed its fears about it.
A: Actually, in the Annual Report we talk about euro area banks in the broader context of globalisation, not about the euro area interbank market per se. By the way, the figures you cite also include lending by UK and Swiss banks and are vis-à-vis all borrowers worldwide, including non-banks.
Q: In 2007, Spanish banks were getting credit from the outside: the Sparkassen in Germany. Then they were shut out. I would like to know if this is still happening in 2017. If it hadn’t been for the ECB, what would have happened?
A: I understand but I’d rather not comment on this specific issue, in line with BIS policy. Let me just say in general that the functioning of the interbank market has been improving, but it differs from pre-crisis. First, once you have a crisis, everyone pulls back; moreover, pre-crisis too little attention was paid to counterparty risk. Second, there is a more technical issue, but with major implications: post-crisis the ECB has greatly increased its presence in the market, flooding it with liquidity, through its QE (quantitative easing) operations. As a result, banks have been more inclined to deal with the central bank than with each other.
Q: Do you think that cross-border operations between banks going from 70% of GDP in 2007 to less than 40% of GDP in 2013 was due to globalisation alone?
A: No, I am not saying that. The argument put forward in the Annual Report is that there had been an overexpansion of the euro area banking system before the crisis and that this extended to the role of euro area banks outside the area. So, what we have seen post-crisis is a cyclical retrenchment, a cyclical unwinding of the previous expansion. That should not be interpreted as a retrenchment of globalisation: it is something that is much more specific – it’s a cyclical factor as opposed to a trend. Now, what the “right” number should be, no one knows. But it is important to understand that if the system becomes overstretched, at some point a correction must follow. And that retrenchment is likely to occur along national lines: you return to your home base, your comfort zone.
Q: Is there any coordination between central banks’ different monetary policies, some sort of “non-aggression” pact?
A: On that I cannot comment.