 Okay. There is a paper. These slides are available on my web page, the address there. First, I'm really happy that I was invited to come here and speak and honor Vitor and his achievements. We have had many fascinating discussions one-on-one that I really appreciate and I hope we will continue and have more in the future. So my assignment is the future of monetary policy and macro-potential policy. I thought that a sensible approach was to check what Vitor has said. What has he said about the future of monetary policy and macro-potential policy? Is there anything to add or to disagree with? And he has, in the last year, he has talked at length about these things. A lecture in Lisbon a year ago, a recent speech in Frankfurt this month, macro-potential policy in Rome in November and in Frankfurt last March. So I read those and then I ran into a problem. And the serious problem I have is first, there is hardly anything to add. Vitor has covered roughly every possible topic one could think of. Thank you. The other thing is there is hardly anything to disagree with. Perhaps I'm a bit more favorable towards price level stability. Vitor is quite skeptic, but I think that's about all. So I face this problem. What I will do is to make a few very selective comments and not try to cover everything. If you want to see everything, you should read the speeches of Vitor. So on monetary policy, I will take the liberty of suggesting some clarification of the ECB's objective and the interpretation of without prejudice to price stability. And I actually propose that the ECB's objective could be phrased as price stability and full employment without prejudice to price stability. Instead of full employment, you could say maximum sustainable employment or full resource utilization or real stability. With a proper explanation, it means for practical purposes the same thing. But the interesting thing is without prejudice to the objective of price stability, how to make an operational definition of that. I think we should just think of average inflation over a longer period being close to target, like a five-year average inflation close to target. If that is the case, then there is no prejudice to price stability. Also, I think it's time to have to modify the quantitative definitions slightly and have a symmetric 2% inflation target. I also suggest somewhat more explicit forecast targeting. In practice, it would involve the staff forecasts being done for several alternative policy rate paths and publish those as a menu of choices. For the governing council, I would love to see the governing council trying the FOMC's summary of economic projections and dog plots. I think that would be very interesting. I say a few things about financial stability as an objective for monetary policy. You can deal with that very quickly if you, as I believe in the assumption or starting point that economic policies should only have goals that they can achieve. Since monetary policy cannot achieve financial stability, hence it should not have financial stability as one of the objectives. I also say something about the very interesting Swedish example of monetary policy leaning against the wind and the dramatic turnaround a few years ago. Something, the summary about cost-benefit analysis of law which has interested me a lot. Importantly, Vitor has done an update of this cost-benefit analysis for the euro area, showing that also for the euro area, the costs are much larger than the benefits. Perhaps even more importantly, he has done a cost-benefit analysis for a macro-potential policy and shown that there the benefits exceed the costs in one case for the substantial margin. That is quite interesting and is one of these speeches. On macro-potential policy, I discuss or repeat the arguments for why I think they are different, very different and should be kept separate. I say a bit about the objective of macro-potential policy. And then I go on to the importance of distinguishing between good and bad credit. I think that's a difficult and important problem in macro-potential policy. And I actually turn to another Swedish example, the current example of macro-potential policy trying to prevent credit growth with arguably the FSA not being able to provide a good case for what they are doing. I think this is relevant for families where authorities worry about the growth and level of debt and housing prices. So in the limited time I have, I will actually only talk about this last point, distinguishing good and bad credit and the Swedish example. The importance of distinguishing good and bad credit, of course not preventing bad credit growth. That may have large cost. We know that very well. But of course preventing good credit growth may also have large cost, considerable welfare and distributional cost. If you prevent people from buying the home, they would like to have. Also when you look at the data, only a third or a quarter depending on the sample of credit booms end in a crisis. Most credit booms are not so bad. Bad credit, what is it? Well, it's excessive relative to fundamentals due to market failures, too low lending standards, insufficient debt service capacity and resilience of borrowers, exuberance, overvalued housing and so on. Good credit is consistent with fundamentals, rising incomes and demand for housing to stay in falls in interest rates, appropriate lending standards, sufficient debt service capacity and resilience of borrowers, realistic expectations, reasonably valued housing and so on. The problem is to distinguish the two. And the practical problem in Sweden is that the Swedish FSA, which is in charge of a macro-potential policy in Sweden, is trying to prevent household debt growth. But when you look at their case, it's a little shaky. However, much is very good with Swedish macro-potential policy. The FSA has been very active and done a lot of things. A series of actions to make sure that the banks are well-capitalized. Currently, capital to risk-weighted assets is about 24%, 22% common equity tier one. And the banks are extremely resilient in stress tests. I mean, they are among the most resilient or the most resilient in the European Union. The FSA is called finance inspection and abbreviated FI in Sweden, but I prefer to call it the FSA in English. They have a commendable mortgage market report, which has, which using individual data on households to do stress tests on households. There's not only stress tests on banks, there's also stress tests on households done by the FSA. And the households have substantial and over time increasing debt service capacity and resilience to disturbances like house price falls, interest increases, and income losses due to unemployment. There is an LTV cap of 85% on new mortgages. The average LTV for new mortgages is 63%. And for the total stock of mortgages, it's 55% pretty reasonable values. There is a structural problem in the housing market. Demand is increasing because of rapid urbanization, rising incomes, falling mortgage rates, lack of a functional rental market due to rent control. Supply is insufficient because of restrictions on land use, building regulations, renal planning problems. So housing prices and debt have risen precisely what we would expect in this situation. Here is a graph of housing prices, disposable income, and mortgage rates. Everything indexed to 100 in 2008. We see that disposable income has increased by almost 50%. Housing prices have increased more, but the interest rates have come down quite a bit. The blue is a 10-year interest rate. Here is price over income. By the way, there is a stock flow issue here. Stocks overflows are problematic. Prices over income, debt over income. Stocks over stocks, LTVs, or flow overflows, debt service to income, user cost, housing cost to income are much better. Here we see price over income, also indexed to 100 in 2008. They have price over income has gone up, but here we see interest expenditures, the product of interest rates and housing prices. They have actually gone down relative to disposable income. Those measures are kind of interesting. That's kind of a simple interest cost of housing or a capital cost of housing or a user cost of housing. If people have Cobb-Dogla's preferences, the user cost of housing should be a given fraction of their consumption. If their consumption is still stable relative to disposable income, it should be a reasonable stable fraction of disposable income. Here in Sweden, they have fallen. The FSA is worried because debt to income has a distance and is now 1.8 times disposable income. On the other hand, we have financial assets which are higher. We have real assets that is housing that is higher. Total assets are now 6.7 times disposable income. Then there is an additional large collective pension claim which is not in there, which is maybe 1.5 or 1.7 times disposable income. Assets are really high relative to debt. Then here is another measure of interest expenditures over income published by the Riksbank. As you see now, the interest ratio is really low, historically low down to 3%. Okay, the FSA is never less worried about household debt growth and housing prices. It tries to reduce household debt growth above income growth by effectively tightening lending standards and thereby reducing credit supply. It introduced amortization requirements in June 2016 and again recently. It has induced or at least welcomed tighter lending standards of the banks. Banks have applied pretty high interest rates, 7% in their discretionary income calculations. And they also apply new or lowered internal debt to income caps. These tighter lending standards now exclude 84% of 25 to 29-year-old individuals in Stockholm from borrowing 85% of the value of an average studio in Stockholm. Here what we see is the income distribution in Stockholm, the cumulative income distribution. Before one can calculate that a monthly salary of 26,000 was enough to just borrow to buy this studio. The krona euro rate is about 10, so it's easy. So that's 2,600 euros per month. Then 55% of these young people had income below that level and could not borrow. Now, with the new standards, you need to have about 36,000 Swedish kronor. And that means about 84% have less income so they cannot borrow and buy this average studio. Then they have to go to the rental market. And the primary rental market is a disaster, not functioning because of rent control. Then they have to go to the secondary rental market with very high interest rates. Okay, but the FSA I think has some difficulties making a case. There is no or little risk to financial stability. The FSA's judgment is that the financial stability risks associated with household debt are relatively small, I'm quoting. This is because mortgage holders generally have good possibilities to continue to pay their interest and amortization, also if interest rates rise or incomes fall. The households have also on average good margins to manage a fall in housing prices. In addition, the Swedish banks are judged to have satisfactory cattle buffers if credit losses never less would materialize. Instead, the FSA argue that there is an elevated macro risk. Instead, the risk presented associated with households that are mainly concerned that highly indebted households may reduce their consumption substantially if either interest rates rise or incomes fall. And this might reinforce a future economic downturn. So high and rising debt to income ratios therefore imply an elevated macroeconomic risks. This is the argument. Okay, first on the interest sensitivity of consumption. Of course, it's right that the households cash flows are more as interest sensitive with more debt. Also in Sweden, we have a high fraction of variable rates. But the interest rates, they are endogenous, not exogenous. In bad times, interest rates are lower. Cash flow is better. This is very different from the 90s crisis when we had a fixed exchange rate and a recession and high interest rates. Now with inflation targeting it's the other way around. Actually, high debt and variable interest rates provide insurance against bad times. It's like an automatic stabilizer if you like. Also, there is a stronger cash flow channel in monetary policies easier for the Riksbank to stabilize consumption and aggregate demand. And you need smaller policy rate changes to have the same effect on aggregate demand. So maybe the risk for recessions actually may rather fall than rise when you take this into account. On the income sensitivity of consumption, the FSA has referred to three studies of the experience in Denmark, the UK, and the US. But when you read these studies, they contradict the FSA. Great paper on Denmark using essentially the whole Danish population as data. So it's not a sample, it's actually the whole population more or less. RSLs do not support any interpretation of the data that impels, involves a negative causal effect on high debt level or high debt level on subsequent consumption growths. Baker, a great paper coming out in the Journal of Political Economy, that has little or no independent relationship with the income elasticity of spending, controlling full liquidity and the ability of households to access credit. The primary reason consumption responses are higher among highly indebted households are credit and liquidity constraints. Tighter lending standards increase credit and liquidity constraints. They may cause the problems that they are supposed to solve. Anyhow, this has been... As far as I can see, when you look closer at the issue, the consumption that fell in Denmark, the UK and the US, was to a large extent or mainly unsustainable overconsumption financed by debt increases, mortgage equity withdrawals. They could not continue when the crisis came, so therefore the overconsumption had to fall. This shows up in low savings rates. So if you have an indication of unsustainable overconsumption in your economy financed by mortgage equity withdrawals, then there is a problem, reason to worry. But there is no evidence of an unsustainable overconsumption in Sweden. And the FSA says themselves that despite optimistic expectations and high margins between income expenses, in income and expenses, households are currently being relatively cautious. The saving rate is high and has increased and the consumption of durable goods is in line with historical average. So there you see the historically high savings rates we have in Sweden. And also consumption of durables is a little below average. This has been debated in Sweden and the director general in an op-ed finally said that household debt is still increasing faster than their income. And housing prices are still high. Consequently the need for action remains. But that seems to imply that all debt growth above income growth is bad. And that doesn't make sense to me. It's natural when housing prices have risen, only a fraction of housing stock is turned over at higher prices each year. So debt growth will lag housing price growth for many years. And housing prices are hardly overvalued and households are hardly over leveraged. Anyhow the consequences of this. You get a large difference between housing payments and user cost. And this large difference is large involuntary saving. Debt service to income ratios become extremely front loaded. I already said the income distribution, the income, the distributional effects. And those excluded may have to go to the secondary rental market. And you can argue that households resilience are actually lower with amortization requirements, fixed payments is a higher, becomes a higher share of income. And also households have to save in the form of more housing equity instead of in the portfolio of financial liquid assets which facilitate consumption smoothing and maybe increases variance. Here I don't have too much time left here but here is the large difference between housing payment and user cost. Remember the Swedish euro rate is about 10. So this is what you have to pay monthly payments if you buy the studio, a fee to the cooperative, the normal ownership form for apartments in Stockholm and then you pay interest after taxes and then you pay amortization. The actual user cost which is the fee to the cooperative is the real housing, the real mortgage payments after tax and the cost of own capital is quite low, only 260 euros per month. The difference is a large saving, involuntary saving that these young people have to do which makes little sense. If they can't afford this then they have to move to the secondary rental market where rents are now 10,000 kr per month and then of course the housing payment and the user cost are the same and there is no involuntary saving. Also what happens, one should remember there is an automatic amortization if you have 2% growth and 2% inflation there is an automatic 4% amortization in the sense that then disposable income grow by 4% and housing prices also probably grow by 4%. So that means that LTVs and debt service to income half in 18 years without amortization for an interest only loan. If you look there, start out with 85%, after 18 years you have 42.5% LTV. That's pretty good. Would the optimal amortization rate be faster than this? I'm not sure. Anyhow, if you don't have an amortization the debt service would follow this blue, debt service to income would follow this blue curve pretty smooth over time. With amortization requirements introduced you actually have to pay over 50% of your disposable income in amortization, they come down. But anyhow, let me conclude here. I think it's important to distinguish good and bad credit growth and I think one should have a good case if one imposes these kind of restrictions that we have talked about. And we all know it's very costly not to prevent bad growth but it is also costly to prevent good growth. So we need to distinguish them. And we require expertise in housing economics, housing market and household finance. And also housing markets and mortgage markets they are very different in different countries. You cannot, no easy generalization, you can't just look at another country and draw conclusions about your own. You have to look at each economy deeply to understand what is going on. And I think you need support over cost-benefit analysis including welfare and distributional effects, cost-benefit analysis of the kind that Vitor is doing in one of his papers. And also what we have in Sweden is a fundamentally structural problem rising housing demand and insufficient supply. Tighter lending standards reducing credit supply is not the right solution to that. Of course increasing housing supply would be the correct solution. Always use the right policy for a given problem. It's easy said, sometimes not so easy done. Okay, there are also governance issues and that's what we could talk about. I think a committee is arguably less likely to make mistakes than a single decision maker. And I also think we need mechanism for evaluation and accountability in macro-potential policy. And we can actually use the governance experience from monetary policy. We know a lot about governance in monetary policy and how to set up accountability mechanisms and evaluate monetary policy. We should apply those same to macro-potential policy. And of course we have already some experience of how it works at Bank of England with their financial policy committee and their hearings and so on. So there I think we should remember macro-potential policy is a very new policy. When we compare to economics we have an enormous amount of experience when it comes to monetary policy. And also monetary policy is arguably the simplest of all economic policy. It's a simple objective, simple instruments, well understood transmission mechanism. We are on much looser ground and we have much less experience when we talk about macro-potential policy. But I think there are quite a few things, experiences from monetary policy that we can apply in macro-potential policy. For instance regarding governance and also make sure that we have the right accountability and minimize the risk of mistakes. Let me end there. Thank you, Lars. Good afternoon. I'm in overtime but I'm also in Germany so the football players know you can still have calls in overtime. Peter, it's a pleasure to be here especially because you've been leading the work on macro-potential policy at the ECB including your drive to not only look at resilience but also financial imbalances. And I think we are there in a minority but I hope you keep it up in your block. So we have a very rich paper of Lars on monetary policy on macro-potential. The first bit is on monetary policy and we didn't discuss it but we both agreed implicitly to focus on the second part. I will be very short on the monetary policy. I only note that we are far away from when the ECB started. The monetary and the economic pillar and we only talk about the economic pillar now. When I read about what we are doing, on the economy, economic activity, transmission mechanism, I agree that financial stability should not be a call of monetary policy itself. However, what I really miss is that the banking and the financial system is crucial for monetary policy. If you don't understand it, then you miss a lot. So we have to be careful with all our inflation targeting and we forget about the importance of the banking and the financial system. We discussed at the start of the afternoon the role of banking capital and that kind of thing. Yu Xin has written on it. Moving on to the main topic, different, separate, from monetary policy and then we need to have different instruments, the famous Tinbergen Rule, the first Nobel Prize winner and lucky to be at the Rasmus University where he studied and he has a famous quote, the profit comes from the distribution, that's the quote I see each day if I enter the building. But this is, I think, very sensible. So if we really want to stay in financial stability, we need separate instruments. Separate decision-making bodies, I also agree, but then how to organize? I will come back to that. The big issue is what I already said is that only resilience and that's what I hear, especially among central bankers as the main thing or should macro-potential policy also target, moderate the financial cycle and I strongly believe yes, but I'm in a minority. So I will show you some crafts. The next thing is, can be distinction between good and bad credit growth. That shows already that central banking is an art because that is extremely difficult and I think that's a key issue. And the new idea last put forward is let's take an interest course or an interest expenditure to income rate show rather than the current price-to-income ratio and if we knew this new measure, then everything looks fine. But is that a good idea? That's the question I would like to put to you. So first on the separation, so I'm educated a bit in the Dutch tradition but also in the British tradition and in both traditions we look up monitoring financial stability as two sides of the same coin and financial stability has always been in the remit of the Dutch central bank and the Bank of England and the big overlap is the role of the financial system which is relevant for both sides. They're both macro, they're about the system and finally you need independence. So if you think there is an issue with imbalances and if you need to stop imbalances we all know if you are linked to the political cycle you're always too late in the game because politicians, they like the voters so you're always too late in the game so you need an independent authority if you want to be in time or try to be in time. At the advisory scientific committee at the European systemic risk board we did a paper and Andre was also, Andre Sapir was also involved and the first conclusion we had was quite strongly should not be a government or a government agency and in some countries the supervisors climbed close or even part of the government. Secondly, you really need the macroeconomic focus and in a paper 16 years back with Charles we looked at the type of staff central banks at supervisors and then we found very strongly that supervisors, micro-supervices had far more lawyers, accountants and only a minority of economists and look for example at the SEC in the US is a very clear example they did prudential policy of investment banks but they didn't do anything because they didn't understand it and central banks on the other side they have a bunch of economists and that is the leading culture. So if we think macro prudential is macro oriented you need an economic focus body to be successful and my favorite is the bank of England it's a pity that Mark left with two separate bodies on that I fully agree with Lars but then both at the central bank we have the macro focus. Now to the objective and my main issue so only resilience not imbalances and to be truthful to Lars he mentions excessive credit code that he agrees with that that we should address that and in the recession interest rates are low so if you get the recession and the financial crisis and interest rates are low so anyway then it is less a problem that assumes that the business and the financial cycle are related and I will show you in it that it's not always the case and you have this tendency that if you see that you find the problems at the structural side and I'm not saying they are not there but lack of housing supply this lack of housing supply was also there during the recession only people knew they couldn't buy a house they wanted a house but they couldn't buy it so this lack of housing supply is not unique only people have now money and want to buy a house so and remember Greenspan the new economy this new economy is completely different because the structure of the economy is new different because of the internet and then a few years later when the internet bubble burst we found out that the business cycle was still alive and we thought earlier with the new economy that the business cycle is no longer anymore and I would call this that this time is different fallacy that if you are an observer in the current time then it's always more difficult to take a distant picture and that you have a tendency to say that today is unique so today credit growth is fine and in the past it was wrong but this time the house prices are fine and credit growth is fine however the financial system is known to have become extremely pro cyclical and not really related to the business cycle so there's not to say in a financial crisis interest rates will easily come down and no problem so I take two elements from that so I would say there is really a need to dampen to moderate the financial cycle like we moderate the business cycle with monetary policy and fiscal policy and I think in the earlier sessions that has been extremely clear there's a strong feedback loop from the financial system to consumption and there's an extremely strong effect which I will also show so we're now coming to the graphs to keep you awake the first time when I show this picture you all know it from Claudio Borio the business cycle is red the financial cycle is blue that this housing prices and credit which are of course correlated when I showed it back in the Netherlands my colleagues of the free universities thought oh Dirk is wrong I had the source BIS so they looked it up and the picture is true first the cycles are not related second business cycle policy second order financial cycle policies first order I mean that's quite clear from this picture so not to wish to moderate the financial cycle would be really a pity and I did together with Peter Wietz at the Dutch Central Bank some simulations that like in the earlier papers at the start of the afternoon that the level of equity in Dutch it is a level of equity matters so the blue one with 10% equity then you have a moderate mortgage credit cycle if you go to 5% where we are today you are already a bit worse if you are 2 or 3% like before the crisis then you get a huge cycle and we looked for a long time what was really driving it and in the end the answer was extremely simple the profit maximizing behaviour of banks so in good times you earn more on your assets than on your debt the leverage and in bad times you earn less on your assets than in debt so the bigger leverage the lower equity the stronger this incentive to maximize your profit so really the normal behaviour nothing to do with create or with wrong things it's just a normal commercial behaviour of a commercial bank and for the instruments that shows that higher equity is not only useful for micro purposes but would also dampen the cycle in addition to loan to value and other measures now to housing and consumption that's my last point and then we can all go to dinner so the blue one, the light blue one is house prices they fall and then you see consumption also going down and why do I show this one this is the Netherlands in 2011-2012 we were really thinking what is going on, the economy keeps on going down because of government spending curts and all kinds of things and we didn't know what was happening if you look at this picture in 2011 the light blue is Germany this is about consumer expectations and we know if consumers are happy they spend, if they are not happy they don't spend and you see this 2011 really rock bottom thing and it's really related to the housing one out of six households had negative equity in Germany you all know it people don't own houses there was no negative equity so there was no consumption effect because our economies are quite similar we are not as good in handmade industry as Germany but otherwise we are quite close but this is really a huge difference and in the start of the cycle when house prices were going up we had higher consumption because we were consuming like the UK and the US the overvalue in our houses so this is a very strong effect so I'm not worried about banks they don't lose money because of full recourse and it's really the consumption effect of financial instability and this is really a nice graph also from Klasknots people at the Dutch central bank it shows at the Y axis you have the correlation between house prices and consumption and at the X axis you have home ownership so the more home ownership you have like in the UK in the US, in the Netherlands then you get a really strong correlation between house prices and consumption if you don't have low home ownership then you don't have at all this relationship so that's why in Germany it is not an issue and in high home ownership country and I think Sweden is part of that you have this connection and I think that fits in at the start of the afternoon where we really say we have to look at the heterogeneity the differences between countries and this is a strong one and that shows that it's very dangerous to ignore this channel and concluding so we have full agreement with last separate objectives but the financial system is connecting the two and that's why the banking and the financial system is always very close to the central bank the only one who gets liquidity support if needed and not the wider economy if you want to dampen financial imbalances you need a macro approach a micro approach wouldn't do and Sweden looks fine on this static interest expenditure to income ratio like all countries look fine on the government debt on the current low interest rates what happens if interest rates co-op what happens or we wouldn't increase as child set interest rates because we would get insolvency so we are in a trap then and at least in the Netherlands if you have a contract of business supervisors doing a dual test can you pay your mortgage today can you pay your mortgage at the fixed rate of 5% and only when you meet both tests you can get the mortgage and that's far more sensible than calculating with today's cheap money and then interest mortgages are cheap you can buy a huge house and if you have this 5% and that is one mechanism and if we will do that then I think Sweden and Amsterdam with 15% increase in house prices for 3 years and nothing happens then it looks less rosy the picture thank you