Full text of Federal Open Market Committee Meeting Minutes, Transcripts, and Other Documents : Meeting, February 2-3, 1999 : Transcript

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Meeting of the Federal Open Market Committee
February 2-3, 1999
A meeting of the Federal Open Market Committee was held in the offices of the Board
of Governors of the Federal Reserve System in Washington, D.C., on Tuesday, February 2, 1999,
at 2:30 p.m. and continued on Wednesday, February 3, 1999, at 9:00 a.m.
PRESENT: Mr. Greenspan, Chairman
Mr. McDonough, Vice Chairman
Mr. Boehne
Mr. Ferguson
Mr. Gramlich
Mr. Kelley
Mr. McTeer
Mr. Meyer
Mr. Moskow
Ms. Rivlin
Mr. Stern
Messrs. Broaddus, Guynn, Jordan, and Parry, Alternate Members of
the Federal Open Market Committee
Ms. Minehan, Messrs. Poole and Hoenig, Presidents of the Federal
Banks of Boston, St. Louis, and Kansas City respectively
Mr. Kohn, Secretary and Economist
Mr. Bernard, Deputy Secretary
Ms. Fox, Assistant Secretary
Mr. Gillum, Assistant Secretary
Mr. Mattingly, General Counsel
Mr. Baxter, Deputy General Counsel
Mr. Prell, Economist
Messrs. Alexander, Cecchetti, Hooper, Hunter, Lang,
Lindsey, Rolnick, Rosenblum, Slifman, and
Stockton, Associate Economists
Mr. Fisher, Manager, System Open Market Account
Mr. Ettin, Deputy Director, Division of Research
and Statistics, Board of Governors
Mr. Winn, 1/ Assistant to the Board, Office of
Board Members, Board of Governors
_________________________
1/ Attended Wednesday’s session only.

2/2-3/99

Messrs. Madigan and Simpson, Associate Directors, Divisions of
Monetary Affairs and Research and Statistics respectively,
Board of Governors
Mr. Reinhart, Deputy Associate Director, Division of Monetary
Affairs, Board of Governors
Mr. Dennis, 2/ Assistant Director, Division of Reserve Bank
Operations and Payment Systems, Board of Governors
Messrs. Reifschneider 3/ and Small 3/ Section Chiefs, Divisions of
Research and Statistics and Monetary Affairs respectively,
Board of Governors
Ms. Kole, 4/ Messrs. English 4/ and Rosine 4/ Senior Economists,
Divisions of International Finance, Monetary Affairs, and
Research and Statistics respectively
Ms. Garrett, Economist, Division of Monetary Affairs, Board of
Governors
Mr. Evans, 2/ Manager, Division of Reserve Bank Operations and
Payment Systems, Board of Governors
Ms. Low, Open Market Secretariat Assistant, Division of
Monetary Affairs, Board of Governors
Mr. Conrad, First Vice President, Federal Reserve Bank of
Chicago
Messrs. Beebe, Eisenbeis, Goodfriend, Hakkio, and Rasche,
Senior Vice Presidents, Federal Reserve Banks of San
Francisco, Atlanta, Richmond, Kansas City, and St. Louis
respectively
Messrs. Altig, Bentley, and Rosengren, Vice Presidents, Federal
Reserve Banks of Cleveland, New York, and Boston
respectively
_________________________
2/ Attended portions of meeting relating to the examination of the System Open Market
Account and changes to the domestic securities lending program.
3/ Attended portion of meeting relating to the Committee’s consideration of its monetary
and debt ranges for 1999.
4/ Attended portions of meeting relating to the Committee’s review of the economic
outlook and consideration of its monetary and debt ranges for 1999.

Transcript of Federal Open Market Committee Meeting of
February 2-3, 1999
February 2, 1999 – Afternoon Session
CHAIRMAN GREENSPAN. Good afternoon, everyone. Before we get started, I
would like to welcome Bob Rasche, the new Director of Research at the St. Louis Bank.
Also, it has been brought to my attention that today is Bill Conrad’s fortieth anniversary
of service at the Chicago Bank. [Applause] It is a rare event when somebody can
tolerate the rest of us for that long!
We will start off, as we do in the first meeting of each year, with the election of
officers. I will turn the gavel over to Governor Rivlin to bring us a Chairman and Vice
Chairman for this year.
MS. RIVLIN. Thank you. It is not clear by what authority you were making
those announcements! [Laughter] I would like to open the floor for nominations for
Chairman of the FOMC. Governor Kelley.
MR. KELLEY. After long and careful consideration, I nominate Alan
Greenspan.
MS. RIVLIN. Is there a second?
SEVERAL. Second.
MS. RIVLIN. Any discussion at this point?
MR. BOEHNE. I would like to hear Governor Kelley’s long and careful
reasoning!
MR. KELLEY. That’s confidential!
MS. RIVLIN. All in favor say “aye.”
SEVERAL. Aye.
MS. RIVLIN. Opposed? I believe we have a Chairman.
CHAIRMAN GREENSPAN. I thank you, everyone.
MS. RIVLIN. Since I still have the gavel, I will entertain nominations for Vice
Chair of the FOMC.
SPEAKER(?). I nominate President McDonough of the New York Fed.
MS. RIVLIN. Good idea.

2/2-3/99

SEVERAL. Second.
MS. RIVLIN. Any further nominations? All in favor say “aye.”
SEVERAL. Aye.
MS. RIVLIN. Opposed? We have a Vice Chair.
VICE CHAIRMAN MCDONUGH. Thank you all.
CHAIRMAN GREENSPAN. Democracy works again. I would like to ask
Normand Bernard to read the list of potential staff officers.
MR. BERNARD.
Secretary and Economist, Donald Kohn;
Deputy Secretary, Normand Bernard;
Assistant Secretaries, Lynn Fox and Gary Gillum;
General Counsel, Virgil Mattingly;
Deputy General Counsel, Tom Baxter;
Economists, Michael Prell and Karen Johnson;
Associate Economists from the Board: Lewis Alexander,
Peter Hooper, David Lindsey, Larry Slifman, and
David Stockton.
Associate Economists from the Federal Reserve Banks:
Stephen Cecchetti proposed by President McDonough;
William Hunter proposed by President Moskow;
Richard Lang proposed by President Boehne;
Arthur Rolnick proposed by President Stern; and
Harvey Rosenblum proposed by President McTeer.
CHAIRMAN GREENSPAN. Would somebody like to move that list?
VICE CHAIRMAN MCDONOUGH. So move.
CHAIRMAN GREENSPAN. Is there a second?
SEVERAL. Second.
CHAIRMAN GREENSPAN. All in favor say “aye.”
SEVERAL. Aye.
CHAIRMAN GREENSPAN. The ayes have it. The next item on the agenda is
the selection of a Federal Reserve Bank to execute transactions for the System Open
Market Account. That, of course, has traditionally been the New York Bank, and unless
I hear objections, I will presume that the Committee has again authorized the same
Bank.

2/2-3/99

Next is the selection of the Manager of the System Open Market Account. Our
incumbent is Peter Fisher. I assume that his nomination is acceptable to the New York
board?
VICE CHAIRMAN MCDONOUGH. It is indeed.
CHAIRMAN GREENSPAN. If there are no objections except mine, [laughter] I
will assume that the Committee is again authorizing our incumbent, Peter Fisher.
We have an examination report on the System Open Market Account. Are there
any questions on that report? If not, I will assume that it is acceptable without objection.
Next we have a review of the System’s security lending program and a related
proposal to amend the Committee’s Authorization for Domestic Open Market
Operations. Peter, would you like to review briefly what this involves?
MR. FISHER. Yes, let me mention a few points, if I may. The Committee
reviews the Authorization for Domestic Open Market Operations at the first meeting
each year. I have proposed an amendment to the Authorization for the Committee to
consider. You have a rather lengthy memo from me on the subject. There are a few
points I would like to try to cover again. I think I could have done a better job in the
memo in providing a clearer contrast between the existing program and the proposed
program. Let me try to do that very briefly.
Under the current program we lend securities for one week. We do it in
accordance with the rather low limits on bills and coupon securities as described in the
memo. We charge a flat fee of 150 basis points, regardless of the security’s current
value in the market, and there is a prohibition on short selling. That is, we are not
supposed to lend to dealers who have sold the security short, even though that is
virtually impossible for us--and in some cases for the dealer--to determine with
accuracy.
A major change we are proposing is that instead of charging a flat fee, we would
hold an auction with a minimum reserve fee of 100 basis points. In effect, for the
dealers who want the security, this provides an allocation mechanism that involves
something other than a first-come first-served, helter-skelter basis of who calls the Desk
first. An auction is also a mechanism that allows us not to waste public assets, so to
speak. That is, when we charge a flat fee of 150 basis points for something that is worth

2/2-3/99

more than 150 basis points “on special” in the Treasury financing market, we are giving
away System property, if you will, by accepting less income than the System could
receive.
The second change that we are proposing is in the limit structure. Instead of the
rather low limits, higher limits are proposed, as explained in the memo. The most
important shift is from the one-week term of the lending to overnight lending. I don’t
think I underscored quite enough in the memo what an important shift that would be. I
have previously described to the Committee my concern that any change in this program
should not simply move both the demand and supply curves out equally. That would not
accomplish very much. The longer we thought about this, the more we realized that the
willingness of dealers and market participants to short securities is really based on term
financing. No one shorts a security and plans to finance it through a series of overnight
borrowings. They think in terms of one-week or longer time periods when they are
trying to finance their short sales. That is where the demand and supply for shortening
securities come from. By limiting the System’s lending to an overnight program where
transactions occur only at noon, which is rather late in the day, I think we would do a
better job of providing a lending facility that addresses clearing issues and keeps the
market moving, without encouraging dealers to short securities. We are trying to stand
apart from short selling activities.
On the 25 percent limit that I suggested we begin with, we would actually be
lending less than we are lending currently, as a rough-cut calculation. I noted that I
would clearly expect to raise that limit somewhat as I saw increased demand, but I
certainly would not do so without talking to this Committee. As I also indicated in the
memo, it is hard to imagine raising that limit at any one auction to anything higher than
50 percent of our total holdings in an issue.
We think of the proposed program changes as enhancements that will better
enable us to achieve our existing objectives. These changes are designed to provide a
little grease for the clearing mechanism and to enable the dealers to avoid some of the
games that go on when a security is so scarce that borrowing it has essentially the same
cost as failing to cover. Then people may not try to cover shorts at all and that creates
some perversions.

2/2-3/99

That is the nature of the proposed program. As we have designed it, it is still
intended to be a limited program. I would be happy now to answer any questions about
what I have just said or about any aspect of the memo that I circulated to the Committee.
MR. BROADDUS. Peter, thank you for the memorandum. It was very detailed
and complete. In a sense, the memorandum presents this as an operational issue, but it
seems to me that this proposal is closely related to broader issues about how we manage
our portfolio and where along the maturity curve we operate and so forth. There is
something of the sentiment of the old “bills only” controversy that the Committee had
around this table many years ago. I have a comment that is perhaps half observation and
half question. The memo seems to imply that in order to run monetary policy effectively
and engage in open market operations effectively, we have to operate all along the curve
including issues of longer maturity. I am not sure that is true. It seems to me that we
could run monetary policy entirely, if we needed to, by operating only at the short end of
the curve.
The question I have is: Doesn’t this proposal raise some potential problems and
issues? It seems to me that by getting involved in this kind of operation, we could
become engaged in arguments as to why we decide to make a loan of securities in a
particular case and not make one in some other case. Whenever we conduct one of these
operations, especially if it involves a thinly traded issue, we are likely to affect prices.
And in such a situation, somebody is going to lose and somebody is going to gain. I
don’t think it is too much of a stretch to think that conceivably some moral hazard issues
might arise if we conduct lending operations on a regular basis. I just wanted to throw
that out and ask if you have thought about some of these broader issues in addition to the
more detailed technical issues that you covered in the memorandum.
MR. FISHER. It certainly was not my intent that any aspect of the program
would suggest anything about the maturity structure of the portfolio; and if the memo
suggests that, I want to correct that misimpression. In fact, the demand for borrowing
comes as much from the bill sector as the coupon sector. So, I really think this proposal
is independent of the structure of the portfolio. We could talk about that, and we have in
the past, but I think this proposal stands apart from that debate.

2/2-3/99

On your second point, we have tried to establish a set of rules that will avoid our
using any discretion. I am not sure I would characterize the issue as involving a moral
hazard problem. I want a set of rules that will enable this procedure to operate every day
without our having to make any discrete decisions affecting prices or changing our
behavior. If we were to change any of the program’s major features, the big change
would be to move from overnight lending to, say, one-week lending. I would imagine
that in a situation like the Drexel crisis, we would have to come back to the Committee
to propose such a change. I would add that we probably have had a rather bizarre, if
marginal, impact on current prices by giving away the scarcity value of the securities to
whatever dealer happens to call the Desk first in the morning. So, while the lending
program clearly has some impact on the price-setting process, I think the open auction
feature of the new program will result in a more transparent and less egregious impact
on prices.
MR. BROADDUS. I think the proposals in the memorandum are an
improvement over present procedures. There’s no question about that. I am really
asking the basic question of whether we need to engage in this activity at all. According
to market analysts, the market is rather liquid and deep. Do we really need to do this?
MR. FISHER. The short end of the market currently is not very deep, though I
know we all come with habits of mind about “relative to what.” In fact, the bill market
is as tight as a drum, and the Treasury supports the proposed change. As I have noted a
couple of times, we have a bit of “after you, Alphonse” on whose proposal this is. They
do not want to be telling us to do it or not to do it. Their interest in this actually is
focused on the short end because the bill sector is so tight that if we decide to lend a
little more in that area, they think that would be helpful. That conclusion is based only
on my conversations with Treasury officials. As I pointed out in the memo, if we had
wanted to get out of this business in the late 1980s and early 1990s on the grounds that
the market was very deep and liquid and we really didn’t need to lend securities, that
could have been a compelling argument. But we have seen a decline of 25 percent in the
number of primary dealers in the last two years. Issue size is contracting. As I will
discuss later in the day, the spread between the on-the-run and the twice off-the-run 30year bond is still at 20 basis points. This moment in history, when the markets are going

2/2-3/99

through this supply-shock transition, would seem to be a rather awkward time to say
these markets are deep and liquid and can take care of themselves. That would be my
counter to the argument you are making.
MR. KOHN. I think one could also argue that an entity that holds a huge
portfolio of Treasury securities and is not allowing them to be lent on the market in itself
creates potential distortions. We are a very large part of the market. Certainly, the
Federal Reserve has been concerned about price distortions in the Treasury securities
markets, dating most recently at least from the Salomon Brothers case. There is the
potential for shortages in that market to distort and reduce liquidity over time. This
proposal is in my view an attempt through the auction technique to keep the Fed out of
the price-setting process except when the shortages are quite acute and fairly large
overall. I view it as a public policy objective to keep those shortages low and to
promote liquidity in the market in which we operate.
CHAIRMAN GREENSPAN. President Poole.
MR. POOLE. I don’t have any objection to the proposed program, but I might
understand it better if the following were explained: What difference would we see in
the marketplace if we did no lending at all versus lending as outlined under this
proposal? Tell me what I would look for with and without the lending.
MR. HOENIG. May I add to that question? You mentioned in your memo,
Peter, that at a time of market crisis it had proved helpful in the past to raise the lending
limits above those that were in place. It would help me to understand this issue if you
explained that.
MR. FISHER. At the extreme, one is worried about--I hope I can get all of my
signs right--the problem that arises when the specials price approaches zero. That
occurs when a security is so scarce that to get me to lend the security to you, you have to
give me the cash collateral for free. At the Desk, we lend bonds against other bonds as
collateral because cash collateral would drain reserves from the banking system and we
want to avoid that; however, the market practice is to use cash collateral. Once the
specials rate goes to zero, everyone in the market who has a short position in that
security is encouraged to “fail.” There is then a tendency for the problem to escalate
because those who have an obligation to unwind short positions and deliver the

2/2-3/99

securities to others have an economic incentive to do nothing and create a series of
“fails”--failures to settle. The multiplier effect kicks in and worsens the adverse market
impact. The result is a series of fails that generally make life miserable and divert
attention from price discovery to back offices and lead to settlement shortages and
unsecured positions. And this situation creates other perverse incentives. I imagine you
get the picture. So, at the margin we are trying to reduce the frequency with which that
happens.
When I first brought up this issue at the May meeting last year, I think it was
President Hoenig who asked me whether we could do all this discreetly. Could we just
lend securities when it looked as if we were on the cusp of a problem? But as we
thought long and hard about it, that seemed to pose the moral hazard issue that we are
now discussing. What is the right specials price? That is, we would be deciding when
to interject ourselves, and we would end up intervening in the Treasury market in a
manner similar to our intervening in the foreign exchange markets, trying to pick a level
and decide when to intervene. We don’t want to go that route. That would create big
problems for us. So, if we want to have some impact on the margin, we think the better
way is to use this lending device and do what we can with our large supply of securities.
Now, in the spirit of this, let me be clear. This relates to what Don said. With
the hats that the Committee and the Treasury have given me, I am engaged in market
surveillance to prevent the next Salomon Brothers episode from happening. I go around
giving talks to dealers, telling them that if they have a long position, they ought to be
lending it back into the market. Well, we have a long position and a lot of securities. If
we are concerned about the smooth functioning of the markets and if we think good
behavior of portfolio managers involves being on both sides of the market to help keep it
deep and liquid, then I think we have a duty in that regard, too. We really live in the
financing market. That is where our operations are. And I think we owe something to
the market to help keep it as liquid as possible. So, that is the positive reason. I have
also described what we will be trying to avoid, which is the risk of fails and specials
rates approaching zero. That happens from time to time, and I don’t want to say that we
are going to prevent that from happening. But at the margin should we be leaning
against that with our large portfolio? I’d say “yes.”

2/2-3/99

Let me go back to the bills we now hold in our portfolio. Having sat still and not
bought a bill in a bill pass for a couple of years, we have gone from owning 23 percent
of bills outstanding to 30 percent. That illustrates how much Treasury issuance has
contracted. So, in the bill sector in particular, we may really have something to offer in
terms of trying to keep the market liquid. And that market does indeed have a very close
connection to the money market in which we operate.
MR. BROADDUS. Peter, I certainly am not trying to give you a hard time. I
understand the kind of pressure you are under, dealing with these people day in and day
out, and I know you feel you have certain obligations. The difference is that we are the
central bank, and there are sensitivities that I hope we will be very aware of. The memo
tries to deal with some of these problems; it really does. But it makes me nervous that
we have this kind of relationship with this market. I think it is risky over the longer run.
MR. PARRY. But, Al, you do think this proposal is preferable to what we do
now?
MR. BROADDUS. Yes, absolutely. I was raising a broader question.
CHAIRMAN GREENSPAN. Further questions for Peter? Would somebody
like to move the Treasury securities lending authority he is requesting?
VICE CHAIRMAN MCDONOUGH. I move the authority that is requested,
Mr. Chairman.
CHAIRMAN GREENSPAN. Without objection, so ordered. The next item on
the agenda is the review of the Foreign Currency Authorization, the Foreign Currency
Directive, and the Procedural Instructions with respect to Foreign Currency Operations.
Are there questions for Peter on the memorandum, which I assume all of you have read?
If not, would somebody like to move approval of the changes he has proposed in the
memorandum?
VICE CHAIRMAN MCDONOUGH. Move approval.
CHAIRMAN GREENSPAN. Without objection. Finally, before we get to the
main substance of the meeting, you have received a memorandum from Don Kohn on
proposed changes to the Program for Security of FOMC Information. Are there any
questions for Don?
MR. KOHN. I would like to clarify a few points, Mr. Chairman, if I may.

2/2-3/99

CHAIRMAN GREENSPAN. Why don’t you go ahead.
MR. KOHN. I was trying to respond to a series of issues that have arisen over
the last couple of years in the course of the investigations of leaks of confidential FOMC
information and in conversations with Committee members and staff. In this proposal
we are trying to accomplish a couple of objectives. One is to make clear that
confidential information extends beyond the physical or even electronic documents that
people possess into what one learns at FOMC meetings--the informational content of
meetings. Our intent is to define better what is confidential.
Secondly, we want to facilitate policy discussions among staff at the Reserve
Banks and here at the Board and between Committee members and the staff. So we
recommend reclassifying the Bluebook to Class II to give it a wider circulation and
clarifying that Committee members and Board and Bank senior staff can have
conversations about research issues that arise at meetings with staff who do not have
Class I clearances. Conversations like that can occur. Those conversations would not
be about what happened at a meeting, or who said what, and would not be about the
results or the vote but about research issues that need to be pursued between meetings.
Thirdly, we want to protect the confidentiality of the decisionmaking process and
the policy decision itself by continuing to classify the directive, the transcript, and the
draft minutes as Class I documents. We would continue to limit the number of people at
each Reserve Bank who can see those documents to about 4 or 5--basically the people
who are authorized to attend meetings. There is some confusion about this in the
document itself, and we will try to clear that up. The new program also avoids a
confusion that apparently existed about the distinction between what is currently Class I
and who is authorized to see the directive. So, we are shrinking the number of Class I
documents by taking the Bluebook out of that category. We are making Class I more
clearly aimed at what goes on at FOMC meetings, the most confidential information that
people have.
Finally, after reading through this and getting comments from several people
around the room, I recognize that some points still need to be clarified. We weren’t as
clear as we could have been in every case. I would appreciate having comments either
at this meeting or after the meeting from Committee members and the staff around the

2/2-3/99

room. I would suggest that the best way to proceed would be for us to incorporate your
comments, clean up the draft, and get back to the Committee in March with a cleaner
draft that says what we want it to say. If you don’t vote on this today, the current
program will continue in effect for another seven weeks until the next meeting. So, it’s
not as if we wouldn’t have something in effect. That would be my recommendation, Mr.
Chairman.
CHAIRMAN GREENSPAN. It seems reasonable. Does anybody have any
objection to that? Let’s do that then.
MR. KOHN. If some people have comments, I’d be happy to hear them.
MR. BROADDUS. I have just one quick comment. I worry a little about the
degree of restrictiveness being proposed for what is now going to be Class I, which
involves any discussion of what actually happens at a meeting. If we limit Class I
information to four people, as I understand the proposal, at my Bank that would mean
me along with Marvin Goodfriend and only two other people. The way we operate now,
I involve five or six people in discussions that get into this kind of detail, and this
change would be limiting for me. There is also the matter of involving new people,
giving them incentives to get interested in policy and become more knowledgeable
about what goes on at these meetings, because at some point they will have to take over.
I would hope that in reviewing this proposal, you will think about that aspect. That is
the one point that I really choke on.
CHAIRMAN GREENSPAN. Is there a way to have a general set of rules and
then, under certain conditions, to have temporary alterations or rules that endeavor to
address a specific situation? That strikes me as the type of issue you are raising,
President Broaddus.
MR. KOHN. There is already in these rules, Mr. Chairman, the authority to
make exceptions for specific situations. I think President Broaddus is raising a broader
issue about how many people normally have access to these documents.
CHAIRMAN GREENSPAN. What I am trying to say is that I can conceive of
the number of people given access as being flexible, depending on the particular
situation and circumstances. We might be wise to be sensitive to that because the
purpose is not to withhold information from those who should have it within the System

2/2-3/99

but to classify it. The obvious way in which we could restrict information is to make
certain that nobody knows about it by passing out amnesia pills at the end of each
meeting! There has to be some balance here. It is a tough balance to determine.
MS. MINEHAN. I talked with Don about this a couple of times. Based on your
comments, Don, I assume you plan to do away with the current arrangement in which
the same group of people who brief a Reserve Bank President also know the outcome of
a meeting. You are proposing wider access to the Bluebook which in the future will no
longer include the directive; a smaller group would be allowed to read the directive; and
an even smaller group could discuss what went on at the most recent meeting.
MR. KOHN. I think that second distinction would go away. Its inclusion was an
error and is one of the things I would clean up. The directive would be Class I, so the
distinction between Class I and directive access would go away. The second
investigation by the Inspector General pointed to the limited access to the directive as
one of the rules that a lot of people were confused about. I think that if we can correctly
set the number of people who should be given access to information about policy
discussions at meetings, that distinction would not be necessary and should be
eliminated.
MS. MINEHAN. Let me then put in a request that we take one step back and ask
ourselves how much sense this makes. How many people in most Banks get access to
all the policy alternatives and are part of the very intimate group of people who are
advising the President on what he or she should do at the meeting? Can we ask
ourselves whether it makes sense not to let all those who advise the President know what
went on at a meeting? I know the rule limiting access to the directive existed before and
a lot of us didn’t know about it. I am questioning that rule.
MR. KOHN. Obviously, it is up to the Committee whether it wants to give more
people access to information about what goes on at meetings. Maybe that’s a choice we
could give the Committee at the next meeting: How many people do you want to have
access to Class I information? You know the risks on both sides. We were trying to
make a distinction between knowing what went on at a meeting in the sense of what
issues arose and what staff work needed to be done versus knowing what went on in
terms of information that could cause a problem if leaked to the newspapers.

2/2-3/99

MS. MINEHAN. I know. Obviously, there have to be restrictions. But to treat
as outsiders, in terms of what went on at a meeting, people one considers part of one’s
close circle or people who are involved in discussions of the issues one brings to the
table really does not sit very well with me.
CHAIRMAN GREENSPAN. Communicate to Don whatever issues you would
like to raise. We will try to close this at the next meeting, but it may spill over to the
meeting after that. The ultimate requirement is to get general agreement on all of this.
We will now go to the substance of the meeting and our regular meeting format.
Would somebody like to move approval of the minutes of the December 22 meeting?
VICE CHAIRMAN MCDONOUGH. Move approval.
CHAIRMAN GREENSPAN. Without objection. Peter Fisher, you have the
floor.
MR. FISHER. Thank you, Mr. Chairman. I will be referring to the package of
colored charts that begins with the chart on 3-month deposit rates. You should have that
package somewhere in front of you on the table. 1/
The charts depict forward rate agreements as they have traded over
the past seven months since July 1st. As you can see in the top panel for
the United States, the 9-month forward 3-month rate has been rising rather
consistently since November. That trend continued in January, with some
fits and starts, and that rate is now above the current 3-month rate, while
the 3-month forward rate is trading just below current rates. The point of
interest here, I think, is that the positive spread in the 9-month forward
over the 3-month forward is now the widest since June 1997. For the first
time in the period shown in this panel, expectations are upward sloping,
with the 9-month forward trading above rather than below the 3-month
forward.
Looking at the middle panel--for Germany or the euro zone--I think
the continued fall in both the current and forward rates reflects the
expectation that the European Central Bank will respond to signs of
weakness in the continental economies with further easing, even if that
occurs a little later this year. The weekend conversion to the euro went
quite smoothly, as did our conversion of 32.4 billion of System and
Treasury holdings of deutschemarks into 16.4 billion of euros. There were
some problems related to volume in the European payments system and
the target system that links the national payment systems. This caused
1

/ A copy of the material used by Mr. Fisher is appended to the transcript. (Appendix 1)

2/2-3/99

some frustration with lags in moving funds around the euro zone, but
those seem to be working themselves out now after a month of transition.
Looking at the bottom panel for Japan, the 9-month forward Japanese
3-month rate backed up a fair amount in late December and into January
and has been trading as much as 20 basis points over current 3-month
rates. There is much going on in Japanese money markets that I don’t
pretend to understand, but the simple point I have come to understand is
that since October Japanese government T-bill rates have backed up
approximately 30 basis points. Having come off their zero and below zero
levels, they have backed up 30 basis points in a period when the Ministry
of Finance has increased bill issuance by about 20 percent. This increase
in bill rates seems to have flowed through to the money market.
On January 11, with dollar/yen having broken through 109, the
Japanese authorities intervened by purchasing
This seemed to
provide some stability to the exchange rate at around 111 or 112. Last
week the yen traded up toward 115 on the increasing perception of the
U.S. economy’s robust performance. It is back down to around 111 to 112
this week. There is still a fair bit of noise in the dollar/yen market but,
again, I wouldn’t pretend to understand why.
Instead of focusing the rest of my remarks on the Japanese economy
or the U.S. equity markets or Brazil, which have received a fair amount of
attention in our own written reports and elsewhere, I thought I would
focus on two areas that have been getting less attention. Nevertheless,
these areas have generated a great deal of uncertainty as well. The first is
how to price the dollar/euro and the second is how to price fixed income
markets in general.
On the second page you will see a 10-year history of a synthetic euro,
based on a methodology developed by Hong Kong Shanghai Bank. There
are a number of methodologies out there, but over the 10-year history
there is not much difference among them. This chart is for broad-brush
purposes. A basic fact to keep in mind is that the euro is quoted in dollars
per euro. The dollar was weakest--at the top of the chart--in 1992 when it
traded at almost $1.50 per euro; it was strongest--at the bottom of the chart
--in 1989 and 1997 when it traded through $1.05. To give some
perspective, we have indicated weeks in which the U.S. authorities
intervened in dollar/mark, but we did not include dollar/yen interventions.
They are not marked here, but clearly there were occasions when we
intervened in dollar/yen, and that intervention had a rather strong impact
on dollar/mark as well. This just lists the dollar/mark interventions to
provide reference points.

2/2-3/99

My point in bringing this to your attention is, first, that there seems
to be relatively little awareness in the markets about how strong the dollar
has been recently; it traded around 10-year highs during most of 1997 and
1998. Secondly, it is not at all clear what people mean when they say that
they expect the euro to be a strong currency. Do they mean that they
expect it to trade above its launch rate of about 1.18? Do they mean that
they expect it to trade up into the 1.30s during a normal cycle? Or do they
mean that they expect its average rate over the next 10 years to be higher
than its average rate over the past 10 years? The fact of the matter is that
people speak rather loosely about this. That reflects, in my view, a great
lack of conviction on how to price this currency. Even though people
think they know about dollar/mark, dollar/euro is really a different animal.
To give you some flavor of that lack of conviction and the jitteriness in the
markets: In the first week of the euro’s existence, it lost a cent because the
target system closed one hour late.
Mr. Chairman, at the last meeting you asked me if I could explain the
differential between U.S. and U.K. swap spreads and those on the
continent, and I feebly mumbled something about central bank trading.
On some reflection, I would not want to hang my hat on that answer. I
know that staff at the Board and New York have tried to come up with
some explanation, but we are still uncomfortable with our lack of
understanding of that phenomenon. One reason I am uncomfortable is that
the longer I look at the relationships, the more I realize how much I don’t
understand about the underlying relationships among government bond
yields, which are shown at the top of page 3.
The top panel shows the yields since last July 1 on U.S., U.K., Italian,
French, and German 10-year government bonds as well as the yield on the
U.S. Treasury inflation-indexed 10-year bond, which is depicted by the
dotted red line. You can see that there was quite a rally in European
government bond markets in November and December. That brought us
to the moment on January 5, which was the day before an auction of new
U.S. inflation-indexed 10-year securities, when the yield on both the
outstanding Treasury indexed security and the outstanding Italian
government nominal 10-year bond stood at an identical 3.94 percent.
While the U.S. inflation-indexed yields have fallen about 15 basis points
since then and the Italian nominal yields have traded sideways, both
German and French government 10-year bonds have traded through the
U.S. inflation-protected 10-year rate.
We know that the pricing of our inflation-indexed security is
problematic, but the problems are likely measured in tens of basis points.
I don’t pretend that I have a magic rule here, but we know we are dealing
with a premium for illiquidity and in particular for the tax treatment that
the dealers complain about quite a bit. But there must be something

2/2-3/99

happening on the European side of this equation as well. The optimists in
Europe would like to explain the low government bond yields as a
reflection of the rush of capital into the euro and the confidence that global
investors have in the ECB. Others, perhaps more plausibly at least so far,
suggest that the declining yields reflect the deteriorating outlook for the
continental economies. But if the euro’s own economies are as weak as
these yields suggest, where are the tax receipts going to come from to help
these governments avoid issuing a lot more bonds?
To make the general point, let me offer what is clearly my own
opinion: That fixed income markets have been slow to recognize that the
so-called short-run supply/demand dynamics have recently been having a
much bigger impact on yield curve movements than have changes in
inflation expectations, at least over the most recent period. Consider the
U.S. Treasury market over the last 2 years with decreases in new issuance
and the flight-to-quality buying. Consider the Japanese government bond
market over just the last 2 months.
There is also considerable unease in the U.S. government bond
market. In the bottom panel, you can see that the Brazilian shock had
relatively little impact on spreads that were greatly affected by the events
of last August, September, and October. The traders in U.S. Treasuries
are still not at all comfortable. Liquidity remains at a very high premium.
The bottom red line shows that the twice off-the-run spread is still around
20 basis points, as contrasted with the 2 to 5 basis point spread that had
prevailed for most of the last 5 years until the events of the past few
months.
Again, let me make clear that I am offering my own opinion. While
the events of last fall effectively knocked fixed income markets off their
then-existing equilibrium, markets have not yet found a new equilibrium
that they are comfortable with. I don’t mean to suggest that markets are
fragile, but I do mean that they are still groping and do not have a great
deal of conviction about their pricing of fixed income instruments. It
would be fair for any member of the Committee to point out that I may be
having problems getting supply and demand right in my own backyard of
the fed funds market!
Turning to the next page, let us go over our operations in the domestic
markets for the last several reserve maintenance periods. In the two
periods depicted in the top two panels, we set out to be quite generous--to
lean against the tightness that we had been seeing during the late fall and
in early December and against the normal year-end funding pressures. We
then found that our generosity in supplying reserves was heavily
supplemented by unanticipated weather-induced float from the winter
storms across the Midwest and the Northeast. In the first maintenance

2/2-3/99

period, our generosity was aimed at avoiding a firm end-of-period
settlement date, which we feared might set us up for a tight year-end that
coincided with the first day of the settlement period. We debated at some
length among ourselves how best to prepare the market for the December
1999 year-end Y2K event. We debated whether we should be
intentionally generous to calm anxieties or intentionally stingy to engineer
a tight close this year in order to encourage everyone to think they had
better lock in their funding in advance for the next year-end. We chose
the generous route and were rewarded at least in one sense with a
noticeable decline in the 1999 year-end premium, as evidenced by futures
prices over the last couple days of the year.
In the second period, bad weather contributed to especially high,
unpredictable float and we drained reserves aggressively on several
occasions. In the third maintenance period, I have only myself to blame
for the softness. I thought we had room to do a term operation of modest
size and thereby avoid the need to operate each day with overnight RPs of
different sizes and still leave room for any additional float or change in
other factors. I was either wrong or unlucky or both, and we ended up
with a little more float and a soft rate through the middle of the period.
Overall, I am glad that we now have the long-term RP in our tool kit,
and I thank the Committee for adding it. We are going to continue to
reflect on our experience of this past year-end as we begin to plan for the
coming year-end.
Mr. Chairman, I will need a vote to ratify our domestic operations,
and I would be happy to answer any questions.
CHAIRMAN GREENSPAN. Are there questions?
MR. JORDAN. Peter, I probably should know this from something else you
have sent us, but when you converted our holdings of securities denominated in
deutschemarks to securities denominated in euros, what did you do?
MR. FISHER. We did a series of foreign exchange trades with ourselves, which
is really what everyone did to convert. These are not done with any counterparty. We
just take out all the assets denominated in marks and then put them back in denominated
in euros, asset by asset. Then the challenge is to confirm that with all our counterparties
and custodians down the supply chain, if you will, to make sure we get to the same
numbers. But essentially the conversion involved a rather rudimentary process of doing

2/2-3/99

a foreign exchange trade with oneself; that was the case for us and for all the market
participants.
CHAIRMAN GREENSPAN. You leave us hanging. Did you check to see if it
worked?
MR. FISHER. Yes. I meant to cover that. I did say that our conversion went
quite well; maybe I covered that too quickly. We were among those who experienced
some of the glitches in the new European payment system. We were holding our breath
sometimes to make sure we did not fail on our delivery of euros while waiting for the
queue to work through, but we avoided any problems with that. The real issue came in
the first week when the German payment system did not seem to be up to the volume,
which is curious because the volume was much reduced. The Germans had put through
some changes in their own domestic payment system simultaneously with the shifted
target. That taxed the capacity of the German payment system and there were some
bottlenecks.
MR. JORDAN. The portfolio now has euro-denominated deposits at the
Bundesbank, euro-denominated German government securities, and euro-denominated
deposits at the BIS?
MR. FISHER. Yes, and euro-denominated repo agreements with our
counterparties, with Deutsche Bank as our custodian.
MR. JORDAN. Okay. Under what circumstances would we wind up having in
the portfolio euro-denominated obligations of other taxing authorities besides the
Germans?
MR. FISHER. We will not, until this Committee discusses whether I should buy
some other assets. That is, we are limiting our repos to German government
instruments. I think I mentioned some months ago that we will see how the markets
develop, and we may come back to the Committee to discuss diversifying. In particular,
the French short end is much more liquid, so the issue is whether we would want to
migrate there if we want our holdings to be more liquid. The Desk will not make that
decision--the Committee will make it--though we may recommend going one way or the
other. I already have a list of people from non-German EU government finance
ministries who want to come visit with me. I am very much looking forward to being

2/2-3/99

able to say: “I’m happy to consider your proposal but I have to talk to the FOMC before
we make any decisions.”
MR. JORDAN. Thank you.
CHAIRMAN GREENSPAN. Incidentally, I noticed that the issuance of euro
bonds in the international market in January exceeded the issuance of dollardenominated bonds and exceeded the combination of those denominated in the previous
currencies that went into the euro. Is this a short-term, one-time adjustment? Or are we
looking at the longer-run impact of presumably narrowing bid-ask spreads on the euro
vis-à-vis the predecessor currencies? In short, does the introduction of the euro
represent a major development in the international bond markets or one that is going to
fade once the one-time impact dissipates?
MR. FISHER. That is a well phrased question. My view is that a lot of issuers
want to test the market quickly. That is, we are talking about the issuance of new bonds.
Everyone wants to get in because in their sector of the euro market they can be the
benchmark, just as all European governments are debating who is going to be the
benchmark. If Philips or whatever corporate entity can get in and create a name for
itself, will it have a leg up on others? Some corporations may find it works very well for
them, and they may expand their program. I have a sense that it’s a matter of “Let’s try
it and see how it goes.” If I followed your question, I’m not quite sure I see a smooth
transition from that to how the euro/dollar exchange rate will trade and what its bid-ask
spread will be.
CHAIRMAN GREENSPAN. No, I wasn’t referring to the exchange rate. I was
referring to the notion that as the aggregate issuance of euro securities increases, one
would expect the bid-ask spread in the single euro currency to be narrower than the
average bid-ask spread in its predecessor currencies. Even though differences still exist
with respect to taxes and even though there are all sorts of other concerns, the liquidity
differences presumably will narrow the average euro bid-ask spread, and that should
increase both the demand for and the supply of new issues in the euro market. The
question I am putting to you relates to my recollection that the predecessor currencies
accounted for 30 to 35 percent of the issuance market but in January issuance in the euro
market was 40 percent of the total. Now, that can be either an indication of the

2/2-3/99

broadening of the market or the mere novelty of it. Perhaps people who had been
thinking of doing an offering in deutschemarks decided to wait another few weeks to do
it in euros. That’s the nature of my question.
MR. FISHER. Right. I think the January effect we are observing is a curiosity
effect. Your other hypothesis may well prove right over time, but I don’t expect a
smooth movement from where we are now, given what happened in January.
CHAIRMAN GREENSPAN. Is the increase in the offerings, which effectively
means going short in euros, a factor in the euro’s decline from its initial offering price?
MR. FISHER. That is something some of us have speculated on. I don’t think
we have firm evidence that there will be a lot of borrowing by people who don’t live in
the euro zone, since they have to take their proceeds out. So, the first-run effect of such
borrowing would be to weaken the euro. Some of us have been telling ECB and NCB
officials for some years that having a deep and liquid capital market is not a one-way-up
ticket for the exchange rate, and I think there may be some of that. I still ascribe the
thinness of the market and most of the tentativeness in exchange rate trading to that.
CHAIRMAN GREENSPAN. Further questions?
VICE CHAIRMAN MCDONOUGH. Mr. Chairman, I move approval of the
domestic operations.
CHAIRMAN GREENSPAN. Thank you. Approved without objection. Let’s
move on to the Chart Show and Messrs. Prell, Alexander, and Stockton. Gentlemen.
MR. PRELL.

Thank you, Mr. Chairman. I will start the presentation. We will

be referring to the package of charts entitled “Staff Presentation on the Economic
Outlook,” which may be at the bottom of the pile of papers in front of you. 2/
Chart 1 provides a basic summary of the staff economic forecast. As
you know, the BEA published its advance estimate of fourth-quarter GDP
on Friday; it was a half point above our 5 percent guess. However, given
that the BEA numbers do not really point us in a different direction--and
they are only tentative, in any event--we have stuck with the Greenbook
figures for our presentation today.
The top panels of the chart show our projection for real GDP, on a
four-quarter percent change basis. We are looking for a substantial
deceleration, from 4 percent in 1998--4.1 percent, according to Friday’s
2

/ A copy of the material used by the staff is appended to the transcript. (Appendix 2)

2/2-3/99

report--to around 2½ percent this year and next. As the bars in the left
panel indicate, the slowing is entirely accounted for by domestic spending:
The negative contribution from net exports is expected to diminish
considerably over the forecast period.
Though the growth of output is much slower, it is pretty close to our
estimate of the trend rate for potential, so that the unemployment rate--the
red line in the middle panel--changes little from where it was last quarter.
The chart also shows the movements in the factory utilization rate. Over
the past year, we observed the odd pattern of the plant utilization rate
dropping somewhat below its long-term average even as the jobless rate
was reaching a 28-year low. We are anticipating that these indicators of
supply pressures will continue to send conflicting messages, each of them
changing little on net over the next two years.
As Dave will be discussing later, we have given greater weight to the
labor market tautness in arriving at the inflation forecast shown in the
bottom panel. We foresee a noticeable pickup in the pace of price increase
over the next two years--about a percentage point as measured either by
the CPI or by the GDP chain index.
Chart 2 presents some of the financial backdrop for this projection.
The basic assumption is that the federal funds rate will be unchanged
through next year. As you can see in the top panel, given our forecast of
rising inflation, this implies a further decline in the real funds rate-
proxied by the nominal rate minus the recent inflation rate. In fact, it is
not at all clear that inflation expectations have come down as far as actual
inflation in the past year, so this measure may overstate the current level
of the real rate--and consequently may exaggerate the decline that lies
ahead. But I would characterize our forecast as incorporating a real shortterm interest rate that is at the low end of the range since 1995.
The real short rate is scarcely an unambiguous or comprehensive gauge
of the financial impetus to demand, so it is worthwhile to look at a few
other indicators of credit market conditions. The picture is mixed. In the
middle left panel, you can see that nominal yields on Treasury and
investment grade corporate bonds have fallen over the past couple of
years--probably by more than longer-term inflation expectations. Junk
bond yields also eased a bit until the market was jolted by the Russian
default shock last summer. The junk rate index plotted here is now up
roughly 1½ percentage points from a year ago, suggesting that the typical
low grade firm is facing a considerably increased long-term borrowing
cost.
The senior loan officer survey--the basis of the right panel--indicates
that businesses also are facing a little less friendly reception at commercial

2/2-3/99

banks, with loan underwriting standards having been tightened some in
recent months, especially in the case of larger firms.
Credit has not dried up, however, as may be seen in the bottom panel,
which plots the real debt growth of households and businesses combined.
Through the fourth quarter, the end of the solid portion of the line, debt
growth remained rapid. We are forecasting a moderation, but this is
scarcely a credit-crunch scenario--as might have been feared for a while
last fall. Lenders have turned a bit more cautious, but most of the
explanation for the projected slackening in debt expansion comes from the
demand side.
If there are corresponding signs of increased caution in equity markets,
they are rather difficult to find. Share prices have continued to soar, and
where they are headed from here clearly is a crucial question for the
economic outlook. The top panels of Chart 3 don’t answer that question,
but they provide some provocative perspective by comparing the
performance of the U.S. stock market in the 1990s with a couple of other
impressive bull markets of yesteryear--Japan in the 1980s and Wall Street
in the 1920s. Given the amount of artistry involved in choosing the base
periods and thus aligning these curves, I would recommend taking the
pictures with at least a grain of salt. But, if you have a taste for salt,
perhaps they are suggestive of some of the risks confronting us.
The middle panels come at the issue from an angle that is perhaps a bit
more analytical. At the left, I have repeated an exhibit I’ve used in prior
presentations to demonstrate why share prices had to peak soon. Perhaps
you should stop me before I do it again! [Laughter] The latest
observations are rough approximations, given yesterday’s prices and our
estimates of earnings through year-end. The S&P 500’s aggregate P/E is
at an unprecedented level. And, to underscore the abnormality, it has
achieved this multiple not when earnings had a cyclical upturn in front of
them but rather, as you can see in the bottom left panel, when the profit
share of GNP has risen to a multi-decade high. And, if the S&P multiple is
remarkable, that for the NASDAQ composite--at the right--is astonishing,
moving rapidly toward triple digits!
Although some analysts would say--indeed, have said--that earnings
are beside the point in valuing stocks in this New Era, we have not been
able to break the habit of looking at them--which may help to explain our
forecast errors. I have shown, in the bottom right panel, annual percent
changes in NIPA corporate profits--the black line--along with those for
S&P 500 earnings per share. The accounting concepts and coverage of the
two series differ, but they have tended to move together. In that light, the
divergence between our forecast of NIPA profits in 1999 and the S&P
predictions of Wall Street strategists is worth noting. It suggests that the

2/2-3/99

market will continue to have to cope with earnings disappointments in the
months ahead. If we had an adequate sample of the Wall Street
expectations for 2000, I suspect that the story would be reinforced. Such
disappointments have not stopped the market to date, but we have
assumed that--from such high valuations--it will be difficult to extend the
uptrend in prices.
If we prove to be anywhere close to right about the trend from here, the
market will cease to provide the impetus it has to spending over the past
couple of years. Chart 4 summarizes some of the key considerations in
our forecast for domestic demand. Because these are familiar stories,
I shall just run through the list very quickly.
The upper left panel highlights the fact that though we have the wealthincome ratio declining, the lags in the effects on spending lead us to
expect that consumer outlays will continue to grow faster than income for
a while longer, pushing the saving rate still lower. But spending growth
does moderate.
In the housing market--depicted at the right--we expect that with
mortgage rates remaining near current levels, the monthly payment burden
of home-ownership will remain attractively low. This should keep
housing demand from falling sharply as income and wealth growth
weaken.
In the business sector, the slowing of output growth, depicted in the
middle left panel, will turn the accelerator effects on capital spending
negative. But that should be partly offset by the favorable effect of a
continuing decline in the relative price of equipment--charted at the right.
Inventory investment, the red line in the bottom left panel, is expected
to move pretty closely with final sales. Not much of a story there.
And, finally, government purchases are projected to run a little stronger
on balance over the next two years. In the state and local sector, surpluses
are growing to the point where we would expect to see some greater
expenditure on infrastructure. In the federal sector, real purchases are
expected to come closer to bottoming out, with some obvious upside risks
in defense and other outlay categories. The pickup in government
spending, however, is only a modest offset to the slackening in the growth
of private demand that we foresee.
Lew will now shift the focus from the domestic side to the external
sector.
MR. ALEXANDER. The main changes in our forecast on the
international side have been driven by events in Brazil. Our previous

2/2-3/99

assumption that Brazil’s exchange rate peg would hold has obviously been
proved wrong. Your next chart focuses on recent developments in Brazil
and highlights the key policy problems facing the Brazilian authorities at
this time. As can be seen in the upper left panel of Chart 5, the Brazilian
real depreciated by more than 40 percent in the last three weeks through
last Friday. The real has appreciated about 8½ percent following the
announcement that Arminio Fraga, a former intern in the Division of
International Finance, [laughter] would replace Francisco Lopez as the
head of the Brazilian Central Bank.
Early in January, a dispute over intergovernmental debts between the
state of Minas Gerais and the Brazilian Federal Government, though
relatively insignificant by itself, highlighted the lack of consensus on the
need for fiscal consolidation within Brazil’s complicated federal political
system. As the dispute threatened to widen to other states, pressures on
the real mounted. On January 13, the Brazilian authorities attempted a
controlled devaluation of 8 percent. Significant capital outflows continued
and the real was allowed to float two days later.
The anticipated fallout from the collapse of the real--including
continued high price-adjusted interest rates, heightened economic
uncertainty, a greater burden of foreign currency denominated debt, and
pressures on the financial system--is projected to cause a severe downturn
in economic activity in Brazil this year. As shown at the upper right, we
now expect the downturn in Brazilian GDP to steepen in the first half of
this year and to extend, although at a diminishing rate, for the remainder
of the forecast period.
Brazilian policymakers now face a dilemma. With the change in the
exchange rate regime and the sharp devaluation of the real, it may be
necessary to keep interest rates high to contain inflation. But Brazil’s
fiscal position remains precarious, owing in part to high debt service costs.
The red bars in the middle left panel indicate that Brazil’s public
sector borrowing requirement was over 8 percent of GDP last year, even
with a modest surplus on the non-interest, or primary, part of the budget.
Fiscal policy actions enacted since last fall are expected to increase
Brazil’s fiscal balance by about 3 percent of GDP this year. But a
substantial overall deficit will remain unless interest costs decline.
Over the last year, domestic short-term interest rates in Brazil--the
black line in the middle right panel--averaged about 28 percent, while
inflation--the red line--was only about 3 percent. Roughly two-thirds of
the government’s domestic debt carries an interest rate that is linked to
overnight interest rates. At the end of last year, Brazil’s federal debt was
already equal to about 50 percent of GDP, up from just 35 percent in 1996.

2/2-3/99

Persistent high interest rates could put Brazil’s government debt on an
explosive path. On the other hand, high interest rates may be necessary to
contain inflation expectations. One of the critical uncertainties facing
Brazilian policymakers at this time is the degree to which the recent
devaluation will pass through to inflation.
The bottom two panels offer some perspective on this problem. The
bottom left panel shows the exchange rate, domestic interest rates, and
inflation for Mexico over the period from 1994 to 1996. In this case the
sharp devaluation of the peso at the end of 1994 was followed by a burst
of inflation that peaked at an annual rate of over 100 percent. The lower
right panel shows the same three variables for Korea over the last two
years. The devaluation of the Korean won at the end of 1997 was roughly
the same magnitude as the devaluation of the Mexican peso, but in the
Korean case inflation only reached a peak of about 30 percent. Before
1997 Korea had less historical experience than Mexico with either large
exchange rate depreciations or high inflation, and this may have helped to
limit the passthrough of the depreciation to inflation expectations and
hence to domestic inflation itself.
Our forecast for Brazilian inflation over the next year falls somewhere
between the Korean and Mexican examples. Empirical analysis of
exchange rate passthrough in Brazil, covering the period before the
adoption of the real plan in 1994, would have suggested a pattern more
like the Mexican example. But over the last four years the Brazilian
government has pursued policies intended to limit inflationary dynamics.
For example, explicit indexation of wages was banned in 1995.
Obviously, there is considerable uncertainty about our inflation forecast
for Brazil.
Your next chart focuses on our outlook for other emerging market
economies. Our current forecast does not anticipate extreme contagion to
other emerging markets from recent events in Brazil. To some degree, this
represents a change in our thinking, in that in our previous “worse case”
analyses we assumed that a failure of Brazil’s program would likely
generate severe financial pressures on other emerging market economies.
This change in our thinking reflects the fact that over the last several
months financial markets have increasingly distinguished Brazil from
other emerging market economies.
The top left panel of Chart 6 shows yield spreads relative to U.S.
Treasuries, for Brazilian, Mexican, and Argentine Brady bonds. In the
immediate aftermath of the Russian devaluation and default in August,
these spreads generally moved together. But since December Brazilian
Brady spreads have moved up more sharply than those for Argentina and
Mexico. Domestic interest rates in Mexico and Argentina, shown in the

2/2-3/99

top right panel, are up some since the middle of January, but these
increases are modest compared with those seen in August and September.
Perhaps not surprisingly, the financial contagion from Brazil to Asian
emerging markets is even more modest. Credit spreads for Korean and
Thai sovereign bonds--the black and red lines in the middle left panel-
ticked up when the real was devalued, but those increases have been
reversed. The events in Brazil raised some speculation about the viability
of currency pegs in Hong Kong and China, but these pressures were
relatively mild. As shown in the middle right panel, the 1-month Hong
Kong interbank interest rate has increased about 150 basis points since
mid-January. Interest rates implied by offshore forward contracts for the
Chinese renminbi moved up somewhat more but this was probably related
to the Chinese government’s decision not to give preferential treatment to
foreign investors in the bankruptcy of GITIC, a failed Guandong
investment company.
Our outlook for GDP growth for emerging market economies, other
than Brazil, is shown in the lower right panel. The repercussions from
Brazil’s economic difficulties are expected to have negative impacts-
lower exports to Brazil and some spillover of financial market pressures-
on other Latin American countries. Consequently, we expect the rest of
Latin America to suffer a recession in 1999, with real GDP in these
countries falling about 1½ percent before rebounding 2¾ percent in 2000.
In contrast, real GDP in all five of the so-called “Crisis Asia” emerging
market economies--all of which experienced serious financial crises and
sharp declines in output in 1998--is expected to bottom out and return to
positive growth by the end of 1999. As you can see in the lower left
panel, industrial production has already returned to positive territory on a
year-over-year basis in Korea, and Thailand is not far behind. We
continue to assume that Hong Kong’s peg to the dollar will hold, but at the
expense of somewhat higher interest rates. Accordingly, real GDP in
Hong Kong is now expected to decline 2 percent in 1999. Real GDP in
China is expected to grow by about 6 percent in 1999, a slower pace than
estimated for 1998.
Your next chart presents the staff’s outlook for Japan and Canada.
We continue to be very pessimistic about the prospects for recovery in
Japan. Since mid-November long-term Japanese government bond yields
--the red line in the top left panel--have moved up more than 1½
percentage points, and the yen--the black line in that panel--has
appreciated about 6 percent relative to the dollar. The sharp increase in
bond yields followed the mid-December announcement of a 1999 budget
and associated financing plans that envision somewhat greater fiscal
expansion and greater bond issuance this year than most market
participants had expected. The change in Japan’s cyclically adjusted fiscal

2/2-3/99

deficit, shown in the upper right panel, is expected to exceed 2 percent of
GDP this year and to be about ¾ percent of GDP next year.
There are few signs of a reversal in the downward trajectory of
underlying private domestic demand, however. In the fourth quarter of
last year, housing starts and new car registrations fell to their lowest levels
in over a decade, and the labor market remains depressed. Since the
beginning of the year Japanese bank stocks--the red line in the middle left
panel--have been buoyed by potential mergers and the perception that the
Japanese authorities may be beginning to move more aggressively to
reform the banking system. Such optimism may be premature. In late
January, Moody’s downgraded the long-term credit rating of five large
Japanese banks, including Bank of Tokyo-Mitsubishi.
We project that Japanese real GDP--the black bars in the middle right
panel--will continue to fall over the forecast horizon even with significant
fiscal stimulus. Japanese consumer prices--the red bars--are expected to
fall 2 percent in 1999, but part of this decline is the reversal of special
factors that artificially elevated consumer prices at the end of last year.
Our outlook for other industrial countries is not so gloomy. Declines
in commodity prices over the last two years put downward pressure on the
Canadian dollar, shown in real effective terms as the black line in the
bottom left panel, particularly in the first three quarters of last year. Given
that we now project a pickup in both oil and non-oil commodity prices this
year, we expect the Canadian dollar to appreciate somewhat against the
U.S. dollar this year and next. Our outlook for Canada is shown in the
lower right panel. Canada should register growth near 2½ percent this
year and in 2000--the black bars--with only a modest uptick in inflation, as
shown in the red bars.
The staff outlook for Europe is presented in Chart 8. Exchange rate
strength has been an important factor in a tightening of monetary
conditions in Europe over the past year and a half. The black line in the
upper left panel shows a constructed series for the real effective exchange
rate for the euro area, using the DM as an historical proxy. Although the
real value of the effective euro has fallen since October, this move
retraced only partially the steep appreciation that occurred following the
outbreak of the Asian crisis in mid-1997. The real value of sterling--the
red line--experienced an even greater appreciation, which started earlier,
before it, too, eased somewhat in recent months.
The latest data suggest that momentum behind European growth has
ebbed lately and the outlook for Europe has softened. This has been
especially apparent in Germany, but has started to show through in other
key European countries as well. As shown in the middle left panel,

2/2-3/99

indicators of business confidence in both Germany and France--the black
and blue lines, respectively--have fallen steadily since the spring of last
year. Despite a sharp uptick in late 1998, U.K. business confidence--the
red line--still is well off its early-1997 level. Euro-area labor market
conditions improved somewhat last year, as illustrated by German and
French unemployment rates in the middle right panel. But the declines in
unemployment rates have been modest and, as activity has slowed
recently, those rates have leveled off.
The upper right panel shows three-month interest rate futures in
Europe. In view of increasing signs of deceleration in Europe, markets
appear to have factored in a modest decline in euro-denominated shortterm interest rates over the near term. Futures rates also suggest that
market participants expect the Bank of England to cut rates further
following substantial declines already implemented since October.
The staff forecast is consistent with this pattern.
The European outlook for real GDP growth is summarized in the
lower left panel. Growth in both the euro area and the United Kingdom
should slow by about ½ percentage point this year before recovering
somewhat in 2000. U.K. inflation, the top line in the lower right panel,
should stay on target at 2½ percent. Although inflation rates will vary
across the euro area, they should stay low on average and well within the
target range of zero to 2 percent specified by the ESCB.
My next chart outlines the influence of the foreign outlook on the
U.S. external accounts. The top left panel depicts trends in the value of
U.S. exports to our principal export markets. U.S. exports to Europe--the
black line--and to Canada, the green line, advanced steadily last year.
Exports to Latin America--the blue line--remained relatively strong
throughout 1998 as well. U.S. shipments to Asia--the red line--fell sharply
last year. The jump in shipments in the fourth quarter to Asian markets
was largely due to aircraft.
U.S. exports have been restrained by weak economic activity abroad
and a strong dollar. Turning to line 1 in the table in the top right panel, we
estimate that economic growth in the foreign industrial countries was
below trend last year; we project that growth will pick up only slightly
during the next two years. Strength in Canada and Western Europe will
only partially offset the effect of the recession in Japan.
As shown in the middle left panel, the dollar had appreciated sharply
in real terms during the past two years, by nearly ten percent against major
currencies--the red line--and by more than 15 percent against those of our
other important trading partners--the black line--before depreciating
during the past few months. In our forecast, the dollar remains essentially

2/2-3/99

flat in real terms, appreciating against the currencies of Latin America and
depreciating with respect to most of the currencies of Asia and Europe.
This assumption reflects a balancing of risks, with the prospect of strong
U.S. growth tending to support the dollar and concerns about the widening
U.S. external deficit tending to depress the dollar.
The middle right panel depicts the arithmetic contribution of U.S.
exports and imports to U.S. GDP growth. Real exports of goods and
services, the black bars, fell during the first half of last year before
rebounding last quarter. We project that exports will decline again during
the first half of 1999 and then stage a slow recovery. We project that
imports--the red bars--will continue to advance at a slower, but still strong,
pace this year and next.
The U.S. external accounts are plotted in the lower left panel. We
estimate that the nominal U.S. trade balance--the red line--in current
dollars in the fourth quarter recorded a deficit of $182 billion, nearly $70
billion larger than in the fourth quarter of 1997. With accumulating U.S.
current account deficits and the resulting deterioration in the U.S. net
international investment position, net investment income--the blue line--is
projected to continue its decline through 2000. We project that the deficit
in the current account--the black line--will increase to $374 billion in 2000
which, as shown in the lower right panel, is equivalent to 4.0 percent of
U.S. GDP, noticeably above the peak reached in the 1986-87 period.
In recent months financial market participants have focused
increasingly on the widening U.S. external deficits and their implications
for the exchange value of the dollar. Your next chart considers the
relationship between the value of the dollar over the medium term and the
likely path of the U.S. net investment position--that is, U.S. foreign assets
less U.S. liabilities owed to foreigners. The top right panel shows the ratio
of the U.S. net investment position to GDP, on an inverted scale, over the
next 20 years under two assumptions about the path of the real exchange
rate. The black line shows the path under the assumption that the dollar
depreciates, in real terms, at a rate of about 1½ percent per year after the
end of the Greenbook forecast period. The projection is based on an
updated version of the long-term model the IF Division used in its Current
Account Sustainability project two years ago. One of the conclusions of
that project was that a steady depreciation of the dollar of 1½ percent per
year was sufficient to stabilize the net investment position. As the black
line in the upper left panel indicates, that conclusion is no longer valid.
The red line on the chart indicates that even a 3 percent rate of
depreciation is not quite enough.
The main factors that change the basic result are shown in the upper
right panel of the chart. First, the dollar has appreciated substantially

2/2-3/99

relative to the baseline that was used in 1997. Second, the current account
deficit over the last two years has been above the 1997 baseline. Finally,
our new higher estimate of U.S. trend growth makes a substantial
difference because the income elasticity of U.S. imports is estimated to be
substantially higher than the income elasticity of the demand for U.S.
exports.
The bottom two panels of the chart show the net investment position
to GDP ratio and real effective exchange rates since 1980 for Canada and
Australia. These are the only two major industrial countries that in recent
decades have had negative net investment positions of the size we are
projecting for the United States. In both cases their exchange rates have
trended down but with considerable variation.
That these projections have changed so much over the last two years
is indicative of the fact that in such long-run projections initial conditions
matter a lot. Moreover, there is no way to know, as the experience of
Australia and Canada suggests, exactly how large a country’s net
investment position could be, nor when the prospect of further
deterioration would prompt a financial market reaction. But these
projections do suggest that the potential problem for the dollar has gotten
worse rather than better over the last two years, and the prospect of
widening U.S. current account deficits is likely to exert downward
pressure on the dollar at some point in the future.
Dave will now continue our presentation.
MR. STOCKTON. Your next few charts focus on the aggregate
supply side of the economy, beginning with a question raised at a number
of recent meetings, “Just how low is inflation?” The upper left panel of
Chart 11 displays the four-quarter changes in two alternative measures of
consumer prices, the PCE chain price index--the black line--and the CPI-
the red line. Prior to 1995, these two indexes increased at roughly the
same pace, on average. Since then, however, PCE inflation has been
running consistently below that of the CPI, with the difference last year
about ¾ percentage point. For the measures excluding food and energy,
shown at the right, the gap was even larger last year--about 1¼ percentage
points.
The middle left panel highlights the major differences between the
CPI and the PCE indexes. One important difference is the aggregation
formulas. The CPI uses fixed weights at both the detailed and more
aggregate levels, while the PCE index uses weights that vary with
spending patterns from period to period. All told, these aggregation
differences account for about three-fourths of the gap between the reported
inflation rates over the past couple of years.

2/2-3/99

A second difference is the scope of the two indexes. The CPI covers
the out-of-pocket expenditures of urban workers, while the PCE measure
covers total consumer spending. For example, PCE includes third-party
payments of medical costs by insurance companies and the government.
The PCE index also includes services provided to individuals by nonprofit
institutions.
A third difference is that, while the PCE index is mostly constructed
using components of the CPI, the BEA occasionally uses price data from
other sources. Most prominently in recent years, they have been using a
measure of medical service prices that has increased considerably less than
the corresponding measure in the CPI.
Finally, the weights used in the indexes are developed from different
source data; the CPI uses spending reported by households in the
Consumer Expenditure Survey, while the PCE weights are derived from
various economic censuses. One notable difference is that the weight of
housing in the CPI is considerably larger than that in the PCE index.
Because housing prices have been rising relatively rapidly, this has given
an added push to the CPI. To get a sense of the importance of the
weighting and price differences, we recalculated the CPI using weights
derived from the PCE data and the PCE’s measure of medical service
prices. As you can see by comparing the red and blue lines in the middle
right panel, this exercise suggests that about another ¼ percentage point of
the difference can be accounted for by these factors.
The lower left panel lists some of the pros and cons of the PCE price
measure relative to the CPI. Clearly, as a measure of the cost of living, the
chain formula of the PCE index is superior to the fixed-weight structure of
the CPI, because it avoids substitution bias. In addition, the PCE program
has been more flexible in introducing new measurement techniques, and
the index can be revised to incorporate new source data. For example,
unlike the PCE index, the historical CPI data will never be revised to
incorporate the geometric mean weighting that will be introduced this
year. That said, the PCE price index does have one significant
shortcoming and that is its reliance on imputed measures of service prices
for some major components. These imputations, about 3½ percent of the
total index, are not actual measures of market prices, but rather are BEA
constructions, most using input cost information of questionable
reliability.
These imputed prices have had a pronounced effect on the pattern of
PCE inflation of late. The deceleration in core PCE over the past year is
almost entirely attributable to the slowdown in imputed service prices.

2/2-3/99

As shown in the lower right panel, excluding these prices, the increase in
core PCE was 1¼ percent last year, the same as in 1997. The bottom line
of our analysis is that consumer price inflation almost certainly has been
running below the rates suggested by the CPI, but the extent of the
slowing suggested by core PCE is considerably less certain.
Of course, we think that even these low rates of measured PCE
inflation are still biased upward; PCE prices suffer from many of the same
deficiencies of quality adjustment that afflict the CPI. The shaded band in
the upper left-hand panel of Chart 12 shows a one percentage point wide
confidence interval around our estimate of the upward bias in the PCE
price index, which we put at about ½ percentage point per year. By this
assessment, we may well have been at price stability last year.
As may be seen, our projection anticipates some turnaround in
inflation over the next two years. By 2000, total PCE inflation picks up
about one percentage point and the change in core PCE increases by about
½ percentage point. Some of the acceleration reflects our expectation that
imputed service prices will rebound from the unusually low increases of
the past year. More important, we are anticipating some reversal of the
favorable influences that have been helping to hold down inflation over
the past couple of years. Food prices--the black bars in the middle left
panel--are expected to be a neutral influence. But, retail energy prices-
the red bars in the middle left panel--are projected to post modest
increases over the next two years, after plunging in 1998. Moreover, core
non-oil import prices--the middle right panel--are expected to register their
first increases since 1995.
As Mike noted earlier, indicators of tautness in product and labor
markets, as measured by capacity utilization and the unemployment rate,
continue to diverge. As shown in the lower two panels, that difference
does not appear to be a statistical artifact. Business reports on vendor
delivery performance, the left panel, reveal few if any signs of production
bottlenecks. In contrast, the percent of households that perceive jobs as
plentiful--the black line on the right--exceed those reporting that jobs are
hard to get--the red line--by a very wide margin.
Turning to the upper panel of Chart 13, tight labor markets have been
an important factor driving an acceleration of real wages, measured here
as ECI compensation per hour deflated by the nonfarm business price
index. However, owing to low price inflation, hefty real wage gains have
required only a modest acceleration in nominal compensation over the past
couple of years. To be sure, favorable price shocks provided an extra boost
to real wage growth that we are expecting to recede as these influences are
partly reversed in coming quarters. But more fundamentally, we expect

2/2-3/99

the legacy of the recent low inflation to put a lid on inflation expectations,
and thus nominal wage demands, despite the tightness of labor markets.
The simulations presented in the middle panel highlight some risks
surrounding this outlook. A model in which wages are a function of,
among other variables, past consumer price inflation and the
unemployment rate--the dashed blue line--projects increases in ECI
compensation per hour that are even lower than the staff projection.
The picture is quite different, however, if wages are determined not
by past price inflation or inflation expectations, but rather by the past
momentum in wage increases. A model with these characteristics--the red
line--projects a substantial acceleration in compensation per hour over the
next two years. Barring a further increase in productivity growth, a stepup in labor costs of this dimension likely could not be absorbed entirely in
business profit margins and would result in more serious upward pressure
on price inflation than is envisioned in our forecast.
As you can see, the staff projection--the black line--is shaded toward
the wage-price model. We read the statistical evidence as favoring this
specification, but our projection is above this model because we are
skeptical that inflation expectations have fallen as much as suggested by
this model’s straight reading of lagged prices. And, we remain impressed
by the anecdotal reports that wages are under upward pressure.
In the lower panel, the staff projection for core PCE is compared with
two reduced-form models in which price inflation depends on lagged
prices and alternative measures of resource utilization. The staff price
projection--the black dashed line in the lower panel--is a bit higher than a
reduced-form model using the unemployment rate--the red line--largely
because we do not believe some of the special factors that have held down
core PCE, most notably the imputed services, will carry forward. An
identical model that uses manufacturing capacity utilization as the
measure of resource utilization, the blue line, suggests a more serious
departure from the staff projection. Over the longer haul, the empirical
evidence provides a slight edge to the unemployment rate formulation,
though in the past couple of years the capacity utilization model has been
closer to the mark. Taken together, the model results suggest that, while
there may be some upside risk to our wage projection that could well feed
through to prices, the low expected rate of factory utilization implies that
the inflation risks are not one-sided.
Your next chart reviews the productivity projection. As you know,
last spring we revised up our estimate of trend productivity growth--line 3
in the upper panel--to 1.8 percent at an annual rate. As may be seen on
line 4, the investment boom of recent years has lifted considerably the

2/2-3/99

pace of capital deepening--that is, the increase in capital per worker. And,
the available data hint at some step-up in the growth of multifactor
productivity--line 6.
The recent behavior of labor productivity appears to have conformed
reasonably well with our estimate that some acceleration has occurred in
the underlying trend. The middle panel shows our estimate of trend
productivity (the thin black line), actual productivity (the thick black line),
and a simulation of a productivity model that attempts to capture cyclical
variation around that trend (the red line). As can be seen, simulated
productivity closely matched actual last year.
The recent behavior of the unemployment rate also provides some
support for our productivity assumption. The lower panel shows the
actual unemployment rate--the black line--and a simulation of Okun’s law
starting in 1990--the red line. As you can see, the simulation using our
estimate of potential output growth of 2.1 percent in the first half of the
1990s and 2.8 percent since then has tracked the unemployment rate
closely in recent years.
While the recent data have been kinder to this aspect of our projection
than many others, it’s far too soon to feel confident. It should come as no
surprise that putting additional breaks in our estimate of trend productivity
improves the fit of these equations, and it is certainly possible that after we
complete this business cycle our upward adjustment of the trend will have
proven too optimistic. On the other hand, if we are in the midst of an
ongoing improvement in the pace of technological advance or
organizational efficiency, further upside surprises could be at hand. The
greater productivity implied by last Friday’s GDP release, if it should hold
up through revisions, would provide some support to that view.
In assembling this Chart Show, we briefly contemplated dispensing
with a discussion of our baseline Greenbook forecast in favor of
presenting only alternatives, given that the experience of the past year
might suggest that this Greenbook’s alternative will be next Greenbook’s
baseline. [Laughter] In the end, we decided not to take that approach, but
your next chart is offered in that spirit. Perhaps one of the largest risks
surrounding the performance of the domestic economy is associated with
the course of the stock market. In this chart, I consider the consequences
of both a continued boom and of an abrupt bust in the market.
In the boom scenario, we have assumed that the gains in the stock
market continue apace, with a decline in the equity premium sufficient to
boost the stock market--the blue line in the upper left panel--by 20 percent
in each of the next two years, all else equal. Monetary policy is assumed
to respond to this shock according to the Taylor rule. In contrast to our

2/2-3/99

baseline assumption of no change in the federal funds rate through the end
of 2000--shown as the black line in the upper right panel--the stronger
activity prompted by the booming stock market results in a 100 basis point
increase in the funds rate by the end of next year. This tightening of
policy leans against the strengthening of activity, but it is not prompt
enough or large enough to prevent the unemployment rate--the lower left
panel--from falling below 4 percent. Despite that, the inflation rate--the
lower right panel--remains about unchanged from that in the baseline; in
the model’s view, the increase in interest rates would boost the exchange
value of the dollar enough to offset the inflationary effects of tighter labor
markets.
In the bust scenario--shown as the red line in all four panels--we have
assumed an increase in the equity premium that would, all else equal,
produce a 40 percent decline in the stock market by the third quarter of
this year. In this simulation, the Taylor rule calls for a cut in the funds rate
of 125 basis points by the end of next year. The unemployment rate rises
to 4¾ percent by the end of next year. Again, the inflation rate is little
changed from the baseline path.
In constructing these scenarios, I have deliberately kept the
alternatives simple. In neither case have I allowed for any special
disruptions to financial markets or to the real economy not already
accounted for by the structure of the model.
However, if either scenario were to materialize, it is not difficult to
envision a considerably more challenging policy environment. For
example, a continuing boom could, at some point, spill over more
noticeably into other asset markets--adding extra stimulus to activity and
inflation. On the other hand, a steep decline in the stock market could
have more adverse consequences for the behavior of capital markets and
for credit provision than is embodied in these simulations. Moreover, if
such a decline were to trigger an abrupt shift in the economic outlook of
businesses or households, the risks of a sharp cyclical downturn would
increase markedly.
The final chart presents your economic projections for 1999. The
central tendency of your forecasts shows expected growth of nominal
GDP of 4 to 4½ percent this year. This is accompanied by growth of real
GDP between 2½ and 3 percent, leaving the unemployment rate in the 4¼
to 4½ percent range. The increase in the CPI is projected to be in the
neighborhood of 2 to 2½ percent. Mr. Chairman, that concludes our
presentation.
CHAIRMAN GREENSPAN. That was a particularly impressive performance by
the three of you. I thought it most interesting in all respects. I found the results of the

2/2-3/99

forecast scenario that assumes the largest change in stock market prices somewhat
startling in the sense of how restrained the secondary effects were. If we look at the end
result, we find that a stock market “bust” brings the unemployment rate to just over 5
percent. Five or six years ago we thought 5 percent was just terrific, an exceptionally low
number. The inflation rate in that scenario is under 2½ percent. Again, that is something
history would suggest is fully acceptable. So, if someone were to say that a bust in stock
market prices would leave us with 5 percent unemployment and an inflation rate of 2½
percent, some might say “Bring it on!” That tells us something about our inability to
create, with a limited number of equations, some of the history of past extraordinary
events. As Dave Stockton pointed out, the endeavor to hold everything else the same is
clearly misleading. Were these types of shocks to occur, I suspect the end results would
be quite different. But it is very difficult to capture exactly what those results would be.
Nonetheless, I think the forecast exercise is suggestive of the types of issues that are
involved. Questions for our colleagues?
MR. MEYER. I have a couple of questions. First, on Chart 1, the inflation rate as
measured by the CPI goes up almost a percentage point over the forecast horizon. In
your opening comments, Mike, you said that you were leaning toward giving somewhat
greater weight to labor market tightness in the inflation picture. Yet, as I look at the
projected performance of the CPI, it seems to me that it is mostly driven by the
dissipation of the declines in energy prices. The effect of labor market tightness is there
only marginally in the next year. This year the labor market tightness is still being fully
offset by other forces. Is that a fair statement?
MR. PRELL. The rise in oil prices that we have assumed is accounting for a very
substantial portion of the overall CPI increase in the forecast. In thinking about the story
in terms of underlying resource pressures, as Dave indicated, we have leaned more
toward the unemployment rate as the indicator of resource pressures rather than capacity
utilization. But the core price measures definitely do not accelerate the way the overall
CPI measure does.
MR. MEYER. I had a question, too, about Chart 10, entitled “U.S. Current
Account Sustainability.” You talked about the factors that have affected the
sustainability of the current account deficit since 1997, including the appreciation of the

2/2-3/99

dollar and the deterioration of external accounts. It seems that many of these
developments reflected cyclical phasing and the crisis in Asia, for example. So, I am
wondering why at some point in the future--when the cycles in the United States, Japan
and Europe become more in phase and we get to the other side of the Asian crisis--we
won’t recoup some of the losses in the current account. Are these transitory and related
to the cyclical crisis or are they long-term structural changes?
MR. ALEXANDER. I think in part what you are picking up are the initial
conditions and compounding effects. If we have a cyclical swing that makes the situation
worse now and if the other economies come back later, we will in the interim experience
a further increase in our net external debt position. And because of the compounding,
that matters. An underlying assumption in the Greenbook forecast is that there will be
some strengthening in foreign economic activity during the forecast period, but fasterthan-trend growth abroad is delayed until the period beyond the Greenbook horizon. So,
part of the problem is that if we have a boom and therefore a bigger trade deficit in the
current account now, we will accumulate additional foreign debt; and the interest cost of
that will mount. In part, therefore, the growing external deficit is the effect of the
compounding.
MR. MEYER. That’s a very good point. Just one last comment: I was a little
surprised in Chart 15 by the limited impact on inflation of whether there was a positive
demand surprise or a negative demand surprise. The Taylor rule, with the exchange rate
consequences, just offsets that. Was that not surprising to you?
MR. STOCKTON. Yes, it was.
MR. MEYER. Okay. Thank you.
CHAIRMAN GREENSPAN. Incidentally, as I think you are aware, the implicit
assumption for the crude oil price embodied in the forecast does exceed the price implied
by the forward futures market for the rest of this year and next. If you had substituted the
market’s estimate for the staff’s estimate, would that have made much of a difference in
the CPI forecast?
MR. STOCKTON. It would have taken a couple of tenths off the forecast. Our
price is just a dollar or so higher than the market’s, so the effect would be a tenth or two.
If one wanted to assume that oil prices stay flat going forward, that would take another

2/2-3/99

couple of tenths off, roughly speaking. The assumptions we have made on both the
import price side and on the oil price are critical factors in explaining the upturn we have
forecast in inflation.
CHAIRMAN GREENSPAN. There is a wage/price interaction, which means
that the change in the crude oil price gives us a multiplier in the inflation rate. Are you
counting that in the 0.2 and 0.3 or are you doing just a direct accounting translation into
the product prices in the CPI?
MR. STOCKTON. We are doing the full accounting. On PCE price inflation,
flat oil prices and, let’s say, 1 percent slower non-oil import price inflation would take
about 0.2 off the forecast we are showing in 1999 and 0.4 off in 2000, including all of the
feedback effects.
CHAIRMAN GREENSPAN. That is what I am getting at. It is a full simulation
effect. I would presume that if the price of crude oil went down, it would have even more
of an effect. So, in a sense a not insignificant part of the forecast is dependent on an
endeavor to make judgments about a very difficult issue. We do the best we can, but it is
important to realize that we could be wrong on either side of that. We could very readily
get an increase in West Texas Intermediate to $15 a barrel, and that would make a
difference; or it could go down to $9 a barrel and that would throw the whole forecast off.
It is important for us to recognize that so we can monitor this and make certain we make
some good judgments about where we are. President Parry.
MR. PARRY. Mike, I want to ask you a question about the performance of
equity markets over the last several years. Typically, one talks about that in terms of
earnings and P/E multiples. Does it make sense to focus on risk premia in equity markets
and look at it from the viewpoint suggested in some of the studies by Ibbotson and
Sinquefield? Those studies indicate that over the longer term there is a significant
premium in equity markets in the sense of a greater return on equities than on risk-free
Treasuries that is mainly a reflection of the risk premium. Is there reason to think that
over this period as a result of persistent growth of earnings and volatility--perhaps almost
all in one direction--that the market’s perception of the risk in equity markets may have
changed? If that risk premium is coming down, obviously a given trajectory of earnings

2/2-3/99

could have a powerful impact on the level of equity prices. It’s another way of talking
about P/E multiples that has a little more analytical basis, I think.
MR. PRELL. It certainly has an analytical basis. It is an abstract notion that is
unmeasurable.
MR. PARRY. That is why it’s worth spending time talking about it.
MR. PRELL. Yes. One of the active hypotheses is that investor perceptions of
the holding periods over which they need to assess the volatility in different assets may
have changed, so that investors are not as preoccupied with the greater short-run volatility
of equities as they were historically. People may have recognized that they had been
giving up excessive amounts of yield historically by avoiding equities, given this new
perception of risk. That might make higher valuations sustainable for given earnings
trajectories and interest rate levels. That said, we are really squeezing these things down
by some estimates to very, very narrow risk margins. And there is a certain euphoria that
one senses in the markets; people may be exaggerating the safety of equity investments.
So if a shock occurs that jars investors out of this complacency, if that’s what it is, we
could see a reversion to wider equity premia that, on top of more realistic earnings
expectations, could result in a very marked adjustment of equity prices. I might note that
the simulations that Dave Stockton presented were concocted basically by assuming
changes in the equity premium. So it is essentially a totally autonomous shock of this
sort that leads to these numbers.
MR. PARRY. Thank you.
CHAIRMAN GREENSPAN. President Stern.
MR. STERN. Thank you. The bottom panel of Chart 13 shows a set of
projections, one of which uses the capacity utilization equation. I believe you
commented that recent history would seem to be a bit more consistent with this view of
the world. That raises in my mind the fact that manufacturing capacity focuses on a
relatively narrow part of the economy. Have you thought about other factors that this
measure might be picking up? Is it a proxy for something? If you wanted to explain why
it seems to work, what explanation would you offer?
MR. STOCKTON. I would say it works principally because the manufacturing
sector is the most cyclical component of the entire economy and therefore involves

2/2-3/99

enough variation to pick up some correlation with what is going on in product markets
more generally. It certainly amplifies, in some sense, what is happening in product
markets more generally. But your basic observation that this is focused on a very narrow
portion of the economy is precisely what makes us hesitant to go the full distance toward
this model. We know that in the current circumstances, given what is happening to our
trade accounts, there should be special stress in the manufacturing sector and that,
therefore, it might not be as reliable an indicator of overall economic conditions as it has
been in the past when we have not seen such disparate behavior between the goods sector
and the rest of the economy.
MR. PRELL. I doubt that there is much that I can add to Dave’s expertise on this,
but one thing we have talked about among ourselves is whether the models that we use
really pick up the importance of the dollar and import price movements. Given the
confluence of events recently and the way they have impinged on the manufacturing
sector, this alternative is perhaps picking up a bigger import-price effect than is manifest
in the model simulation we have used for the unemployment rate version of the Phillips
curve.
CHAIRMAN GREENSPAN. What happens if you use the GDP gap instead of
the unemployment rate?
MR. STOCKTON. The outcome is quite close to the unemployment equation
results.
CHAIRMAN GREENSPAN. Doesn’t that address President Stern’s question?
MR. STOCKTON. It does that except one has to remember that the way we
construct the GDP gap is basically a version of Okun’s law. It is a little more
complicated than that, but it does not bring much additional information to bear relative
to what is already incorporated in Okun’s law.
CHAIRMAN GREENSPAN. In order words, these are two independent
methods, and the combination of the two doesn’t add anything?
MR. STOCKTON. Right.
CHAIRMAN GREENSPAN. President Poole.
MR. POOLE. I have a quick question of fact on financing conditions in the U.S.
corporate bond market. I should know the answer, but I don’t. Roughly speaking, what

2/2-3/99

fraction of the new issues is at the lower-rated end and what fraction is at the higher-rated
end? Your chart emphasizes the corporate junk bond rate, and that is up substantially.
How important is that quantitatively? I’m just asking for a ballpark figure on that.
MR. PRELL. To be honest, I don’t have a good number in mind. It has varied
considerably over time. The table on page III-2 in Part II of the Greenbook shows the
gross issuance of bonds by U. S. corporations in 1998. Sales of investment grade issues
averaged $14 billion a month, whereas those at the speculative grade were $9.5 billion
per month.
MR. POOLE. So the percentage breakdown is roughly 60/40.
MR. PRELL. If one wants to take into account the degree to which these data
represent refinancing, this becomes a tricky question to deal with. One of the problems
in trying to examine the changes we’ve seen in credit market conditions over the past
year is that the variations depend not only on a firm’s credit rating but on which
institution it turns to for financing and also the maturity range. Take, for example, a nonprime borrower looking at bank loans, where the money market rates have been coming
down considerably. Even if the spread widened a bit--and it may not have yet--the firm
might still be borrowing more cheaply at a bank as opposed to trying to sell securities in
the bond market. The cost may not be unambiguously higher.
It is a very difficult situation to assess. My judgment would be, taking the
totality of this picture going forward with our stock market assumption, that we do have
perhaps a modest degree of overall financial restraint. But it is hard to read at this point,
and thus you might not have discerned a clear-cut bottom line in my presentation. I think
it is a mixed bag.
CHAIRMAN GREENSPAN. We have fairly detailed information on the
outstanding stock of debt by credit ratings. I think the median is just at the edge of
investment grade versus noninvestment grade.
MR. PRELL. I suspect so.
CHAIRMAN GREENSPAN. Governor Gramlich.
MR. GRAMLICH. I am back on the bottom chart on Chart 13, and my question
is an offshoot to Gary Stern’s question. There is a lot of uncertainty about the NAIRU. I
believe your equations have it in the low to mid 5 percent area, perhaps 5.3 or 5.4

2/2-3/99

percent. This chart actually helps to get at a question that I have worried about, which is:
How important is that uncertainty? I think one might take the capacity utilization
equation as a proxy for things about as they are now, as if we are roughly at the NAIRU.
Your wage projections would be based on the notion that we are substantially below
NAIRU. So I think the answer to my question is that over a period of a couple years we
get about l percentage point more in inflation depending on whether this NAIRU
assumption is 5½ or 4½ percent. Is that roughly correct?
MR. STOCKTON. Over this two-year period that is exactly in line with the basic
rules of thumb, yes. It is certainly true that there is enormous uncertainty in all these
estimates. Indeed, in terms of the capacity utilization type equations, we are in essence
below the model’s estimate for the natural rate of capacity utilization. It might even
suggest, as this simulation indicates, that there could be some further downward pressure
on inflation if that in fact were the measure of overall resource utilization.
MR. KOHN. In the Bluebook, Governor Gramlich, there is a simulation with a
4½ percent NAIRU, which does exactly what you said. The economy is at equilibrium.
Nominal and real rates are where they need to be and inflation ends up about ¾ of a
percentage point lower than in the baseline.
CHAIRMAN GREENSPAN. President Minehan.
MS. MINEHAN. I found the paper that you circulated on the PCE versus the CPI
measures very interesting, and the charts are interesting as well. What I found myself
asking, however, was this: If price stability is characterized by people not taking
inflation into account when they are engaging in decisions--along the lines of the
Chairman’s definition of price stability--it is not clear to me which of these measures
does a better job of measuring the inflation that people perceive and care about. For a
long time we thought it was the CPI; in fact, I thought that was what the CPI was
constructed to do. There obviously are technical as well as nontechnical reasons to prefer
the PCE. But some aspects of that measure seem very arcane to me in terms of whether
people really see what is going on. Have I missed the point here somehow?
MR. STOCKTON. I think the clearest area where the PCE has that kind of
problem is that it includes prices associated with nonprofit institutions such as religious
organizations and other charitable organizations. If prices in those areas are changing

2/2-3/99

because the statisticians are imputing prices for labor costs and so forth, I don’t think that
affects anybody’s expectation very much. Clearly, one might argue that the nonmarket
price portion of the PCE is not so relevant.
On the other hand, I think that most people probably are not looking only at their
out-of-pocket expenditures in considering what is happening with the overall price
situation. For example, in the area of insurance costs they might very well care about
third party payments or the part that the government is picking up. I view the weighting
structure in the PCE as clearly superior and probably closer to what most people
perceive. It does not involve a fixed market basket of goods that people were consuming
five or six years ago but is more a period-to-period type of weighting structure that
probably is much closer to what people perceive is happening.
CHAIRMAN GREENSPAN. Are you saying that the CPI or the PCE weighting
is better?
MR. STOCKTON. The PCE weighting is probably more appropriate and closer
to most people’s thinking than the fixed-weight structure of the CPI.
CHAIRMAN GREENSPAN. Well, let’s assume that the CPI weighting is wrong,
because it clearly is a sample and people make very poor judgments as to what the
weights are with respect to what they spend. But that is indeed what they think they are
doing. The point is that their perception of the inflation rate may be wrong from an
economist’s point of view. But if you view it in the context of the point President
Minehan is raising, it is an interesting question of whether their perception is the relevant
one or not. Remember that in the University of Michigan Survey, for example, their view
of the rate of inflation is usually higher than it is in reality. The broader question is
whose perception matters--the perception of the business community or the consumer
community? I thought the memo demonstrated conclusively that the true rate of inflation
is far better measured by the PCE. But the question you are raising is a different one.
MS. MINEHAN. Yes, it is a different one.
MR. STOCKTON. I would add just one comment on that. Part of the difference
in reporting in the consumer expenditure survey that is used for the CPI is probably
deliberate. It is not necessarily a misconception. The fact is that consumers significantly

2/2-3/99

underreport their consumption of alcoholic beverages and tobacco. Now, it could very
well be that they just don’t remember. [Laughter]
CHAIRMAN GREENSPAN. There is some evidence that suggests that the more
you drink, the less you remember!
MR. MCTEER. And the less you care!
MR. PRELL. People certainly remember how much they paid for that last pack
of cigarettes. I suspect that even if they are underweighting it in the survey, it is on their
minds.
CHAIRMAN GREENSPAN. President Jordan.
MR. JORDAN. I want to ask about the last chart because it goes into the
Humphrey-Hawkins report and does get some attention from the public. I was struck that
the central tendency on the real GDP numbers and on the CPI numbers stayed the same
as those we had last summer but the nominal GDP range went down. That implies, I
suppose, that our idea of what deflator is appropriate was revised down compared to what
we thought last July.
MR. PRELL. We don’t get the members’ estimates of the deflator. I must say
that, when I look at the numbers you give us, I can see that in some cases people are not
distinguishing between the CPI and the deflator in getting to the nominal GDP number.
People may have been on top of that more this time than at a prior time. There is, I think,
a considerable looseness in these numbers. Also, we simply eliminate the high three and
low three from the individual distributions to get the central tendency and there can be a
lack of coherence across the components in some instances. I don’t know whether that
was the case, but that distortion can creep in.
MR. JORDAN. Last Friday’s report on economic activity in the fourth quarter
had nominal GDP growth at 6½ percent?
MR. PRELL. Yes, it was 6½ percent.
CHAIRMAN GREENSPAN. Incidentally, speaking of Humphrey-Hawkins, I
ought to remind you that Mike Prell will accept revisions to your individual forecasts
through the close of business on Monday, February 8. Are there any further questions for
our colleagues? If not, would somebody like to start the go-around? President Parry, you
have been drafted.

2/2-3/99

MR. PARRY. Mr. Chairman, the Twelfth District economy expanded at a strong
pace in 1998 and entered the new year with substantial momentum. Initial estimates
indicate that total payroll employment in the District grew by 2.7 percent last year, nearly
½ percentage point faster than the nation. Following a slow third quarter, District
employment regained lost momentum during the fourth quarter, when net hiring stepped
up noticeably in the retail trade and construction sectors. However, the District’s
employment growth rate in 1998 was about one point below its 1997 pace and our growth
rate advantage over the nation fell substantially. The key weak spot was the durables
manufacturing sector, which was hindered by deterioration in the District’s East Asia
export performance.
Among subsectors, a sharp contraction in the aircraft and parts manufacturing
sector has begun, with 3,600 jobs lost during the fourth quarter in Washington State and
another 1,800 jobs lost in Los Angeles County during the past year. Manufacturers of
computers and other high-tech equipment also had to face a slowdown. As a result,
employment in the San Jose metropolitan portion of the San Francisco Bay area was flat
during most of 1998.
Despite the slowdown in durables manufacturing in California, wage and salary
growth in the state has been sustained by the creation of high-wage jobs in other sectors
of the California economy. The largest contribution came from finance, insurance, and
real estate, which grew rapidly in 1998. Among other states, Nevada and Arizona were at
the top of the national employment growth ranking in 1998, and growth during the fourth
quarter surged in Oregon, Utah, and Idaho, where it had fallen nearly to a standstill
earlier in the year.
Turning to the national economy, our outlook has changed little since the
December meeting. We continue to forecast a slowdown in real GDP growth to around
2¾ percent this year and unchanged core CPI inflation of 2½ percent or slightly less both
this year and next. Along with the expected slowdown in growth, we have an assumption
of an unchanged stock market price level, reflecting the random-walk nature of stock
prices. In fact, the waning of the expansionary effect of past increases in stock market
prices contributes to slower consumption growth this year. I suppose I should mention

2/2-3/99

that we forecasted a similar growth slowdown last year, along with an unchanged stock
market.
It is not hard to think of developments that could alter this year’s outcome.
Recent problems in Brazil serve to remind us of the potential downside risks stemming
from fragility in many Asian and Latin American countries, as well as their implications
for international and domestic financial markets.
But on the other side, last Friday’s data on economic activity in the fourth quarter
followed the familiar pattern of strong output growth and low inflation. It seems more
and more likely that our economy is benefiting from a more rapid expansion of potential
GDP, probably due to advances in technology. In fact, our staff has looked at a
consumption-based measure of potential GDP, which has the potential to pick up shocks
to the supply side of the economy. It shows a considerably faster expansion in potential
output last year than do conventional measures. This factor also would help to explain
why inflation has been so well behaved in the face of strong labor markets and why the
stock market has been so strong. Thank you.
CHAIRMAN GREENSPAN. President Moskow.
MR. MOSKOW. Thank you, Mr. Chairman. The Seventh District economy
generally continues to show trends similar to what I reported in December, namely
strength in consumer spending and housing activity, mixed signals in manufacturing,
tight labor markets, and a few signs that inflation will accelerate in the near future.
Consumer spending remains healthy, boosted in part by robust activity in the housing
sector, including refinancing activity.
Two of our directors, one a major bank credit card issuer and the other a major
retailer, reported that consumers have been paying down outstanding credit balances, in
part reflecting strong mortgage refinancing activity. Most of our retailers experienced
better-than-expected sales during December, and that continued into January. Sales have
been particularly strong for big-ticket items such as appliances, consumer electronics, and
cold weather products. The blizzard we had in early 1999 forced some merchants to
close stores temporarily, but the impact on sales for the month was said to be minimal.
Light vehicle sales were in the stratosphere in December but came back to earth in
January. Some early January sales were included in the December figures. Inclement

2/2-3/99

weather crimped sales a bit in January, but sales continued to be boosted by high
incentives. Automakers have revised 1999 sales forecasts upward, although they still are
not quite as high as in the Greenbook.
Continued strength in the motor vehicle industry, including heavy trucks where
there is a 13-month order backlog, stands in sharp contrast to the weakness still being
reported in our steel and agricultural equipment industries. In the steel industry, two
small firms declared bankruptcy last year. Three more may do so in the first quarter of
this year. A couple of recently opened mini mills are now up for sale. Among the major
steel producers, half were still profitable in the fourth quarter, but all are likely to lose
money in the first quarter. This will make it easier for the industry to demonstrate injury
from steel imports. Production in the farm equipment industry this year is expected to be
down 25 percent. Moreover, since Brazil is a major soy bean producer, the devaluation
of the real will likely put downward pressure on prices and exacerbate conditions for our
soy bean farmers and ag equipment makers.
District labor markets generally remain tight, although some easing is reported in
locales affected by weakness in certain manufacturing sectors and in agriculture. In
terms of prices, firms continue to report that as a result of intense domestic and foreign
competition they lack the ability to raise prices. Firms continue to press suppliers for
lower prices, and we had several reports recently of larger firms, both within the same
industry as well as across different industries, establishing purchasing alliances to
increase their bargaining power with suppliers. But some firms have managed to increase
prices. Magazine advertising prices have been raised 5 percent. Also, several of our
directors expressed concern about higher prices for construction projects.
I asked our directors to report on 1999 capital spending plans. It appears that a
few of our District firms altered their capital spending plans in response to last fall’s
financial market turmoil. Changes that have been made appear to be driven by
prospective business conditions. Y2K issues are not driving changes in their current
plans at the moment.
Turning to the nation, our outlook for economic activity in 1999 has strengthened
somewhat since our December meeting. Our analysis suggests that real GDP growth will
be around 2.6 percent this year and that CPI inflation will be about 2.3 percent. Our

2/2-3/99

forecast is similar to the Greenbook’s, although the Greenbook has traveled a greater
distance to get there than we did.
Last week we met with our Academic Advisory Council. They continue to look
for a strong economy in 1999, with inflation remaining around 2½ percent. Though they
indicated that the current setting of monetary policy is about right for now, they expect
inflation to accelerate somewhat in the year 2000. I share this concern about accelerating
inflation. As we have discussed before, the special factors that led to the 1998 CPI
inflation rate of 1½ percent are not likely to recur this year. Since we see aggregate
demand decelerating only to the growth rate of potential output, resource imbalances are
unlikely to diminish soon.
Last fall there were very real concerns that the financial turmoil could negatively
affect creditworthy borrowers and institutions. This Committee’s aggressive actions
were an appropriate response to those shocks. But if and when those risks diminish
sufficiently, an appropriate response would be to move toward a more neutral policy
setting. The uncertainties about the Brazilian situation probably mean that it is still too
early to tell if its financial crisis has passed, but we may not be too far from knowing.
CHAIRMAN GREENSPAN. Incidentally, I have noticed that I am losing a
number of you for coffee. Is that an indication that we ought to take a short break?
MR. POOLE. Do we deserve it?
CHAIRMAN GREENSPAN. Let’s keep it to 15 minutes.
[Coffee break]
CHAIRMAN GREENSPAN. President Minehan.
MS. MINEHAN. Thank you, Mr. Chairman. The situation remains about the
same in New England. Labor markets are tight, with the regional unemployment rate at
3.2 percent in December. Connecticut and Rhode Island reported the two largest declines
in unemployment among the nation’s states in the month, and both states achieved new
lows for this decade. Job growth continues to be slower than that for the nation as a
whole, as it has been for most of 1998, and a wide array of businesses complain that a
lack of workers, skilled and unskilled, hampers growth. Even manufacturing firms that
are shrinking noted that labor markets are tight. Terminated workers do not remain out of
work for long. One firm used furloughs around the holidays to avoid layoffs. They

2/2-3/99

feared that layoffs would cause them to lose workers to other businesses and that they
would face search, recruitment, and training costs to get new employees when they
needed them.
Consumer prices rose measurably faster in Boston than for the nation as a whole
throughout the year, with prices of food and medical care rising markedly faster--at a
pace at least twice as fast in Boston as in the nation. However, tight labor markets and
local price pressures did not seem to lead to rising wages generally, at least in 1998. But
we continue to see large premiums being paid for workers in various skilled occupations.
At least part of the success in holding wages down has involved increased
investment in capital goods. A wide range of firms, from dairy farmers to jewelry
manufacturers, reported that they were increasing capital spending as well as engaging in
in-house training and offering more incentive pay to offset the rising cost of labor and to
make such labor more productive. Interestingly, while both retailers and manufacturers
expect a slower 1999, most retailers and about half of the region’s manufacturers expect
significant capital expansions in 1999.
Real estate and credit markets are healthy. Residential real estate indicators
exceed those of a year ago, with the market for newly built homes in the greater Boston
area very strong and prices up smartly. On the commercial side, the speculative wind
was taken out of the sails of developers and financers starting in the spring of 1998, after
the supervisory warnings on REIT lending. Major developers and lenders both tell me
that securitization may now be playing a role in stabilizing real estate cycles. The speed
with which the market corrects the cost of financing makes projects less feasible more
quickly than the more traditional process used by commercial mortgage bankers, who I
am told never saw a building project they didn’t love.
More broadly, credit spreads, while wider than at earlier points last year, do not
seem to be shutting out borrowers. Moreover, lenders across a broad range of financial
firms in Boston report very good conditions for their own profitability. One large insurer
reported very solid yields on lending activity and the lowest rate of delinquencies in at
least a decade.
Finally, amid the doom and gloom of reports about the U.S. agricultural industry,
there is a bright spot. New England’s dairy industry is reporting the best year ever. This

2/2-3/99

is due to higher milk prices and lower costs for feed and other--I should hesitate to use
this term--inputs. [Laughter] Even in dairy farming, however, there is a shortage of
workers. One contact, a cheese manufacturer, is running his cut-and-wrap operation on a
7-day-a-week schedule and is speeding up installation of labor-saving devices.
On the national scene, the data we developed for our Humphrey-Hawkins forecast
for 1999 differ very little from those in the current Greenbook, although we, like
Chicago, had a little less far to travel. We, too, project a flat stock market and stable oil
prices, and we see consumption and business spending slowing to about half the 1998
pace. Our forecast of overall GDP growth is a bit higher than the Greenbook’s and the
unemployment rate drops a bit even from its current low level. The continuing pressure
in labor markets and the flattening of oil prices produce a modest rise in inflation, with
both the core and the overall CPI rising to just below 3 percent by year-end 1999. In
view of this, for the Humphrey-Hawkins forecast we have projected a modest tightening
in mid to late 1999 and probably would see another modest tightening in 2000 as well.
Even with this, it is hard to imagine a more sanguine forecast than either ours or
the Greenbook’s. One has to wonder whether either is just too good to be true. We have
all talked about upside and downside risks to Greenbook forecasts for the past year.
However, the risks that have materialized all seem to have been on the upside. We have
seen lots of growth, which is good, but lots of upward pressure on asset prices and labor
markets, which could be bad. Moreover, since the financial panic last fall, credit and
capital markets seem to have resumed financing just about anything, albeit with greater
spreads than earlier last year and with increased volatility. Some fragility in these
markets remains, but it seems quite small compared with the problems we seemed to be
facing in October.
Arguably, monetary policy is stimulative, given the available liquidity in markets
and the reduction in real interest rates brought about by the 75 basis points of easing in
the fall. The question we have to ask is whether we really want to stimulate the economy
right now or whether it might be prudent to bring policy closer to neutral, recognizing
how difficult it might be to measure where neutral is. I do not say that because of a nearterm threat of inflation. I say it because whether one thinks we face risks on the downside from a large market break or from an international situation, or on the upside from

2/2-3/99

continuing pressures on labor capacity and wages, stimulative policy right now seems to
run a greater risk of making things worse later in terms of a big drop in the market--a
bursting of the asset bubble, if there is one--or more price pressures, if and when they
begin to build.
CHAIRMAN GREENSPAN. President Jordan.
MR. JORDAN. Thank you, Mr. Chairman. The regional economy was very
strong in the fourth quarter of last year, but that no doubt reflected the weakness at
midyear associated with the auto strike. But it is creating what I think of as a ski jump
effect. Inevitably we are going to reach a point where people’s expectations are going to
be disappointed.
The anecdotal reports suggest that people are expecting 1999 to be a stronger year
than 1998, but that may just be a reflection of how strong autos, construction, and other
sectors were as we finished the year. We are told that the backlog of presold, not-yetbuilt houses is at record levels and was rising as of year-end. Construction employment
this year is expected, by the construction trade unions anyway, to exceed the levels
recorded in 1998.
Overall, the labor markets continue to be very tight. One of the large regional
banks that operates in a number of states in the Great Lakes region said that as of the end
of last summer they had 3,000 open positions. In order to cut that, they significantly
increased their use of retention bonuses. For example, any teller who was on the payroll
on September 1, 1998 will get a $1,000 cash bonus if he or she is still on the payroll on
February 28, 1999. I’m going to call in the early weeks of March and find out what
happened after tellers get their bonuses! The company also was planning an across-theboard increase of 4 to 5 percent in base wages for 1999.
Reports from people in the retail sector almost never square with the subsequent
data. That may reflect overcapacity in that sector or it may just be unrealistic
expectations by the people in the sector. The retailers complained in December about
warm weather hurting sales; then they complained in January about cold weather hurting
sales. Even in the last week of December they were saying that retail sales were soft or
disappointing and they subsequently reported that it was the best Christmas since 1984.

2/2-3/99

A director from the retail sector says that no matter what happens in 1999, we will hear
retailers saying that it was not as good as 1998.
Bankers report that C&I loan demand is very strong. One of the regional banks
said that at the end of the year loans in the pipeline were at the highest level ever, and
they are now able to improve their profit margins. Earlier in the year they felt their
margins were being squeezed.
Even the hard-hit steel industry struck a note of optimism recently. Steel prices in
the fourth quarter of last year were said to have been down 6 percent on average from a
year earlier but some view that as the bottom. That may be wishful thinking. And it was
asserted that steel imports, especially from Japan, had peaked sometime last fall and are
now declining. With consumption last year at a record and expected to be as high or
better this year, steel industry executives are starting to turn optimistic. I agree with what
Mike Prell said about the earnings numbers. LTV reported a big loss for the fourth
quarter, but even they are less bearish going forward.
Turning to the national economy, I have the same problem with this Greenbook as
I have had with other recent ones, especially for the Humphrey-Hawkins projection
period, in that I like the Greenbook forecast; I just don’t believe it. I continue to want to
believe it because it looks so good.
For the first meeting of the year, I look back over the forecasts of the last several
years. This forecast is now the highest current year forecast for real growth that we have
had, at 2.6 percent on a fourth-quarter-to-fourth-quarter basis. That looks really good.
It’s higher than last year’s 2.4 percent. The year before it was 2.4 percent and in 1996 I
think the forecast at the first meeting of the year was only about 1.8 percent for the fourquarter period. Growth substantially exceeded that initial forecast in each of the last
three years and inflation came in lower than projected. It’s hard to beat that. It’s easy to
say: I want more of the same--faster growth, lower inflation. This is the first time in the
last couple of years, though, where the unemployment rate is not rising in the current year
and the out year. I also like that outcome, but I doubt that, given what we inherited from
1998, we have conditions in place that will produce that result again. Certainly in my
region, but I think nationally as well, a significant portion of the surge in economic
activity reflected an acceleration of final demand--fueled by what I consider to be very

2/2-3/99

rapid growth in all of the reserve, money, and credit measures--to a pace that was
associated in part with large and growing current account deficits. That, as the Chart
Show illustrated, gave the appearance of a virtuous cycle of rapid growth, strong demand,
and low inflation that cannot be sustained. Some components of that are going to start to
reverse on us and will give us a disappointing result of less output and more inflation.
The longer we wait to start to rein in some of the nominal aggregate income
growth and spending growth, the longer we are subsequently going to have to endure a
period of very weak growth in output and employment in order to lean against the rise in
inflation. So, at some point we have to contemplate an adverse transitory tradeoff, and I
think the sooner the better.
CHAIRMAN GREENSPAN. President Broaddus.
MR. BROADDUS. For the sake of variety let me start off by saying that not
everything is great in our District. Agriculture is hurting in our region as elsewhere. The
dairy farmers, as in Cathy Minehan’s District, are doing better than the others, but overall
the agricultural sector is very weak. Currency depreciations in a number of competing
countries have really hammered apparel manufacturers in the Carolinas, and there have
been some significant layoffs in that industry. So, we have a few holes in our region.
But, by and large, our District economy is probably as robust now as at any time I can
remember in my career. Even in manufacturing, which apart from agriculture is the
weakest sector of our region, there are pockets of strength. The furniture industry in
Virginia and the Carolinas, which has been in the doldrums for a long time, has been
revived to a remarkable degree by the strength of new and existing home sales.
Elsewhere, consumer sentiment and spending remain very strong essentially
across the board. We had a good holiday selling season, according to the anecdotal
information anyway. There is a lot of speculative building in the Washington region,
especially around Dulles Airport, and in several other cities in our District, notably
Charlotte. Labor markets remain very tight except in the areas I mentioned a little earlier
where there are some layoffs. State unemployment rates are below 4 percent in four of
our six jurisdictions. Generally, the picture is pretty strong.
Let me make my comments on the national economy in the context of our Bank’s
Humphrey-Hawkins forecast. We are asked to submit a forecast based on what we

2/2-3/99

regard as an appropriate monetary policy. I have always had a little difficulty with that
procedure because in some situations I am not at all sure that the Committee, in its
wisdom, is going to adopt what I personally think is appropriate monetary policy. It
usually does not do that. [Laughter] In that case, my projection would be misinterpreted
if somebody thought it was a statement of what I think is the most likely actual outcome
for the economy over the year. So, a little less ambiguous and perhaps more informative
way to proceed in my view would be simply to assume that there is not going to be any
change in policy, and that is what I have done in generating our forecast this year. As it
turns out, of course, that is the same assumption as the one underlying the Greenbook
projection, but we do use a different model to derive ours, namely a small six-equation
VAR model. We get broadly similar but somewhat different results: Our real GDP
growth projection is 3 percent as against 2½ percent in the Greenbook; and we have a
somewhat higher inflation rate of 2½ percent. At first glance, as we have been saying,
both our numbers and those of the Greenbook look pretty good. But the 2½ percent CPI
inflation rate we are projecting, while obviously a big improvement over what we had not
too many years ago, still worries me. That is really the main point I want to make today.
The Labor Department has now corrected about ½ percentage point of the 1
percentage point upward bias in the CPI that is usually associated with the Boskin
Committee Report results. So our 2½ percent CPI projection is equivalent to an unbiased
projection of 2 percent, which I personally believe should be the upper bound of our
tolerance range for inflation going forward. That seems to me as good a place as any to
draw the line on inflation, given where we are now.
In this regard, it is worth noting also that the core CPI rate increased from 2.2 per
cent in 1997 to 2.6 percent in 1998 on a consistently measured basis, which is not an
inconsiderable move in the wrong direction. The lower PCE inflation numbers certainly
are comforting to some degree, and I enjoyed the discussion about that today. But the
fact is--and this is related to what Cathy Minehan was saying--the CPI is still the nation’s
inflation standard. I think it, more than any other measure of inflation, drives people’s
expectations and perceptions of aggregate inflation in the country. So for now, at least
until that changes, it seems to me that the CPI is what we should focus on.

2/2-3/99

I would add just two quick final points. First, last fall’s ¾ percentage point
reduction in the funds rate was done because unsettled financial market conditions
threatened to restrain aggregate demand in the future, but the negative shock did not
really materialize. I am not sure that the implications of this are captured in our own
model’s estimation and hence in our forecast that I just summarized. If not, then our
forecast may understate and underestimate the potential lagged stimulus that could still be
coming in the months ahead from last fall’s easing. That could hit the economy at a time
when it is already in full stride. That, to me at least, represents a clear upside risk.
Second, M2 growth last year overshot the top of its target range by 4 percentage
points; if I am correct, that is the biggest overshoot by a substantial margin since the late
1970s or maybe 1980. I recognize that some of this problem is due to anomalous factors
affecting money demand. But keep in mind that our money targeting procedure--and
here is a blast from the past--still allows base drift. So last year’s overshoot will now be
ratified by the targeting procedure. That big potential monetary impulse will still be out
there as we move through 1999, even if we are able to keep M2 growth within whatever
target range we set for 1999. So, bottom line, I think the balance of risks in the outlook
has now shifted back to the upside.
CHAIRMAN GREENSPAN. President Boehne.
MR. BOEHNE. I could just say that the District economy is doing well and that
the risks to the national economy are balanced, but I will bow to Committee tradition and
elaborate some.
The regional economy in the Philadelphia District remains strong and, if anything,
has picked up a little strength recently. Manufacturing has accelerated in recent weeks,
following several months of deterioration. Retail sales have held up, with notable
strength in autos. Home building has been on the rise. Commercial construction is doing
well without the emergence of the kind of boom/bust signs that have characterized the
industry in past expansions. Labor markets remain tight. There are layoffs among some
of the larger firms, but smaller firms are hungry for employees. Large signing bonuses
for some job specialties such as programmers are common, but general wage pressures
are not showing through in the form of higher prices.

2/2-3/99

Looking ahead, most business people feel confident about the outlook for 1999,
but on the whole they expect a less robust year than in 1998. The risks to the national
outlook are still broadly balanced with perhaps a shade more upside than downside risk
because of the momentum from domestic demand coming into the new year.
Nonetheless, the downside risks are there, mostly in the form of potential turmoil
emanating from international developments. With inflation low and not likely to
accelerate quickly, we have the luxury of watching and waiting.
CHAIRMAN GREENSPAN. President Guynn.
MR. GUYNN. Thank you, Mr. Chairman. The Sixth District has begun 1999 on
a moderate growth path, consistent with a healthy and balanced expansion. Residential
building has moderated, at least for the moment, but that should be offset by a healthy
commercial real estate sector. Factory activity was sluggish again in December, and new
orders and production are soft compared to a year ago. However, the outlook indicators
from our latest survey are positive. The strength is in the high-tech area. The exceptions
are apparel and paper, which continue to be weak, but that is not a new story.
As for our important tourism industry, in recent meetings I have noted some
sluggishness in future bookings but the outlook is promising this time. While there was
some disappointment regarding the traffic over the year-end holiday season and some
continuing concern about a falloff in Latin American visitors, the Super Bowl provided a
big shot in the arm. And domestic tourists with money to spend fleeing the cold weather
in the North should ensure a strong first quarter. Airline flights to Florida and the Florida
resorts are fully booked through February, and the cruise industry is bullish on 1999 due
to strong bookings.
Low energy prices continue to pose a problem for Louisiana. The rig count
declined still again in December and now stands at 147, the lowest since August 1995.
The only clues about capital spending plans in the region come from our December
manufacturing survey. It showed that expected capital expenditures six months out were
more positive than in November. About one-third of our respondents said they expect an
increase in capital outlays over the coming period.
Consistent with the national price data, price inflation in the District generally
remains subdued. However, as others have said, labor markets remain tight; and we are

2/2-3/99

now getting reports of growing wage pressures, with increases in the 15 to 20 percent
range being necessary to retain key employees with certain specific skills. It is not clear
to us whether these wage gains are being fully picked up by the ECI data because many
of the increases are in areas other than regular wages, such as incentive pay, premium
pay, and bonuses. The only other area where we continue to get reports of price increases
is health care, as others have noted.
Brazil’s problems have worsened the outlook for the District’s exports to Latin
America, damping earnings prospects for regional companies investing heavily in that
part of the world. Although not huge in the total scheme of things in 1997, companies
based in the Southeast exported $2.6 billion to Brazil, accounting for about 20 percent of
the value of the nation’s shipments to that country. We think Florida is the state with the
largest exports to Brazil, with about two-fifths of the value of the state’s shipments
attributable to computer equipment, electronics, and transportation.
At the national level concerns clearly remain about the international sector. I will
turn to that in just a moment. I suggested last time that I saw a danger in having our
attention diverted from the stronger-than-sustainable growth and the inflationary
pressures that may now be building within the domestic economy. I am no less
concerned now than I was at the last meeting. Although our judgmental forecast, like
that of the Greenbook and most others, sees some slowing in the pace of growth over
1999, as of yet there are few measurable signs of that slowing. Our concern that growth
may once again turn out to be stronger than expected is further influenced by the
projections of our Bank’s econometric model. That new VAR model, which has tracked
the performance of the economy quite well over recent years, is now suggesting a
significant upturn in inflation during 2001 to a rate approaching the 4 percent level as
measured by the CPI. Importantly, that model also suggests that a significant increase in
the federal funds rate--to a level above what we may be comfortably willing to
contemplate--will be required to keep inflation at even a 3 percent pace.
While I am not campaigning for a monetarist label, I would also note that the
growth of the monetary aggregates shows no signs of abating and in fact continues to
accelerate. The longer that continues, the greater the risks may be of an outbreak of
inflation. I expressed the view last time that with the lags that we know exist, the

2/2-3/99

implications of this relatively long period of strong money growth may not yet have
shown through in our inflation measures.
Governor Rivlin also made some observations about lags at our last meeting,
suggesting that for a number of reasons we may now have the luxury of waiting longer to
institute needed policy changes because the economy has become more responsive to
shocks than it was previously. That hypothesis piqued our curiosity and we attempted to
investigate it a bit. We did so by examining differences in the impulse responses in our
model, estimated over different periods of time. Indeed, as was hypothesized, that
modeling work does suggest a faster response recently to real side shocks than was the
case previously, although the quantitative response to those shocks appears to be smaller.
At the same time, we found no evidence that there was any change in the economy’s
response to monetary policy shocks, which continue to work with long lags, requiring
over two years for 80 percent of the total response to be reflected in measures of
inflation. Obviously, we have a limited number of recent data points from which to do
such modeling but, in addition to the other reasons I suggested for caution, the results of
that work are flashing caution lights for me.
Turning very briefly to international developments: We have followed recent
problems in Brazil with special care, given the ongoing bank supervision work we do in
Latin America, working with central banks and other supervisory authorities. Early in
the year my staff was telling me that the major key to possible developments in Brazil lay
in whether authorities could engineer an orderly depreciation of the currency and whether
it could be accommodated peacefully by Brazil’s multilateral creditors. We did not
expect that the depreciation would come so soon or that the value of the real would
decline by 40 percent or more. The good news, it seems to us, is that there has not yet
been much spillover effect to other countries as might have been expected. After initial
declines, the Mexican and Chilean pesos have recovered somewhat. Argentine
authorities have assured that peso/dollar convertibility would be maintained, and
Venezuela does not appear to be under substantially more pressure than it was prior to the
Brazilian problems. Moreover, trade ties with Brazil and the rest of Latin America,
except for Argentina, are relatively limited.

2/2-3/99

In summary, the international situation certainly remains fragile, and we could
have to contend with the shock of worsening problems in Brazil, added contagion from
the Brazilian problem, or both. Having said that, I don’t think we should give undue
weight in our policy deliberations at this meeting to what could happen in that part of the
world and too little weight to the still very strong domestic economy, given inflationary
tendencies that have been identified by both the Greenbook and the independent work at
some of our Reserve Banks. Thank you, Mr. Chairman.
CHAIRMAN GREENSPAN. President Hoenig.
MR. HOENIG. Thank you, Mr. Chairman. The Kansas City District continues to
do well and its economy is basically sound. Our weak areas continue to be agriculture
and energy, and there has been some slowdown in certain areas of our manufacturing
sector. Of course, there is some concern on the manufacturing side about the Brazilian
situation, particularly if that were to spread to Mexico, on which we are much more
dependent. Having said that, the economy otherwise is doing very well even in some of
the manufacturing sectors as well as in the housing sector and construction generally.
Retail sales continued strong, even in January. Overall, our major cities are still
booming.
On the national front, I expect the economy’s growth to moderate toward trend
this year with low inflation, as others have said. In other words, basically we see the
economy continuing to grow. The issue we face continues to be very strong domestic
demand set against the backdrop of weak international demand, which is further
complicated by the possibility of adverse shocks. That has not changed much in the last
several months. Nevertheless, compared to last fall I believe that the risks to the outlook
are now far more balanced because of the earlier rate reductions taken to offset the global
financial turmoil. As a result, as others here have indicated, I am becoming more focused
on the potential upside risk to inflation if we continue to maintain our current monetary
stance. To me, a key issue for this Committee at this meeting or very soon is whether,
when, and how fast the policy actions of last fall should be unwound. Thank you.
CHAIRMAN GREENSPAN. President Stern.
MR. STERN. Thank you, Mr. Chairman. The economy of the Ninth District
remains strong outside of some parts of agriculture and mining. But aside from that,

2/2-3/99

consumer spending is continuing to expand, construction is very strong, and labor
markets remain very tight. The Twin Cities economy without question is booming, and
that is a term I do not use often. As just one indication of that, the unemployment rate in
the Twin Cities metropolitan area is now about 1½ percent, which obviously means that
everybody who wants a job has one or probably several. So that area is going along very
well. There are wage pressures but selective ones. They tend to be for some entry-level
positions, for some information technology professionals, and so forth, but they by no
means seem to be generalized. When we ask business people whether they are seeing
inflation or deflation, the answer we get back most frequently is “neither.”
The one thing that gives me real pause about the District economy--and it is a little
disconcerting--is that bankers still seem to be chasing deals very aggressively. That is
my general sense from conversations with them and with some of their customers.
As far as the national economy is concerned, let me say first that I was impressed
with both the number and the variety of the scenarios presented in the Greenbook, the
Bluebook, and in this afternoon’s presentation. I believe they help us think about the
risks and about how we ultimately might want to be positioning monetary policy. But an
equally intriguing question is the reasonableness, or the accuracy, of the baseline
forecast. In that regard, I would say that I am a bit more optimistic about real growth in
1999. That is as a consequence of both the momentum behind aggregate demand as we
go into the year and also because I am more positive about productivity trends and,
therefore, the supply side. So, I think we might see somewhat greater real growth than in
the Greenbook. Another reason for being positive about the real outlook is that our VAR
forecasting model is quite positive. It has been reasonably accurate in 1996, 1997, and
1998, so it is harder to dismiss the model forecast than it used to be since it is building up
a bit of a track record.
I agree with the Greenbook’s view of a modest uptick in inflation mainly because
I think there will be some unwinding of the effects of the favorable shocks that we have
experienced in recent years. But I must say the anecdotes about essentially no inflation
give me some pause as I think about that.
CHAIRMAN GREENSPAN. Governor Rivlin.

2/2-3/99

MS. RIVLIN. I am glad I stimulated research in Atlanta. I am interested in
seeing it.
A couple of meetings ago I referred to the cheerful little elves who run the U.S.
economy and get their kicks out of proving the cautious forecasters wrong. The only
thing that can be said about the economic news since our last meeting is that the elves
have scored again. When I heard that the estimate of the 1998 fourth-quarter GDP was
likely to be about 5 percent, I said, “Wow!” Then the official estimate turned out to be
5.6 percent with the possibility of an upward revision to perhaps over 6 percent.
The elves clearly have been working overtime and they have gotten more
ingenious. They have figured out how to control the weather, at least temporarily, and
how to keep productivity growth increasing when any reasonable elf would suspect that
all the reengineering and restructuring and computerizing that could be done had been
done already. Most amazing of all, they seem to have figured out how to keep
unemployment rates lower than what the NAIRU enthusiasts have said for a long time is
the drop-dead rate, while wage increases actually have decelerated and inflation does not
seem to be a danger at present.
The elves are clearly beginning to undermine the confidence of their opponents,
the Greenbook forecasters, [laughter] who are starting to doubt their own models, or at
least are getting a little defensive about them, and perhaps even doubt their common
sense. To be sure, the Greenbookers are too sophisticated to fall for the elves’ more
brazen gimmicks like the weather ploy, or the bounceback from the auto strike, or
perhaps even the extraordinarily low oil price. They think those cannot be sustained, so
they are boldly forecasting a relatively quick slowdown to trend and some acceleration in
prices.
But after pointing out for some time that stocks were overvalued and that the
stock market likely would come down, the Greenbookers are now saying that those
stocks are even more overvalued than the last time they looked. And although they
expect profits to decline further, they do not expect stock prices to fall. They will only
move sideways. What can be the explanation of that? It’s only the fear that the elves
have been right so often that the Greenbookers would be embarrassed to be caught wrong
again. Besides, nobody can predict the stock market.

2/2-3/99

The question now is: Have the elves run out of tricks or can they beat the game
one more time? If it were not for the rest of the world, I would bet on the elves.
However, I am not sure they speak Portuguese. I am not sure they can move the
Brazilian political system into higher gear fast enough to reassure a nervous world
market. I am not sure the elves, for all their ingenuity, can control the mood swings of
international investors or control the contagion that might flow from a collapse in Brazil
or from some other major negative event as yet unpredicted.
So despite history, I am left thinking that the Greenbook may actually win this
time, though I must admit that I also thought that about the Atlanta Falcons. [Laughter]
There is some risk that the economy may prove stronger in the early part of the year than
the Greenbook forecast. I also have my doubts about whether the increase in the oil price
will be as much as they think. On the other hand, the Greenbookers may have
overreacted to the elves on the stock market and could be understating the chances of a
negative wealth effect. On the domestic front, I see the risks of upside and downside
surprises around the Greenbook forecast as approximately balanced. However,
internationally, I see the risk as principally on the downside and possibly quite serious.
As for the FOMC, I believe we should watch the game very carefully from our
comfortable seats on the sidelines but resist getting into the action. Right now, any
action, or even any indication of possible action, is likely to do more harm than good.
CHAIRMAN GREENSPAN. I kept hearing green eye shade when I know that is
not what you said! President McTeer.
MR. MCTEER. Do I have to? [Laughter]
CHAIRMAN GREENSPAN. No!
MR. MCTEER. I will be very brief. I think Alice ought to write a poem entitled
“The Elves versus the Greenbookers.”
SPEAKER(?). It doesn’t rhyme.
MR. MCTEER. Little has changed in the Eleventh District since our last
meeting, so I will be very brief. As I said then, the growth rate in the Eleventh District
economy slowed throughout 1998, and that has continued in recent weeks. With oil
prices hovering in the $12 range and not expected to pick up appreciably in the near
future, consolidations and layoffs in the energy sector will likely continue for quite some

2/2-3/99

time. Mexico, which accounts for nearly 40 percent of Texas exports to foreign
countries, continues to weaken due to lower oil prices and the higher interest rates they
experienced in the weeks following Brazil’s currency devaluation. This does not bode
well for the Texas economy in the months ahead.
On the other hand, our contacts in the semiconductor industry, which also has
been a source of weakness recently, indicate that there are signs that demand has
bottomed out and that shipments have begun to improve. We recently learned that some
of the workers who have been laid off in the semiconductor and communications sectors
have been rehired as contract workers. Other segments of the economy that continue to
show strength are all the areas of construction.
Labor markets remain exceptionally tight, and the only group that we hear has
pricing power is computer programmers; everyone else says they have none. Office
rents, which were rising very steeply in the first half of last year, have been increasing
much more slowly as new supplies have come on the market. Other than in these two
areas, there is little or no talk of inflationary pressures in our District.
Regarding the national economy, I am looking for 1999 to be pretty much a repeat
of 1998. I expect the expansion to slow only slightly, due primarily to labor force
constraints, but for inflation to maintain its downward tilt. Our contacts in the oil industry
are considerably more pessimistic than the Greenbook staff regarding oil prices over the
next year or two. I guess outside of Texas that pessimism would translate as optimism!
CHAIRMAN GREENSPAN. President Poole.
MR. POOLE. Mr. Chairman, in the Eighth District the story is primarily an echo
of recent months. The Eighth District was not much affected by the credit market
disturbances last fall and, therefore, we have not been much affected by the weakening of
those disturbances. Rather than repeat the echo of what we have seen, I would just like to
make one comment: It seems to me that more and more firms are learning to live
successfully with what they had regarded as labor shortages. They still do not see the
pricing power needed to raise prices and, therefore, they are learning to get along with
what they previously regarded as a short-staff situation. We just do not hear so much
about that any more, although when we ask people they tell us about all their unfilled
positions.

2/2-3/99

On the national level, my contacts at UPS and FedEx say that they see steady
growth in the United States. They see the situation in Europe as solid. They see a
bottoming process in Asia and they see Latin America as weak, though they do not have
extensive business there.
I would comment with regard to Brazil that the market response to the Brazilian
upset has seemed to me very sensible and measured. The markets have made distinctions
that make good sense; that is completely unlike the response in August to the problems
faced in Russia. I think the situation is entirely different.
The staff forecast makes good sense to me. We could all quibble with a tenth or
two here or there. I think the staff has done a fine job of incorporating everything that is
reasonably forecastable. What is going to get up and bite us at some point is something
that we cannot foresee, but I am not going to criticize the staff for not forecasting the
unforecastable. That’s all I am going to say at this point.
CHAIRMAN GREENSPAN. Governor Meyer.
MR. MEYER. Thank you, Mr. Chairman. The forecast revisions in the
Greenbook are very similar to my own and in my view constitute somewhat of a sea
change in the economic outlook. It doesn’t seem to me that most of you share that view,
so let me explain why I think that is so. Previously the Greenbook, the consensus
forecast, and my own forecast projected what I refer to as a reverse soft landing. Growth
slowed to below trend and the unemployment rate gradually increased, unwinding some
of the exceptional tightness currently prevailing in labor markets and reducing the
inflation risk posed by the above-trend growth and the very tight labor markets. The
revised forecast has the economy slowing, but now just toward trend. And the result is
that the unemployment rate stabilizes at a lower level than for any quarter in the previous
Greenbook forecast.
My first question is: Should such a sea change in the outlook have a counterpart
in monetary policy? When I talk about policy here, I really am not referring to the
targeted funds rate for the intermeeting period that we will talk about tomorrow. I am
talking about policy in terms of the path of the funds rate that we think would be
consistent with our outlook over the next year or two and whether or not that should
change in light of the changed forecast.

2/2-3/99

Let me just say a word about the inflation forecast. It does not seem to have
changed dramatically, but it has changed a bit. First of all, we have mainly what I call a
convergence story in 1999. That is, although core CPI does not really change, we have
an increase in both the overall inflation rate and in GDP chain inflation as these measures
converge toward more normal relationships to the core rate, with overall inflation rising
because of the dissipation of the favorable supply shocks, particularly in the energy area.
But thereafter, with labor markets very tight, the Greenbook and my own forecast would
expect continued increases in inflation going forward. I think that is what we have to
worry about.
I pay a lot of attention to the policy prescriptions from the Taylor rule. Sometimes
the different rules that are in the standard packet yield quite different implications for
policy. Today, while they offer a variety of prescriptions for the current rate setting,
there is one commonality. Whether one looks at the CPI or the chain GDP price version,
at the versions with imposed or estimated coefficients or with the backward- or forwardlooking specification, the prescription for the federal funds rate in the current quarter is
higher than the prevailing target. For example, it ranges from 5.1 to 6.3 percent for the
rules with imposed coefficients.
So, I ask myself: How have we ended up departing so aggressively from the
Taylor rule prescriptions? First, we hesitated to tighten in the face of global instability
following the crisis in Korea and then again following the turbulence after the
devaluation and default by Russia. But neither of those events to date has slowed the
expansion. Second, our forecast generally called for a slowing to trend growth just
around the corner, so we waited for the spontaneous slowdown rather than imposing a
policy-induced slowdown. Of course, while we waited for the slowdown, continued
above-trend growth kept pushing the unemployment rate lower until we ended up at a
4¼ percent unemployment rate. And now we find ourselves at that rate with one of the
highest Greenbook growth forecasts in some time and many other forecasts also are
pointing to relatively robust growth. Another reason we have ended up there certainly is
the possibility of a structural change suggested by the combination of declining inflation
along with a declining unemployment rate. That provided a good reason for not slavishly
following any historical regularity.

2/2-3/99

Let me end with a suggestion on how we might want to think about the strategy of
monetary policy going forward. We ought to follow and think about what I refer to as an
incremental asymmetric Taylor rule. To start off, I would consider setting the initial
specification of that Taylor rule by calculating what NAIRU would have to be to make
the current setting of the funds rate the Taylor rule prescription. The answer is that in the
Taylor Rule with imposed coefficients, the NAIRU would have to be about 4½ percent. I
think that is very much on the low side, but some might find it a plausible number. That
is fine. It seems to me that going forward we should at the very least follow the Taylor
rule incrementally, raising the funds rate if continued above-trend growth was pushing
the unemployment rate even lower. That is, going forward we should be careful to begin
again to lean against a cyclical wind; otherwise we will continue to accommodate and
indeed reinforce any and all positive demand shocks.
On the other hand, if the unemployment rate were to rise modestly, given its
already very low level, I would resist easing immediately. Hence the asymmetry in my
approach. If inflation increases--and here I mean increases in the core CPI rather than a
convergence of the overall rate to the core CPI--then I think we should respond as in the
Taylor rule with more than proportionate increases in the nominal funds rate. On the
other hand, if inflation declines modestly further, I would passively accept an increase in
the real federal funds rate in light of the very low level of the prevailing unemployment
rate. That again is an asymmetric response. So, I offer that as food for thought as we
turn to discuss policy strategy in more detail tomorrow.
CHAIRMAN GREENSPAN. Vice Chair.
VICE CHAIRMAN MCDONOUGH. Thank you, Mr. Chairman. The Second
District’s economy retained strong momentum going into 1999, with price pressures
largely in check. Private sector job growth in New York and New Jersey accelerated to a
3.1 percent annual rate in the fourth quarter. New York City registered its strongest
annual job growth on record, 2.7 percent; the previous high had been 2.2 percent in 1969.
Retailers report brisk post-holiday sales in January, buoyed by a later-than-usual cold
snap, a relaxation for eight shopping days of the sales tax in New York City, and I
suppose the view that since it got cold people needed winter coats after all.

2/2-3/99

Availability of office space in the New York City area was tight but stable in the
fourth quarter. Vacancy rates stopped falling and rents rose at a less frenzied pace. Our
District’s housing market showed more signs of strength in December as indicated by
rising home construction, brisk home sales, and sturdy price appreciation. A
benevolently warm, delightful December might have had something to do with that.
Surveys of Purchasing Managers indicate continued weakness in the region’s
manufacturing sector. Hotel occupancy rates in Manhattan remained exceptionally high
in December, with room rates continuing to run about 10 percent ahead of a year ago. In
spite of all that, the CPI in the metropolitan New York City area rose only 1.6 percent in
1998. That is the lowest since 1964, and I assume that the mayor is extraordinarily
unhappy that he cannot run for reelection.
At the national level, our forecast for growth is very much like that of the
Greenbook. The Greenbook suggests 2.6 percent in 1999 and we have it at 2.5 percent;
we have growth at 2.3 percent in 2000 as compared with 2.4 percent in the Greenbook.
We are a little more concerned about--or at least our forecast shows a higher increase in-
inflation. We have the CPI at 2.8 percent in 2000 compared with the Greenbook’s 2.4
percent.
The question of the balance of risks is an interesting one at the moment. If one
were looking only at the domestic economy in the United States, there is no question in
my mind that the balance of risks would be on the upside. That is, the economy is likely
to grow faster than we anticipate. On the other hand, the risks on the international side
would have a negative effect. But I think they are likely to be understandable more
quickly than has been the case in the past. Japan is weak and could get weaker because
of the combination of

and terribly low consumer and business confidence.

Continental Europe and the United Kingdom are slowing down a bit, but probably not to
a point where some policy action couldn’t turn those economies around.
The interesting case, obviously, is Brazil. Brazil is a country that I think is very,
very different from just about every place else. It has immense natural resources and a
very diversified high quality population. They are given to considerable swings of
confidence and to periods of believing that the national leader is capable of great and
wonderful things. For a quite understandable reason, they thought that President Cardozo

2/2-3/99

would be able to continue to lead the economy and to make good his commitment to the
exchange rate design, which was both the economic and perhaps more importantly the
psychological base of the real plan. The law of gravity prevailed, or markets prevailed,
and that was no longer possible. The leadership in the country had been so committed to
what they were doing that they did not have contingency plans ready. That is not
surprising; most people don’t have contingency plans ready. Therefore, the Brazilians
are still going through a period of trying to figure out what to do. However, just in the
last couple of days, it looks as if people are beginning to find a more positive view of the
situation. Now, that could reverse because as one bumps along a psychological trough it
is difficult to know whether the next bounce will be down or up.
The important thing is that between now and our next meeting, the Brazilians
either are going to restore their sense of confidence in their leadership and their country
or they will not. If they do, the Brazilian shock probably will be less great than we now
think it is likely to be. On the other hand, if they don’t regain their confidence, the shock
will be a bit greater and the contagion effect on Argentina and Mexico a bit more severe.
Therefore, at the present time watching and waiting clearly makes a great deal of
sense for us. We do have to be careful not to fall into the trap of thinking that we are the
central bank of the world because we are not. With any luck, we will have a much better
opportunity at our next meeting to calculate more accurately what Brazil is likely to do.
Then we will be in a position to make a more accurate judgment about what policy path
we should follow and how quickly we might need to adjust our current policy.
CHAIRMAN GREENSPAN. Governor Gramlich.
MR. GRAMLICH. Thank you, Mr. Chairman. I felt the fourth-quarter growth
statistics were significant news. They pushed the slowdown further into the future.
There is still a slowdown in both the Greenbook and the Blue Chip forecast but it is
smaller. The Greenbook now has only one quarter of growth below 2.5 percent. It seems
to be nearly a perfect soft landing in both forecasts without requiring policy changes.
The question is: Is that too good to be true?
The slowdown in the Greenbook forecast hinges on three factors. I would like to
discuss each one briefly. The first is the international situation. I am beginning to
change my view about that because in a way we have taken the full shock. We are now

2/2-3/99

on the pessimistic outcome track for Russia, Japan, and Brazil, along with having a
modest slackening in Europe. The effects are noticeable, but as we saw in the Chart
Show earlier, they have done relatively little damage to U.S. demand growth. The U.S.
economy may not be an oasis of prosperity, but at least it seems to be a cluster of palm
trees. [Laughter]
On the stock market, even the Greenbook seems to have given up waiting for the
market to drop and now projects only a sideways movement. Actually, I more or less
agree with that. My own reasoning is similar to the point that Bob Parry’s question
brought out earlier, which is that the earnings/price ratios are not out of line with falling
real interest rates over the past few years. So I think the Greenbook forecast on stock
prices is roughly accurate, and that has actually changed the staff’s forecast some.
The third factor is something we have not talked about much here; it is the
investment accelerator. Investment is slowing down from the rapid pace of a few years
ago, but it is still growing more rapidly than output in the forecast. With the prevalence
of information technology investment, it’s not clear to me how much of an acceleratorinduced slowdown we should expect in investment any more.
In summary, my view is that each of these negative factors that could generate a
slowdown is either partly digested, weakened, or in some doubt. There are still grounds
for expecting some slowdown, but as far as I am concerned the balance of risks may be
shifting. At the same time, I think it is too early to change policy because I also see very
few signs of price acceleration yet. There is some in the Greenbook forecast, but that is
partly due to special factors such as oil and partly due to an estimate of NAIRU that I am
growing uneasy with, though I don’t want to write down my own number. There is not
much acceleration of inflation in the Blue Chip forecast. So, all things considered, it is
difficult to proclaim that we see an acceleration yet.
The second factor is productivity growth, which Dave Stockton illustrated in his
charts earlier. That growth may be stepping up by a fairly noticeable amount, which
actually gives us more room for vibrant demand growth.
The third factor is the lags in monetary policy, the subject that Governor Rivlin
put on the table at our last meeting. Unlike the Atlanta Fed, I was not able to do any
research on this in the last month, so I accept what she said. I do believe that monetary

2/2-3/99

policy operates more quickly these days, and I think that does give us the luxury of
waiting for evidence. We still have to be alert. We have to move quickly. But we don’t
have to move in advance of real evidence of acceleration in inflation which, as I said, is
not available yet.
All this adds up to me as an argument for sitting tight. But as a result of the
continued strong growth in the U.S. economy, I do think the balance of risks is changing.
Thank you.
CHAIRMAN GREENSPAN. Governor Ferguson.
MR. FERGUSON. Thank you, Mr. Chairman. As we start the new year, we are
clearly faced with a different situation from the one we faced last year, and we probably
will have some difficult decisions to make in the not-too-distant future. Last year, our
forecast of slowing obviously was being driven by external events; now we have a
forecast that suggests slowing will be driven to a large extent by internal forces, namely
private domestic final purchases. It is obvious that just as domestic forces proved much
stronger than the drag from deteriorating net exports last year, so too the forward
momentum of the economy may prove to be stronger than the forces for domestic
slowing featured in the current forecast.
As the Greenbook notes, the near term will probably reflect some unwinding of
special factors that led to the upside surprise in the fourth quarter, namely unusual
weather and an auto strike rebound. Certainly there is some truth to that. But the major
factor for the long-term slowing, as I read the Greenbook, appears to be a forecast that the
bull market has run its course and that the household wealth-to-income ratio will decline.
I must admit that to me, as to others, the stock market does seem to be levitating above
what one might think of as reasonable levels. But it is also true that we simply do not
know enough about the forces that drive the stock market to put significant weight on a
forecast that has a flat stock market as a prominent feature. Therefore, it is really unclear
to me that we are going to see the slowing that we all seem to have in our forecasts. The
risks seem to be mainly to the upside. It is easy to determine what those risks are and
what may drive economic growth above the forecast: plentiful jobs, accommodative
credit conditions, and upbeat consumers.

2/2-3/99

The external sector was a major downside risk last year. This year much of that
has been resolved with the obvious exception of Brazil. Brazil is still a large and
troubling question mark, but the contagion thus far has been contained and market
reaction has been muted. Most of the other issues that we were concerned about last year
seem to have resolved themselves. They are reasonably well understood and probably
have already had their strongest impact on the U.S. economy. Emerging Asia seems to
have bottomed out. The most credible forecast for Japan is that while it will not
strengthen in the next several years, the deterioration will be moderate from this time
forward. Europe shows some slight weakening, but there are reasons to believe that
monetary policy there can offset that. Therefore, despite the uncertainty created by the
Brazilian situation, we have probably seen the worst of the risks from the external sector.
While citing these forces for faster growth, one must also admit that there are few
signs yet of emerging inflation. The economy has achieved this benign price picture
through a combination of special factors and possibly through increasing productivity
growth. However, there are some risks to the inflation outlook. One is that the special
factors will unwind more quickly than we currently expect. Another is that labor market
tightness will worsen at a faster pace than businesses can offset. It is not only that the
unemployment rate is at a generation low, we also have fewer excess reserves, if you
will, in the labor market. The number of people of working age who are not in the labor
force but want a job has been decreasing at a fairly steady pace and is now at its lowest
level since 1970, when that statistic first began to be tracked.
Given the string of surprises we have had, including the strength of the fourth
quarter, like many others here I approach a forecast of slowing with some skepticism and
see the risks both internationally and perhaps even domestically as shaded slightly to the
upside compared to the baseline forecast. I think the appropriate response for us is to rely
less on a future as predicted by models and more on inferences, both quantitative and
qualitative, that come in with the latest data. Obviously, for those of us who take this
approach there are some challenges. By necessity we are likely to have a shorter time
frame for action. However, the corollary is that once the evidence is more clearly in
hand, we should not hesitate to move decisively if the data so warrant. We are not
precluded from acting preemptively if new information were to tip the balance of risks to

2/2-3/99

the outlook much more decisively toward an unacceptable probability of higher inflation.
Such a probability, if it were to emerge, would in my judgment warrant a policy response.
We are not there yet in my view, but we may be soon. We must be careful not to
misinterpret the signals that we receive.
CHAIRMAN GREENSPAN. Governor Kelley.
MR. KELLEY. Thank you. Mr. Chairman, I would not have thought it possible,
but somehow the Committee’s policy dilemma continues to deepen rather than to show
signs of beginning to resolve itself. The economy continues to surge ahead. We all, or at
least most of us, believe that it should and will slow. But we have had good reason to
believe that for some time and yet the pace, if anything, has accelerated and there are few
signs of slowing so far. If the pace of growth begins to slow as we expect, we likely will
have to address the possible need for further easing at some point. If, however, the
present pace or something close to it should continue, we soon will almost surely need to
consider at what point policy will have to lean into that strength in the interest of
sustainability. I say this realizing fully the difficulty that such a policy move could
present internationally. But even if one adopted the most optimistic reasonable scenario
of the strength and durability of the current virtuous cycle--and I am pretty much in that
camp--it is still necessary to realize that it has limits beyond which unsatisfactory
conditions begin to gain momentum. At the pace the economy has been growing, we are
very likely getting close to those limits.
In the Humphrey-Hawkins forecast made at this time last year for 1998, we were
well off the mark. Everything turned out to be much better than we expected. Growth
was stronger, unemployment lower, yet inflation quiescent. The outcome was literally
wonderful. Can we reasonably expect a repeat? I doubt it. At the moment, working with
the available data, the risks appear to be decidedly on the upside. But I can envision the
possibility, perhaps the likelihood, that this could be reversed very quickly for any of a
combination of reasons arising from the domestic or international, real or financial
economy.
I continue to be uncertain as to which direction our next move may be or when we
should take it. But my sense is that the upside risks will have to begin to dissipate soon
or they are going to need to be addressed.

2/2-3/99

CHAIRMAN GREENSPAN. Thank you. We are running a little behind
schedule but not by much. We will adjourn until 9:00 a.m. tomorrow.
[Meeting recessed]

2/2-3/99

February 3, 1999 -- Morning Session

CHAIRMAN GREENSPAN. We will turn to Tom Simpson for a presentation on
the Humphrey-Hawkins ranges for 1999.
MR. SIMPSON. Thank you, Mr. Chairman.
The Committee must decide whether to adopt the provisional
money and debt ranges it chose last July or to modify them. In the
Bluebook, we did not present alternatives to the provisional ranges
because of the Committee’s previous skepticism about the reliability of the
relationship between these aggregates and economic performance and
hence their usefulness as guides to policy. Instead, the Committee has for
some time chosen ranges for money that it viewed to be benchmarks under
conditions of price stability and of velocity behavior that conforms to
typical historical experience. The Committee has not found that this
practice has impeded its ability to extract and use information from the
monetary aggregates as an input to its decisionmaking. In the case of debt,
the Committee has chosen ranges based on expected growth--not the price
stability benchmark--but the inconsistency between the money and debt
ranges has not been a problem. Today, I would like to review the staff’s
projections and to discuss recent experience with money as an economic
indicator.
Your first exhibit presents the provisional ranges and the staff’s
projections of money and debt growth for 1999 consistent with the
Greenbook economic forecast. 3/ Also shown are actual outcomes in
1998. The staff foresees growth in both M2 and M3 slowing this year, but
remaining rapid with respect to their provisional price stability ranges and
with respect to growth in nominal GDP. Debt of domestic nonfinancial
sectors also is projected to decelerate this year but to finish the year
around the 5 percent midpoint of its provisional range.
We believe that the behavior of money market mutual funds is an
important element in understanding recent and projected declines in
monetary velocity. As shown in the lower panel of the exhibit, money
funds in both M2 and M3 have grown at double-digit rates in recent years,
and both types of funds registered a pronounced acceleration last year. Of
course, if growth in these components came solely at the expense of the
other components in M2 and M3, their strength would not have
implications for overall growth in money. However, we are of the view

/ A copy of the material used by Mr. Simpson is appended to the transcript. (Appendix 3)

2/2-3/99

that much of the expansion in money funds in recent years reflects other
influences that have had the effect of boosting growth in M2 and M3.
At the M2 level, we think that money funds may have been
benefiting from efforts by households to diversify portfolios swollen by
the surge in equity values. Illustrative of this process is the top panel of
your next exhibit showing shares of mutual fund assets. Even with the
hefty double-digit increases in money funds, the money fund share of total
mutual fund assets had been drifting down until the middle of last year. It
turned up in the third quarter amid the market turmoil and heightened
demand for liquid and less risky assets and then turned back down in the
fourth quarter. Looking ahead, in an environment of essentially stagnant
equity prices, as assumed in the staff forecast, growth of total assets will
diminish. But investors are expected to view money fund returns as more
attractive in relative terms than they have in the recent past and to seek to
rebuild somewhat the money fund portfolio share. In short, we foresee
money funds continuing to grow fairly rapidly, lifting M2 growth again
this year.
As a consequence, M2 velocity should decline further this year even
though stable short-term interest rates imply little change in opportunity
costs, as shown in the bottom panel of the chart. The 2 percent expected
drop in velocity and the staff’s forecast of 4 percent growth in nominal
GDP imply the 6 percent forecast for M2 in 1999.
Turning to M3 and Exhibit 3, money market mutual funds are
believed to have contributed to growth in this aggregate beyond their
impact on M2. In particular, those funds in M3 only, so-called
institution-only funds, have become an ever-popular cash management
instrument, substituting on business balance sheets for direct holdings of
money market assets outside M3, such as Treasury bills, as well as other
balances in M3. Firms can outsource this function to money funds and
thereby dispense with in-house time and effort required for the frequent
placing of liquid balances in the market. We believe that this trend will
continue for the foreseeable future, boosting M3; however, with no further
declines in money market interest rates, and thus no appreciable spreads
favoring money funds reemerging, money funds in M3 only should
decelerate from last year.
Another factor that boosted M3 growth last year was a surge in
bank credit and associated funding needs, shown in the bar chart in the
middle of Exhibit 3. This was caused in part by the disruptions to
financial markets beginning late in the summer that led some businesses to
tap bank lines instead of market sources and banks to hold rather than
securitize loans and to acquire securities having unusually large spreads.
In the less turbulent market setting anticipated for the current year, we are

2/2-3/99

forecasting that growth in bank credit will slow some, contributing to a
slowing in growth in overall depository credit, the bottom panel. As a
result, growth in depository credit should move closer to that of total debt
of domestic nonfinancial sectors. With the distribution of funding between
M3 and other sources similar to last year, we are led to a still considerable
8 percent expansion in M3 this year, implying more than a 3½ percent
drop in its velocity, which is smaller than in 1998.
The anticipated slowing of debt growth this year shown in the chart-
to 5¼ percent--is accounted for both by a larger run-off of federal debt,
owing to a $100 billion projected fiscal surplus, and to some moderation
in business and household borrowing. Nonetheless, growth in total debt
again exceeds that of nominal GDP by about the same margin as in 1998,
as spending remains tilted toward credit-intensive consumer and producer
durable goods and housing and as debt-financed merger activity stays
brisk.
Last year, both money and income growth exceeded staff
expectations, resurfacing the question of whether the monetary aggregates
may have become more reliable as indicators of economic performance.
As I noted earlier, some of the unusually strong growth in M2 and M3 last
year reflected not only income growth but also outsized declines in their
velocities, much of which had not been anticipated. Still, the surprises in
money and income were large and correlated, however loosely. Certainly,
the rapid money growth in the fall was suggesting that the banking system
was able to intermediate credit without serious strains.
To examine the question of whether the monetary and debt aggregates
may have become more useful as indicators, I have presented some
evidence on Exhibit 4--evidence that should be regarded as illustrative and
not definitive of indicator properties. Each panel contains the coefficients
and t-statistics for sixteen-quarter rolling regressions that relate growth in
nominal GDP on a quarterly basis to growth in money or debt on a
contemporaneous and one-quarter-lagged basis. Large, and positive,
values for the coefficients and t-statistics imply value of the aggregate as
an indicator of quarterly growth in nominal GDP growth. As shown in the
top panel for M2, the last time the t-statistics approached the important
level of 2 was in the mid-1980s. In recent years, coefficients have been
very small and statistically insignificant at standard levels, suggesting by
this metric that M2 has had little value as an indicator of nominal GDP.
Moreover, the story doesn’t change much for M3 and debt, shown in
the middle and bottom panels, respectively. It is worth noting that these
are quite simple statistical exercises and there may be times when
surprises in the monetary aggregates, allowing for special factors that may
be affecting velocity, are giving off clearer signals that the economy is

2/2-3/99

departing from expectations, especially when those surprises are
corroborated by other indicators. Thus, at this point we are hard pressed
to suggest that you treat the broad monetary or debt aggregates any
differently as indicators than you have in recent years.
CHAIRMAN GREENSPAN. Questions for Tom?
MR. GRAMLICH. In Exhibit 4, what does a negative t-statistic mean? Does that
mean that there’s a negative sign?
MR. SIMPSON. The coefficient has turned negative.
CHAIRMAN GREENSPAN. Tom, is there any evidence to suggest that the big
surge in velocity that occurred, contrary to expectations, from opportunity costs in the
early 1990s may be in the process of reversing itself?
MR. SIMPSON. That’s a possibility, but it is rather hard to come up with the
economic intuition as to why that might be happening.
CHAIRMAN GREENSPAN. Any more so than was the case with respect to the
economic intuition in 1990? What happened subsequent to 1990 was far more of a
surprise than a reversal would be today, if I may put it that way.
MR. KOHN. I think that by 1991 we were beginning to get some sense of what
was going on--specifically, that banks were getting into the mutual fund business.
Mutual funds in particular were much more freely available to households, and
households were diversifying their portfolios out of deposits. That was encouraged at the
time by a steeply upward sloping yield curve which, to be sure, is reversing. So there
were a lot of purchases-
CHAIRMAN GREENSPAN. We knew about that but we didn’t forecast a
change in the gap between opportunity costs and M2 velocity as a consequence, as I
recall.
MR. KOHN. If you look back at our forecasting in 1990 and 1991, that
development caught us somewhat by surprise. But toward the end of 1991 and in 1992
we were doing a better job of forecasting.
CHAIRMAN GREENSPAN. You were putting in add factors.
MR. KOHN. Yes, because we knew that the world had changed, so the old
equations weren’t working. Exactly.

2/2-3/99

CHAIRMAN GREENSPAN. I must say that I have not changed my view that
inflation is fundamentally a monetary phenomenon. But I am becoming far more
skeptical that we can define a proxy that actually captures what money is, either in terms
of transaction balances or those elements in the economic decisionmaking process which
represent money. We are struggling here. I think we have to be careful not to assume by
definition that M1, M2, or M3 or anything is money. They are all proxies for the
underlying conceptual variable that we all employ in our generic evaluation of the impact
of money on the economy. Now, what this suggests to me is that money is hiding itself
very well.
Don, do you think we ought to discuss the Humphrey-Hawkins issue before we go
into this or after?
MR. KOHN. It’s up to you. You need to do two things here. One is to decide
whether to readopt the ranges you had on a provisional basis. If people have views on
that, maybe while they are giving those views they can also comment on-
CHAIRMAN GREENSPAN. Let me suggest the following: If we are all of the
same view as we were the last time--to stay essentially with the noninflationary or price
stability ranges for M2 and M3--maybe we can get that out of the way and then discuss
what to do with debt. If it turns out after we go around the table that we have significant
differences of view, it may be desirable to discuss both issues together. That way we can
have a single conversation covering not only what we want to do today but what, if
anything, we want to recommend to the Congress with regard to potential revisions in our
reporting requirements under the Humphrey-Hawkins Act. So, I would appreciate
getting a quick sense from everybody as to whether they would like to stay with the
preliminary M2 and M3 targets. We will learn very quickly whether or not to go on to
the next discussion. The simplest way to do that is to start with you, Governor Rivlin.
MS. RIVLIN. I would stay with the preliminary ranges because I don’t feel that I
have any basis for a new set of targets. [Secretary’s note: In the subsequent go-around,
all Board members and Reserve Bank Presidents indicated that they concurred with Ms.
Rivlin’s statement.]
CHAIRMAN GREENSPAN. On the debt range, we have a little inconsistency.
What would be your recommendation on what to use for debt?

2/2-3/99

MR. KOHN. I think you’ve done very well with what you have been doing. No
one has really noticed this inconsistency. [Laughter] Seven months ago I gave up on
this!
CHAIRMAN GREENSPAN. What specifically are we using for debt?
MR. KOHN. The preliminary range was 3 to 7 percent, and the projection is that
growth in debt will be right in the middle of that range.
CHAIRMAN GREENSPAN. Does anybody disagree with adding the debt range
to the vote on the M2 and M3 ranges? Hearing no objections, I suggest we vote on the
three of them.
MR. BERNARD. Do you want me to read the directive language?
CHAIRMAN GREENSPAN. Yes, please.
MR. BERNARD. The language is shown in the Bluebook on page 23 under the
heading “1999 ranges”: “The Federal Open Market Committee seeks monetary and
financial conditions that will foster price stability and promote sustainable growth in
output. In furtherance of these objectives, the Committee at this meeting established
ranges for growth of M2 and M3 of 1 to 5 percent and 2 to 6 percent respectively,
measured from the fourth quarter of 1998 to the fourth quarter of 1999. The range for
growth of total domestic nonfinancial debt was set at 3 to 7 percent for the year. The
behavior of the monetary aggregates will continue to be evaluated in the light of progress
toward price level stability, movements in their velocities, and developments in the
economy and financial markets.”
CHAIRMAN GREENSPAN. Call the roll.
MR. BERNARD.
Chairman Greenspan
Vice Chairman McDonough
President Boehne
Governor Ferguson
Governor Gramlich
Governor Kelley
President McTeer
Governor Meyer
President Moskow
Governor Rivlin
President Stern

Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes

2/2-3/99

CHAIRMAN GREENSPAN. Let me take a minute to explain this issue of the
Humphrey-Hawkins reauthorization. Unbeknownst to 104 percent of the world,
[laughter] a piece of legislation went through the Congress a few years ago which
effectively sunset virtually every report required to be issued by various governmental
agencies. One reason it happened that way was because the legislation said that all
reports listed in some obscure source would be “included under this Act,” and that list
included absolutely everything. Nobody here caught it except Don Winn, who does that
sort of thing for a living. It came as a great puzzlement to everybody. So it turns out that
under law the reporting requirements in the Humphrey-Hawkins Act--not the Act itself
but the reporting aspect of it, I gather, Don--will expire at year-end.
MR. WINN. It is the reporting requirement that expires, not the goals of the
Federal Reserve Act. This doesn’t have anything to do with the goals of maximum
employment, price stability-
CHAIRMAN GREENSPAN. I understand. Do you know offhand if our
testimony is included in the reporting requirements? How is that stipulated?
MR. WINN. The way the Federal Reserve Act reads, we are supposed to submit
this report on a semi-annual basis and then we are to consult with the Congress on the
report. The “consulting with the Congress” has meant the testimony. If there is no
report, it is hard to see how there is any obligation to testify on anything.
CHAIRMAN GREENSPAN. So, effectively, one can read the sunsetting as
applying to both the report and the testimony?
MR. WINN. That is my understanding.
CHAIRMAN GREENSPAN. Clearly, when that information rolled its way onto
somebody’s agenda, it created a few sparks. The general presumption was that there
would be some automatic rollover or reauthorization of the Humphrey-Hawkins reporting
requirements and that would be that. Someone asked the newly ensconced Chairman of
the Senate Banking Committee, Phil Gramm, whether in fact that would be the case and
when he was going to put reauthorization on the Committee’s agenda. He said that it
wasn’t clear to him that such legislation was needed, and that response opened up a host
of repercussions. Within hours, Chairman Leach of the House Banking Committee
indicated that he was adamantly opposed to any change. We are now in the position

2/2-3/99

where, as best I can judge, Phil Gramm isn’t saying he is going to fight this to the death
but just that he thinks we ought to discuss it. It is clear, however, that it’s an issue about
which he doesn’t feel all that strongly.
But there is a secondary issue here, which is not irrelevant to the consideration of
this reauthorization. And that is that we do have an opportunity, if we so choose, to offer
recommendations on changes in the nature of what we report and how we report it. So,
rather than just having an extension of the existing legislation with no alteration, it is
quite conceivable that recommendations from the Federal Open Market Committee
would have some influence on the Congress in terms of how the reauthorization would go
forward.
Since this is not anything urgent, there is no need to come to any conclusions
today. But if you have some suggestions that you would like to raise today, that would
be useful. And if you have a suggestion at a later time, communicate with Don Kohn. In
the process, when the time for the actual reauthorization rolls around and we are
requested to give our views--indeed, it is conceivable that I will be asked at the
Humphrey-Hawkins hearings on February 23-MS. RIVLIN. I think it is inconceivable that you will not be asked.
CHAIRMAN GREENSPAN. I guess that’s right. So, it would be helpful to hear
from you on any change that you think might be useful. Obviously, the sooner you bring
it up, the better, but it need not be today. Having said that, if anyone has any thoughts at
the moment triggered by the memo that was sent to you earlier, it might be useful to get
them on the table so that we can get a sense of the Committee’s tentative views on this.
President Boehne.
MR. BOEHNE. I’d just like to comment on the broad issue rather than make
specific suggestions for change. One of the criticisms traditionally leveled at the central
bank is that we are a peculiar institution in a democracy in that we are not accountable
and somehow are out there on our own. While the Humphrey-Hawkins reporting has not
eliminated that criticism, it does seem to me that it goes a long way toward saying that
indeed we are accountable. It gives us an opportunity to explain what we are doing and
to consult with the Congress. So, I think Humphrey-Hawkins has turned out to be very

2/2-3/99

useful, not only for us as a central bank but also because it probably has enhanced
understanding to some extent on Capitol Hill.
In terms of the broad issue, I think we ought to come down foursquare in support
of a continuation of the basic nature of the Humphrey-Hawkins report and the testimony.
That said, there probably are some modifications that might be put on the table that could
improve it or bring it more into line with some of the things we actually do. I’d like to
give some further thought to what those modifications might be specifically. But on the
broad question of its reauthorization, I think we ought to be for continuing it.
CHAIRMAN GREENSPAN. Yes, I think the few people I have talked to on this
subject would fully agree with you on that. If anything, we ought to over-emphasize that
particular point because that is the crucial part of the Humphrey-Hawkins Act, even if
there are elements within it, which I suspect there are, that are just fillers and do not
represent any useful communication from the Federal Open Market Committee to the
Congress. Those elements really ought to be eliminated; they are a waste of time for a lot
of people here in Washington and throughout the Federal Reserve. President Minehan.
MS. MINEHAN. I am totally in agreement with what Ed Boehne has said. I
think there is a real downside risk if we do not report on a frequent basis to the Congress.
We do have our autonomy to be concerned about. If we are not accountable, that can be
taken away. We have to demonstrate that we think about the right things and that our
actions are well motivated. And the Humphrey-Hawkins report is one way of doing that.
I have one question, though. I’m not all that familiar and I don’t know whether
my staff is all that familiar with this “filler”--to use your language--that is required by the
Humphrey-Hawkins Act. That might not be all that obvious to us. Is it just the estimates
on GNP, inflation, and unemployment that we prepare or are there a lot of other things
the staff has to put in the report?
MR. KOHN. Brian Madigan has copies of Section 2A of the Federal Reserve
Act, which covers that. Why don’t you pass them out, Brian.
CHAIRMAN GREENSPAN. I think a lot of it relates to our interpretation, as
distinct from what is statutorily required.
MR. KOHN. Right. Obviously, the monetary aggregates ranges are in there, but
so is the requirement to take account of past and prospective developments in

2/2-3/99

employment, unemployment, production, investment, real income, productivity,
international trade, payments, and prices. And there is the point about discussing the
relationship of all that with the goals in the President’s Economic Report.
MS. MINEHAN. So we have to weigh that Report, which we get after it is given
to Congress. And as far as the staff is concerned, almost all of that is required as opposed
to things we have added?
MR. KOHN. To a considerable extent we feel they are required. The Act has
been amended over the years to include additional emphasis on international
developments. In response to that, we have beefed up the international sections.
MS. MINEHAN. Obviously, our treatment of a lot of things included in that list
probably could be modified in some way or another.
CHAIRMAN GREENSPAN. That is the type of recommendation that would be
helpful for me to get.
MS. MINEHAN. I figured that might be what you had in mind.
CHAIRMAN GREENSPAN. I have my own ideas, but it would be very helpful
to me to get a good sense of the Committee’s thoughts, as distinct from my own, on any
specific issue. I might be unaware of strong support for including certain things that I
thought people might not want to include. So I would like to get a sense of the
Committee’s views.
MS. MINEHAN. I can’t comment on all the details, but there is one thing I’m
sure we ought to do. We ought to convey the view that while there may be real
information in the monetary aggregates conceptually, we believe the specific ranges
suggest a precision about our knowledge of the relationship of those ranges to economic
performance in the upcoming years that is somewhat lacking. We have told Congress
that. I think some of the detailed information we are required to produce on that may not
be serving us well and may not be conveying anything about what we really think is
going on.
CHAIRMAN GREENSPAN. President Broaddus.
MR. BROADDUS. I don’t have a lot of detailed suggestions, Mr. Chairman.
More generally, a key point to me about the Humphrey-Hawkins reports, which we do
twice a year, is that they are very useful both internally and externally from the

2/2-3/99

standpoint of developing a strategy. Naturally, at our FOMC meetings we are focused on
current developments and the immediate policy issues; this semi-annual reporting
requirement gives us a chance to stand back and think a bit more strategically. I think
that helps us internally and it helps us in terms of communicating to the public by giving
them a background against which to interpret our short-term actions and statements. I
think that is a key value of this reporting requirement.
In terms of the changes, I agree with what has been said about the ranges for the
monetary aggregates. It seems to me that, in terms of communicating our strategy, we
need to supplement those numbers with something. As I have said in these meetings a
number of times before, my own feeling is that something like inflation targets would be
a particularly promising direction in which to move. We now have some experience with
other central banks doing this. I would hope, now that the issue of the report is going to
come up in any event, that one of the things we might consider seriously is adopting an
inflation target of some sort.
I would suggest one other change. One of the key parts in the report is the
summary of the individual projections we provide, which are encapsulated in the central
tendency projections in the report. To clarify those, it would be helpful if we all made
them contingent on a uniform assumption of no change in policy. That would in a sense
put us all on the same page and eliminate any ambiguity or confusion or lack of clarity
that might result from the fact that different members of the Committee may be using
different policy assumptions in generating their own individual forecasts. It seems to me
one of the advantages of that would be that it would help to signal the undesirable
consequences of not taking policy actions in one direction or the other that we need to
take. It might put us in a position to be more proactive in presenting our policy stance
and perhaps less defensive in public discussions of monetary policy. I believe that is the
procedure the Bank of England uses in its current reporting approach. That is another
suggestion I think we ought to look at.
CHAIRMAN GREENSPAN. President Parry.
MR. PARRY. I agree with much of what has been said. Just to add to a point
that Cathy Minehan was making: This does seem to present us, as a Committee, with an
opportunity to discuss in the testimony what we think the role of the aggregates should

2/2-3/99

be. We have spent a great deal of time over the years talking about this issue, and we end
up giving the targets more prominence than most of us think is warranted. This may be
an opportunity to adopt a procedure that would more closely reflect what we in fact do
and what we believe in terms of the weight these statistics versus others should be given.
CHAIRMAN GREENSPAN. Vice Chair.
VICE CHAIRMAN MCDONOUGH. Mr. Chairman, it seems to me that
everybody here is in agreement that the Humphrey-Hawkins testimony is a great
opportunity and should be continued. I have a little difficulty, just procedurally, hearing
people say they like this, that, or the other thing because that results in a disorderly
debate. People make suggestions, some of which I think are grand and some of which I
think are quite bad. If people gave their ideas to Don Kohn, as you suggested, then at the
next meeting we could have an orderly discussion of the ideas that have been offered
rather than have somebody announce an idea and others by their silence give the
impression that they agree with it. In some cases I agree and in some cases I don’t. But
we could spend the rest of the week here if we debated every idea brought forward.
CHAIRMAN GREENSPAN. The difficulty is that the February 23 testimony is
before our next meeting. One possibility is for Don Kohn to send around a questionnaire
and list the various suggestions that would involve changes to the statute. Changes on
things not mandated by statute we can make any time we want.
VICE CHAIRMAN MCDONOUGH. I was of the impression that at the
testimony on February 23 you would be asked what you thought and you would give a
generic answer that the testimony is appropriate and that we will be thinking of
suggestions on the content of the reauthorizing legislation.
CHAIRMAN GREENSPAN. Yes, I could do that. In fact, that is what I would
plan to do. What do you think, Don?
MR. KOHN. I think this could be a two-stage process. That is, you could make it
very clear on February 23 that the FOMC strongly supports the continuation of a
reporting regime of some sort embedded in law. And you could say that we would be
more than happy to work with the Congress over time on designing a reporting regime
that meets their needs and that we think is reasonable. So, I don’t know that we need to
reach a conclusion before February 23 about what we want in the legislation except that

2/2-3/99

we want the regime to continue and that we will be more than happy to work with
Congress on how that process will go.
CHAIRMAN GREENSPAN. That strikes me as a quite reasonable approach.
MS. MINEHAN. Yes.
CHAIRMAN GREENSPAN. Why don’t we leave it there, then?
MS. RIVLIN. May I raise a technical point as a word of caution? If we start to
indicate what we would like future Humphrey-Hawkins reports to include, in terms of
what we would want specified in the law, we will not get the last word. They will. Every
member of the Banking Committee is going to have an idea about what ought to be in the
law, and the easiest thing for the Chairman of that committee to do is to put them all in. I
think we should not encourage a rewriting of the law, even though I personally would
like to get rid of the monetary ranges, because once we start down that road we might end
up a lot worse off than we are now.
CHAIRMAN GREENSPAN. That is indeed a thoughtful comment.
MR. BOEHNE. That’s a good point.
CHAIRMAN GREENSPAN. The more I think about it, the more it strikes me as
being subtly obvious.
MR. BOEHNE. Yes, I agree.
CHAIRMAN GREENSPAN. We may actually be better off with the same
legislation.
MR. STERN. We can make some changes without changing the Act.
CHAIRMAN GREENSPAN. We can make a lot of changes ourselves because
the statute itself is not all the explicit.
MR. BOEHNE. I think Alice has spoken wisely; I think she’s right.
CHAIRMAN GREENSPAN. You earned your pay today! President Moskow.
MR. MOSKOW. I think Alice’s caution is very well stated. In looking at this
though--and this is really the first time I have looked at it carefully--I assume this is the
entire reporting requirement.
CHAIRMAN GREENSPAN. That is the Act.
MR. MOSKOW. The section is entitled “Monetary and Credit Aggregates.” All
the reporting is under monetary and credit aggregates, which is rather ironic given the

2/2-3/99

discussion we just had about the limitations of the aggregates themselves. I recognize
that there is some risk in trying to come up with changes to this, but it really is
misleading to the American people to have the reporting requirements in a section
entitled “Monetary and Credit Aggregates.” I would hope that we could do something
better than that in terms of how we are going to communicate.
MS. RIVLIN. If none of us has read it, you can rest assured very few other
people have. [Laughter]
MR. KOHN. I’m not sure whether that title is something we put on the document
or-MR. MADIGAN. Yes, Virgil Mattingly says it is.
MR. KOHN. It is our title, not what is in the law. So the next time we reprint the
Federal Reserve Act, we can change the title.
CHAIRMAN GREENSPAN. Can we legally do that?
MR. MATTINGLY. Yes.
MR. PRELL. It is just an editorial-
CHAIRMAN GREENSPAN. That is not in the literal Act itself? I am learning
things that I did not know.
SPEAKER(?). Let’s do it tomorrow. Why wait?
CHAIRMAN GREENSPAN. I am learning all these things, though I’m not sure
of what conceivable value they are. [Laughter] Why don’t we leave this issue for now. I
think Alice has raised a more fundamental question. Let’s think about it for a while and
then Don Kohn can poll us before the hearing to get a sense, at least, of how we would
like to answer the question at the hearing. And then we can go on from there. Is that
satisfactory to everyone? Let’s do it that way. Let’s move on then to Don Kohn.
MR. KOHN. Thank you, Mr. Chairman.
I will begin my briefing today with a discussion of the simulations
presented in the Bluebook, and I will be referencing the charts in the
Bluebook in the process. These exercises were designed to shed light on
the economic and policy environment embedded in the staff’s Greenbook
forecast, and they may have implications for the Committee’s policy
strategy.
The baseline scenario, shown on Chart 3 following page 8, extends
the Greenbook forecast. It was designed to be consistent with the

2/2-3/99

underlying logic and economic trends of that forecast, and hence preview
what could be in store after 2000. The staff’s assessment is that the
economy is now in disequilibrium in that the unemployment rate is below
its sustainable level; moreover conditions are not in place to correct this
situation. Real interest rates are lower than their natural levels and fall
further over the next two years as the nominal interest rate is held constant
while inflation and, by assumption, inflation expectations rise. Decreasing
real rates do not produce an even greater intensification of inflation
pressures because some of the economic strength of 1998 is seen as
transitory and because the equilibrium real rate is falling. Over the long
run, the wealth-to-income ratio declines as the stock market levels out,
government debt is paid down, and foreign indebtedness grows, raising
household saving and depressing the natural interest rate.
This disequilibrium implies that policy must firm at some point and
that the economy must experience subpar economic growth to limit
inflation. In the baseline, the policy response was assumed to be delayed
and the Committee satisfied with capping inflation at 2¾ percent. To hold
inflation down by more, as in the price stability scenario, the Committee
must begin to tighten sooner and be willing to continue to firm policy even
as the unemployment rate rises.
The disequilibrium in the baseline arises from a judgment that the
surprisingly favorable inflation performance of recent years has been the
product in part of transitory factors depressing prices and labor costs. The
lower NAIRU scenario in the second set of simulations, shown on Chart 4
after page 9, considers the consequences if instead the disinflation is
assumed to have been a result of lasting changes in the structure of labor
and product markets. If the NAIRU is as low as 4½ percent, the economy
is not in disequilibrium at present and nominal and real interest rates are
close to sustainable levels. Under these conditions, keeping the federal
funds rate at 4¾ percent over the next several years will be associated with
a slight uptick in inflation as oil and import prices turn around, but only to
a steady state rate of 2 percent.
As it happens, a 4½ percent NAIRU also would help to reconcile the
current stance of monetary policy with the results of Taylor-type rules.
Governor Gramlich noted at the last meeting and Governor Meyer
yesterday that the versions of this rule the staff calculates all tend to show
that the federal funds rate is too low. This undershoot results from the
existence of a large gap of actual over potential output, by standard
calculations. If the NAIRU is at the lower 4½ percent level, however, the
gap about disappears, and the current funds rate is more nearly consistent
with the Committee’s past pattern of reactions to actual and forecasted
levels of output and inflation and with Taylor’s rule. Of course, it is
possible to look at the results from both types of experiments from another

2/2-3/99

perspective: For the current level of the funds rate not to be too
accommodative over time, structural changes in labor and product markets
must be very substantial and persistent.
The surprises over recent years have been in aggregate demand as
well as aggregate supply, and the simulations shown on Chart 5 after page
10 look at the implications of situations in which aggregate demand
deviates from the baseline. We chose alternative paths for the demand for
producers’ durable equipment because that has been an important source
of unanticipated strength in aggregate demand in recent years and feeds
back on supply as well. Greater capital spending does raise the
productivity of labor and the level of potential output over time, but its
more significant effect in the short run is on demand. Thus, policy must
be appreciably firmer if demand surprises on the upside, even if it is
productivity-enhancing spending that constitutes the surprise. The
simulation follows the Taylor rule to tighter policy; perhaps the more
general point is that with labor markets already stretched, policy needs to
respond promptly to unexpected overshoots in demand to hold down the
rise in inflation.
Against this background, the decision at this meeting would seem to
rest in part on whether the Committee agrees with the basic message in the
staff forecast that existing and prospective pressures on resources and the
likely recovery in oil and import prices point toward higher inflation. If
the Committee views the risks as strongly skewed in this direction, it
might want to consider a firming of policy, if not at this meeting, then in
the near future--as might be indicated by a tilt toward firming in the
directive.
Growth has yet to slow from an unsustainable pace, and labor
resource utilization is higher than expected when you eased in November.
From a somewhat longer perspective, over the past year both the
unemployment rate and the federal funds rate have been reduced
significantly. Other things equal, policy interest rates and unemployment
rates ought to move in opposite directions, since the lower unemployment
rate generally raises inflation risks. Of course other things haven’t been
equal. For one thing, inflation and, to a lesser extent, inflation
expectations were falling in 1998, raising real short-term rates. But a good
portion of the decline in nominal short-term rates since last summer
probably has fed into real rates, offsetting much of the previous increase.
In addition, inflation expectations may be less likely to fall going forward
to produce such a passive tightening with the nominal funds rate
unchanged. Even if the staff has given too little weight to the possibility
that the economy is much less "inflation prone," transitory factors surely
have played some role in holding down prices. Absent additional serious
problems abroad that reduce commodity prices and put upward pressure

2/2-3/99

on the dollar, those factors will abate, giving underlying cost pressures
more of a chance to show through to prices and price expectations over
time.
Another thing that wasn’t equal last year was the financial market
disruptions that led to a tightening of credit conditions for a time.
However, the expansion of economic activity has been supported by
reasonably well functioning financial markets of late. Growth of money
has been strong, and credit has been readily available in markets as well as
at banks. While risk spreads remain high in some sectors of the markets,
they have stabilized for the most part; despite higher spreads, the cost of
credit for many borrowers has declined on balance since last summer,
owing in part to Federal Reserve easings. And increases in equity prices
have boosted wealth and reduced the cost of equity capital. Improved
conditions in credit markets have been signaled by their resilience to yearend pressures and to the difficulties of Brazil. Indeed, to the extent your
easing last November was undertaken to protect against the possibility of
major financial disruptions from these particular events, that protection
might seem to be less needed now.
Despite the ongoing risks of greater price pressures, however,
inflation has remained very well behaved. Moreover, threats persist to
U.S. economic performance from developments abroad. In these
circumstances, the Committee may be inclined to leave policy unchanged
today, and also to consider the odds on tightening in the near future remote
enough to justify retaining a symmetrical directive.
The benign inflation news, especially when coupled with a lack of
acceleration in nominal compensation, might suggest that the potential for
higher inflation has not yet been demonstrated. Uncertainties about the
supply side of the economy make preemptive policymaking especially
difficult. Although the Committee may not want to await the "smoking
gun" of a string of higher inflation numbers before it firms, it might want
more indicators that current levels of pressures on resources were raising
costs than it now has. And in this price environment, it may have the
scope to wait for confirmation that growth remains above potential before
leaning against tightening labor markets.
Moreover, our economic performance may be viewed as still subject
to downside risks from developments overseas. A highly unsettled
Brazilian situation has the continuing potential for financial contagion to
other countries beyond that assumed in the staff forecast. And industrial
economies aside from the United States appear to be weakening. In Japan
and Europe, the risks to growth prospects may be tilted to the downside
partly as a consequence of constraints on monetary policy from the zero

2/2-3/99

bound in Japan and from the desire to gain credibility and impose
discipline on fiscal policy in the Euro zone.
Financial markets looking at the information on the U.S. and foreign
economies have taken out their previous expectations of ease, but they
have not built in any odds of a tightening in the foreseeable future. With
inflation low and expected to remain contained for some time, and with
economic prospects around the globe still more likely to be dismal than
encouraging, the potential benefits of sitting tight may be seen as
outweighing the potential costs of risking a possible future buildup of
inflation pressures.
As a reminder, the implementation of the new announcement policy is
slated to begin in March, after it has been published in the Minutes and
explained by the Chairman in his testimony. Should the Committee
change the symmetry in the directive or otherwise feel that it has made a
significant change in its view of the risks, an announcement of that
without this preparation might provoke an especially sharp market
reaction and an expectation of imminent tightening. In any event, the
Committee will have an opportunity to give a nuanced view of its
assessment of the economic and policy outlook in the monetary policy
report and testimony on February 23.
CHAIRMAN GREENSPAN. Questions for Don? President Parry.
MR. PARRY. Don, I have a couple of technical questions and then a comment.
On page 10 of the Bluebook there is a reference that says “the real funds rate is now at its
natural rate.” I am familiar with what an equilibrium real rate is; I don’t know what a
“natural” funds rate is.
MR. KOHN. I was using it as a synonym for the equilibrium real rate, as in the
“natural rate” of unemployment--the same thing.
MR. PARRY. Turning to Chart 4, the top right-hand chart, I was a little surprised
at the difference in real funds rates in the simulations, particularly the one called “Lower
NAIRU.” Why would that produce analytically such a different real equilibrium rate?
The equilibrium rate has to be there somewhere. Do these converge in some year some
time out in the future at something different from 3 percent? I’m asking because I don’t
think of the equilibrium real funds rate as being 3 percent. Do you?
MR. KOHN. Let me address the first question first: Why is the rate with the
lower NAIRU lower than the other two?
MR. PARRY. Or why is the nominal rate so low in that simulation?

2/2-3/99

MR. KOHN. The reason is that the economy can produce at a substantially
higher level with a NAIRU that is one percentage point lower than assumed in the
baseline. As a consequence, interest rates need to be lower to stimulate the demand to
use that additional production. That is a level adjustment of production out into the
infinite future. Production can be higher by a couple of percentage points than at the
higher NAIRU, and a lower interest rate is needed to stimulate the demand to use the
extra production that is now available at the lower NAIRU.
MR. PRELL. Don, I might suggest that another way to think of this is as follows:
If the NAIRU is really more like 4½ percent, we might be around the NAIRU now and
looking at the prospect of a steady inflation path, which is the sign that we are essentially
around the natural rate of interest. That’s different from what is in our forecast, which
suggests that we are headed toward an acceleration of prices at the prevailing level of
interest rates. I think that’s a very simple way to look at that.
MR. PARRY. I see.
MR. KOHN. As for the level of the natural rate, it is higher than it has been in
some of the past forecasts, although as I noted it drifts down over time as the wealth-toincome ratio drifts down. I think the height is a result of the fact that the wealth-toincome ratio is a lot higher than we thought it was--or thought it was going to be a year or
two ago--given what has happened to the stock market. Demand has been much stronger.
In effect the experience, in terms of the level of the wealth-to-income ratio and the
strength of demand at previous interest rates, has led us to think that the natural or
equilibrium rate is a lot higher than we used to believe and a lot higher perhaps than it has
been in history. If the strength of demand for producers’ durable equipment and so forth
persists, the saving rate, even if it is creeping up, is going to be lower than it has been
historically, and then the natural real rate will be high relative to history.
MR. PARRY. Interesting. I have a comment about the simulations. They
suggest that we are going to have to make decisions in the next year or so that indicate
whether or not we are serious about price stability. Secondly, if I can go back to one of
my favorite topics, which is opportunism--clearly locking in the gains to date--there are
not many simulations where we lock them in, are there?

2/2-3/99

MR. KOHN. Except for that price stability one, which does a little more than
lock them in, that’s right. But that is the underlying premise of the staff forecast that you
all talked about yesterday: That the economy is producing a bit beyond its potential.
How much beyond is unknown, but it’s somewhat beyond. In the staff forecast the
assumption is that potential is at a level of unemployment nearly a percentage point
higher than where it is right now and that ultimately, over time, that is going to show
through in inflation. If you believe that analysis, the simulations show you the
implications of that analysis and those assumptions. And then in fact, yes, nominal and
real interest rates will have to rise from here.
MR. PARRY. It is very troubling that PCE inflation is 2¾ percent in all of the
alternatives but one. In that one it is 2 percent.
MR. KOHN. Right.
MR. PARRY. Thank you.
CHAIRMAN GREENSPAN. The same simulations five years ago had a wholly
different level with the same model, as I recall.
MR. PARRY. That is true. You only know what you know at a given time.
CHAIRMAN GREENSPAN. Vice Chair.
VICE CHAIRMAN MCDONOUGH. Mr. Chairman, during my years in
Chicago, I know I chatted with Milton Friedman about the natural rate of unemployment,
but I have managed to spend about 58 years of my life without ever hearing that the
NAIRU was confusing economists by its existence. So, taking advantage of not being a
Ph.D. economist, let me tell you what I see in comparing Charts 3 and 4, or at least give a
possible interpretation.
Chart 3 says that this is the world that used to exist and, based on the last couple
of years of experience, probably does not exist any more. But central bankers are
supposed to be hardboiled fighters against inflation, so let’s make believe that the world
is the way it used to be and launch ourselves into a fight against inflation.
Chart 4 says that this is the world that we in fact have been living in rather
successfully for the last couple of years. It doesn’t appear to me that we have to make an
act of faith that it will continue forever in order to think that this world, which produces a
much better climate for our citizens, could be allowed to continue. We may get some

2/2-3/99

evidence that in fact this is not the real world--that it has involved a series of lucky breaks
and that the real world is going back to something more like what is depicted in Chart 3.
If we do get such evidence, we would not want to be in a position of believing that Chart
4 will last forever because that would lead us to make mistakes that would be quite
dangerous if in fact the world of Chart 3 were the real world. But based on what we have
been seeing, even though we do not fully understand it, the world of Chart 4 seems to me
closer to the world we now appear to be living in. And I’m not sure why we shouldn’t
give that world a chance to endure. Am I missing something?
MR. KOHN. I wondered whether that was a question or not! [Laughter]
VICE CHAIRMAN MCDONOUGH. I had to put in a question mark at the end
because this is the time for questions. My question is: Would the gentleman not agree?
[Laughter]
MR. KOHN. The gentleman would agree. A key element in interpreting what
has happened over the last few years is this: To what extent has there been a fundamental
shift in the structure of labor and product markets that will persist? And to what extent
do the favorable results of the last few years stem at least partly from factors that are not
fundamental to labor and product markets such as the very sharp appreciation of the
dollar through the middle of 1998 that lowered import prices, the collapse of economies
abroad, the collapse of commodity prices and similar developments? These factors not
only lowered inflation directly but helped keep inflation expectations down.
In effect, the staff has chosen to assume it was a little of each. We used to have,
not that long ago, a NAIRU of 6 percent or perhaps a little more than 6 percent. We have
interpreted the good inflation numbers as reflecting in part a lowering of the NAIRU by
½ or ¾ percentage point, depending on what day we talk to people about it, and in part as
the product of some of these other factors. But I totally agree that it is in large measure a
judgment call. The reason we showed Chart 4 was to indicate that if you had the other
interpretation--that the favorable inflation results reflected almost entirely a change in
structure and that the declining oil prices and rising dollar really were not important
factors--then you would come out with the lower NAIRU path as shown in Chart 4.
These two charts were an effort to give you a way of doing what you just did, which is to
compare across charts and make a judgment. But the key judgment relates to what

2/2-3/99

accounts for the good price performance of the last three years and what is the mix
between a change in structure and one-time price effects.
MR. PRELL. In a sense there is some link between what you just said and what
the Chairman said earlier. I may not have known exactly what charts the Chairman had
in mind from prior years’ simulations, but if you view what has happened as having
benefited from some supply-side shocks, basically those things have become engrained in
the inflation expectations. And as long as the Committee follows a policy that doesn’t
lead to excessive pressures in the market, it can consolidate that progress and continue
the lower path. But if these are transitory developments and you pretend or hope that
they are changes in the structure, then you could be faced with a very severe turnaround.
And what we could be looking at a year or two from now is a much-elevated set of
simulations of inflation paths going forward. It is a judgment call that each of you has to
make at this point as to just what has led to the positive surprises on inflation. It’s not at
all clear that it is entirely due to permanent structural changes. I think one can see clear
risks in the medical care market, for example, that the benefits from some structural
changes there may have run their course and that costs pressures are turning things
around in that area.
VICE CHAIRMAN MCDONOUGH. If I could make a comment: I think an
appropriate degree of skepticism is no doubt appropriate. [Laughter]
MR. PRELL. As always.
CHAIRMAN GREENSPAN. That is by definition a safe statement! President
Hoenig.
MR. HOENIG. Don, having looked at the analysis in the Bluebook and at the
various charts, including Chart 3, and having listened to your comments this morning, I
want to talk a little about something I mentioned yesterday. That is the possibility of
wanting to unwind the actions we took earlier. Last fall we were projecting growth in the
fourth quarter to be less than half the growth we apparently got, and we were expecting
even lower growth in the first quarter of this year. We took actions to ease partly because
of those projections and the financial turmoil that was occurring. We have moved into a
much more stable financial environment. You did mention the idea of perhaps
unwinding our easing moves. I assume for the sake of discussion that one reason we are

2/2-3/99

not looking more favorably at unwinding at this juncture is the continuing concern and
uncertainty about external factors. Is that why you touched on it but did not really
develop an alternative for unwinding our earlier actions?
MR. KOHN. I mentioned it as a possibility. That is, to the extent that you were
buying insurance against a hurricane and the hurricane season has passed and you don’t
expect another season, then you don’t need the insurance. But in part the decision
involves a judgment, in terms of the risks, as to whether the hurricane season really has
passed, if I can stretch the metaphor beyond its breaking point. [Laughter] Certainly
most of us were pleasantly surprised by how well the financial markets, not only in the
United States but in other Latin American countries, weathered the Brazilian crisis. But
the crisis is not over yet, so that is still a potential issue.
The other thing that has happened over the last couple of months is that we all
were probably once again surprised not only by the strength of the economy but by how
well behaved inflation has been excluding the tobacco price increases. In particular, the
ECI and nominal wage compensation have exhibited very good performances. That is
another factor you might think about going forward. We’ve not only had positive
surprises on the real side, but we have some pleasant surprises still coming in on the price
side to balance the other concerns.
MR. HOENIG. That’s a good point.
CHAIRMAN GREENSPAN. Governor Gramlich.
MR. GRAMLICH. I’d like to address the point that Bob Parry brought up on
whether we have locked in progress on inflation. Let me refer everybody to Chart 13 that
Dave Stockton showed yesterday. The bottom panel of that chart shows that if we take
out food and energy in the PCE and use the capacity utilization approach, which is in
effect a proxy for the lower NAIRU--or in Bill McDonough’s terms the issue of whether
the present situation can go on--in fact, we have locked it in. The only reason there is a
little uptick in Chart 4 is because of food and energy prices. It may be that we want even
more aggressive action to get to lower inflation. But if it is true that something like the
present regime can go on, which means either that capacity utilization could be a standard
or that NAIRU is at about the current level of the unemployment rate, then arguably we
have locked in our gains.

2/2-3/99

MR. PARRY. I don’t understand that.
MR. GRAMLICH. Well, in Chart 13 the inflation rate actually goes down.
MR. PARRY. What about its behavior in terms of the long-run simulation on
Chart 4 of the Bluebook?
MR. GRAMLICH. Chart 13 is a long-run simulation of PCE without food and
energy, and the inflation rate actually goes down in that simulation. Apparently the only
reason it goes up in Chart 4 is food and energy. So if we lock in the endogenous part of
prices, leaving aside food and energy, then I think we could arguably say that this regime
does lock in the progress on inflation.
MR. PARRY. But that’s not true in terms of the long-term simulation shown on
Chart 4, which goes beyond the year 2000.
MR. GRAMLICH. They give slightly different results.
MR. KOHN. If you thought that the capacity utilization equation was the more
accurate equation in terms of predicting PCE, then as you remarked yesterday, Governor
Gramlich, it would be very much like the lower NAIRU situation. It would indicate that
the structure of the economy had changed and that capacity utilization was a better
measure than the unemployment rate. Then you would be in the lower NAIRU world
rather than the unemployment rate world.
MR. PRELL. You don’t even need a structural change. If somehow, for reasons
we don’t understand, capacity utilization really is the true measure rather than the
unemployment rate, then that simulation reflects the fact that the level of economic
activity is actually below the natural rate of capacity utilization. There is slack in the
economy, putting downward pressure on inflation. If you were looking at the
unemployment rate and you thought the NAIRU was 4½ percent or a little below, it
would be a close call. We are in the neighborhood, but that means we would not have the
depressing effect that we would have by looking at capacity utilization. What we noted
yesterday is that, as an approximation, that approach may be telling us roughly the same
story and not that we are a point below somebody’s notion of the NAIRU at this juncture.
MR. GRAMLICH. So, for these purposes any differences are small, they are hard
to predict in the long run, and they may not concern us too much.

2/2-3/99

MR. PRELL. I would just add a note of caution on the basis of bitter experience,
which does not have to be repeated admittedly. When one studies the history of the past
few cycles, it is not unusual to see a tendency to reinterpret capacity constraints because
the inflation turnaround isn’t in sight. I think it is manifest in the current episode that
there have been extraordinary shocks in terms of world activity, exchange rate
movements, declining import prices, and structural changes in various markets. And
there may be still more structural changes on the horizon after the electricity deregulation
and so on. A lot of these developments are, in the abstract, in the nature of one-time
shocks, which we can’t depend on being repeated. But we can consolidate the gains we
have made if we are able to figure out where that level of sustainable resource utilization
is in this model.
CHAIRMAN GREENSPAN. Governor Rivlin.
MS. RIVLIN. This is not so much a question as a plea for perhaps a different
structuring of the problem next time around. I think the difference between the world
depicted in Charts 3 and 4 is very interesting, but it becomes dramatic only if one
assumes that we stay with the policy implied by one or the other for a very long time. I
don’t know whether we are in the world of Chart 3 or Chart 4, and asking me to decide is
silly. There is no way I can know that. What we have to focus on, given that we don’t
know and that the truth is likely somewhere in between, is what the costs are of making a
mistake and how to move back from one to the other if we do make a mistake. I don’t
know how you could help us think about that, but that is what I believe we need to be
thinking about for the next several meetings.
CHAIRMAN GREENSPAN. President McTeer.
MR. MCTEER. When I raised my hand, I didn’t know there would be so many
people talking after Bill McDonough. I basically wanted to associate myself with his
remarks. I would associate myself with Alice Rivlin’s comments, too. We now have had
three years during which economic growth has been faster than most people’s estimate of
potential and the unemployment rate has been below most people’s estimate of NAIRU.
Three years is a long time. So, we have a real economy that we can match up against
simulations, and I’m not sure why we wouldn’t use the real economy rather than some of
the simulations. I don’t know if you remember the comedian Richard Pryor, but if he

2/2-3/99

were here he would say something like: Who are you going to believe--me and my
simulations or your own lying eyes? [Laughter]
CHAIRMAN GREENSPAN. Governor Ferguson.
MR. FERGUSON. I am struck as I look at Chart 4 that we are being asked to
make an either/or decision. That is, we have to think either that the NAIRU has dropped
or that adverse price shocks are going to hit us. Is there no simulation that actually
suggests some of both--that the NAIRU may indeed be below 5.3 percent, and perhaps at
4½ percent, but also that these other special factors have been hitting us? It strikes me
that that does not necessarily end up being your baseline because your baseline has a
somewhat different set of assumptions in it. Even though I understand what Bob McTeer
had to say about believing your own lying eyes in some sense, is there a simulation that
does a little of both? I’m sure that is doable, but is there one that has been done?
MR. KOHN. We have not done that, though the baseline does do a little of that in
the sense that the NAIRU in the baseline is considerably lower than it would have been a
couple of years ago. That’s part of what the Chairman was talking about. But the
baseline does take some of the favorable price and output news as stemming from supply
shocks. We have not done an interpolation with an assumption, say, that the NAIRU is 5
percent. But we could easily do that. One can do that visually by looking between the
two charts. Then in fact the interest rate would be a little too low, but not a lot. The
Committee would have to tighten at some point, but the urgency of a move would
perhaps be less than indicated in the baseline. One can do an ocular regression and figure
out that we are somewhere in the middle of these two scenarios.
MR. FERGUSON. Okay.
MR. KOHN. I should note that we did the simulation showing the adverse price
shocks, the dotted lines in Chart 4, in part to illustrate the power of supply shocks. Those
are not very big supply shocks relative to the baseline assumptions. There is a $5
difference in the price of oil and a 1 percent faster growth in benefits costs, which
translates into only ¼ percent faster growth in compensation. Those supply shocks
noticeably raise the inflation rate. And with the Fed leaning against increasing inflation-
not in a really aggressive way but at least by raising the funds rate right away--we get a
higher unemployment rate and a higher inflation rate simultaneously with these adverse

2/2-3/99

100

shocks. Think about the flip side of that. We have had a lot of favorable shocks over the
last couple of years, and that has given us both lower unemployment and lower inflation
than we would have expected. The dotted line was put there in part to let the Committee
members think about that and about what might have been if we had not had the
favorable price shocks over the last few years. These price shocks are very powerful in
shaping economic performance.
MR. FERGUSON. Let me ask one other thing on the locking-in point. Ned was
going back to Chart 13, but doesn’t the bottom panel of Chart 4 get to the issue of the
long term? That strikes me as suggesting that even if one believes the NAIRU has
dropped to the 4½ percent you have assumed there, we will end up in the long run with a
PCE excluding food and energy of about 2 percent. I think that goes back to Bob Parry’s
question.
MR. PRELL.

If I understand the construction of this--please correct me if I’m

wrong, Don--basically this represents the staff’s baseline assumptions naturally extended,
at least for the near term. Those assumptions include a rebound in oil prices and the turn
of the dollar and so on. So in a sense this is already the lower NAIRU path. It adjusts
down the NAIRU, but it doesn’t remove that modest reversal of the favorable supply
shocks that we have experienced.
MR. KOHN. Including most importantly the depreciation of the dollar.
MR. FERGUSON. Okay, thank you.
CHAIRMAN GREENSPAN. President Minehan.
MS. MINEHAN. I am attracted somewhat to the construct that Alice talked about
of not really knowing where we are between Chart 3 and Chart 4, because I really don’t
know how much of the good news we have been hearing for the last couple of years is
temporary or permanent. But I am drawn to thinking about where we were last August or
even last July. At that time, with the fed funds rate 75 basis points higher than it is
currently, many of us were thinking--using the usual constructs and equations and so
forth and given the resource constraints--that we were facing a real risk of an inflationary
increase over the forecast horizon. Some of that, not a lot, was built into the Greenbook
forecast as well. At that point, with the fed funds rate 75 basis points higher than it is
now, I think we were still betting that we were somewhere between Chart 3 and Chart 4.

2/2-3/99

101

We thought we faced an adverse shock stemming from developments in financial markets
that we expected to have a big impact on the real economy. We moved the funds rate
lower, but the expected impact on the economy did not materialize. So where is our
betting now? Our betting, it seems to me, is much more weighted than it was last
summer toward the view that everything has changed. A lot of us were not comfortable
with that bet last summer. So why isn’t Tom Hoenig’s logic in favor of reversing a little
of the easing not good logic now as a tactical matter, if not as a worry about near-term
inflation? I would view that as a way of expressing where we think we really are
between Charts 3 and 4. Is that a question? I am trying to find a question! [Laughter]
MR. PRELL. One reaction, though, is that in a sense you are beginning to bring
Chart 5 into the picture.
MS. MINEHAN. Yes.
MR. PRELL. That raises the question of whether some of the demand-shock
surprises we have been experiencing will be sustained. My sense of the discussion
yesterday, as we were thinking about the coloration of the Humphrey-Hawkins report, is
that this group might see greater risks that we will have stronger demand, other things
equal, than is assumed in the current staff forecast. It does raise the question of whether
you want to blend in some of the flavor of Chart 5 in terms of the policy content.
CHAIRMAN GREENSPAN. President Boehne.
MR. BOEHNE. This is really a question, even though it might sound like a
comment! [Laughter] If we try to choose between the world in Chart 3 and the world in
Chart 4, I think we have to be somewhat agnostic and have a fair amount of humility.
But as long as we’re being humble and as long as we’re being somewhat agnostic, how
do we know there isn’t a Chart 5 that represents a different world? How do we know that
the NAIRU couldn’t be 4 or 3½ percent? If one goes back, for example, to the 1960s
there was a lot of talk that we could have price stability and unemployment rates of 3½
percent or certainly 4 percent. I think the term used was “3 to 4 percent.” I don’t know
whether that is realistic or not. I believe if any of us had been asked two or three years
ago if we thought we could have a 4½ percent unemployment rate and inflation still
falling, we would have said that was a pretty nutty idea. How do we know that the string
has run out? How do we know that we can’t have a still better world? The question is

2/2-3/99

102

whether we have looked at that possibility analytically. I ask because I think we do have
some historical experience where unemployment of 3 to 4 percent without rising inflation
was considered realistic.
MR. PRELL. That proved to be an error, in a sense. The economy got very
overheated partly because of the optimism about how low an unemployment rate could
be sustained without inflation. To go back to what your eyes have been perceiving, labor
markets clearly have been tight and wages have been accelerating. Pressure has been
manifest in the labor markets. Now, that hasn’t fed all the way through to prices,
depending on what measure one examines. If one looks at the core CPI adjusted for the
technical changes, that has been accelerating. The PCE measures and the GDP measures
look a whole lot better. The answer is not entirely clear. We may have had a
productivity surprise that helped. We could get another. I think the case for a
substantially lower level of unemployment that is sustainable without a buildup of
inflation is perhaps a bit wishful, though I can’t rule it out. But we on the staff resist that
notion pretty strongly at this point. I would not be as resistant to the notion that the
NAIRU might be below 5 percent or maybe significantly below 5 percent. It is hard to be
strongly opposed to that notion based on recent experience.
CHAIRMAN GREENSPAN. President Poole.
MR. POOLE. Chart 3 has a price stability simulation, which Chart 4 does not
have. I have what I think is a simple question. Roughly speaking, what would the
federal funds path be in Chart 4, assuming the NAIRU were lower but the same end
result--converging on ¾ percentage point less inflation per year than the baseline--is
desired? To get that in Chart 4, presumably we would have a federal funds rate
somewhere between the baseline and the dot-dash line. Is that correct?
MR. KOHN. Right. On Chart 3 there is a difference of about 2 percentage points
between the inflation rates, and you’re talking about a scenario that might result in an
inflation rate roughly one percentage point lower than the baseline. So I would take
about half the difference between the real funds rates of the two simulations on Chart 3
and load it into Chart 4 to get the inflation rate for that scenario.
MR. POOLE. So Chart 4 is built under the assumption that we are satisfied to
converge more or less on the existing rate of inflation. That addresses the question of

2/2-3/99

103

what we have to do if the NAIRU is lower, if we’re satisfied with where we are on the
inflation rate. But for those who have an objective of a lower rate of inflation, we’d need
to add that to Chart 4 in our discussion of the implications for policy of Chart 4.
MR. KOHN. Right. Looking back at Chart 3, you would have to tighten by
about ½ percentage point, let’s say, in the near term, but run with a substantially higher
real funds rate over the next couple of years, though not as high as in the price stability
case.
MR. PRELL. A simple rule-of-thumb exercise using our model is that a 100 basis
point higher funds rate implemented immediately and sustained would knock about ¼
percentage point or a little more off the inflation rate in the first year. The inflation rate
in the second year would be about ¾ percentage point lower. I suspect there would be
overshooting in terms of getting that inflation rate to flatten out.
MR. KOHN. A hundred basis points sounds too-MR. PRELL. I think that would be more than enough to get you there very
promptly, so Don’s response of perhaps ½ percentage point sounds about right.
MR. POOLE. At the beginning of this trajectory, where we are right now, even if
the NAIRU is as low as 4½ percent we need a higher funds rate to start to work the
inflation rate down or we are at risk of actually seeing the inflation rate rise. That is what
the baseline tells us even at the current unemployment rate.
MR. KOHN. Right, that’s the lower NAIRU scenario.
CHAIRMAN GREENSPAN. Anybody else? If not, let me get started and try to
focus on some of the same issues that many of you raised.
The Committee discussion that we had yesterday elicited general agreement
around the table that, with the exception of energy, agriculture, and some manufacturing
and mining industries, the economy has been exhibiting substantial, and what many
analysts regard as unsustainable, strength. The strength is especially evident in interestsensitive areas such as housing and light motor vehicles. That, of course, would suggest
that our monetary stance is too loose and unless economic growth slows quickly, the
failure to tighten will reaccelerate inflationary pressures. Many of you have noted that
the sequence of crises and decided weaknesses abroad clearly have had no broadly
negative impact on the rate of growth in the United States, at least not yet.

2/2-3/99

104

I have not heard it argued specifically, but our 75 basis point action last fall was
directed at countering a freezing-up of financial markets, which constituted a
demonstrable threat to the stability of our economy, and arguably we have largely
succeeded. It is true that one can still observe some residual impact of the liquidity
problems that we have experienced, with yields on junk bonds remaining significantly
above Treasuries and even obligations rated A and AA still running spreads against
Treasuries that we haven’t seen for a very long time. If it is correct that we have
succeeded, then one could argue that we ought to reverse at least part of our easing
moves. There has been a legitimate concern about stock prices that, as an inadvertent
side effect of our easing actions, have returned to record levels--a development that
clearly has augmented effective demand.
All of this makes a compelling case, indeed. There is little question in my mind
that, unless we see a pronounced slowing in the expansion of economic activity, some
form of preemptive action may be called for. I say “preemptive” because there is
remarkably little evidence that inflation pressures are building. There is none in the data,
and I heard very little commentary about price pressures in our round-table discussion
yesterday. Everyone was arguing that the pressures were there, but no one was saying, at
least that I heard, that anything was happening to prices as a consequence. Indeed, the
CPI change for December, if we take out the tobacco price increase, was slightly negative
for the total CPI and only slightly positive for the core CPI. The same is essentially true
for the PCE data. If this is inflation, something is wrong with our data system. The point
is that we are at the tail end of a series of years in which, by all our historic measures,
growth has been above trend. Price pressures should be mounting at this stage, but
instead they are going in the other direction. This involves, in my judgment, a major
issue that we need to understand before we move forward with a policy shift. The
discussion that has just been joined relates precisely to this issue.
I find the resumption in the fourth quarter of the steep decline in the number of
those seeking work to be the most compelling evidence of an unsustainable, and perhaps
an unstable, economic expansion. That statistic, of course, includes the unemployed plus
those not in the labor force who nonetheless say they would like a job if they could get
one. You may recall that through the first three quarters of last year we had a sharp

2/2-3/99

105

divergence between the still strong growth of payroll employment and the apparently
stagnant growth indicated by the household data. The surge in household employment
and the decline in unemployment in recent months have now fully closed the gap.
Our great dilemma is that although the labor market has tightened, and tightened
quite appreciably in a statistical sense, gains in compensation per hour have slowed! It is
not that they haven’t risen, but their growth has slowed. Doubtless, the ECI change for
the fourth quarter is biased downward. As I have indicated before and as a number of
you have argued, it is very likely that to some extent wage increases are being masked as
promotions in one form or another. But even if the ECI is biased in that regard,
compensation per hour is not. Preliminary data for compensation per hour in the fourth
quarter, at least for the nonfinancial corporate sector that I believe probably mirrors the
total, indicate an average annual growth rate of 3.7 percent. That is down from an
average of about 4 percent in a number of previous quarters. Indeed, for the nonfinancial
corporate sector, it was running 4.4 percent in the middle part of last year on a fourquarter-change basis. For the fourth quarter of last year, year-over-year, it was estimated
at 4 percent.
Moreover, despite our tightened labor markets, when we disaggregate the data,
there is only very weak evidence of significantly greater compensation gains in areas
where the unemployment rate is demonstrably below the national average. The national
average unemployment rate currently is 4.4 percent. Gary Stern was mentioning
yesterday that Minneapolis had a 1½ percent unemployment rate. Their manufacturing
wage increases over the past two years have been close to the national average. If we
look at other cities, Boston with an unemployment rate of 2.2 percent has had average
hourly earnings increases over the last two years only somewhat higher than the national
average. Indeed, the only two cities on my list with low unemployment rates that have
experienced wage increases significantly above the national average are Charlotte and
Richmond. What are you doing, right or wrong, Al? [Laughter] San Francisco has a
2½ percent unemployment rate, and its average hourly earnings growth over the last two
years is about average. Denver unemployment is 2.6 percent, and its average wage
increase is well below the national average, as indeed are the increases for Phoenix and
Dallas. So, there is very little evidence, granted all the qualifications we want to make

2/2-3/99

106

with regard to these data, that the NAIRU is alive and kicking. It may exist, but it
certainly is in hiding, no matter how we look at it. Using NAIRU in our structural
models is in effect like using a phantom. Chart 4 in the Bluebook is the real world; the
NAIRU got lost somewhere.
If we look at compensation gains, they are virtually fully matched by the
acceleration in productivity. Total unit costs over the four quarters of 1998 rose just 0.2
percent. Indeed, unit labor costs for nonfinancial corporations over the latest four
quarters, as estimated by Larry Slifman who never makes a mistake, increased only 0.2
percent. That estimate is comprised of compensation per hour of 4 percent, as I
mentioned before, and productivity growth of 3.8 percent. The productivity growth
figure for the fourth quarter is 4.8 percent, preceded by 4.7 percent in the third quarter.
Clearly, something is happening. As I said before, absent the cigarette price hike,
we cannot find inflation in either the CPI or the PCE index for December, and we surely
do not find it in the pipeline data anywhere in the system. I submit that interpreting these
results requires a fundamental reassessment of how we look at the world. I will take a
shot at this and try to describe what I think we know and what we don’t know.
How is it possible, first, for hourly compensation growth to be flat or falling in
an ever-tightening labor market? Let me begin by suggesting what does not explain it.
You may recall that two or three years ago I was arguing that fear of job obsolescence
was a major factor suppressing the nominal increase in compensation per hour. That
factor clearly has not gotten worse; if anything, it has eased. The International Survey
Research Company is the source of the data that I was quoting back in 1995 and 1996, as
you may remember. When workers were asked whether they frequently were concerned
about being laid off, 46 percent responded “yes” in 1995 and 1996 compared with figures
in the teens or in the twenties throughout the 1980s. The 46 percent number is now down
to 37 percent. Statistics on job leavers, another indicator I would use, likewise do not
indicate any significant change. So an increase in uncertainty and the fear of job loss
amongst workers cannot account for this extraordinary combination of low
unemployment and no acceleration in hourly compensation.
The evidence, anecdotal and otherwise, suggests that the explanation lies in
pressure coming from employers, who have apparently lost virtually all pricing power--

2/2-3/99

107

an issue that a number of you raised in our discussion yesterday. We saw quite similar
episodes during the long period of the gold standard, which produced price stability on
average prior to the 1930s, although obviously there was a lot of price volatility. During
that period, wage increases were limited by the exogenous price capping of the gold
standard.
The technologically driven process that is breaking down barriers to cross-border
trade today has apparently created an environment that simulates the old gold standard
forces. One way of looking at this is that in earlier decades when there may have been
excess capacity or excess potential in one part of the world, it didn’t matter because that
excess could not be moved to another part of the world. But in the current more
technologically advanced environment, as barriers come down we get an increase in
potential supply relative to total actual physical capacity in the world economic system.
And it is conceivable, but by no means provable, that globalization--the major force that
people are talking about--may be having an impact on the price level, and our price
measures may be reflecting that. It may also be, with regard to my previous discussion of
compensation gains, that the data are capturing that phenomenon, although the argument
I am making is global as distinct from a specific manufacturing issue.
The argument is basically that tradable goods prices are being significantly held
down by excess world capacity and that the arbitraging into the nontradable goods areas
that occurs within economies, largely through wages, is the reason why service price
inflation, which arguably has very little in the way of direct international globalization
components, also has been restrained appreciably. In the United States this process has
been augmented by a dramatic increase in the backlog of new technologies, which is an
issue we have discussed in the past. This really gets down to the question of whether the
synergies that have evolved over recent years have created a large pool of potential
capital investments that firms can dip into to obtain a rate of return in excess of the cost
of capital. We have seen considerable evidence of this in the sense that rates of return
everywhere seem to be moving up.
There has been a very interesting pickup since 1994 in the average rate of price
decline in the high-tech area of the economy. Through the early 1990s, the deflator for
computers, communications equipment, and other high-tech goods was going down at an

2/2-3/99

108

annual rate of about 4 percent. Starting in 1994, the rate of price decline fell off the
chart, and the most recent data suggest that high-tech prices are dropping at an annual
rate somewhere in the area of 17 to 18 percent. Thus, even though that sector’s share of
GDP is only a few percent, these price declines are having an appreciable influence on
the overall inflation rate.
What this implies is that we are getting a rapid increase in opportunities for
investment in new technology. It is overwhelming the expansion of demand, and the
acceleration in the downward adjustment of prices suggests that we have a very large
backlog of unexploited investments that, as they are implemented, are displacing labor
and effectuating a very significant increase in multifactor productivity. That in turn has
spilled over into labor productivity. Indeed, estimates produced by the staff’s
econometric model suggest that we have seen a fairly dramatic pickup in recent years in
multifactor productivity consistent with this process.
I don’t believe that transitory factors can explain the failure of models to forecast
successfully in recent years. I suspect that what we have here is a missing variable, if I
may put it that way. Certainly, judging from the slowing in the rate of PCE inflation,
supply generally appears to be overwhelming demand despite the evident continued
decline in unemployed job seekers. I might say parenthetically that the level of these job
seekers--that is, the sum of unemployed plus those not in the labor force but seeking a
job--is currently ten million; and it has been falling by almost one million annually. We
have no way to judge how far down that number can go before there occurs an inevitable
acceleration in nominal wage demands. Unless the laws of supply and demand have
been repealed, there has to be some level of labor market tightness at which nominal
wages begin to accelerate. The truth of the matter is that we don’t know where that is;
we only know that the number of job seekers has been reduced by almost a million in
each of the last several years, and nothing has happened. That is like my falling off a
building and saying when I reach the fifth floor on my way down that I am in great shape.
Something has to give, but we don’t know when. I think the presumption that we do
know is very difficult to maintain.
The Greenbook forecast of a renewal of inflation statistically rests in part on a
marked decline in the growth of multifactor productivity, and hence in the growth of

2/2-3/99

109

output per hour. Indeed, without a slowing in multifactor productivity growth, we
obviously could not have a deceleration in labor productivity growth unless we
significantly lowered the forecast of capital deepening or labor quality, and that is very
unlikely to be the case. So what we have is a projected slowdown in multifactor
productivity growth, which is based in turn on a substantial slowing in the pace of overall
economic growth. I am not saying that the internal workings of the model actually have
that linkage; they do not. Nevertheless, if you plot the multifactor productivity that is
implicit in the staff’s forecast along with the change in economic activity, you will see a
fairly high correlation. I assume that the reason we have this dramatic falloff in
multifactor productivity growth in 1999, after very strong gains in the two previous years,
is the forecasted slowdown in economic growth.
But if the slowdown in economic growth does not occur, then presumably
productivity growth will not slow much either. And unless there is a really dramatic rise
in nominal wage increases, unit costs will not move very much. I am having difficulty
knowing where this higher inflation in the model forecast would come from if I leave that
out. That is especially the case when the pairing of multifactor productivity growth and
low unit costs is essentially combined with an oil price increase to get the CPI increase.
As we discussed yesterday, the oil price increase is based on the presumption that we
know more than the market about the long-term direction of the price for West Texas
Intermediate. Now, that may well be the case, but history has been less than helpful in
that regard. I would not bet the ranch that one could buy December futures for WTI and
know for certain that it would result in a profit.
Our structural and VAR models that are projecting this price acceleration are not
properly specified, in my judgment, unless some means are found to capture the
technologically driven price-capping variable. Lagged dependent variables do not do it.
VARs may appear to give us good results, but they are begging the question on the
crucial missing variable.
Moreover, it is evident that whenever nominal wage pressures have surfaced,
producers have chosen to dip into the available technology to substitute profitable capital
for labor. This has made the growth of potential output per hour variable; indeed, it’s a
function of nominal wage increases. The reason is that if nominal wage increases pick

2/2-3/99

110

up, there is clear evidence in recent years that producers will endeavor to dip into that
untapped pool of technological capital projects. We have had many anecdotal
conversations on this subject. We had a discussion yesterday in which I mentioned
talking about this issue with the head of British Telecom. He was saying that the
availability of new technologies and synergies is accelerating at such a pace that they
don’t know what to invest in. Everything is terrific. It is a question of which of the
individual rates of return is higher. They are all above the cost of capital. Larry Meyer
mentioned a discussion he had with a senior Microsoft researcher who was confirming
the same thing. Something different is happening here, and our models are missing a
crucial variable.
As we all know, when econometricians get regression results that appear out of
line with the real world, as Bob McTeer was saying, they have to look for the missing
variable. I submit that there is a missing variable, and we are learning more about the
nature of its characteristics. I think it may be about time to try to substitute this variable
for NAIRU. Let me put it this way: Neither one is an observable phenomenon, but
neither was the planet Pluto before 1930. Scientists figured out that there had to be
something there, given the extent to which Uranus and Saturn were deviating from their
forecast orbits. Well, I submit that at some point we are going to come to the conclusion
as statisticians that the simultaneity of a falling inflation rate and an ever tightening labor
market is trying to tell us something. My plea is that we try to do something to see
whether we can get some answers that do not give us Chart 3 and Chart 4, which in their
present form are not terribly helpful.
There is very little evidence at this stage that the expansion is slowing. I know it
is supposed to decelerate significantly in the first quarter, and I don’t deny that the motor
vehicle segment is going to reduce it by some 2 percentage points. I also gather that we
will get some further deceleration from government and probably from trade. But these
are forecasts. As of the moment, we have a 5.6 percent annual rate of growth in the
fourth quarter, and the impression I have is that that number will be revised up, not down.
In the first quarter, some points may be taken off the possible 6 percent fourth-quarter
growth number, but it has to go down by a significant amount. I recognize that
productivity increases cannot explain all of the growth because if they could, then we

2/2-3/99

111

would not be getting a reduction in the total number of people who are either unemployed
or out of the work force looking for jobs. So, something has to give here at some point.
The inflation rate, as we measure it, is at zero currently; if we got the
measurement right, it probably would be negative. Given that there is still some bias in
these statistics, we certainly can wait a while. But it is conceivable that the data we are
going to see in the weeks ahead will confirm something that we have not seen to date,
namely underlying inflationary forces. I don’t mean strong economic growth. That tells
us little except that productivity is accelerating. But what is happening in the labor
market may tell us something. We may indeed find that the labor market issue is
becoming significant. We may find when the data are published at the end of the week
that average hourly earnings will show a spike, and we may see another spike the month
after. Clearly, that would suggest that something different from what we’ve been seeing
is happening.
What I would recommend after this long discussion is that we stay where we are
at the moment. I would be disinclined to go asymmetric toward tightening at this stage
because the data may turn out to be softer than I suspect, and flipping the symmetry back
and forth would not help us. If it turns out, however, that the labor market data we get on
Friday really tighten up or if the data we get on the CPI or the PPI show that something is
stirring, I think a fairly strong signal in the Humphrey-Hawkins testimony that we might
tighten would be wholly appropriate. Indeed, I would say it would be highly desirable. I
don’t think we have reached that point yet, but I can readily conceive of the possibility
that we will.
I would suggest that we wait a few weeks before we show our hand and leave it
for the Humphrey-Hawkins testimony to suggest some asymmetry toward tightening if
that seems appropriate. If I were to sum up the general discussion at this meeting, I
would say that the consensus is probably a shade in the direction of asymmetry. I request
that we do not adopt asymmetry, but that really depends on our strategy. I don’t sense
any pressure to move today, but I do detect some inclination on the part of a significant
number of the Committee members to move in the direction of asymmetry toward
tightening. I frankly would prefer not to do that. Whichever way we go will probably
have zero impact on when and to what extent we move or do not move in the weeks

2/2-3/99

112

ahead or on what we decide when we meet again on March 30. If inflation were not
actually dead in the water at this stage, I would say that we would have to be far more
concerned about being preemptive. I don’t see the necessity for that at this moment.
Vice Chair.
VICE CHAIRMAN MCDONOUGH. Mr. Chairman, I fully support your
recommendation. Clearly, it seems inappropriate to change policy. And with inflation
dead in the water, I also believe this is not the moment to go asymmetric toward
tightening. I agree that the Humphrey-Hawkins testimony gives you, representing the
Committee, an opportunity to discuss that. With the additional data becoming available
between now and that time, I think that’s the right thing to do.
CHAIRMAN GREENSPAN. Governor Rivlin.
MS. RIVLIN. I also strongly support your recommendation, Mr. Chairman, for
the reasons ably stated by the Vice Chair. But I would be a bit cautious about sending a
signal in the Humphrey-Hawkins testimony. We have seen so little inflation that I’m not
sure we will see anything in the near term that could convince us that we should move up
soon.
CHAIRMAN GREENSPAN. President Stern.
MR. STERN. I certainly support your recommendation, Mr. Chairman. Let me
make a few somewhat random, but hopefully not too random, comments. First of all, I
enjoyed your analysis, especially the point about the importance of global capacity and
the arbitrage between the manufacturing and service sectors. It is something I have been
trying to convince my staff of without a great deal of success to this point, and perhaps
you have aided in that effort.
I must say I am a little puzzled by the fascination with NAIRU. I thought work
done by Stock and Watson and others suggested that if there is a NAIRU, it lies
somewhere between 4 to 7 percent or some range like that, which is quite wide. If that is
right, it is not a terribly useful concept for policy. So, I have been reluctant for some time
to go down that path very aggressively.
Let me also add a thought about wages and compensation. At least as far as our
employees are concerned, and this is confirmed by the labor leader on our board as well,
the issue we hear most about is the scarcity of time. What many employers are doing--I

2/2-3/99

113

know we are--is offering more flexibility in work hours to make it easier for people to do
the routine tasks of running a household and taking care of family obligations that
obviously take a lot of time. My guess is that at the end of the day this is probably
productivity enhancing. If it shows up in wages or compensation at all, it may not affect
unit labor costs adversely if it has a positive effect on productivity.
I am a little suspicious about attributing a lot of the recent favorable inflation
performance to special factors. Back in my youth when I was doing bottom-up
forecasting, one could always find special factors to explain virtually everything. It
seems to me that this subdued inflation behavior has gone on long enough now that it
probably reflects a lot more than special factors. And I am struck by the fact that most, if
not all, of the major central banks around the world are committed to a low inflation
policy. The fact that we are getting low inflation and some surprises on the downside is
perhaps not so surprising, if you get my drift. That to me is the policy regime that we
probably have in the world today.
The one thing that concerns me in the current situation is a communications issue.
I have the perception that financial market participants and others feel that we cannot
change policy in either direction in the current circumstances and that we will not. I
certainly don’t feel that we should any time soon, but I do think we should make an effort
to communicate that indeed we are prepared to change policy if and when it is
appropriate to do so. As for what we might point to as a rationale for changing policy in
either direction, right now I don’t think there is any such rationale, which is why I came
out where you did. But it seems to me that we would want to put people on notice that
indeed we are prepared to move under some circumstances. We might want to call
people’s attention to money growth, exchange rates, nominal interest rates, the bond
market, and so forth, and say they have informational content and that we are going to be
paying attention to those variables among other things in assessing the appropriate stance
of policy going forward.
CHAIRMAN GREENSPAN. Governor Ferguson.
MR. FERGUSON. Mr. Chairman, I support both halves of your
recommendation. I also want to comment a bit on the strategy, which I strongly endorse.
As I said yesterday, in periods when models do not seem to be working very well--and

2/2-3/99

114

we seem to be in one of those periods--I think it is quite important to weigh the incoming
data more heavily. To some extent that is what you were suggesting. That doesn’t mean
that our hands are tied in any sense. It does mean that once the data become clearer, we
should be prepared to move. And I sense that that is where we are.
I’d say one other thing. You put some emphasis, appropriately, on productivity.
But we are aware that productivity is quite cyclically related, and I’m not sure we have
parsed out those cyclical differences well enough yet. There may be some studies of the
data that do that. One of the things we have to continue to look at is the degree to which
we can pull apart the cyclical components from the underlying changes in productivity.
I also would agree with President Stern’s point with respect to NAIRU. It has
never been measured very accurately, and I’m not sure it is going to be measured any
more accurately in the future. It is an interesting concept, but we should not put too
much weight on it for policy purposes.
CHAIRMAN GREENSPAN. President Poole.
MR. POOLE. Mr. Chairman, I support your recommendation. I would like to
talk a bit about what I regard as a very important issue, which Larry Meyer emphasized
yesterday. And that is that we need to be talking not just about the setting of the federal
funds rate, but about the policy strategy from which the current rate setting arises.
It is clear that the central fact occupying us is this ongoing surprise in the
relationship between real activity and the rate of inflation, which we call the Phillips
curve. The Phillips curve framework has been around in its current form essentially since
the late 1960s. We have worked with what we call the expectations-augmented Phillips
curve. I believe there is far too much emphasis on the NAIRU and not enough emphasis
on the expectations part of the behavior in these markets. To me the experience over the
last couple of years indicates that expectations trump the unemployment rate in
determining the current rate of inflation--that the expectation of continuing low inflation
is the factor dominating this outcome.
Firms say over and over again that they don’t have pricing power, and that is
because their competitors won’t follow any price increases they have tried to put into
place. Competitors behave that way because they think the inflation rate will remain low

2/2-3/99

115

and, therefore, if they can retain control over their costs, they are happy to grab more
market share. So prices remain very much under control.
I agree with you that emphasizing the exchange rate and oil prices as special
factors is not the way to look at those variables. Those above all are expectational
variables. Oil is a storable commodity. Oil is behaving so well in good part because it is
expected that prices will remain low in the future. Oil is not a good store of value any
more with the rate of inflation down. The same is true of the exchange rate. It has
behaved so well because of the sustained low rate of inflation in the United States.
Now, that does not mean that the unemployment rate is not below what is
sustainable in the long run. There are so many stories about the existence of labor market
pressures and shortages and so forth that we probably are operating below the NAIRU. It
is just that the expectations part of it is dominating the outcome in terms of the current
rate of inflation.
The ideal situation, if we knew how to do it, would be to peg the rate of inflation
at a very low level and go for price stability, letting the real economy run wherever it
wants to run. What we would all like to see happen is to have the maximum possible
employment and productivity growth consistent with price stability. The problem is that
there seems to be so much evidence that the real economy leads the cyclical process, with
inflation coming later. So, to say that we can ignore the real economy in setting current
policy seems to me likely to be a mistake in the long run.
Therefore, I come down to saying that we are going to have to pay attention to the
fact that the real economy is growing very vigorously and that the unemployment rate is
low. We should not ignore the anecdotes about very tight labor markets. Credit
conditions and monetary growth are both very clearly on the stimulative side. The
financial turmoil that motivated this Committee last fall appears to be mostly behind us.
Over the course of this year, starting sometime this spring, I think it is going to be
appropriate to be tightening up on the funds rate.
I would like to come back to the original point that I was making about the
importance of expectations. I think we need to do everything we can to seal in this
expectational environment. That is what makes all this possible in my view. That means
that if we get some bad news on the inflation rate, we need to be prepared to pounce in

2/2-3/99

116

order to demonstrate that we are not going to let that continue. To show that we take
inflation very seriously is to me the most constructive thing we can do. Whether or not
we want to act before we get some uptick in inflation is another matter. But we are not
faced with that issue immediately at this meeting. Thank you.
CHAIRMAN GREENSPAN. President Boehne.
MR. BOEHNE. I support your “B” symmetric recommendation. I believe
watching and waiting is the right approach at this point. I think most of us feel that we
are sitting on a situation that is really very positive; or at least our experience over the last
several years has been favorable. We may have some inner sense that it will not last or
that we are going to have to do something over the next few months. Maybe we will or
maybe we won’t. I have a completely open mind as to what the next move on policy
might be. I do think that when reality tells us something different from the models that
we ought to take a new look at the models. Your contribution this morning was helpful
in that regard.
As far as NAIRU is concerned, my personal view is that it is a useful analytical
tool for economic research but that it has about zero value in terms of making policy
because it bounces around so much that it is very elusive. I would not want our policy
decisions to get tied all that closely to it, especially when most of the NAIRU models
have been so far off in recent years.
As far as your testimony goes, I think it would be appropriate to lay out the
possibility that we may have to tighten and for you to describe the kinds of circumstances
that might lead us to that. On the other hand, that discussion ought to be balanced in the
sense that there may be a set of circumstances in which we do not have to tighten. So,
the possibility of tightening ought to be raised, but the possibility that we might not have
to tighten also ought to be discussed.
CHAIRMAN GREENSPAN. President Hoenig.
MR. HOENIG. Mr. Chairman, I am satisfied with your recommendation. As I
mentioned yesterday, we have put some additional stimulus into the economy over the
last several months and I think there is a case to be made for some unwinding of that,
perhaps in the not-too-distant future. Depending on unfolding developments, that might

2/2-3/99

117

be appropriate to mention in your Humphrey-Hawkins testimony, as you suggested. So I
am fine with your recommendation.
CHAIRMAN GREENSPAN. President Guynn.
MR. GUYNN. Mr. Chairman, I like your recommendation, including the
handling of the symmetry issue. I hope, like others, that you will use the HumphreyHawkins setting to put people on notice that we are in fact prepared to tighten if
circumstances suggest that is appropriate.
I was particularly pleased that you resurrected the notion of “preemptive”
policymaking this morning, something we have gotten further and further away from as
we have gotten caught up in the euphoria of good experiences recently. Last night I was
reflecting on yesterday’s discussion and I was struck by how many of us have fallen back
on our models recently. I know we have all been using them, but we talked about them
more yesterday than in the past. I think we did that perhaps as a second check on our loss
of peripheral vision, as a check against just going on our instincts. Those models, even
with their flaws and possible missing variables--and you gave them another licking this
morning--do serve to remind us that there is a danger of making the mistake of waiting
too long to tighten policy and that there is in fact a price associated with that. So I am
pleased that we at least have left the door open to a preemptive move. Thank you.
CHAIRMAN GREENSPAN. President Minehan.
MS. MINEHAN. As my comments over the last couple of days probably have
indicated, I can accept not changing policy but I am a little more uncomfortable with not
changing the symmetry.
In our meetings last summer we discussed--though you were much more
expansive in your discussion today--the various aspects of what is going on and how
difficult that is to measure and to capture in our equations. But it is easy to see, if one
looks at it closely, that in terms of global capacity and changes in technology and so forth
you could have given that same speech last July. To some extent you did. We had
forecasts then that looked toward no change in policy going out over a period of several
years and the trajectories didn’t look terribly different from what we are looking at now,
in that growth was a little slower and we had some small pickup in inflation. As I recall,

2/2-3/99

118

your Humphrey-Hawkins testimony reflected that in July. Some of us were uneasy then,
and we had a bias toward tightening at that point in time.
What has changed? We experienced some financial turmoil. We changed
monetary policy to provide protection to the real economy. The need for that has largely
dissipated. We have an easier monetary policy now, and we have the same kinds of
concerns that were disturbing me at least in the spring and the early summer. I’m talking
about credit conditions, monetary growth conditions, stock market conditions, and asset
market conditions more generally that seem to be stoking things to a point where, no
matter what one feels about the future, we could be facing some real problems down the
road. The higher everything flies, the farther it has to fall.
For me anyway, it is not a situation of when we see the “eyes of inflation” but of
what we are building in with regard to overall conditions. I think that relates to the
expectations issue. It feeds into what people expect. I don’t have any problem with it
right now. But if people continue to see not just the same market conditions as last year
but those same market conditions and an easier monetary policy, there is a point where
they will begin to think that we aren’t going to change our policy--that nothing can
convince us to change in a preemptive fashion. That could feed into a negative
expectations process that will defeat us in the end. So I am very worried about where we
are. I do not have a vote so I can’t dissent, but I am really concerned.
CHAIRMAN GREENSPAN. Governor Meyer.
MR. MEYER. When I think about the inflation process and the inflation
dynamic, I always point to two things: excess demand and special factors. I don’t know
any other way to think about the proximate sources of inflation. When I think about
excess demand, I think about NAIRU. If we eliminate NAIRU and that concept of
excess demand, it moves us into very dangerous territory with monetary policy.
I would remind you that in the 20 years prior to this recent episode, the Phillips
curve based on NAIRU was probably the single most reliable component of any largescale forecasting model. It was very useful in understanding the inflation episode over
that entire period. Certainly, there is greater uncertainty today about where NAIRU is,
but I would be very cautious about prematurely burying the concept.

2/2-3/99

119

I was also surprised to hear how many people believe that special factors--the
appreciation of the dollar, health care cost developments, and so forth, which I feel have
been so important over this recent episode--have been so immaterial in their views of
inflation. Declines in oil prices from $26 to $12 per barrel seem to me quite relevant in
understanding the recent inflation experience. So, I think we have to keep some balance
in this.
Even if productivity growth has moved upward, we know it has been above trend
because we know the unemployment rate has been declining. So I come back to my story
about the incremental Taylor rule. If you think NAIRU is as low as 4½ percent, fine.
But that still means to me that if the economy is growing above trend, there is a
considerable prospect that it will continue to do so and that we should put some discipline
in monetary policy by leaning against the cyclical winds. Otherwise, every time there is
an upside demand shock, we will allow monetary growth to accelerate to accommodate it
and wait for the day of reckoning.
In terms of symmetry and asymmetry, the risks seem reasonably balanced around
trend growth. Does that mean we should have a symmetric or an asymmetric directive?
The problem I have is that I believe the initial conditions are such that we are already
operating beyond the point of sustainability in labor markets. If we have balanced risks
around trend growth, that tells me that we should have an asymmetric directive. Growth
a little lower than trend would not call for any easing of monetary policy, but any growth
above trend would call for a tightening of monetary policy.
Mr. Chairman, I can certainly accept your recommendation and wait a little longer
on the asymmetry story. But I hope some of these concerns will be reflected in your
testimony. Thank you.
CHAIRMAN GREENSPAN. Governor Gramlich.
MR. GRAMLICH. Mr. Chairman, I am happy to support your recommendation
for sitting tight. I think Larry Meyer made a useful contribution yesterday. I have been
thinking in my own mind about how to resuscitate the Taylor rule. I may not agree with
everything he said, but I view it as a reasonable strategy. One thing it means is that as
soon as we see evidence of any acceleration in inflation, we have to move against it. I’d
say that almost everybody around the table would agree with that.

2/2-3/99

120

Models have taken a hit this morning, so let me say a good word for them. For
many years they have been a useful framework for thinking about the economy.
Actually, I think the staff should be commended for this week’s presentations because
they have employed models very well in illustrating the issues, even though the models
did not resolve all uncertainties. The models say that we ought to be a little nervous
about the current situation, and I am, even though I am happy to sit tight and follow your
recommendation. On the other hand, you and others have made some very telling
criticisms of the models, and a lot of you set your staffs to work on points that came up at
the last meeting. Some points have come up at this meeting that would be worth
exploring. Bill Poole raised a point about expectations. For a long time, how to factor
expectations into the models has been a real puzzle. The treatment of expectations has
always been fairly haphazard, and it may be that expectations are much more important
now in a different way from what has been built into the models. We ought to take a
serious look at that.
You also raised a very important point on the price-capping issue. I don’t know
exactly what adjectives to use to describe it. I consider it important; I have been mulling
it over myself. It may be time, and there may be enough data, to try to work that into the
models. Let’s try to have the models step up to the challenge and see if they can deal
with some of these issues that clearly ought to be dealt with. Thank you.
CHAIRMAN GREENSPAN. President Broaddus.
MR. BROADDUS. Mr. Chairman, in terms of the earlier discussion about the
way the world works, I want very much to believe my lying eyes but I am having
difficulty doing that. I don’t doubt that there is a missing variable out there. Your point
on that is very well taken. But the fact is that we really don’t know what it is exactly.
We do not yet have a very clear understanding of its strength or its likely persistence.
That is the background against which I will present my own way of approaching the
current situation.
The comments that Larry Meyer and Cathy Minehan made, and especially Bill
Poole’s comments on expectations, really resonated with me in terms of the way I view
the world today. It seems to me that even if we knew what this missing variable was and
could somehow specify it and analyze it, we still would very likely conclude that

2/2-3/99

121

domestic demand is growing at an unsustainable pace. I think we would also suspect that
in the months immediately ahead it may be sustained and even boosted by the lingering
and lagged effects of the easing actions that we took last fall and possibly also by the
strong growth in the monetary aggregates. I worry about where all of this is going to take
us. What worries me the most now is that the markets seem to be discounting almost
completely--others have made this same point--any tightening action on our part, despite
these continuing upside surprises in the data. At a minimum we need to send a clear
signal that there is at least a possibility that at some point in the near term we might
tighten. I think that would help sensitize the markets. There is a ball game going on here
and we just are not in it. We need to get back in it in some way so that if we have to act
on short notice, or maybe on no notice, at least we will not be so much of a lightning rod.
I am probably the only person in the room who feels this way, but I believe one
could make a case for undoing at this meeting some of the easing that we did last fall.
Actually, in the Bluebook on page 19 the staff made a good case for that. Of course, they
made some other cases as well. That’s the problem. [Laughter] Speaking for myself, I
would strongly prefer an asymmetric directive this time. If we don’t do that, I hope, as
you have already suggested, Mr. Chairman, that you will strongly signal in your
testimony that there are upside risks in the current situation.
CHAIRMAN GREENSPAN. President Parry.
MR. PARRY. Mr. Chairman, I support your recommendation with regard to the
federal funds target. But I share many of the concerns that have been expressed by my
colleagues, particularly President Minehan and Governor Meyer.
I also believe that it is less clear, certainly less clear than at the last meeting, that
the risks to the outlook are symmetric at this point. I think a case could be made for
asymmetry, but we will be getting lots of information between now and the next meeting
and perhaps we can probe that issue more deeply at our next meeting.
CHAIRMAN GREENSPAN. President McTeer.
MR. MCTEER. I agree with and support both parts of your recommendation.
CHAIRMAN GREENSPAN. Governor Kelley.

2/2-3/99

122

MR. KELLEY. Mr. Chairman, I certainly support your recommendation. But I
also would like to associate myself with the many concerns and reservations that have
been expressed by a number of members of the Committee.
Let me add just one additional brief caution. We frequently speculate about
whether or not this Committee should or should not be preemptive. If we believe, and I
think we all do, that policy affects the economy with long and variable lags, then we
should keep in mind that we are always, every day, unavoidably being preemptive. Even
taking no action at a meeting is preemptive, in the sense that any decision is going to
impact the economy a year or two years out, or however long those lags may turn out to
be. I think it is useful to keep in mind that we are always being preemptive.
CHAIRMAN GREENSPAN. President Jordan.
MR. JORDAN. Thank you. Quite a few references have been made to the role of
expectations. I did not come into this meeting with any expectation that there would be
an adjustment in the funds rate at this meeting. I didn’t know what the reasons for not
making an adjustment might be, but I did not expect an adjustment. However, I still
stand by my dissent on the cut in the funds rate taken in November, and I think it should
be reversed. I don’t think this would be the time to do it, with due respect to Al Broaddus
and others whose views are that perhaps it should be done now.
A reason for acting at this time that I would not find desirable would be that the
economy is growing too fast. In fact, there is a potential that the economy would not
grow so fast if we tightened at this meeting. There is the potential for an unfortunate,
undesirable misinterpretation, coming after last Friday’s report on real growth, that the
Committee thinks that there is too much growth and too much employment and that we
are unhappy about that. That would not be desirable. In fact, I believe we ought to go
the other way by trying to emphasize how delighted we are with this strong growth, how
delighted we are with the low unemployment rate, and how delighted we are with low
inflation and, therefore, we did not take action.
Nevertheless, expectations going forward are going to be a dominant concern to
us and to everybody who watches us. As much as we and the world are happy that the
average level of current consumable goods and service prices is not changing very
rapidly, that is the necessary condition for successful monetary policy but it is not a

2/2-3/99

123

sufficient condition. If it were a sufficient condition, we would still be under the gold
standard. In fact, the Federal Reserve would not exist if it were a sufficient condition.
We have had cases in our history and cases in other countries’ history where the present
price of claims to future consumption moved very substantially--sometimes too far up
and then sometimes very rapidly down. This creates financial instabilities that are
unhealthy for the performance of an economy. Otherwise, 1907 would not have
happened; 1893 would not have happened. Central banks and monetary policy are
designed to try to correct some of the flaws inherent in a pure gold standard type of
environment. That means that we have to take into account things, such as the present
price of future consumption, that are missing in our usual measures of goods and services
prices.
For some time, people have been purchasing equities in the expectation of selling
them to somebody else at a higher price, with not a thought as to what the earnings
potential or real value is--and it’s probably more than just my three kids. [Laughter]
There are people making real estate investments for residential and other purposes in the
expectation that prices can only go up and go up at accelerating rates. Those expectations
ultimately become destabilizing to the economic system.
CHAIRMAN GREENSPAN. President Moskow.
MR. MOSKOW. Thank you, Mr. Chairman. It is clear from our discussion that
there is a lot we don’t know. I think we all feel this conflict--I certainly do--between
what economic theory and economic models are saying on the one hand and what we
have seen as the reality in recent years. We all have been over-forecasting inflation and
under-forecasting growth for at least three years. Someone raised the question as to what
has changed. The way I look at it, the longer it goes on, the more it becomes such a
puzzle and the more one questions it. So I certainly agree with the thrust of your
comments that we should be questioning our forecasting models with regard to the help
they give us in policymaking.
But having said that, I am concerned that our current monetary policy stance may
be too loose because of the actions we took last fall. I don’t think the risks are strongly
skewed toward the upside, but I believe they are somewhat skewed to the upside at this
point. I agree with the comments that have been made today that we have to be prepared

2/2-3/99

124

to move very quickly when we see some evidence of inflation in the pipeline. It doesn’t
have to be in the actual numbers, but we have to be sure that we see some evidence.
Clearly, we have not seen any evidence in recent years. So, I support your
recommendation at this time. But we have to be very, very vigilant.
CHAIRMAN GREENSPAN. There is a substantial majority in favor of “B”
symmetric and I would like Norm to read the directive accordingly.
MR. BERNARD. This language begins at the bottom of page 23 of the
Bluebook: “To promote the Committee’s long-run objectives of price stability and
sustainable economic growth, the Committee in the immediate future seeks conditions in
reserve markets consistent with maintaining the federal funds rate at an average of around
4¾ percent. In view of the evidence currently available, the Committee believes that
prospective developments are equally likely to warrant an increase or a decrease in the
federal funds rate operating objective during the intermeeting period.”
CHAIRMAN GREENSPAN. Call the roll.
MR. BERNARD.
Chairman Greenspan
Vice Chairman McDonough
President Boehne
Governor Ferguson
Governor Gramlich
Governor Kelley
President McTeer
Governor Meyer
President Moskow
Governor Rivlin
President Stern

Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes

CHAIRMAN GREENSPAN. Thank you. Our next meeting, as you all know, is
on Tuesday, March 30.
MR. GRAMLICH. I have just one question on something Don Kohn said earlier.
Next meeting we will be under the new regime where if some significant change in the
Committee’s consensus develops, in the sense that we talked about at the last meeting, we
would announce that in the afternoon. Is that right?
MR. KOHN. If it involves a significant shift in the Committee’s consensus, that
would have to be considered. Yes, the new regime would be in effect.

2/2-3/99

125

SPEAKER(?). But we are not required to say anything?
MR. KOHN. We are not required.
MR. GRAMLICH. But the regime is in effect?
MR. KOHN. The regime is in effect.
MR. PRELL. Just as another minor housekeeping matter: Could everyone let me
know whether or not they want to change their individual forecasts? That way, even if
there is no change, I will know that we haven’t missed anyone. That would be helpful.
END OF MEETING

Full text of Federal Open Market Committee Meeting Minutes, Transcripts, and Other Documents : Meeting, February 2-3, 1999 : Transcript

FRASER