 Welcome back for the final part of the conference. After the concluding panel in a short award ceremony, President Draghi will hand over the price to the winner of the Young Economist poster session and then we'll have the group picture of everyone in the courtyard. So please don't run away, but at the same time please take the time to answer the feedback survey on your iPad. It's essential for us that we can improve the conference for you next year. I've also been asked about the dinner tonight. Buses will depart at 6.30 and the dinner is hosted by Banco de Portugal. And now without further ado, let me introduce Governor Fluke, who chairs the concluding panel with Governor Kani, Governor Corroda, Governor Polos and President Draghi. And after the panel we'll have a short Q&A as well. Over to you. Thank you before Governor Fluke starts. Just wanted to thank all of you. Also name of my ACB colleagues, Banco de Portugal, all the staff and all the presenters and the panelists for the insights they have provided and for the lively debates we had in the last two days. The discussions about the relationship between investment growth and productivity in advanced economies have indeed been illuminating and thought-provoking and will continue having a conversation on these issues just in a few minutes. Governor Fluke. Thank you very much. So first let me thank you President Draghi and the organizers from the ACB for inviting me to moderate this final panel. Before I invite you to actually present your opening remarks, let me share with you some insights about the experience of Israel, which has been named Startup Nation, it was already mentioned, due to the very active startup scene that we have in Israel. We have some 5,000 startups in very broad range of areas from cybersecurity to fintech, clean energy, agrotech and so on and so forth. The high density of startup activity has been attributed to a number of factors, some of which are captured by various measures of the degree of innovation in an economy and others may have to do with the way the country has evolved and the challenges it has faced. The World Economic Forum and Bloomberg published international rankings of innovation where Israel ranks high. These rankings measure and show the level of universities that provide the economy with a scientific basis and technological abilities, the close link between the academia and industry, the extent of government support at least earlier on to commercial R&D and the high developed venture capital industries as factors that made key contribution to the innovation based economy. Adding to that, I would add the need for sophisticated intelligence unit in the army where young talented people gain experience in various digital environments, including for example in the area of cybersecurity or the need in civilian areas such as water supply for example, for a country that is characterized mostly as desert which led to important innovations in areas such as drip irrigation, water desalination, water recycling and so on. So basically you can say that necessity is the mother of invention. To all of that I would add a cultural element, some call it chutzpah, and that is the willingness to explore new ideas, fail, learn from your failure and then try again. As the vast majority of startups actually fail and only a small fraction end up being great success stories such as Waze or Mobili. However, this is only the bright side of the Israeli economy. A more complete picture is that Israel can be characterized as a dual economy. One economy that is knowledge and innovation based, that is dynamic, internationally competitive and attracts highly educated individuals and the rest of the economy that is characterized mainly by lower productivity and low wages. Innovation is limited mainly to the high-tech sector which at its broadest definition accounts for only about 10% of the workforce and actually it's among the highest among developed countries but it's still only 10%. The rest of the economy is benefiting from innovation only to a small degree. There is limited diffusion of innovation into the other sectors and it is not adopting cutting edge technologies developed in the high-tech sector or elsewhere. This economy is unable to compete for talent that would facilitate actually the adoption of advanced technologies in these other sectors. So the outcome is that of a polarized dual labor market resulting in high income inequality. The top-notch skilled labor being absorbed and very well compensated by the high-tech sector as I said about 10% of the employment and the other 90% are mostly low-skilled employees, less productive and earn lower wages. How to facilitate and provide incentives for the adoption of innovative technology by the rest of the economy? How to encourage diffusion? What skills need to be provided by the educational system that would make it possible to adopt sophisticated cutting edge technology in these sectors? How to help the diffusion of innovation to the rest of the economy? These are some of the questions and challenges that need to be addressed in order to reach an inclusive and sustainable growth so that the benefits of innovation will be shared by the society at large and not only by privileged few. So this is the Israeli experience and I want to turn now to the questions that the panel will address. The first question is really what does the issues that have been discussed in this conference have to do with central banking and with monetary policy or to what extent? It's clear that the basic solutions to the challenges that were discussed here lie outside the policy domain of central bankers. These challenges have to be addressed by the composition of government spending, by the composition and structure of taxes, by structural policies and clearly trying to focus monetary policy on these issues will further overburden monetary policy. But still, the question is if monetary policy and more broadly central banks policies can have a role in supporting higher productivity and inclusive growth. So this is the issues that I'm sure will be covered by our distinguished panel and let me invite President Draghi for his opening remarks. Thank you. First of all, I agree with you that monetary policy should not be overburdened by other tasks besides the many it has now. But having said that, still we can't underestimate the importance of innovation of productivity and of growth and they are very important in informing our monetary policy. We've been seeing in the last several years, probably 15 years, a decline in productivity both on cyclical accounts and on trend accounts. I'm sort of relatively confident that as the economy will improve in the Eurozone and we are entering an upswing in the business cycle, we'll also see a cyclical improvement in productivity. We've seen a pickup in private investment after many years being subdued and that's the relatively recent evidence in the last few quarters. So we do expect an improvement on the cyclical front. Now the structural front is more complicated. There we have to deal with issues that pertain more to innovation and the diffusion of innovation. An innovation, clearly certain policies that foster R&D, the investments in human capital are the first ones that come to mind. But then we ask ourselves what is really, if we look at countries that are very successful with innovation, what sort of environment they have made them to be successful. We had that very interesting presentation by Professor Schmidt at lunch where he described the ecosystem of MIT. So one very first indication is that we, especially for applied research, to have an ecosystem like the one was described by Professor Schmidt or aiming at that is, for me it would be probably the most interesting policy prescription. Now, however, to get there one has to have in place policies that enhance private entrepreneurship, individual initiative and a cultural environment that is conducive to this. So to get all this, now another distinction is actually quite important, whether we talk about applied research or pure research. And while we had this, we had this in this beautiful presentation by Professor Schmidt, all for applied research really, what matters most is individual private initiative. For pure research, there is a very strong case to have government support. I would suggest that there is a marvelous small book called The Usfulness of Useless Research, which is being written by the director of the Princeton Institute for Advanced Study. He's a renowned mathematician, Professor Geigreff. I hope to say it correctly, it's a Dutch name, so it's always a big problem. But, and there he makes the case that good, pure research will produce the applied results that everybody wants to get. We don't know how many years it takes, and he shows it makes a very good case showing that many of the current, of the current inventions today rely on research has been carried out many years before. I mean, this Princeton Institute is the place where Einstein used to be for Neumann, and it's the place where arguably there are probably the highest concentration in the world of physicists and mathematicians in the last 20 years. So, that's for innovation. But in Europe, we, I hope I didn't insult anybody saying that. I didn't say at each point in time, okay? So, but when you go to diffusion, things are in a sense, as far as Europeans are concerned, easier. And there we can see and we have estimates where the largest measurable progress in increasing productivity come more from diffusion than innovation. That's in Europe. And so we have to have proper policies that foster this diffusion. And again, we, these are policies that deal with basically product markets, enhance competition that enhance private investment, that make the environment to use words being used many times growth friendly, growth friendly. Incidentally, was something that thought it came to mind listening to the paper by David Otto last time. He showed, you remember, he demonstrated that increase in productivity within each industry may have a depressing effect on employment. But the spillovers to other industries more than offset the negative effects in one specific industry. Now, I was thinking that to get, to actually get the benefit of these spillovers, one has to have labor mobility from one sector to another. If there is no labor mobility, either you don't have the productivity increases because they are sort of kept down, or you have, you're bound to have more unemployment and without the benefits of higher employment in other industries. So that suggests what, that we have to have policies that, as I said before, they increase the labor mobility and the competition in the product market. Now, more specifically to our, to our own environment in the Eurozone, there are many policies or many things we can, we can sort of do and think and imagine for the coming months that enhance confidence. That's the other thing, it's very difficult to have investments that are bound to take several years in an environment where there is no confidence. And fortunately, we are seeing that the environment is changing for the better there. I would expect that also on that ground we will be seeing improvement. Thank you very much. Thank you very much. Governor Karmin, please. Thank you. I brought some pictures, so I'll just speak from up here if I may. First off, let me thank you President Draghi, Governor Costa, colleagues from the Euro system for this invitation. It's an extraordinary gathering here at CENTRA and it's been my pleasure to be back. What I'm going to do is a little more pedestrian because a lot of the discussion has been about innovation, cutting edge type technologies. And I'm going to talk about investment. Investment of the cycle, more than innovation, TFP per se. And I'll use the UK as an example. There's basically two parts of the presentation. Use the UK as an example of some broader trends and some of the explanations for why investment has been weak and just provide a few perceptions on that. And try to answer Karnit's question at the start, which is what can central banks do. So first thing, I think we all know this, good to put it in context. G4 has represented investment has been weak since the crisis. The UK has done relatively well, but it's nothing to be proud of because actually this is the weakest investment recovery in half a century in the UK. And why could that have been the case? Well, one explanation is well it was really poor recovery. It was also the weakest recovery since the Great Depression. But actually the investment share is kind of hard to see on this slide. But the investment share has actually fallen over the course of that time by about two percentage points of GDP. It's also the case in other advanced economies. Another common explanation of course is that investment following a financial crisis is particularly weak, not least because of the scarring effects and the restrictions on credit supply. And we certainly had, and we can go into detail if anyone doubts it, but plenty of examples of that in the early days of the recovery or the recession in the UK. Misallocation, Chad Jones and others talked about miscalocation this morning quite rightly. This is just trend capital to output ratios, different scale, which gives you a sense in terms of orders of magnitude of investment in the UK versus other major advanced economies. Basic point, we see the capital overhang being worked off in the UK over the course of the next couple of years. So some of the conditions are coming into place for a pickup in investment. Zombie firms, often an idea floated for why there's been poor investment. Actually, both as the regulator of the banks, but also looking at macro, we've seen the proportion of zombie firms fall if you just do them on a basic who's not covering their interest with their EBIT in the UK. The BIS sees them rising across advanced economies. The BIS sees lots of things that we don't see. So we're more confident in our data there. Now, I'm very sorry to have missed Ben Bernanke's presentation the other night. He's taught us many things and one of the things that he's taught us, of course, is the option value of waiting in investment. And one of the things that is an unfortunate reality in the United Kingdom is that firms have faced high degree of uncertainty, not just economic, geopolitical, and policy uncertainty. We all know these measurements are imperfect, but directionally tell us something. And we think that it probably helps tell us why hurdle rates have been so stubbornly high. I'll show you what I can go ahead. That's a survey of hurdle rates in the UK firms at present that we have done at the bank. So think mid-teams, hurdle rates. And one of the questions is how do you reconcile that with a slower, secular growth outlook if you believe anything about what various tenure rates may be telling us? A lower real return environment. This is just using a simple Black-Scholes model, and I've plagiarized it. I've cited it actually, so it's not plagiarism from Ben Broadbent. And just looking at if you have a slightly higher uncertainty around future returns, it's easy to swamp the reduction in hurdle rates that would come from lower overall growth outlook. And we do think that is what explains some of what I just said, which is stubbornly high hurdle rates. Just talked about diffusion, important points made by Mario a moment ago. We do see this issue quite strongly in the UK data. There's many ways to cut this. This is value added per worker in various firms. The magenta are frontier firms, and the blue and the green are the rest, and you can see that the gap is widening quite significantly. I think labor mobility, which is quite high in the UK, can be part of the explanation here. There's a variety of other reasons. And one of the puzzles in the UK right now is that despite very strong quantities in the labor market, labor mobility is quite low. So people are staying in their jobs for much longer than you would expect with a 4.6% unemployment rate. Possibly that goes back to the uncertainty question, although that's speculation on my part. But elements of diffusion go with workers. There's a lot of evidence around that. And if you join my mother and my two thesis advisors, you'll find a very tortured explanation of that for my graduate days. Now, secular forces that are there. Demographics, importantly mentioned this morning. I'm not going to go into detail, but that is part of the demand element that we think could be there as well. And also the shift into more intangible investment that's happened over time. I'm sure you talked about it when I wasn't here. I didn't hear much about it this morning. But just with intangibles, I think a number of you know that the investment in accelerator effect is dampened, and there is quite a bit of evidence around that. It's much harder to borrow against intangible assets, and that's part of the potential solution set here, which is, of course, outside the remit of the central banks. Okay. So in terms of, I'm getting ahead of myself. Yeah, there we go. All right, so I want to turn now to Karnit's question, which is what can we do as central banks to support a belated recovery in investment, recognizing that other policies will be much, much more important. First, I think we need to recognize that, well, the cost and availability of finance matters, internal cash flows, profit outlook, and uncertainty are far bigger determinants of investments than 25 basis points here or there. Monetary policy, its impact comes from affecting companies' profits, and cash flows through its effects on domestic demand and by the exchange rate on external demand. And given the importance of internal finance for investment and those high hurdle rates I talked about a moment ago, such indirect effects are much more important for investment, I'd argued, than the effects on the cost of capital. Now, the current situation in the UK can help illustrate that point. Right now, in the UK, output is approaching potential, and that capital overhang I mentioned a moment ago looks set to be eliminated over the course of the next few years. So if companies want to expand, they're going to need to invest, particularly in a tightening labor market. The strengthening global economy should tempt them to do so, particularly since UK companies are generally competitive given the fall in sterling. We have a broad-based recovery, more than our estimate is more than three quarters of the global economy is growing above potential, and we're seeing the circumstances in place for a rotation, some rotation from consumption to investment, or at least the indicators of that, whether you look at capital goods orders, you look at global trade, or investment intentions, all of this suggests that there is that rotation. And if those stronger investment intentions are realized globally, there is the prospect of the global equilibrium interest rate rising somewhat, making it a given monetary policy setting more accommodative. The extent to which it does, of course, depends on other secular factors that have been holding it down, including demographics again, dead overhangs, and the capital intensity of production. Now in the UK, as I mentioned, there are a series of uncertainties that are pushing the other way, uncertainties about how consumers will adjust to a period of weaker real income growth, uncertainty about market access, access post-Brexit, and uncertainty about the potential risks in the transition to new arrangements with the EU and the rest of the world. And so the best thing we can do is to pursue the right policies as the bank within clear frameworks to promote monetary and financial stability. With respect to the latter, financial stability, basically what we've been trying to do is remove any lingering uncertainties about access to credit, access to finance. Good times are bad, and this is just one illustration. This is common equity for Core Tier 1 UK banks. We've also taken steps, such as we took a number yesterday around raising the counter-cyclical buffer, tightening underwriting standards, and recalibrating the leverage ratio. So a series of things, including contingency planning around Brexit, which we do for a variety of reasons, but really are, from an investment perspective, is to take one uncertainty off the table. And I could show you lots of illustrations of the relative success of this. I'm not sure where the US-Iyield bond market is at present. It's the best illustration of that, but maybe you can take my word for it that capital is available and it's keenly priced. So let me turn to monetary policy and then finish. So we, as you all know, just like the ECB, operate within a well-established framework. For us, it's anchored in the inflation target. And we set out in advance of the referendum how we would apply that framework to monetary policy. And we emphasize that the process of leaving the EU would have impacts on demand supply and the exchange rate, and that the impact on monetary policy, therefore, would not be automatic because you need to balance those factors. In addition, what we have done is clearly, as clearly as possible, set out a reaction function consistent with our remit. And this is the important point is that under our remit, under our mandate, when we are in, quote, exceptional circumstances, and Brexit is an exceptional circumstance because it has brought an exchange rate move that is itself bringing forward an expectation of a real fundamental shock, a fundamental hit to real income, we are required by that remit to balance a period of above-target inflation with a period of weaker growth. So the primary objective of monetary policy remains inflation control, and since that is the case, any overshoot of inflation above target can only be temporary in nature and limited in scope, and therefore we've been clear that we have limited tolerance for that above-target inflation. So we've been in a situation where since Brexit emerged, financial markets have marked down the UK's economic prospects were in a situation where we can't prevent that weaker real income growth that's likely to accompany Brexit, but we can determine how that hit is distributed between job losses and price rises, and we can support households and businesses as they adjust. And what this chart shows is just in effect the trade-off that we've been managing between the output gap and inflation, at least in projections, two years out in the light purple is back in a year ago, effectively, if there had been no stimulus. We felt it would be a very large output gap with a modest inflation overshoot farther out, and then the path we've charted, and it's going, as you see, back towards the origin. And key point is that some removal, sorry, key point before that, is when, as spare capacity erodes, so as you chart that, chart those dot plots, the trade-off that we must balance lessens and all else equal are tolerance for above-target inflation falls. Now different members of the MPC will have different views about the outlook for the economy and exactly where that trade-off will be, but not so much different views of what the Lambda should be. In other words, the interpretation of the remit, and so they will have different views about the timing of any bank rate increase, but all expect that any changes would be limited in scope and gradual in pace. And so when we met earlier this month, my view is that, given the mixed signals on consumer spending was too early to judge, with confidence how large and how persistent any slowdown in growth would prove, and moreover with domestic inflationary pressures, particularly wages and labor costs still subdued, it was appropriate to leave the policy stance unchanged at that time. Now some removal of monetary policy stimulus is likely to become necessary if that trade-off that the MPC faces continues to lessen, so you can see the chart and you can plot your view on where it's headed, because as it lessens the policy decision becomes more conventional, or would become more conventional. The extent to which the trade-off moves in that direction will depend on the extent to which weaker consumption growth is offset by other components of demand, including business investment and importantly business investment, whether wages and labor costs begin to firm and more generally how the economy reacts to both tighter financial conditions and the reality of Brexit negotiations. And those are the issues, some of the issues that the MPC is debating, but what we're trying to do, and we consistently tried to do, is frame them within a clear framework, a reaction function so different people have different views and they can anticipate accordingly how it would change, but it doesn't become a factor into business investment because there are far more important considerations that should determine them. And so with that, I will conclude and leave you with a stylized picture of the Bank of England to enjoy. Thank you very much. Thank you, Mark, and let me invite Governor Koroda, please. Yes. Thank you, Chair. My development among advanced economies during the recovery from the global financial crisis is the weak momentum in business-fixed investment and in corporate investment in human resources or wage growth in spite of strong corporate earnings. This situation is sometimes called corporate saving grant. This is particularly stuck in Japan. After the introduction of the QQE or quantitative and qualitative monetary easing four years ago, economic conditions have improved and firms have experienced record high profits. However, firms are still cautious about increasing investment, spending including investment. In terms of the saving and investment balance, the corporate sector has been in a surplus since the latter half of the 1990s, and now corporate cash and deposits have kept increasing, amounting now to 250 trillion yen or almost 50% of nominal GDP. Looking more closely into the development in the corporate sector, business-fixed investment is on a moderate increasing trend, but it remains sluggish. Investment has been mainly on maintenance and repairs and less on capacity expansion. In terms of the corporate investment in human resources, wage growth has been low, standing at less than 1% on an annual basis. This is despite the tightening of the labor market with unemployment rate falling below 3%. The wage growth for full-time workers has been slower than that for part-time workers. While the wage growth of the part-time workers is in the range of 2.5 to 3%, as it is more sensitive to labor market conditions, the base pay for full-time workers is growing at less than 0.5%. These developments seem to suggest that firms are still hesitating to make long-term commitments. As often pointed out, business-fixed investment has a feature of irreversibility and wages have a characteristic of downward rigidity. For firms to decide on making new investment or raising base wages, they need to become sufficiently confident about growth prospects. It seems that firms have not yet regained that confidence. The question then is how to achieve stronger growth expectations. The key would be an effective combination of structural policies and macroeconomic policies that lead to a sustained increase in economic growth. Structural policies include regulatory and institutional reforms and these will raise growth potential. The main components of structural policies in Japan currently are so-called work-style reform and investment in human resources. These are expected to improve labor productivity and increase labor participation, especially among women and elderly. This in turn will make growth not only stronger but also more inclusive. As many studies suggest for economic growth to be sustained, inclusiveness is really essential. Turning to the role of macroeconomic policy, let me mention three points. First, macroeconomic policy can support the implementation and the progress of structural policies. Structural policies could be accompanied by unfavorable economic impact in the short-term, such as an increase in employment. Macroeconomic policies can alleviate such unfavorable impact by increasing demand. This would support the implementation of reforms that lead to higher productivity in the long run. Second, in order to improve business sector confidence in future growth prospects, it is important to achieve a sustained period of strong economic activity. This is particularly important in the case of an economy that has struggled for longer periods of stagnant growth like Japan. Once strong economic activity continues for an extended period and labor and production capacity shortages linger, firms are likely to more actively seek to increase business fixed investment and investment in human resources. Third, in Japan, since the latter half of the 1990s, weak growth expectations have accompanied with low inflation expectations. The Bank of Japan introduced QQE in 2013, four years ago, and the policy was intended to change the deflationary mindset entrenched among public. At the same time, the government has pursued fiscal policy and structural reform with a so-called three-pronged approach. As a result, economic and price situations have improved in Japan. That said, there is no magic wand to quickly fix the weakness in growth expectations and inflation expectations. As I mentioned earlier, an active combination of macroeconomic policies and structural policies is essential. This would lead to an environment where firms upgrade long-term growth expectations and more actively make commitments toward the future. Let me stop here. Thank you very much, Governor Koroda. And we move to the Governor of Bank of Canada. Well, thank you very much. Thanks, Karnit. Well, I'm delighted to be here. It's obviously a wonderful setting. I don't think I'll get a debate on that. It's been a great conference, and I want to thank the ECB very much for the invitation. So I'm not going to try to summarize everything or even many things that I've learned in the last couple of days. I don't have enough time. But what I would like to do is talk about what central bankers actually do when they learn something at a conference like this. You know, something that calls into question one of their basic assumptions or maybe one of their economic models or otherwise changes their thinking in some way. Now, central banks can't go around embracing every new shift in academic thinking. The role we play demands a certain degree of continuity of framework. In fact, we're much more likely to go home and stress test our framework with a new idea than maybe shade our judgment as a result. Our framework, I think of it as like a nice old house, right? We might add a room or redo the kitchen, but we rarely tear down the house and start over. So I like to think of this as sort of a gathering of new ideas as a Bayesian process. So there I put that in for the aficionados who are here, of which there are many. It's where our priors, the priors that we bring to the table are summed up by the results given by our standard framework. And then we confront those priors with a compelling new insight from new research. Now, there are two dimensions to Bayesian updating. And the first dimension concerns point estimates. The new insight or the new model convinces us to tilt our judgment from the old result toward a new result. Probably not all the way, but toward it. And to illustrate, say we start with a forecast of the economy based on our standard framework. Then we take a new experimental model and we use it to construct an alternative scenario. Well, then as policy makers, we have a debate on how much weight to put on that new alternative. And in the end, essentially, we would choose some sort of judgmental blend of the two scenarios, possibly do us the small weight on the new one, but that's the kind of exercise I have in mind. That's the point estimate. The second dimension of Bayesian updating is less commonly practiced, but I think no less important. That dimension takes into account that there's now a new range of possible outcomes that's now introduced into the exercise by that new research. The fact is, by tilting your forecast, you know, by putting some weight on an alternative model, also means that you're embracing a wider distribution of possibilities than your original model suggested. And the question is, how do we as policy makers acknowledge that wider range of possible outcomes in our process of policy formulation? Now, there is a literature, of course, on decision making in the face of uncertainty, but it's not that well-developed, at least for the purpose that I have in mind here. So what I'm offering today is just some early thinking on this issue. So in practical terms, acknowledging the risk that in the light of new research, there's now a wider range of possible outcomes means moving monetary policy further away from a mechanical rule or a reaction function. It turns into a bit more of a risk management exercise. So it means asking yourself questions like, okay, if I base my policy decision on my standard model, but this new model turns out to be right, what will be the consequences for the economy? Now, if those consequences would be undesirable through your policy lens, then you need to adjust your policy plan to hedge against that risk. In effect, you lean in that direction to some degree. So to illustrate, let's consider our discussions at this conference around innovation, investment in growth, and where monetary policy fits. So for one, one session, I came away with an understanding that technological change will eliminate some jobs in the sector where the change is happening. It'll demand difficult adjustments from those affected in that sector, but it will gradually create jobs in other sectors. So in effect, according to the evidence that David Ohtor presented, generally has been positive eventually in the end. Now as policy makers, certainly we wouldn't want to disrupt those natural forces. Rather, we'd rather facilitate them. So the process of technological change is inherently unforecastable, but the risk that it will happen is great enough that we should at least consider allowing for it in our policy formulation. So this would mean doing enough experimenting with the models to understand the risks that you would be taking in ignoring the possibility and then hedging against those risks. So as a second example, we had some discussions at this conference around the global weakness in investment. Mark has just given some good insights there from the UK. So based on what I heard here, we should be considering whether we have the right model of investment in mind. So globalization and the relative growth in the service sector have made intangible investment far more important than in the past. And some of this is measured well, but there's a risk that much of it is not measured very well. So for instance, the switch to cloud computing is clearly complicating the measurement of investment for statistical agencies. So our models of investment industrial capacity are quite traditional, right? And they're very central to our understanding of the inflation process. So if another model of investment dynamics is growing in importance, we may need to allow for a wider range of possibilities around that part of our forecast and adopt a policy that reflects that uncertainty. So those are just two possible sources of new thinking that I take away from this conference. Actually, Mario raised a couple of others in his introductory speech around properties of the Phillips curve. The same kind of exercise one could conduct for those kinds of issues. So this Bayesian thought process may be applied to any assumption in our policy framework that looks vulnerable to us. And that would be based on new research. The assumption about potential output, about R star, so on. So managing policy risks in this way clearly moves us further away from mechanical reaction functions or policy rules. Importantly though, it's not an argument for pure discretion in policy decisions. Now, complementary to such a shift is increased transparency around those judgments and of course, clear and regular communication, all of which is still anchored by our inflation targets. So what I'm offering today is just some early thinking on how best to incorporate alternative views of the world and monitor policy formulation along the way. Our goal is to formalize that process so as to ensure policy accountability through time. Now that's an active research area at the Bank of Canada right now. So thank you very much for that. So taking from your last comment, what I understand is that having to continuously update our thinking about policy, a Bayesian process and getting away from a mechanical rule based policy reduces the probability that we will soon be replaced by robots. So in that terms, we're safer than... We can move now to some questions or comments from the topic. So we have about 35 minutes, so we'll have probably two or three rounds. Why don't you start? Thank you very much. I enjoyed all introductions in particular the last one. Yesterday we had a beer in the bar discussing all the issues which were discussed that day and we didn't agree about every thing to put it mildly but we agreed with the colleagues about one thing and that we are confronted with 19 uncertainty. So my question to the panel, not only to Mr. Polos but to everybody, how are you going to deal with that 19 uncertainty which we are confronted nowadays? Thank you. Thank you. We'll take two more questions before I let the panel answer. Yes, please. Richard Baldwin from the Center of Economic Policy Research. I think one of the themes especially on the first day about innovation is that there may be an increasing gap between utility and GDP and unemployment. So in some sense the central bank is skating between helping people's lives improve and keeping inflation from going up. But so much of the innovation we were talking about is essentially saying that not only inflation may be measured because of zero price goods and whatever but that many of these zero price goods are providing utility and benefit to people's lives even though it's not showing up in GDP or potentially employment. I wondered if you think about any of those things in doing your monetary policy or more prosaically do you think seriously about the GDP mismeasuring economic activity and do you take that account in your monetary policy discussions? Thanks. Maybe one more on this side. Yes, please. Over the past two days we've talked a lot about investment but not a great deal about how it's funded. Besides the fact that there's an important role for the public sector in primary research. But what about private savings? And what about the main source of private savings in Europe in particular life insurance companies or pension funds? Another panel concerned that recent regulation of these things actually encourages them to invest more in government bonds say than in equity, public equity, private equity. We've learned over the past couple days if you want to invest in intangibles, in green revolution, climate change, we do need more equity type instruments and more long term investing. Thanks. Stephen would you like to go first on the night uncertainty? Well, why sure. So, well, anyway, I love the question about nighty and uncertainty. I resisted the temptation to actually put that in my remarks. So thank you. Because it's true we do face some risks that appear to be sufficiently profound to fall into that class. Usually when we think about risks, and that's what I meant when I said that the literature of decision making in the face of uncertainty is kind of in the zone. It doesn't really take into account that some more fundamental wrong model versus paradigm shift kind of uncertainty, which gives us its limitations. But I do think that the, I do think about that. We face some things today. They're mainly geopolitical in nature that would be so rattling that you'd need to start, we would have to knock down the house and do something quite different. But I think that we do have at least the equipment to think about those things. Whether we have the nerve to actually incorporate them is your question. So it's a little bit like what we do in financial stability analysis, right? We have these tail risks that we think they're low probability, but they have major consequences. We focus on vulnerabilities in the system, like a crack in a tree. Someday the right wind comes along and down and goes, but it's a one in a thousand event. And so you still have to understand those implications, whether it actually goes to the stage where you are changing policy in some way is a second question. But I think we have the equipment in the way I described to actually understand them. So for me it would be like, well, what if, you know, NAFTA were changed in a really big way? How would that affect the Canadian economy? That would be a really big shock for us to deal with as opposed to modernizing it or tweaking it, right? So that would be the kind of break that we would need to analyze in that separate way. But I would use the same methods. I'll leave the other questions to my colleagues because I know they're already now. Now that I've ragged the puck. Okay, Mario? Well, just a word on a second question. Measurement errors have always been with us. But clearly with the better and better technology, we're discovering more and more measurement errors and better and better ways to measure things. Now, there is a transition time. These better measurement systems haven't translated yet, by and large, into hard core measures that you can use to change, to use his words before, to change your framework, to redo the house. So you basically take into account that maybe the assumptions, or some of the figures you're using are not 100% exact, but that's not enough to change the framework of a monetary policy. For example, in my speech yesterday, just to make an example, we have quoted one of the problems with Phillips Curve is the measurement of the real unemployment. So there is this use six new measures where you take into account more. Labor participation has increased in ways that we understand up to a point, especially in a continent where there have been large migrations, large inflows. How much of the new jobs have gone to migrants, how much they've not, and so on and so forth. Is that enough to make your framework change in a substantial way? I wouldn't say so. Rather the question we asked is, suppose that these measures were correct, would they hint at a, would they suggest that the reasons why inflation is not responding as fast as we wish to the improved economic conditions are permanent or not? We conclude that they're not permanent because these measures, even though not appropriate, these measures of unemployment may actually go down as the economic conditions will improve. So by and large the direction of our monetary policy is not going to change. I think that is what I feel like saying on that. Well, if I may say some words about... Now, that question has caused into mind really the special financing conditions that we have in Europe. It's a bank-based continent, no matter what. It has been a huge improvement in the school of shadow banking. It has been a huge increase in shadow banking, but we are still far from being a capital market-based economy. And so it's true that many private initiatives don't have the same easiness of access to markets that they have in other parts of the world, namely in the United States first and foremost. One measure that could certainly foster progress in this direction is the, say, the progress on the capital market union. That's key. That's very much on the agenda of all governments for the next few years. Just a couple of comments on each. One on the... I was in here yesterday, so for me that's a known unknown, that discussion. And I associate myself with much of what Steve said around thinking about some of these alternative scenarios, and that has to play in. I think part, if I may, Steve, to suggest, part of what, or at least I interpret what you're saying as well, is you can't be incapacitated by these potential tectonic shifts. We all always think they're living in interesting times, and to some extent they are. And so we do have some big potential changes to the labor market, which may be akin, global labor market may be akin to, and I look down to Charlie Bean over there, who recall the work you did about the integration of China into global prices, and you and the Bank of England I think were relatively early on understanding the deflationary force of that in the early 2000s and calibrating policy appropriately like that. The BIS, to be positive about the BIS brought this out, I think this past weekend, about potential shifts in the global Phillips curve. I know we have to work in the microfoundations for that, but... Professor, but we'll go there. So I think the point not to be incapacitated, think through and not overreact, and Steve's provided a bit of a framework for that. Second thing, on this mis-measurement point, I agree with what Mario just said. But a couple of things. One is, if this is so great, why is everyone so unhappy? And I recall, not everyone's so unhappy, but I recall one of our colleagues pointing out hedonic pricing and mis-measurement in the CPI because the iPad 2 had come out to an audience in Queens, and there was a furor in the audience, and the headline in New York Times the next day was, let them eat iPad. There is an element here, we're going to tell people they recognize how well they have it because we haven't been measuring GDP properly. I think where it goes to the framework, you can't change your framework on a mis-measurement issue, as Mario said. There's always been mis-measurement. If it becomes much better, if it is measured much better, and we're in a much sharper disinflation environment, then one potentially has, if we're quite confident about that, the good deflation, bad deflation discussion or whether you have a mix of it, but that is well down the road and that is not a single anyway. The last point I'll make on Avinash's important question. I would make a positive point about the new regulatory framework, particularly with respect to life companies because the matching adjustment, which is hugely important to the incentives for life companies, provides not equity investment necessarily, but infrastructure and long-tailed investment and those who have the capability of making those types of investments have actually been, and I use this term positively even though we're the supervisor, have been quite aggressive in building up those portfolios. So that is a positive regulatory development. Thank you. On the first question, how to address this uncertainty? I mean, if it is technological development uncertainty, I think this is inherent and nothing can be done, probably. But from the central bank point of view, we tend to classify various uncertainties into external ones and internal ones. And external uncertainties, yes, there are quite a lot, including geopolitical problems here and there, but of course these are external, nothing can be done by ourselves. External uncertainties, I think central banks can at least try to reduce. For instance, policy uncertainty could be reduced not just by the central bank, but also the government as well, but I think there's room for reducing uncertainty in this case. And also, by the way, the central bank, like the Bank of Japan, is always involved in securing financial stability. And financial stability is kind of buffer to avoid some uncertainties actually creating huge problems. And so I think uncertainty is difficult to address, but certainly internal uncertainties can be reduced by our efforts. Now GDP and measurement issue, by the way, in Japan, even GDP measurement is a big issue and recently the government decided to substantially improve GDP statistics. Because we found that some irregular movement of GDPs in the last few years and probably R&D investments, software investments are not well represented in the current GDP statistics. So GDP statistics could be further improved, but so-called gross national happiness or something like that developed by Bhutan. Quite interesting, but whether it's quite useful for us, I'm still a bit skeptical. Thank you. Thank you for another round. So let's start with you, then you, then you, and we'll take four questions. Richard Port is London Business School. I suppose this is a question for the Europeans on the panel because it's about the European international system. The issue is this. If Capital Markets Union doesn't really move along at the pace that we would like to see, what alternatives are there? The banking system, bank assets in Europe have fallen by, I don't know, 6% or so in the past five years. The shadow banking system has expanded by about 30%. Now, if we're thinking about financing innovation and growth, is it really sensible, are there possibilities for the shadow banking system to play a significant role in financing innovation and growth? Or is it just too risky? Do you not want them to go there? I think that's the question. Thanks. You're next. Yeah. Andre Sapir from Brussels. Now, the topic of this conference has been investment and growth in advanced economies. And I think it's fair to say that much of the discussion has been investment and growth in all countries in the aftermath of the crisis that started basically 10 years ago. But 10 years ago there was another event that occurred that we have not that much discussed in this conference. It's the fact that advanced economies that had represented the largest share of global GDP have assumed since roughly 10 years ago less than 50% of global GDP measured in purchasing power parities. And that share obviously will continue to go down as the years go on. So advanced economies are still very important and all of the countries represented here are very, very important, but they are less and less important. China and other emerging economies and some developing countries are assuming a bigger role. So that raises in my mind two questions. One is that there are more actors. And yes, there is not just a G7, there is no a G20 to deal with some of those matters. And how much of a coordination can there be in the context of the G20 as they had been in the past, in the context of the G7? If there are some global matters including for monetary policy that does require very coordinated action. How much of coordination is there going to be as far as monetary policy is concerned among these more varied actors? Thanks. Thank you. That was you and then you. Okay. Jim Bullock. Jim Bullock, St. Louis Fed. My question is about model uncertainty. So the panel has already addressed sort of issues about measurement uncertainty, but it seems like the most profound issue we face is model uncertainty. Really entire frameworks that are very different would give very different policy advice. And it's even our leading theories probably only have, you know, 30% of the chance of being the true model as compared to all the other models. The other models are you have even less weight on them. So you have a situation where you have leading theories, but they're not really theories that we want to really put really high weight on. So it seems to me one way to deal with this is to have your baseline model, but then have other corners of the central bank that are working on other models that may be useful in the future, but maybe aren't ready for prime time today. And my question to all of you is how much of this is going on at your central bank? Thanks. Please. Right here. Yeah. In the first row. And then you. And then. Robert Hall Stanford. So there's one person missing from this panel. That's Janet Yellen. And to take advantage of that fact, I'm going to ask a question. No, please. I would otherwise ask. Thanks. That question is the Fed has consistently over forecast inflation for at least the past six years. That's well known from their published forecasts. Is the Fed's policy rate today too high as a result of failing to update its Phillips curve from the sample evidence? And I require an answer from each of the four of you. Okay. Can you pass the mic? I'd like to. John Morgan. I'd like to follow up the excellent question that was just asked about modern uncertainty. Question for Stephen Pollos. Bank of Canada is one of the banks that's leading the way in developing models. And you've got two models. You've got Totem, which is a DSG model with a particular focus on terms of trade. And Lens, which is an Fribus type macro econometric model in which balance sheets and asset prices can play some role. I'm just wondering, and I think Totem probably helped you in dealing with trade shock that you've had. Whether Lens had enough attention on terms of trade, I'm not quite sure. But the question is whether either of these models adequately addresses the housing market mortgage debt issue that hangs over Canada, which the IMF has flagged up in the past. You know, Canada was one of those economies supposedly ripe for correction in the housing market, which has never happened. So do either of those models adequately represent the risk factors there? Okay. So now we go this way. And Governor Koroda, please. I would make my comments on the increasing role of emerging economies in the world economy. And then I would make a few comments on model uncertainty and what kind of econometric models the Bank of Japan has been developing. Now, G20 was created in 1999 in the aftermath of the Asian currency crisis. Because G7 countries felt that the G7 framework is not wide enough to include many important emerging economies whose impact on the world economy were increasing. So the G20 was created. And then after the Lehman crisis in 2008, G20 summit was created. I mean, from 1999 through 2008, G20 meeting was always Finance Minister's and Central Bank Governor's meeting. We were discussing various issues, but more or less technical ones. But after the Lehman crisis, particularly the 3G20 summit, first in, I think, in Washington, DC, and then in London and Pittsburgh, these 3 initial G20 summits really made decisive contribution to combat with the upcoming financial crisis and the upcoming global recession. So I think already in this kind of sense, the international community has responded to the changing or increasing role of emerging economies including China and India and so forth in the world economy. Second, model uncertainty. I think this is theoretical as well as practical issue. Like the Federal Reserve, Bank of Japan has developed two sort of econometric models. One is typical DSGE, Dynamics Stochastic Generic Regulatory Model, which is theoretically beautiful but not always used for our policy simulation. Rather than DSGE model, we frequently resort to a typical Neocansian model. Maybe theoretical not so beautiful but quite practical and quite useful. But I'm quite sure that this model uncertainty will continue to linger and not easy to resolve. I mean, this is in some sense at the center of economic science and I don't think we can easily reach a consensus or the kind of best model applicable to all countries and used by all central banks. Thank you, Governor Corot. I ask the families now to be brief as we finish on time. Okay, very quickly I'll just associate myself with Governor Corot's comments on the G20, say that there still is an ability if necessary to have the right conversations between the right central banks. I wouldn't overemphasize the coordination versus understanding, but that's an entire other conference topic. Secondly, on the Jim Bollard's question, the only thing I'll supplement because I think others will speak to it as well, is one of the areas in financial stability land. You do need different models for exactly, I think, somebody picked up the point on housing debt and how it's incorporated. And it's one of the ways that we have thought about and modeled out a bit as a financial stability committee. Some action on mortgage, on restricting high loan to income mortgages because of the amplification effects through the macro cycles on the monetary horizon. I'll pass on your question to Janet on the Phillips Curve. And I think that was a test to see if we'd answer something. And maybe John Mueller had raised it, so that was my response to him. Thank you, President Corot. Thank you. That's the first question asked by Professor Portas. Right now the system is still bank-based. In answering your question, we may want to make a distinction between projects that need to access capital markets for their financing and projects that don't need to do so. Most private equity, for example, in Europe today doesn't need to access capital markets. To that extent, it's either financed by banks themselves, incidentally. The role of banks, even though the assets in the shadow banking system have increased spectacularly over the last three, four years, banks will continue playing a decisive role in financing. But basically many private equity, for example, are financed by banks, are actually created by banks themselves. On the other hand, and the more innovation, the more new projects will need financing, the more it will be a pressure to have a broader capital markets than we have today. Today they are quite fragmented. They are fragmented by regulation. They are fragmented by supervision. Each country has their own market supervisor. And they are fragmented by legislation. Each country has its bankruptcy law, for example. So it's really very, very hard to satisfy the capital needs on any reasonable scale, because you have to have a market which is bigger than the single country. Even worse, very often the selection of where-to-issue shares to finance is made not so much on the financing needs basis, but just rather on regulatory arbitrage or legal arbitrage. So in other words, you're looking for to establish a company which has very dubious governance standards. So you look for around and you sample for that. And so that's very inefficient. I think whatever we want to say, we should work hard to create this capital market union. It wouldn't make much sense to continue in the way we have today. In the meantime, there will mostly be known capital market-based financing for most of the small, we're talking about small, medium-sized projects in highly innovative technologies. I think that's the, finally, the expansion of the shadow banking system, and here the term of the FSB will certainly have something to say, should be, I'm pretty sure they are monitoring quite closely, but also should be accompanied by an extension of the regulatory perimeter beyond the banking system, of course. Because this migration from banks to non-banks will continue in the future. On the G20 and coordination, last year, I remember, I think now you made me remember, the speech I gave was about enhancing understanding of monetary policies in different jurisdictions, enhancing reciprocal understanding, I wouldn't call it coordination, certainly not, or even cooperation because each of us is bound by a national mandate. And so it would be very difficult to imagine a cooperation. It's very difficult. No question about that. We've discussed this quite extensively. It's very, very difficult. But there is still, there is need for that because if one reflects today, one of the major sources of uncertainty in the markets is exactly the different positions that different countries have in the recovery, and therefore the different positions of the monetary policy stance in the recovery. So that is one source of uncertainty for the markets. It's quite clear. So the more we can sort of talk to each other and also communicate to markets in a way that doesn't increase uncertainty, that would be very helpful. So the need is there. How to do it is not simple. Now, one thing about Bob Holt's question. Let me say that the Fed isn't unique in over forecasting inflation. Good company. I think they're in good company. Okay. Well, a lot of the questions have been covered very well. I would say on Bob's question, it's part of the code of conduct of center bankers that we never comment on each other's policies. And so thanks for reminding us of that. But I'll deal mainly with the question about modeling. Yeah, it's true the bank of Canada has invested heavily in models over the years. And certainly Totem and its predecessor were pathbreaking DSGE models that Totem just extended to include the terms of trade, which is a really big investment, but important for economy like Canada. And all that work paid for itself in 2014-15 when the price of oil fell. And we understood instantly just how that was going to work. And very proud of the folks that built that and the staff that run it for having said, this is what's going to happen. No doubt about it. You need to cut rates. It's all this debate raging. And so now, you know, two years later, we're happy that it happened that way. We're able to cushion the blow and speed up the economy's adjustment. So that's been a positive thing about models. We do have though lots of other models. You mentioned one other, which we use a smaller, simpler, more basic, which is used to complement Totem because Totem frankly is very sophisticated, very hard to play with. Actually, you need a double PhD to, it seems, to play with it. But of course, it does have the, both of those models have the equipment to allow us analyze housing price declines, for example, if you're interested in a shock like that. And we have other models that we use for financial stability risks. So in there, we've got all the financial institutions modeled and their interconnectivity captured. Here we do the actual shock that we talk about in the financial system review to kind of capture what kind of GDP impact might be happening from a basic shock, how much magnification there would be. And finally, we have launched the work on the next generation of model, which is to replace Totem someday. And I do this because I know that Totem's predecessor was started well, actually in about late 80s, around 1990. And so it took well over 20 years for us to go from there to what we use today as Totem. And so I know that I'm never going to see this completed, but it is my job to give people a license to invest now for something that we may only begin to use in production say 10 years from now, hopefully sooner. But there's no doubt that there's lots of new things going on that we need to build in and elaborate upon. So thanks for your question. Thank you very much. We're coming to a close. I think we're right on time, one minute late. So thank you very much to the panelists. I think it has been a great panel. Thank you. Good. Thank you. Thank you. Thank you. Not so fast. We're now coming to the final part. And we have 10 students who have worked a hell of a lot and who will now come on to the stage to be awarded one of them with the final prize. So I'm going to call you all on to the stage. Starting with Andrea Dabila, Martin Derrida, Miguel Fariai Castro, Alexandra Fotiu, Elena Gerco, Gregory Howard, Mengqiang Huang, Jean-Marie Meyer, Pierre-Lucca Panela, and Tomaso Sonno. And now I have the honor to ask the president on to the stage and he will explain and enlighten us as to who won the award. So now I have now the pleasure to announce the winner of the prize for the best paper in a Young Economist poster session. I hope you had the time to take a look at the posters and I wish to thank those of you who helped the selection committee to rate them by casting an electronic vote. Before I announce the name of the winner, let me say a few things about. All papers included insightful research analysis and reached interesting conclusions. Selecting the best among these 10 posters was certainly a difficult task. So let me thank the members of the selection committee, Philip Hartman, the chairman, Ines Cabral and Peter McAdam, and two external leading academics, John Mulbauer, professor of economics at Knufffield College and Oxford and Ricardo Rice, professor of economics at Columbia University and the London School of Economics. Posters have been rated along two dimensions, academic quality and policy relevance. Given that they have been chosen from over 100 submissions, all of the 10 short-listed papers were excellent. However, there must be a winner. And I'm particularly happy to say that the recipient of the 10,000 euros prize for the best poster this year ECB Forum in Sintra is Miss Venkyam Waung from the Timbergen University in Amsterdam with a paper, Central Party Capitalization and Misaligned Incentives. Before concluding this fourth ECB Forum, let me take the opportunity to thank all of you for your contributions towards making it a success. Over the past few days, we've had many valuable insights from Ben's thought-provoking speech on Monday night that the analysis of papers presented yesterday and today. To the lively and inspiring debates that followed during the different sessions, all of this provides food for thought for our work over the coming weeks and months and can inspire our future research and policy work. Finally, let me thank the organizing team of the ECB, including the staff from the Bank of Portugal for this very hard and very good job they've done. And I look forward to see you again next year.