Scott Sumner on Fed Performance since the Great Recession

Scott Sumner is the director of Monetary Policy Program at the Mercatus Center at George Mason University and a blogger at The Money Illusion. In this episode taped in front of a live audience, Scott returns to Macro Musings to share his thoughts on the Federal Reserve’s performance from the Great Recession to the present. Scott explains how forecast targeting and price level targeting could have mitigated the economic decline in 2008 and 2009. He also discusses his thoughts on how the cognitive biases of central bankers can cause them to make mistakes in evaluating the stance of monetary policy and offers some solutions to address this problem.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Scott, welcome back to the show.

Scott Sumner: Thank you, David. Good to be back.

Beckworth: You're our first guest on the podcast, we now have 62 episodes out and a few more in the pipeline. We've come a long way, and we've got things rolling, and we're looking forward to this one as well. Let's go back and evaluate what the Fed has done, look back to the past. Let's go back to 2008. At that time, you argued that the Fed turned what would have been maybe an ordinary garden variety recession to the great recession by failing to act appropriately. There's a number of arguments you've made, and some of our listeners and audience may be familiar with them. I won't repeat them all, but if you could go back in time and do something different at the Fed, so that it didn't make the mistakes that you think that it made, what would you have done?

Correcting the Fed’s Past Mistakes

Sumner: Okay. You can think in terms of both strategy and tactics. In the area of strategy, I'm someone known for promoting nominal GDP targeting, but I think it probably will be unrealistic for them to suddenly leap to such a radically different regime right in the middle of a downturn. More likely, they could have done something a little more modest like, announce that since we are at the zero interest-rate boundary in late 2008, we're going to temporarily do level targeting of the price level where we have in mind a 2% increase over time in the price level and if we fall short or overshoot, we promise to come back to that. That would have made monetary policy more effective when interest rates were at zero, but it would have been a little bit more reassuring that's such a radical shift because you'd still have 2% as an inflation number.

Sumner: On the tactical and that's where I have more suggestions, I tend to favor and approach that Lars Svensson’s calls targeting the forecast, that is, you set your policy instruments at a level where you expect them to succeed, which sounds like it's not very controversial. Right? Central banks are often quite timid, so they might do a gesture, but knowing it's not really enough to prevent a steep downturn. I think in late 2008, in 2009, they weren't actually doing enough aggressive policy, first cutting rates and then QE, to where they were forecasting that they would be successful in terms of inflation, unemployment, and so on. Let me just give you one example. Right after Lehman failed in September of 2008, they had a meeting and they refused to cut interest rates below 2%.

Sumner: At the time, they were still worried about inflation, but the markets were signaling actually the real risk was that inflation was going to come in well under 2%. In addition to targeting the forecasts, I think the Fed should pay more attention to the market forecast. I think if they'd done so in 2008, they would have been more aggressive and then in addition to that, do whatever it takes approach, don't just do a gesture of QE, but do enough monetary stimulus in a deep slump, to where you actually expect inflation or spending growth to be on target. That aggressiveness, I think, would have made the recession much milder.

Beckworth: It's worth mentioning that, not just in September, but that whole period between April and October, the Fed was really concerned about inflation and they were signaling and they were concerned about it. If you look at Fed fund future targets, so you look at the one year ahead forecast, the market was making for the Fed funds rate in June, I believe that that year it was at 3.5%. They're currently at 6%. The market was actually forecasting an increase, because of the way the Fed was talking. It wasn't just as a 2%, it was also that they were signaling, they were just as likely to increase those rates, which made things worse. Well, let's move forward to the Fed did QE, and we're now reaching a point where they're talking about unwinding the balance sheet, so looking back at the Fed's QE programs, QE1 QE2, QE3, were they successful, were they a failure or somewhere in between?

Reflecting on the Success of Quantitative Easing

Sumner: Well, I think somewhere in between. Again, this is the distinction between tactics and strategy. QE is a tactic that can have some positive effects on and spurring increases in spending. If it's not embedded in the proper strategy, then by itself, it's not going to be enough and I think we saw that. The markets responded to QE, as if it was going to have somewhat of a positive effect, number one. Number two, when you compare the US to Europe, which did less during those early years, clearly, the recession was milder in the US starting around 2011, ‘12, ‘13. That's partly because the Fed did more than the ECB around that time. If it's not embedded in the proper strategy, it's not going to be enough. I think what people who follow monetary policy casually tend to do is, overrate the importance of tactics and underrate the importance of strategy, having the proper target and the proper willingness to do whatever it takes to hit that target.

Beckworth: Along those lines, the Fed has a 2% inflation target, as you mentioned, it became explicit in 2012, but lots of research shows that they were implicitly doing something similar to 2% for many years leading up to that. If you know that, that there was both an implicit and explicit 2% inflation target, number one, the number two, if we started the time of the recovery about June 2009, and go forward, the Fed's preferred measure inflation is averaged about 1.5%. It's persistently under shot 2%. Is this a big deal?

The Fed’s Inflation Target

Sumner: Well, yes and no. The fact that it's been a little bit below 2% is perhaps not that big a deal for the macro economy. We're still recovering, we're still seeing job growth and so on. The economy can adjust to any trend rate of inflation within reason. On the other hand, you could argue it's a problem for the Fed's credibility if they're not hitting their target. Right? Let me just point out that, there's an argument you often hear that I think is a little bit misleading and it's like, well, if the Fed isn't hitting its 2% target, why do you think the Fed should change its target? What good would that do if they're not even hitting the 2% target?

Sumner: One reason they're not hitting the 2% target is arguably that it's the wrong target. There's other targets that would be easier to hit. In fact, if we went back in time, 20 years ago, and the Fed had to decide what's the proper target, knowing what they know now, the Fed almost certainly would not have picked a 2% inflation target. I'll tell you why they wouldn't have. One of the justifications for that target was, it'll keep the United States above the zero interest-rate boundary. That was one of the justification. We adopted that policy so that We wouldn't be like Japan with zero interest rates. Well, we now know that at a 2% inflation target, you can end up with a prolonged period of zero interest rates. Clearly, if they went back in time knowing what they know now, they would have picked a target that avoids the zero interest-rate boundary more easily, makes it more easy, makes it easier to do effective monetary policy, and makes it less likely, as you said, they persistently undershoot that target, like they have the last five or six years.

Beckworth: It's been interesting to follow the discussions on this. Fed Chair Janet Yellen, at the last meeting here in June, she said it was a one-off event that they undershot. The thing is, there's been a number of one-off events because you, again, have eight, nine years of undershooting. At some point, you have to attribute that to some revealed preference, some bias. Right? What do you think?

Sumner: That's a good argument. It's probably a better argument that I'm willing to cut them some slack as making mistakes rather than intentionally deceiving us as to what their true target is. I would point out that, they're relying somewhat on this Phillips Curve model that I think is flawed. They look at the labor market as a signal of where inflation is going. I think that's really a flawed model. I think what we're seeing recently, and we saw this also in the late 1990s, is sometimes, the unemployment rate can fall to relatively low levels without triggering the sort of inflation you'd expect. I think the Fed actually did believe that inflation was going to be a little bit higher. I would point to the fact that a lot of private sector forecasters also did.

Sumner: However, the markets, again, going back to the market forecast, which I put more weight on, has correctly signaled that the Fed was likely to undershoot inflation so that, spread between conventional and index bonds has been generally somewhat lower than 2% in these years. I think the markets have been more aware of this problem than either the Fed or some of the private sector forecasters.

Beckworth: One more question on the topic of the Fed not hitting its inflation target. There's been some recent research done that indicates that if you're someone who lived through the great inflation of the 1970s, that really was a jarring experience, it scarred you, it shaped who you are, your views, and many of the Fed officials lived through that. This basically says, look, they're fighting the last war. They're concerned about a 1970s inflation as opposed to hitting their target. One, maybe, anecdotal point would be like Neel Kashkari, for the youngest member, I think is the youngest member of the board of the FOMC, he's the president of the Minneapolis Fed, he has been very vocal voting against these interest rate hikes because they're not hitting their target. He doesn't seem to be as fearful of overshooting. One other point I'd mentioned. Janet Yellen at one of her press conferences was very explicit, "We don't want to repeat the mistakes of 1970." Do you think there's any kind of a generational issue here as part of the story?

Sumner: Yeah. For years, I've been teaching a similar generational story about the great inflation. The argument I would present to my students was that, the top policymakers in the 1960s, came of age and were young in the 30s, so for them, unemployment was the big problem, the Great Depression, and their bias was towards pushing up the Phillips Curve towards higher inflation and lower unemployment. Some of that bias fed into the great inflation in the 60s and 70s. Then, of course, you had my generation that came of age and so I became an Inflation Hawk early on. Yes, I do see this bias and that's why I think it's important to develop a monetary regime, where biases don't have the sort of persistent effect on policy.

Sumner: One reason I like this notion of level targeting, is it imposes some discipline on monetarism. You have this target path and if you overshoot, or undershoot, you promise to come back to that target path. I think that discipline helps prevent sort of cognitive biases from distorting monetary policy. We're all human, we've all lived through certain events. We tend to, I think, misinterpret them based on our previous life experience.

Beckworth: What's important is having a rule in place, of course, the right rule? Have the right, rule number one, number two, have a rule that will get us through biases, generational issues, latest fads. Okay. One more thing I said the last one, one more thing, but one more thing, it stirred some thoughts. Larry Summers and others have talked about secular stagnation. Part of the story is... it's not all. The part of the story is this persistent shortfall of demand. They've seen the demand hasn't been as robust and this ties into the understanding of inflation. They argue that persistent understanding of demand has led to kind of an atrophy in the economy, so potential GDP is lower as a result, whereas, some of us might think supply side issues are important as well in that story.

Beckworth: The analogy I would use to explain their view is, at some point in your life you used to bench press 300 pounds. Right? Then you quit going to the gym, a couple years later, you go back and you only bench press 150. Right? That's because you haven't been doing your work. That's, I think, the Larry Summers story. There's another story as well. Maybe that's true, but maybe you're just getting old. You can't bench press 300. The supply side is hurting you as well. Where do you come down on that?

Sumner: I disagree a little. I think there's some truth in that the severity of the Great Recession, left some permanent scars, but I really think it's both a global story and mostly a supply side story. When people ask me to forecast growth or interest rates and so on, I'm not really a forecaster, but the thing I noticed is that, we've been seeing some of these trends now play out for several decades, and we're seeing them at a global level. That doesn't give me much reason to think that any of this is going to change. Let's take an example. One of the first countries to experience this slow growth and zero interest-rates was Japan. Right? People cite demographics, slow population growth, aging population and so on.

Sumner: Well, Japan looks like today, what much of the developed world will look like in the future, slower population growth, aging population. Another factor cited for the low interest rates, for instance, is high savings rates in Asia. Well, in the future, Asia will be a greater and greater share of the global economy. If you imagine a future, where say, China builds out its infrastructure and housing and becomes relatively developed, and then slows a lot in investment, but remains a high saving country, you could imagine the same forces being much more powerful in terms of putting downward pressure on interest rates.

Sumner: People talk about the productivity slowdown reflecting the move towards a service economy. Well, we're likely to become even more of a service economy in the future. I'm not really a forecaster, I wouldn't rule out a sudden surge in productivity due to developments that I cannot foresee, but for the time being, and then finally, most importantly, again, getting back to the market monetarists angle, markets like the 30-year treasury bond are seemingly predicting fairly low growth in nominal spending. We have about a 3% yield on 30-year treasury bonds. That means the markets think there's a new normal of lower interest rates, it'll cycle up and down with the business cycle, but around a lower trend probably than during the 80s or 90s.

Beckworth: Okay, I had a guest on, recently on the podcast, Matt Yglesias. He's a progressive, left of center guy, but he really follows Fed policy avidly. We had an interesting conversation and he was surprisingly critical of President Obama for not appointing more governors to the Board of Governors of the Fed, for not appointing people. He thinks President Obama undersaw the importance of the Federal Reserve and in creating a more robust recovery, and that, had he done things differently, had he made quicker appointments, appointed the right people, there would have been a more robust recovery. What are your thoughts on that?

The Federal Reserve under Obama

Sumner: Well on my blog, I did criticize President Obama in 2009 for leaving seats at the Fed empty. I felt it was an important issue. I believe he was getting some advice from some of his advisors to the effect that monetary policy was out of ammunition. I think that was incorrect. I think, by the way that Christine Romer ran into that problem when she first met President Obama. She was dismayed by the fact that he seemed to not take the Fed very seriously as a possible tool for recovery. In that sense, I agree.

Sumner: On the other hand, I do think that people underestimate the extent to which the Fed is this large inertial institution that is hard to change. If you go back in time and ask me, what would I have expected Ben Bernanke had to do as Fed Chairman, I would have looked at his academic writings on Japan, and expected him to do what I've been proposing in my blog over the last nine years. In fact, Ben Bernanke, went along with the consensus view at the Fed and did some things that were more aggressive than other central banks like the ECB, but not as much as what I would have liked. I think there's a limit to how much effect one or two appointees have on a very inertial institution like the Federal Reserve.

Beckworth: You mentioned a minute ago about the markets indicating, long-term real interest rates are going to be lower than they were previous. There's been a big debate about that. We call this the natural interest rate to rate driven by the fundamentals of the economy. It's also called R star, our little astrobotic. R star is the buzzword they use. There's been debates about this. If you do go back and look, you'd do see a downward trend, than is real rates for the reasons you mentioned, demographics and other factors.

Beckworth: John Taylor has made an argument that, in addition to this trend, there was a sudden drop in the natural rate. For example, there's this measure that Lubbock Williams, R star measures, it's John Williams, who's the president of the San Francisco Fed him and a fellow economists, they've estimated measures of the natural interest rate. One example here in 2007 Q4, fourth quarter, it was 2%, delivered 2%. A year later, it fell 2.3%, 4%. In a pretty short time it fell almost two percentage points. John Taylor's argument is, that's not just part of the trend. There's something else going on there and it could pop back up. Any credence to that argument?

The Natural Rate of Interest

Sumner: Well, I think there's two things. Certainly, that's partly true. The severity of the Great Recession, and changes in the lending practices of banks in response to the problems they had earlier, probably depressed the central, sorry, the natural rate of interest quite a bit in a very short period of time. You can view that as an artificial or temporary decline. It also depends how you define the natural rate. I think of it as the interest rate, where if the market rate was at that level, we would be on target in terms of our macro goals, whether it be inflation or whatever. In that sense, it's partly a great recession story, but I still go back to the fact that, we've seen a downward trend for about 30 years.

Sumner: If it were just the Great Recession, there'd be really no reason for interest rates to be so low, in say, the 30-year Treasury bond market. That bond market is looking out for a long period of time, we're already down to less than 4.5% unemployment. There's no reason why looking out over the next 30 years should be that much affected by what happened in 2008 and 2009 at this point. Therefore, I'm inclined to think that these factors I talked about earlier really do represent a kind of a new normal. Now, that doesn't mean that it's inconceivable that we could not get a lot more growth and a higher natural rate of interest. I think a lot of people tend to underestimate what it would take. Many free market economists and I include myself in that camp, talk about how deregulation could help growth. Right. I think there's some truth to that or tax reform, et cetera.

Sumner: However, to make a big impact, I think you have to do some really dramatic things, like maybe completely deregulate real estate development, so you could build more easily all across the country. That would actually put upward pressure on the natural rate of interest. That's a very hard sell, given that a lot of that's at the local level, or a truly radical tax reform, going to a simple consumption tax that might spur some investment. Those are not the kind of reforms that I think are that politically feasible in the current environment in Washington. My best guess is that, growth is likely to stay relatively slow. It may pick up a little bit, but I don't think we should expect too much in that area in the near future.

Beckworth: Unless you're a techno optimist, you believe at some point, technology is going to push up productivity growth at a rapid rate, which hasn't happened yet, but I want to believe.

Sumner: Let me just say one point about technology. I'm not sure if this is right, but it seems to me that, a lot of modern high tech industries require less capital. Compare something like Facebook to General Motors. When I was young GM would spend a lot of money building factories and so on and that would absorb a lot of the world savings, those kind of investment projects. We know some of these highly successful companies now are able to get going with a relatively small amount of capital and labor. I wonder whether the new economy in some sense is affecting this natural rate as well.

Beckworth: Well, I will refer our listeners to the recent podcast we did with Mike Mandel and Bret Swenson. They responded, they would argue that, IT technology has been dramatic in raising productivity in certain sectors, finance, IT sector itself, but they argue 70% of the economy is the physical part and that hasn't received the gains from productivity from technology and they're hopeful anyhow. It just comes down to a forecast. We don't know. Let me bring up a question I know some of our listeners are going to have, you're talking about these low rates and I think it's great that you point to the market, itself, forecasting low rates 30 years out. Many people will say, "It's all the Fed’s fault. The Fed is the reason rates have been low." How do you respond to that?

Are Low Rates the Fed’s Fault?

Sumner: Well, I recall, you did a blog post a couple years back showing that, although the Fed has bought a lot of treasury bonds, it still owns a relatively small percentage of the total. In addition, since we had such large deficits over the last decade or so, the total amount of treasury bonds outstanding that have been purchased outside the Fed has actually gone up a lot. If it was just the Fed artificially depressing the yield on treasury bonds, you'd expect the Fed to own a much, much larger share of the total amount of treasury bonds outstanding, but that just has not occurred. In addition, I think that people overestimate the extent to which the Fed affects interest rates and I believe Jeffrey is going to talk about that later.

Sumner: Basically, if you think about the Fed artificially depressing interest rates, what do people have in mind? They have in mind an expansionary policy, monetary policy. Right? Where you have an open market purchases, you buy bonds, you depress interest rates. That expansionary monetary policy should also result in higher inflation, and in the long run, interest rates would actually go up during that higher inflation. If we're looking at really long-term interest rates, it's likely not reflecting Fed policy. You might argue, they're artificially depressing short-term interest rates, which is what they focus on, but the long term interest rates really do reflect market forces.

Beckworth: Okay. Well, let's move along to another heated topic that's now on the horizon and that is the Fed's balance sheet. Beginning this year, and more recently, there's been a lot of talk about the Fed shrinking its balance sheet, back at the June FOMC Meeting. They actually provided detailed plans to how they would do that. Is this something that should be done? Should we see that that reduces balance sheet? What implications does that have for interest on reserves?

Shrinking the Fed’s Balance Sheet and IOER

Sumner: Well, I do favor reducing the balance sheet, although, i have to say, I don't think it's a really big issue, to be honest. Raising the balance sheet was the last thing they did after they cut interest rates. Logically, it should have been the first thing they did, is they tighten monetary policy. Interest on reserves is something that I think was a big mistake when it was instituted in 2008, not because the policy is something that can't work or doesn't have some good justification, but rather, it's actually a contractionary policy, relative to not paying in just some reserves, and the policy was introduced in October of 2008, which was the worst possible time to introduce a contractionary monetary policy.

Sumner: Now, the Fed's argument is, well, the interest on reserves was a pretty low interest rate. That's really a weak argument and even Ben Bernanke, as an academic, has pointed out that interest rates are very, very misleading as an indicator of monetary policy. Even small increases in interest rates can represent highly contractionary monetary policies. I'll give you two quick examples. There was just a very small increase in 1937, in America, quarter point or so. It was enough to drive us into a double dip depression in the 30s. The European Central Bank just did a half a point increase in 2011. It was enough to push Eurozone into a double dip recession.

Sumner: People don't understand this because they see the small change in interest rates and think, "Well, it can't have that much effect." It's not a linear system and you can have a small change in interest rates, it accumulates, it changes the macro economy in a way that its impact over time, is much greater than anticipated. For that reason, I'm opposed to the policy of interest on reserves. If it is done, it has to be done in a more sensible way than what the Fed did in 2008. My preference, actually, is to go back to the old system of smaller balance sheets and no interest on reserves.

Beckworth: Okay. Now, one thing I've noticed at FOMC meetings, press conferences that follow, is there seems to be at least to me more noise to signal because there's so many levers now. We've got what's happening to the balance sheet, what's happening to some reserves, what's happening to the RRP, their reverse repo program they have, what's happening for the guidance, a lot of different things to look at. If you're looking at through a meeting, you'll often get the summary of economic projections. There's a lot of noise to sort through to get to the actual signal. Tim Duy, he's written about this before, and I can see the advantage in going back to the simpler approach if it reduces some of that noise. Put aside your preference, do you think it's actually going to happen?

Sumner: Reducing the balance sheet?

Beckworth: Yeah. Are they going to get rid of into some reserves, go back, because many other central banks already do something like interest on reserves.

Sumner: Right.

Beckworth: Is this the next phase of central banking?

Sumner: I think it is. Right now, I think I'm losing this battle, so to speak. Let me explain it this way. Yeah. They're going to probably reduce the balance sheet somewhat, but not all the way back to where it was before 2008. They'll maintain interest on reserves and I think that can work fine in environment where interest rates, market rates never fall to zero. You're right. Other central banks have done this. My objection is hard to explain. My critique of modern macroeconomics is, there's too much focus on interest rates as both the tool and indicator of monetary policy. It's used as a tool by the Fed and it's used as an indicator of the stance of monetary policy.

Sumner: Even though macro economists know that it can be misleading for various reasons, I still think they get lulled into making mistakes like in 2008, not understanding that money was too tight because they were focusing on the low and falling level of interest rates. If we do interest on reserves, it moves us even further away from a monetarist way of thinking about monetary policy, towards an interest rate-centered way of visualizing monetary policy. If I'm right that there are these cognitive illusions where we look at interest rates and misjudge the stance of monetary policy, I see that becoming more likely to occur as we make interest rates more and more the only way in which we do monetary policy.

Beckworth: Well, on a more promising note for you, Larry Summers recently came out in favor of nominal GDP level targeting. Is he part of a trend you see more interest in nominal GDP level targeting?

Increasing Support for Nominal GDP Targeting

Sumner: Yes, I think we've seen a number of pretty well known economists, either endorse the idea or suggested has some merit. I think there's reasons for that. I know the blogger Nick Rowe mentioned that, what moved him from inflation targeting to nominal GDP targeting, was he looked at the situation in the Great Recession, and to him, it looked like nominal GDP was just giving a much clearer signal of where monetary policy was, relative to where it needed to be. Ken, he's a Canadian blogger and Canada was doing inflation targeting and he felt that inflation wasn't giving as clear a signal, as to what was wrong in monetary policy. Especially in 2008, for much of the year, inflation was actually above target, but nominal GDP growth was below normal in 2008.

Sumner: I have this critique of inflation, I call it, never reason from a price change. A price rise can be due to either demand or supply side factors. Unless you know why the price has changed, you don't know what it's telling you about the economy. If you're doing inflation targeting, you're implicitly assuming that any inflation signal is a demand shock signal, it's due to changes in aggregate demand, and you're acting on that basis. Now, of course, the Fed knows this, and they do some adjustments for supply shocks and so on. I do think that signal is less clear and the Great Recession to me was almost a textbook example of where nominal GDP gave a clear signal about the real nature of the economy, the situation the economy was in, then did inflation.

Beckworth: Yeah. You mentioned the example of the ECB in 2011. They raised interest rates twice, because inflation was going up. It was going up because of commodity price shock, supply side shock.

Sumner: That's right.

Beckworth: Clearly the Eurozone was going down, the demand was falling, it wasn't demand shock.

Sumner: I would add, there were two things in Europe, I believe. One was their oil prices in 2010 and 11 went up because of China importing more and so on. In addition, there was this push to raise value added taxes, this push for austerity, because they did have some deficit problems, especially in southern Europe and so value added taxes were raised. Well, that's inflationary, but it's also a contractionary fiscal policy. That fits into my never reason from a price change and you wouldn't really want to interpret inflation due to a tax increase, the same way as inflation due to a booming consumer spending occurring. Right?.

Beckworth: Yeah.

Sumner: Again, economists know this, but I think there's a mistake, a tendency to overlook the nuances that are necessary in interpreting inflation just like with interest rates, and that's why I prefer nominal GDP.

Beckworth: Last question along these lines. There's been a recent push for a higher inflation target. Even though folks like Larry Summers and a few regional Fed presidents have endorsed nominal GDP targeting, there's been some push for higher inflation target, recent letter by some progressive economist, latest example. What is your argument against a higher inflation target?

Arguments Against a Higher Inflation Target

Sumner: Well, first of all, I do understand the logic behind it. I think you can make a pretty good argument that it would be better than the system we've had in place for the last 10 years. As I said, if they knew about the zero-bound problem 20 years ago, they would have picked a higher number. The argument in favor of it is that, it avoids the interest rates getting stuck at zero when monetary policy is harder to operate using conventional techniques. The argument against it and the reason why that's not my first choice is that, I think there are better alternatives and, again, I think that nominal GDP targeting combined with this level targeting concept. It's a little hard to explain this in words, but the upshot of level targeting is like putting sand under tires on an icy road. Level targeting gives monetary policy more traction or more ability to hit its target when your interest rates are zero. For that reason, I don't think we actually need to have a higher inflation target.

Sumner: On the other hand, I do understand the argument in favor. In retrospect, if we'd had a 3% inflation target over the last 20 years, the Great Recession probably would have been milder. It's not that far-fetched an argument, it's a question of, are there better alternatives?

Q&A Section

Beckworth: Okay. We'll now do Q&A. There's a microphone up front here, so if you have a question for Scott or for me, please come up now and we're going to roam around. Okay. We'll have one mic roam around.

Audience:This is for both of you. You blame Obama for the vacancies on the board of governors, but remember the vacancies first occurred during Bush when the democrats were blocking republican nominations. I had the impression that the republicans were doing the same under Obama and that affected Obama's policy, so that we have a situation in which, not only our supreme court justices is their choice, the choice of supreme court justices is politicized, but now the choice of members of the Fed Board of Governors has become politicized.

Sumner: In my case, I started blogging right at the beginning of the Obama administration, two weeks after he took office. That's something I noticed and there's some truth to what you said, but during the first year, the Obama administration, the Democrats had very strong control of the House and Senate, 60 senators. He could have gotten anyone through, but for about 12 months, I believe, he didn't even nominate anyone for those two empty seats. Later in 2010, when he finally got around to nominate some people, one of them, I think it was Peter Diamond, was held up by the Republicans in the Senate.

Sumner: By that time, there's been a special election, and the democrats had lost their 60 vote majority in the Senate, which weakened their position. You're right. A lot of this is gridlock and both parties involved in this gridlock. Even during that window of those first 12 months, when it's important to get off to a good start, there was nothing really preventing it. It really a lack of interest in the topic within the Obama administration, other than Christine Romer, who was pushing Obama to move more aggressively on that issue.

Beckworth: I would mention, there's a famous clip that President Obama said, "The Fed has shot its ward." That was his perspective. Why even try? He didn't nominate both for the gridlock reasons, but also he didn't see any point. Why spend political capital on something that may not make any difference?

Audience:Yes. I enjoyed your comments. You tended to focus mainly on what the role of monetary policy with respect to what contributed to the recession and now coming out of the severe recession that we've had, my question relates to the housing price boom and bust, which occurred prior to the severe recession. Indeed, I would argue that there's substantial evidence suggesting that the severe recession was largely due to the housing price boom and bust, which led to a broader financial crisis, including the banking crisis, and that led to the recession that began in December 2007 and then ended in the summer 2009. You tended to focus on, if I got you correct, in the middle of 2008 and late 2008, in monetary policy.

Audience: I'd ask you to back up and say something about the role of monetary policy with respect to the housing price boom and bust. Did it play a role, a major role? What about the role of monetary policy with respect to dealing with asset price bubbles including, say, the housing price bubble?

Sumner: Okay. In my view, the role of monetary policy in the housing bubble or other financial bubbles is overrated. I'm not saying there was no role. A different monetary policy might have made the housing boom a little bit smaller. Remember, the Fed has to hit certain macroeconomic targets and there's only really so much they can do. Housing is only 3% to 6% of GDP depending on where in the business cycle. If the Fed had done enough to sort of kill the housing boom or make it much more moderate, then probably, we would have pushed the US into a pretty deep recession earlier on.

Sumner: If you think about the fact that the Fed has to look at stabilizing the whole economy, there's only so much they can do to control any single sector. That doesn't mean nothing should have been done, you could point to regulatory areas where the government was either not restraining, or maybe encouraging the housing boom, depending on your perspective. We may have made a lot of mistakes in how we handled both housing and banking, but in my view, those mistakes were more likely to be in the regulatory arena rather than monetary policy. I would also add that, when I said monetary policy mistakes turned what would have been a mild recession into a great recession, basically from January 2006 at the peak of the boom housing to April of 2008, housing construction fell in half in the United States, from 2 million to 1 million annual rate.

Sumner: The unemployment rate barely budged. It went from 4.7 to 5.0. Why is that? Because it was a small percentage GDP and workers losing jobs there were getting jobs in other growing areas the economy. Then after April 2008, we had not just a decline in housing, but a broader decline in the overall economy, exports, manufacturing, commercial construction everywhere. Now, unemployment doubled from five to 10%. That's where monetary policy plays a bigger role in affecting spending all across the economy rather than one sector. A healthy market economy should be able to absorb a severe downturn in a sector as big as housing, with only a modest hit to the economy, not something like the Great Recession.

Sumner: I would argue that period from January 2006 to April 2008, is the sluggishness, or slowdown that you'd expect as we move resources out of commercial, sorry, out of residential housing into other sectors, reallocating resources, rather than having the whole economy tank. That's where I think monetary policy had the failure of not propping up total spending in the economy as a whole, but I don't think it could do much about any single sector. I'll give you an anecdote from the Great Depression. Governor Strong who headed the Fed effectively in the 1920s, was pressured to do something about the stock market bubble. He said something to the effect of, "If I spank one of my children, do I have to spank them all?" Meaning if I try to punish Wall Street with tighter money, that's going to punish Main Street as well, all sectors of the economy. He was reluctant to do anything. He died in the middle of 1928. His successors at the Fed were more aggressive in trying to stop the stock market bubble with tighter and tighter monetary policy.

Sumner: Eventually in October, 1929, they succeeded, but it did end up taking the whole economy. Monetary policy is not a scalpel that can just deal with one issue like a stock bubble or a housing bubble, it only deals with those by affecting spending all across the economy and It's a very blunt instrument, which is why I favor a more intelligent regulatory approach, to deal with either housing or banking.

Beckworth: Let me just add to that. One counterfactual that sheds lights on Scott's argument is Australia. It had a massive housing boom, massive run up in household debt and it too got hammered by the Great Recession, but nothing happened there. They didn't really have the Great Recession. They were exposed to the global financial shocks. They had nothing like the US. On top of that, they were also hammered by commodity price shocks that we didn't have. Australia provides a counter example of what it could have been. They never hit the zero lower bound. In fact, if you look at nominal GDP or nominal demand, it was kept on a stable path, Scott's ideal there.

Beckworth: I think, Scott and I agree with your point is that, you don't see the big panic until really like mid-2008. That's when the Fed fell asleep at the wheel. I will say, though, I do think though, the fact that the US is highly leveraged, I think that doesn't matter. It makes you more susceptible to when the Fed makes mistakes. The economy, all this debt that the US was primed for a Fed error. That's the critique we often, I've often gotten. Well, yeah, the Fed set on interest rates at 2%. From April to October, that's not a big deal, what's 2%? But it's a big deal when you're highly leveraged and all these other things are going on. I do think it amplified maybe the process. I do think we could have been more like Australia and less like we actually had.

Sumner: I would just add that, in many ways, the US has a political regime and I think it has throughout history, that's very pro debt. We have all sorts of public policies that either explicitly or implicitly encourage the creation of more debt, everything from backstops to banking, like FDIC and too big to fail. We have tax deductibility of interest. We have regulators encouraging more lending during the housing boom. Everywhere you look, public policy has always been oriented as being sort of pro debt. We do have an issue of our financial system occasionally going to extremes and producing too much debt, but, again, I don't think that's something that can be dealt with through monetary policy. It's to blunt an instrument.

Audience: This was that and David's mentioning of Australia as a good segue to my question. I want to think about, the Fed as a monetary policy innovator has a bit of an uneven history. There been times when the Fed ran through different innovations and I think the 1950s, the monetarist experiment and in late 1979, our examples. A lot of the discussions that we're having today about NGDP targeting and changing the inflation rate, sound a lot like debates we're having in the mid-1990s, in the United States, where timidity and caution, were the watchwords and we're looking to see how these things were happening elsewhere. Same could be said for negative interest rates today.

Audience: If that's true, who is the New Zealand of NGDP targeting in 2017 and 18? Would it have to be someone like Australia or New Zealand, where low inflation is a real problem because the economies are relatively developed and yet they're not dominant economies where they're going swashbuckling into under the rise and could have real global impact? Did you see something like, for the Central Bank of Kenya, which is emerging economy, a lot of infrastructural problems, institutional problems, but at the same time competently-led and somewhat experimental? Or is this simply something that's only going to be able to function from an ECB or a Fed?

Audience: In other words, what advice would you give to Kenya or maybe to urge some of the Southeast Asian economies off of $1 peg to try this experimentation? Would you say no, this is better something that's going to have to be done at the really big economies and central banks?

Sumner: Yeah, good question. I believe that nominal GDP targeting may not be appropriate for a smaller commodity exporter, because swings in commodity prices can really distort nominal GDP. They might want to look at something like total labor income in their economy as a target. Paradoxically, what makes it tough is that, it probably works best in large, diversified economies and those are the least likely to experiment. You're right. New Zealand was the first one with inflation targeting. My vision is more that this change will take place incrementally. On several fronts, they'll be more interested and hopefully in academia, and more research on it. Also, it may develop through smaller steps along the way. I've been advocating accountability framework where the Fed would look back, say at each meeting, or every so often, to its previous decisions made one and two years earlier, and it would report to Congress as to whether in retrospect, its previous policy stances were too expansionary, too contractionary or about right. It would also provide some numerical metric to justify that.

Sumner: Now, they have this dual mandate of inflation and unemployment, but one good way of consolidating that into a single number is nominal GDP, so they could tell Congress, "Well, we set policy at such and such a year ago, but nominal GDP growth has been too high, say 7% and so it was too high to hit our inflation target. In retrospect, policy was to expansionary." Or vice versa. If they become comfortable with using nominal GDP as a single measure of total aggregate demand in the economy, and as a measure of whether policy was to expansionary or contractionary to hit their dual mandate of inflation and unemployment, it starts to develop a comfort level with that variable as being a sensible, single variable to encapsulate what they're trying to do. I see it gradually evolving, where nominal GDP takes a greater and greater role in the discussion, and the appraisal of the stance of monetary policy, and works its way in that way, not just that they'll suddenly jump to it all at once.

Beckworth: Right. One more question.

Audience: Hi, just a question, I guess, getting to Scott's issue of the difficulty in measuring monetary policy stands by interest rate. Do you think there's any contemporary confusion about interest rates and our star because I've noticed a little bit about people who seem to, even sometimes Janet Yellen, even sometimes in the Lubbock Williams literature, who seems to confuse our star, which if you think of it in terms of the tail role, is on the right side of the equation with the I on the left side of the equation that, our star, if we're going to look into an equilibrium rate, the Fed should've actually set, it should include things like deviation of inflation from trend, of course inflation adjusted, but also the output gap. You see a lot of people say, "Well, our star's 2% therefore, inflation adjusted, interest rates should be 2%." As opposed to also including the output guard on that, which would be more of also a step towards closer to GDP including both output inflation and the Fed stance of monetary policy?

Beckworth: I think it's a great question. I've been critical of the Fed, [inaudible] was talking about this last night. I wish that they would put out its own estimates of our star. Janet Yellen will often talk about the neutral rate or the natural rate, your laugh and you want to know, what is it? She keeps talking about, "Relative to the neutral rate, we're here or there." I think they'd do a public service if they'd actually released their estimates. This is undoubtedly something hard to measure. There should be a range of estimates. Going to your question more directly, the Lubbock Williams estimate as a medium term estimate of the natural rate. There's a lot of confusion in the press. Absolutely, it's another issue. That's, over the next four or five years on average, where do we think the neutral rate will be? What Janet Yellen is talking about is a short run. Often she talks about short run, and she's actually released a chart, one of her talks where she shows that. There's this confusion between a short run, natural rate, our star and then medium run one, which goes into a Taylor rule.

Beckworth: I think there's a lot of confusion over this. It goes back to the point that I made earlier that there's so many moving part in the Fed policy today. I think there's a lot of noise, the signal here and it makes it hard for us to understand what's happening. I'm not sure they always completely understand what's happening and that's why I think the push for a rule will minimize the reliance in all these unobserved latent variables.

Sumner: I have one quick comment. Someone needs to ask Ben Bernanke key this question. In 2003, Bernanke said, "Interest rates in the money supply are not good indicators of the stance of monetary policy. For that you need to look at inflation and nominal GDP growth." In 2009, and 10, this Fed Chairman, he said, "Monetary policy was extraordinarily accommodative." If that was not based on interest rates, and money supply, what was it based on because inflation and nominal GDP were both falling sharply, which would indicate contractionary. There's a disconnect between his views as an academic and how he described monetary policy as head of the Federal Reserve. I've never seen anyone asked him that, but I'd be curious as to how he would explain that.