 Hello and welcome to the second day of the ECB forum on central banking this year dedicated to the role and policies of central banks in a shifting world Yesterday we heard the introductory speech from ECB president Christine Lagarde We discussed implications of global changes for central banks and had a panel discussion on inflation objectives Structural forces and communication in case you missed it all of yesterday's content is available on our website Today we will start by discussing macroeconomic stabilization frameworks in the new economic environment before turning to monetary policy instruments and financial stability and We will conclude the forum with the policy panel Let me remind you that all sessions and panels during the forum are on the record and our webcast live We encourage you all to join our social media conversation using the hashtag ECB forum Our participants will be able to interact live with our speakers For that you can raise your virtual hand already ahead of the Q&A time in each session I would request again today that participants keep their questions to 90 seconds as we have a full program and time will be limited But now let me welcome Fabio Panetta member of the executive board of the ECB who will chair the first session today the floor is yours Thank you, Terry. Good afternoon to everybody and Welcome to this second session on macroeconomic stabilization frameworks This session brings together two topics that are widely debated in the current environment of low inflation and low interest rates The first topic is the interaction between monetary policy and fiscal policy Which according to many has become more important for macroeconomic outcomes The second topic refers to the monetary policy challenges that are associated with the decline in the natural rate of interest At the root of both issues the monetary fiscal interaction and the new challenges for monetary policy Lies the fact that central banks are more likely to hit the lower bound of nominal rates than in the past as a consequence Non-conventional monetary policy measures have become more important for avoiding a low inflation equilibrium and fiscal policy in this phase has become an important channel for transmitting those measures as the fiscal multiplier increases at the effective lower bound Before giving the floor to our speakers, I wish to make one observation The challenges related to the low interest rate environment are important not only in the short term for the conjuncture But also from a longer-term perspective for two reasons The first reason is that the factors that are associated to the full of the natural rate of interest such as productivity growth age in population or scarcity of safe assets are slow-moving Hard-wired and persistent Therefore reversing their impact on interest rates appears to me to be challenging at least in the short to medium term Second if we look at long historical time series of interest rates We would soon realize that low interest rates are not that unusual On the contrary what seems to be unusual is the high level of interest rates of the 1970s and The 1980s this may imply that if and when central banks will be able to normalize Their instrument set again They will still have to confront a high risk of hitting the lower bound in fact Policy normalization may not mean returning to high rates But rather dealing with the reality of a word with low R star This is the focus of this session in which we have two outstanding economists Professor Klaus Adam from the University of Mannheim will have 20 minutes To present his paper entitled monetary policy challenges from falling natural interest rates We will then have a 10 minute discussion by Arjas Bordone from the New York Fed I will then open the floor for questions which will be allocated 90 seconds each to ask for the floor. You should raise your electronic hand with this Klaus the floor is yours So thank you very much Fabio. It's a pleasure to be here and to speak at this event on monetary policy challenges from falling natural interest rates Let's see. This is not working. No Here we go. So central banks in advanced economies have been confronted with a number of adverse macroeconomic trends amongst which were generally declining trend growth rate over the past decades and I have strictered by that Corresponding decline in the natural rate of interest, which is the real interest rate consistent with stable inflation Of course, these concepts being theoretical concepts at any point in time somewhat uncertain what the level of the trend growth rate is of the economy and also what the natural rate of the economy is but there is a wide agreement amongst Economic observers that there are downward trends in both of these over time and And This downward pressure has obviously put some downward pressure on nominal interest rates over time To the extent that many center banks are now pushed for extended periods of time To what is called the effective lower bound for nominal rates That is a situation where policy rates are either close to zero or slightly below zero So for the European center bank, this has been the case Since July 2012 when the main refinancing operation rate Has reached zero federal surf had seven years of zero interest rates up to 2015 and then again following The pandemic at the start of this year Japan has had earliest, you know decline of a natural rate and also has seen the lower bound virtually without interruptions being reached ever since 1999 So this is the situation where we are We find ourselves increasingly at the lower bound I think it's important To recognize that the effective lower bound is in fact a real constraint for monetary policy And then there are two pieces of evidence that we can put forward to show this So the first if you look at center banks that have reached the effective lower bound These are the center banks that have persistently undershot their inflation targets and They have done so despite the ample deployment of a large variety and No, and in big terms of quantitative easing measures and in some times they have done this even Despite the joint deployment of a quantitative easing measures and fiscal policy measures And and the euro area is no exception to this if you look at the HICP inflation average Since the point where the main refinancing operation rate has reached zero in July 2012 The average inflation rate since then has been zero point eight percent Which clearly is below two percent, but not anywhere near is close to two percent The second piece of evidence that sort of speaks to the effective lower bound being A real constraint is if you look at What these measures do the quantitative easing measures actually do While we have sort of reasonable good estimates of how they affect financial market prices like long-term bond yields Risk premium default premium of various kinds. There is quite a considerable degree of uncertainty How these measures affect macroeconomic outcomes in particularly inflation And this is also something that Falka Villand has addressed yesterday So there's a reason to believe given this uncertainty that we are not entirely sure to what extent these measures actually are suitable and effective in substituting fully for normal interest rate movements Now I'm afraid to say that these two trends that I've just mentioned are not the only adverse trends that have, you know Come up over the recent decades Besides declining long-term growth rates and declining natural rates of interest There's evidence that asset price volatility has also risen over the same, you know period In particular, we have seen a number of large and very persistent housing price cycles in many of these economies in advanced economies And we have seen what is the near back-to-back repeat cycles in equity markets So these are of course challenging for monetary policy because these asset price cycles have all kind of real implications and may generate also sorts of credit market distortions of various kinds and Perhaps triggered by that and there is additional evidence that as the average level of the natural rate has come down Its volatility has increased at the same time Which is of course particularly bad news if you're confronted with a lower bound constraint on normal interest rates Now let's have a look at the evidence of asset price volatility Of course, it's generally hard to pin down the volatility of asset prices for two reasons The volatility is large and therefore uncertainty about its precise level is large and asset price swings are very persistent So that over time one doesn't get a lot of independent observation and it's even harder You know to measure changes in asset price volatility over time So the only thing one can really do to deal with these issues is to compare long time spans and To see whether over those time spans there's a trend and this is exactly what I'm going to do I'm going to look at asset price volatility over the past 30 years and then compare it to the prior 30 years to spot some trends Let's have a look at housing markets first So this is a graph that is showing you the standard deviation So the measure of volatility of a basic housing market valuation ratio the price to rent ratio so the blue bars are the price the standard deviation of the price to rent ratio prior to 1990 and The red bars are the standard deviation of the price to rent ratio post 1990 and the set of countries is Canada France Germany Japan UK USA So you see that the point estimates have increased in all those economies and in some of them quite substantially But because of the high uncertainty associated with these point estimates, but all of these increases statistically significant at all times But some of them are as you can see by the p-values shown at the bottom of the table now if we compare, you know, the joint evolution of Average natural rates pre and post 1990 and the standard deviation of The changes in the standard deviation of the price rent ratio we get at this scatter block on the x-axis You have the change in the average natural rate in these economies pre and post 1990 and on the y-axis you have the change in the standard deviation of the price rent ratio in housing markets And you can see that all these economies fall in the upper left quadrant They experience quite substantial changes in natural rates and an increase in the price rent ratio volatility over this time period If we then look at the same picture for equity markets We see something very similar also not as stark as in housing markets So again, we have the change in the average natural rate pre and post 1990 on the x-axis And now the change in the standard deviation of the price dividend ratio a basic valuation measure for equity market prices on the y-axis and you can see most of the economies with the exception of France and Japan fall into the upper part Japan obviously being an outlier because the sample split here occurs around the peak of the late 1980s Japanese stock market boom and So that the run-up is in the first phase and the The decline of asset prices in the second sample period But overall the message I would say is very similar to that of housing markets So now What to make out of this Observation that asset price volatility seems to have increased Well one very natural reaction by Congress would be to say well It could all be quite efficient because in the world in which Real rates and natural rates are lower on average It is efficient that asset market volatility increases because any persistent movement fundamental now gets discounted less And therefore should have a higher effect on current market prices So this is all fine But I would caution to this because there is mounting evidence that asset price volatility Cannot be entirely efficient In particular if you look at investor surveys that ask investors about their capital gain and return expectations and markets There is very clear evidence of the presence of investor optimism and pessimism and in particular of quite significant deviations from the assumption that these investor expectations are consistent with rational forecasts and Perhaps even more importantly the patterns of optimism and pessimism of form that they actually amplify as a price volatility So to give you an example Let's have a look at Expected and the actual capital gains In housing markets, okay So we can run two very simple regressions that will give you a flavor of the kind of results that are around So the first regression is Regressing on the left-hand side investors expectations about the capital gains to be made in housing markets on a constant a a coefficient C and then a basic Valuation measure of how richly the market is valued here at the price rent ratio in the market So this tells you, you know, the coefficient C is going to tell you how Investors expectations co-move with how richly the market is valued The second equation is very similar but puts on the left-hand side The realized capital gains as they mature Down the road and represses those on the valuation measure the price rent ratio and the very typical finding in is that we can make is that in actual investor expectations of Investor expectations are prosyptical so that coefficient C is positive Which means that investors are optimistic about future capital gains when current market valuation is high Realized capital gains, however, tend to be counter-syptical So, you know, somewhat puzzlingly The realized capital gains tend to be particularly low when the market is valued richly To give you an example of this kind of result I've run this regression on the Michigan household survey over the period 2007-20 Where we regress expected capital gains and actual capital gains on the price rent ratio And you see the positive coefficient on the expected capital gains you see the negative coefficient on the realized capital gains and A test that takes into account that asks are these coefficients actually equal as they should be a forecast to a rational Is strongly, you know, rejecting the notion of rationality and of course the fact that expectations are high When market valuation is high strongly suggests that the reason market valuation is high is because Potentially investors are too optimistic about the future returns in these markets Now it's important to note that this phenomenon is not confined to housing markets We have shown and documented the same pattern in stock markets with general stock market investors And they are of course additional dimensions along which expectations display biases that I have no time getting into But if you ask me about, you know, overall looking at all those servers over time How would what's the simplest way to explain the expectations the capital gain expectations in of housing investors and stock market investors retail stock market investors Well, if this evidence is all consistent with these expectations, these investors weekly extrapolating past capital gains into the future It gives rise to the sort of expectation of patterns that I've just documented Now this is boring, but it is even more boring in a world where interest rates and real interest rates are low And the reason is that the sensitivity of asset prices to any given amount of investor optimism and pessimism That is not justified by subsequent outcomes on average Is higher when the real interest rates are low So asset prices respond more to this Optimism and press pessimism When the natural rate and the interest rate is low on average And if investors then extrapolate from past asset price movements into the future This is going to increase the likelihood of belief driven boom and bust cycles And this is one way to explain how asset price volatility has increased at the same time as natural rates of interest have come down Now what of course the question is what are the monetary policy implications of this Asset price movements that are potentially inefficient And in particular if these inefficiencies increase in size as the natural interest rates fall Well, we know that inefficient volatility and asset prices can generate all kinds of distortions For instance, if can have direct effects on resource allocation You get a housing price boom that is driven by housing price optimism You get over investment in housing as we may have seen in some of the columns in the past You get credit misallocation You could affect some balance sheets and all kinds of effects What we have shown is that these effects have the potential To increase the volatility of the natural rate So this is also an avenue not only to explain higher volatility of the asset prices But also to explain why simultaneously Low levels of the natural rate have been associated with a higher volatility And that is of course very bad news because now the effective lower bound becomes more springing because of two reasons Because its average level is falling and because the volatility around that average level is increasing If we look at the empirical evidence on the volatility Of the natural rate, there is some evidence that suggests That volatility of natural rates have in fact increased at the same time as the average level has fallen So here you see again a scatter plot where I plot the change in the average natural rate on the x-axis And the change in the standard deviation of the natural rate on the y-axis And you see that most of the economies again fall in the upper left quadrant Which suggests that volatility of the asset price of the natural rate has in fact done up as the natural rate fell on average So then what are the monetary policy implications that come along with a more stringent Effective lower bound? Well, I think our model suggests it would be optimal to regain some room for conventional monetary interest policy By increasing the inflation target or the average inflation rate implemented by monetary policy Of course, the increase in inflation target doesn't come without cost Higher inflation is going to have welfare costs But these welfare costs are going to have to be traded off Against the increased ability to stabilize the economy when the inflation target is somewhat higher Now the quantitative resolution of, you know, these two effects, the higher welfare costs of inflation and the increased ability to stabilize Depends in important ways on while one interprets the observed increase in asset price volatility In particular, if one thinks the asset price increase, the increased asset price volatility is efficient Then it is, you know, the welfare cost of inflation that is the dominant concern If you think that the increase in asset price volatility is not entirely efficient Then the increased ability and desire to stabilize more takes more of an effect So let's illustrate this Let's first have a look at the world where the increase in asset price volatility is entirely efficient And this graph shows you the optimal increase in the target due to the presence of a lower bound On the y-axis as a function of the average level of the natural rate So for average levels of the natural rate around 2 to 3 percent, you know, the lower bound doesn't do a lot And, you know, there's not a great desire to increase the inflation target because of the presence of the lower bound Because you're far away from it But as the average natural rate falls, you see some effect But the overall effect is relatively muted. It's around 0.4 percentage points So this was the case where asset price volatility was efficient and therefore the only fact that is present as the natural rate falls is that you get closer to the bound But there is no effect coming from the volatility of the natural rate increasing at the same time So what happens to this picture if you sort of think that at least partly asset price movements are not entirely due to fundamentalists Well, this curve shifts up quite considerably and also becomes steeper. It becomes more responsive to the average natural rate So now as the average natural rate falls from a level of say 3 percent to a level of 1.8 percent, the inflation target increases by a full percentage point So this shows you how, you know, the optimal increase in the target is actually the average natural rate falls from a level of, say, 3 percent to a level of 1.8 percent So this shows you how, you know, the optimal increase in the target depends, to a large extent, how you interpret the increase as a price volatility Now, of course, and this is something that has been raised in the discussion yesterday, the undershooting of the current target is a problem. Is it sensible to raise the target in a setting where you currently undershoot And I think there are two views to this. There's the somewhat optimistic view that by purely raising the target alone inflation expectations will follow suit and will do largely the job But there is a pessimistic view, which is probably, you know, of some relevance that By raising the target, the only thing you effectively do Is increase the distance between the target and the actual inflation rate, so the target shortfall And it was also Mentioned yesterday that this will have potentially a reputation of damage So how, if one is concerned or is convinced that some increase in the target were warranted, could one get from here to there Now the alternative approaches to deal with these pessimistic concerns One of it has been mentioned yesterday by Jorica Lee. He said that we should act opportunistically Once we have increased the target, we have reached the current target We can sort of think about it and increase the target to where we think it should be But I think this has costs and it has Opens the center bank to some criticisms down the road that it's raising the target at the time It's potentially overshooting its previous target So there could be some value of pre-announcing that the target will be revisited once the existing target is reached And the reason being that maybe you will look less opportunistically And also you're going to potentially at least get some of the expectation and anticipation effects Without suffering the reputation of damage of increasing the distance to your current target I have no time to talk about This I think I'm largely out of time. So let me wrap up So center banks confronted with lower average levels of the natural rate And higher volatility of asset prices and also higher volatility of the natural rate Should certainly rethink their inflation targets The welfare costs of higher targets should be traded off against the increased gains to stabilize In the presence of an effective lower bound constraint And quantitatively these trade-offs going to depend amongst other things On how one is going to interpret increased asset price volatility Thank you Thank you Klaus We now have the discussion by Arge as Bordone Arge you have the floor Seems to be some problem in the connection Maybe I can take this a few seconds to encourage you to raise your virtual hand to ask questions to Klaus Why we are waiting to be connected with Arge Hello Arge, yeah Yes, thank you Thank you Fabio and thanks for inviting me to participate in this distinguished forum And giving me the opportunity to discuss Klaus paper The whatever I say is my own opinion and don't reflect The opinion of the Federal Reserve of the Bank of New York So the Paper addresses as Klaus said that dual challenges come into monetary policy from the decline in the natural rate of interest An increased volatility of asset prices or housing prices and in this discussion I will just bring a little bit More up front the analytical background to obtain the conclusion that Klaus Very clearly set out and discuss a little more the implication for the central bank strategic review The motivating observation of Klaus were very simple There is a decline in the natural rate of interest in many countries You see on the graph on the left the estimate of the natural rate of interest According to Alston, Labak and Williams that we we put You know on an interactive version on the website of the new york fed updated quarterly and you see that the Corresponded this natural rate decline that is a decline in trend growth in most of these countries in The factors that are behind this decline in trend growth and therefore the decline on the natural rate of interest are many structural factors that have been named during mentioned during this Forum already the second motivating observation Is something about the volatility an increase in volatility in housing prices and asset prices here i'm reproducing These graph that Klaus has presented the standard deviation of the price to rent operation I will mostly focus on housing prices in this discussion The pre and post 1990 differ quite a bit I had some quibbles on how this is computed because post 1990 there was a big trend In housing prices determined partly by the decline in the interest rate, but I will defer this The nice thing is that the country cluster in the upper Left quadrant when you plot the change in the average natural rate across the two period Versus the standard deviation. So these are the facts Are challenging because of the vicinity of the elb as a klaus has Shown the many countries have now extremely low or negative short term rates And the inability to use the interest rate for stabilization purposes Risks undershooting the target and unanchoring expectation in the housing price volatility add the major layer to this risk to the Compounds the vicinity of the elb because increased the natural rate volatility as klaus has shown so the analytical framework that brings klaus to the conclusion of The recommendation the two main takeaway if you want of his paper Is that um as an ocassian model has the enough feature to allow to address the The problem at hand so has a new housing sector and most importantly Has a departure of expectation from the assumption on Rational expectation in housing prices or asset prices evolve in somewhat extrapolative ways now klaus In other people have used other type of departure from rational expectations The essential thing is that this departure may give rise to inefficient fluctuations The paper accounts the paper the structure the framework in which results are derives account for a binding elb constraint and in with this model you can set a well-defined optimal monetary policy problem Or maximize the welfare in the presence of a bunch of constraint This picture is very important for two reasons. First of all has the main message That is optimal what the klaus calls optimal inflation target increase due to the zero lower bound is Different In so far as you have different in level in so far as you have subjective expectation versus rational expectation So in so far as you have some inefficient asset price fluctuation but what is important is that the Soon as you As soon as you lower the average natural rate you have a higher probability of hitting the zero bound so the Two main conclusion from drone from the the whole talk and the these pictures Revealing is the optimal inflation target and I put it in quote is generally higher The lower is that the our star the source of price volatility matter and policy should lean against the asset asset Or housing prices now this tuition is very simple. I won't repeat it again because it's been Stressed very well by by klaus. However, I wonder if this really means that we need to have a higher inflation target There's optimal policy calls for a higher inflation target. I would say not really In in particular if we look at how the optimal policy Here I'm saying things that most of you know perfectly well The optimal policy result in models of that kind Is implies a great optimal criterion Which described the trade-off in the absence of ELB between inflation and output gap But when there is a ELB, there is there are terms Involving the constraint and the severity of the constraint represented by this Lagrange multiplier of the constraint so if you Think of a gap adjusted price level, which I call q This constraint it tells you how much delta q has different from zero and the different from zero Depends on the severity of the lower bound So if there is a lower bound if the lower bound doesn't bind you go back to the usual trade-off in the absence of zero bound If it's bind at any time zero then qt has to decline to pick up later The risk of oversimplifying the matter if you on the upper Graph you I plotted the incidence of the zero bound. So the shadow cost of the zero bound that become at some point time t zero Positive so you have a constraint here When the lower bound the binds that the gap adjusts the price level of declines But then there's a promise of pickup. So optimal policy in this context simply represented by this graph Promises high future inflation when current policy is constrained by the zero bound The decline in the price level will be compensated later and the extent of the compensation Is determined by the extent of the binding constraint Importantly after this correction the price level path has moved up But policy returned to target exactly the same Target rate that before the constraint now in with this framework in mind again. This is purely simplified Representation if you have either a lower r star or you have a larger shock hitting on the same level of r star You have a more stringent constraint which determine Bigger decline and a bigger uh catch up later So what is the result of this optimal policy? Is that when you compute inflation rate average over the period of undershooting in overshooting results somewhat higher so, uh, I would say that the correct interpretation of this of this Optimal policy in all the pictures that we have seen is not that optimal policy calls for a change in the inflation target And I think in fact the class was more careful with words talking about average inflation But um, it's uh the implication of conducting optimal policy in this environment. It's a in this model environment It implies that there is a higher Average level of inflation because inflation has to average periods of undershooting and success subsequent corrections And why making this distinction is important? I think is important because it's problematic to talk about an increase in the inflation target as this is understood If it is understood by the public as a long term concept and the long run Target is what the central bank target accepting periods in which Has to deviate from the target and will correct later And I think a commitment to correct policy With no change in the long run target has the advantage versus increasing the long run target Of avoiding the cost of permanently higher inflation as we saw You know claus said in fact you have to balance the cost of higher inflation permanently the welfare cost with the risk of The cost of non stabilizing the economy. I think the feds framework review process and the new policy strategy underscore perfectly well this difference because the feds as issued The what we call the consensus statement that means the statement of longer run goals and monetary policy strategy There was first issues first time in 2012 and has been amended in august 2020 The fed has restated the existing numerical inflation target at the rate of two percent measured by the um PCE the personal consumption expenditure index But opted for a strategy that addresses shortfalls from that target And so as in order to anchor long-term inflation expectation at the level of the target Will make up for the losses with period of inflation temporarily above the target and as the fed chair The fed chair powell has indicated in the jackson hall speech This approach could be viewed as a flexible form of average inflation Targeting and I think this is very much in line with the optimal monetary policy in that that is behind behind the paper that Klaus has presented now um Obviously an important piece is the whole analysis is the role of subjective beliefs subjective expectation as claus has shown may lead between efficient asset price movements and may The natural rate more volatile that under rational expectation By the way, one has to consider the oldest estimate of the natural rate are extremely uncertain. So there is a you know a lot of Bounds around this estimate Now under subjective beliefs that monetary policy ultimately leans against asset price movement in this conclusion That here is reached under in the in the paper that I cite here The claus site that under paper by Keynes and Winkler This conclusion will hold under extrapolative expectation But hold also under more general form of belief distortion as a claus as another paper will be woodford and also When beliefs are extrapolative doing best even when there is a macro prudential tools available So class of models is now exploring also, you know, how much monetary policy has to take On the financial stability if there are other tools available It is important implications call for more empirical analysis both of the volatility of price in expectation formation to conclude The paper addresses key challenges present optimal policy implication The optimal policy results in period in which we have Ultimately higher inflation and this policy rationalize a particularly aggressive form of average inflation target Of course, there are open question and uh claus already mentioned, but I'll stop here. Thank you Thank you, arja I would now give the floor back to claus for A reply to the comments by arja. I would Still encourage you to raise your hands And ask questions. You have the floor claus Thank you very much. Actually for a wonderful discussion. I um I agree with all what you have said and maybe just one remark on the semantics of an average inflation versus the optimal inflation target. So the logic of the exercise is that The fundamental idea is that the monetary policymaker steers inflation and the economy through changes in real interest rates Now at the lower bound this cannot be done anymore through changes in the normal rates And the way to then stimulate would be to promise conditionally in those situations Higher future rates of inflation and this is in fact something that is embedded in the average inflation targeting Concept that has now been adopted by the federal reserve But it's also important to note that there's a fundamental asymmetry in that concept As it comes out of the model you do that at the lower bound when you are constrained But you do not do that if you want to tighten So when you want to tighten you do not have to promise deflation But you can tighten by just raising the normal interest rates So the effect of this asymmetry is To raise the average rate of inflation above what it would otherwise be in the absence of the lower bound And I've thus called this average rate of inflation that materializes under this optimal Strategy as the inflation target. But of course, it's true that this is not the inflation target At all points in time that you would always target at the medium term Thank you Klaus We don't have Questions at the moment Maybe I can can ask a question myself Klaus you you Mentioned in your paper that the four in the falling natural rates Justifies an increase in the optimal inflation target But as you mentioned, this could raise communication challenges as long as inflation remains Below aim, which is the situation that we are in in the euro area You suggest a sequential approach. That is the central bank should first strive to Get to the inflation aim to bring inflation back in line with with its target and then announce a higher inflation target but the other Possibilities have been proposed. One is to enhance the role of fiscal policy another which was proposed by alja is to Revise the existing numerical inflation target Can you elaborate on how your result would hold in case one was, you know, analyzing alternative possibilities? Well, this is a very difficult but also quite interesting question. I mean the models that we typically wrote right down do not have Typically do not have issues that relate to the reputational concerns of the central bank So the question is of course If you increase the target and you are not yet going to be Increasing for a longer time period the increased target Will this be costly for the central bank because it's going to affect negatively The reputation and therefore maybe lead to an unanchoring and we do not have this in the model and it's a legitimate concern to entertain Um, that's why I have proposed that a two-step approach Where maybe a more regular frequency of strategy reviews allows Maybe for a pre-announcement of a regular revisiting of the target And it may be understood then by the private sector that the target may be adjusted in certain directions And this could have beneficial effects without having the reputational cost Now with regard to your question on fiscal policy I'm a I must say I'm a little bit skeptical because the Japanese experience of you know, the last years Certainly where fiscal and monetary policy have been strongly pushing in the same way is not exactly encouraging And this is probably one of the big disappointments of this these experiments and The united states example is perhaps a little more encouraging on on this front But overall it's not It's not clear that this is a panacea Thank you. We now have one Question by harold urlich from the university of chicago harold you have the floor I think you and thank you claus for the very interesting talk. I wonder whether the result about monetary policy having to raise, you know, maybe temporarily its inflation target Holds more generally and let me let me get to that with a with a potential critique of this idea that subjective expectations are really at the heart of it all It's probably, you know, the alternative view to this discrepancy between observed expectations and subsequent market developments is That this is all due to the market discount factor That by appropriately taken to count the You know, the the risk rate of probabilities for certain states in the future. There wouldn't be that discrepancy. It's just when you take actual average You know, realizations that you that you get this mismatch. So I mean, it may be hard to truly you know Get from the data where the subjective expectations are driving this over. This is just movements in the stochastic discount factor And expectations are rational but if I take the fact as given and and you know, there was some discussion on this that indeed as we as we get to the to the lower rates That the volatility increases there were the there were the point that you want to maybe then temporarily at least raise inflation target May still seem to be correct. So the question is for the result for monetary policy Do we really need to believe that this all driven by subjective expectations or is this point more general? Klaus, you want to take up this answer now? Maybe Arja, you also want to comment on harald's question Yeah, let's let's see what klaus first says let klaus go first Okay, so thank you. Thank you harald for this question. So i'm very happy you asked it because You know john cochrane and others have made similar points that say the investor surveys may not be what we take them They may not report Expectations but risk adjusted expectations so in response To this we have actually written a paper which is forthcoming in the journal of monetary economics to get together with stefan nagel and mitri maglev Where we actually looked at this point and there's actually no way You can reconcile The expectation of patterns that we see in survey expectations with the notion that they arise from risk adjustments And so this is in this sense pretty bad news And I think you know there's a big discrepancy between the chicago view which is always that Markets work efficiently and what the general public is doing If you tell we've write down as surprising models in which people are pessimistic about future returns When the market is valued highly to the normal investor this sounds like somewhat unusual And actually this is exactly what we see in the survey. So I think it's very puzzling And it's not explained by discount factors now to your second part I think it is the point i'm making is more general To the extent that as a price fluctuations, they may themselves be inefficient But even if they were efficient They could matter in other ways if there were additional constraints in the economy like boring constraints Collateral constraints maybe default incentives and so on and then there would be also incentives to lean against those As a price movement in these more richer environments, which we haven't yet studied Yeah, if I may add something fabio. Yes. Um, so I want to cite that Some research that has been done some paper published on The survey of consumer expectation that we ran at the fed examining the behavior of housing price expectation And so what this result and this is with this experiment control experiment that's a paper published in 2019 on wristad by fuster in Zafar They conducting this experiment they find that indeed expectation of individual expectation household expectation about Housing prices do not behave like rational especially at the medium term frequency. So they tend to They don't tend to revert as the actual price Do and that's in in line with what claus was saying. So they In the very short period that they Have are directionally. They say directionally rational, but they don't they under Shoot but on the medium term they fail to pick up the reversion in in prices So there is a certainly argument that housing price expectation subjective are quite different from rational Thank you. We have another question by helen rey from the london business school helen. You have the floor Thank you very much I would really like to have the view of claus on the role of Macro potential policy, which has been mentioned by by arger and in particular whether He is mostly concerned about short run volatility in asset prices or more low frequency Latilities leading to asset price bubbles Which are arguably Maybe taken care of at least to some extent by macro potential tools. So we very much want to have is you on on this Oh, yes. Thank you helen for your question. Um, so indeed we are not at all concerned about Daily volatilities In equity markets or about these sort of short-run fluctuations It's more about the long run cycles That have also proved to be, you know, most damaging to the economy now Ideally if you have a second instrument to effectively deal with those Those cycles that would be wonderful. I just think that the macro potential framework In europe is very incomplete And it is also not tried and tested For instance the incompleteness that is worrying for instance is It's mainly geared to banks, but it doesn't address the non-bank financial sector, which is becoming increasingly relevant And the second point is probably then that Monetary policy as jeremy stein has said, you know, it has the potential to get in all cracks While the regulations imposed by macro potential Authorities may potentially be circumvented by certain financial innovations That then take place in the financial sector. It's great if it were to work, but i'm somewhat skeptical Arja, you want to comment on helens question? Well, I think it would be wonderful to have a more macro-prudential Regulation of tools that that allow to, you know, forestall the fragility in the financial system when there are Not looming when there have been excess taking excess risk taking but again I think there should be more discussion on that and I I don't like the idea that The burden has to go on monetary policy alone, but So I would be very much in favor of Pushing this measure of macro prudential regulation all over and not just in the europeans in in the us, of course Uh, thank you. We don't have any other questions. So, uh, I think we can take this session to a close but Before doing so, I would like to Emphasize one implication that I draw from discussion we had Uh, the low level of our star means that we face a conundrum Not only has monetary policy less scope for expansion But it may need to make more use of it given as claus has shown in his presentation The highest price volatility implied by the low rate environment But central banks have less policy space to address the risk of low inflation than the risk of high inflation This means in my view that they should be less tolerant for inflation drifting downwards away from target On this note, let me thank once again Klaus arger and all the participants who intervened for this very rich discussion I now hand over to terry Well, thank you very much. Fabio and speakers for such an engaging session Before a short break, let me take a brief moment to remind all of you of the phd students in this year's ecb young economist competition They are with us participating in this forum and you will find their research papers and posters on the ecb website The jury actually as we speak is now in session And the best paper will be awarded the price of 10 000 euro later today at the end of this forum We will now take a break and I look forward to seeing you all back here at 3 15 sharp