John C. Williams, president and chief executive officer of the San Francisco Federal Reserve, joins Richard H. Clarida, professor of economics at Columbia University, to discuss evolving challenges and approaches to monetary policy, and their implications for fiscal policies aimed at enhancing long-run growth. Williams assesses the condition of U.S. employment and inflation. He goes on to evaluate the influence global economic conditions have had on the Federal Reserve's decision-making.
The C. Peter McColough Series on International Economics brings the world's foremost economic policymakers and scholars to address members on current topics in international economics and U.S. monetary policy. This meeting series is presented by the Maurice R. Greenberg Center for Geoeconomic Studies.
CLARIDA: Well, good morning, everyone. I’m Richard Clarida. And I’ll be moderating this session with John Williams. This meeting is part of the C. Peter McColough Series on International Economics. And so we’ll be pursuing the following format. We’ll begin with about 25 minutes or so of a Q&A session up here. And then the remaining 30 minutes will be question and answer time for the audience.
Well, when I got the request a couple of weeks ago to moderate this session I accepted immediately. I, myself, have been looking forward to it. Just a brief introduction of our speaker, and then we’ll go right into the Q&A. John Williams is the president of the San Francisco Federal Reserve District Bank, and has been president since 2011. Before that, he was the director of research at the San Francisco Fed and really, under his leadership, turned into one of the very top research groups in the Fed system. He began his career years ago at the Federal Reserve Board and was part of the group there bringing the Fed’s modeling efforts into the 21st century. And many of us now actually use the Fed models because they’re now made available to the public. And John had a big hand in developing that, has a Ph.D. in economics from Stanford and studied also at LSE and UC Berkeley.
So let’s get right into the opening set of questions before we turn it over to the audience. So, John, what is your outlook for the U.S. economy—unemployment, inflation, and growth? And if that materializes, how many rate hikes do you think you would be recommending?
WILLIAMS: So I think 2016’s going to look a lot like 2015 in many ways. I expect GDP growth to be about 2 percent—real GDP growth this year, like we had last year. I think it’s basically the same set of issues, strong domestic growth especially in the service sector, some subtraction from growth because of the strong dollar and weakness abroad. So basically this balancing of the domestic economy kind of doing well, but being—facing some headwinds from abroad.
I expect to see pretty solid growth this year, like we saw last year, over 2 million jobs this year and I expect the unemployment rate to continue to edge down from 5 percent today, down a few more tenths. My own view is that we’re basically at full employment or very near it based on 5 percent employment rates. And I feel that’s very good news. And with a strong economy and with the dollar and oil prices and other influences basically turning more neutral in terms of inflation, I view inflation as moving back gradually over the next year or two to our two percent goal.
Like you said, I started this job in March of 2011. It’s been a long slog, but actually I think, you know, on our full-employment mandate we’re basically there. And on our inflation mandate, I do think that we not only made good progress but I’m pretty optimistic that we’ll get to our goal in the next year or two.
CLARIDA: And then if that—if that outlook materializes, what rate hikes would you be recommending?
WILLIAMS: Oh, I didn’t get away with skipping the rate question?
WILLIAMS: Oh, OK. So, you know, I think that our basic strategy is the right one. And that’s, you know, slowly or gradually removing accommodation, meaning raising interest rates over the next few years, and getting our interest rates back to normal. Now, the exact timing of when we have the next rate increase will depend on the data, depend on how we’re doing relative to our goals. But when I look back at the March economic projections that my colleagues and I put out suggesting, you know, over the rest of the year, say, two or maybe three rate increases this year, maybe one or two more next year, so three or four next year, I think that’s still about right. It will depend on the data. We’re still getting another month’s of data before the next—for the June meeting. I want to analyze that and come to a conclusion. Obviously we’ll be meeting numerous times a year, get to keep doing that, and think about it.
CLARIDA: Thank you. One of the bright spots in the economy recently has been the pickup in wages. Not only are wages rising, but labor share of national income has been rising. So how does the Fed factor that into its reaction function for the economy?
WILLIAMS: Sure. And, you know, the wages has actually been the puzzling part of the story. You know, wage growth has not picked—we kind of would expect with unemployment coming down to about 5 percent, wage growth would have started picking up earlier. I think a number of factors have held wages down. One is the very slow productivity growth and some other factors. But we are seeing some signs—encouraging signs that wages are starting to pick up. I think, you know, wages are the biggest cost for businesses. So that’s going to put somewhat upward pressure on inflation.
I think we have to be realistic, kind of what’s the normal for wage growth. My own view is productivity growth probably will average about 1 ½ percent over the next several years. So, you know, we might be seeing wage growth of something like 3, 3 ½ percent of normal. So we’re moving towards that. And I think that’s a positive sign. It shows, I think, a couple things: One is, you know, that we are close to or at full employment, with wage growth picking up that’s kind of a sign of where the labor market is. Obviously, it also is a sign that, you know, we’re seeing some uplift to inflation, which is a good thing.
CLARIDA: It is. Now, speaking of inflation, since the Fed announced in January of 2012 that it was pursuing a formal inflation target, inflation has undershot that target for four years and will likely to undershoot this year. So how would the Fed react to a modest overshoot of inflation relative to target. Core CPI, for example, is already above 2 (percent). Would it welcome that?
WILLIAMS: Yeah, this is probably—when you say, how will the Fed react—is probably the time when I should put in my normal disclaimer that my remarks reflect my own views and not necessarily anyone else. OK.
WILLIAMS: So you know, I think that—you know, you’re right. This has been an undershoot for several years. There is a—there were some signs that inflation expectations have come down a bit. And that’s somewhat worrying to me, because, you know, having well-anchored inflation expectations, people having confidence that the Fed will achieve its goals is very important. So my view—my own forecast, my own view of appropriate policy is we’ll see inflation move back to 2 percent, not overshoot by much, by definitely get to 2 percent. If it overshot 2 (percent) by a few tenths for a while, that wouldn’t be a big problem I think, you know, because, you know, we want inflation to be an average, 2 percent, sometimes a little bit above sometimes a little bit below.
I wouldn’t want us to purposefully overshoot inflation—the 2 percent goal by a lot and for an extended period, and here’s why. If we overshoot inflation, inflation moved to 2 ½ or 3 percent, say, well, then we would have to bring inflation back down and we would have to tighten policy and slow the economy a lot. So I think that there are some risks. There are always risks to overshooting your target and risks to, you know, having to undo that. So, you know, my, you know, kind of perfect landing for the economy would really have inflation moving, you know, basically back to 2 percent. And then from that point, you know, it sometimes would be a little above, sometimes a little bit below, but averaging 2 (percent).
CLARIDA: Mmm hmm. Now, on that point, we were chatting before onstage that you attended a conference honoring Mike Woodford, one of the, you know, great thinkers of monetary economics. And of course, Mike’s theoretical work has pointed to the benefits of a version of inflation targeting, price level targeting, which case if the Fed were price level targeting it would be aiming to overshoot now.
CLARIDA: The Fed’s thinking on price level targeting?
WILLIAMS: So, this is, I think, a pretty important issue not just for the U.S. by for other countries. I mean, the U.S. and most—as you know—most advanced economies, and many emerging market economies and central banks have decided to go some kind of version of inflation targeting, where you basically say you want inflation to be 2 percent and if you’re above you try to bring it down to 2 (percent). If you’re below, try to move it back to 2 (percent). And that’s worked amazingly well across dozens of countries over the last several decades. So it’s been a big success. Price level targeting basically says, you know, I want the price level to grow, say at 2 percent, but any undershoots or overshoots are made up later. So you have this built-in memory that says if you had low inflation for a few years, you’re going to have high inflation for the next few years to compensation for that.
Now, that’s not the strategy we’re doing. We’re doing inflation targeting. I think it’s really important for the Fed, where we are, close to our goals, is to stay—not change, you know, horses in the middle of the river, get our 2 percent goal, reach that and kind of focus on that. I do think that based on the work that Mike Woodford and many others have done, and given the challenges that we’ve seen central banks facing around the globe with achieving 2 percent—I mean, basically very few advanced economies today have reached their goals; many have been below for several years—that economists and policymakers around the world, we think the strategy of inflation targeting, and think hard about whether they should—we should be doing some version of price level targeting. I think there is advantages in terms of situations like the one we’ve been in the last few years. But I think that that’s kind of a research project that should be done for several years. And right now, in my view, is we should finish that job that we set out. But I do think it’s an important issue for economists to be considering.
CLARIDA: OK. I think we’ve all noticed, those of us who follow the Fed, that this Fed has been emphasizing global developments as much or more than other Feds. And personally, I applaud the Fed for that. But does it complicate communications, since the Fed’s mandate is U.S. inflation and unemployment. Does it complicate communication to be discussing global developments?
WILLIAMS: So as I was telling Rich this morning, the answer is, yes, it does complicate our communications. And for the very reason you said. You know, our goals are domestic inflation, domestic employment. That’s what—everything that we talk about at FOMC meetings when we think about the economy and we do our analysis is focused through the lens of, well, what does that mean for employment in the U.S., what’s it mean for U.S. inflation, what’s it mean for the outlook. But in the current situation in the last couple years, several years in fact, global developments have been a big driver of what’s happened with U.S. inflation and with U.S. growth.
I mentioned already the strong dollar and the decline in oil prices. Well, those have had, you know, enormous impacts on the U.S. economy and have been important drivers of U.S. outlook and also risks to that outlook. So at the Fed and the FOMC we’ve been kind of facing this challenge of how to communicate that. We highlight the fact that our goals are domestic, but these global developments are really important to understand the outlook for the U.S. economy and the risks to it. So I have been a supporter of putting in more language in our statements and our descriptions of our policy decisions that reflect global developments, but it has led to some of these kind of questions that you brought up, is are we trying to be the central banker for the world, are we taking into consideration what’s happening abroad in terms of our decisions.
So it is a nuance issue, but I’m a big believer in transparency. If our decisions to, you know, hold off on raising rates in the past were influenced because of global developments and how they impact the U.S. economy, I just personally think we should explain that as best we can just keep working on being better at explaining our logic.
CLARIDA: And just parenthetically, one thing I think has been useful with the Fed providing now—essentially the code to run its model that John and others developed, you know, many of us have been able to run simulations on the Fed model for the effect of things like a 10 percent move in the dollar, or whatever. So you can actually see essentially the same analysis that you’re looking at as well, which is that it does have an impact on both inflation and employment.
There are those who claim that there is a risk to the Fed in reacting to the dollar, for example, or global developments in markets, perhaps overreacting to financial asset prices, which is a general issue in monetary policy. What do you think about that?
WILLIAMS: Right. We’ll you’ve written about this yourself. (Laughs.) This is an old issue about how do the—you know, the Fed obviously or any central bank, but the Fed clearly has a huge impact on financial market conditions, not only in the U.S. but globally. Obviously, financial market conditions then impact the U.S. economy. And you know, sometimes when you’re a policymaker at the Fed you kind of feel like you’re living in a hall of mirrors where, you know, you can’t—you know, how do you—you know, if we do this then that will happen, then that would mean we shouldn’t do this.
So this is why we have economists. This is why we have models, is to try to think about what are the—you know, kind of the endogenous or reactions in the economy to changing economic conditions and the outlook versus some kind of shock or some outside event that then feeds back, and do our best to analyze, you know, what are the financial conditions, how they feed that in—how does that feed into the U.S. outlook, and then what’s the appropriate policy. You know, that’s just part of the world that we have to operate in. And you know, I think it’s really important not to overreact to changes in financial conditions.
I always tell people that I don’t have a Bloomberg terminal. I don’t watch Kramer on CNBC. (Laughter.) I seriously do not—you know, as a policymaker, you want to—you know, monetary policy as Milton Friedman taught us and has been studied for decades, monetary policy takes a year or two to have its full effect on the U.S. economy. We’ve always got to be looking down the road and thinking about what’s the right decision. And I really do try to avoid kind of getting caught up in the day-to-day fluctuations in markets. At the same time, you know, being aware of what’s happening in the world. So it is a challenge, but we try to do the best we can.
CLARIDA: Yeah. In addition to your full-time job as the president of a district bank, you’re also continuing to do research, and in particular research with Thomas Laubach at the board has been quite influential, certainly on me and a lot of others for thinking about this idea of the time variation and the equilibrium or neutral real interest rate. So perhaps you could tell us the current state of that work and, in particular, as I understand it, you’re extending this to some other major economies as well.
WILLIAMS: Right. I mean, this is actually one of those times where being a researcher at the Fed was a great opportunity for actually thinking about issues that other people weren’t. So back in the ’90s—late ’90s Thomas Laubach and I were starting to think about, you know, what’s the normal or neutral interest rate. And it all started with this idea of a Taylor rule, which is some kind of mathematical equation that tells, you know, a central bank how to move interest rates as the economy changes. And that has this—you know, the Taylor rule has a notion of a normal or neutral or equilibrium interest rate.
And so my colleague, Thomas Laubach, and I have developed a model on that. And one of the things that, you know, in neutral interest rates in the past, which tended to be between 2 percent or 3 percent on a real basis, inflation-adjusted basis. So a short-term interest rate of, say, 4 or 5 percent would be normal in the past, in our estimates. Then we saw a dramatic change over the last few decades, but especially since 2007, where our analysis says that at least currently the U.S. economy, the neutral interest rate—again, on a real basis, an inflation-adjusted basis—is about zero percent, which means that, you know, a 2 percent short-term interest rate is a neutral interest rate. This is a huge decline. It reflects slower trend growth in the economy. It reflects other global factors that are pushing down globally kind of neutral interest rates.
And there’s been a lot of research on this, a lot of analysis. But one of the things that we’ve been working on now is looking is this just in the U.S., is this something particular to the U.S.? I mean, often Americans and American economists, because we have all this data, we tend to focus on our country and kind of ignore other countries too much. So we’ve been looking at Europe, Canada, and Britain as using our model. And what we’re seeing is a very similar pattern in all of those countries. A couple things: One is slower trend growth. Basically because of demographics in all these countries is slower population in labor force growth, but also slower productivity growth.
Similar patterns across all the other advanced economies. We’re looking at, and in each case, we’re finding this neutral interest rate coming down significant, and especially in the last decade, to numbers that were much lower than we’ve ever found. I think this is a—you know, we can—there’s a debate among economists about, you know, will this revert back to normal eventually? But it does present a significant challenge to central banks, because if you’re neutral interest, you know, your normal interest rate is, say, 2 or 3 percent, that means when the next recession happens, you know, once we get out of this—you know, for those countries, once they get out of it, they’re going to be starting with an interest rate that’s like, you know, 2 or 3 percent.
Typically, in the U.S., when there’s a recession, the Fed cuts interest rates by about 5 percentage points. So you know, we hit the zero lower bound this time in 2008. Basically, if this kind of research is right, then countries in the future are going to be hitting the zero lower bound and basically finding monetary policy struggling and needing to go to unconventional policies much more frequently. I think it raises a whole bunch of issues that we need to be thinking more about in terms of the future.
CLARIDA: John, were talking right before we came in that one of the things that’s noteworthy, at least to me, about this—about this cycle, of course the Fed hiked in December, but at least, in my case, going back to the 1960s, I think this’ll be the first rate hike cycle for the Fed in which its actually beginning to hike rates when inflation is below the desired level. So typically a rate hike cycle occurs because the Fed wants to reduce inflation or, as in ’94 and ’04, it begins to hike when it wants to keep inflation at the current level. So does this present a communications challenge, hiking when you’re also saying we want inflation to go up?
WILLIAMS: Apparently it does—(laughter)—based on my email that I get and the bloggers comments about some of our—some of the things I’ve said about the need to raise interest rates even though inflation is below 2 percent. So I think—first of all, I think it’s actually a good thing that we’re doing this because of this lag in monetary policy. Again, it takes a—well, based on research, it takes about 2 years for monetary policy to affect inflation rates. So we really need to be thinking where is inflation going to be, not in, you know, mid-2016, but where is it going to be in 2017, 2018?
And you know, as you know, Rich, the history of the Fed has not always been a good one. In the past, there’s been the stop and go policies, especially back in the ’60s and ’70s, where the Fed did wait around too long, allowed inflation to get too high, then had to put on the breaks, cause a recession, and then take—you know, put the pedal to the metal again to get the economy going. We’re trying to avoid that on and off kind of cycle, and just gradually move interest rates over the next several years back to normal in anticipation of where the economy is going to be in the next couple years.
In terms of communication, it definitely does, you know, raise the question that people say, what’s the rush? You know, why are you raising rates when inflation is still about 1 ½ percent on kind of an underlying basis? You know, and my answer is: You know, we were at zero for seven years. We’ve only raised the rate once. I don’t know how the word “rush” fits into that description of the Fed.
WILLIAMS: But again, I think we’re trying very hard to balance the need to be forward looking, but also the fact is that this is a global environment where growth is going to be slow and there’s a lot of disinflationary pressures globally, even though the U.S. economy is doing better.
CLARIDA: So the Fed, of course, has a large balance sheet, about $4.5 trillion, reflecting the quantitative easing programs. And at various times in minutes and statements, the Fed has also—Fed officials have also indicated a desire to have the balance sheet shrink over time. So could you update us on the current state of play of the Fed’s thinking on this, because it has evolved. And in particular, when the balance sheet does begin to shrink, how will that occur?
WILLIAMS: Right. Well, the balance sheet, you know, started at about, you know, 850, 900 billion (dollars) when we started in 2007. And as you said, we’re over $4 trillion now. And most of that increase is due to the quantitative easing programs that—QE programs that we’ve done, which increased our asset side and obviously equal increase in the liability side in terms of reserves and the Federal Reserve system. So the current thinking is that we’re going to use the Fed funds target, the short-term interest rate, as our primary policy tool to adjust to reflect macro circumstances, and leave our balance sheet at least for the time being where it is. And how do we do that? Well, as our assets basically mature or are paid off, we’re reinvesting the proceeds from that to keep the quantity of our balance sheet at the same level. And we’ve been doing that for years now, and we’ll continue to do that for the time being, and basically adjust the interest rate to changing economic conditions and outlook.
Then, once we get well-away from zero, basically. Once the normalization is—we’ve made enough progress on that, then we’re going to allow the balance sheet—the plan is to allow the balance sheet to basically organically decline over time. And what I mean by organically is that, you know, we hold Treasury securities, they mature eventually, when they mature we just won’t reinvest that. And our mortgage-backed security portfolio, those mortgages do get paid off, you know, either through principal payments or people, you know, paying off their mortgages. And over many years—well, actually, many, many years, that side of the balance sheet will come down. Now, based on our own projections and, you know, analysis we do, once we start this process of allowing the balance sheet to shrink over time, it would get back to a more normal level in, you know, about, you know, six or more years or something. So it’s going to be a very gradual process of moving that back to whatever the new normal there is.
Now, I think the point is really about communications in my view, is that the Fed funds target is really the active tool that we’re using, either up or down, to respond to changing circumstances. And the balance sheet is basically just happening in the background. Now, the good news for transparency is every Treasury security, all the securities we owe, is publicly available on the Fed—New York Fed’s website. So you know, people can plan ahead. And I don’t think there’ll be disruption when we start that. So it’s really just a matter of having that happen slowly over time.
Now, the big question, I will tell you, is what’s going to be the new normal for the balance sheet? And that’s something that we haven’t actually decided. This is something we are studying. And the idea is to shrink the balance from where it is today, but, you know, we’ll see over the next few years, we can think hard about should the balance sheet shrink all the way down to where we had just maybe $25 billion of excess reserves, or we want to take a different stance, and what’s the right amount of excess reserves? And this is an ongoing research topic.
CLARIDA: Thank you. One more from me, and then we’ll turn it to the audience. So be thinking of the questions you’d like to ask.
One interesting development in the U.S. labor market is after several years in which labor force participation was declining, it began to pick up about a year, year and a half ago, as some discouraged workers reentered the labor force. So, how do you and others think about that as potentially a positive for the supply side of the economy and for those individuals getting back in the labor market? So how much more potentially do we have to go on that?
WILLIAMS: Yeah. Well, that’s a big question. I mean, you know, we don’t know. We’re not that good at predicting labor force participation. I think that we—during the recession, we saw two big developments. One was dramatic decline in labor force participation because of retirement of the baby boomers and things like that. So that’s kind of the part that we more or less understand. The other was a decline that I think was due to cyclical factors. It was the worst recession in, you know, 50 years, it lasted a long time, people did drop out of the labor force who actually do want to have a job.
So as we’ve seen the labor market get better we’ve been adding, you know, millions of jobs in the last several years, we are seeing some people come back into the labor force, and as we mentioned last year a noticeable pick up there. So my own view is that most of the—you know, where labor force participation is today, most of that is structural. Most of that—the declines that we’ve seen is due to demographics and things. But we definitely could see some further increases in participation as we pull more and more people into the labor force who had basically been sidelined during the recession and during the recovery.
It is a question mark. It’s one of those things that, you know, we have to continually update our estimates. One of the things—you know, when you’re a policymaker, you have to basically be (bazy ?). And you’re always, you know, reassessing your assumptions, you’re reassessing your views based on the incoming data. And if labor force participation does pick up—continue to pick up, then I, you know, would probably be more optimistic about how much further increase we could see. Right now I feel like we’re probably close to the trend, but I have an open mind on that.
So what’s it mean? Well, what we’ve seen in the last several months, even though we’ve been adding a lot of jobs, unemployment has not been coming down. It means more supply side to the economy. And so it’s a positive development. When I look at some of the statistics that they collect, they actually ask a question in the monthly survey of households, and they do ask the question: If you’re not in the labor force—meaning you’re neither working nor actively searching for a job—do you want to work? And that number of people who want to work but aren’t in the labor force has shrunken dramatically over the last few years. So at least from that question, there doesn’t seem to be a lot of people still standing on the sidelines.
CLARIDA: OK. So at this point I’d like to invite members to join in the conversation with their questions, and reminding this meeting is on the record. So if you have a question, raise your hand. A microphone will be brought to you. Please stand, speak directly into it, your name and affiliation. And please limit yourself to one question and keep it a question. So let’s see. I see Willem Buiter there, so we’ll let Willem kick it off.
WILLIAMS: Willem, the way to turn a statement into a question is just to say, don’t you agree, at the end? (Laughter.)
Q: Right. Willem Buiter, Citi and Council on Foreign Relations.
I was somewhat shocked by not surprised to see—hear no reference to the words “financial stability” in the entire exchange here. The Fed has not just a monetary policy dual mandate—in fact, it’s a triple mandate, (model ?) long-term interest rates, but never mind—it has a financial stability mandate. And much of the Federal Reserve Act is, in fact, about this.
Now, we clearly are in the U.S. suffering from late-cycle syndrome, right? Non-financial corporate credit, especially high-yield. We see CLOs backed by subprime car loans in large amounts. We even see, again, CLOs backed by residential mortgages—never thought I’d see those again—subprime residential mortgages. Now, the Fed has not countercyclical macroprudential instruments. Is there a case for raising the policy rate for financial stability reasons, even when it’s not warranted for dual-mandate reasons? Do you agree? (Laughter.)
WILLIAMS: Don’t you agree? (Laughs.) No, it’s a great—it’s a great question. And I agree completely with the premise. The Federal Reserve was created 103 years ago for financial stability reasons. And monetary policy as we’re talking about it is a relatively modern phenomenon.
So the way I view the issue in terms of financial stability is that, you know, given that we are close to our dual mandate goals—essentially at full employment and I believe on the inflation side we’re, you know, not only close but moving towards it, I do take seriously the notion that very low interest rates for a very long period of time does create—as we know from history and from the study of this topic—does create a lot of basically reach for yield, risk-taking channel, and can contribute to I think investors and, you know, I think other households and businesses taking on risks that actually don’t make sense in the long run. They only make sense in the environment of very low interest rates.
And so one of the reasons we don’t want to keep interest rates excessively low and just let’s see where this economy takes us, if you will, is because I believe these risks will grow over time. Right now, I don’t think we’re in a significant risk of a—in terms of financial stability, or risks of—systemic risks right now. But we don’t want those to occur either. So I do agree that one of the arguments of normalizing monetary policy, given where we are in our macro goals, is to make sure that those financial stability risks don’t further increase and become more of an issue.
I don’t think we face really a tradeoff right now, because I think that the macro situation supports raising interest rates, as well as the financial stability issues that Willem raised. So I don’t see this as a tension in terms of monetary policy. In terms of macro-prudential tools, I agree with the point—I made this point in a number of speeches—everyone loves to say our first line of defense should be macro-prudential tools. Well, in the U.S., unlike some countries, we don’t really have explicitly macro-prudential tools. So what we need to do—macro-prudential tools are things, if you think about it, in other countries where the government will just say the loan-to-value ratio on mortgages, they basically lower the level on what’s allowed on new mortgages from, say, you know, 80 percent to 60 percent, that just is required throughout the country.
Well, we don’t have those kind of tools. But we do have some tools that have a macro-prudential kind of use. And I specifically refer to the Fed’s stress tests, or CCAR process, where basically if we’re worried about significant risks in parts of the financial system, we have, and we can build those into our stress test to make sure our largest banks can withstand a major decline in asset values or other events in those areas of the economy. So I think we do have some tools that we can bring a macro-prudential perspective or framework to. But I do think this is something that—you know, whenever I hear someone hear, well, macroprudential’s our first line of defense. In the U.S. it’s really micro-prudential, which means strong regulation and supervision of the financial institutions, plus some of these other tools where you can bring more of what we call macropru perspective.
CLARIDA: Next question, please. Right there in the blue jacket, yes.
Q: Rollie Flynn, Singa Consulting and Georgetown University.
We all heard and read about the New York Fed, the cyber-hacking. Could you comment on what you and San Francisco are doing to bolster your cyber defenses, both technically and in terms of your management structure?
WILLIAMS: Sure. And you know, this issue of cybersecurity is obviously incredibly important, not only for, you know, for banks, for businesses, for all of us, but at the Fed we take it very, very seriously. The Fed does operate the largest payments systems in the U.S. economy, whether it’s Fedwire, whether it’s check or ACH. So we do enormous investment in terms of making sure that our systems are safe and cannot—are bulletproof, as well as we can.
I think a couple things that means is not only investing in, you know, hardware and software and personnel, but also having pretty—very stringent programs around employees, insider threats, thinking hard about, you know, what all those risks are now how those are changing all the time. So this is a big area for us. But it’s also for our own bank and for our own system. And obviously when we think of the Federal Reserve System, there’s 12 Federal Reserve Banks, there’s the Board of Governors. All of us are working together on all these cybersecurity risks to the system, not individually. And San Francisco doesn’t have its own cybersecurity program. It’s a Federal Reserve program.
But then we’re working hard with the banks that we regulate and supervise to make sure that they are putting in place stronger cybersecurity defenses, being much more proactive, much more, you know, modern in their thinking and their investments there. So when a lot of people ask the question of, you know, we’re seeing a lot of—the smallest community banks either disappear or merge, you know, why is that? You know, I actually think cybersecurity is one of the biggest challenges for the smallest institutions because it’s just a fixed—it’s a fixed cost of running a business today.
The last thing that I think we really need to do is to see banks and financial institutions definitely working better together about this, because there’s a competitive advantage if you can say we’re better at cybersecurity than my competitor you should do business with us. And that’s true. But we also want to make sure that we’re sharing information across institutions and working together. In terms of—you know, I think the biggest challenge for us is this is a threat that is becoming more and more sophisticated, with a lot of money behind it, and obviously at times foreign governments. So this is not, you know, the kid in Berkeley living in his parents’ garage anymore. (Laughter.) We are way, way beyond that. And we understand that.
And in terms of our own policies we’ve instituted at our own—at the Fed, but obviously at the San Francisco Fed too—we instituted a number of programs that basically restrict us significantly in terms of, you know, background—we do background checks, we do a lot more work on basically making sure that access to the payment system, to what we call the backbone of the Fed, is very tightly controlled. And one of the key advantages of that, apparently, is that I still use a BlackBerry. (Laughter.)
CLARIDA: Right there in the front table.
Q: Good morning. My name is Nili Gilbert. I’m a co-founder of Matarian Capital. Thank you. Thanks to you both for being with us here today.
Recently at a speech in Sacramento you contrasted the productivity gains that saw after the tech boom in the ’90s with the lack of productivity gains that we’ve seen in the context of current technological innovation. And you observed that perhaps this is because some of the current innovations are more focused on how we spend our leisure time, whereas the prior tech boom affected businesses more. I would observe that some of the current innovations are also affecting our labor markets, in the context of companies like Uber, Airbnb, TaskRabbit. I was wondering if you would be willing to speak further about how you see innovation affecting the U.S. economy at this time, and in particular from your perch in San Francisco, right next door to Silicon Valley where many of these companies are housed and many of these technologies are being used. Is there anything to be observed in the regional data that can help us to understand how innovative technologies are affecting the economy?
WILLIAMS: Sure. That’s a great set of questions that we focus a lot on in the San Francisco Fed. So one of the things I said in the speech, as you correctly pointed out, is that productivity growth in the U.S. economy is actually back to its old normal. This is something that Bob Gordon has written about in his book and many economists have studied, including one of my colleagues at the San Francisco Fed, John Fernald. And basically, normal productivity growth in the U.S. is about 1 ½ percent. And then we get—during our history we get—in the U.S. and other countries there’s these periods of surges in productivity growth often associated with a whole set of fundamental breakthroughs in technology, whether it’s electrification, whether it’s internal combustion engine, or things like that. And they all come together, and they basically transform what we make in our economy and how we make it.
And we had a mini-boom in the ’90s around high-tech, especially around basically computers, the Internet, and all of that, and it lasted for about 10 years. It was the smallest productivity boom—a small productivity boom compared to the others. And the point I made about this one is we’re not seeing it in the productivity data.
So if you live in the—you know, by the way, when you live in—if you’re in San Francisco, you’re not in Silicon Valley, but everybody who works in Silicon Valley lives in San Francisco now. So you—basically, they’re creating all the traffic and everything. (Laughter.) So you feel like you’re in Silicon Valley.
So what we’re seeing is a couple things. I think—you know, why we’re not seeing a big productivity boom? One is, is that some of it’s just—is really that it’s about leisure time. It’s about sitting on your phone if you’re on the train or the bus or, you know, at home, and you’re, you know, seeing all the amazing things that cats can do which you never knew before. (Laughter.) OK, so that’s great. I’m not against it. I love cats. I say this and people say, you don’t like cats. No, love cats. But that’s not really GDP. It doesn’t show up as GDP. (Laughter.) And in fact, it could be negative for productivity when you’re doing it in your workplace. (Laughter.)
So that doesn’t show up in GDP. Maybe it shows up in utility. Maybe it makes us happier and stuff, but happiness isn’t what we’re measuring when we measure output per worker in the business sector.
The other thing is we don’t produce computers in the U.S. very much anymore. Back in the ’90s, we were actually building the computers. The chips were built—you know, made by Intel or other companies. Computers were often assembled here. But a lot of this was actually investment in the United States, in very high productivity jobs in the United States in the hardware industry. Well, the hardware industry has mostly moved out of the U.S., has moved to Asia, and now a lot of the innovation is really in software. And that just doesn’t affect productivity in the same way that all of those jobs—factory jobs that we saw in the ’90s.
Now, in terms of—you know, if you’re in Silicon Valley they say, no, John, you don’t understand; we’re mismeasuring. It’s the accountants that are wrong. It’s the—you know, it’s the economists. It’s the government, you know, statisticians. But again, my colleague John Fernald just did a really good Brookings paper in—a couple of papers now on this. And, you know, if you look at the productivity slowdown, no matter how you slice and dice it, or how you try to, you know, get all the gig economy and Google and Facebook and all these things measured as best as you can, it still does—it doesn’t really change the story. The productivity slowdown is real. Whatever we’ve seen in the last 10 years is likely to be what we’ll see in the next 10 years. And it’s just basically because the newest technologies haven’t fundamentally changed what we produce and how we produce it in the U.S. in a way that earlier ones have.
Now, going back to the gig economy, I’m going to say something else that people in Silicon Valley just shake their heads and say, you’re an economist, you don’t understand. It’s still tiny. Even in the Bay Area, it’s tiny. You know, companies like Uber and Lyft and those, I mean, the tax industry is basically a rounding error to the U.S. GDP; Airbnb, these things. I think they’re important, and I think they have important issues about what employees are, issues about full-time/part-time employment and things like that, a lot of issues that I think our legal system or kind of regulations need to evolve to fully incorporate. But in terms of the economy, it’s still a pretty small part of the economy.
So what are—what is the U.S. economy today? I think a lot of people, you know, I think are thinking about kind where we are in 2016. It’s not only a service-sector economy, it is a primarily service economy. So what are services? If you look at the national income accounts, it’s health care, it’s education. It’s also housing, but you know, so it’s just abstract from that. But where the U.S. economy, you know, creates the jobs? Over 90 percent of the jobs created in the U.S. are in services last year. Where is our GDP growth? It’s in services. Manufacturing is a very small part of our employment. It’s still a part of the GDP. And so when you think about—whenever you think about these changes that are happening, you just got to go back to say how does this fundamentally change the delivery of health care? How does it fundamentally change education? And how does it fundamentally change what we produce in our economy?
So these are great open questions. But right now, you know, we’re studying them actively. I think there are a lot of regulatory/legal issues that need to be thought about. But in terms of the macroeconomy, it’s just not having the impact that I think a lot of people would have expected.
By the way, even if you try to measure this utility thing, some people have studied how many hours people spend on their iPhones and, you know, Android phones, and for me on my BlackBerry—but of course, I can’t do anything cool on my BlackBerry—(laughter)—what we’re doing is we’re basically not watching—well, I should say it right—we’re watching TV on our phones now. So it’s basically a movement of advertising revenue and eyeballs from newspapers, magazines, and TV—you know, old TV, you know, the boxes that you watch in your living room—to these phones. So it doesn’t—again, it’s cool, it’s interesting, people love it, but it doesn’t create a lot of GDP and productivity because it’s basically just moving one activity from one device to another.
And I think I probably out-geeked what I needed to do on that question. (Laughter.)
CLARIDA: Right there at the front right table, yeah.
Q: Thank you. Stephen Blank.
The views—the rather rosy scenario you described about our economy couldn’t be more different than the views on the economy expressed by our emerging political leaders and agreed on with many of their followers. What do you think explains this variance? And, more important, what does it portend for the relationship between the Fed and the emerging political leadership we’ll see in the next—in the next years?
WILLIAMS: It’s a great question. And I think it’s actually pretty easy to understand, because at the Fed we’re interested in our dual-mandate goals, again, which is 2 percent inflation and what we think of as maximum employment. And we just take the supply side of the economy, the structure of the economy, fiscal policy, we just take that as given. This is the economy that we’re handed, and how do we best achieve our goal?
So the way I view it is—the Fed, in my perspective—we’ve done what we can in order to get the economy, you know, basically back to full strength with interest rates. Interest rates can only do so much, and obviously they can’t solve structural or other problems. We’re doing what we can to get inflation back to 2 percent.
So what are the issues that really are the big economic issues in the U.S. economy? They are much longer—bigger, longer-term issues about the supply side, if you will, of the economy: the fact that productivity growth is only 1 ½ percent; the fact that trend GDP growth is 2 percent or even a bit less than that, which is the lowest we have experienced in our lifetimes; the fact that, you know, income inequality is rising, or however you think about these—or economic mobility seems to be declining somewhat. All these issues—the hollowing out of the middle class—these are all real. These have been studied. They have to do with a lot of different dynamics over the last several decades. But they’re not the things that monetary policy can or should address.
And so when I sound optimistic—and I hear—you know, it’s a great question, and I hear it all the time—it’s optimistic within the confines of what can monetary policy do. Monetary policy can get unemployment down from 10 percent, which was our peak, down to 9 (percent), to 8 (percent), to 5 (percent), or maybe a little bit lower. But if we try to do more than that, we just create other problems, whether it’s the financial stability issues that Willem highlighted, or just creating inflation or other problems.
So I think that this is the disconnect. I’m optimistic within the context of the hand that we’re dealt with—dealt. I think the concerns a lot of people raise is that the hand isn’t one that’s favorable. I mean, trend growth is lower than any time we’ve seen before, all these other structural issues. And again, I can’t do anything about that.
So you asked, what’s it mean for the Fed? I think it does, I think, you know, create a lot of these—this tension about, you know, the last several years people have been saying the Fed’s the only game in town, right, you know? So they’re saying, you know, fiscal policy at times has been at loggerheads, other policies, you know, or other issues—you know, we’ve had government shutdowns and things like that—so we’re counting on the Fed to solve our problems. And how I’m hearing more and more of these kind of, you know, claims or, you know, people raising issues: well, can’t the Fed solve this problem? I mean, you guys were amazing, you know, low interest rates, big balance sheet. You got $4 trillion; can’t you use that to deal with these structural problems? And we know from history, both in the U.S. and in other countries, that when you try to solve structural problems, longer-term problems, with monetary policy, it leads—you know, it never ends well. And I think that that is going to be something we’re going to probably see more pressure on us to try to use monetary policy to solve longer-run structural problems, and we just have to fight that every day.
CLARIDA: Right there. Mmm hmm, next to Willem.
Q: Mario Platero, Council on Foreign Relations and Sole 24 Ore.
I just wanted to go back to the short term. And we have seen some retailers that have done pretty badly recently, and we have seen housing in New York that is slowing down. Apparently, there is not such a great sale of the new condos that have been built up. And so the economy seems to be—you know, there are some issues. So the question is, how do you put those items into perspective? And what kind of bias the slowing of the economy would have in your, you know, monetary policy decision-making in the short/medium term? We know Dudley last week said that, you know, maybe by June there could be an increase of rate. So would the economy intervene, or that is pretty much a decided factor? Thank you.
WILLIAMS: So this is—you know, when we say we’re data dependent—and I always say, you know, monetary policy is data dependent—it goes right to your question, right? I mean, wait a minute, what data are you looking at, which data are decisive in terms of decisions. And of course, when we say data, there’s millions of data points that we observe and collect and analyze.
But I, again, go back to what’s it—you know, what’s it mean for our goals of maximum employment and 2 percent inflation, or price stability? So I look at the data, again, from a two-year perspective. I agree with your point that we’re seeing some softening in some parts of the hottest commercial real estate in the—in the—you know, the globe, New York and San Francisco. But remember, this is in the context of, like, all of my employees would love to see some declines in rents in San Francisco. So, you know, it’s in the context of a very hot market.
So we have to read through that data, try to analyze it best we can. That Q1 GDP growth, which was very soft, you know, to what extent does that portend softness in second quarter? So far, I think the data are showing that that was an aberration and anomalous, and the second quarter’s looking better. But we do want to keep watching that. We’ll get employment data for—you know, for this month, and we’ll get more—you know, more data on the economy. We have to take all of it together and try to get this big picture of what’s it tell us where the economy is going over the next year or two.
In terms of the retail sales, this is one of the reasons macroeconomics is not a stagnant kind of subject or field. A lot of what we’re seeing is this mix between bricks and mortar, whether you think about the big department stores like Nordstrom or Macy’s or others, versus online sales. So we—you know, we have to make sure the—you know, the data we’re looking at is not just what’s happening in one part of retail, but it capturing the broader trends. And of course, our statistics have done a—have to do a good job of keeping up with that.
In terms of where we are in terms of monetary policy, since you did ask, it’s the first time I’ve heard the word “June” mentioned this morning, so, OK, I’ll go—I’ll bite. You know, it is data dependent. And what that means is, you know, Rich asked me, you know, what kind of rate increases over the rest of the year. That’s going to—you know, if the data come in softer than expected, you know, all else equal, that tells you we’ll move a little bit slower or maybe a little bit later in terms of the next rate increase. But, you know, my own view is that, you know, the data stream we’ve been getting over the last several months has been—you know, I always compare it to what I was forecasting a few months ago or a few quarters ago. And my view is, in terms of employment and inflation, the data—the data we’ve been getting in the last few months or quarters is basically consistent with the forecast I’ve had. So, you know, there are some ups and some downs, you know, and everyone gets caught up in that labor report or that retail sales or that, you know, manufacturing report, but overall I view that the data have been consistent with, you know, a good 2 percent growth and steady job gains.
CLARIDA: Right there. Straight on. Yes.
Q: My name is Peter Goldmark. I’m an independent consultant to foundations in the climate and energy area.
This is a room where a lot of people will be familiar with Mervyn King’s “End of Alchemy.” And I wondered if you would share with us your reaction to his recommendations to, quote, “avoid the next financial crisis”—the two-track system, wide banking, narrow banking, different reserve policies. What are your thoughts on that? I found it very powerful.
WILLIAMS: So, you know, there’s—obviously there was a lot of reassessment/reevaluation of the appropriate regulatory framework after the financial crisis—a lot of studies being done by universities, by academics, and policymakers. In the U.S., I think that the approach that has been taken so far has hit the most important kind of risks to the financial system in terms of making sure that our biggest banks have adequate capital liquidity, that we put in enhanced supervision. The bigger you are, the more supervision and regulation you get, and the more demands we have for you for capital liquidity—basically putting on, if you will, as an economist, like, a—if you create a negative externality, meaning that by being very large you create a risk of too big to fail, that we put more—basically, more requirements on you, making that a—making you make the tradeoff between should we be big and take on all the extra regulatory requirements, supervisory requirements, or choose not to be big.
So I think that the approach we’ve taken so far has been the right one in general. It’s very complicated. It’s got—you know, nothing’s perfect. But right now I feel that the—you know, in terms of the largest banks, in terms of the, you know, financial system broadly, we’re in a much better place than we were 10 years ago. However, I think that the issue that we should be kind of reassessing or reevaluating over the next five to 10 years, or thinking about in terms of the future, the way—getting to your question, is, what do we do in the U.S. around the shadow banking system? Which is now—because we’re—gotten more regulation, more requirements on the regulated depository institutions, it’s just obviously pushed more and more activity to the non-depository institutions, or the shadow banking. And I think that this is the area where I—where I worry. As it’s grown, it creates a lot of systemic risk. Obviously, we saw that in the financial crisis with, you know, AIG or Bear Stearns or Lehman or, you know, money market mutual funds. So, to me, that’s the question that we need to be focused on, is how do we bring a regulatory framework that isn’t focused—isn’t designed based on what your charter is, but basically is designed more around how do we protect the financial system as a whole.
So when I look at what Britain has done, for example, I mean, I think there are advantages to their approach, and I think there are advantages of thinking about how do we create, you know, a system that makes sure that we just haven’t basically protected one—created a Maginot Line around the banks, but then created an opportunity and actually pushed more activity to areas that create more risk.
Right now, I have to admit that, you know, the Fed, we’re just focused on getting the Dodd-Frank, if you will, regime in place, getting the last few pieces, which has to do with living wills and all of that to the place where, obviously, the regulatory—the Board of Governors, the FDIC are comfortable with that. I think, you know, once we have all of that, then some of these kind of deeper, longer-term questions are ones we should be thinking hard about.
CLARIDA: And the last question will be right there at the front table.
Q: Good morning. Bhakti Mirchandani, One Williams Street. Thanks for a great discussion.
In today’s discussion, you mentioned two developments that may limit the use of tried-and-true monetary policies in future crises. One is the change in equilibrium interest rates to zero percent, and the other one is the quintupling of the Fed’s balance sheet since 2007. Given both of these developments, what types of unconventional monetary policies do you expect the Fed to consider in the next crisis?
WILLIAMS: Not just crisis, I think just even in the next recession, I mean, or next situation.
So let me—let me broaden that out a little bit. So say that we had a—you know, it was called a rosy scenario, but say that we get—for whatever reason, my forecast of 2 percent growth, continuing improvement in inflation doesn’t come true. What tools do we have today to boost the economy, to get us back on track? So I think it’s the same toolkit we used from 2009 till today: basically, interest rates close to zero. You know, we could lower interest rates back close to zero or not raise interest rates. That would be one way to stimulate the economy, if it was needed.
The other is we could go back to QE. I don’t think we’re out of—at all out of ammo on that. I personally think that our asset-purchase programs were successful in lowering long-term interest rates, increasing—basically improving financial conditions and boosting the economy. So we can go back to that.
I also feel the forward guidance that we used in 2011 to 2012, basically saying we’re not raising rates for the next couple of years, was a very powerful tool to clarify what the Fed’s thinking was in that kind of circumstance. We’re not in that circumstance now, but at the time it was—these were powerful tools.
So I think that those are the same toolkit we can go back to. We understand them better. I think we have more confidence in which are the ones that work better under different circumstances, and we can go back to those if those were necessary.
You didn’t ask, but I—you know, we’re out of time. I will say I don’t think negative interest rates is a tool that we would use, except at the—it would be at the bottom of that list. I think the tools that we would use in the future would be very similar to the ones we’ve used in the past.
What I do worry about is, when I look around the globe, it’s not the U.S. economy. I think the U.S. economy’s in good shape. I think we’re in a good place. You look at Japan, you look at Europe, you look at ECB, Switzerland, Scandinavia, I mean, all of these countries are facing persistently low inflation. It’s going to take—you know, they’ve got—not only do they have massive QE and they’ve got negative interest rates, very negative interest rates in a few countries, and they’re struggling to get inflation back on track. So that’s where—if you ask me what keeps me up about thinking about monetary policy, it’s really not the U.S. so much, but thinking about what’s happening there and learning from what they’re doing in terms of the effectiveness of the tools that they’re deploying.
CLARIDA: All right. And on that note, we will conclude, right on time. So thank you all, and have an enjoyable rest of your day. Thank you. (Applause.)