 Three fantastic panelists joining us today who've put together some really excellent papers and I'll just briefly introduce them so we can get going. Robert Hetzel is a senior economist and research advisor at the research department with the Richmond Fed. Thank you for being here. David Papel is the Joel Saylers Professor of Economics at the University of Houston. And then we have Scott Sumner who is the Ralph G. Houghtry Chair of Monetary Policy here at Mercatus Center at George Mason University. He's also the Director of the Program on Monetary Policy and getting us started today will be Robert Hetzel. So two things from my comments. This will be obvious but these are my views not the views of the Fed, the Richmond Fed or the Board of Governors. The second thing is monetary policy is complicated. There's lots of moving parts and I hope you'll read my paper. Mercatus is not posting them on its website so please contact the Richmond Fed website and get my email and give me your email and I'll send you a copy. Thanks. So you can't pick up a newspaper without reading about what the Fed is likely to do its next F-form C meeting, how the choice of the funds rate is going to depend upon the evolution of the economy. And also, you know, presumably we know what our objectives are, a stable economy and price stability but there's something in the middle that's missing. And how does the behavior of funds rate setting translate into achievements of our objectives? There's a strategy that's missing in the middle. So I'm going to try to make some comments today about what I think of as the missing core of monetary policy communication. So you won't spend all your time looking at that pretty young lady. I'm going to go back to this. So academic economists have printily made the argument to central banks to please let us in on what's consistent in your behavior. What's your strategy? And knowing your strategy, can we figure out what works and doesn't work and make the good part into a rule? I'm not going to argue for a particular rule in my comments. What I'm going to argue for is that we're more likely to get some response from central banks if we follow an indirect strategy of pushing them on the issue of transparency in terms of what they actually do. Now central banks have adopted an inflation target but they did so without articulating the strategy for how to achieve it. And even that inflation targeting seems to be in question now coming under attack that while it prevented us from pursuing financial stability goals that we should have followed. So I'm going to try to bridge the gap between central banks and academic economists by arguing for better articulation of what is consistent in terms of what central banks are going to do. On the one side, academics want an arithmetical rule or often they do, not David Laidler, but often a rule that ties down the period by period choice of the central banks instrument, the funds rate. And policymakers contend that they can't follow an arithmetical rule. They need more judgment. I'm sympathetic to that point of view but I'm going to argue that it's somewhat disingenuous that central banks can be a lot more explicit about their objectives and their strategy for achieving those objectives. And I'm also going to take a Burkean point of view that central bankers are not likely to, from one day to the next, say, hey, we're going to adopt a rule and that's going to be a new monetary regime. We're much more likely to get progress again in terms of persuading central bankers to be more transparent about existing procedures. The proposal today is not going to deal with a lot of the current debate about financial stability, but I want to be clear about the point. I think the proposal will make clear what's monetary policy and what's credit policy or financial stability policy. And if we can't get the first part right, there's no hope that we can get the second part right. And I want to also say I'm honored to be in the presence of John Taylor who's done more than anyone in the world to try to kind of beseech, persuade, cajole, whatever central banks to talk about monetary policy in terms of rules. And my comments today are very much in the spirit of his 93 paper to build on what central banks are doing that's consistent and seems to be working. Okay. So I probably shouldn't have to do this. You know, we ought to, since we're all interested in monetary policy, we ought to know what the strategy is that central banks use to achieve their objectives. It's sort of strange that I have to get up here and give you my views, but that's how we're going to have to start. And you may or may not agree with me. That's not the point. The point is we need some kind of consensus over what that strategy is. So at least as I see it, understanding monetary policy starts from the inescapable fact that central banks, like the Fed, use the funds rate as their instrument. Well, think about it. The funds rate's an overnight interest rate. Who cares what the rate of interest is between today and tomorrow? Why should we care about that? So the problem that the central bank has to solve really is how to cause the term structure of interest rates to respond in a way that is stabilizing in response to the behavior of the economy. So that when news arrives, say the economy's growing faster than markets anticipated, you want the term structure of interest rates to rise with all of the rise being in forward real rates and none of the rise being in inflation premium, or ideally term premium, reflecting increased uncertainty. It follows to get that kind of behavior, the Fed has to behave in a systematic way. It has to communicate not only the near term likely path of the funds rate, it's got to communicate how it will respond to incoming news in the future. There's no way to get around that. Effectively, the FOMC has to act in a way such that forecasting the funds rate causes markets to forecast the real term structure of interest rates that causes output to grow along its potential path. So the funds rate and communication about its anticipated path are the proximate instrument and the term structure of interest rates in the way it responds to incoming information are the ultimate instrument. So how does the FOMC go about, I can't speak for the FOMC, these are my generalizations from public comments. So how does the FOMC go about imposing the kind of consistency that causes financial markets to behave in a stabilizing way? Well, over time it assesses the strength of the economy and on an ongoing basis it's got to make a decision about the growth gap, whether the economy is growing faster or slower than potential. And it's got to have some sense of what full utilization of resources is, that is what is the output gap. So if the economy is growing faster than potential, you're in a cyclical recovery, well, that's good, but that's not good if you overshoot the economy's potential long run growth. So again, the FOMC on an ongoing basis has to formulate some sense of what the growth gap is, what the output gap is, and in a judgmental way it makes a decision about whether the term structure of interest rates is appropriate for keeping the economy on its potential growth path. And here I think there's a legitimate difference of opinion that process requires a large amount of judgment. But the point is that it's a continual process of trial and error of trying to discover the natural rate of interest and responding when you don't have it right. So monetary policy then is not only the decision regarding the funds rate but it's the communication by the FOMC that shapes the behavior of the term structure of interest rates. And the FOMC has got to behave that way and it's got to be consistent about how it does it. And so I think the best way to get a rule is to push the FOMC on this issue of transparency. What is it doing that provides this consistency? Now to some extent we have this in the summary of economic projections and those are made four times a year. FOMC participants provide their guesses, estimates, predictions of what real output inflation, core and headline, the unemployment rate are going to be at the end of the year for the current year, the succeeding year, and then for the long run. They also in the form of what's known as the famous dot plot provide forecast for what they think the funds rate is going to be. Okay, well that's interesting especially if you're in the financial markets you can kind of figure out what the chair's dot is and there's information in that. But it's not a lot of use for the average person in terms of trying to understand what the FOMC strategy is. First of all, the participants are told to make their forecast based on appropriate monetary policy to what they would like policy to be. Well, so if you're the hawks and the doves they're both going to forecast inflation is going to come back to target, whatever it is. But that could mean very different paths for the funds rate and the forecast and the funds rates path they're disembodied. You have no way of going from one, you have no way of knowing how the individual forecast connect up with the forecast of the funds rate by the person engaged in the process. And finally, what's important is what the committee consensus is. It's not what the president, the Federal Reserve Bank of Atlanta thinks. It's how the chairman, the chair gets a consensus and that's hard to reveal. So my proposal is really very simple. It's just turn this SEP process, Summary of Economic Projections, which is kind of all 19 participants, constrain FOMC discussion four times a year to provide an FOMC SEP. So that the information we get now, or the path of the funds rate and over the forecast for output inflation and unemployment, those come out as a committee decision and we'll see that here. So how would that work? My lady, so the four times a year you would get three charts, three graphs. The first one's pretty simple. You just start in the current period and you start from the current price level and the FOMC has a target of 2%. So the benchmark path grows at 2%. And you also have a forecast for inflation. Right now inflation is growing below trends. So the path for the fund, for the price level is going to fall relative to the benchmark path, but at some point it'll come back and start growing at 2%, the same rate as the benchmark path. The second graph, the solid line is the FOMC's best estimate of what potential output is. Again, I've argued the FOMC has to operate with some sense of what that is, and we, at least in publicly available teal books, green books, the board staff does make that kind of estimate. So the solid line is the FOMC's best guess of the path of potential output. And that's, the FOMC has to behave in a way that the term structure of interest rates causes output to grow along that path. And so I've drawn a dotted line which shows the forecast for actual GDP growth. And I've done it with two different possibilities. So assuming where we start here at the beginning is where we are now, there's some slack remaining in the economy, can't be a lot, and the growth gap is still positive. So one possibility is for the economy to overshoot, go from ABDEF and create a positive output gap as a way of speeding the return of the price level to 2% growth. The other strategy is ABCEF, which is to conduct monetary policy with the intention of returning output to its potential path, keeping output growing along potential and relying on the way our implicit rule shapes expectations to bring inflation in line with its target. Okay, so I'm not arguing for one or the other, but the point is if we had an FOMC SEP along with a third graph showing the path for the funds rate, over time you could learn a lot about strategy. You know how the FOMC handles the situation where it's missing its objectives on different sides. It's sort of simple if you're missing both on the same side, but when we get to point B, which can't be too far away, what's the strategy for both getting to and staying on the long run potential path and the inflation path? My wife talks about Bob's theory of relativity. Time passes a lot faster when I'm talking than when other people are talking. So I think you get the idea. And again, if you're interested, ask me for a copy of the paper because there's a lot in the paper about how to make this practically implementable in terms of just one step beyond what the FES is actually doing. Thank you, Bob. We will have an opportunity to chew through those ideas a bit more in the Q&A. Up next we have David Papelle. Thank you, David. Okay, thank you first. Let me start by thanking George and Scott for the invitation and also say this is joint work with Alex Nikolsko-Roshevsky at Lehigh and Luxandra Protan from University of Houston. Policy rule legislation was first proposed by John Taylor in the 2011 Cato Journal. And as we saw in the last session, this has gone past House and Senate committees, past the House of Representatives. And one hallmark of this is that the Fed chooses the rule. It's not legislating any particular policy rule in particular, not legislating the Taylor rule. So the question I asked today is, suppose the Fed were to adopt the policy rule. And in this context, it could adopt it because it's legislated, because it wants to preclude future legislation, or because it feels that it's a good idea to do that. The question is, which rule should the Fed adopt? And so in contrast, probably with everybody else, we are going to propose a specific policy rule that the Fed should adopt, which reminds me of the Alexander Pope quote, how fools rush in where angels fear to tread, but nevertheless, we're going to do that. Three criteria. The first criteria is consistency with good, which we left out, long-term economic performance. Second criteria is consistency with Fed policy following the Great Recession. And third is consistency with projected Fed policy going forward. It would be very specific. The first criteria is normative. The second and third are pragmatic. It's just inconceivable to us that the Fed would adopt a policy rule that said, we've done the wrong policy for the last eight years and we're going to do something different than the dot plots are saying over the next five years. The second and third have nothing to do with normative. Those are purely pragmatic. The first is normative. Now to get terms settled, we're going to look at three classes of Taylor rules. The first class is, and so any type of Taylor rule will have the nominal interest rate, equaling the inflation rate, coefficient times the inflation gap, inflation minus target, plus the coefficient times the output gap, plus the equilibrium real interest rate. And in this context, if with those coefficients and a fixed equilibrium real interest rate of two, this is Taylor 1993 rule. Take the same coefficients, but use the time varying equilibrium real interest rate from the work of Thomas Laubach and John Williams. You get Janet Yellen's 2015 rule that she gave at the speech of the San Francisco Fed. Another alternative is output gap tilting rules. Take the original Taylor rule, double the coefficient on the output gap, and you get the Yellen 2012 rule that she gave at Boston Economics Club. Or you can reverse that and have what we call an inflation output gap tilting rule. So the first we call a balanced rule. One thing about a balanced rule is that the balanced rule gives you the same impact on the real interest rate as if inflation goes above target or if the output goes above potential. And so that's balanced is not just the coefficients balanced as in the economics of it. The output gap tilting rule tilts more on the output gap side, or you can have an inflation gap tilting rule in the example we're going to use in this paper would have a coefficient of one on the inflation gap and 0.5 on the output gap. Now, one thing we do here that I can't get into the details on is known as real-time data. Data that was available to FOMC members when they're making their decisions. And that data is different over long historical periods. What you can get, our data is going to go for the long historical period from 65 to 2015, or what you can get more recently. One thing I do want to mention is in terms of the policy rate, we use the federal funds rate up through the end of 2008. After that, using the 0 to 0.25 basically assumes that there's no impact on the economy of quantitative easing or forward guidance or anything. And we need some measure, our measure is we use the shadow federal funds rate from paper by Cynthia Bowen-Doroshiar. This is pretty much in the middle of a lot of estimates of shadow federal funds rates. And then back, starting in 2016, you go back to the federal funds rate. We also use in the paper four measures of equilibrium real interest. Rate two, as in John's 1993 paper, trend growth for the previous 10 years, which is in the spirit of the 93 paper, Lawbuck and Williams data, and that a new paper by Lawbuck and Williams with third author from the Fed, which is very similar to the Lawbuck and Williams work. I'm going to concentrate today on the first, on two and Lawbuck and Williams. So when we think about long-term economic performance, this is based on the paper that the three of us have that's just sort of coming out in the first draft also. And the idea here is to work what we call policy rule deviations. The idea of a policy rule deviation is take the prescribed federal funds rate from any rule minus the actual federal funds rate. And we're going to call those periods rules-based if these deviations are small, discretionary if the deviations are large. Now, to evaluate policy, a standard way is to use quadratic loss functions. Inflation minus the target inflation squared plus unemployment minus the target by the natural rate of unemployment squared. And you want to minimize that loss function. Now, the loss function over time is independent of the policy rule. It's whatever it is. It's whatever it's been for the last 50 years. That's the function. What we do here, though, is you divide the time into discretionary periods and rules-based periods. And so what you want, a good policy rule, is that if you're close to it, you have good economic performance and you have a low loss function. If you're far away, you have worse economic performance and you've a bad loss function. So we have what we call quadratic loss ratios, the ratios of the loss in the discretionary and rules-based periods, and a good rule has a big loss ratio. So let me give you an example. Suppose you take the original Taylor rule 65 to now. The blue line represents deviations. The shaded areas are where the absolute value of the deviation is greater than 2%. The white area is where it isn't. The loss ratio is the average loss in the shaded area divided by the average loss in the white area. So what we do in the other paper is say, let's look at this with 100 different rules. So we're taking coefficients from 0.1 to 1 on the inflation gap and the output gap. There's a benchmark, which is all I'll show you in this paper today, that has equal weights. It has r star 2. It has a threshold of 2 and a policy lag of 6 quarters to incorporate a maximum effective Fed policy between 1 and 2 years. And you can change a lot of these things. So here's an example of what you get by doing this. What you have are quintiles of loss functions. So dark green is the first quintile. Those are the best rules with the biggest loss ratio. Light green is the next best. Yellow is the middle. The brownish-orange is the fourth. And the red is the fifth. And so I want to concentrate on three of these. First, we can say look at the one in the middle, the 0.5 and 0.5. That's the 93-tailor rule that's in the second quintile. The one and 0.5 in the top is the inflation tilting rule. That's in the first quintile. That's better. The one over there to the right on the 0.5 and 1 is the Yellen 2012 rule. That's in the fourth quintile. That's worse. And if you look at this, however you measure equilibrium real interest rate, the inflation tilting rules in the first quintile, the balance to second and third, the output gap tilting is fourth and fifth. So consistency with long-term economic performance tilts towards inflation gap tilting rule. Let's look at the following the great recession. Again, we're going to look at two things. We're going to look at average deviations. And we're going to look at the deviation at the end of the sample. The deviation for second quarter of 2016 because you don't want to have a jump. If you're going to adopt the policy rule, you're not going to say, well, we're going to jump 2% in the next quarter. And so just to look at a couple of these pictures, the upper left is the original tailor rule. And it's a well-known story that the prescribed rule in blue is higher than the shadow rate in red and how this is not consistent with Fed policy since the great recession. That's not a new story, obviously. It's certainly not original to us. Look at the bottom left. That's the Yellen 2015 rule. Same coefficients. Lobach and Williams. Much more consistency with Fed policy following great recession. For example, of one reason why she promoted this. Output gap tilting rule. Just going to look at one. If you look at the top left, that's the Yellen 2012 rule. Again, more consistency with Fed policy than the original tailor rule. Here are inflation gap tilting rules. Just want to show you, look at the bottom left. Lobach and Williams one doesn't do too badly. So here you can look at this in numbers. Look at four things. If you look on the bottom left, you can see these balanced or inflation gap tilting rules with time-varying equilibrium real interest rates do pretty well. The fixed rate does not accept, in that case, with a fixed equilibrium real interest rate, the Yellen rule does better. Next thing we look at is consistency with projected Fed policy. We use output gap projections from FOMC, recent FOMC meetings, output gap projections from CBO. One thing to see here is because the CBO has projected a zero output gap by the end of 2016, the tailor 93 in Yellen 2012 rule gives the same answer for this because it doesn't matter what the coefficient on the output gap is if it's zero. So let's see what you get here. Those rules don't accord with projected Fed policy. If you look on the left with the December 2015 SEP, then you see you have basically a two percentage point difference at the end of 2015, both of these rules. Not just the tailor 93, but also the Yellen 2012 rule is two percentage points higher. They converge over time. You move into further SEPs, the differences get even larger. Now, what about with time varying equilibrium real interest rates? With time varying equilibrium real interest rates, you can't ask the same question exactly because you don't know what the time varying equilibrium real interest rate will be going forward. So the question we ask is, what Lawbuck and Williams time varying equilibrium real interest rate would be consistent with the Fed SEP projections? And one thing to keep in mind is that that number is now about point two, both in first and second quarter of 2016. Did I do two at a time? Okay, so if you look at this, let's look at the December 2015 SEP. December 2015 SEP, the projections were pretty much on target with a Yellen 2015, well not projections on target, but the time varying Lawbuck and Williams rule seemed to make sense. It started at point two at the end of 2016, which actually doesn't look, I'm sorry, was zero at the end of 2016, which looks low now, goes up to 1.5. Look at the inflation targeting version, that's even better. Starts at point two at the end of 2016, that seems pretty reasonable given that's about two right now, and then goes up to the 1.5. You move into the March projections, the June projections, the fit starts to get worse. You get negative Lawbuck and Williams rates by the end of 2016, where those rates have already gone up above zero, so the fit gets worse. Okay, so last thing to talk about is a week ago, Janet Yellen proposed yet another rule at the Jackson Hall conference. This rule is an unemployment gap rule, as opposed to an output gap rule, and it's what I call an unemployment gap tilting rule because it has a coefficient on the unemployment gap, which is consistent with a coefficient of one on the output gap, which would be an output gap tilting rule. Of course, you could have had a smaller coefficient on that and had balanced or inflation gap tilting rules. And so what do you do if you do that? So we just do the same thing we did before for the output gap, tilting rule and see what you get. This Yellen 2016 rule is still in the fourth quartile of long-term policy rules, which means it's no better than the Yellen 2012 rule. If you just take the Yellen 2012 output gap tilting rule and put in an equilibrium real interest rate of one, which she does on the Jackson Hall paper, you get into the fifth quintile, you get even worse. So this rule is not consistent with long-term economic performance. Let's look at, following the Great Recession, let's look at the picture. The picture is the unemployment gap tilting rule said the Fed should have tried to drive the shadow rate down below negative 8% in 2009, huge deviations there, but equally, maybe more importantly, look now, this rule is saying that the federal funds rate should be two or above two. So it's actually giving a bigger gap at the end of the sample than the Taylor 1993 rule in the same direction. So the average deviations are negative, but the end of the sample, it's positive and even larger. What about projections going forward? You take this rule, you feed them into SEP projections for unemployment rate, the Yellen rule has a 4.8% natural rate of unemployment, feed that in, what's the path of the federal funds rate to be in accord with this rule? And it's actually 2.75% at the end of 2016, which seems doubtful to say the least, but then it gets worse because it goes up to 3.25% at the end of 2017, 3.4% at the end of 2019, and then goes three, so it actually gives you a path that overshoots the long run level, and so it's not consistent with... So we have criteria, we say what works best, the inflation gap tilting version of the Yellen 2015 rule with the Lobach and Williams equilibrium real interest rates is consistent with the three criteria, and it's basically the only rule among these promoted by high-fed officials that is consistent with these, and finally the Yellen 2016 rule moves much farther, very far away from consistency with any of these. Thank you. Thank you, David. Next up we have Scott Sumner. Thank you. David Laylor mentioned earlier today about one possible definition of emeritus professor, and I just became an emeritus professor at Bentley University, so I'm reminded of that. I think an alternative definition might be free to write self-indulgent papers on anything you want, and in that spirit, I'm going to talk some about how I see the field as evolving over the last decade before I actually get to my specific rules proposal a little bit later. I want to try to figure out what, in my view, sort of went wrong, how we got off course, and how we can return to a monetary regime that worked more effectively in past decades. So I'll look at three sort of heterodox ideas. Some would say four, although I consider market monetarism, which is my own view, to be much more orthodox than many people assume. The three heterodox ideas I'll briefly discuss are a revival of old Keynesian economics. Second one, a focus on financial excesses instead of simply targeting macro variables and what's called neo-fisherism. Starting with the old Keynesian ideas, all of these three, by the way, have been sort of revived by this extended period of zero interest rates, and I think this is really what's thrown monetary economics into turmoil. People haven't known how to deal with this, how to think about it, what it means, and we've sort of veered off on all sorts of directions. One is not particularly surprising. We've gone back to a lot of old Keynesian ideas, and old Keynesian economics was developed also at a time of near zero interest rates. And some of these ideas, of course, are monetary policy is supposedly ineffective at the zero bound. You want to use fiscal stimulus, and interestingly, the proponents tend to prefer more G, more government spending, rather than less taxes. Most interesting to me is a revival of ideas from what you might call the underworld of economics. Ideas that had been sort of discredited in the classical view, like the jealousy of trade, that might be Germany's trade surpluses, stealing jobs from the rest of the world, paradox of toil, the idea that wage cuts or cuts in unemployment benefits actually reduce the number of jobs. The idea that deficit spending may pay for itself through growth, and just this general idea that everything is different. That is the rules of classical economics and opportunity costs don't hold at the zero bound. Now, of course, this is not my view, and I won't go into detail here about all my objections. Maybe I'll just focus on one of these monetary offset in 2013. So, in 2013, the budget deficit in the U.S. fell nearly in half, from a little over a trillion to a little over 500 billion. And this was done through a combination of tax increases and spending cuts, and at the time in early 2013, a lot of Keynesian economists were predicting a sharp slowdown in the recovery as a result of this so-called austerity. At the same time, many market monetarists like myself were more skeptical and pointed to the fact that monetary policy tended to offset changes in fiscal policy. And at the end of 2012, monetary policy became more expansionary with forward guidance and QE, partly intended to offset this austerity. Well, the experiment was played out, and what happened is growth actually sped up in 2013. So I think one of the big problems in sort of Keynesian fiscal analysis is the tendency to overlook the fact that monetary policy is targeting average demand and largely offsets many of these fiscal decisions. Let me go on to the second one. So this one is more, I think, on the conservative side of the spectrum. There's been sort of a lot of anxiety and concern about how an extended period of low interest rates can lead to financial excesses. And there's a big recent paper out of the Bank for International Settlements that really sort of discusses and encapsulates a lot of these ideas. And some of the concerns are that extremely low interest rates may lead to excessive debt, financial investment in dubious projects, and asset price bubbles, and that ultimately these financial problems will create macroeconomic disturbances in the long run. Now, and by the way, they're very explicit that they're even willing to sort of miss inflation targets for a period of time in order to battle some of these financial market instabilities. So this is one challenge of a strict inflation targeting approach. Again, I have some problems with this view. I think that policymakers aren't good at spotting bubbles. Another one is that the low interest rates that they're concerned about have been really on a downward trend for like 35 years. So it's not at all clear these low interest rates are actually due to an expansionary monetary policy. Now, most interestingly, I'll focus here on this fourth point. There have been a number of attempts to raise interest rates at the zero bound that seem to have backfired. The U.S. in 37, a couple of times in Japan, and most recently Eurozone in 2011. I think one of the things that we can learn from these failed attempts to raise interest rates off the zero bound is that you can't really put the cart before the horse. If your objective is to get a durable increase in interest rates, the economy is strong enough to sustain a tighter monetary policy. And importantly, if it's not, you actually end up hurting your cause. So in Europe today and for the foreseeable future, they will have considerably lower interest rates than the United States. I would argue partly because they attempted to raise interest rates in 2011 and created a double dip recession. Now, the 2015 Fed rate increase so far has turned out better in the sense the economy is doing okay this year. But even there, we can ask a question is whether the goal of achieving higher, longer term interest rates over time to sort of reduce financial market excesses, has that really worked? Well, longer term interest rates have fallen since December of 2015. So even there, it may not have achieved the objectives that the BIS was looking for. For neo-fisherism, this is kind of an interesting theory. And it's a theory that in the end I don't really accept, but maybe should be taken a little more seriously than it is. So they kind of turned things upside down in economics by arguing that a policy of low interest rates for an extended period of time can actually create low inflation. And this course goes against the standard view that low interest rates are an expansionary policy and will create higher inflation. One thing they have going for them is that over the longer term there is of course a positive correlation between nominal interest rates and inflation. And apparently some of the New Keynesian models have predictions that at least under certain conditions you can get these kind of results. Now, that's been criticized by Michael Woodford in a fairly long paper, but what I'd like to do is focus on a different aspect of the problem. I don't think the issue of whether low interest rates create low inflation is really even a question at all that can be answered. It's sort of an under-determined statement because you first have to stand back and ask, well, what is causing the lower interest rates? And I would argue very simply that lower interest rates produced by tight money will lead to lower inflation. Lower interest rates produced by expansionary monetary policy will lead to higher inflation. So what's being missed here I think is that the sort of interest-oriented New Keynesian models overlook the fact that monetary policy works on two levels. And you can see this more when you look at other indicators like money supply or exchange rates. It works on a level of sort of level changes and also growth rate changes. So you can have a one-time change in the money supply or you can have a change in the money supply growth rate. You can have a one-time change in the exchange rate or you can set up a crawling peg where the exchange rate changes gradually over time. And depending on which type of change you do in monetary policy, you may get a different correlation between the change in interest rates and the change in inflation. So this is, I think, what's being missed here. And the one example that's sort of the exception where the Neofisher's view may be correct is a very interesting one in Switzerland in 2015 in January where they cut interest rates sharply to 75 basis points negative. And they're probably going to end up with a lower inflation rate. Now, how did that happen? Normally when you would cut interest rates sharply, your currency would depreciate in the foreign exchange market. But at the same day, the Swiss essentially sharply revalued the Swiss franc higher and that combination of policies is what's going to produce the Neofisherian result in Switzerland. In most cases, though, when you cut interest rates sharply, your currency will depreciate and so you will not get lower inflation. So market monetarism. There's a lot of things that could be said about this, but let me just say that the last item on this list, nominal GDP targeting, I think what's most associated with market monetarism, but is perhaps the least important in my view aspect of this. So the other ideas I have here, I've got a bunch of names, well-known monetary economists that are attached to some of these ideas, and what I'm trying to do in the paper is show you that the way we think about monetary policy has changed a lot in the last 10 years and in my view, our views have often regressed. Many of the economists that I've associated with some of these ideas no longer seem to hold these ideas. At least their recent public statements conflict with what they said before. Here's Frederick Mishkin from his 2008 textbook. The first one is low interest rates are not necessarily easy money. The third one is monetary policy is highly effective in reviving a weak economy even at zero interest rates. Clearly, a lot of economists don't believe either of these things today, but in the case of the third one, even his textbook has changed. It's highly effective and replaced it with just effective. Ben Bernanke in 2003, well, the first quality talked about how policy in Japan was a case of self-induced paralysis and that they should consider switching to level targeting, which I think would have been a good idea, but the Japanese didn't do that and neither did the Fed at the zero bound. The third point, Bernanke says that you can't judge the stance of monetary policy by interest rates or money supply growth and instead you want to look at nominal GDP growth or inflation. Well, it's pretty clear that Bernanke doesn't believe this anymore because from 2008 to 2013, by that criteria we would have had the tightest monetary policy since Herbert Hoover and yet all during that period, Bernanke was insisting that policy was highly accommodative. So there's been a change in how we think about that. Paul Krugman, 1999, I won't go through this whole quote, but he's sort of incredulous that people are suggesting fiscal stimulus for Japan, which is at the zero bound, when obviously the ordinary, boring monetary policy stimulus would be the appropriate thing to do. This is dramatically different from what Paul Krugman is saying today and it's notable that this quote is from 1999, one year after his famous Brookings liquidity trap paper that he often refers to today is changing his mind on the Japanese situation. So how do we make market monetarism operational? How do we bring these views together? Let me back up just a moment here, by the way. On the last few items here, we've got level targeting. Central banks should target the forecast. Lars Vensen is associated with that idea, and that's also a market monetarist view, but the extra wrinkle we put on it is he focuses on internal central bank forecasts and we focus more on market forecasts, so that's why I mentioned rational expectations and asset markets. Now, getting back to this point, one way of doing this is you could create a sort of a tailor rule type reaction function for nominal GDP instead of the current dual mandate. It wouldn't be dramatically different from the current tailor rule being the same spirit, or you could go to a target the forecast approach and internal central bank targets or internal central bank forecasts could be used or market forecasts. The basic idea here is to equate the policy forecast with a policy goal. It's a very simple concept. Two percent inflation target, you should also be forecasting two percent inflation. I've suggested using nominal GDP targeting futures markets as a guide to this sort of policy. Since I don't have a lot of time left, I'm gonna sort of get right to the punchline here and go ahead and talk about what's called the guardrails approach. So we heard about the gold standard earlier. This is very analogous to the gold standard. We're essentially gonna make dollars convertible into nominal GDP futures contracts. The simplest way would be to convert them at a fixed price. And that way, monetary policy would have to be consistent with what the market thinks nominal GDP will be in the future under the current instrument settings. Now, if you want to allow some discretion, constrained discretion, just like under the gold standard, you could create a band of gold prices from, say, 32 to $38 an ounce. You could create a band of, say, three to five percent where the Fed would take a short position on five percent futures contracts and a long position on three percent. Essentially, the Fed would be the counterparty so that anyone that thought the Fed was gonna end up outside this range could make that bet and the Fed would take the other side of that bet. I just want to end up. I don't have time to get into the details here, but to give you what I think is the intuition of how this approach to rules is different from what people usually think of. People usually think of rules as sort of a burden we're putting on central banks that constrains them, prevents them from the free exercise of their choices. I don't think that's the right way to think about policy rules, at least the one I'm proposing. I think central bankers are trying to achieve good outcomes, and we want to have monetary policy rules that help them achieve what they're trying to achieve. That is good outcomes. Think of guardrails on a twisty mountain road going along a ridge where there's steep drop-offs on both sides. You don't, as a driver, view those guardrails as an imposition that prevent you from driving the car in any old direction because you obviously don't want to go over the edge of the cliff, right? So those guardrails are helping you as a driver. These two futures' market boundaries of 3% and 5% would be sort of like warning signals like a truck has when it backs up and it starts beeping if it's going to hit something. If there was a lot of action on either the 5% or the 3% were betters, were betting en masse that they were going to miss in one direction or another, that would provide the central bank with useful information about their policy being off course, and they could nudge it back to a position within that 3% to 5% range. And the Fed could narrow that range over time if they thought it was a successful regime, but having this band would allow you to sort of go into it gradually by setting the band initially fairly far apart, still giving the Fed some discretion, see how it works if it seems to be working well. Over time, gradually narrow that band. Ultimately, you would hope if it worked well you would end up right at a point like 4% nominal GDP growth where the Fed would take bets in both directions and the Fed would basically be trying to end up with a situation where it wasn't going to lose a lot of money either way, that the bets were roughly balanced on both sides. So think of that as something like a gold standard structure where money is convertible into these contracts, but think of it as assisting the central banker in imposing a burden on the central bank. Thank you. Thank you, Scott. Thanks to all three of you. To start the discussion off, I think I'd like to ask kind of a big philosophical question. It's one that the last panel sort of got into a little bit and that is, this panel is entitled Monetary Rules and Monetary Stability. How certain are we that stability is an achievable goal? We see over time that monetary regimes tend to work for a bit and then fall apart. I can imagine lots of reasons why that might be the case, the relationships between variables change or things of that nature. Are we sure that we can actually create a rule that's going to be durable? Scott, I'll start with you. So my way of looking at this is all these rules are provisional and the way we should think about it is not creating a single rule that lasts till the end of time, but a series of rules where each one is better than the one before. If we take sort of the wig view of history, we had, you know, gold standard, we had Bretton Woods, we had this period of pure fiat money with no constraint. So we've learned lessons from each mistake. We've learned how to avoid the mistakes of the Great Depression, we've learned how to avoid the mistakes of the Great Inflation, but we make smaller mistakes under these improved rules like the 2% inflation target rule and so the way I think about nominal GDP targeting is just sort of a further iteration. It's still in the spirit of the 2% inflation rule, but does a little bit better job of dealing with cyclical fluctuations and after a few decades of that, someone will see some problems with that that could be improved even further by another rule that tweaks it a little bit further. So I think of it as a process, not a single rule that will end the discussion for all time. So I'm an old line monitorist. I've been accused of being a high church monitorist. And the monitorist hypothesis is that as long as the central bank provides a stable nominal anchor in terms of the domestic price level and then follows a rule that allows the price system to work freely to determine real variables, the price system does a pretty good job at stabilizing the real economy. So here I'm going to put in an advertisement for my paper. Recessions are unusual events. They're not standard. So central banks have to have a rule that allows the price system to work basically as a baseline rule. And the question is, what happens with recessions? Do those recessions occur when the central bank departs from its baseline rule? Or do they simply come from the animal spirits of the private sector and the central bank departs from its rule in a stabilizing way? And so the advertisement for my paper is that we're never really going to learn, we're never really going to settle these issues unless central banks become clear about the rules they're following so that we can identify departures and then we can ask, well, are those departures causal in terms of recessions or are they constructive responses? I guess when I think about this, one thing to think about is that there's a lot of evidence that economic performance is better in what John has called rules-based errors than discretionary errors. Alan Meltzer has very exhaustive historical evidence of this. John's talked about this in his paper in JMCB. Alex Roxandra and I have talked about this earlier more quantitatively. And I think when you think about this, there's progress that's been made. One thing where the progress is made is that every policy rule that we talk about adheres to the Taylor principle, that when the inflation goes up, that raises the nominal interest rate more than point for point, and you need that in order to stabilize inflation in whatever old Keynesian models, new Keynesian models, whatever type of models you have. And so I think this has been a big step forward we did not do that in the 1970s and there was a very high price to be paid for that and I think that sometimes we emphasize the differences instead of the similarities. That's interesting though. I think I sense a little bit of disagreement here between Scott and the other two. When we think about the failures that led to the Great Recession, was that an intellectual failure? Was it that we hadn't learned the lessons yet and had the wrong rule? Or was it an operational failure and that the Fed failed to do what it ought to have done given the information it had? Well, I've written on that critically of central banks. Again, I think you need to distinguish clearly between monetary policy and credit policy and most of the discussion deals with credit policy and monetary policy gets a free pass. As I see it, central banks around the world behaved in the same way. We had a huge commodity price shock which began in the summer of 2004 and continued until the summer of 2008 and even though the Fed started following its usual lean against the win procedures in September 2007 lowering the funds rate, it became concerned early on in 2008 that it was going to lose credibility because inflation was so high above its target. So it backed off and sent signals that the easing cycle was over effectively. Again, I'll argue this in my paper. You can look at it. Central banks were trying to stabilize nominal GDP growth in a way that would trade off creation of a negative output gap as a way of lowering high headline inflation and try to get it Goldilocks just right. Maybe it would have worked. Then you had the financial crisis which as I see it created enormous uncertainty, part of which was abetted by policymakers that reduced the market clearing natural rate of interest. Central banks were very slow to realize that and a moderate recession turned into the great recession. I think John Taylor answered that question better than anybody else has in 2007 where you look at starting in his work 2003 in our work maybe a little bit earlier, 2002-2001, the Fed went from a policy where you were approximately in line with the Taylor Rule to a policy where interest rates were well below the Taylor Rule prescriptions and that what did we see? You saw a search for yield. You saw more risky behavior. You saw developments in housing markets and you can say there are regulatory issues there but I sort of look at to say, take Barry Eichengreen's phrase in a different context, original sin that you wouldn't have had those if you didn't have the very low interest rates and the rates on safe assets being so low and you're searching for yields on riskier assets and that's where those came then. We don't really know where they're coming now which is something that worries me. We don't know where the search for yield is going then. This is something we learn after the fact. Scott, do you agree with those assessments? Well, I put a little more of the weight on mistakes that were made in 2008 not to discount earlier mistakes but the way I would think about it is if you read Bernanke's critique of Japanese policy written like a decade earlier and what he thought the Bank of Japan should be doing I think those kind of suggestions if adopted in the United States would have prevented the Great Recession and we would have had a mild recession instead. Now I admit to being complacent I assume that the Fed was going to do those things because Bernanke was chair of the Fed but I think at the time I didn't really understand the institutional complexities of the Federal Reserve and just because as an academic you favor something doesn't mean you can move the institution and maybe Bernanke's views themselves changed or were not what I assumed but a couple specific points there wasn't enough emphasis on targeting the market forecast the markets were way ahead of the Fed in late 2008 but I don't think the Fed was even targeting its internal forecast. In other words, by late 2008 the Fed's forecast for the economy was much worse than they wanted and that combined with a failure to do level targeting the zero bound for economic stability you really need level targeting to come back to the trend line when you deviate above or below it and once the market sensed that the Fed was not committed to come back to that trend line the bottom fell out and I think there was a tremendous loss of confidence in the fall of 2008 in monetary policy as the markets correctly saw that the regime they thought the Fed had and the one that I thought the Fed had was not in fact the Fed's regime and they didn't have a way to prevent a deep and persistent crash in aggregate demand. We've talked a little bit about credit market failures and sort of poor decisions made by investors it seems to me like there's a little bit of a time consistency problem where you might have with central bankers where you might have an excellent rule but once you get to a certain point and asset prices are doing crazy things they can't resist the pressure to step in and to pull the punch bowl away and I'm wondering should there be some incorporation of financial stability variables into a rules-based framework to help sort of protect against that? Jeremy Stein had talked a little bit about trying to systematize the way you look at financial stability variables is that something we should consider? Bob, do you have? Well, the focus of monetary policy before World War II was on financial stability the assumption that policymakers could identify bubble's asset speculation and then they could make a trade-off by slowing the economy in a manageable way that would quench the speculation and then the economy would continue to grow again and we have a lot of experience and that period between when the time Fed was created and the Treasury Fed accord and it was a disaster and I have no more confidence that policymakers today can judge the correct level of asset prices than they could in the 1920s to me it's a slippery slope on the way to some hot place. Should we then just take away financial stability duties from the Fed and hand that entirely to a regulator? If we get monetary policy right I have every confidence in the world that those greedy people out there on Wall Street will figure out ways to get money from savers to investors. Now, the Fed doesn't currently have a strict rule that it's following it does have a target though and it made that target public and hasn't done a very good job hitting it since then. How precisely does, what is the mechanism by which the adoption of a rule fixes that problem? I mean, Scott, as you said there are institutional complexities at the Fed. Are we confident that the rules are going to be robust to those? Oh, well that's a couple questions. I think it to some extent it depends on the type of rule and so I think if the rule is set up as a level targeting type rule either for prices or nominal GDP it actually constrains the Fed a lot more than the current type of growth rate rule and David Glasner referred to that this morning. It really puts a lot more constraint on than you might think. Think about it this way. Suppose you have a level target for the price level and it's rising at 2% a year and let's suppose that every single year the Fed falls 1% short, okay, year after year after year. Today you'd have 1% inflation every year. Under a level targeting regime you'd still have 2% inflation under every single year except the first year when you fell 1% short. So it's very difficult for a central bank to diverge to any significant extent from their long run promises under a level targeting regime. So I think if you ask what extent does this constrain them, it depends on the type of rule you have and some rules allow a lot more wiggle room than others do. I think when you think of inflation targeting the way the Fed does and a specific 2% inflation target I think that's one step towards a policy rule but only one step and this is what I mean. If you have an inflation target that means that you have to have the Taylor principle because if you have an inflation target if you don't satisfy the Taylor principle if when inflation goes up you only raise nominal interest rates point for point you don't raise real interest rates you never bring inflation down. So inflation targeting implies the Taylor principle or in terms of my equations implies a positive coefficient on the inflation gap or more than one coefficient on inflation. Add to that the dual mandate that adds a positive coefficient on the output gap in the Taylor rule context. And that's Ben Bernanke's constrain discretion that we have an inflation target we have a dual mandate and that's all. The policy rule goes beyond that and says we're going to have specific coefficients and that can hold the Fed accountable. The Fed can, if they deviate they have to say why they'll say why they're deviating so I think that's a step forward from that but I view the inflation target as a first step. Let's take some questions from the floor. There's a microphone over here waiting for you all. Would you like to walk to the mic? Say who you are and who do you like the question to be answered. Yes, to all of you. John Burla, competitive enterprise institute. My question is do you think some of the new bank regulations like Dodd-Frank and Basel III may be working at cross purposes with an expansionary monetary policy? Preventing loans including loans from qualified borrowers is contractionary while the monetary policy quantitative easing low interest rates are meant to be expansionary. Steve Hankey who is an adjunct fellow at Cato also teaches at Johns Hopkins makes the distinction between state money created by the Fed and then bank money not available in the private sector and believes they are working at cross purposes so I'd like your views on that. So Dodd-Frank is an unfortunate consequence of the fact that Congress is unwilling to limit the financial safety net by guaranteeing the deposits of banks because it's a subsidy it can offer that doesn't appear on the books. The political imperative is just overwhelming and so if you're not going to let the market discipline banks the regulators got to do it and you got to trust the regulators so it's like Fadisha in the early 90s we know that worked beautifully and now we've got Dodd-Frank and the question is whether it'll become so onerous that it'll force financial intermediation off the regulated sector into the unregulated sector. My own feeling as I've expressed just a few minutes ago is the key thing is for central banks to get monetary policy right. If they get monetary policy right then we can go on and have a discussion do they need these extra ad hoc tools for intervention in credit markets to allocate credit toward housing and so on. I don't think so but at least we can have that discussion. Yes, Chris Ezracaglou, investment advisor again. Two questions if I may. One, I just was curious if the panel has any thoughts on the repealing of the Glass-Steagall Act and any role that that might have played in the 2008-2009 financial crisis we went through and then a question for Scott Sumner. In your talk you had referenced how interest rates were falling over the last 35 years and I just am wondering why you picked that particular timeframe because in 1981 we had the highest 10-year Treasury bond rates that I believe in the last 100 years and I believe that was an anomaly as opposed to a more normal interest rate environment which I believe is in a 5-6% range so if you just maybe comment on why referencing the peak and all-time peak in interest rates. Thank you. Yeah, so good question. So the decline over 35 years in interest rates has been two factors. As you point out we had high inflation back then so part of the high interest rates was reflecting the Fisher effect but it's also true that most estimates of real interest rates have been declining most of this time for 35 years and I guess I should also say that fact by itself doesn't tell us much but if you juxtapose that fact with the fact that inflation and nominal GDP growth have slowed down at the same time there's really no monetary model that can get you 35 years of falling interest rates due to easy money and at the same time that long a period of declining inflation and declining nominal GDP growth. There have to be some fundamental factors beyond so-called easy money that are doing that otherwise by now we would have seen a pickup in inflation in fact long ago we would have seen a pickup in inflation so there's some not well understood things going on in the global credit markets which seem to be reducing what's called the Vexellian equilibrium interest rate over time. You can talk about many theories of why that is but I think we just observed that we're not seeing higher inflation that's pretty powerful evidence that market forces more than the central banks are actually driving these rates down. Glass-Steagall? I'll make some comments. I think it's important to put it into a longer run perspective. In the late 1980s regulators got all concerned that American banks were in a death spiral that they couldn't compete with international universal banks and one problem was that American banks were having to hold too much capital relative to Japanese banks and so on so regulators encouraged or at least turned the other way when bank holding companies took financial intermediation off the books of their banks and put it on to these SIVs structured investment vehicles financed by commercial paper short-term lending financing long-term mortgages and it seemed like the right way to go to make American banks more competitive. The problem was it was a fragile system when you got the layman in September of 2008 the situation was that markets just assumed 30 years of history from Franklin National to the present to leverage the fail there was no way regulators would allow a Wall Street firm to not make good on its debt and when Paulson Bernanke said no we're going to let one of these firms fail the cash investors who had been in this sort of artificially created shadow banking system all jumped the line to the too big to fail insured banks and so it was that stampede from one side of the line to the other where the whole line was shifting that caused all the chaos and the difficulties it would so again just to summarize what I said I don't think regulators realized that they had created a system that was so fragile that when they changed the rules of the game in terms of the financial safety net that the cash investors would jump from one side to the other so I don't think Glass-Steagall per se its repeal was the problem I think it was this interaction between the financial safety net and thinking you could change it from one day to the next and not have chaos I think that was the problem Next question Hi this is Usmar Barton I'm from Georgetown Law School my question is on Ryan's comment on whether financial stability is an achievable goal or not I'm asking in context to the proposal in the senate about establishing a monetary oversight commission or some body the question is to Robert and maybe Scott as well can comment on that whether establishing or putting up a committee for monetary oversight would add to financial stability Well the issue is it gets to independence versus accountability and it's a difficult situation under a constitution Congress is responsible for the monetary system and it's jealous of its prerogatives it doesn't want to give that control over to the treasury but the Congress can't at the same time the Congress can't run monetary policy so basically it's delegated creation of the monetary standard to the central bank and said go run it and yet the central banks haven't told us what the monetary standard is that they've created so how do you have accountability and I know there's only two ways you can go and to deal with that one is to increasingly give control over monetary policy to the treasury and then say you've got accountability because the president is elected or to move toward the current system where Congress maintains its constitutional progry for the control of monetary policy but to move in the directions of rules so that you actually can have accountability because you know what the central banks are trying to do just a brief comment so I've suggested increasing accountability and transparency in monetary policy by instructing the Fed to set up its own procedure for doing so in this way each meeting or periodically the Fed would produce a report saying in retrospect the policies we implemented a year ago were too expansionary, too contractionary, or about right so they would tell us whether they made mistakes in previous years and they would do this every so often that would make it easier for us to see what they're trying to accomplish because as part of this report they would point to specific statistics or metrics you know it could be inflation, GDP, whatever unemployment that led them to the conclusion that in retrospect previous decisions were either on target or off course now today all they do is they go back decades and they say yes we screwed up in the Great Depression or the Great Inflation but they don't tell us how they screwed up a year or two years ago and I'm not doing this to or suggesting this to embarrass them I'm suggesting this because I think that it would actually force the central bank to sit down and this is sort of in the spirit of your paper too Bob it would force them to sit down and come up with a better explanation of what they're trying to do so that we could see how they're judging success and failure right now they take credit for the Great Moderation when things go bad they say well it was outside forces it wasn't our policy mistakes and I just don't think that approach is satisfactory in terms of accountability okay thank you my attorney suggested I go to law school I'm not thinking about it I'd like to address my question in terms of Scott Robert any of you and you go back a little bit in history and you think about the SNL bailout and how that impacted banks today and what you're saying there's no accountability there's no transparency the central bank does what it wants so what I would like to know what impact if any did the SNL bailout have on the current banking and financial system and if interest rates are increased and raised does that impact not only stocks and shareholders and the stock market is that a plus is or is that a negative people who have stocks they don't want dividends the interest rate goes up people start buying inflation comes down so it's kind of a mixed bag you've been talking about accountability is fine transparency is fine but who's going to make the decision and who's going to make the regulation that you all keep saying you won't change every couple of years because then there's no clearly defined guidelines or criteria I'm not an economist but I know how to manage money thank you yeah on the SNL bailout this is actually related to what I said about Congress being unable to discipline itself to not offer off budget subsidies it's just too compelling and so housing man how how the cost to the economy the social disaster of our housing policies has got to be evident to anybody who has any sense of history and has followed the SNL SNL debacle and then the more recent policies which started in 1993 to create increased home ownership which ended up with double digit increases in house prices from 1997 through 2006 it made it irresistible to create these new instruments to finance housing it's been a disaster and Congress just can't keep itself from offering off budget subsidies that look like a freebie and then eventually the crows come home to roost so somehow or other it's not going to happen but if we could get Congress out of the housing business that would be just extraordinary on your second point on low interest rates and there are going to be times when interest rates have to be low because you've got to encourage people to save less and consume more that's just basic unicyclical policy but if you have a financial safety net which encourages people to borrow short and lend long to finance portfolios of subprime stuff with commercial paper because they know the government is going to be on the other side and bail them out well you know people are going to follow the incentives you create so all I can say is if we understand the problem as economists maybe we can do something at least if it's not an election year I don't know Scott you can have the last brief word it seems like Congress has two conflicting goals they want a lot of cheap credit for people to buy homes and they don't want banking crises and the way things seem to be going now is that the only one way to achieve both these goals is to have the federal government take over most of the role of providing credit to homeowners and I'm no expert here but it seems like more and more of the credit is being funneled right through the GSEs and other federal FHA and so on and this is reflecting these goals that really are not compatible I'm afraid that's all the time we have for this panel thank you all very much if you could all give our panelists a round of applause