A Conversation with Jeremy Stein

Friday, June 28, 2013

Jeremy Stein discusses current monetary policy developments.

The C. Peter McColough Series on International Economics is presented by the Corporate Program and the Maurice R. Greenberg Center for Geoeconomic Studies.

RICHARD CLARIDA: Hello, I'm Richard Clarida, the Harriss Professor of Economics at Columbia, and it's a real thrill to be here this morning to moderate this session with Federal Reserve governor Jeremy Stein. This is part of the CFR's Peter McColough Series on International Economics, and, again, I remind you it is on the record.

Just a brief bio on our speaker -- I've known Jeremy for years and he, of course, has been a member of the Federal Reserve Board since May of 2012. Prior to that, he was a professor of economics and finance at Harvard University since 2000 and also a senior adviser to the secretary of Treasury and the National Economic Council in 2009 in the Obama administration.

He has had many recognitions in his career. He's widely regarded as one of, if not the leading, scholar in the intersection between finance and macroeconomics. President of the American Finance Association, and, without further ado, I will now turn it over to Governor Stein. He'll begin with some remarks. We'll then have a conversation. We'll have plenty of time for question-and-answer. Thank you.


JEREMY STEIN: OK, thanks, well, thanks, Rich. Thanks very much.

Well, it's a pleasure for me to be here at the Council and I look forward to our conversation. Just to get things going, I thought I would start with some brief remarks on the current state of play in monetary policy. As you guys know all too well, we had our meeting last week. At the meeting, we opted to keep our asset purchases going at the -- at the current rate of $85 billion a month, but there's been an awful lot of discussion about recent changes in our communication both in the formal FOMC statement, in Chairman Bernanke's press conference that followed the meeting, and what I'd like to do in the next few minutes is offer my take on these changes as well as my thoughts on where we might go from here.

But before doing so, let me note that I am speaking for myself, and that my views are not necessarily shared by my colleagues on the FOMC.

So, I think it's useful to start by discussing the initial design and conception of this round of asset purchases, and two features of it in particular are of note. The first is, you know, it was designed to be flow-based and state-contingent. That is to say, we intended from the beginning to continue doing asset purchases until the labor market improved substantially in a context of price stability. But the second -- and this is sort of key -- is that in contrast to the forward guidance we've been giving with the federal funds rate, we chose at the outset of the program not to put a concrete articulation on what "substantial improvement" meant. We didn't put a numerical threshold. So, on the one hand, the program was designed from the beginning to be data-dependent, but we didn't spell out the nature of this data-dependence in a formulaic way.

So, to be clear, I think that was the sensible thing to do. I think that made a of sense, particularly at the outset. So, if you think about where we were back in September, it would have been hard to predict with any confidence how long it would have taken to reach any given milestone. So, if you had said, you know, back then at 8.1, how would it have taken, say, to get to 7 percent unemployment, there was a fairly large standard error around that. And if you think of the fact that as time passes the balance sheet is growing and there's some uncertainty regarding the costs of an increasing balance sheet, I think it was a reasonable thing to not want to, you know, to -- it was sort of prudent to preserve some flexibility on that dimension.

So, I, again, I think that was the right design choice at the time but, of course, the flip side of maintaining this flexibility is that it meant that we provided to you, to the market, less specific information about the nature of our -- of our reaction function on the asset purchase side. I mean, it -- why it made sense on the other to do it is we were not with the -- with the funds rates, you give the concrete guidance; you're not accumulating a stock of something that might be -- that might be costly.

All right, so, where do we stand now? We're nine months in. I think with respect to the economic fundamentals, both the current state and future outlook, things are improving. When we started, the unemployment rate was 8.1 percent and trailing nonfarm payroll growth over the prior six months was close -- just under 100,000 -- 97,000. Now we're at 7.6 unemployment and trailing six-month payroll growth is closer to 200,000 -- 194,000. If you look at our forecast going forward, back then, in September, the forecast for year-end 2013 was for the unemployment rate to be about 7 and three-quarters (percent). This is the forecast made by -- this is essentially the midrange of the forecast made by the FOMC members, and now it's -- that forecast -- so at a fixed point in time, that forecast has come down about half a percent.

So, again, there's clearly been some progress. And it's difficult to do the attribution, but it's my belief that our policy has at least played some supportive role in generating that improvement. I mean, one thing I'd point to is that some of the clear progress has been in areas that are typically response to to monetary accommodation -- housing market, autos and so forth.

And, of course, asset purchases also bring with them various costs and risks. I've been particularly concerned about risks relating to financial stability, but thus far I would judge that the cost-benefit test has been -- as been met.

All right, but the fact that we've made progress in some sense has brought to the fore communications issues because clearly the further down the road we get, you get closer to the end, in some sense, and market participants need to know -- or are going to have a stronger demand to know what are the conditions that are going to be -- that are going to lead you to wrap up the program? And that's the context in which I think the current messaging should be interpreted. So, in particular, I view Chairman Bernanke's remarks at the press conference -- in which he suggested that if the economy progresses, essentially, as we anticipate, asset purchases might be expected to wrap up when unemployment falls to something like the 7 percent range. That's an effort to put some more specificity around the heretofore less well-defined notion of "substantial progress."

OK, now, it is important to stress that this added clarity is not a statement of unconditional optimism, nor does it represent a departure from the data-dependent philosophy of the program. Rather, it involves a subtler change in how data-dependence is implemented; that is to say, rather than just saying "substantial progress," it's a greater willingness -- it's sort of less reliant on a "we'll know it when we see it" approach.

I think that's a very natural evolution, which is as we make progress toward our objectives, the balance of the trade-off is going to shift between wanting to maintain flexibility -- as we get closer, it's -- we have less balance sheet uncertainty about how long it will take, how much we'll have to accumulate to get to a given end-point than we did back in September. OK, so the balance sheet becomes more manageable at the same time that the market's demand for specificity goes up.

So, in addition to guidance about the ultimate completion of the program, market participants are also interested to know -- or eager to know about the conditions that will govern interim adjustments to the pace of purchases. Here, too, it makes sense for the decisions to be data-dependent. But something I'd like to stress is that when we -- as we approach an FOMC meeting where an adjustment decision looms, it is appropriate to give relatively heavy weight to the accumulated stock of progress that we've made and to not be too excessively sensitive to the sort of near-term momentum captured, for example, by the last payroll number that comes in just before a particular meeting.

Now, in part, this just reflects good statistical inference. That is to say, you don't want to put too much weight on one or two noisy -- on one or two noisy observations. But there is more to it than that, which is not only do our actions shape market expectations; the converse is true as well. Market expectations influence FOMC actions. So, it's difficult for the committee to take an action at a meeting if that -- if that action is wholly unanticipated and we don't want to create a lot of market volatility. But once you recognize that there's this two-way feedback between financial conditions and FOMC actions, an initial perception that noisy data play a central role in the policy process has a potential element of becoming somewhat self-fulfilling, so -- and can itself be the cause of extraneous volatility in asset prices.

So, in an effort to make reliable judgments about the state of the economy as well as to reduce the possibility of an adverse feedback loop, the best approach is for the committee to be clear on the proposition that when we're making a decision, say, in September, we're going to give primary weight to the accumulated stock of news that;s come in since the inception of the program and to not be unduly influenced by, say, whatever data releases come in, in, you know, the weeks before the meeting, as salient as these releases may appear to be to market participants.

Now, let me emphasize that doesn't necessarily mean that we're going to abandon -- it doesn't mean that we're going to abandon the premise that the program as a whole is data-dependent and forward- looking. So, for example, even if a data release comes in, in September, and it doesn't overly color the decision to make an adjustment in September, that data remains valid for the remainder of the program. And if, you know, the revisions substantiate it and the further data in October, November, substantiate it, since we have in some sense a fixed end-point goal, if the news is week, we would tend to go further -- you know, the program would be extended further even if an adjustment is made based on the data.

OK, so I think the principle of articulating, sort of, where we're going in many ways takes a lot of the fray -- takes a lot of the heat out of the exact pace -- whoops, there you go -- takes a lot of heat out of the exact pace of when we do it because, I mean, this is in some sense the clarity we're trying to provide to the market. If you have a better sense of where we're going, the exact, you know, speed at which we're driving when we pass Chicago becomes a little bit less -- becomes a little bit less important.

So, in sum, I think there's basically the following key principles for effective communication. First, I think it's central for us to reaffirm the idea that the program is effectively a data- dependent program. Secondly, as the chairman has tried to do, give more clarity on the type of data that will determine the endpoint of the program. Third, when we make interim adjustments, as we will, those interim adjustments should be made primarily on the bases of accumulated progress toward our goals and should not overemphasize the most recent momentum in the data.

So, all that's about -- all that's about asset purchases. With respect to federal funds rate, we have had explicit guidance in place since December. That explicit guidance stands. We've reaffirmed it, and a number of -- the chairman and a number of other FOMC members have come out and basically reaffirmed that in recent days -- that we have the 6.5 (percent) threshold.

It's a threshold, so that means it's -- that's when we'll first begin to consider a first increase in the federal funds rate, and as Chairman Bernanke and others have emphasized, that gives us further flexibility in terms of reacting to the incoming data. So for example, if inflation continues to be softer than we anticipate, it would be very natural for us to extend the period longer beyond the 6.5 (percent) -- I mean, we'll be doing what's effectively optimal at that point in time by extending the duration of time at which we keep the federal funds rate very low.

Now, of course, there are limits to what you can do even with good communication -- even with good communications in terms of -- in terms of market volatility. I think the best that the FOMC can do basically is to help market participants understand how we're going to make decisions about the policy fundamentals that the FOMC controls, that is to say, the path of the short rate and the total stock of assets that we -- that we're going to purchase.

Now, you know, a perspective that's useful to keep in mind and that this sort of really recurs in academic finance is if you look at variability and asset prices, be they stock prices, long-term bonds, you can really only explain a relatively small fraction of that variability by appealing to changes in fundamentals. So this is, you know, Robert Shiller many years ago showed you can only explain a small fraction of the variability in stock prices by, you know, changes in dividends, changes in earnings; a similar thing holds true for long-term bond prices as well.

The bulk of the variation comes from what finance folks call changes in discount rates, which is essentially a fancy way of saying stuff we don't understand as well. (Scattered laughter.) Right? In other words, it could be changes in investor sentiment, in risk aversion, in levering, de-levering, people getting off sides and having to sell, market dynamics broadly speaking, OK?

So that's just a general statement, but it sort of reminds you -- I think it often doesn't make sense when faced with the large price move, when you say, oh, prices have moved a lot, it doesn't make sense to sort of have a temptation to say, I can explain that based all on a particular set of fundamentals, so in --- to say, oh the price has moved a lot; what was the change in expectations about the path of policy is not necessarily sort of the right -- the right way to think about it.

So while we've obviously seen very significant increases in Treasury yield since the FOMC meeting, I think it's a mistake to infer from those movements that there must have been essentially an equivalently big change in policy fundamentals.

You know, just to be specific, nothing we have said suggests a change in our reaction function for the path of the short-term policy rate. I mean, we've been very explicit that that reaction function with the 6.5 percent threshold stands.

And with respect even to the asset purchase program, if you think about the specificity that we've put around it, the clarity that we've put around it, my own personal reading of it is it puts us very close to where market expectations were before. You know, we look at these primary dealer surveys and others, and a program that you know, under sort of, you know, the scenario we're envisioning, wrapping up something in the neighborhood of the 7 percent unemployment rate, seems very much in line with market expectations.

So again, I wouldn't try to back out from the market moves that there has been a substantial change in policy.

Now, at the same time, I don't -- I don't mean to suggest that the market movements are inconsequential or should be dismissed as noise; quite to the contrary, I think they teach us a lot about the dynamics of markets and how our communication policy interacts with -- interacts with that. The only point I'm making is that if, you know, businesses and consumers are looking to market prices for clues about the stance of monetary policy, they should take care not to overinterpret the market moves.

So we've -- you know, we've attempted in recent weeks to provide more clarity about this -- the nature of our reaction function, but you know, I just want to stress, I view the underlying fundamentals of the policy stances being broadly unchanged.

Thank you. Thanks very much, look forward to the -- look forward to the conversation. Thank you. (Applause.)

CLARIDA: OK, Jeremy, well thank you. Thank you very much. You covered a lot of ground there, and I definitely want to give our audience plenty of time to ask you questions. So I think I'll focus on two or three points at the -- at the outset.

One of the things you mentioned right towards the end of your opening remarks is that the reaction in markets provides you with information that you were learning about. One thing that occurred to me is there's been a big discussion in academia and at the fed about the stock versus flow effects of QE.

So has the market reaction taught you or the FOMC any information about the stock versus flow argument?

MR.STEIN: So that's an interesting question. I mean, you hear this all the time, that market folks say, you know, flows may matter more than stocks. I think that that's certainly right. I think that's certainly right, not more -- that the idea that flows matter to some extent. That certainly has some truth. I think it probably has a little bit more truth in the NBS market than in the Treasury market.

In the current thing, when I hear flows, I'm not thinking narrowly just of the flows of our purchases, but flows related to, you know, redemptions from open-end funds, things like that. I think that's probably more, because it's hard to say -- it would be hard to say, well, jeez, we had a surprise that was not a surprise about stock but a surprise -- you know, I don't think that there was a big surprise about either, really, the stock or the flow of our own stuff. So there's something going on that's not the stock or flow of our stuff that could be flow-related, but I would sort of be more inclined to think of it as market-type flow-related stuff.

CLARIDA: Got it. You mentioned, as did the chairman last week, the 7 percent guidance, which was the first time the Fed had given us a number for substantial improvement in the -- in the labor market. The chairman spent less time last week talking about inflation. And of course, as you know, inflation is running well below 2 percent now. So does that convey any information about your or the board's thinking about the relative waits that those two are receiving right now?

STEIN: No, I mean, I think -- look, I think inflation is obviously central to what we do, central to our mandate. I'll say -- I'll say something about the inflation forecast in just a sec, but, you know, the thing to maybe stress more so than our forecast or anybody's forecast again, is the central data dependence of everything we do both in terms of the asset purchases and as well as the forward guidance. So I think that the chairman emphasized this in his press conference, even if it maybe didn't get quite as much -- quite as much play is, you know, for example, you know, we think inflation is likely to mean revert at least somewhat. Could be wrong, could be wrong. If that turns out to be wrong, when we get to thinking about decisions for the funds rate, that will obviously enter in a very first order way. That's, again, central to everything that we do. So, you know, the likelihood that we'll start to raise the funds right around 6.5 (percent) is going to be much lower if inflation is -- so I mean, that's just --

CLARIDA: And you mentioned that today.

STEIN: Yeah, yeah.

CLARIDA: One of the things that I've admired about -- and for those of you on the Fed's website, Jeremy's given a number of speeches -- I think three in the last several months -- that are really excellent overviews of some of the issues the Fed faces as it thinks about the interplay between policy and markets. There are those who interpret the chairman's comments and the board's decision last week as being in part influenced by your work in terms of frothiness or potential excesses appearing in markets. Would you like to discuss that, to the extent that was a factor?

STEIN: So I would be happy to discuss my stuff. (Laughter.) But to focus -- let's just focus on the -- I think the decision that was made at the most recent meeting was really first, foremost and almost entirely a decision about clarity. As I said, I don't think it was an effort to change the stance of monetary policy. I think we were coming to understand, as we moved forward in time, it was just not acceptable to leave, you know, unanswered or to put no color around what substantial improvement means. And so I think really the essential focus was on trying to -- you know, now, how the market was exactly going to react, you know, there's a set of issues. But it was not -- I can tell you with some confidence, at least speaking for myself -- and, you know, and I'm sort of obsessed with these issues, I was not thinking, oh, well, this will move market volatility in a particular way. This will move asset prices. Really the goal was a feeling that market participants need to better understand what substantial improvement.

CLARIDA: Understood. You mentioned -- and I myself have been thinking about these issues as well -- the two-way interplay between your policy and market reaction and the concerns potentially both theoretically and in practice about avoiding self-fulfilling issues. And I think you introduced -- I've not seen another Fed official mention it quite this way -- this idea of looking at the accumulated evidence.

STEIN: Right.

CLARIDA: And so perhaps since I think that is sort of a new thought here, could you perhaps elaborate on the merit of that?

STEIN: Yeah. So one of the things that I think is potentially counterproductive is -- and by the way, let me step back.

I think we've done a lot by starting to put some clarity around the end point. I think this helps a lot. You know, in a world where there's no clear end point, where we're just saying substantial progress and people are kind of -- different people have different views of what that means.

I think as you start running up to a meeting, you have a sort of speculative dynamic that is maybe not the most productive thing in the world. Where, you know, maybe market participants start thinking, you know, what exactly is the Fed looking at? Well, maybe they're looking at the latest jobs number. Maybe the market conjectures that the Fed is looking for a number like 200,000. Maybe that's not how we're thinking about it, but maybe that's how the market is thinking about it.

Well then, if the number is not 200,000, it's 170,000, and the market concludes, well, there's no way the Fed is going to do anything at this meeting, that at least -- I won't say -- you know, we're not a prisoner of the market, but it circumscribes a little bit your flexibility if there's a very strong market presumption that you're not going to act at this meeting.


STEIN: I think once you have pinned down the -- not pinned down -- but once you've articulated the end point a little bit better, it takes the heat off any particular meeting, right? Before if we didn't act in September, or didn't act in December, the market was, in some sense, free to draw the conjecture, well, that's three more months longer the program will go on, that's X number of billion dollars of increased asset purchases.

Now, well, it seems more like a decision to do it now versus do it later, but the total scale of the program has a -- has a tighter connection to economic fundamentals. So I think there's a desire to sort of minimize or to reduce sort of speculative stuff that's around, you know, guessing the -- guessing the month.

CLARIDA: One point which may sound a bit technical, but I think it is relevant here, is that in -- I think was December of 2012 when the 6.5, 2.5 numerical thresholds were introduced. That was a -- that actually appeared in the FOMC statement. In this context, the 7 percent number appeared not in the statement but in the chairman's press conference. Should we infer anything from that?

STEIN: Well, I think -- I think the one thing you should infer, very clearly, is it reflects in some sense the extent to which the FOMC was thinking about this not really as a change in policy. I mean, I think we were thinking we had the same essential policy as before and we're trying to clarify a little bit what "substantial improvement" meant. And the press conference is a natural venue -- certainly a natural first venue where the chairman can explain.

And as he said, you know, it's a little delicate because I don't think we want to be having the 7 percent number heard as the formula. I think the word he used was indicative. In other words, we're trying to be a little bit -- you know, on the one hand, provide more clarity but not be, you know, here's the equation and it's 7 percent unemployment divided by the square root of the labor force participation rate or -- you know.

So it's indicative of the sort of labor market conditions you might want to have, but that sort of means that other things will be in line with it. It's not 7 percent on the back of very weak labor force participation or weak GDP growth or something like that. So I think giving him the opportunity to sort of put a little bit of color around it seemed like a sensible way to proceed.

CLARIDA: One more question from me and then I will let our audience ask you questions. There are sort of different flavors of Fed guidance. And in the summer of 2011, the Fed introduced for about a year and a half the notion of what has been called calendar date guidance. And of course, going back to '03, the Greenspan Fed had a version of that with considerable period. Any thinking on your fault or the FOMC's about the relative advantages or disadvantages of calendar date guidance versus what you've laid out here very clearly today -- the macrothreshold approach to guidance.

STEIN: Yeah. I think we've learned -- I think we learned basically -- you know, calendar guidance was the sort of -- I think a natural first place to turn. It's extremely clear. It communicates easily. It's unconditional. But I think we learned as we were going sort of some of the limitations of calendar. You know, because it's unconditional, you know, how strongly can you say: We're going to go to, you know, end of 2014, when obviously it has to depend on -- (inaudible).

So if you want to ultimately put a little bit of -- bit more force behind the statement, it has to be data dependant in some way. Now of course, it's impossible -- and, you know, here are the trade- offs we've been playing with -- it's impossible to summarize the state of the labor market exactly by any particular number. But I think, you know, the evolution in our thinking has been it's better to attempt to build some contingency in rather than to have sort of a flat thing.

And again, just to -- just to be very clear -- you know, when the chairman spoke he said something like, you know, we might reach the 7 percent mark middle of 2014 with respect to the asset purchases, but he's not putting a date on it. He's not saying, we're going to buy until. He's saying if the sort of data-dependant thing works out such that we kind of hit the median of the forecast, that might be where it winds up.

But I think we've become sort of very, you know, sensitized to the limits of putting a fixed -- a fixed calendar date.

CLARIDA: OK, I think it's now time for questions from the audience. I see Mr. Fisher there, hi, Peter.

QUESTIONER: Hello, good morning.

STEIN: Hey, Peter.

QUESTIONER: Peter Fisher of BlackRock.

STEIN: Yeah, hi.

QUESTIONER: An observation and a question, I mean, both you this morning and Chairman Bernanke have actually played with the language a little bit. The committee's always spoken about the outlook for the labor market improving substantially, which would refer to the committee's forecast. The committee's forecast for the labor market has improved dramatically from last September, as you alluded to. And so the shift to the 7 percent number or something like that is actually moving from a forecast, that is, the committees in long range, to a point -- a point in time, a point number.

And I just sort of leave that for you as the backdrop to asking you, putting aside whether it's calendar-based or data-based, is forward guidance here to stay in central banking? Are we ever going to go back to central bankers commenting about the state of the world they see in front of them and doing whatever they do today? Or is forward guidance commenting on something about time horizons and data in the future going to be with us, or are we ever going to exit from forward guidance?


STEIN: So that's a very good question. I mean, I would -- the only thing I would try to unpack a little bit is transparency -- is the transparency aspect versus the how-far-forward aspect. So I think and I very much hope that the transparency aspect is with us to stay.

Some aspects of forward guidance were clearly designed to deal with zero lower bound problems, in other words, you know, we couldn't lower the current funds rate, we wanted to add accommodation; one way to do that is to talk about what's going to happen, you know, a year, two years hence.

I think as we move away -- and there's an element of commitment clearly to that, right? In other words, it's pretty clear that the threshold, you know, is in some sense, an effort by the committee to commit itself to that.

As we move away from the zero-lower bound, while I think the imperative for transparency is still there, the idea that you're always in some sense going to be saying, here is where the funds rate, or suggesting, here's where the funds rate will be two years hence, that seems to be less obviously an optimal thing to do when you have room to maneuver on both sides, and in some sense, the uncertainty is more two-sided.

So that part, I think, as the committee's been clear, has been a sort of an attempt to deal with the special problems that arise at the -- at the zero lower bound.

CLARIDA: Right, right there, mmm hmm.

QUESTIONER: Thank you -- (name inaudible) -- Pace University.

I'd like your comments on alternative interpretation. The (90 ?) forecast would be on the short-term real rate of interest, 2 percent, which -- FOMC forecast, as well -- and the long-term real rate of interest at 3 percent. Is it possible that all of a sudden, the markets realized after the events of mid-May that eventually, we need to reach that and that explains the market reaction? At the end of the day, the TIPS rate in the beginning -- at the beginning of May, was around 40 basis points. Eventually, it's going to have to reach 300 basis points on the third year.

STEIN: Right, I mean, you know, you can -- but this is some sense the challenge, right, which is that that's probably a reasonable statement about eventually. What we understand less well is, you know, why does the change happen when it happens? So in some sense, it's easy -- it's easier to say, oh yeah, this is where fundamentals should be or this is -- you know, but I think -- as I was trying to stress before -- what I think is hard to explain based on fundamentals -- what is hard to explain based on fundamentals is you know, the size of the change, not the level, but the size of the change relative to the amount of news that was news about the path of policy, OK?

So again, you know, I would submit, based on my own understanding, that the news that was released about the path of policy was small essentially, you know? We basically reaffirmed the policy -- prices changed a lot, why they changed a lot now, you know, maybe people sort of -- markets converged around a new thing, maybe there was something having to do with flows and so forth. But I think it's sort of not a sufficient explanation of it to just say, well, eventfully, the rate will -- if you want to understand the dynamics, to say eventually, the rate will have to wind up -- will have to wind up there.

CLARIDA: Joyce, and please do introduce yourself.

QUESTIONER: Yeah, thank you, Joyce Chang from JP Morgan. Thank you for your comments.

I was wondering what your thinking is in what work the Fed has done on the impact of financial regulatory reform on market conditions and market dynamics and how you're viewing just the implications for, you know, the current market conditions and what longer-term implications are?

STEIN: Well, there's been a lot -- I mean, you know, there's been -- there's a lot of aspects to regulatory reform, so there's been a lot of work. And so for example one of the things that has been given a lot of attention is with respect to the liquidity regulation. You know, what's that going to do to the supply of safe securities, what's that going to do to, you know, premia on T-bills and things like that? So you know, we have quantitative impact studies of that thing.

I mean, you're asking, I guess, about, you know, the kind of thing that we hear about sometimes from practitioners which has to do with, you know, liquidity in the market now and whether flows have a bigger affect on market prices because of liquidity in the market. It is true -- I mean, it seems to be true that dealer balance sheets are smaller than they were before the financial crisis.

It's hard for me, at this point, to do any attribution to that to regulatory stuff. I mean, it may well be. I haven't seen any data that really sort of makes a -- makes a tight connection to that. But I mean, it's something we're clearly paying a lot of attention to, whether it's a regulatory affect or something else -- something else in the market.

CLARIDA: Jim. Jim Grant.

QUESTIONER: Thank you. Good morning, Governor. James Grant of Grant's Interest Rate Observer. Could you help us understand the economic difference, not the legal one, but the economic distinction between the private manipulation of LIBOR on the one hand the public manipulation of markets on the other doing business as ZIRP, QE, Twist and Portfolio Balance Channel? What ever happened to the price mechanism?

STEIN: You know, that's a hard question for me to answer because -- (laughter) -- I don't see the connection between these two whatsoever. I mean, obviously the LIBOR set -- you know, is a set of -- you know, of criminal and near-criminal activity, which is a, you know, very substantial policy concern, a lot of effort is going to go into trying to, you know, both reform LIBOR itself, look at other benchmarks, see if they're more resilient. That's a whole set of issues. To be frank, I just don't see the connection to that and the monetary policy -- and monetary policy side.

CLARIDA: Let's see. Back there.

QUESTIONER: Thank you. My name is Jeffrey Rosen from Lazard. First of all, an observation. You speak with remarkable clarity for an economist, perhaps because you --

STEIN: (Laughs.) I don't know how to take that.

QUESTIONER: -- taught at the -- (inaudible) -- school. As a compliment. The question is this: You keep -- you talk about the employment rate -- the unemployment rate as being an important data point in policy formulation, but the unemployment rate essentially has -- or the employment rate has two components -- a numerator and a denominator.

There's been a lot of commentary about the construct of the denominator and the concern that frequently it doesn't adequately take into account the number of people who have dropped out of the labor force. So I'm curious, and it's another way of saying, a 7 percent unemployment rate in one economy may be very different than a 7 percent unemployment rate in another economy. How do you take that into account in thinking about the policy formulation?

STEIN: That's a great question. So let me preface by saying, you know, the chairman in his press conference talked about 7 percent as sort of an indicative goal, and I think it's indicative in some sense for exactly the sorts of reasons -- you know, it -- one the one hand we would like to have some ability to have some specificity, right?

And to do that, you have to pick a number, even if -- you know, as I said, it's be hard to have a formula that involves the employment rate, the labor. So it's a desire to have some specificity. And maybe the unemployment rate is about as good as any single number we could come up with. But we absolutely recognize that it's not a perfect summary statistic.

And so the reason it's an indicative goal is -- I think the spirit behind is really very much tied up with what you said, which is, you know, if it's a 7 percent unemployment rate in some sense of the type that you would expect is associated with substantial improvement in the labor market, that's kind of what we're looking for.

If it's 7 percent that gets there because of sort of an aberrant move in the labor force participation rate, that's not what we're looking for. So I don't think you want to read -- again, it's an attempt to provide clarity. I don't think you want to read from that its meaning that we're going to shut out the other very sort of relevant data on the labor force.

CLARIDA: Right there, up front -- front table.

QUESTIONER: Craig Drill, Craig Drill Capital. Has the committee ever measured how good it is as a forecaster, how good it is at forecasting economic activity and inflation?

STEIN: I'm not sure that the committee has. I'm sort of aware vaguely of some academic studies that have done. And I'll take a guess, which is probably not completely accurate, but I'll take a guess. I realize I'm on the record.

I think -- put it this way: Beyond a certain horizon -- you know, maybe a very near term -- beyond a certain horizon -- let's call it a year or two -- my guess is that the Fed -- as talented as the Fed staff is -- probably does not in a meaningful way out-forecast other collection of private forecasters.

STEIN: OK. Well, of course, that -- the wear-ness of that limitation is central to the way we think about policy. That's why we do stuff in a data-dependent way. You know, we have to make a forecast; you have to make a forecast to have a sort of central tendency to think about what you're doing. But we recognize, you know, the staff is unbelievably good about showing us the large standard-error bands around these forecasts when they make them. And that's -- you know, we talked about why we moved away from calendar guidance and towards data-based -- you know, data -- it's an appreciation of exactly this thing.

So, while, again, you get in the SEP our sort of -- you know, we all have to do this. It's an exercise in humbling the governors. We each have to, like, write down a forecast every quarter and then see how wrong we are -- (laughter) -- going forward. But, you know, we understand that problem and that's the reason we're sort of trying to be data-dependent.

QUESTIONER: Wilson Ervin from Credit Suisse, good morning.


QUESTIONER: Governor Stein, you've always spoken a lot recently about financial stability, and I wanted to ask a couple questions about how we're driving and whether we've passed Chicago yet in that department. (Scattered laughter.)

Really, two questions. One, as you said, we've been making good progress in "too big to fail" but we're not satisfied yet. What two or three things would make you more satisfied in that department?

And a follow-on question: How important is that to you in the broader picture of financial stability, and are there other top-of- the-list priority reforms for financial stability beyond "too big to fail"?

STEIN: So, on "too big to fail," look, my view is sort of two-fold. On the one hand, I think we've made, really, very meaningful progress. In fact, we're -- I think we've just made public the fact that we're going to have an open board meeting next week to talk about the final -- the final rule -- Basel III rule. So, you know, if you think about the collection of things that've been done in terms of increased capital via Basel, entrain is -- the so-called (g)Civ. (ph) surcharges; stress testing, all that. That's really quite meaningful stuff in terms -- I mean, as you know better than I do -- in terms of the capital position of the largest -- of the largest financial institutions.

More -- that's (and trained ?), you know, resolution work is very -- the idea of making large institutions resolvable under Title II is another very, very important thing. And we, in conjunction with the FDIC, have been working around, you know, thinking about having a senior debt requirement that'll -- that'll facilitate all that. So, you know, I want to be sort of mindful and appreciative of, I think, the very substantial progress that's been made. At the same time, my own view -- and I've talked about this -- I don't think -- I would not characterize myself as saying job done, we're sort of finished, we should be kind of satisfied with where we are. I think there's potentially a further road to go on large institutions.

And part of this is because, you know, people talk about the "too big to fail" subsidy -- and you've seen these numbers throw around and I think Bloomberg had a number. It's, like, 83 billion (dollars) is the "too big to fail" subsidy. I think that's a part of the debate. I think it's -- maybe gets more attention than it -- than it should be. You know, even if there was no "too big to fail" subsidy, even if we completely committed -- as we will and we should -- that the government will not bail out an institution, there's still the problem that a failure of a large financial institution has systemic spillovers and consequences for the rest of the -- for the rest of the financial system. And I think we want to get to a place where we feel like capital and the other stuff that goes with it is sufficient to basically attenuate those systemic -- those systemic spillovers.

So, I've said -- and I know Governor Tarullo has said on occasion -- we should at least be open to, you know, as we get all this stuff that we've talked about done, we should be open to considering doing more. Maybe that would take the form of some kind of a capital surcharge or something like that, but I think that that's stuff that should at least be on the table. That we shouldn't, again, be in a mode that we're kind of done and we should be complacent. Then, you know, what happens aside from all of that? I think you make a great point, which is one can get a little too focused on just the largest institutions and say, well, if we sort of put enough capital and we do resolution right and we get to a situation where the largest institutions are essentially in good shape, the job is done.

As you know, a very important part of the crisis had to do with problems associated with short-term wholesale funding anywhere in the system. So, even if we've totally stabilized all the large institutions, there can be runs essentially anywhere you have illiquid securities, you know, very heavily levered against -- with short-term funding.

So the -- all the issues that go along with, you know, repo funding, you know, with -- of illiquid positions, those, you know, some of that stuff lives in the large broker dealer firms, but of course, it transcends. So I think a second very important round of work that we have to do, and it's -- you know, will move to center stage as we -- as we progress on the -- on the biggest institutions, is going to be dealing with short-term wholesale funding in a kind of comprehensive and coherent way, as opposed to just, you know, where does it -- you know, where does it reside in the biggest -- in the biggest institutions?

So I think that is -- in some sense, you've given us the blueprint for, you know, the next -- the next year's worth of work.

CLARIDA: Next question, right there.

QUESTIONER: Nancy Truitt, Truitt Enterprises.

I'm not an economist, so I may have missed something somewhere along the way, but your remarks have been positive. We're all going to reduce employment (sic), et cetera, that's the way we're headed. But I don't see where the growth is going to come from. China's not doing well; Brazil is not doing well; Europe is not doing well, and we have to sell products. So where is the growth coming from?

STEIN: OK, so I will give you just my own kind of take, again, with the caveat that I've now made a number of times, that, you know, we may be too optimistic and you may be right, in which case, you know, again, the policy will be data-dependent. If the growth doesn't come, the asset purchases will go on longer and so forth.

But if I had to tell you a story for why things will be, you know, better or you know, at least somewhat better going forward, you know, we've had now this past -- you know, we're sort of averaging around 2 percent growth -- a lot of the headwinds that we sort of saw before, Europe and others, have dissipated, but we're running really pretty strongly into the teeth of strong fiscal restraint this year. So between the -- you know, the -- some of the changes that were made at the beginning of the year with the phase-out of the Bush tax cuts and the sequester, estimates are on the order of 1 1/2 percent of GDP of headwind coming from tax.

So the fact that we have, whatever, call it 2 percent growth against that headwind, and the fact that we pretty know that that headwind will dissipate over the coming year, suggests that, you know, you just do in some sense, an additive adjustment, right? And you see the underlying strength in parts of the economy. I mean, you see the private sector sort of trying to push forward against this headwind; the housing market is clearly stronger than it was a year ago, right? And then that spills over to other consumer behavior; as I mentioned, you know, autos. The consumer generally seems to be showing some signs of -- some signs of strength.

So I think we're going to do it to some extent, internally, in -- you know, in the face of some of the external stuff. Again, we may not, we may -- we may disappoint. We may disappoint and then the policy will be -- will be have to be adjusted accordingly.

CLARIDA: All right, right there, mmm hmm.

QUESTIONER: Elizabeth Bramwell, I'm a retired portfolio manager.

I was wondering if you could explain how asset purchases are really helping small business. So you know, I live in a city, New York. It's largely geographically made up of small business, and you know, they're dependent on actually, savings more than going to a bank for a loan. So when interest rates are, you know, zero percent, it's very hard to increase your savings.

So I wonder if low interest rates, maybe we're destroying our seed corn?

STEIN: Well, so, you know, there's clearly the effect of low interest rates on savers. It's sort of important to remember, of course, that people tend to wear many hats; in other words, many of the same people who are, you know, putting their money into a CD at the bank and therefore, not earning a high interest rate, are also borrowing to buy a house, borrowing to buy a car, all that kind of stuff..

So the basic hope -- as I've said -- the basic hope is that the stimulus is sort of in first pass, is running through those sectors. And that builds a stronger economy and then the demand, you know -- I mean, you see this -- as housing goes up, so do all the tertiary industries associated with housing, you know, roofing and plumbing and contracting and furniture and all that, and a lot of that is small business, a lot of that is small business.

So it's not to say that there are some people who are not on the wrong side, in some sense, and are feeling more the saver effect than the other, but I think, you know, as you lift up the totality of the economy, you know, one would hope that for the most part, you're lifting up small business, as well.

CLARIDA: Any more -- I'm going to invoke a moderator prerogative and ask you a question. As we mentioned, a very distinguished academic career, now a policymaker, what has been the biggest surprise to you in making the transition from the chalk board to the FOMC?

I guess -- you know, I maybe underappreciate -- it's not that I didn't appreciate, but I think I still probably underappreciated the interdisciplinary aspect of it that - you know, it kind of made me wish I had been to law school and had been -- you know, spent several years on a trading floor and had spent some time in politics and all of that, especially on the regulatory side.

I other words, you know, I came in -- you know, I had done -- a lot of my own research was on banking regulation as opposed to macroeconomics. And I thought that might be, in some sense, easier for me. But when you think about regulation, not in theory but in practice, it's sort of both deeply legal and has a tremendous amount of sort of interagency negotiation, international negotiation.

All that is stuff that is essentially new to me and where I'm kind of scrambling to come up the learning curve. And I'm surprised -- I've been a little bit surprised by how much I've been scrambling to come up the learning curve on that. So that's one of many -- one of many sort of --

CLARIDA: And a related point, as a faculty member you were obviously on committees in your time. Tell us about some of the dynamics of monetary policy by committee.

STEIN: So, yeah, as an academic one of my big successes was avoiding a lot of committees. It was like all you try to do is hide under the desk when somebody asks you to. You know, I really -- I really appreciate the process, right? I mean, you've got to sort of go into this with the view that, you know, individual people are going to have strong opinions about stuff and there's going to be some collective wisdom that comes out of it. And I think I've been very sort of impressed by that part of it.

CLARIDA: Interesting. We have time for one more, right there, and then we'll finish right on time.

QUESTIONER: So you have been, I guess, the leading discussant of financial excesses as a cost to QE. And yesterday a colleague in his speech sort of highlighted that, you know, perhaps in early April we were at levels in financial markets that were perhaps a bit excessive and it was beginning to become a bit more money constrained on the program. But then, and not to penalize what was a great speech, within two paragraphs he gave fairly strong language guiding interest rates in the front end, the path of policy expectations down.

And to some extent, you know, this notion of accumulated evidence, very appropriate means of talking on volatility, but you know, at some point are you trying to have it both ways with the market, of sort of talking down the markets with regard to policy expectations, talking down the markets with regard to volatility, but then sort of asking for a reasonable amount of risk premium to be introduced into things like emerging market bonds or high-yield bonds?

STEIN: OK. So I'll restate something I said a moment ago. I said, and I sincerely believe, you know, my understanding of what we were trying to accomplish at this -- at this meeting was about kind of clarity around the nature of the -- of the large-scale asset purchase program. It was not about trying to, you know, manipulate volatility or to -- or to do something particular to the level of asset prices.

So let me clarify a little bit why I think that's the case, in spite of the fact that I sort of have, you know, reasonably strong views about financial stability. So what I had meant to say when I talked about financial stability and what -- you know, what I believe is, I do believe that in a general monetary policy framework, when one thinks about monetary policy, it belongs in the framework. One should be thinking about, what are the financial stability consequences?

So if I -- you know, abstract away from today. If I'm just thinking about, should I raise the funds rate, should I lower the funds rate? The financial stability consequences of that belong -- in my view, belong in the decision. I think that's just part of the dual mandate. In other words, we care about output and employment. We don't literally just care about the mean, we care about variability.

So for example, if I thought that lowering the funds rate was going to increase output but was going to create kind of more variability of that, that belongs in the decision. So I meant that as a sort of general proposition about our analytical framework. Now, take that analytical framework to today -- to today's environment, OK? So, you know, maybe there's a trade-off that, you know, more accommodative policy is helpful, obviously on the -- on the usual grounds, but it creates some financial stability risks at the margin.

You've got to do the trade-off. The point being, we're very far from our objective on the unemployment rate or on the labor market more generally. So when you do the trade-off, which I believe you should always be doing, but when you do it under today's conditions, I think the trade-off sort of still, you know, boils down -- comes down to one in which you want to end up being accommodative.

So I think we have sort of the appropriate -- here's a way to say it: If the funds rate could go lower, we would have it minus two, OK? Maybe if you were really concerned with financial stability you'd say, it shouldn't be minus two, it should be minus one, OK?

But we're going to be at zero either way. We're at a corner solution. So I think -- that's my -- I think you can simultaneously believe it's very important to think about this in the general framework, and we shouldn't get lazy and kind of dismiss it, because when we get closer to full employment, the balance may tilt.

But for the -- for the time being, I think the policy was sort of the appropriately-calibrated policy, and that's why, again, I think what we were -- my own interpretation is, the goal this time around was to clarify the policy as opposed to sort of modulate it, you know, left or -- left or right.

CLARIDA: OK, I think on that note, we will conclude to stay right on time. Jeremy, thank you so much -- a wonderful morning and -- (applause) --

STEIN: Thanks, everybody, thank you.

Top Stories on CFR


The partnership India has forged with the United States, Australia, and Japan appears to be gaining momentum, but some challenges remain.

Pharmaceuticals and Vaccines

The swift development of effective vaccines against the new coronavirus was an unprecedented scientific achievement. But production challenges, vaccine nationalism, and new virus strains have all presented hurdles.


Africa continues to have more peacekeeping missions than any other continent. As conflict-stricken countries increasingly look outside the United Nations for support, experts say reforms are necessary to improve peacebuilding.