Transcripts of the monetary policymaking body of the Federal Reserve from 2002–2008.

  • Good morning, everybody. Let’s begin our meeting by calling on Dave Stockton to report on the GDP data.

  • Thank you, Mr. Chairman. My dictionary defines a miracle as an event so improbable that it appears to defy the laws of nature. Along those lines, we distributed a GDP report that compares our forecast with the actual number that was published this morning, and they are exceedingly close. [Laughter] As you can see comparing the Greenbook and advance estimate columns, in fact, it wasn’t just close on the top line, but it was really quite close in terms of the various components. Personal consumption and business fixed investment were very close to our estimate. Residential investment was not quite so weak in the advance release as it was in our forecast; that might reflect either a different estimate by the BEA about cost per start or something they know about additions and alterations that we don’t. In the opposite direction, federal spending—in particular, defense spending—was weaker in the advance estimate than we have incorporated in the Greenbook estimate. At the bottom of the table are the price indexes. Both total PCE and core PCE were spot-on with the forecast. Really, I didn’t see anything in this report that would alter our outlook at all going forward. I would just note, obviously, that both the advance estimate and our Greenbook estimate are based on partial data. There is still a lot left to be learned about the fourth quarter, and even more about the first quarter going forward.

    One other piece of information that became available this morning was the ADP report for private nonfarm payroll employment. That report showed an increase of 130,000. Looking at that report and thinking about our forecast of 20,000, we’d probably up our forecast to about 50,000 for the month, following our normal rules of thumb in responding to that. Quite frankly, I don’t think that particular piece of information would alter my view about the state of the labor market going forward. As we noted yesterday, we have seen a fair number of other indicators suggesting that there has been some softening in the labor markets. I think that is probably still the best bet going forward, so I don’t think at this point we would change too much our employment forecast.

  • Thank you. Are there any questions for Dave? If not, Brian, would you like to talk now?

  • Thank you, Mr. Chairman. I will be referring to the package labeled “Material for FOMC Briefing on Monetary Policy Alternatives.” Your policy decision today takes place against an unusually complicated and uncertain backdrop. Over recent months, reverberations from the contraction in the housing sector have spread to the subprime mortgage market, to financial markets and institutions more generally, and now evidently to the broader economy. It seems clear, based on the Senior Loan Officer Opinion Survey and numerous other sources of information, that these factors are interacting, with mortgage-related writedowns and concerns about the economic outlook triggering a broad and substantial tightening of credit to businesses and households and the tightening of credit feeding back onto the housing market and the economy. At the same time, recent inflation performance has been disappointing, the near-term policy tradeoff appears to have deteriorated somewhat, and inflation compensation has risen. Meanwhile, investors have been rather skittish and, in the environment of heightened uncertainty, have sought more clarity in your communications and more assurance about policy prospects than you can reasonably provide. The combination of all these factors has left financial market prices unusually volatile. In these circumstances, the Committee faces a difficult task in reaching judgments as to the outlook and the balance of risks, gauging the appropriate policy course, and communicating your views to the public.

    To provide some assistance in your policy deliberations today, a box in the Bluebook explored the possible implications of addressing pronounced downside risks to growth through monetary policy, and the charts from that box are replicated in exhibit 1. The panels in the left-hand column consider the possible benefits of risk management under the assumption that a recession does develop. The recession considered in this exercise is the same as that examined in an alternative scenario in Part 1 of the Greenbook. As you may recall, that scenario was calibrated to match the typical degree of weakness relative to fundamentals in expenditures other than those for housing in six past recessions. The dotted line shows the results, as gauged using the FRB/US model, if the Committee were to respond to emerging evidence of an unfolding recession by adjusting policy in a manner similar to its past behavior, as captured by the staff’s estimated outcome-based policy rule. In this case, the Committee lowers the federal funds rate gradually to about ¾ percent by mid-2009. The unemployment rate peaks at about 6.1 percent, and inflation eventually falls below 1½ percent. The blue line shows the results that would be predicted if, instead, the Committee responded more aggressively. Here we have arbitrarily assumed that the Committee lowers the federal funds rate to 1½ percent this quarter and holds the rate at that level through the second quarter. From that point forward, it follows the outcome-based rule. If the recessionary conditions do eventuate, the timely policy stimulus moderates the downturn in activity, trimming about ¼ percentage point off the peak unemployment rate, and helps keep inflation from falling toward a region that some of you might find uncomfortably low.

    Although aggressive policy easing would help mitigate economic weakness, it would also raise the risk that policy could add unduly to inflation pressures should recessionary weakness not develop, a possibility explored in the right-hand column. The simulations underlying this column are carried out under the baseline Greenbook scenario. As shown by the blue line, the cost of an aggressive near-term easing in the absence of a recession could be limited if policymakers were to recognize quickly that the economy was not weakening to the degree feared and boosted the federal funds rate rapidly. In this simulation, the Committee increases the funds rate about 2½ percentage points in six months—that is, at a pace of more than 50 basis points per meeting—to around 4 percent. This prompt reversal brings the funds rate 1 percentage point above the baseline, providing an offset to the earlier period of excessively easy policy. In this case, core inflation would run slightly higher than baseline for a couple of quarters—as shown in the bottom right-hand panel by the distance between the solid blue line and the dotted black line—but would subsequently return to baseline. In contrast, as shown in the dashed red line, if policymakers were slow to recognize that no recession was in train and so reversed policy more gradually after the first two quarters—that is, in line with your historical behavior—considerable momentum could be imparted to economic activity, with the unemployment rate, the middle panel, running roughly ¼ percentage point below baseline into 2011. The cost of responding more aggressively to the risk of a recession in this case could be inflation that remains 0.1 or 0.2 percentage point above the baseline path for several years, the bottom right-hand panel. Concern about such an outcome could give you some pause as you consider how much insurance you wish to seek at this and coming meetings.

    Your concerns in this regard may also be exacerbated by the fact that the inflation picture seems to have deteriorated somewhat in recent months. Your current projections reflect some worsening of the inflation–output tradeoff, with more slack apparently required over the next three years to push inflation down toward what would appear to be your preferred outcomes. Also, as was discussed yesterday, five- year-forward inflation compensation increased notably over the intermeeting period. While there are reasons to question whether much of that rise represents an increase in inflation expectations, it might be due in part to a higher inflation risk premium, perhaps signaling that inflation expectations are becoming more loosely moored and are hence more prone to drift in response to various shocks.

    If you were particularly concerned about a possible unmooring of inflation expectations, you might be inclined toward alternative D, shown in the right-hand column of table 1, which is included as the next page. Under this alternative, you would hold the stance of monetary policy steady at this meeting but would indicate that the risks to growth remain tilted to the downside. However, based on the interest rate assumptions provided with your forecast submissions, the recommendations of most Reserve Banks to reduce the discount rate by 50 basis points, and your comments yesterday, it appears that most if not all of you see some easing today as appropriate. Consequently, I will dispense with a detailed discussion of alternative D. Rather, the questions for today’s meeting would seem to be not whether to reduce rates but how much and what to say about the outlook for growth, inflation, and policy.

    Under alternative C, the Committee would reduce rates 25 basis points today. The rationale section of the statement would note that financial markets remain under stress, cite the tightening of credit availability, and mention the deepening of the housing contraction and the softening of labor markets. With regard to inflation, the Committee would note its expectation that inflation should moderate, but it would also cite a range of factors that could put upward pressure on inflation. The statement would conclude by indicating that the policy action, combined with actions taken earlier, should help promote moderate growth over time, but it would also warn that downside risks to growth remain. The Committee might see alternative C as appropriate if it believes that some further easing of policy is warranted by the deterioration in the economic outlook and the associated risks but feels that a moderate move is preferable today, given the policy steps you have taken to date, the possibility that the fiscal stimulus could still turn out to be larger than currently envisaged, and questions about how firmly inflation expectations are anchored. Market participants currently put considerably greater odds on a 50 basis point move today than on a 25 basis point action. Moreover, market participants reportedly expect you to issue a statement similar to the January 22 announcement, but the addition of the sentence on inflation risks in the third paragraph would likely suggest to investors that you have significant concerns about inflation. The combination of a smaller-than-expected policy move and the statement drafted for alternative C would likely suggest to investors that monetary policy had shifted into an incremental mode. Shorter-term interest rates would likely rise somewhat, and equity prices could decline noticeably.

    Should the Committee see an outcome along the lines of the Greenbook forecast as most likely, it might be attracted to the 50 basis point easing of alternative B. Under alternative B, the Committee would issue an announcement that closely resembles the January 22 statement. However, the fourth paragraph would include a sentence indicating that the policy actions to date should help to promote moderate growth over time and to mitigate downside risks to growth. In the second sentence of that paragraph, the word “appreciable,” used in the January 22 statement, would be dropped from the characterization of the risks, presumably in recognition of the significant further adjustment of policy under this alternative. As discussed yesterday, the Greenbook-consistent measure of r* has declined more than 1¼ percentage points since the December meeting, to about ¾ percentage point. A 50 basis point move today would leave the real federal funds rate just below 1 percent, gauged on the same basis as our r* measure. Thus, by the r* metric, policy would remain slightly restrictive under alternative B, consistent with applying modest downward pressure to inflation going forward. Policymakers might view such a policy stance as appropriate if they saw the downside risks to growth as roughly balanced by the upside risks to inflation following such a policy action. However, given the views expressed in your economic projections—in particular, the assessment expressed by many that the trajectory for the funds rate would probably need to be a little lower in the near term than assumed in the Greenbook—it seems more likely that you would prefer the approach of alternative B if you believed that further easing will likely be warranted but should be implemented somewhat gradually, perhaps to provide time to assess the effects of the policy easing already put in place. With investors placing greatest weight on a policy choice at this meeting along the lines of alternative B, this approach would likely prompt relatively little reaction in financial markets this afternoon. Given recent developments, the Committee may place a larger premium than usual on avoiding policy surprises.

    If you have lowered your economic outlook even more sharply than the staff or if you see the downside risks as particularly large, you might not see moderation as a virtue and prefer to ease policy aggressively by 75 basis points at this meeting, as in alternative A. This approach would seem consistent with the concerns and policy assumptions expressed by many of you in your forecast narratives. Indeed, some of you have espoused a risk-management approach to policy in which the federal funds rate is lowered sharply in the near term to provide greater assurance of continued economic expansion and then that policy easing is reversed quickly once it becomes clear that risks are diminishing. As suggested by the risk-management simulations that I discussed earlier, this approach relies heavily on your ability to recognize promptly that the stimulus is not needed and to reverse field rapidly. Binary options on target fed funds futures suggest that, over the past few days, market participants have essentially priced out the possibility that you will take such aggressive action today. Thus, money market rates would likely decline sharply in response to implementation of this alternative, downward pressure on the foreign exchange value of the dollar could increase, and equity prices could rise somewhat, although any rally could be damped by a sense among investors that you were motivated to take this action by concerns about pronounced economic weakness. Obviously, a significant concern with a step of this magnitude today would be the possible implications for inflation expectations and inflation uncertainty. Although the relative contributions of various factors to the recent increase in inflation compensation can be debated, last week’s policy move seemed to be well understood at the time as a response to incoming evidence of a sharply deteriorating economic outlook. By contrast, the motivations for a larger-than-expected move today might be less clear to market participants, potentially elevating the risk of loosening the inflation expectations anchor. Thank you, Mr. Chairman.

  • Thank you. Are there questions for Brian? Governor Mishkin.

  • I would like to ask a question and possibly make a comment about the risk-management strategies. Am I correct in understanding that the treatment of expectations is basically backward-looking?

  • That is correct, Governor Mishkin.

  • So even though you know I am a fan of—well, let’s put it this way, you said moderation, and I am not a fan of moderation in many dimensions. But there is a potential cost here that is not articulated, which is that if there is a very aggressive easing and it has an effect on inflation expectations, then the more benign scenario that we have in terms of cost may not be there. This is one of the reasons that, if we think about a risk-management approach, we have to think about whether we can do it more systematically. What kind of communications would be attached to it? Also, what kind of information would we have to watch out for to make sure that we don’t unhinge inflation expectations? So although I think this is a terrific box and it was very, very useful, there is a bit of a qualm here. It is also very hard to model the issue of what would happen to expectations. So I am not complaining in any way about the usefulness of this; but we should have another concern, which is how we would manage inflation expectations if we pursued an approach like this. Thank you.

  • Thank you, Mr. Chairman. My question is related to what Governor Mishkin just asked about. The risk-management analysis was very helpful; it was a very good addition, and it captures a number of things that we have talked about. One is sort of a quick policy reversal or the potential for that if things go well. My question is, How do you see the path of communications likely to evolve? What is it going to look like? What will it sound like if we are thinking about reversing policy, and we get to the bottom, and then we start taking it back a little more quickly than normal? A lot of times when we have done this before you heard the phrase that markets need to be prepared for this and they have to understand it. Would that somehow get in the way of the stimulus? I think this gets a bit at what Governor Mishkin was talking about in terms of expectations. In particular, do you have any thoughts about how the slope of the yield curve might help inform how people are going to be thinking about this? Are we going to move from something that is potentially inverted or flat to a rising yield curve that could inform a better outlook, and then raising rates could maybe in a parallel fashion move the yield curve up? I am just curious whether that could be helpful for the communications strategy.

  • I think my response is that it is very difficult to say how your communications should or will evolve going forward at this point because of the very substantial uncertainty in the economic outlook, which shows up very clearly in the Committee participants’ economic forecasts. Presumably, what you say will depend on evolving circumstances. In terms of the yield curve, again, I think that is difficult. That will depend on the interaction of market participants’ perception of incoming economic information with their sense of your own policy reactions.

  • I’m just curious if, in the path of this simulation, the yield curve behaves in that fashion or not. This gets at what the mechanism is for the stimulus. Is it to improve market liquidity, market functioning, long rates, or an expectation of short rates? If they start expecting a tightening, is that going to get in the way of something, and so we are going to view that as more difficult to move at that time? I think everybody is quite committed to the idea that we need to do the right type of policy reversal when it is called for, but it is going to be very difficult to identify in the moment. So the yield curve could help.

  • In these simulations, the standard channels of monetary policy are working through things like the yield curve, asset prices, the foreign exchange value of the dollar, household wealth, et cetera.

  • Thank you, Mr. Chairman. Brian, I want to compliment you and the staff for presenting these risk-management strategies. I think it is really useful for the Committee to see and think about these in terms of strategies rather than sort of one-off policy choices, and thinking through a sequence of actions like this is very, very helpful. Governor Mishkin asked about expectations and noted that they were backward-looking, and I wanted to ask about that. You guys have described to us before that there are firms and wage setters and then there are financial markets. Is there a forward-looking part in the financial markets, or is that backward-looking as well?

  • The financial markets are forward-looking in these simulations.

  • As you know, in applied macroeconomics, there is a range of approaches to handling expectations, and the rest of the range is filled out with more forward- looking approaches. In this instance—in the instance of a recession—you hear and read a lot about recession. In fact, there was a story in the paper today about businesses preparing for a recession. So you get the sense that market participants might be understanding that this is a special episode of economic dynamics called a recession and look back to past recessions and think of it in sort of a forward-looking way. That to me, more than in the usual circumstances, heightens the value of thinking through or at least exploring the implications of a forward- looking approach to expectations. I say that in particular thinking about these strategies because we get only a few trials on recessions. They are pretty rare, and so the inferences a set of market participants are going to make about recessions is going to be influenced by their past recessions. This is going to be the last chance to influence their perceptions about how we behave in the next recession. So I am really interested in thinking through our strategy for this recession from the point of view of, well, what if from now on market participants expect us to adopt the strategy that we do adopt in this. That is why I would like to see some analysis that says, “Well, all right, if this is the strategy and it is understood by agents that this is our strategy, how does the economy behave?” So we adopt a strategy, and it is one that we think we can sustain and one that we would like to choose again in the future if it is understood.

    Related to that, Governor Mishkin mentioned inflation and inflation expectations, and I was a little confused about the discussion yesterday because I always thought of us as wanting to minimize the distortion in the rate of return on money, and that suggests minimizing the raw- return gap between nominal and real securities. So I am not quite sure that we don’t care nearly as much about the variance in inflation or the risk premiums due to inflation as we do about expected inflation. Is that your understanding when you guys think through optimal monetary policy? I know you do scenarios like that. I mean, how do the fluctuations and the variance in inflation fit in?

  • I don’t think they fit in in a direct way. We still have a lot of work ahead of us, despite the progress that we have made in modeling in recent years, to consider things like the distribution of different economic outcomes and how skewness in those distributions, both with respect to output and with respect to inflation, would interact with risk- aversion in monetary policymaking. I don’t think our science has quite gotten to that stage yet, but I agree that these are worthwhile things to study further.

  • One thought I had in that discussion about expected inflation, you noted that the one-to-five-year rise in inflation hadn’t been affected much, but the five-to-ten-year rise had. If it is well understood by agents when going through something like a recession, which comes around only every once in a while, maybe it is ratifying current inflation and not bringing it down. You might expect that the inference would change about how inflation behaves in the next recession, which would come five to ten years from now. Wouldn’t it?

  • I see the point you are making, and that sounds plausible. You know, again, our thought was that, if investors were concerned that monetary policy makers were making an inflationary mistake, it would begin to affect inflation in the near term, meaning within several years rather than seven or eight years ahead.

  • I have just a technical question on inflation expectations. You presented to us at some point a variation of the model in which the policy action itself was fed into the model for inflation expectations. Is that part of the model at this point?

  • No. It is not part of the baseline model that we are running off of.

  • That is operating here very much. It is the same mechanism, but it works differently in the two cases, because in the case where the recession actually shows up, whether you take the gradualist approach or you take the risk-management approach, they see the recession, and the easing in monetary policy doesn’t surprise them very much, although we can take our risk-management approach and respond a bit more than usual. But then, things behave as they expect, and so not too much happens to inflation expectations. In the other case, in which recession doesn’t emerge, they say, “Oh, this was an easing that wasn’t expected,” and then it becomes critical how long you hold it. If you hold it and you get rid of it only gradually, then inflation expectations start to shift up. They think the Fed’s inflation goal has changed. In the case where you take it away quickly, they say, “Okay, you took it away quickly,” and so not much happens to long-run inflation expectations.

  • So the answer is that the funds rate is affecting inflation expectations in these simulations.

  • In both cases. But the question is whether it is viewed in the context of what is happening to output and inflation, whether they say, “Yes, that fits with the historical pattern of stabilization,” or whether, because the recession didn’t emerge, this looks as though it was after the fact an ease in policy that was not warranted. Then the question is, Do you take it away quickly, or do you take it away slowly?

  • Okay. So in the case where we are doing the risk-management approach, do they understand that we are more likely to reverse course, or do they just mark down their whole expected path for policy?

  • No, they don’t know that you are more likely to reverse course in that case. So when you think of a communication strategy, which is not in these simulations, it says, “Oh, we’re going to do this, but in the event that the recession doesn’t occur, we will take it away more quickly than we might have based on historical averages.” But it doesn’t make too much difference in the simulation, because it is taken quickly. Just because the funds rate is down surprisingly low for a quarter or two, that is not long enough to have much of an effect on long-run inflation expectations in that simulation. What it would do in the real world is different.

  • We are trying to model the real world.

  • There is a “castle in the sand” quality in this discussion. Let me make just three points on what I took from this and see if I got it right. One is, in the staff’s judgment, at 3 percent—in that view of the world, which is not way off from where the discussion was yesterday—policy is not providing any degree of insurance or accommodation against the adverse outcomes that may be ahead.

    The second thing I took away is that this framework seems like a good way of thinking about how you think about moving lower, how you think about what mistake we’re going to make or we want to make. We are going to make a mistake, but it is good to think about what mistake is more costly and what mistake is easier to correct. It seems to me it tells us almost nothing, frankly, about when we are going to be in a position where we are going to want to start to take it back. It also probably tells us very little about how quickly we are going to want to take it back. I would have thought that the most remarkable thing in these paths is how little difference the variance is in the outcomes a year out. I mean, they must be so far within the range of uncertainty in this context, and for us to start thinking now about how to design the optimal exit and communicate it, not just for President Evans’s reasons but many others, just basic humility, it seems to me kind of premature. I have a lot of sympathy for the fact that we will want to make sure that we are choosing which error we are going to make. I have a lot of sympathy for thinking through how we correct for that error. But for us to assert, because we are worried about the implications of that, that we are going to start focusing now on the design of the exit, sounds to me a bit like a statement of the obvious. Of course we are going to move with alacrity to take back and make sure those inflation expectations stay anchored at a reasonable level. Of course we are. To suggest otherwise and to assert it would look kind of insecure—a lack of confidence in what we are here to do.

    The third thing I took away from it is that it seems to say that, if you make policy in a way consistent with the response of this Committee over the past decade and a half or so, you have pretty good outcomes in the strategy in which you take out enough risk against the downside error. Now, life could be uncertain. Maybe that is too charitable to the past, but it seems to be kind of reassuring in that way.

  • I think that it is hard to generalize. You say there are so few recessions, so the specific communication that is undertaken—for example, in the last episode there was actually communication that leaned against the rapid reversal—you know, “considerable period” and so on. So rather than trying to generalize from two examples, I think it would much more have to do with how we present the outlook and communicate our intentions. Other questions for Brian? Governor Mishkin.

  • Yes. I just want to make a comment about what President Geithner said and follow up on President Evans’s question. One thing that the simulation doesn’t get at is the adverse feedback loop. It’s just not built into the model. So the benefits of pursuing a risk- management approach are not as clear as they would be if they were included in the simulation. However, I would like to focus on President Evans’s issue because I think it really is critical. He asked about how we might know whether or not we should reverse and talked about things like the yield curve. But one thing that is important to recognize in this scenario is that I think it is very unusual in terms of this Committee’s past behavior. But I don’t know. Don, you are my man who is the ancient mariner here. Have we ever reversed this quickly in past episodes?

  • By “this quickly,” what do you mean?

  • What the path of the prompt reversal is.

  • I think it depends. As the Chairman was saying, there hasn’t really been a typical recession. Right? We have had two recessions in the past twenty years, and I think in both cases we stayed down longer because of the circumstances. So in 1992-93, we had a zero real funds rate while the economy was growing for two years. Then when we started raising, we did raise very quickly. So I think that was mixed. I can remember presenting to the Board, as a staff member, charts in 1990 showing how interest rates always reversed with the trough of the recession, a couple of months this way, a couple of months that way. And that was just wrong. I mean, it was just a different circumstance because of the 50 mile an hour headwinds. So it is going to be very hard to set expectations about what we are going to do that will influence people in the future because each episode is so different. As the Chairman was saying, we were very worried in 2003-04 that people would generalize from 1994 and see us reversing rather quickly, so we got into this verbal gymnastics to try to modify their expectations because we thought the rapid reversal in that case would not be consistent with macroeconomic stability. This is a case in which we are just going to have to judge when we get to the circumstances what is necessary without knowing ahead of time what the pattern is going to be.

  • So let me turn back to President Evans’s question because there is an issue about what kind of information might tell us that we took out insurance and then it wasn’t needed. I think some pieces of information that the staff has been looking at are relevant. Particularly important in this regard is that there are a lot of indicators of financial stress. What is different about this episode is that the reason we are so worried about downside risk is the financial strains, and there are a lot of pieces of information that the staff has been producing that could be very important in terms of looking at this. They have a financial—I don’t know what you call your indicator, but you have something that you have been using in your model, Dave, that takes up spreads and volatility and—

  • Principal components.

  • —principal components of these things.

  • There are a lot of financial variables.

  • Also an issue that the Chairman raised yesterday was that the housing market is a big component of our downside risk. The market’s concerns about future declines in housing prices are causing a very sharp decrease in demand for housing. That could turn around very quickly as well. Even now we should be thinking about these issues, and Governor Kohn’s use of the word “nimble”—I like “flexibility,” but I think “nimble” is probably a better word—is really I think key here. It is somewhat of a departure from normal—exactly what the Chairman said. This episode is different from past episodes. So we do need to start thinking about this, and the staff will need to think about exactly these issues.

  • President Poole, you had an interjection?

  • Yes. I am older than Governor Kohn, so I can answer. [Laughter]

  • It is not age. It is how long you have been in the System.

  • 958 was just about like that.

  • Like that prompt reversal.

  • It was very much a sharp V.

  • I just wanted to point out that the reason I asked the question about how you think about the exit strategy is that, since we are putting a lot of discussion weight on the reversal, it is going to color how you think about the next move potentially, if not today then another day. So I think it brings the future into the present very quickly.

  • If I could just continue the dialogue with Governor Kohn. It has been observed by some that the past two recessions differ from previous post-war recessions in some key dimensions, and some have drawn the conclusion that it is the changing nature of the business cycle after the Great Moderation. Both have been characterized as shallow and longer lasting. I take the point that one’s inferences about a third instance are going to be very tenuous, naturally. But it is also natural to look back to those episodes and see what we can draw out of them. You mentioned headwinds in 1990, and we had some special stuff that kept us from raising rates with nimbleness in 2003 and 2004. To what extent do you think that is likely coming out of this episode, that the types of things that kept us from raising rates in the 1990s and in both of these episodes are likely to recur?

  • Well, I asked Governor Kohn, but if he gets to respond.

  • I think we are really getting into the policy discussion right now, and the issue, really, will be the headwinds, the financial market restraint. So it is reasonable to think that we are facing at least 25 mile an hour headwinds. I also think that it is going to take a while for the headwinds to abate. It is going to take a while to rechannel these credit flows, to restart many of the secondary markets in mortgages. So my personal view is that I wouldn’t be surprised to see us having to run with a real federal funds rate that is below the historical average for some time. But as everyone has been saying, it is highly uncertain and very dependent on how the financial markets evolve and how the real economy responds to those markets.

  • For follow-up before President Geithner responds, to the extent that we place greater likelihood on that, should these simulations suggest that we should take that into account in how rapidly we cut now?

  • I don’t think so. I would just point out that with the 1990-91 episode, inflation after the recession—after the whole episode was over—was significantly lower than in the period before it. So if there are headwinds that are bringing the economy below potential and are causing high unemployment, for example, there has to be a mechanism. Expectations have to be tied to something, and there has to be some way in which excessively low interest rates are stimulating inflation, through either commodity prices or wage pressures or some other mechanism. If those pressures aren’t there because of headwinds or some other factor, then unusually low interest rates will not by themselves create inflation. Vice Chairman, did you have anything else you wanted to add?

  • I really couldn’t tell, President Lacker, what inference you were going to draw from that. But I would just reinforce the point that, if you are more worried and uncertain now about the magnitude of the headwinds and the duration, I think it has to mean that you err on the side of going lower sooner. But the main point is that we just don’t know much about it, and I think it is worth a lot of humility. I mean, think how surprised we have been by so much over this period, even with all our thinking through three years ago about alternative paths for housing. So I would just vote for humility. But the basic point is that we have to err on the side of being worried about reducing the risk that you end up with 75 mile an hour headwinds rather than 25 for a long period of time.

  • Could we start our policy round? President Pianalto.

  • Thank you, Mr. Chairman. Like the Greenbook, my projection for the real economy incorporates a sharp decline in the equilibrium real fed funds rate. Given the large risks facing the real economy, I think we need to take precautions against having a restrictive fed funds rate target. I think a 50 basis point cut in the target fed funds rate today may be large enough to eliminate that possibility, although there is plenty of uncertainty around that estimate, as we have been discussing. Based on my analysis, comments from my business contacts, and what I have heard from all of you at this meeting, I feel very comfortable supporting this position today. The economic environment has been volatile and highly uncertain, and I realize that my outlook could change appreciably in the weeks and months ahead. I can imagine that my economic projections will evolve in a way that supports even further reductions in the fed funds rate target.

    At the same time, as I said yesterday, I am concerned about inflation risks and that they may now be elevated. I can also imagine scenarios that would lead me to want to pursue a more restrictive policy than would be appropriate based on the downside risk to growth alone. At some point, on the margin, inflation concerns could become my dominant concern. We know that inflation expectations play a crucial role in determining the inflation outlook. We have been talking about that. But, unfortunately, we don’t really know all that much about what it is going to take to unanchor inflation expectations. It is hard to know for certain how far out on the ice we can skate without needing to worry that the ice has become too thin. I know that we are bringing our best thinking to bear on this issue by developing diagnostic tools such as the decomposition of inflation compensation into its component parts, as we talked about yesterday and this morning. I hope that we are going to come to learn that these tools are useful guides to policy, but we just don’t have enough experience with them to know how much confidence to place in their estimates.

    Yesterday Governor Kohn told us about his conversation with Paul Volcker and that Paul Volcker told him that unfortunately we do have experience of seeing the erosion of public confidence in our ability to meet our price stability objective, and we know from this experience that it makes the attainment of price stability more costly. But today I support the policy directive expressed in alternative B. My concerns today are more with the downside risk to economic growth. Given what I know today, I think it is the right course of action. I have discovered during the past couple of weeks that I can be very nimble when it comes to the reduction in our fed funds rate target. If inflation developments require, I want to be just as nimble in the other direction. Thank you, Mr. Chairman.

  • Thank you. President Hoenig.

  • Thank you, Mr. Chairman. Given the discussion I have heard so far, it is apparent to me that we will most likely move to a 3 percent fed funds rate today. If we do so, I think we should be prepared to stop at this point and let the market work through the disturbances that only time and market adjustments will correct. While I would not pre-commit to no further cuts, I would give no hint of them either. I think that is important. The discussion we just had around this table is that we cannot know the future, there are always headwinds, and we will make a mistake. Therefore, we should limit ourselves to some extent. As I have expressed in this Committee before, I think we should narrow the range to which we are willing to move the fed funds rate. My judgment is that to move to levels outside the reasonable range that, yes, I know would be debated, but a reasonable range of neutral, is as likely to introduce disturbances into the economy as it is to moderate the economy’s performance.

    I find an approach to policy that focuses on the real equilibrium rate compelling. I think it is good theory, and it is common sense and can be explained to the public. I understand that we cannot observe the equilibrium rate, but we cannot observe the NAIRU, potential GDP, or most other criteria. The real rate is at least related to the policy tool that we focus on, which is interest rates. When you systematically move below or above the neutral rate, you systematically stimulate or restrict the economy. With that in mind, and since monetary policy works with a lag, I would recommend that we follow a policy in which we stop moving rates and let the market work once we are notably below or above the long-run neutral rate based upon a reasonable estimate. This is a credible approach to policy. Today, a nominal 3 percent fed funds rate would provide for an actual inflation-adjusted rate that is less than 1 percent. That is quite stimulative. That will serve to support the real economy as the financial sector adjusts, and it has to adjust. To stimulate the economy further at our current inflation rate is to invite, or at least to increase the probability of, higher inflation or encourage the next asset bubble or both, and it will undermine our credibility in the long run.

    Turning to the press statement, I have three brief comments. First, in paragraph 2, I would begin by putting the economic rationale before the financial market reference, which would most appropriately align our recent actions with our concern for the downside risk to the economy and place our concern for financial market stress in better perspective. Second, I believe that a detailed discussion of inflation risk may not be helpful at this time, and the less said the better. Therefore, I prefer the language in paragraph 3 of alternative B rather than any other of the options. Finally, I believe we should modify the risk assessment to provide more flexibility for future decisions. Most important, it should place less emphasis on the possibility of future rate cuts and better position us to return to a neutral position later this year and early next. Therefore, I would rewrite paragraph 4 along the lines that President Plosser suggested. To summarize, I suggest that we act at this meeting, explain our action as I suggested, and hold at that point. Thank you.

  • Thank you. President Rosengren.

  • Thank you, Mr. Chairman. I support alternative B, though I think a case can be made for alternative A. The Boston model indicates that even after a 50 basis point reduction, we still need more easing to return to an economy with both full employment and inflation below 2 percent. Taking out insurance against more-severe downside risks would imply even more easing than our baseline forecast. Given our recent move and the additional easing in alternative B, I am comfortable waiting to take more aggressive action only if incoming data warrant it. However, I will not be surprised if we find further action is indeed needed.

    What would be the arguments against taking an aggressive tack? Certainly, one argument might be that elevated oil and commodity prices and core inflation currently above 2 percent warrant a more restrained approach. However, I would note that in previous recessions the inflation rate has declined significantly, even in the 1970s, in the midst of historic surges in energy and food prices. Whether we skirt a recession or experience a recession, I expect core inflation to trend down. A potential second argument is that we have responded too slowly to the need for tighter policy in the past, so we should be more reluctant to ease in the present. While it may be true that we raised rates too slowly at the onset of previous expansions, I see no reason for this Committee to behave in a manner that it believes is suboptimal. As a Committee, we seem to have consensus on the importance of maintaining low inflation rates, and I am confident we have the will to raise rates with the same alacrity that we reduced them, should economic conditions warrant such action.

  • Thank you, Mr. Chairman. Since September, this Committee has lowered the federal funds rate 175 basis points. My estimate is that the real funds rate before any action today is 1 percent or slightly below that, and that is very low by historical standards. The slowdown in growth suggests that the equilibrium real rate really has fallen, and the Committee has appropriately allowed the nominal funds rate to fall as well. Do nominal rates need to go lower and, if so, about how much? Part of this depends on what you think the equilibrium real rate of the economy is now. The Bluebook indicates that estimates of r* vary considerably by the model and the process they use to calculate them. The 70 percent confidence interval around them is 3 percentage points. Moreover, the estimates of r*, as we talked a bit about yesterday, can be quite volatile. It troubles me that the estimate of r* consistent with the Greenbook has changed by 140 basis points from December to today. I am uncomfortable using an estimate that is so variable and so sensitive to stock markets as a guide to setting policy. I also want to note that the Bluebook indicates that the appropriate funds rate, based on a range of Taylor rule specifications, is anywhere from 40 to 120 basis points above where we currently are today. That includes forecast-based versions of the rule that rely on the weak forecast found in the Greenbook.

    While I don’t want to suggest that such guidelines are definitive, they do suggest that the current level of the fed funds rate is clearly accommodative and that we have taken out insurance against downside risk. When do we stop taking out more insurance? If we do cut 50 basis points today, which is the amount the market is expecting, it would bring, to my mind, the real funds rate down to below ½ percent. That is based on expectations of about 2½ percent inflation, which in fact may be conservative. To my way of thinking, that is a very accommodative policy by any standard. Moreover, I don’t believe that enough time has elapsed for us to realize the full effect of the cuts that we have already put in place. I share President Hoenig’s concern that only the market can solve many of the problems that we see out there, and we must give the market time and patience to do so. The last time real rates were this low was in 2003-04, when the real rate was in fact apparently negative. But that was different. Inflation was running around 1 percent or less, and our concern was possible deflation. Today, we are not worried about deflation in the near term. We are worried about inflation; inflation has been moving up. Lowering rates too aggressively in today’s situation would seem to me a risky strategy, fueling inflation; possibly setting up the next boom-bust cycle, which I worry about; and delaying the recognition of losses on bank’s balance sheets but not eliminating them. The main effect of the rate cut will be after the first half of the year, if the economy begins to recover. I think we need to be very cautious not to get carried away in our insurance strategies with lowering rates too much. In my view, we are on the verge of overshooting, and I worry about the broad range of consequences for our credibility and the expectations of our future actions such behavior may have. That is closely related to President Lacker’s comments about what people interpret that behavior to mean about what we may do in future episodes. But two things are even more important, in my mind, about what we may do and what we do today.

    First, we need to be very careful about our communications and not to excessively reinforce the market’s expectation that further rate cuts are coming. In particular, I would feel much more comfortable with supporting a moderate 25 or 50 basis point cut if the statement language today were more agnostic about the balance of risks, as I suggested in my memo before the meeting. The market interprets our saying that there are downside risks to growth as that we are planning to cut rates again. I do not think we should encourage those beliefs. I worry that the balance of risks portion of our statements has come to be a code for predicting the path of our federal funds rate. I think that is not a good position for us to be in, nor should we condone it. Given the Greenbook forecast, I don’t believe that negative real rates are called for, and signaling further cuts clearly sends the message that negative real rates are on the way, if not already here. When our forecasts are released, the public will get our assessments of the risks in our outlook. We don’t need to say anything more about it in the statement. That, of course, does not preclude us from cutting rates again if our forecast deteriorates further. But until it does, I am reluctant to encourage the perception that more rate cuts are forthcoming.

    Second, as I said in the last go-round, we need to be able to better understand how we are going to unwind these cuts that we have implemented as insurance against the macroeconomic effects and financial disruptions. Of course, this was the theme of the discussion we had before the go-round. Unwinding those cuts too slowly not only risks our credibility on inflation but also risks setting up the next boom–bust cycle. Hindsight, of course, is always 20/20, but as we discussed in Monday’s videoconference, the Fed’s being slow to raise rates back up after the deflationary scare was over in 2003 may indeed have contributed to the conditions we are facing today. Thus, it is crucially important, to my mind, that we do have a plan for unwinding the significant cuts we have implemented as insurance against the financial turmoil. If the market turmoil subsides, I believe this Committee needs to have clear signals as to what we are going to look at and what has to happen before we start to remove the accommodation. I believe that the Committee must undo the accommodation as aggressively as we put it in play. We need to determine what indicators we will be looking at to determine when that process should begin. When we know ourselves, we want to help the markets and public understand what our process will be as well. I strongly believe that we must be both credible and committed policymakers, and our communications must signal not a particular funds rate path but articulate and focus on the contingent nature of that path and help the public understand and appreciate the systematic part of our policies and our policy decisions. Thank you, Mr. Chairman.

  • Thank you. President Lockhart.

  • Thank you, Mr. Chairman. I think the best course of action today is to lower the fed funds rate by 50 basis points as in alternative B. I’ve heard cogent arguments that 50 basis points would be restrictive and likewise accommodative. Yesterday I had a chance to look at the disabilities-related display in the elevator lobby on the Concourse Level, and I took some comfort in the fact that many great people are or were bipolar. So whether it’s restrictive or accommodative, I can be convinced either way. I can live with the idea that this does not incorporate a great deal of insurance against the downside, provided that the language doesn’t preclude further timely action as we have been using the word “timely.” So regarding paragraph 4, my preference is for language that, first, serves to signal that the Committee is fully aware of the situation and not behind the curve, a signal that’s reassuring and confidence enhancing, and, second, preserves our nimbleness, our flexibility, meaning that it doesn’t preclude any options to respond to developments, including further moves and intermeeting actions. I think the straightforward phraseology “downside risks to growth remain” as opposed to what was suggested in alternative A, which included “may well,” is preferable because the alternative A language strikes me as a bit too clever and risks appearing out of touch. I also think that repetition of the “in a timely manner” language from the January 22 statement preserves the options to move in an intermeeting action if necessary. So I think alternative B language is acceptable because it is spare, straightforward, and familiar, and it’s the least likely to confuse the markets in the near term. Thank you, Mr. Chairman.

  • Thank you. Governor Kohn.

  • Thank you, Mr. Chairman. Like President Lockhart, I agree with the action and the language of alternative B. As I noted a few minutes ago, I don’t think a real interest rate of around 1 percent, which is where we would be after this action, is really all that aggressively low, given what we’re facing in financial markets and the uncertainty that’s constraining business and household behavior in terms of spending. I guess I wouldn’t obsess so much—a loaded word there—about where we are exactly relative to some estimate of the neutral federal funds rate. As President Plosser pointed out, our band of confidence around that is huge. It moves around a lot when it’s defined in the short-term, three-year window that the staff has been using, but I think we just need to concentrate on the forecast, where we think this path of rates would have us go.

    The forecast central tendencies that we looked at yesterday look like a pretty good set of tradeoffs between getting back to full employment and damping inflation. I think one message from this longer-term forecast exercise is where the Committee thinks the Taylor curve is and where we would like to end up on it in terms of trading off inflation and output variability. If I heard Brian correctly, these results encompass not only a fairly universal assumption of 50 basis points at this meeting but at least ten of the cases see further rate cuts after that. In my case, I assumed further rate cuts but that they would start to be taken back in 2009. So it seems to me, heading where we’re heading, and maybe even moving a little lower if circumstances permit over future meetings, is perfectly consistent and is consistent in the view of the majority of the Committee with some pretty reasonable outcomes for the economy given the shocks we’re facing and the circumstances we’re in, whatever it implies for where we are relative to some long-run neutral real rate that might pertain over ten or twenty years. Implied by that—and the Vice Chairman said this a few minutes ago—is that it’s not clear that the 1 percent real rate has much, if any, insurance built into it relative to the kinds of headwinds we’re facing. I agree that a 1 percent real rate is not sustainable indefinitely, but as I pointed out before, we ran with a zero real rate for two years in ’92 and ’93. I don’t think that had any adverse effects on inflation expectations at the time. We explained why we were doing it. We explained the circumstances that were forcing us into that. We kept our eye on inflation expectations. You might remember that at the time there was a lot of political pressure that we should lower the rate even further, and we resisted that pressure in part by keeping our eye on inflation expectations and making sure they weren’t moving higher.

    So if circumstances dictate, I think we could sustain, as the Greenbook has us doing, a 1 percent real federal funds rate for some time without any adverse effects on inflation expectations going forward. I don’t think 50 implies an unacceptable inflation risk. I think it’s consistent with the gradual reduction in inflation that we’ve outlined in our forecast. In these circumstances, we need to concentrate on addressing the economic and financial stability issues that we’re facing. That’s the bigger risk to economic welfare at this time than the risk that inflation might go higher, and the 50 basis points in my mind is just catching up with the deterioration in the economic outlook and the financial situation since the end of October. We are just getting to something that barely takes account of what has happened, with very little insurance. Now, I agree that if we got into a situation where we went lower, then these subtle tradeoffs between risk management and the inflation outlook would come into play, but I don’t really see them in play at the level of interest rates that I’m suggesting we be at at the end of this meeting under the current circumstances.

    I do agree with President Plosser. We need to think about the circumstances under which we would begin to take back the easing. In the staff forecast we don’t have to think about this for two years, if they’re right. But they may not be right, and if we go further and have insurance, then I think that’s a more important issue. One point that I took from President Evans’s discussion was that there might be a risk in talking about taking it back right now because it would undermine the effects of the ease you put in place. So as a Committee we need to think about the circumstances. It’s a very subtle and tricky issue. The Chairman can perhaps cover this in his testimony, and obviously we’ve talked about it. It must be reflected in the minutes. But going out front with hammering in public how we’re going to take it back is going to undermine the effects of the ease itself. So we can talk about those circumstances, but I think we need to be careful about overemphasizing the “taking it back” idea, particularly from the current level.

    I’m comfortable with the language of alternative B. I think the first sentence of paragraph 4 does help to say that we think we’ve done something considerable that’s going to be helpful. It is a change from the last thing we put out. Removing “appreciable” in terms of downside risk is also a change, and I agree with that. But I do think there are downside risks even after we move today, and it would be a mistake to avoid that topic. Those risks are still going to be there, even after the funds rate is 3 percent, until the financial markets begin to stabilize and for more than a few weeks. We had a bit of a head fake in October, right? They seemed to be stabilizing. We said the risks were roughly in balance, and then the financial markets collapsed. I think we need a period of stable financial developments so that we can gauge the effect on the economy before we go to a balanced risk statement. Right now the risks are still tilted toward the downside on real activity, and that should be our focus. Thank you, Mr. Chairman.

  • Thank you. President Yellen.

  • Thank you, Mr. Chairman. I support alternative B in the Bluebook. If we cut the federal funds rate a further 50 basis points today, we will have done a lot in just a week and a half, but I think these actions do represent an appropriate response to a substantial deterioration in economic conditions. As I said in my comments on the economic situation, I basically agree with the Greenbook forecast for this year and perceive the risks to be to the downside. With the fiscal package, a funds rate of 3 percent will likely promote growth in the second half of this year that’s moderate after a brush with recession in the first half. I’m comfortable with this action because I believe our inflation objective is credible, and I do have confidence that we will be able to reverse the accommodation we’re putting in place when it’s appropriate to do so. But our discussion does highlight the important point that alternative B seems to be appropriate monetary policy in the context of the modal forecast. It just brings the real funds rate down to the Greenbook estimate of neutral, around 1 percent. If the economy were to go into a recession, additional easing would be needed and would be appropriate. Also, as Governor Kohn and others have emphasized, alternative B still doesn’t seem to incorporate much of anything for insurance against recession. So, indeed, there is a case for doing more than B. There is a case for A, but I wouldn’t go there today. I think we can wait, and I think we can watch as developments unfold and monitor data.

    With respect to language, the skew in the risks toward a downside surprise and the possible need for insurance against that possibility, especially if we see some further deterioration in financial conditions—that strongly inclines me toward the assessment of risk sentences in alternative B with their asymmetry toward ease. We need to be absolutely clear, to state clearly today, that we recognize the continued existence of downside risk and communicate that we stand ready to cut further if necessary. Therefore, I would definitely retain the sentence in alternative B, paragraph 4, that states, “However, downside risks to growth remain.” However, I could see a case, following President Plosser’s suggestion, to substitute the wording from the last sentence of the December 11 statement to the wording in the last line in alternative B. That is, we could substitute the words “will act as needed to foster price stability and sustainable economic growth” for the words “will act in a timely manner as needed to address those risks.” This seems to me to be a small change that would, taken together with the new first sentence in paragraph 4, slightly dial down the perceived odds of further cuts relative to the proposed wording in B. Even without the change that President Plosser suggested, though, the fact that we have added the new first sentence in paragraph 4 does seem to me to change the wording of the assessment of risk enough relative to our intermeeting statement to communicate to markets that we will view future policy moves after the one today somewhat differently going forward. Today’s move and the intermeeting move are essentially catch-up, to put us where we think we need to be, and moves going forward will respond to the evolution that we see in the markets.

  • Thank you. President Poole.

  • Thank you, Mr. Chairman. I can support alternative B, down 50 basis points, although as I’ll try to argue in a moment, I don’t think it makes a whole lot of difference whether it’s down 50 or down 25. I would support the Plosser recommendation on paragraph 4, and I would also note that the statement as it’s written recognizes the credit market problems and the housing problems, but it seems as though a critic from the outside might say it’s oblivious to the fact that inflation rose somewhat last year. So I would propose adding a sentence at the beginning of paragraph 3 along the following lines: “Although the inflation rate rose somewhat over the past year, the Committee expects inflation to moderate.” It seems to me that we should recognize the fact that inflation rose because it’s parallel with the recognition of what has been going on in the credit markets and in the real economy.

    Now, let me make a couple of fairly general comments. I’m very much of the view that the natural state of the U.S. economy is full employment and output growth at potential. That’s where the economy tends to gravitate, and firms and markets respond relatively quickly on the whole to shocks. A particularly good example was the shock of September 11. In the following three months, the companies shed a million jobs out of nonfarm payrolls, responded pretty quickly, and slashed production inventories, and in fact, it was that shock that ended up creating the case for that being labeled a recession. Without that, I don’t think it would have been labeled a recession. Anyway, the shock last August did not lead to a sharp retrenchment in the real economy. Putting housing aside now, it aggravated the housing problem that was already under way. Firms and markets are making many necessary adjustments. Housing starts are down. House prices are falling, which I think they have to do. Banks are raising more capital. Risk spreads are rising from abnormally low levels, and lots of other kinds of adjustments are occurring that need to be made and are ongoing. I just do not see in the data today the sort of “in your face” data suggesting that the economy is in recession. It might happen, but I don’t see it right now. We may not have the luxury of a policy that avoids recession. We don’t know that. Our choice may end up being whether we accept a mild recession this year or whether we follow a policy that is so expansionary that we end up with some further increase in inflation and a deeper recession later. I have no doubt where I come down. I would rather take the risk of a mild recession now rather than the risk of an increase in inflation and a larger recession later.

    Now, I said that I didn’t think that it made all that much difference whether we choose down 50 or down 25. A number of people have commented on the critical role of communications going forward to try to stabilize the situation, particularly the market’s expectations about the Federal Reserve’s monetary policy, and the burden here is going to fall primarily, 98 percent worth, on the Chairman. I think that we need heightened attention to longer-run concerns. So much of the market commentary seems to revolve around the possibilities of a recession, recession, recession. I don’t see very much attention to longer-run concerns. I think those longer-run concerns are important, and we need to emphasize them. I think monetary policy needs to be based on our best estimate of what is happening. We spin out alternative scenarios so that we are prepared for things that might happen, but the actual policy decision has to be based on our best guess as to what is, in fact, happening. I think that we are at risk that inflation expectations might rise. We monitor them closely, but once we start to see inflation expectations rising, it’s going to be difficult and costly to rein them in. It’s going to create a big problem for us. So we can take some comfort that it hasn’t happened yet—I don’t think it has happened—but that doesn’t say that we can leave that issue alone. I think that the recent policy actions and policy statements have not adequately balanced the near-term concerns over economic weakness or potential weakness and the longer-run inflation risks.

    Now, this is the question. Should we try to indicate to the market that it is now time for a pause in the funds rate target changes, assuming that we go down 50, given that we’ve now come down 225 since September? Should we try to give that direction? Well, let me recount a bit of history here, and some around the table, particularly Don, will remember this and may correct what I’m saying if I have the details wrong. The FOMC has really been struggling with this question of forward guidance since it first began to issue statements at the conclusion of every meeting. The first time we did it, we were using the bias or tilt language. The very first time we did it, there was a reaction in the marketplace that we regarded as excessive and undesired. That led to the communications committee that Roger Ferguson chaired. I sat on that committee, and Don was fully involved with that effort, which led to the balance of risks language. Now, my judgment is that, over time, the forward guidance that we’ve used in that balance of risk language has not been successful. There’s always a problem of putting the language in, and there’s always a problem of taking it out when you want to take it out without giving impressions that you really don’t want to give. We have always understood around this table that incoming data can change the situation, but the market doesn’t know when the incoming data will trump the guidance that is in the statement. Hence, I think that the statement or a more formal tilt or balance of risks language has created some of our communications problems. That was clear last fall, when we tried to give an indication that we thought we had done enough for the time being and it was overtaken by incoming data, and that created a lot of problems in the market in understanding where we were coming from. So it seems to me that we’d be better off focusing on as clear an explanation as we can of why we have taken the actions that we have. That’s the best chance we have of providing guidance to the market of how we might respond under similar circumstances in the future that we can’t predict. I think we need to concentrate on explaining the policy responses. The Committee’s task is—and again, as I emphasized, I think it falls principally on the Chairman—I’ll say to “restore,” because I don’t think that that’s an unfair word to use here, a greater degree of policy predictability. That cannot mean unconditional because we don’t know how the data and how events are going to move the world that we have to respond to. But we need to create a better understanding in the market of how the policy changes are conditional on incoming information, given the policy objectives that we share and that are given to us by the Congress. Thank you.

  • Thank you. President Evans.

  • Thank you, Mr. Chairman. I favor 50 basis points today. I think that a 3 percent funds rate may be reasonably positioned for the economy given our forecast. That’s the Greenbook path assumption. That certainly seemed reasonable, and today’s data gave us no indication that reasonably changed that path. I, like President Geithner, did find it remarkable that risk-management paths were not really that different either way. Now, surely risks abound, and it’s tough to capture that in models, and I’m all in favor of humility. I just don’t know which direction humility argues for in this case—more action or less action. Anyway, those are my preferences. We have talked a good bit about the potential of reversing our policy action sooner than we often have done in the past. I just think that it’s going to be difficult in the moment. The Bluebook path that I’ve been looking at has this reversal starting when the unemployment rate is peaking. I think it’s unknowable that it is actually the peak. There will be a lot of uncertainty, but still, that’s the situation that we face, and so I sort of accept that. But because of that, I would prefer to be a little more careful about how far down we go. Three percent could be the bottom of a funds rate cycle. It’s very difficult to know. From here on out, I’d like to be hopeful for positive news but mindful of further deterioration that we would certainly act on.

    In terms of the statement, I guess from some discussions over the weekend, I wasn’t expecting to hear much movement in that. But the suggestions of President Plosser and President Yellen seem to make a lot of sense from my standpoint. So I think it would be nice if there were a little more discussion of that, and I thought President Poole’s suggestion on inflation was also helpful. Thank you, Mr. Chairman.

  • Thank you. President Lacker.

  • Thank you, Mr. Chairman. Our Vice Chairman urges humility. I strongly support that. I agree with President Evans that it’s not obvious that the greater one’s humility, the greater one should favor ease. I think we should be humble about the path of inflation going forward, whether it’s likely to fall on its own. I think we should be humble about our understanding of the output gap. I think we should be humble about whether that’s even a sensible way to think about how real and monetary phenomena interact. The Phillips curve itself embodies a relationship. It is uncertain, but it embodies expectations of our future behavior. I think we should be humble about what those expectations are in the present circumstance. Times in the past when we’ve gotten in trouble on inflation have often been when we were oversolicitous about weak economic growth, and I think we should be humble about whether we’ve completely gotten past those inflation dynamics or not. I think we should be humble about the willingness of our future selves to reverse course, and a lot has been said about that. For some it seems to counsel greater ease now, but I think the opposite argument can be made that the extent to which we think we may be hindered or feel impeded in raising rates, even if we think it’s warranted in the future, should cause us to be more cautious about lowering rates now. It always seems in recoveries that there’s always something that looks fragile, that looks likely to threaten economic growth. You know, one month it will be the commercial paper market and CDO writedowns, and this month it’s monolines. There will be headwinds. I predict we’ll be talking about headwinds a fair amount in the next couple of years. But if those are genuinely going to impede us, we need to be realistic about that, and I think we need to take it on board now. I agree with President Poole. We need to be humble about our ability to prevent a recession. I think we should also be humble about the extent to which what we see in terms of both growth and financial markets is presumptively inefficient and needs remedial action on our part. You spoke several meetings ago, I think a year or two ago, Mr. Chairman, about our need to retain a concern about inflation but not be seen as inflation nutters. I think we need to care about financial fragility but not be fragility nutters.

    If left to my own devices, I could have chosen a 25 basis point cut today. I recognize that would surprise markets, but judging from the policy paths that you all have submitted for your projections, a whole lot of us think that we’re going to have to surprise markets some time in the year ahead. Having said that, I can support a 50 basis point cut now. But more broadly I think inflation is just going to have to keep influencing our policy path. I don’t think that episodes of weakness in real growth mean that we put part of our mandate aside. I think it has to temper the extent to which we lower rates and temper our behavior on the other side as well. As for semantic tactics, I support President Plosser’s suggestion, seconded by President Yellen, and I support President Poole’s suggestion as well.

  • Thank you. President Stern.

  • Thank you, Mr. Chairman. We had a fairly extended discussion yesterday about concerns about the condition of various financial markets and some financial institutions and their implications for the economic outlook, and I certainly share those. In this environment, I favor alternative B—a 50 basis point reduction in the federal funds rate target. My own forecast is below the Greenbook forecast for this year and next, and so my guess is that we’re going to wind up moving further before we’re done. But that said, I don’t have enough confidence in my forecast to advocate that we do more right now.

    As far as language of alternative B is concerned, I am certainly happy in general with what we have as written, and I share President Hoenig’s view, although perhaps for a different reason, that the less we say about inflation right now, the better. The reason is that, in some sense, we haven’t adequately expressed to at least some market participants that we understand where the risks lie right now. In trying to remind people that we are not inflation nutters but are serious about maintaining price stability, we’ve kind of garbled our message, in my opinion, and I think people in the marketplace know this central bank is committed to price stability. I don’t think we have to remind them with every statement. Thank you.

  • Thank you. President Fisher.

  • Well, Mr. Chairman, I’ve seen the discount rate tally. I’ve listened carefully to all my fellow Presidents and to Governor Kohn. I suspect I know what your fellow Governors are going to recommend. I’m in a distinct minority at this table. This weekend, by the way, I searched the newspapers for something to read that didn’t have anything to do with either a rogue French trader or market volatility or what the great second guessers were blabbing forth at the chat show in Davos; and in doing so I happened upon a delightful article. I hope you saw it in the Saturday New York Times on the search for a motto that captures the essence of Britain. My favorite was nemo ne inclune lacet, which very loosely translated, I think, means “never sit on a thistle.” [Laughter] Well, that’s where I am. I’m going to risk sitting on the thistle of opprobrium for my respective colleagues by making the recommendation that we not change the funds rate and that we stay right where we are.

    Now, for the record, I would have supported last week’s 75 basis point cut for the reasons that it would put us ahead of the curve and bought adequate insurance against a recession. I told you that directly, Mr. Chairman, and I mentioned it to Governor Kohn as well. Judging by the policy rules on page 21 of the Bluebook, as well as by the adjusted rule that our economist Evan Koenig has developed in Dallas, we are, indeed, ahead of the curve from the Taylor rule standpoint as we meet today with the rate of 3.50. As was mentioned earlier, we have not been docile. We have cut rates 175 basis points in a matter of months, and we’ve taken some new initiatives that I think are constructive and useful. I’d like to see more along the lines of the TAF. To be sure, in the discussion that we had in that emergency meeting, I had the same concerns that President Hoenig expressed in the call, but with the wording change that was put forward by Governor Kroszner I ended up where President Hoenig did. I regret not voicing my discomfort with the penultimate sentence in the statement—the one dealing with appreciable downside risk after the move we took—as I felt that it undercut the potential effect of our decision. During that call, you may recall that I pointed out the pros and cons. I began my intervention on that call by saying that there’s a very fine line between getting ahead of the curve and creating a sense of panic. I also expressed concern of the need to be mindful of inflation, as many have at this table today. There are some critics who say we panicked in response to the market sell-off of that Monday. I do not believe that’s the case, and I don’t believe it’s the case because I find it impossible to believe. As I’ve said repeatedly in this room, other than in theory, markets are not efficient, and on the banks of the Hudson or the Thames or the Yangtze River, you cannot in practice satisfy the stock market or most other markets, including the fed funds futures market, in the middle of a mood swing. When the market is in the depressive phase of what President Lockhart referred to as a bipolar disorder, crafting policy to satisfy it is like feeding Jabba the Hutt—doing so is fruitless, if not dangerous, because it simply will insist upon more. But attempting to address the pathology of the underlying economy is necessary and righteous, and that’s what we do for a living, and I think we are best sticking with it. We’re talking about the fed funds rate. I liken the fed funds rate to a good single malt whiskey—it takes time to have its ameliorative or stimulative effect. [Laughter]

    But I’m also mindful of psychology, and that’s what I want to devote the remainder of my comment to, and then I’ll shut up. My CEO contacts tell me that we’re very close to the “creating panic” line. They wonder if we know something that they do not know, and the result is, in the words of the CEO of AT&T, Randall Stephenson, “You guys are talking us into a recession.” To hedge against that risk is something to them unforeseen, even after they avail themselves of the most sophisticated analysis that money can buy. CEOs are, indeed, doing what one might expect. They are tightening the ship. They’re cutting head counts to lower levels. They’re paring back cap- ex where they can beyond the levels they would otherwise consider appropriate after imputing dire assumptions of the effects of housing. I’m going to quote Tim Eller, whom I consider the most experienced and erudite of the big homebuilders, which is Centex, who told me, “We had just begun to feel that we were getting somewhat close to at least a sandy bottom. Then you cut 75 basis points and add ‘appreciable downside risks to economic growth remain’ in your statement, and it scares the ‘beep’ out of us.” He didn’t use the word “beep.” These are his words, not mine. Imagine scaring a homebuilder already living in hell. The CEOs and CFOs I speak to from Disney to Wal-Mart, to UPS, to Texas Instruments, Cisco, Burlington Northern, Southwest Airlines, Comerica, Frost Bank, even the CEO of the felicitously named Happy State Bank in Texas, repeated this refrain, “You must see something that we simply do not see through our own business eyes.” They do see a slowdown. They are worried about the pratfall, as I like to call it, of housing. They’re well aware of California’s and Florida’s economic implosion and broader hits to consumer welfare across the national map. I recited some data points from those calls yesterday. But they do not see us falling off the table. They worry aloud that by our words and deeds we are inciting the very economic outcome we seek to cut off at the pass by inducing them to further cut costs, defer cap-ex, and take other actions to hedge against risk. They can’t fathom it but assume that we can.

    Our Beige Book contacts and the respondents to the business outlook survey in Dallas say pretty much the same thing. One of those actions is to fatten margins by passing on input costs. Now, I mentioned the rail adjustment factor yesterday, and I’m troubled by the comment that I quoted yesterday from the CEO of Tyson Foods. “We have no choice but to raise prices substantially.” I mentioned that Frito-Lay has upped its price increase target for ’08 to 7 percent from 3 percent. Kimberly-Clark notes that it is finding no resistance at all to increasing prices in both its retail and institutional markets, and I mentioned that Wal-Mart’s leaders confirm that, after years of using their price leadership power to deflate or disinflate the price of basic necessities— think about this—from food to shoes to diapers, they plan in 2008 to apply that price leadership to accommodate price increases for 127 million weekly customers. This can’t help but influence inflation expectations among consumers.

    I experienced a different kind of price shock two weekends ago, when I went to buy a television so I could watch President Rosengren’s football team demolish President Yellen’s. [Laughter] I was told that they had doubled their delivery and installation fees because of a “fuel surcharge.” Well, I reminded the store clerk that I had been there about the time of the Army-Navy game, around Thanksgiving, and that gas prices had not doubled since the Army-Navy game, and he said, “Mr. Fisher, we’re selling less, and we will take what we can get away with however we can get away with it.” With one-year-forward consumer expectations, according to the Michigan survey, already above 3 percent, everyone from Exxon to Valero to Hunt Oil and our own economists in the Greenbook telling me that oil is likely to stay above $80, and the national average price therefore above $3, this mindset really worries me. I’m going to add one more very troubling little personal anecdote. Driving home from work last week I heard a commercial for Steinway pianos. The essence of the advertisement was that manufacturing costs had increased and that you could buy a piano out of their current inventory at the “old price” that was in place in 2007; but come February 1, there would be sizable price increases, so you’d better purchase your piano quickly. It has been thirty years since I have seen advertisements to go out and buy now before the big expected price increases go into effect. Now, this is an isolated, little bitty incident, but I fear this may be just the beginning of the more pervasive use of this tactic.

    Everyone in this room knows how agnostic I am about the predictive value of TIPS and the futures instruments comparing TIPS with nominals, like the five-year, five-year-forward. I’ve sent around an eye popping chart that shows the predictive deficiencies of the professional forecasters that were tracked by the Philadelphia Fed. I know that dealers are telling us that inflation is contained, but I have spent many years in the canyons of Wall Street, and I would caution against their disinterest in the predictions that they offer. When I see that every measure of inflation has turned up, learn from studying the entrails of the last PCE that 83 percent of the items therein experienced a price upswing, consider the shortcomings of the few tools we have for evaluating expectations of future inflation, and then hear from microeconomic operators of the economy that, by golly, we’re going to take what we can while the getting is good, I can’t help but feel that we cannot afford to let our guard down by becoming more accommodative than we have already become with our latest move.

    Mr. Chairman, you know because we’ve talked about this that I’ve anguished over this. In fact, to be politically incorrect in a government institution, I have prayed over it. It is not easy to go against the will of the people you have enormous respect for, but I have an honest difference of opinion. I truly believe we have it right at 3½ percent right now. I think that, even with some important language changes, we risk too much by cutting 50 basis points at this juncture and driving the real rate further into what I perceive, even on an expectations-adjusted basis, is getting very close to negative territory. Mr. Chairman, I think we’ve gone as far as is prudent for now, and that 3½ percent, together with the other initiatives we’ve taken to restore liquidity, is sufficient. So I ask for your forbearance in letting me sit on the thistle of recommending no change.

    I do want to say as far as the language is concerned, since obviously we’re going to go with alternative B despite my vote against it, that I strongly recommend you consider the changes that were given by Presidents Plosser, Yellen, and Poole, and I would strongly advocate particularly at the end adding that we will act as needed to foster price stability and sustainable economic growth. I thank you for paring back alternative B, paragraph 3, in terms of getting away from discussing only energy, commodity, and other import prices. Thank you, Mr. Chairman.

  • Thank you. Why don’t we take a coffee break and come back at 11 o’clock? Thank you.

  • [Coffee break]

  • Okay. Why don’t we recommence. Governor Warsh.

  • Thank you, Mr. Chairman. I support alternative B for several reasons, not least of which because it actually seems to reasonably capture what we talked about yesterday. So I think it has that benefit. To state the obvious, we are in a very tough spot. There are clearly risks on both sides of the mandate, and for folks who are of differing opinions, I don’t think that they would either deny the tough spot we’re in or say that we have a lack of worries on either side. I think President Geithner said in the Q&A at the beginning of this round that we’re choosing which policy error we’re prepared to make or, more charitably, we’re deciding where our worries are the greatest. It does strike me that alternative B does that reasonably well. To step back for just a minute, I think that the tough spot we’re in is a function of issues that were very long in the making and will take a long time to work out. That probably has two implications. First, as to what Governor Kohn said, it’s easy for me to feel more comfortable with our judgment by saying that we’ll be able to undo it really, really soon; but if it is going to take a long time for this to work out, I don’t want to take any false comfort in that. The other implication is that, when I say that this is long in the making, I mean particularly for financial institutions. It has taken them a long time to dig themselves into this hole for the credit-intermediation process to be as deeply disturbed as I take it to be at this point. So when I think about what policies we would make on the monetary policy front, I wouldn’t want to excuse or somehow allow them to get off scot-free; by going with alternative B, I don’t think that would be the case. Financial institutions have far more work to do than we have yet to do on monetary policy. I think they are further from recognizing where they need to get. That will take some time.

    In listening to other people’s perspectives on the elixir of a 3 percent fed funds rate, I think it would be a nice luxury to give 3 percent a chance, but it’s probably not practical for the following reasons. I have tried to convince myself that the monetary policy moves in the last nine or ten days were a one-time step-function change, by which we were setting a level based on where we think the real economy is, and we need to get back to normal monetary policymaking. I think dropping the word “appreciable” as in alternative B is one way to do so. The goal would be for monetary policy to get back to sort of normal business in tough times.

    Another point, just of commentary, is that, by taking this action along the lines of alternative B, it would be nice if we weren’t going to be lowering the value implied by the markets of where the rate will end up. So my support of alternative B is nonetheless with this worry as well—that the markets might think that we have more-dramatic actions and that we will have to go lower longer than is currently implied. I don’t mean to give them additional credit there, but I don’t think that we have a great alternative. As to the point about market expectations, I would note that we will—and necessarily at some level should—during the course of our next several meetings disappoint market expectations, and that is not something I think we need to run from. I think that will be part of the discipline function, but this is probably not the right time to do it.

    As a final point, alternative B is consistent with the narrative that we’ve started based on the data we saw in mid-December, based on the Chairman’s speech earlier this month, and based on our meetings over the past several weeks. It strikes me that it would be prudent at this time of financial stability risks and uncertainty in the markets for us not to add volatility—not to throw out a different nuance or try to be too clever. So while I have sympathy for some of the ideas and amendments that have been suggested in this discussion, we’ve come to a pretty tough place, as I said at the outset. It’s a tough fork in the road. We should all feel to some degree uncomfortable about the choice we make, but the choice we’re making today needs to be as clear as it can be. Even if we’d get some comfort in being a little cute, a little clever, and a little nuanced, I’m afraid we might be undoing some of the clear bet that we have to put on the table. So with that, I support alternative B as written. Thank you, Mr. Chairman.

  • Thank you. Governor Kroszner.

  • Thank you very much. I also support alternative B. I think by any measure 125 basis points of easing within a month is a lot and perhaps it is a nearly unprecedented level of insurance that this Committee has purchased in such a short period of time. But I think it makes sense to have done it in these circumstances because we have the risk of a substantial tightening of credit and financial market conditions. There are still lots of shoes that may be left to drop and, in Nellie Liang’s great phrase yesterday, “unplanned asset expansions can occur.” Just today we heard that UBS increased by $4 billion the report of their expected losses from what they reported last month, and that is not their final report. They’re not going to come out with the final numbers for another month, and so it’s certainly conceivable that more could be there. The SocGen situation is obviously another thing that is potentially unsettling, and so I think it’s very important to be buying some insurance, particularly when the slow-burn scenario that I’ve talked about for a long time is the one that I see as the most likely. There is not going to be just some immediate credit crunch or some immediate problem of impairment of capital at an institution. But just given the increase in the cost of raising capital, given the difficulty of getting things off the books, given that other things have to come on, or given just that the old securitization machines that at a given level of capital sustained a very high level of activity, which can’t be done now because you can’t get the things off the books, the machine can’t churn at the same rate that it did even with the same amount of capital. Then with all these other challenges, we have to be wary of that. So the fragility is still there. As we talked about before, we had some dramatic improvements in October, through early November of last year, and suddenly they reversed, and I just don’t feel comfortable that I understand where those went.

    That said, however, I think the very recent data that we have gotten provide just a glimmer of hope. I mean, we’ve had relatively low claims numbers. The ADP number was relatively low. I’m not sure how much information content is in that, but with the claims number it makes a not- implausible case that there wouldn’t be a dramatic negative number in the employment report and potentially it could be on the positive side. I don’t think GDP is quite so miraculous, as we heard the very humble Dave Stockton tell us that they nailed it. But also look at some of the places in which there was some weakness now that may come up next quarter. Over the next couple of quarters probably we’re going to see a payback in government spending, whether it’s through direct stimulus or through more expenditures on the military. The advance durable goods number, which no one really mentioned, was on the positive side. I don’t want to put too much weight on that, but it suggests that there are at least some mixed signals going forward. So I think that means that we need to leave our options open. We need to worry about those downside risks, but we shouldn’t dismiss the possibility that the forecast that Dave and the team have put out not only has nailed it for this quarter but also has not done such a bad job for the quarters going forward.

    That means we have to think a lot about what we want to say in the statement. How do we get that balance right? Clearly, it’s important that we take out the word “appreciable,” and the markets will see that. The concern that President Fisher raised is a real one. If we continue to talk about “appreciable risks” after 125 basis points of cutting in a month, I think that would be unsettling to market participants. But I think the markets are expecting us to take that back. Actually if I could just for a moment get clarification from President Yellen. My understanding was that what she suggested about the change to the fourth paragraph and the assessment of risk was to change only the last sentence back to the December 11 statement.

  • That’s correct. I was suggesting changing only the very last sentence.

  • Which is different from what President Plosser proposed.

  • He proposed, as I recall, changing two sentences, eliminating the one that referred to downside risks and also changing the last sentence. I would not want to see the downside risk sentence eliminated; I would change only the very last sentence.

  • Yes, I would agree that I would not want to see that sentence changed. I’m open to the possibility of simply changing the last sentence to go back to December 11. Now, that clearly would be taken by the market as a signal that we were moving back.

    Another proposal that has been put on the table is to add something about inflation concerns or acknowledgement of the inflation situation in the rationale, starting off with something like “although inflation pressures remain.” I’d probably end up coming down with President Stern on that—at this point, I’m not sure we gain that much by saying that. That is something we may need and want to say the next time that we undertake a policy action. I’m not sure it’s necessary to do so now, although if there were a strong desire to say something like “although inflation pressures remain,” I would not be opposed to that. But I do think it might be reasonable to at least consider moving back to the December 11 formulation for the last sentence of paragraph 4, but that’s certainly a step back. I don’t think that closes off options for us going forward, but I’m certainly open to those concerns because I don’t want to close those options off.

    Finally, on the broader issue of communications, particularly with respect to reversals, something that worries me is that there has been a lot of discussion about Japan. After they would provide more liquidity or a rate cut, they would say, “Well, we really didn’t want to do that, and we’re going to take this back as soon as we can.” So they kept providing more and more liquidity, and it had less and less effect. It was the classic pushing on a string. I think that shows the very important role of expectations. So at this point, given that I think we want to leave our options open, I don’t think we want to be emphasizing that. It’s extremely healthy for us to be discussing that internally, but until we see how things play out, I’m not sure it’s wise to do that because we could really undermine the effectiveness of our monetary policy actions. That is the last thing we want to do because that might force us to go down further than we would otherwise like and cause more volatility in the markets, more uncertainty with more potential for unanchoring expectations. Thank you, Mr. Chairman.

  • Thank you. Governor Mishkin.

  • Thank you, Mr. Chairman. As you know from my discussion of the economy, I think that we’re skirting a recession, but I see very large downside risks. When we looked at the Greenbook, I agreed with the view that the equilibrium fed funds rate has dropped substantially, at least 125 basis points from the last FOMC meeting in December. When I think about a 50 basis point cut in this meeting, which is clearly the consensus of the Committee, my view is that it is consistent with the downward revision in the modal forecast but it does not take out any insurance; and this is something that the Vice Chairman alluded to before.

    So all else being equal, I would actually advocate a 75 basis point cut at this meeting because I do think there is a need to take out insurance. However, there are mitigating factors, so I do not feel as strongly as I did at the full FOMC meeting in December, when I took the view that we should take out insurance and aggressively cut more than the consensus of the Committee, but felt that dissent would not be helpful in terms of the perceptions of the Committee and the Federal Reserve and so was willing to vote with the majority. In this case, I don’t feel as strongly, so I am more comfortable with 50 basis points, although everything else being equal, I think that insurance is warranted, and I get nervous that we are not getting sufficiently ahead of the curve.

    So what are the mitigating factors that allow me to be more supportive of the consensus of the Committee? One is that I do worry that markets might think we are panicking. This is the issue that President Fisher raised. A 50 basis point cut would be a cut of 125 basis points in a month. There has been a lot of discussion in the media about our panicking on this, so there is a danger in this regard, and that is a mitigating factor for my not pushing hard for a 75 basis point cut. The second issue is that we saw a sharp move up in inflation compensation after the 75 basis point cut at the last conference call meeting. Actually there’s no evidence at all that this was a result of higher inflation expectations, but I do think it illustrated that there was more uncertainty about inflation and where our long-run inflation goals might be. This is a reason that the Committee needs to get to more clarity on our inflation objectives. You know, it is no surprise to people that I’m not completely happy with where we are now in terms of the range. Exactly in a situation like this, just as in 2003, you may drift up to the top of the band and be happy to stay there or drift to the bottom of the band and be happy to stay there. I do think there is a cost. This illustrates a cost to our communications strategy, and I hope we would reconsider that at some point in the future.

    The third mitigating factor is that there is an issue about taking out insurance when the Committee is not completely prepared and we have also not prepared the markets to think about how we might think about reversals if they were necessary in the future. The whole insurance strategy is one in which you can’t take out insurance and then have it be seen as always easing and thus unhinge inflation expectations and lose the strong nominal anchor. So the Committee has to think more about conditions under which it would reverse course and actually prepare the markets for that. As I said, there are two kinds of information that I think we get fairly quickly, along with others—particularly information on inflation expectations and information on whether financial market strains are easing. I think we need to clarify that in terms of the public. It is a point that Governor Kohn made, and I want to return to it in a second. But the bottom line is that I support the 50 basis point cut in the funds rate, so I support alternative B.

    On the statement, I am probably the only person who likes the alternative A, paragraph 3, rationale. But I am sympathetic to President Stern’s concerns—the less said the better—so I’m not going to push it. But I am comfortable saying that inflation expectations are currently well anchored and that there is some expectation of easing of slack in the economy, and so my view is that the upside risks to inflation are not the serious concern right now. I’d be willing to state that if it were me personally, but the issue that President Stern has raised that maybe the less said the better is fine with me. Again, I’m perfectly comfortable with alternative B as it is written now.

    One last point, which is to address an issue that Governor Kohn mentioned: I think he’s absolutely right that we should not be saying or thinking that we expect to reverse right now. Even if we actually did a 75 basis point cut at this meeting, which we’re not going to do, I would not take the view that it would mean that we plan to reverse it. When you put these projections out it’s tough to indicate that you would be willing to reverse but the reality is that we’re taking out the insurance because we think that the financial strains will last for a long time. That is our best prediction, and on that basis, to give an impression that we want to take it away would create problems. This is the issue that Governor Kroszner talked about in terms of Japan, where clearly expansionary policy was not working because they kept saying, “We’re going to take it back.” As a result, from the viewpoint of expectations, the policy was not expansionary. I do think policy needs to be expansionary at this point. However, I’m not sure that I agree that we should not talk about the issue of reversals. It’s a question of the way we talk about it. It’s not that we expect that we would be reversing quickly, but conditions could change, which we’re not expecting, but we really would like that to happen. We’d be hopeful, in particular, that the financial strain is turned around quickly, which in the past we have seen happen in some cases. In some cases, it takes a very long time. The episode in the early 1990s was one in which those financial strains took a very long time to dissipate. Similarly, we don’t talk about it as much, but in the early aftermath of the 2000-01 recession, there was a sequence of financial strains, particularly things like Enron and the lack of confidence, that lasted a heck of a long time. So I would say that we do need to think and talk about it, but it is not that we want to reverse it. We are ready to reverse, but the conditions must be such that we’re actually getting information that tells us that reversals are appropriate. Thank you very much.

  • Thank you. President Poole.

  • Just very quickly, given the number of people who have talked about reversals, it seems to me that the minutes need to say something about it because the minutes have to be true to the ultimate transcript as published.

  • Thank you, Mr. Chairman. Let me just say that I am completely comfortable with the way you just framed the case, not just for thinking about and talking about strategies for how we take this back at the appropriate point, but I also don’t have any problem with the way you talked about how you’d frame it in public. I thought that was fine. When I talked about the need for humility, it was just in our capacity to design optimal strategies for the timing and the slope of the path to reversal at this stage. I think it’s going to require more time.

    A few things about the policy action and the statement. Obviously I support 50, and I support the language in alternative B, paragraph 4, as written. The Chairman laid out in public earlier this month a basic strategy that said that policy would be directed at providing an adequate degree of insurance against the downside risks, given the nature of those risks. It was said with more nuance and eloquence, but basically that was the thrust of the strategy. I think that the balance of evidence we have—even with all the uncertainty about what equilibrium is, what the appropriate level of the real rate is, and what it actually is—suggests that we have not gotten policy to that point. Therefore, getting that strategy in place is likely to require further action. It’s very hard to know when or how much. It’s very hard to know what should frame that choice now for us, I think. As many of you said, markets have to go through a further set of adjustments, and I think that has to work through the system. Our job, again, is not to artificially interfere with that process or to substitute our judgment for what the new equilibrium should be in that context. Our job should be to make sure that adjustment happens without taking too much risk that it tips the financial system and the economy into a much more perilous state that would be harder for us to correct and require much more policy response. There’s a big difference between a world in which housing prices fall 20 percent and one in which they fall 40. If everybody thinks they have to prepare for a world in which they fall 40, we’re going to take much greater risk that we have a scale of financial trauma and credit crunch that would produce the odds of a deeper recession.

    What we face now is not the choice, as I think President Poole said, between a mild recession now and higher-than-expected inflation over time. The risk we face, as the Chairman said several times, is the choice between a mild, short recession and a deeper, more protracted outcome. The scale of financial market fragility we now face, you could even say solvency in parts of the financial system, is a function of the confidence we create in our willingness to get policy to a point that provides meaningful protection against an adverse outcome. I think the experience of the past five or six meetings suggests that we cannot carefully enough design a message that can lean against expectations in the market about the likely path of policy that we judge to be excessive and that makes us uncomfortable without taking too much risk that it will just undermine confidence in our willingness to get policy right. We just don’t have that capability. We thought about it and tried it in lots of different ways, but everything we experienced over the past four months justifies the judgment that we can’t lean against those expectations without taking too much risk that we undermine confidence in our capacity to get policy to the point at which we’re giving some insurance.

    It would be a mistake to try to recalibrate expectations now relative to the stance of policy as stated in the Chairman’s statement and our statement last week. It would be a mistake to recalibrate back to a point that looks more like October. Everything we know suggests that it’s probable that the data will get still worse from here and that the financial markets will be in a state of considerable fragility and tenuousness for some time. To sort of zig at this meeting, to recalibrate more toward a sense of balance in the face of that reality, just means that we’re likely to have to zag back again. I just think that would risk our looking as though we’re ambivalent and have to correct again because of that. That would just be uncomfortable. It’s hard to know—I think Governor Kroszner said it right in many, many meetings—whether we face the risk of a grinding, downward, self-reinforcing set of pressures on balance sheets that raises the risk of deeper trough in housing and creates more caution and deleveraging and the economy moves slowly down or whether we face the risk of much more acute cliffs in asset prices with much greater consequences for confidence and the fragility of the financial system. It is hard to know, but I think both would be very uncomfortable for us.

    Again, I strongly support the language in B as it is. Even though I understand the rationale, I’d be very uncomfortable with dialing back that statement as it now exists to something that suggests we’re closer to balance.

  • Thank you, and thank you all. Let me make a few comments. The recent period has been more ragged in terms of policy and communications than I would have liked. In particular, during the past few months I think a perception has developed that we are tentative and indecisive and are not communicating clearly enough to the markets and to the public. Now, whether that’s fair or not, that perception is there and has to be addressed. I think we have begun to address it with our recent communications and the strong policy action that we took last week. For my part, I plan to speak more often and more clearly in public about the economy and our policy strategy, and to do that I will need your continued thoughtful input and support. All that has happened notwithstanding, I think we now have the opportunity to get policy to about the right place. As we have all noted, the pace of economic activity appears to have slowed sharply around the turn of the year. I am very hopeful that the labor market report this Friday will be positive. However, as I mentioned yesterday, there is as yet no evidence whatsoever that the housing market is stabilizing, as the new home sales data show. So long as house prices keep falling, we cannot rule out some extremely serious downside risks to our economy and to our financial system. We need to be proactive and forceful, not tentative and indecisive, in addressing this risk.

    I recommend a 50 basis point cut at this meeting, to 3 percent. This is the value that the Greenbook forecast requires to achieve modest growth later this year and price stability by next year. A decline of 225 basis points since last August fits well with both constructed indicators, like the staff’s medium-term r*, as well as market indicators such as short-term nominal interest rates. A 3 percent funds rate is lower than most Taylor rule prescriptions, but the Taylor rule makes no allowance for current unsettled conditions in financial markets, including spreads and bank capital constraints, which make a given value of the funds rate less accommodative today than it would be in normal circumstances. In contrast, the Greenbook analysis attempts to take those financial conditions into account. Even at the 3 percent level, assuming that the Greenbook analysis is in the right ballpark, we would be making little or no allowance for risk-management considerations, as several people have pointed out. For that reason, it’s instructive to note that the Greenbook has a flat path that doesn’t reverse, and the reason is precisely that they’re in a certainty-equivalent framework without any insurance having been taken out. If you accept risk-management principles, then we should be considering the possibility of cutting below 3 percent and then reversing when that insurance is no longer needed. However, 125 basis points of ease in two weeks is certainly a good amount, and I don’t think that going further at this time is advisable.

    Regarding the statement, I recommend alternative B. It is very similar to our statement of last week in the description of the economy. There are two main differences, both in paragraph 4. First, and very important, we acknowledge our two recent policy actions, including today, and note that they should lead to moderate economic growth and to mitigation of risks. The latter part is strictly true because a cut will certainly mitigate risks, but it may be read to say that we think we’ve taken out a little insurance, which is, as I said, somewhat debatable. That’s something we can discuss in the future. Second, the downside risks are no longer described as “appreciable.” I think both of these steps will serve to moderate expectations of further rate reductions by indicating that we think that our actions thus far go a substantial distance in addressing the downside risks.

    I strongly counsel, however, against taking out the sentence referring to downside risks. First of all, it’s not plausible that downside risks that were appreciable just last week are no longer an issue today; and indeed, in the projections that we’ve submitted, the Committee overwhelmingly saw a downside risk to economic growth. We have to be honest about that. Second, although I realize what the intent of that change would be, which would be to say that we’re not promising further cuts, it could actually be read in quite the opposite way, which is that we are saying we’re not making further cuts even if the conditions warrant, and that would put us back in the situation we were in at the end of October, when the market was confused about whether or not we were willing to respond to circumstances. We have to maintain a posture of flexibility and responsiveness in this fluid situation, and drawing a line in the sand is not the way to do that. I’m also reluctant to make the other changes that have been suggested. I understand the purposes behind them, but I’m very conservative about changing wording that has just been used a week ago in a statement and less than three weeks ago in my speeches. It’s just so risky to give the impression that we are changing course when we have just simply now identified the downside risk and have taken aggressive action in that direction.

    Now, I understand that several of you will be quite unhappy with the nature of that statement, and let me try to reassure you in the following way. Rather than trying to convey all the subtleties of our policy strategy in the statement—and here I agree with President Poole, who has pointed to the limitations of the statement—I propose to supplement the message of the statement, as I said, by commenting more on the economy and the policy in public speeches and in testimonies. Between now and the next FOMC meeting I have a testimony the week after next, I have the Humphrey-Hawkins testimonies, and I have a speech, and in each of those I will talk about the economy and the policy strategy that we are taking. In particular, subject to your counsel and individual, perhaps even collective, consultation, in my public speaking I will try to make three main points.

    First, I will try to reinforce the message of the statement, which is that the Federal Reserve recognizes the downside risks to economic activity and we are on the case. We have already moved aggressively and proactively, and we are prepared to respond in a timely manner as needed to mitigate the risks of very bad outcomes. Conveying the message that the Fed will be active and willing to mitigate tail risk is critical for achieving financial stabilization, which in turn is necessary for achieving economic stabilization. We have to show that we’re in touch. But—and let me now address these points in particular to President Plosser and others who have raised concerns—the second point I will make is that the Fed is not on autopilot for further rate cuts. Monetary policy takes time to work, and our recent actions will do little for the economy over the next few months. Thus, some weak economic news is to be expected in the near term and is not a prima facie case for additional easing. Instead, we will be data-dependent in the specific sense that we will respond to the incoming information that affects the medium-term outlook and to our assessment of the risks. The third point I will make in my public comments is that, to the extent that we are being proactive in addressing downside risks to the economy, we must also be proactive in removing accommodation once the economy is on a sustainable recovery path, and we must be clear about the circumstances that will prompt that reversal. Again, in saying this I reiterate that it’s not clear that we have yet taken out a great deal of insurance; nevertheless, I think that point needs to be made. The need for making a policy reversal at an appropriate point, which may not be the next six months but when the conditions warrant, is important; moreover, I think it’s actually credible. If you look at the dealer surveys or the fed funds futures, they show a hockey stick. They do show a response and then change, coming back later this year or next year.

    So that’s what I’d like to do. I’d like to be conservative with the statement. Given some of the problems we’ve had, I don’t want to do any right turns or U-turns or add any confusion. I want to continue on message in terms of where we have been in the past few weeks. But recognizing the subtleties of our policy approach, the concerns that we have about inflation and about financial conditions and economic growth, I will be making a concerted effort to try to explain in public the subtleties of our strategy and of our economic outlook. I hope we can present a united front to the public behind the strategy to the extent possible. Again, I appreciate your support and understanding of the very difficult pressures and conflicting forces that we face. So that’s my recommendation. I would be happy to take any further comments. If none, Debbie, can you call the roll?

  • Yes. The vote is on the language for alternative B as it is in the Bluebook and in Brian’s handout and on the directive from the Bluebook, which I will read.

    “The Federal Open Market Committee seeks monetary and financial conditions that will foster price stability and promote sustainable growth in output. To further its long-run objectives, the Committee in the immediate future seeks conditions in reserve markets consistent with reducing the federal funds rate to an average of around 3 percent.”

    Chairman Bernanke Yes
    Vice Chairman Geithner Yes
    President Fisher No
    Governor Kohn Yes
    Governor Kroszner Yes
    Governor Mishkin Yes
    President Pianalto Yes
    President Plosser Yes
    President Stern Yes
    Governor Warsh Yes

  • Thank you very much. We’ll take just a very quick recess for the Governors to meet, and then we’ll come back for our presentation.

  • [Meeting recessed]

  • Okay. Let’s reconvene. We have a special presentation on policy issues raised by financial crisis. Let me turn to Pat Parkinson to introduce the presentation.

  • Thank you, Mr. Chairman. The first exhibit provides some background on efforts to analyze the policy issues raised by recent financial developments and an overview of today’s briefing. As indicated in the top panel, in response to a request from the G-7, at its meeting last September the Financial Stability Forum (FSF) created a Working Group on Market and Institutional Resilience. The working group’s mandate calls for it to develop a diagnosis of the causes of the recent financial market turmoil, to identify weaknesses in markets and institutions that merit attention from policymakers, and to recommend actions needed to enhance market discipline and institutional resilience. The working group has been asked to prepare a report for consideration by the FSF at its meeting in March so that the FSF can complete a final report to the G-7 in April. The President’s Working Group on Financial Markets (PWG) is conducting its own analysis along the same lines and will ensure coordination among the U.S. members of the FSF working group. Chairman Bernanke, Vice Chairman Kohn, and President Geithner asked the Board’s Staff Umbrella Group on Financial Stability to organize and coordinate staff support for their participation in the FSF working group and the PWG’s effort. Specifically, they asked the staff to analyze the nine sets of issues listed in the middle panel. Subgroups of staff from the Board and the Federal Reserve Bank of New York were formed to address each set of issues, and work is well under way on all of them. The first four of these issues will be discussed in today’s briefing.

    As shown in the bottom panel, today’s briefing will consist of three presentations. I will start by presenting a diagnosis of the underlying reasons that losses on U.S. subprime mortgages triggered a global financial crisis. This diagnosis will suggest that among the most important factors were (1) a loss of investor confidence in the ratings of structured- finance products and asset-backed commercial paper (ABCP), which caused structured-credit markets to seize up and ABCP markets to contract, and (2) the resulting losses and balance sheet pressures on financial intermediaries, especially many of the largest global financial services organizations. In the second presentation, Mike Gibson and Beverly Hirtle, to my left, will present an analysis of issues relating to credit rating agencies and investor practices with respect to credit ratings. Then, further to my left, Jon Greenlee and Art Angulo will make the final presentation, which will focus on risk-management weaknesses at large global financial services organizations and the extent to which bank regulatory policies contributed to, or failed to mitigate, those weaknesses. I should note that also at the table today we have Norah Barger, who worked with Art Angulo on the regulatory policy issues, and Brian Peters, who worked with Jon Greenlee on the risk-management issues.

    Turning to the next exhibit, the diagnosis begins with the extremely weak underwriting standards for U.S. adjustable-rate subprime mortgages originated between late 2005 and early 2007. As shown by the solid line in the top left panel, as housing prices softened in 2006 and 2007, the delinquency rate for such mortgages soared, exceeding 20 percent of the entire outstanding stock by late 2007. In contrast, the dashed line shows that the delinquency rate on the stock of outstanding fixed-rate subprime mortgages increased only 2 percentage points over the same period, to around 7 percent. Nearly all of the adjustable-rate subprime mortgages were packaged in residential mortgage-backed securities (RMBS), which were structured in tranches with varying degrees of exposure to credit losses. The top right panel shows indexes of prices of subprime RMBS that are collateralized by mortgages that were originated in the second half of 2006. The blue line shows that prices of BBB minus tranches already had fallen significantly below par in January 2007 and continued to decline throughout last year, falling to less than 20 percent of par by late October. Prices of AAA tranches (the black line), which are vulnerable only to very severe credit losses on the underlying subprime mortgages, remained near par until mid-July, but they slid to around 90 percent of par by early August. After stabilizing for a time, they fell more steeply in October and November and now trade at around 70 percent of par.

    As shown in the middle left panel, from 2004 through the first half of 2007, increasing amounts of subprime RMBS were purchased by managers of collateralized debt obligations backed by asset-backed securities—that is, ABS CDOs. High-grade CDOs purchased subprime RMBS with an average rating of AA, whereas mezzanine CDOs purchased subprime RMBS with an average rating of BBB. The middle right panel shows the typical ratings at origination of high-grade and mezzanine CDOs. In the case of high-grade CDOs, 5 percent of the securities were rated AAA, and a further 88 percent were “super senior” tranches, which would be exposed to credit losses only if the AAA tranches were wiped out. Even in the case of the mezzanine CDOs, the collateral was perceived to be sufficiently strong and diversified that 14 percent of the securities issued were rated AAA at origination and 62 percent were super senior. As delinquencies mounted and prices of RMBS slid well below par, the credit rating agencies were forced to downgrade (or place on watch for downgrade) very large percentages of outstanding ABS CDOs. The bottom left panel shows that such negative actions were quite frequent throughout the capital structures of both high- grade and mezzanine CDOs, even among the AAA-rated tranches. Moreover, the downgrades frequently were severe and implied very substantial writedowns, even of some AAA tranches. When this became apparent to investors, they lost not only faith in the ratings of ABS CDOs but also confidence in the ratings of a much broader range of structured securities. Likewise, sophisticated investors who relied on their own models lost faith in those models as writedowns significantly exceeded what the models led them to expect. As a result, large segments of the structured-credit markets seized up. In particular, as shown in the bottom right panel, issuance of all types of non-agency RMBS declined substantially over the second half of 2007. Although comprehensive data for January are not yet available, conversations with market participants suggest there has been very little or no issuance.

    Your next exhibit focuses on two other markets that were affected by a loss of investor confidence—the leveraged-loan market and the ABCP market. The top left panel of that exhibit shows that spreads on credit default swaps on leveraged loans (the solid black line) already had come under significant pressure in June. By July these spreads had widened about 150 basis points. Investors had become concerned about a substantial buildup of unfunded commitments to extend leveraged loans (the dashed blue line), which in the U.S. market eventually peaked at $250 billion in July. As shown in the top right panel, as many segments of the structured-credit markets seized up, issuance of collateralized loan obligations dropped off significantly in the third quarter, adding to the upward pressure on spreads on leveraged loans. Nonetheless, the CLO markets continued to function much more effectively than the non-agency RMBS and ABS CDO markets. As shown in the bottom left panel, from 2005 to 2007, the U.S. ABCP market grew very rapidly. Much of the growth was accounted for by conduits that purchased securities, including highly rated non-agency RMBS and ABS CDO tranches, rather than by more traditional “multi-seller” conduits that purchased short-term corporate and consumer receivables. As investors became aware that some of the underlying collateral consisted of RMBS and ABS CDOs, they pulled back from the ABCP market generally, even to some extent from the multi-seller programs. Between July and December, total ABCP outstanding declined about $350 billion, or nearly one-third. The bottom right panel provides additional information on the growth of ABCP by program type. As shown in the first column, during the period of rapid growth from 2005 to July 2007, ABCP issued by structured-investment vehicles (SIVs) and CDOs grew far more rapidly than any other program type. The second column shows that, when investors pulled back from the ABCP markets, those program types shrank especially rapidly. The only program type that declined more rapidly during that period was single-seller programs. The single-seller category included a significant amount of paper issued by nondepository mortgage companies to finance mortgages in their private securitization pipelines, and this paper has almost completely run off.

    The next exhibit focuses on how the seizing-up of structured-credit markets and the contraction of ABCP markets adversely affected banks, especially many of the largest global banks. As you know, a combination of balance sheet pressures, concerns about liquidity, and concerns about counterparty credit risk made banks reluctant to provide term funding to each other and to other market participants. The top left panel of exhibit 4 lists the principal sources of bank exposures to the recent financial stress: leveraged-loan commitments, sponsorship of ABCP programs, and the retention of exposures from underwriting ABS CDOs. The top right panel shows the banks that were the leading arrangers of leveraged loans in recent years. The three largest U.S. bank holding companies (BHCs) head this list. As spreads widened and liquidity declined in the leveraged-loan market, these banks became very concerned about potential losses and liquidity pressures from leveraged-loan exposures. Although these exposures were smaller at the U.S. securities firms, those firms were even more concerned because of their smaller balance sheet capacity. However, to date the adverse impact on banks and securities firms from these exposures has been relatively modest and manageable. The middle left panel shows the leading bank sponsors of global (U.S. and European) securities-related ABCP programs—that is, programs that invest in asset-backed securities, including SIVs, securities arbitrage programs, and certain hybrid programs. As the conduits that issued the ABCP encountered difficulty rolling their paper over, many of these banks, fearful of damage to their reputations, elected to purchase assets from the conduits or extend credit to them, which proved in many cases to be a significant source of balance sheet pressures. This list is dominated by European banks. Indeed, the only U.S. bank among the top nineteen sponsors is Citigroup. However, Citigroup was the largest sponsor of SIVs, which, in addition to issuing ABCP, issue substantial amounts of medium-term notes. Citigroup, like nearly all SIV sponsors, eventually felt obliged to provide full liquidity support for all the liabilities of its SIVs, which amounted to around $60 billion at year-end. The memo item shows that some other U.S. banks sponsored securities-related ABCP programs that were relatively small in absolute terms but significant as a share of their total assets.

    But losses from leveraged-loan commitments and conduit sponsorship have paled in comparison to the losses some banks and securities firms have incurred from the retention of super senior exposures from ABS CDOs. These include exposures that the underwriters never sold, exposures that originally were funded by ABCP issued by the CDOs that was supported by liquidity facilities provided by the underwriters, and relatively small amounts of exposures purchased from affiliated money funds for reputational reasons. The middle right panel shows the leading underwriters of ABS CDOs in 2006-07. Merrill Lynch, Citigroup, and UBS head the list. Each of those firms has suffered very large subprime CDO-related losses, and Citigroup and UBS still reported very significant exposures at year-end. As shown in the memo items, among the other very largest U.S. bank holding companies, only Bank of America has suffered significant losses or still has significant exposures from underwriting ABS CDOs. I should note that the exposures shown in the exhibit are net of hedges purchased from financial guarantors, and most of these firms have hedged a significant portion of their exposure. As you know, there are concerns about the ability of the guarantors to honor their obligations under the hedging contracts. Indeed, some firms have begun to write down the value of their hedges with the most troubled financial guarantors.

    The bottom left panel shows total risk-based capital ratios for the four largest U.S. bank holding companies. All four remained comfortably above the 10 percent minimum for well- capitalized companies at year-end. Of course, Citigroup was able to do so only by raising substantial amounts of capital at a relatively high cost, and each of the other companies also announced capital-raising efforts. Moreover, Citigroup’s year-end ratio of tangible common equity to risk-weighted managed assets was 5.7 percent, well below the 6.5 percent target ratio that several of the rating agencies monitor and that Citigroup seeks to meet. The bottom right panel shows credit default swap spreads for the three largest U.S. BHCs. On balance, spreads for all three have moved up about 60 to 70 basis points since the market turmoil began. The spread for Citigroup has been elevated since October, when investors began to become aware of its subprime CDO exposures. The spreads for Bank of America and JPMorgan were in line with a broad index of bank spreads at year-end but jumped in early January on Bank of America’s announcement of its planned acquisition of Countrywide and JPMorgan’s announcement of substantial additions to its loan-loss reserves. Spreads for all three companies fell back more in line with the index last week. Thank you. I will now take your questions on this presentation before you proceed to Mike and Beverly’s presentation.

  • Are there questions for Pat? President Lacker.

  • You characterized underwriting as weak, and I guess the document circulated had some heavy criticism for the credit rating agencies. I want to understand more what the nature of that assessment involves. Basically is it ex post regret, or do we have objective evidence about the quality of the decisionmaking ex ante? That evidence, of course, would involve assessments of the probability they should have placed on things we saw.

  • Mike is going to address that in his briefing. I guess I’d prefer to delay and just simply say that I think we can point to aspects of their methodology that look fairly weak so that it wasn’t simply an ex post result but one that should have been foreseen at least to a degree if they had had a stronger methodology ex ante. Obviously that’s Monday morning quarterbacking, but still you can point to specific things that were weaknesses.

  • I have a question on exhibit 2, the top right corner: Are those AAA examples rated as of January 1?

  • I think that was the rating at origination.

  • At origination, not afterward. Okay. The other question I have is really a comment. On exhibit 1, the middle panel, it seems to me something that needs to be explored here, given that this is not the first time that banks have made a lot of bad bets, is the way in which the management incentives are designed. I just get the impression—and there has been some stuff in the paper recently, the Wall Street Journal or somewhere—that a lot of people are really given incentives to push these products and to make these deals. They walk away with big bonuses, and who in the heck cares what happens some time later. So I think that the management incentives are something very important to investigate in order to have an idea of how that works, and we might be of some assistance in designing something that would be helpful here.

  • Yes. Your exhibit 4 and your risk-based capital ratios, fourth quarter, are those as of December 31?

  • Okay. And do these reflect losses already taken in capital rates?

  • Yes. They reflect the developments in terms of the changes in their financial statements during the fourth quarter until the year-end.

  • Do you have a question, President Yellen?

  • I wanted to support President Poole’s comment. I remember very well back at Jackson Hole in 2005 that Raghuram Rajan presented a paper in which he emphasized the misalignment of incentives between investors and managers and the fact that almost everyone down the line right up to the investors themselves should have had incentives here. I don’t know what they were thinking, but everybody was rewarded for the quantity and not the quality of originations. He warned us before any of this happened that this could come to no good, and I think he did have some suggestions about compensation practices. These were not popular suggestions. So I think this is worth some thought. I don’t know what the answer is in terms of changing these practices. Maybe the market will attend to them, but it seems to me that we have had an awful lot of booms and busts in which this type of incentive played a role. Your presentation and the paper started from the fact that you note the deterioration in underwriting, but we should go one step backward. I suppose another issue here is what we saw in our supervision and whether we acted appropriately given what we saw. That raises a number of issues that I won’t go into at the moment but that I think we need to be sensitive to.

  • Just a question on exhibit 4, the bottom two panels. You have the four banks there, and when you look at the capital ratios, it doesn’t look that discouraging. But when I look at the credit default swaps, it looks a little less encouraging. So if I put Wachovia on this, I believe Wachovia’s credit default swaps now are up to 166, which would be much higher than your scale is right now. So it would be interesting actually to add Wachovia to that list. The second thing is what you take from the fact that the capital ratios don’t look so bad but the credit default swaps and what the markets are looking at indicate that, in the past month and a half, people— despite the capital infusions—are actually exhibiting more concern about the default experiences that might occur.

  • Right. Well, particularly with respect to developments in the last month or two, obviously the economic outlook is cloudier. As you know, I don’t think we try to reflect that fully in the capital ratios, and there is an element of stress testing and whatnot, but your other important effects are essentially looking through the cycle. So that may be part of it. Also, if you look at the credit default swap spread behavior compared with that for lots of other financial intermediaries out there, these look like very modest changes. I don’t know—if Mike or someone could help me out—to what extent these would translate into significant increases in actual implied default probabilities, but I wouldn’t think it would be all that large an increase.

  • Thank you. I agree with President Poole and President Yellen about the need to focus on compensation structures and incentives, but just two observations. One is that, if you look at compensation practices among the guys who actually look as though they did pretty well against those who didn’t do so well—I’m not talking about in a mortgage-origination sense but in the major global financial institutions—the structure of compensation doesn’t vary that much. What varies a lot is how well people control for the inherent problems in the basic compensation structures. Remember Raghu’s presentation was mostly about hedge fund compensation, and I think he is mostly wrong when you think about that and the incentive structure. The difference really is how you design your limits to make sure that your traders’ incentives are more aligned with the incentives of the firm as a whole. The biggest errors and differences are in the design of the process of the checks and balances to compensate for the inherent problems in the compensation structure. That’s important to know because a lot of these things, if you look at the formal attributes of the risk-management governance structure across these firms, don’t look that different. What distinguishes how well the guys did is much more subtle around culture, independence, and the quality of judgment exercised at the senior level, and this is important because, when you think about what you can do through supervision and regulation, to affect that stuff is hard.

    I have a question for Pat. Pat, not to overdo this, but where do you put in your diagnosis of contributing factors the constellation of financial conditions that prevailed during the boom and what those did to housing prices? You know, there’s a tendency for everybody to look at regulation and supervision and the incentives that they have created or failed to mitigate, but there is a reasonable view of the world that you wouldn’t have had the pattern of underwriting standards of mortgages without the trajectory of house prices that occurred. Sure, maybe what happened in the late stage of the mortgage-origination process contributed to the upside, but if you look at a chart, the rate of house-price appreciation started to decelerate about the time you had the worst erosion in underwriting practices. Anyway, my basic question is, Where do you put the constellation of financial conditions, not so much just what the Fed was doing but what was happening globally that affected long rates, expectations of future rates, et cetera?

  • Well, partly what I would say, in general, about the pricing of risk is that many, many people, including people in the Federal Reserve, were concerned about how narrow spreads were, were concerned about some of the slippage of practices, and were predicting that trouble lay ahead. But—and I’m certainly speaking for myself—I never expected this magnitude of trouble. What I’ve been focusing on are some of the factors that essentially made a bad situation much worse than we expected it to be. But there is no question that we entered the period with risk being priced very cheaply and a fundamental reassessment of risk. Again, I think that shouldn’t have surprised anyone, but almost everyone except the most extreme pessimists has been surprised by just how much trouble that repricing of risk has caused. Some things that we have focused on certainly were not anticipated, and we think they made the situation markedly worse than we expected it to be.

  • Are there other questions for Pat? Okay. We can continue the presentation.

  • As noted in the top left panel of exhibit 5, we would like to stress two key points on the rating agency and investor issues. First, credit rating agencies are one of the weak links that helped a relatively small shock in the subprime mortgage market spread so widely, though certainly not the only one. This is not just our staff working group’s view—most market participants have also expressed the opinion that rating agencies deserve some of the blame. Second, the way that some investors use ratings for their own risk management has not kept up with financial innovations, such as the growth of structured finance. These financial innovations have made a credit rating less reliable as a sufficient statistic for risk. The top right panel provides a roadmap to our presentation. To start, I’ll expand on some of the points that Pat made on the role of rating agencies in the financial crisis. My aim is to show why credit rating agencies were a weak link, which will lead naturally to our recommendations on rating agency practices. As we go, I’ll point out several places where the rating agency issues link up with the investor practices issues that you’ll hear about next from Bev. We feel strongly that the ratings and investor issues are really just two angles on the same underlying issue.

    The crisis began in the subprime market, the subject of the next panel. The subprime mess happened—and keeps getting worse—in part because of the issues associated with rating agencies (though as I said earlier, there is plenty of blame to go around). Our staff working group was asked whether the rating agencies got it wrong when they rated subprime RMBS. The answer is “yes”—they got it wrong. Rating agencies badly underestimated the risk of subprime RMBS. Last year, Moody’s downgraded 35 percent of the first-lien subprime RMBS issued in 2006. The average size of these subprime RMBS downgrades was two broad rating categories—for example, a downgrade from A to BB—compared with the historical average downgrade of 1⅓ broad rating categories. As indicated in the exhibit, the rating methodologies for subprime were flawed because the rating agencies relied too much on historical data at several points in their analysis. First, the rating agencies underestimated how severe a housing downturn could become. Second, rating agencies underestimated how poorly subprime loans would perform when house prices fell because they relied on historical data that did not contain any periods of falling house prices. Third, the subprime market had changed over time, making the originator matter more for the performance of subprime loans, but rating agencies did not factor the identity of the originator into their ratings. Fourth, the rating agencies did not consider the risk that refinancing opportunities would probably dry up in whatever stress event seriously threatened the subprime market. Of course, the rating agencies were not alone in this. Many others misjudged these risks as well. Some have suggested that conflicts of interest were a factor in the poor performance of rating agencies. While conflicts of interest at rating agencies certainly do exist, because the rating is paid for by the issuer, we didn’t see evidence that conflicts affected ratings. That said, we also cannot say that conflicts were not a factor. The SEC currently has examinations under way at the rating agencies to gather the detailed information that is needed to check whether conflicts had a significant effect.

    In the next panel, I turn to the ABS CDOs that had invested heavily in subprime. Rating agencies got it wrong for ABS CDOs. The downgrade rate of ABS CDOs in 2007 was worse than the previous historical worst case, just as it was for subprime. AAA tranches of ABS CDOs turned out to be remarkably vulnerable: Last year, twenty-seven AAA tranches were downgraded all the way from AAA to below investment grade. As indicated in the exhibit, the main reason that rating agencies got it wrong for ABS CDOs was that their rating models were very crude. Rating agencies used corporate CDO models to rate ABS CDOs. They had no data to estimate the correlation of defaults across asset-backed securities. Despite the many flaws of credit ratings as a sufficient statistic for credit risk, the rating agencies used ratings as the main measure of the quality of the subprime RMBS that the ABS CDOs invested in. And the rating agencies did only limited, ad hoc analysis of how the timing of cash flows affects the risk of ABS CDO tranches. As a result, the ratings of ABS CDOs should have been viewed as highly uncertain. As one risk manager put it, ABS CDOs were “model risk squared.” A final point on ABS CDOs is that the market’s reaction to the poor performance of ABS CDOs makes it clear that some investors did not understand the differences between corporate and structured-finance ratings. Because structured-finance securities are built on diversified portfolios, they have more systematic risk and less idiosyncratic risk than corporate securities. They will naturally be more sensitive to macroeconomic risk factors like house prices, and by design, downgrades of structured-finance securities will be more correlated and larger than downgrades of corporate bonds.

    Turning to the bottom panel, as Pat noted, in August of last year the subprime shock hit the ABCP markets, especially markets for ABCP issued by SIVs. Rating agencies also got it wrong for the SIVs. More than two-thirds of the SIVs’ commercial paper has been downgraded or has defaulted. The problem with the ratings was that the rating agencies’ models for SIVs relied on a rapid liquidation of the SIVs’ assets to shield the SIVs’ senior debt from losses. While this might have worked if a single SIV got into trouble, the market would not have been able to absorb a rapid liquidation by all SIVs at the same time. Once investors began to understand the rating model for SIVs, even SIVs with no subprime exposure could not roll over their commercial paper. Investors who thought they were taking on credit risk became uncomfortable with the market risk and liquidity risk that are inherent in a SIV’s business model.

    The next exhibit presents the staff subgroup’s recommendations for addressing the weaknesses in credit ratings for structured-credit products. A common theme of our recommendations is drawing sharper distinctions between corporate ratings and structured-finance ratings. First, we recommend that rating agencies should differentiate structured-finance ratings from corporate ratings by providing additional measures of the risk or leverage of structured-finance securities to the market along with the rating. We don’t make a specific recommendation on exactly what measures of risk or leverage because we believe rating agencies and investors should work out the details together (on this and the recommendations to follow). Second, rating agencies should convey a rating’s uncertainty in an understandable way. The ratings of ABS CDOs were highly uncertain because the models were so crude. This is what I call the Barry Bonds solution—put an asterisk on the rating if you have doubts about the quality. [Laughter] Third, we recommend more transparency from rating agencies for structured-finance ratings. What we need is not just a tweak to the existing transparency, but a whole new paradigm that actually helps investors get the information they want and need. For example, why can’t the rating agency pass on to investors, along with its rating, all the information it got from the issuer that it used to assign the rating? Fourth, we recommend that rating agencies be conservative when they rate new or evolving asset classes. Fifth, the rating agencies should enhance their rating frameworks for structured products. For example, when they rate RMBS, they should consider the originator as well as the servicer as an important risk factor. Our last recommendation is addressed to regulators, including the Federal Reserve. When we reference a rating, we should differentiate better between corporate and structured-finance ratings. Sometimes we do that already, but we could provide some leadership to the market by doing more. Now Bev will discuss the work on investor practices.

  • Mike has described how the rating agencies treated structured- credit products; a closely related issue is how investors used these ratings. Did investors rely too much on ratings in making their investment decisions? Did they take false comfort from ratings and not really appreciate the risks they were assuming, leading to excessive growth of the market for subprime structured credit? As noted in the top panel of exhibit 7, our approach was to examine these questions through the lens of one representative type of institutional investor: public pension funds. Public pension funds are an informative example of investor use (and misuse) of credit ratings for several reasons. First, public pension boards of directors are composed largely of representatives of the employees and retirees covered by the pension plan and have only limited financial expertise in some cases. Second, survey evidence suggests that a high portion of these funds use credit ratings in their investment guidelines. Finally, relative to some other investors, many public pension funds provide significant public information about their activities. While we need to be cautious in generalizing, we believe that practices in the pension fund sector reflect the tensions faced by other institutional investors in making risk assessments and investment decisions. We examined the investment practices and fund governance of 11 public pension plans. These plans ranged from the largest fund—CalPERS, with $250 billion in assets—to six much smaller plans with assets of $6 billion to $11 billion. We used the funds’ 2006 comprehensive annual financial reports, which were generally the most recent available, and the funds’ websites to generate our information. We focused specifically on the funds’ fixed-income segments, since this is the asset class in which structured-credit products would likely be held and for which credit ratings are used.

    The middle panel lists some key conclusions from this review. The first is that these funds have developed workable market solutions to address inexperience or lack of financial sophistication among their managers and board members. These include hiring professional investment managers to make investment decisions on their behalf and, perhaps as significantly, hiring investment consultants to structure asset-allocation strategies, to select investment managers and develop mandates guiding their actions, and to monitor and assess fund performance. While these funds clearly obtain significant professional advice in managing their investments, our review suggests several ways in which these arrangements could be improved in light of recent financial innovation. Specifically, the mandates guiding investment managers have not always kept pace with the growth of structured-credit markets. These mandates typically require managers to meet or exceed returns on a benchmark index or of a peer group of investment managers, while constraining the risk the managers may assume. Credit ratings play an important role in these risk constraints—for instance, by imposing a minimum average rating for the portfolio or a minimum rating on individual securities. However, few of the funds we profiled made significant distinctions between structured-credit and other securities in these credit-rating-based constraints, although there were sometimes other limits on the aggregate share of asset-backed positions. The failure to make this distinction provides scope for investment managers to generate higher returns by moving into structured-credit products, without raising warning signals about the additional risk these positions entail. This is not necessarily a “naïve” use of credit ratings by investment managers, as they could well have recognized that higher-yielding structured-credit products embodied significant additional risk relative to similarly rated corporate debt. Instead, it reflects a previously effective mechanism used by fund boards to convey risk appetite to these managers falling out-of-date with the emergence and rapid growth of a new form of credit instrument.

    As indicated in the bottom panel, our key recommendation is that the pension fund industry and other investors should re-evaluate the use of credit ratings in investment mandates. In a narrow sense, these mandates should distinguish between ratings on structured credit and those on more traditional corporate credit. However, the more fundamental point is that mandates would do a better job of enforcing desired risk limits on the overall portfolio if they acknowledged differences in risk, return, and correlation across instruments rather than relying on generic credit ratings. A second important point is that investors should ensure that their investment consultants have independent views of the quality and adequacy of credit ratings for the types of positions in their portfolios. This is particularly important if mandates guiding investment manager behavior feature credit ratings as a key risk constraint. That completes our prepared remarks. We would be happy to take your questions before we proceed to the final presentation.

  • Questions? President Lacker.

  • I am still confused. How much of these weaknesses would have been identified by an impartial observer in January 2006, say, without knowledge of what has happened since then?

  • Are you talking about the credit rating agency weaknesses?

  • There are plenty of investors who said, “We are staying out of that ABS CDO market because we don’t trust the ratings and we don’t think we have an ability to model it.” But there are also enough who were willing to take on that risk. So I think it is a combination.

  • Going back to exhibit 5, the second panel, your point that we have no evidence of conflicts of interest having an effect on ratings, I am thinking about the investor- practices presentation. There is an inherent conflict of interest because the issuers pay the raters. By one account, Moody’s earned 44 percent of their revenue in 2006 from rating structured products. I have always wondered—and by the way, I have been on the side of the table that has worked a rating agency and have gotten what I wanted—what about just a common sense solution, which is that the investors rather than the issuers pay the rating agencies?

  • That used to be the way it was before 1970—the investors paid. I think the common explanation for why that model faded was that, once photocopy machines came into existence, there was no way to constrain the information that they were providing just to subscribers.

  • I am well aware of that. But is there an inherent conflict of interest in the process?

  • There is. We are not denying that there is a conflict of interest. We looked at some of the mechanisms that rating agencies have put in place to address conflicts of interest, and that is something that every rating agency is aware of. The only question is whether the mechanisms that are in place to address conflicts of interest were working or whether they need to be enhanced. Our conclusion was that we don’t really have the detailed information to know whether there were specific conflicts of interest, and we are looking to the SEC to provide that detailed information. But lots of other things seemed to go wrong with the rating agencies that don’t rely on conflicts of interest as an explanation. I think people who have looked at the question of the issuer’s paying versus the investor’s paying feel that, given how costly it is to rate these things even in the mediocre way that it has been done, it would be difficult to generate enough revenue through a pure investor-pay model. That is not to say there couldn’t be more competition between the two kinds, and there are some proposals out there to do that.

  • Again, remember how costly it has proven to investors to have that built- in conflict of interest. That’s all. That’s my say.

  • Actually, I wanted to focus on the same issue—conflict of interest. This is something I did a little work on in the past. When you look at the standard corporate ratings market, the move to have the issuer pay really did not create a problem in the market. There clearly is an inherent conflict of interest, but there are things that can counter that—in particular, reputation.

  • We could talk about Arthur Andersen, too, because I am going to talk about the more complicated issues of conflicts of interest. With plain vanilla conflicts of interest, if there is enough information, the market frequently can solve the problem because if you know that if you do what the issuer wants and you give a good rating, then you lose your reputation. Then, if it has no value, issuers won’t pay for it. What is interesting here is that for the subprime market, you didn’t find any evidence of conflicts of interest, and I am not surprised by that, because those securities are much more straightforward. Where I really do worry about the conflict of interest is in the structured products because one thing that happened was that it became less plain vanilla. You actually had consulting practices inside the credit rating agencies; these structures are very complicated, and you need to slice here and dice here, and consultants were providing advice on structuring them and making a lot of money, and then it was much less transparent. What I wondered about here is that you didn’t say this for the first one, subprime RMBS. You said you didn’t find the evidence. I buy that. But what about the CDOs and the SIVs, for which I would expect that this problem would have been more severe? In the book that I wrote with others on conflicts of interest in the financial services industry, we actually said that there was not a problem with the plain vanilla products because the markets have the information, but we worried about exactly this issue in terms of the structured products. I am just wondering whether or not it was an accident that you said for the plain vanilla that there was less problem. Could there have been an issue here? The reason this gets complicated is that the standard view of conflicts of interest in Arthur Andersen was in the firm’s compensation scheme. Actually, the conflict of interest was that the Texas unit did not worry about and weakened—not their ethics, but what is it? The center has rules for its branches so that they don’t screw the overall firm, and that is what got weakened during the fight between the consulting part and the auditing part.

    So do you have any information on these very complicated elements, particularly the nontransparent parts? Was it an accident that you said for subprime that you didn’t find evidence, and for these is there more possibility that there was a problem? That really does have important implications for the nature of the regulation and accreditation agencies and also their ability to give good ratings for these very complex nontransparent products. You talked about investor practices later, Bev, when you said that we should differentiate between plain vanilla and this very complex stuff. I don’t know whether or not you have views on this.

  • There is certainly a possibility that conflicts of interest were occurring in all these areas. We are taking a somewhat neutral position because we don’t really have the detailed information to say more.

  • Right. But this is the typical Federal Reserve cautiousness, and I am pushing you a little harder.

  • I know, but I am a Federal Reserve economist. I would agree with your point that separating out these nonrating businesses from the rating businesses is important, and rating agencies have announced some changes along these lines. When we looked at our recommendations, we didn’t really feel as though, even if they separated out rating businesses from nonrating businesses and even if they did all the things that people who are concerned about conflicts of interest want them to do, it would solve the problem.

  • There would still be problems.

  • We think the real problem is that they didn’t differentiate well enough between structured-finance ratings and corporate ratings, and that is where we would like to put the focus. Securities regulators are already focused on conflicts of interest, and there are codes of conduct regarding how rating agencies have to behave, and those are monitored. So, coupled with the SEC’s already doing examinations, that wasn’t an area we chose to focus on, but we really can’t say for sure what was going on.

  • One quick follow-up on that—is one implication that we might take from this that structured products are just so complicated that they may never get good ratings or sufficiently good ratings, so the market really has to shrink? Could the rating agencies just fix themselves up so that they actually could do decent ratings? I’m trying to get a feel for this. It really relates to the issue that Bev raised, which is you want to differentiate between the two. But is there something inherently problematic so that maybe people should just realize that these structured products are just not such a great thing. Financial engineering can go too far.

  • I agree. In fact, ABS CDOs are disappearing or have disappeared. Investors don’t have any appetite for them. SIVs are disappearing as well.

  • But they could come back. I think subprime lending will come back under a different business model.

  • Really, the question is, What sort of market forces would produce that outcome? It certainly seems as though things should move in that direction, and we feel that some of the recommendations we are making on the differentiation between structured-finance securities and corporate securities would encourage the rating agencies to put more scrutiny on the structured-finance side of it. Yes, those ratings should be lower, especially when you factor in things like the complexity and the uncertainty.

  • My comment is on the conflicts of interest as well. The only thing I would like to add is that this is not the first time the rating agencies have miscalculated the risk and put that out there. The incentives are designed to do exactly that, and it will occur again. I think this statement is too generous to the rating agencies. The incentives are driving them to do this, they did it, and they will do it again in the future. It is just inherent that, when you are going to make that kind of money and if you can get it down to working with them and pushing this stuff out, it really is a matter, for those who use them, that you get what you pay for.

  • I didn’t know anything about SIVs before last summer, but I must say when I began to learn something about them, I was astonished because it would seem to me that, in Finance 101, you don’t finance long-term, risky assets with short-dated paper. There’s a maturity mismatch, and there’s a rollover risk, and we’ve known about that kind of thing for 100 years, I suppose. So my question is, Could these things have been marketed as standalone entities without the banks’ sponsorship, which was there in, I guess, most cases? Second, if they wouldn’t really fly as standalone entities, it would seem to me that the regulators should have insisted that they be consolidated on the books of the bank.

  • A lot of the SIVs were standalone entities managed by independent asset- management companies that were set up to do that.

  • And they collapsed?

  • All the SIVs are in the process of winding down now, and in the immediate aftermath of what happened in August, that was when there was a crunch in the ABCP market.

  • But when you say the asset-management companies, are they providing at least in principle some liquidity support or something?

  • SIVs had partial liquidity backup lines from banks covering one week’s maximum withdrawal or two weeks’ maximum withdrawal. That was the way to a rating. The SIVs were operating under the regulation of the rating agencies. The rating agencies modeled their portfolios, and they modeled the inflows and outflows, so I think it’s fair to blame the rating agencies at least partially for allowing the SIVs to grow to $400 billion or whatever it was.

  • There is a large, independent SIV not affiliated with a bank, named Gordian Knot, that is still in existence. It’s under severe pressure, but it has been weathering the storm. One can argue it might have been more conservative than some of the other SIVs.

  • Thank you, Mr. Chairman. My question is on exhibit 6, and the question is, Are we making any progress here? In particular, are the rating agencies stepping up to the process of doing different ratings, different nomenclature for the structured finance? My impression is that there is a lot of resistance, at least there was as of a few weeks ago.

  • Your impression may be more informed than mine, but I would say that the rating agencies are doing a lot of self-examination now. I haven’t seen them willing, in my opinion, to go far enough in the directions that we’ve outlined. A lot of their recommendations focus on managing conflicts of interest and doing a better job of separating compensation from the rating and things like that. I would say that I haven’t seen enough on the sort of recommendations that we are pushing.

  • Did you have any questions?

  • I’d add only that I think where the rating agencies are now is trying to come up with cleaner boxes and better governance—Sarbanes-Oxley types of structures, ombudsmen, liaisons, Chinese walls. The core issues that Mike and his team bring up seem highly resistant to change, but you know, there will be nothing like three months of public hearings, if not hangings. [Laughter] That could change that dynamic, but that’s in the early stages.

  • A few different points. One, on the conflict of interest issue, with respect to traditional corporate credits, it seems that credit rating agencies do pretty darn well and certainly have done very well before this. So it’s clear that it’s not just fundamental to the model that it can’t work. I think it’s clear enough to remind people of that and that people trusted those ratings throughout all the turbulence during the summertime. But that said, it gets back to some of the other points that have been raised. What was special about these particular areas that led to a breakdown? The question is, Well, why did they have bad models? If they were able to develop good models for these other things, in principle they are capable of developing good models, but they seem to have developed poor models here. Now, that could be succumbing to a conflict of interest, or it could be something specific in this area.

    I am reminded that another area in which they are perceived to have done a very poor job was in sovereign ratings back a couple of decades ago. I don’t know whether it’s worthwhile to drill down into the characteristics of where this model seems to be successful and doesn’t seem to be subject to significant conflicts of interest or where it does, whether it relates to particular information issues. Also, one thought that I had with respect to these structured products—and you should tell me whether this is true or not—is that most times when a corporation comes to get a rating they’ve decided to issue a particular security—that is, if GM comes, their CFO has decided to issue ten-year debt. If you come to a credit rating agency, however, the credit rating agency can say, “Well, why don’t you issue six securities rather than one? If we chop it up in these different ways, I can make more revenues off that, so I have more of an incentive to work with you on this.”

  • I definitely agree with you that it’s more difficult to rate structured-finance securities than corporate securities because with structured-finance securities you have a lot of quantitative modeling of future cash flows with a lot of uncertainty. Clearly the ABS CDOs are very complicated structures because it’s a two-layer securitization—a securitization that invests in securitizations—and the rating agencies didn’t drill all the way down to model the ultimate underlying loans. They relied on simplifying assumptions and aggregations that were very crude. But because they were branded as CDOs, people usually understood them to be corporate CDOs that invest in 100 corporate bonds, which is a simpler, one-layer structure. So the rating agencies were willing to rate these much more complicated things and investment banks were willing to market them to investors who were willing to buy them because they all were willing to believe that a CDO is a CDO is a CDO, and, in fact, that wasn’t really true.

  • Yes, it might just be interesting to drill down and see where the successes are and where they haven’t been successful and whether it has something to do with the model.

  • Oh, it’s definitely not a random sample of asset classes that we chose to look at. The rating agencies have done a great job in many other places, I agree.

  • Right, but just thinking about those differences, I think sometimes the rating agencies may be excessively maligned. They actually do provide something that’s very useful, and we don’t want to throw the baby out with the bath water. But also on that, there is a severe lack of competition among rating agencies, and we’ve tried to address that through some legislation, which doesn’t seem to have been very successful at addressing that issue. This is getting back to Governor Mishkin’s point and some other points about why competition doesn’t get us to a better solution. It seems that part of it is a regulatory structure that strongly discourages competition. Did you think about addressing that issue?

  • The regulatory structure has been changed now as a result of the law that was passed in 2006, at least if you’re talking about getting the regulatory stamp of approval from the SEC to become a nationally recognized statistical rating organization. That’s now just a notification process by which, if you meet some minimum requirements, the SEC is required to okay you. So entry is now easy whereas before it used to be restricted by the SEC. In some sense there’s a bit of a natural monopoly going on because, if you really had a couple of rating agencies that you trusted, it’s not clear why someone would be willing to transfer from an established to a new one. But if you were to take some of our recommendations to the next level and talk about how you would actually implement this and how you would go about making this have some success, one thing people have suggested is that maybe we should help investors set up their own rating agencies— maybe they don’t want to do that now—and maybe more transparency from the existing rating agencies and from the issuers. It would then be easier for a new rating agency that’s funded by investors to get going and get some traction if there were more transparency around the whole process. That’s just one possibility

  • For sure, although I think we tried something like that with the Sarbanes-Oxley issues of independent research—because this in some sense is a parallel to independent research—and even required funding of that, and that doesn’t really seem to have taken off. So it seems as though the model, even with its flaws, is the only one that seems to be sustainable. But drilling down more into where those flaws are, in some sense we’re seeing a privatization—I’m not quite sure what the right word is—with some of the very sophisticated investors effectively building their own internal credit rating agencies. That’s what hedge funds do, and so you’re getting in some sense a loss of the public rating—a kind of free rider problem or whatever is the right way to characterize it. In some sense there may be a loss because you’re getting this to be just purely private, but maybe that’s a gain because those are the only guys who should be in this game. It certainly makes it difficult to sustain the credit rating agency or industry.

  • We have kind of a dual problem in that we want investors to do a better job of evaluating the risks, and maybe better ratings would help on that. At the same time, the existing rating agencies did a bad job, so we have to criticize them. We have to find some way to reconcile those.

  • I have President Evans for a two-hander and then President Lockhart and President Lacker. I also see that President Plosser has a two-hander, and I think at that point we should probably go onto the last presentation.

  • I can pass if it gets too long. That’s okay.

  • Okay. All right. So President Evans, President Lockhart, and President Lacker.

  • I had an elementary question that Governor Kroszner’s question kicked off. The background briefing was very well written and very clear, which made me think that I understood things, and I probably don’t. But you used the term “get a rating,” and at other times you talk about it as if it’s a sufficient statistic and how the agencies were warned that it’s not a sufficient statistic. There are transition matrices to downgrade, and yet they violated that themselves because you said that for the ABS, they thought of it as a sufficient statistic themselves. Is a lot of this inherently a multidimensional risk model that’s required for some of these structured- finance vehicles, whereas for the corporates it’s closer to a single dimensional risk model?

  • I’d like to ask as a practical matter what weighting we should put in our expectations or even our recommendations on reform of the rating system versus the expectation that investors would actually reduce their reliance on ratings. My impression is that many institutional investors, especially public pension funds, are notoriously understaffed. Then they work for, as you pointed out in your presentation, boards that either represent the beneficiaries or in some respects are quite political in nature and not necessarily financially sophisticated. That would lead me to the conclusion that, as a practical matter, they’re going to be highly reliant on ratings. The ratio of professional employees to the volume of investment is so low that they have to choose their restaurants by stars because they don’t have time to do the tasting themselves or they’re not given the budgets to do that now. So I’m curious about this tension between getting the rating system right versus having the investors reduce their reliance on ratings.

  • Well, I think you’ve identified the key problem that many of these funds face, and the important entities in many ways are these investment consultants. At least the funds we looked at all hire one or more of these investment consultants. Many of them are from large, globally active firms, and the role that those consultants play is precisely to address the issues that you’ve raised. They help them select the asset manager who is doing the investing. They help them design the mandates or instructions. They help them monitor what the asset managers are doing and assess their performance. So to some extent those consultants are a kind of sweet spot for these funds in terms of where the additional layer of sophistication that doesn’t run just off the rating could come from.

  • But is your impression that those supplemental parties, the consultants and the asset managers themselves, have the credit analysis and valuation capability to address sophisticated structured products?

  • We didn’t do an evaluation of them, but some of them are from very, very large firms. They are advising funds in the trillions of dollars across all their different clients. That is the service they are selling, and so they should be—they had better be—in a position to do that.

  • May I just suggest that you reexamine that proposition? They usually evaluate performance principally after going through initial screening devices. They almost never, that I’m aware of, provide the kind of analysis that you’re assuming. So I would reexamine that proposition.

  • My memory of Finance 101 is that you don’t buy long-term, risky assets funded by short debt without sufficient compensation. So the neglected aspect of the whole episode we went through was how people would have been compensated in other circumstances, but that’s a different discussion. I think these recommendations make sense, and confirmation of that is that the market is moving in that direction anyway. I don’t know how many of these recommendations were put forward by those market participants you cited who two years ago were not trusting these ratings—I think that’s a key question. But the broader thing here is that the world of finance now is a world of competing econometric analysis, and we’re never going to rate the model risk if you ask the question, What would you expect the behavior of a market like that to display? You’d expect occasional big misses from erring assumptions. You’d expect those misses to be corrected by people reflecting and improving practices going forward. A sort of Schumpeterian process in this is going to generate model risk and big model misses, and so I think it’s being careful to keep track of whether the insights we’ve gleaned are hindsight or should have been known ahead of time.

  • Okay. Thank you. Why don’t we continue with the final portion?

  • Thank you. To support the Financial Stability Forum’s Working Group on Market and Institutional Resilience, as noted in the top panel of exhibit 8, supervisors from France, Germany, Switzerland, the United Kingdom, and the United States formed the “senior supervisors group” in late October. Participating U.S. supervisors included the OCC and SEC as well as the Federal Reserve. The group’s goal was to develop a common understanding, through a series of interviews with selected firms, of how the risk-management systems of core financial institutions performed during the financial market turbulence. The top right panel of exhibit 8 shows the 11 banking firms that supervisors interviewed. This effort was not a complete review of all firms and events. For example, we did not meet with Bear Stearns or Morgan Stanley as part of this effort. Rather, it was designed to inform supervisory authorities about the general effectiveness of risk management at global financial institutions. The supervisors have prepared a paper detailing their findings, which will be conveyed to the FSF and released publicly. The bottom panel lists some observations about the firms’ overall performance. Most large financial services firms, while affected by market developments, generally avoided significant losses. Although most firms’ risk-management processes worked as intended, there were some definite outliers. Some firms recognized the emerging additional risks and took deliberate actions to limit or mitigate them. Others recognized the additional risks but accepted them. Still other firms did not fully recognize the risks in time to mitigate them adequately. Moreover, the risk-management practices varied by firm and by strategy, as did the range of outcomes to date. I should note that the primary risk-management weaknesses observed here are not new. They have been observed in past episodes and are thoroughly discussed in existing risk-management literature and supervisory guidance.

    As noted in the top panel of exhibit 9, the senior supervisors group identified four primary factors that differentiated the organizations that suffered larger losses from those that did not: (1) the effectiveness of senior management oversight of balance sheet, liquidity, and capital positions; (2) the effectiveness of communications among senior management, business lines, and risk-management functions; (3) the sophistication, diversity, and adaptability of risk measures utilized; and (4) the attention devoted to valuation issues. With respect to senior management oversight, as indicated in the bottom panel of exhibit 9, the more effective firms were more disciplined in measuring and limiting these risks in advance of the crisis and proved to be more agile in reducing exposures or hedging when the crisis occurred. These firms focused on maintaining a strong balance sheet with strong capital and liquidity positions throughout the entire organization. Senior management of these organizations had established adequate capital and liquidity buffers that could sustain the firm through a period without access to the market for funding. They have created and effectively enforced internal pricing mechanisms, capital allocation methodologies, and limits that provided effective incentives for individual business line managers to control activities that might otherwise lead to significant balance sheet growth or contingent liquidity demands. Conversely, the less effective firms were not as focused on the overall strength of their balance sheet across all legal entities and thus operated with more limited liquidity and capital buffers. These organizations had weaker controls over their balance sheets and were more focused on earnings growth or defense of a market leadership position. These firms did not have limit structures that were consistently or effectively enforced, which allowed business lines to grow balance sheet exposures rapidly and increase contingent liquidity exposures. They did not properly aggregate or monitor off-balance-sheet exposures across the organization, including the exposure to contingency back-up lines of credit to ABCP programs and generally did not have in place effective financial controls, including capital allocation processes, commensurate with the business strategy.

    The top panel of exhibit 10 provides additional detail on the importance of effective communications among senior management, business lines, and risk- management functions. The more effective firms emphasized a comprehensive, firmwide, consolidated assessment of risk. Senior managers of these organizations were actively engaged and had in place a disciplined culture and well-established processes for routine discussion of current and emerging risks across the business lines, risk management, and the corporate treasury function. Senior managers at these organizations collectively made decisions about the firm’s overall risk appetite, exposures, and risk mitigation strategies rather than relying solely on the judgment of business lines. They were able to effectively leverage the assessment of risks from one business line to consider how subprime exposures, for example, might affect other businesses. As a result, the more effective firms had a more timely and well- informed perspective on how market developments could unfold. In some cases, senior management had almost a year to evaluate the magnitude of the emerging risks from subprime mortgages on its various business lines. This, in turn, enabled them to implement plans for reducing their exposures while it was still practical and more cost effective to do so. Conversely, less effective firms were siloed, did not effectively share information across business lines on emerging risks, and were comparatively slower in taking actions to mitigate exposures as each business line had to assess and consider emerging risks on their own without the benefit of views or actions taken by other business line managers.

    With regard to the risk measures utilized, as shown in the bottom panel, the more effective firms used a wide range of risk measures and analytical tools to discuss and challenge views on credit and market risk broadly across different business lines within the firm in a disciplined fashion. These firms have thought more thoroughly about the interplay of their risk measures than the other firms and used a combination of different risk measures and scenario analysis to understand risk exposures. It also appears that the more effective firms had committed more resources to risk- management and management information systems. As a result, they had more timely and scalable management information systems and in large part did not have to create new management reports to understand risks and exposures. Conversely, the less effective firms were too dependent on a single quantitative risk measure, and they did not utilize scenario analysis in their decisionmaking and tended to apply a “mechanical” risk-management approach. Management information systems also were not as scalable, and there was a need to develop a number of ad hoc reports to help senior management understand the risks and exposures of the company.

    The top panel of exhibit 11 elaborates on the fourth factor that proved critical, which is the attention devoted to valuation issues. The more effective organizations were more disciplined in how they valued the holdings of complex or potentially illiquid securities. They employed more-sophisticated valuation practices and had invested in the development of pricing models and staff with specialized expertise. These organizations were skeptical of and less reliant on external ratings and emphasized mark-to-market discipline in their businesses in ways that others did not. Less effective organizations in some cases did not have key valuation models in place prior to the market disruption, relied heavily on third-party views of risk, and tended to have a narrower view of the risks associated with their CDO business as mainly being credit risk and did not actively seek market valuation information. The bottom panel explains how supervisors are planning to address the specific deficiencies. As I mentioned earlier in my presentation, the risk-management deficiencies identified during this exercise are not new, and existing supervisory guidance addresses these issues. Therefore, supervisory efforts will include addressing risk-management deficiencies at each company through the supervisory process and re-emphasizing the importance of strong, independent risk management through a series of speeches, industry outreach, and possible re-issuance of existing guidance. In addition, supervisors plan to complete the work already under way within the Basel Committee on Bank Supervision to update liquidity risk management guidance to strengthen industry practices. A review of existing Federal Reserve guidance on market and liquidity risk management is under way to ensure that it effectively outlines the need for banks to use a number of tools to include multiple ways of viewing quantitative and qualitative risk analysis, including VAR, stress tests, and scenario analysis. Finally, supervisors plan to develop, on an interagency basis, guidance related to the management of the originate-to-distribute model to ensure that banking organizations effectively manage the credit, market, and operational risks of this activity. I will now turn it over to Art Angulo to discuss related regulatory policies.

  • The top panel of exhibit 12 sets forth the question we sought to address—namely, to what extent did regulatory incentives contribute to or fail to mitigate weaknesses exposed by the recent turmoil? In doing so, we defined the term “regulatory incentives” to encompass both regulatory capital and financial reporting requirements. In addition, we distinguished between policies that may have mattered for the buildup to the market turmoil and those that have made managing the turmoil more challenging. Our conclusions are summarized in the middle panel. Not surprisingly, incentives to minimize regulatory capital are a much more important driver of bank behavior than financial reporting incentives. Moreover, the current regulatory capital framework is not neutral as to how banks structure risk positions. Both the leverage ratio and the Basel I risk-based capital framework have encouraged banks to securitize low-risk assets and, importantly, to support securitizations of higher-risk assets through instruments with low mandated capital charges, such as 364-day liquidity facilities. Financial reporting incentives were not critical to banks’ decisions, although certain financial reporting issues—particularly disclosure practices—have been a factor in how the turmoil has been unfolding.

    Before I turn to specific recommendations, it is important to emphasize that improvements were already in train even before the market difficulties emerged last summer. Most significantly, the Basel 2 advanced approaches and related improvements greatly enhance the risk sensitivity of the regulatory capital framework and create incentives for better risk management. In the bottom panel are three examples relevant to the issues we have been discussing this morning. First, capital charges for most unused short-term credit and liquidity facilities have been increased to more adequately reflect the risk exposure. Second, new standards for banks to hold capital against the default risk of complex, less liquid credit products in the trading book are being finalized, as the Basel Committee is currently seeking public comment on principles for calculating a so-called incremental default risk charge against such products. I should note that the work to develop the incremental default risk principles was done by a joint Basel Committee–IOSCO working group co-chaired by Norah Barger. Third, the securitization framework in Basel 2, which builds on previous unilateral U.S. enhancements, establishes a more risk-sensitive capital treatment for securitization exposures—it bases capital charges on estimates of underlying risk rather than on an exposure’s legal form.

    I will now turn to recommendations in exhibit 13, beginning first in the top panel with those related to regulatory capital. First, notwithstanding the improvements brought about by Basel 2, there is scope for reassessing the treatment of securitizations involving ABS CDOs. The risk weights and resulting capital charges for these “re-securitizations” were calibrated under the assumption that loss correlations within a pool of securitized assets would be no greater than those exhibited by CDOs backed by traditional corporate bonds. It is now apparent that the actual loss correlations within such re-securitizations—especially for the most highly rated tranches—may be much higher. At its December 2007 meeting, the Basel Committee agreed to take up this issue.

    The second recommendation addresses the issue of “reputational” risk. There have been several occasions over the last 6 months in which banks have elected to purchase assets from, or extend credit to, off-balance-sheet vehicles they had organized and money market and other investment funds they managed, even though they were not contractually obligated to do so. Whether a bank management will provide support in excess of its contractual obligations is a business decision. Thus, it is not practical to attempt to design an explicit capital charge for reputational risk. Instead, supervisors should exercise supervisory oversight to ensure that banks sufficiently consider reputational risk and its implications for capital and liquidity buffers.

    The third and fourth recommendations deal with the issue of the procyclicality of capital regulations. The existence of fixed capital requirements (as well as rating agency expectations for capital ratios) discourages banks from drawing on their capital cushions in times of stress. It is therefore very difficult to devise changes in capital regulations that would allow capital to function more effectively as a shock absorber without compromising safety and soundness objectives. Nonetheless, these last two recommendations are aimed at mitigating the potential procyclical effects of a more risk-sensitive capital framework. The third recommendation is based on elements of the Basel 2 framework that were designed to allow supervisors to address both safety and soundness and procyclicality concerns. The advanced internal ratings based approach of Basel 2 requires that banks’ loss-given-default estimates reflect economic downturn conditions and that stress tests of their advanced systems include consideration of how economic cycles affect risk-based capital requirements. Rigorous supervisory evaluation of these two elements can help ensure that (1) regulatory capital embeds forward-looking forecasts of recovery rates and (2) banks manage their regulatory capital positions in a manner that enables them to accommodate variations in the amount of minimum required capital over an economic cycle. The fourth recommendation is to explore ways to encourage inclusion in the regulatory capital base of debt instruments that mandatorily convert into equity when a banking organization is under stress. The automatic conversion of debt instruments into equity under such circumstances is appealing from a supervisory perspective as well as from a broader macroeconomic policy perspective. This concept has not yet been examined in depth by the staff, but we believe it merits further study, including an assessment of the past experiences of banks with such instruments, to determine its feasibility.

    Let’s now move to the recommendations in the bottom panel. These deal with improving disclosure practices, which can help to reduce the uncertainty that has been a key feature of the recent turmoil. First, financial institutions—especially those in the United States—improved their disclosures about subprime-related exposures as the turmoil wore on. Nonetheless, supervisors and regulators should continue to push market participants to make timely and detailed disclosures about the size and composition of subprime-related exposures. The second and third recommendations focus on disclosures related to asset-backed commercial paper programs. In view of the potential exposure associated with reputational risk, market participants appear to desire additional details about banks’ dealings with ABCP programs. Disclosing information about the distribution of assets underlying such programs by type, industry, and credit rating would bring the disclosures on par with those provided for on-balance-sheet assets. Similarly, in view of the extent to which investors in asset- backed commercial paper have retreated from rolling over or purchasing paper that they suspect may be supported by subprime-related assets, banks and asset managers that sponsor ABCP conduits should improve disclosures to investors, particularly for conduits other than the traditional multi-seller programs. That completes our prepared remarks. We would be pleased to take your questions at this time.

  • Thank you. Questions? President Rosengren.

  • It is great to see some bank supervision people at this table, and I would just highlight one of the comments that you made about silos. It is interesting that this morning we have been discussing issues of bank balance sheet constraints and how that would occur, and it might be useful to think structurally within our own organization whether there are ways to do a better job of getting people in bank supervision to understand some of the financial stability issues we think about, and then vice versa. Maybe having some bank supervision people come to FOMC meetings might be one way to actually promote some of this.

    In terms of the things that you were talking about, another issue that I think has been important is that we’ve been talking about the effect of dropping housing prices. I know a horizontal stress review was done about a year ago, and I’m sure Brian or Jon remembers that the large financial institutions did the stress testing. When they did that stress testing, what was striking was that there were four institutions—I think it was Citigroup, JPMorgan Chase, Wachovia, and Bank of America—in that stress test, and all four concluded that a housing-price reduction of between 10 percent and 20 percent would affect earnings but wouldn’t affect capital. Obviously, in retrospect that doesn’t seem to have been a good forecast. One, since we do think that stress testing is useful, maybe understanding that and going back to those same four institutions and understanding their stress testing might be a useful exercise to do. Two, if you did that exercise, we’d learn something about how they’re thinking about housing prices and the indirect effects that might occur because one of our concerns around the FOMC table is that there may be unintended consequences if housing prices drop more than they have historically. Just as the banks have to think about those kinds of stress tests and what they have already learned from the fact that they didn’t pick up some of the indirect effects, our own knowledge would be supplemented if we thought about some of those indirect effects as well.

    So the horizontal stress testing was interesting in that I think some of those institutions are on the good side and the bad side of your things. The horizontal stress testing isn’t what generated the decisions—and I do agree with your conclusion that most of the decisions were made by senior management. Some organizations didn’t do subprime mortgages at all. I’m not so sure it was generated from the stress testing as much as a gut feeling by senior bank management that they weren’t going to engage in subprime mortgages. I think that’s true for a lot of these activities. It’s interesting who’s at the top of the list and who’s at the bottom of the list for a lot of these activities. Ideally, over time, both bank supervision and bank risk management would get to the point that it’s not just a gut decision by a person at the top but is a little more systematic and that the risk- management process does that. Are you thinking of ways that we could actually encourage that kind of interaction so that it becomes less gut from senior management and more integrated into the risk management? I know that was kind of a long entry.

  • Well, I would just comment on the first point that you made in terms of silos. I guess we would fully agree we’re having a great time here, so we would love to see more interaction between supervision and the FOMC.

  • Be careful what you wish for.

  • You know, when people debate whether to have supervision with a central bank or not, one of the key arguments is that only we can bring to bear the resources of the entire central bank and also provide input from the supervision side. So clearly, that’s something I think we need to do more of. In terms of the second question, do you want to take a fresh crack at that, Brian?

  • Yes. My perception is that the people who were further down on the list made active decisions to get out of the business or reduce their scale of the business, not necessarily just on gut feeling but because they had invested earlier on in the staff and the models to evaluate the product space. So they understood that the economics of the subprime business had deteriorated and that the risk relative to the return was rising, and they made an active decision, usually, to retract a bit from that space. Now, they may have made those decisions gradually and incrementally over the year from mid-2006 forward. It wasn’t one stress test, but it was a thorough understanding of the economics of the actual business that drove their decisionmaking.

  • I would also agree that your general point about banks being more exacting in their stress testing is a good one because it’s something we and other supervisors, to be honest with you, continually look at, and many times the response that banks get internally from either owners or managers or through supervisors is that, well, if it’s too extreme, it’s not really plausible, and you can’t really act on it. The other argument you sometimes hear is, well, it costs a lot of money to integrate that across the firm and do it the right way, and that’s a tough sell. So I agree with your point that we need to do more and then push harder on stress testing. The way that you suggest is kind of novel: You do a back test and show it to them. That’s an interesting concept.

  • I might add one thing to that. In interviews we did with some firms, one point that they made was that stress testing was informative and important, but a next step they’re trying to take is what it would take to cause that stress event to happen for different types of assets. So they are trying to anticipate—not so much that they just assume a 10 percent drop but how they would get to a 10 percent drop. What would be the events that would cause that to happen? Then they think that through in a more systematic way. I know some companies have learned what their highly rated things are that are viewed as stable-value assets elsewhere on their balance sheets and are thinking about how those could start to unwind or to deteriorate.

  • We need to be careful with the nomenclature. Stress testing right now in the industry encompasses a lot of things, and people will tell you that they do stress testing, but they are really doing static shocks. What we are talking about is not a historical analysis but a kind of forward-looking scenario that builds in your view of the world. Those are the firms that, again, made more-active decisions to reduce their exposures.

  • I think this was very interesting, and I have a suggestion and then a question. The suggestion is that this morning we had the Federal Open Market Committee meeting. This afternoon we should have the Federal Market Oversight Committee meeting, and that is what we’re doing right now in looking at some of these institutions. My question is, As you look at these and you compare the most effective and least effective, are you in the process and shouldn’t we be in the process—because we go back and look at these horizontal reviews—of looking at what lessons we’ve learned about more-effective institutions and less-effective institutions during the horizontal review so that we can be more proactive in terms of the outcomes that we’re now seeing? If we do that in a systematic fashion over time, we could anticipate not all—we learned a long time ago that you cannot anticipate it all—but some of these differences and put more pressure on some institutions that were not doing good risk analysis to step up to it before it becomes a crisis. I don’t know if we’re doing lessons learned for ourselves in that regard.

  • I would say we’ve looked at a lot of that. The one place I would highlight most articulately is the degree of international coordination and cooperation we’ve had under way. The core firms in the United States are one portion of this system now, with a number of very significant global competitors. So the coordination between us, the OCC, the SEC, the U.K. FSA, and the other senior supervisor groups has been really necessary for everyone to get a good understanding of practices. It’s perhaps more valuable to some of the foreign supervisors who have only one or two large firms under their jurisdictions. But even from our vantage point of what our direct supervisory responsibility is, the consolidation within the industry has collapsed the number of firms that we view directly.

  • I might add that we are trying to look at this. It’s not just an issue of guidance, but we are going back through the supervisory process in terms of what we’ve learned through these interviews and challenging our beliefs and prior assessments.

  • I think that is good because, if you think about it, we have the Basel capital standards coming forward, which involve the advanced method—relying on those institutions to a great degree in terms of judging their capital and their capital levels. When they are under pressure, they will tend to work the model. It is natural. As we have said before, the incentives are that they will work the model. So if we are not taking lessons learned from this in terms of anticipating those behaviors, I think we will repeat history in the surprises that we get.

  • Just very quickly, you did a terrific job. I think it’s important, though, to recognize that this isn’t done yet, and we’re not going to know fundamentally how we feel about the relative strengths and weaknesses in the system until we see how this plays out. Don’t let these initial presumptions—either the diagnosis or the prescription, particularly your list of prescriptions—harden too much because there are some judgments that we’re just not going to be able to make until the dust settles and we have a little time for reflection in that context. I think a lot of damage has been done to the credibility of our financial system. It’s not clear how much damage because we don’t know how this is going to play out. But damage has been done, and we are going to bear a lot of the burden of figuring out how to craft a compelling policy response, recognizing of course that regulation may be part of the problem and won’t necessarily be part of the solution. Anyway, mostly I just meant to compliment you. You did a great job, and I think it’s helpful really to have this much work done early on in getting us to the point where we know what we’re going to do to the system to make it less vulnerable to this in the future. Even as we manage the crisis, I think it’s good to have made that investment and a good tribute to the strength of the system that we were able to devote these quality resources even though we’ve all been busy managing the storm.

  • Let me also thank the staff for an excellent presentation. This is just the tip of the iceberg in terms of the work that’s being done on all these different topics, and in turn we’re collaborating with our colleagues here in the United States and abroad, and I guess we’ll keep working and hope to find some valuable lessons out of this experience.

    I’ve been asked to remind people that you have until tomorrow close of business if you wish to revise your projections. Brian Madigan raised a few questions about consistency. If you want to think about that, of course, feel free to do so. Our next meeting is March 18. I look forward to seeing you then, and there’s a lunch available for those of you who can stay. Thank you very much. The meeting is adjourned.