The following is an edited transcript of an interview between Sam Fleming of the Financial Times and Eric Rosengren, president of the Federal Reserve Bank of Boston on Thursday, November 12, 2015.
Q. Your recent speech took a more optimistic tone. Are you feeling a little more optimistic about the economy and what are the main changes that you see?
A. The employment report certainly came in quite strong. That has been the most obvious new data, 271,000 jobs. Unemployment at 5 per cent. U6 [an alternative measure of unemployment] coming down below 10 per cent. The U3 [unemployment rate] now at 5 per cent. Frequently there are enough things between the payroll and the household that there are some things that are up and some things that are down; this was pretty unequivocally positive. It was positive after a couple of reports that hadn’t been quite as strong. And looking across industries, no matter how you cut it, it seemed to come out a good bit better. That was good news.
Average hourly earnings up is good news as well. It is early to say that we are seeing that much evidence in wages going up, but the fact that it was at least in that report was certainly a positive sign. Then when you combined that with the GDP report for the third quarter — the headline number read GDP being only 1.5 per cent did not seem particularly good news.
But when you dug into the report and took out inventories, took out net exports, we have been growing pretty consistently in terms of the domestic spending in a way that is consistent both with the recovery, and when you average the good domestic economy, despite weakness occurring from abroad both because the dollar is strong and because Europe, Japan, and many other parts of the world are a little weaker than we would like to see at this stage, but when you put the two together we are still getting enough GDP growth that it is causing downward movement in the unemployment rate, the labour market slack is definitely coming down.
Our own forecast is expecting growth a little bit above 2 per cent; we have only averaged 2.1 per cent since the beginning of the recovery and that has been sufficient to bring the unemployment rate down from 10 per cent to 5 per cent. So at that kind of growth you would hit a much lower unemployment rate unless you expected to bring people back into the labour market. That is embedded in our forecast — we should see some people that are currently not in the labour market actually come in. And that is how you square the growth a little above 2 per cent with still seeing the unemployment rate not get down so low that you would be concerned that it was going too low too quickly.
Q. So you do think there is a fair bit of potential for people to be brought back into the labour market?
A. What is striking is how much the U6 has come down. That line is actually a good bit steeper than the U3. That means we have already been pulling people in who are part time for economic reasons or discouraged, back into the labour force. That is a real positive sign. One of my concerns had been that the labour market slack wasn’t fully captured by the U3 measure. As we see that U6 measure come down, and it has been coming down pretty rapidly, I get more confident that as long as we have growth above potential over the course of next year, a lot of that labour market slack will be reduced.
Q. What kind of number represents a satisfactory reading for you on non-farm payrolls to feel that slack is continuing to diminish at a reasonable pace?
A. That is really tied to what you think the participation rate is going to be doing. It is probably under 100,000 jobs if you are assuming participation, if you are not pulling people into the labour force and just looking at the demographics, it is probably under 100,000 a month. If you are including pulling people into the labour force that is probably 125,000. Anything above that should be enough to have a gradual diminishing of the labour market slack. That being said, if we were getting reports of 126,000 that would not be giving me a lot of satisfaction. There is not a lot of precision in these numbers. I do think that the numbers if you average through what we have been seeing are substantially better than that. It shows that both the U6 and U3 measures have come down relatively rapidly.
Q. The Fed all year has been trying to shift the debate from timing of the first move to the pace. We may be on the cusp of that debate being concluded if things do go ahead in December, which your speech seemed to suggest you thought was a reasonable case. That then takes us to what we mean by gradual. What is your notion of gradualism?
A. This is where Summary of Economic Projections actually is useful. It hasn’t been completely clear that the Summary of Economic Projections [SEP] always provided information that provided additional clarity to the market. But I do think you are going to get both the range and the median view of what gradual means. December is an SEP meeting. That is going to define what gradual means among all the participants. Already, and I had this chart in the talk, relative to the 25 bps in each meeting that we had at the previous recovery, both the SEP in September — and the markets seem to think it is going to be much flatter than that; I mentioned in the talk that I thought the markets were probably a little bit closer than the September SEP. The September SEP was closer to 100 bps a year. The market is a little bit flatter than that. My own view now, and obviously it could change depending on how data comes in . . . is that the market path is probably about right. We will continue monitoring the data and if the economy speeds up and some of the international offsets become diminished then we would move a little bit faster, and if it turns out the headwinds coming from the international side, from other areas, are more substantial we can go more slowly.
A. Or a little bit lower. Whether you think the fed funds rate is going to be trading in the bottom, middle, or top of the range does matter for how you come up with these forecasts. But roughly in that range.
Q. So that would be the sort of range you would currently — not pre-empting what you would say next month . . .
A. Subject to seeing how the data comes, somewhat flatter than what we had in the September SEP would be closer to my own view.
Q. Do you think there is any merit in formalising the “gradual” expectation in the statement itself?
A. A lot of things can happen . . . you want to have monetary policy be flexible to incoming data. If the data were to come in strong you wouldn’t want to be trapped by your language. And if the economy were to weaken much more than we are expecting you wouldn’t want to be trapped by the language. So this may be an area where the SEP actually is a pretty effective communication mechanism, which is saying as of the data we have today this is what we think the path is going to be. But it is subject to change and we don’t want to be overly precise in what that is, because we don’t know how all the economic data is going to unfold. But this gives you some assurance that we are expecting it to certainly be slower than the previous recovery, and the markets seem to have gotten that message. What seems to be implied in markets is a pretty gradual increase as well . . .
Q. You are not talking about scrapping gradualism as a message?
A. Bill Dudley [of the New York Fed] was today talking about gradual. Charlie Evans [of the Chicago Fed] was yesterday talking about gradual. I have talked about gradual. There are plenty of Fed speakers that are emphasising gradual. The markets have heard that message. Our SEP gives a ballpark of what we are thinking. So between all those communication mechanisms we have the communication about right . . . My own personal view is we should have a flexible approach to thinking about the path with gradual being the important consideration, but we are still not near 2 per cent inflation. By the core PCE at 1.3 per cent we are still pretty far away. What gives me reasonable confidence about the path of inflation is the fact that the labour market slack seems to be diminishing relatively quickly. But I would want to continue to see progress on wages and prices moving up. If we weren’t seeing wages and prices moving up over time our willingness to keep raising rates would go down . . .
So it is partly conditional on whether that reasonable confidence, as your rates get higher you should probably want a standard that is a little higher than reasonable confidence. I would not expect to continue to see 1.3s for the core PCE. If we continue to see 1.3 [per cent] for the core PCE we would have to think about why is inflation not picking up towards our 2 per cent goal.
Q. So reasonable confidence should be behind us after the first move?
A. Reasonably confident applies to what it is appropriate to start raising rates. The closer you get to your target, or not getting to the target, should influence the confidence you need to be able to continue to make moves.
Q. What kind of linguistic test would you like to set next?
A. The linguistics are up to the committee to sort out, what the best communication is. As with any general statement you want something that is understandable and you can get agreement across the committee on. I am not going to guess on what that eventual language will be. We will come up with appropriate communication tool to talk about the path and what affects the path.
Q.l If we got unemployment down to 4.5 per cent or something that would be pretty low by historic standards in the US. What are the risks of going into a high pressure economy and are they less than the benefits?
A. Looking at SEP path for the unemployment rate it was 5 per cent for Q4. And then going down to 4.8 per cent and a straight line for 4.8 per cent. If you look at where the consensus view was for long-run unemployment rate it is above that. So in some sense we will be running the markets a little bit tight. That is one way you get up to a 2 per cent inflation rate from where we are now. The difference though between a 4.8 [per cent] -- and my own estimate of the full employment would be at 4.8 [per cent] -- but for someone who had 5 per cent or 5.2 for full employment that would start being a fair amount of pressure already.
So it depends a little bit on where you think full employment actually is. But I think the main advantage is it brings people that are less attached to the labour force back into the labour force. If we want to be sure U6 continues to come down, and that we pull more people in to labour force running it a little bit tighter would help. We have not seen much improvement in wage growth. Wage growth at 2 per cent seems low if you think that you have an inflation target at 2 per cent and positive productivity — wage growth should be faster than 2 per cent. We haven’t seen much evidence yet. We have seen a little glimmer of evidence in the average hourly earnings. But it would be nice if we started to see more substantive evidence that wages and prices were clearly moving up consistent with the 2 per cent inflation target. To date we really haven’t seen strong evidence of it in the data yet . . .
Q. And the risks — the flipside of that. One might be what you referred to in your speech — there are a lot of cranes going up around Boston although I can’t see any out of your window . . .
A. If you walk in that direction I guarantee you will see plenty of cranes . . . I wasn’t lying in my speech! If you walk in that area you will see a lot of cranes, that whole area of Boston is the innovation district. That area if you had come here five years ago was all parking lot. All those buildings are new . . .
It is getting at one of the trade-offs is a financial stability trade-off. I made the analogy in the speech at 10 per cent unemployment when you are greatly undershooting your inflation target, that ought to be the primary focus. As you get much closer to where you want to be for employment and inflation there is some cost to having people reach for yield and wanting to take a little bit more risk with the kind of financial positions — and this is households or firms. That means when you do normalise they are going to be more susceptible. So you want balance not only now but in the future.
Commercial real estate is a good example where you could make the case at least in some markets that it seems to be growing pretty rapidly. Boston is not a city where there is a big influx of population, so when you think about the number of buildings being built you have to figure how are you going to get enough people that will both fill the office space and fill the apartment buildings. And particularly since a lot of the apartment buildings are for people with fairly high incomes — so it is not just creating jobs, it is creating high income jobs for the kinds of buildings they are building here. So I think it is a warning sign if we allow that to go for too long . . .
Another example is the tabletop exercise that we did actually gave an example on one of the parts of the economy that was going a little bit off track was commercial real estate. Another area to look at is the shared national credit program . . . There was a release from the board last Friday — it looks at large loan participations. It looks at how many of those loans are classified and how the examiners view the quality of those loans. As you would expect, when you go into a recession the number of troubled loans goes way up. When you come out of the recession people tend to be pretty conservative in their underwriting standards, that number goes way down. What is striking is that number hasn’t gone down nearly as much this time as it has in previous periods, which is a way of saying that maybe people are reaching enough for yield that they are willing to take a little bit more risk on at least the loan participations that are captured in the shared national credit, relative to this point in previous cycles.
Some of that is tied. So it is lending in large loans, commercial real estate are two areas that I think bear some watching. It would give me a little bit of pause if we were continuing some of the trend lines to go up at the same pace. It would be a reason to maybe think about raising rates a little more quickly than I otherwise would, given the same unemployment and inflation rate.
Q. I take the message that people don’t see the policy rate as a tool to rein in asset booms or prevent financial stability problems. They see that is the realm of financial regulation. And yet you there is an argument for tightening monetary policy?
A. The challenge of financial regulation is it is not that easy to turn those knobs.
Q. Especially in this country.
A. More so in this country than in your country to be quite honest. You [in the UK] have created a governance structure where financial stability can be discussed and where there is a group that is charged with thinking about — they both have tools and are supposed to respond to those tools . . . A number of speakers at [a recent Boston Fed] conference highlighted that we don’t have that same structure in the United States. So that it depends on what sector of the economy was starting to be buoyant.
If it is an area of bank regulation, that is a shared responsibility of Federal Reserve, OCC and FDIC, particularly if it affects something other than the largest systemically important banks. So that requires a fair amount of co-ordination and Congress tends to get involved in bank supervisory issues. When the chair testifies on bank supervision issues it is a very animated discussion with members of Congress. Probably you get more questions on bank supervision than you do sometimes on monetary policy even when the hearing is on monetary policy. So it highlights that people do view the supervisory process a little bit differently than the monetary policy process, and that the independence that has been given to the Fed on monetary policy — that same independence may not be felt as strongly when it comes to supervisory policy.
So I do think our whole conference highlighted some of these issues. It is not that we can’t change things, but it probably takes a little bit more time. It probably takes getting more organisations on board at the same time. So supervision may be one of the ways you can address this concern. But I wouldn’t rule out using monetary policy tools, particularly if right now your short-term interest rate has been at zero for quite some time. It is something to at least consider as a cost to having low rates for a very long time.
We have a lot more flexibility in moving some of the monetary policy tools, so we have more ability to change it. Some of these things are interest sensitive — so with commercial real estate the kind of interest rate you are paying does affect the profitability of doing more large construction projects. I don’t want to over-emphasise — it is not the only tool but it is an important tool we should think about.
Q. So because of the institutional set-up you might want to be quicker on this than perhaps a central bank like the Bank of England which has this whole apparatus that has been created since the crisis.
A. We created the FSOC [Financial Stability Oversight Council]. The FSOC was designed to designate organisations. In some sense it was more to address too big to fail than it was to address macroprudential supervision. The approach that was taken with the Bank of England was more of a macroprudential approach. The United States has been a little bit more reticent than other parts of the world to fully integrate macroprudential tools. We are putting in capital buffers, it is not that we don’t have any tools. But I would say other counties, including the UK, may be a little bit further along in thinking about how and when to use macroprudential tools and what the appropriate governance around that is.
Q. Regarding the Fed’s portfolio; do you have a sense of how high you would want the policy rate to be before starting to think about starting runoff of the portfolio? Do you want a comfortable buffer in a sense before starting to reduce the size of the balance sheet?
A. That is the way I would describe it: I want a comfortable buffer. The logic there is that short-term rates are something we have a lot of experience moving up or down. Not that we have perfect knowledge about the impact on the economy when we move the federal funds rates. But we have enough experience that we can rely on our models to think about: ‘Gee, if we have a big negative shock how should we respond; if we lower rates this is the anticipated impact that it would have on the economy,’ and a rough idea of what the transmission mechanism would look like.
Quantitative Easing and some of the non-traditional tools we have utilised during the financial crisis and have been utilised in Europe and Japan as well: we don’t have a lot of experience with exactly how it works. I think it has been effective, I think it is an important tool in the toolbox. But given the less understanding of how exactly it works, and given we haven’t had countries exit using these tools, the preferred tool to be using to manipulate if we get negative shocks would be the short end of the curve.
That would argue for getting the fed funds rate high enough that we would be able to address reasonable negative shocks before we think about altering our balance sheet. The degree of comfort will be something the committee has to wrestle with as a committee, but I think the principle, which is we would rather get the short-term rate up so we have a tool that we have more practice working with, that if the negative shock comes in, not that I am anticipating a negative shock, but if one were to occur, that we would have the flexibility to be able to respond in a way that we would probably be more comfortable with . . .
Q. And how do you define this comfortable buffer?
A. It partly depends on what you think the probability of a negative shock is, how fast is the economy growing. There are a lot of variables that are going to go into that thought process. If you are seeing no problems at all on the horizon you may view it differently than if there is some problem that is at the forefront of your mind. At that time we can evaluate what is happening in Japan, what is happening to Brazil, what is happening to Europe. Do we think some of the headwinds we were concerned about have diminished or possibly are worse — hopefully not . . .
With relatively small perceptions in changes around China we had pretty big moves in financial markets at the end of August, so it highlights we are still in a sensitive region. It is not just the US, the UK is still at the zero lower bound, Europe is at the zero lower bound, Japan is at the zero lower bound. The ability globally to respond to negative shocks has been diminished. It is not that we don’t have tools, but I think we don’t have the tools we have been most comfortable using in the past.
So we not only want to get the US economy up and running if there is a negative shock but the rest of the world is going to want to get up and running if there is a negative shock that is of a global nature. So if only a few countries have lifted off the zero lower bound, that just means globally we have a less straightforward way to address a global negative shock . . . less firepower than we would otherwise have. We have other tools we can use but as you are aware there is some uncertainty around those tools. It is certainly globally the politics around quantitative easing programmes tend to be different than the concern people express around moving short term interest rates up or down. It is partly economic, it is partly what we know about these tools, it is partly the political economy around the tools as well.
Q. Would you be comfortable going negative [with interest rates] in the US?
A. It has been an interesting experiment in Europe. I am not sure we have enough evidence to fully evaluate the value of negative interest rates. The fact that we are talking about additional stimulus in Europe tells me it has not been fully successful. It is not a panacea, It is another one of the tools. I would hope we are not in the situation where we have to do that in the future. But we will learn a lot from the experience that Europe is currently trying this tool out.
Q. Are there US-specific reasons why it would be harder here?
A. There are some US-specific reasons. One of those would be that our money market funds are much more developed than in Europe. I know there are some money market funds in Ireland and France and some other places. But they play a larger role in the short term funding markets in the US than they do elsewhere. They aren’t the only ones affected by very low short-term rates but because they are restricted to only hold short-term assets they can’t really diversify out of that position in a way an organisation with a more flexible regulatory or internal rule setting mechanism would have. So that is an institutional difference between the US and for example Europe. There are probably some other institutional reasons that you might think would be a little different. I am not sure the institutional difference are going to be as important as just how effective is this strategy. We will get a sense with what the European experience is. We will have to think about how that would work in our own domestic situation . . . Hopefully this is far in the future — if ever. It is certainly not something that we are contemplating now.
Q. Does the divergence between policy expectations in the US and other countries — which could crystallise next month — worry you? Or does the world know this, and if there were to be ramifications around the dollar or whatever we would already have seen those?
A. One of the headwinds we are currently facing is from our net exports. That is partly a reflection of the dollar and the dollar is partly a reflection of the fact that monetary policy in the US is at a different position from in many other countries in the world. The hope would be that while we may be the first to start raising short-term rates, that other countries would in a position to follow in the not so distant future. We will see if that is actually true. It looks like we need to see more progress in Europe: the inflation data has been a little disappointing, but I would say there is some real economic data that have been a little more positive, so we will see what happens in Europe as time goes on.
I do think we will get some sense of what happens when monetary policy diverges among major developed countries. Normally we are a little more synchronous than it looks like we are going to be at this time. My hope would that some of the other developed countries would pick up a little more quickly over time and that divergence would be a temporary phenomenon. To the extent that divergence manifests itself in a . . . stronger dollar, then that means our net exports are going to be a bit weaker, so we need to have stronger domestic demand to offset the weaker foreign sector. So far our domestic demand has been strong enough to offset that weaker foreign sector. If the divergence were to become large and have a big impact on the dollar, it probably means we would follow a more gradual path than we otherwise would.
Q. This is an argument for gradualism, rather than putting off the day entirely of raising rates?
A. It is an argument for why it would be gradual. We don’t know exactly what the full impact of those countervailing forces will be. We will get a sense after it becomes appropriate to lift off.
Q. You will have the ECB policy decision before December 16 — so you could see a big new stimulus potentially.
A. We’ll see. But he [Mario Draghi] has already said that, so hopefully that is already embedded in the markets . . . A lot of that expectation will already be embedded in the markets before we have our meeting.
Q. There was a lot of criticism of Fed communications after the last meeting. Are there any lessons from that episode that the Fed could learn?
A. Many people who speak about communication have a position in the market already. And when their expectations don’t get realised they are unhappy with that. That is different from not communicating reasonably accurately . . . The way I would read what happened at end of August going through September is we got a surprising event. Globally we were seeing emerging markets much weaker than people expected, and financial market reverberations reflecting that . . . We didn’t know whether there would be spillover to the US economy. So it was perfectly appropriate to say we need more time to assess whether there is going to be more spillover. The data has come in since that meeting and the spillover hasn’t been that substantial. Our domestic economy is still quite strong. From a risk management standpoint we should react to large movements that are occurring globally . . . It is not all that surprising to me that where there is a judgment call not everyone is going to be happy.
Q. We talked earlier this year about raising the inflation target. That doesn’t seem to have gained traction. Are there any other changes to central bank mandates, strategy, that you would want to see?
A. In the short run getting off the zero lower bound here and abroad should probably be the primary focus of central banks in most of the developed world. We have yet to have a real success story with lifting off the zero lower bound and staying off the zero lower bound. My guess is in the next year that is where we should focus our attentions.
There have been a few developed countries that have lifted and then had to go back down. I am hoping that is not the case. I am hoping that the US is one of the first to start going up and that others follow and we start having a more resilient global economy. But we will see if that is the outcome . . . Let’s make sure this is actually a successful lift-off, let’s not tighten so much that we weaken the economy and create a negative shock, and lets make sure we don’t delay for so long that we get built up inflationary pressures that causes us to react more in the future.
If there was going to be one lesson I would additionally take it is if you look at the SEP for where the long run nominal fed funds rate will be, it is pretty low rate by historical standards. It is also a low rate relative to how we have reacted when there were negative shocks in a recession. If we are at 3.5 per cent on the federal funds rate that doesn’t give us a lot of ammunition if we have a shock that is the size of a traditional recession in the US. That is something we will have to think more about in time.
If the reality is we are at 3.5 per cent and that is where we are likely to be in the longer run, then the probability of hitting the zero lower bound is going to be higher than in the past. And that means some of the non-traditional tools we have used are not going to just be a feature of an unusual financial crisis, but may become more a feature of how we have to react when we have recessions . . . unconventional may become more conventional unless we end up at higher rates than what is currently being predicted.