Past President's Breakfast
Alan Greenspan
Chairman of the Board of Governors
Federal Reserve System
Alan Greenspan's speech to the National Association for Business Economics, October 7, 1998. Chairman Greenspan, who was NABE president in 1969-70, was introduced by Kathryn Eickhoff, who was NABE President in 1980-81, and a long-time colleague at Townsend-Greenspan & Co. Afterwards, she presented him with a cassette version of Mahalia Jackson singing "He's Got the Whole World in His Hands."
I'm here to talk about the economic outlook for the United States because that's what we do for business. And I mean that in several different ways. It's pretty obvious, I think, that the outlook for 1999 for the U.S. economy has weakened measurably in the aftermath of the Russian devaluation and debt moratorium. The building sense of stability in international markets has been undermined as the belief that the Asian contagion had moved into remission has been proved quite wrong. The result of this, as I'm sure you're all acutely aware by now, has been a very dramatic change in the whole risk profile of the world. Clearly the size of the Russian economy, either in real terms or in financial terms, can scarcely be an element which would engender the types of consequences that have occurred subsequent to that moratorium.
So what we're looking at is a really far more fundamental shift in attitudes for which the Russian episode essentially -- for which the Russian episode was essentially the triggering mechanism. What we have seen, as I'm sure those of you who watch these data daily are acutely aware, is really starting in mid-August almost concurrently with the Russian moratorium so-called stripped Brady bonds -- that is, dollar-denominated issues whose basic obligation is that of the host country -- escalated very sharply in yield relative to U.S. Treasuries, all representing, essentially, credit risk in the issuing country.
If you look across the spectrum beyond dollar-denominated issues in the United States and look, for example, at the 10-year swap rates, which, as you know, is the spread between the yield on -- or I should say the cost to banks in particular areas less the sovereign credit-rated issue of that particular country, they've all gone up. So it's an issue of yield spread expansion, which really cuts far more across the world spectrum than merely dollar-denominated issues, which we, as our general won't would have us do, would be our fundamental concern.
When you get beyond the rest of the world and come very specifically to the United States, we see a phenomenon that I suspect that none of us in the room have seen before, at least since I have more gray here, what's left of it, than most, and have been effectively looking at the American economy on a day-by-day basis for almost a half-century, I've never seen anything like this. This is really quite an unusual phenomenon, and it shows up in a variety of different ways. The most obvious one is that we're seeing a fairly dramatic increase in spreads of junk bonds, so to speak, and even though the investment-grade issue -- say B-double-A down to triple-A, or up to triple-A, I should say -- while those yields have actually gone down, they've gone down nowhere near the extent to which U.S. Treasuries obviously have declined, and it's basically the spread itself which is the ultimate measure of risk-aversion, because what the episode that comes out of the Russian experience has created is a major shift towards risk-aversion pretty much throughout the world, as exhibited mainly in the financial markets.
If you go further and you look at the number of different other spreads, they're all showing the same thing in the United States: Commercial paper or euro-dollar yields vis-a-vis the Treasury Bill rates have opened up, all, incidentally, at the same time, reflecting the mid-August sort of quantum shift. Having seen this phenomenon emerging, we at the Fed put in a special survey of senior loan officers to see whether in fact the exceptionally benign results we have been getting pretty much through the summer had changed. And indeed, as I'm sure you know from publications we've put out, we have seen a major shift amongst the larger banks, specifically towards tightening credit standards, but it is essentially against larger companies.
As far as we can judge, the standards and the credit availability through smaller companies, at least as exhibited in this survey, doesn't show anything of considerable significance. Anecdotal evidence, however, indicates, that there has been a fairly dramatic increase in collateral requirements pretty much across the board in every area where collateral is a crucial issue. And as one might expect as a consequence of all of this, the issuance of long-term debt or intermediate debts have all gone down fairly dramatically. Obviously, in the junk bond area, it's dried up really quite significantly. But it has even gone down in investment-grade issues, despite the fact that yield spreads have net on balance gone down from investment-grade, even though obviously far less than corporates.
As a consequence of this, we're seeing a good deal more of the financing in the United States moving into the shorter end of the market. The commercial loans out of banks, for example, in September just tilted up sharply, and commercial paper issuance has been fairly significant. So that what we are looking at is a marked moving back into shorter-term financing, which is what ordinarily one would expect to see when you see a significant degree of risk-aversion emerging in the context of the type of markets that we are looking at.
I had argued, in fact quite extensively over the last two or three years, that the markets were becoming too complacent about risk. That is, I argued, using these same data, that yield spreads had gotten down to negligible proportions, and that the usual normal pricing of risk in the debt markets was obviously lacking. Well, that has changed -- and it has changed -- if there was a dime to turn on, it did. And it has been a fairly dramatic, broad and as yet seemingly incomplete adjustment. I mean we -- there are fluctuations up and down in the Brady yield spreads, but here, junk bond spreads are continuing to open up, all of the short-term measures of yield -- of risk aversion, I should say, are still opening up. So this thing hasn't come to an end yet. But even so, we've reproduced some of the spreads that we haven't seen for years. In fact, you go back to 1991 and earlier in certain cases.
The thing which I think I want to emphasize here, even though this is not the way we usually do the calculations, is that the appropriate calculation which invariably uses Treasury issues as the denominator or as the base against which the spreads are being calculated, obscures a very important trends, and that is the distinction between risk aversion on the one hand and the desire for increased liquidity on the other. Specifically, we ought to be measuring risk aversion solely by using the particular instrument, whether it's a junk bond or commercial paper or whatever, against an illiquid, riskless U.S. Treasury equivalent. The reason I say that is one of the other aspects of this post-Russian adventure has been a fairly dramatic shift towards pure liquidity-protection desires, as distinguished from risk aversion.
And we know it in a number of ways, but very specifically, there are, as a number of you people in the market are aware, what we call "on the run" U.S. Treasury issues; that is, those are the issues that for, say, 30 years, have just been issued by the U.S. Treasury. They have a fairly significant amount of transaction and volume going on, and, indeed, the bid-ask spreads are relatively narrow because the activity is fairly heavy.
"Off the run," meaning, say, just the immediately preceding issue, which is effectively the same maturity, will be yielding, normally, three to five basis points higher and that differential is a liquidity premium; pure liquidity -- no risk, because obviously, all those issues are riskless instruments.
We see a similar story when we look at the so-called "TIPS," that is, the Treasury inflation-protected securities whose yield, when -- whose yield -- remember, it's a real yield. It's essentially, supposed to be, a nominal yield less an expected projection of what the BLS will publish in the Consumers Price Index. I basically say -- I didn't use the term "inflation" because inflation is getting ever more difficult to measure in all its various respects, but very specifically, what the instrument is, is an endeavor to get a forecast of the CPI.
Now, if you match the nominal equivalent of, say, 10- or 30-year maturities and you take the nominal versus the real, one would expect you're getting an inflation forecast. Well, I suspect that when you look at the 30-year gap and you see that the real yield is about 3.6 percent -- that's what the actual security sells at currently -- whereas the nominal is 4.8 -- actually, it's 4-3/4 this morning, the last time I looked -- you get 1.2 percent difference.
Risk aversion is a positive, conscious, calculated adjustment that the risks of security A have risen or fallen relative to the risks of security B. Liquidity protection, however, results not from a judgment, some conceptual insight, but rather a loss of knowledge; that is, a destruction of what one perceived to be that they knew about.
As a consequence, what is occurring is a broad area of uncertainty or fear. And when human beings are confronted with uncertainty, meaning they do not understand the rules or the terms of particular types of engagement they're having in the real world, they disengage. When you're crossing the street and you're uncertain as to whether a car is coming, you stop. You disengage from that particular process. And when you are uncertain about commitments in the marketplace, you disengage. And since most of our markets are net long in the sense that they're not a zero-sum game, disengagement, of necessity, means that prices fall.
But what is crucial about this distinction is that the individuals who were moving from, let's assume, the illiquid U.S. Treasurys to the liquid, on-the-run liquid issues, are basically saying, "I want out. I don't want to know anything about whether a particular investment is risky or not. I just want to disengage." And the reason you go into these liquid instruments is that that is the vehicle which enables one to disengage as quickly as possible.
The importance of this very specifically in the international markets is that it has tended to flatten the risk-aversion judgments. In the abstract, if there was only liquidity protection in market actions, all investments would sell equally with respect to risk, because no one would be making any distinctions. And indeed, there are a number of senior policy-making officials in a number of emerging economies who are basically arguing effectively, "We do everything exactly right. We haven't changed what we have done in the last six months. In fact, we've improved our policy. And what's happened is we've been hit by this tsunami coming from Asia, and we are under very significant stress and we cannot borrow in the private markets."
The notion that that is true is clearly not demonstrated by the figures. We do know that if you look at the various different grading spreads, they've all moved up. But the levels are quite different, and indeed there are certain countries which seem to be far less subject to the problems of risk aversion than others, so that the argument that there's been some flattening is partially true, but only partially. But the important issue is that what we have involved here is a more virulent form of erosion in the sense of liquidity protection than risk aversion. Risk aversion may be exceptionally difficult for the people who are at the top of the risk, but it's still a rational market-calculated market-adjustment process.
A major shift towards liquidity projection is really not a market phenomenon. It's a psychological -- it's a fear-induced, psychological response. And markets cannot effectively function in an efficient manner in that environment, because obviously if we're looking at market and consumer preferences working their way through the price structure, that's the way one effectively creates a sensitized system in which the allocation of goods and services basically are maximized relative to consumer preference schedules. So what we are looking at here is still a major shift in risk and a budding rise in illiquidity protection, which still hasn't taken on a significant dimension.
Having said all of this, we are far short of anything that could resemble a credit crunch in the United States. To be sure, there are all sorts of difficulties lots of people are having borrowing. But it is by no means evident that this is having as yet a significant impact on the real variables. Indeed, the virtual drying up of the junk-bond market, one would assume, would have a significant impact on lesser credit-risk capital investment. But it turns out a substantial amount of junk-bond issuance was largely for the purpose of refunding at lower interest rates what had previously been booked, and as a consequence, prior to this big surge with those relatively low junk-bond interest rates, a lot of the volume had nothing to do with the financing of capital investment.
So as a consequence of that, we don't see any really material pulling back. In fact, if you look at the physical aspects of the economy as a whole, it is very difficult to find any real variable impact as yet. To be sure, the payroll data that was released on Friday showed significant slowing, but both the initial claims data and anecdotal evidence still indicates that we are still confronted with fairly tight labor markets and labor markets which are still putting some pressure, if perhaps not accelerating, on the overall wage structure, an issue I'll get to in a minute.
It is true that we're having a fairly pronounced negative impact on the manufacturing area, but this is not a psychological issue. It's basically the result of the arithmetic of our net real trade balance falling. And indeed, we're displacing parts of domestic demand effectively with imports. I mean, in fact, the only statistic in the last three weeks which really shows -- I mean, a real statistic, as distinct from the financial data -- the only statistic which shows a real dip is production, and that's one of the reasons -- we haven't had that weekly, so there's a lot more available.
But the point at issue here is that we do not yet see the impact of what is this whole looming, sort of somewhat scary psychology which occurs as a consequence of all of these data. And if you read the newspapers in the morning, which are filled with all the awful stories of what is happening, one gets the impression that the economy has collapsed and we might as well all go home and go fishing or something like that.
The truth of the matter is that we are still seeing a fairly significant continued momentum that is occurring as a consequence of previous major expansions in specifically the wealth effect and a whole variety of areas related to this extraordinarily extended bull market. It is the case that we are beginning to see anecdotal evidence because of the increased perception of risk that some capital projects are being scaled back a little it, delayed. We do see a flattening but very little deterioration in non-defense durable goods orders, which essentially have over the years matched capital appropriations.
To be sure, orders for non-residential building have been going down rather appreciably in -- oh, since close to the beginning of the year. And that is something which is really, in a sense, unrelated to this new episode. What is related and which is going to be relevant is that this risk aversion is showing up as increases in equity premiums and the cost of capital for capital investment. If it's going to filter its way into our econometric models which spew out presumed outlook for the rest of the next two years, the key area that is impacted is clearly in the cost of capital.
And what we have seen is a fairly dramatic increase in the cost of equity capital. And indeed, if you believe these really quite non- believable expected earnings-per-share increases over the next three to five years, which have not changed through all of this particular period, then you get a pronounced increase in equity premiums; that is, the premium in the market for equity issues over the riskless rate of return. And that has risen quite dramatically in the most recent period.
Obviously, it's the result of the stock market going down dramatically while neither riskless interest rates nor expected long- term earnings have changed all that much. The decline since the stock market peak probably reflects, offset in part by some increase in capital gains -- let me put it this way. The net wealth effect for all financial variables -- for all financial products, I should say -- is about one and a half trillion dollars since the stock market peak. This is a calculation based on trying to make judgments of what the losses are in equities by U.S. holders, less the relatively modest but not insignificant gains in bonds.
Some of this net is coming out of pension funds, which, since they have no liabilities, theoretically don't have any really important impact. But the biggest share by far is in households and mutual funds net. And that's one of the reasons essentially is that the concentration of equity instruments relative to debt instruments in these areas puts a greater weight on equity decline as distinct from the rise in overall, essentially private bond capital gains, which, remember, for investment-grade issues, still reflect some decline in interest rates, despite the huge increase in spreads.
The interesting thing about all of this is that we're talking about a number of one and a half trillion dollars. It's got to show up somewhere. This is not a zero-sum game. It is essentially a measure of what the perception of one's assets are, and we obviously are seeing a number of indications in losses, in corporate income statements. But we're bound to see the major impact in personal consumption expenditures and housing. I don't know how all of your models essentially capture the so-called wealth effect, but it's pretty obvious for any type of evaluation that the sharp rise in stock prices over the years, up until the summer, has been a major factor galvanizing consumer expenditures and holding up housing sales to levels which seem beyond anything the demographics would have suggested could be supported.
So what we are looking at is an economy which is far more driven by balance sheet and asset values. And just as we had a virtuous cycle going up on the up side, we have seen that virtuous cycle disappear as we run to the other side. And as a consequence of that, one can be inclined to say that there's a major contraction in the process. I must tell you that if I read some of the reports in the newspapers about all the things that are going on -- and remember, good news is really page 35; bad news is right smack up front. For those of you who want to get an objective view of what is going on in the world, it's probably wise to put your newspapers in your in box and leave them there for about a week, and then you can read them, you know, a week later. You'll probably get a more objective view of what is going on.
But the truth of the matter is that we have got an economy which, as of now, just looking at it as economists like to do, with the real variables, which is really still quite an impressive sight, the labor markets are, to be sure, tight. And some of the anecdotal evidence is that they've actually tightened slightly recently. But the data with which we are dealing in this area is really quite problematic.
For those of you who know, when you take a look at the household data on employment on the one hand, and payroll employment on the other, payroll employment is going straight up and household employment is going straight out, flat. And the near 600,000 increase in the published payroll number -- I'm sorry, published household number -- on Friday sort of closed the gap only slightly. That gap is not explainable definitionally. In other words, if you take all the different conceptual adjustments, it's still there. And it is a fundamentally different view of what's going on in employment. And one can argue that with the sample size of the household survey, that the monthly data indeed, or the changes in the monthly level, have a very high variance associated with the calculation. But when you pool all the numbers over a period of time, the variance falls very dramatically, and it is very hard to argue what is the difference.
I mean, we've made adjustments such as we presume that the underlying population data are being underestimated by inability to get the right data on illegal immigrants. And obviously if you're doing a sample survey, the universe against which you're making the calculations is crucial. And if the population growth is being underestimated, then clearly so would employment. Having done all of that, we just do not explain this very material difference.
What is important about this, obviously, is a judgment of how tight these markets are. But having said that, if you take a look at all of the variables on average hourly compensation, which is what we are critically interested in while we're looking at the tight labor markets, we can't come out with a clear picture.
The reason I say that is if you look at the employment cost index, it looks relatively benign. But there has got to be, in the data for earlier this year, a significant amount of real wage increase which doesn't get picked up, largely because a lot of employers gave wage increases not across the board, but basically by phantom increases in job classifications. In other words, a lot of people were elevated to better titles, with new levels of income, and it was only a means to get their wages up without giving to everybody else. So the ECI doesn't capture that, and it probably is deficient.
The average compensation data are tough in the sense we're not terribly clear what the mix is. But from the point of view of doing an evaluation of inflation we don't care, because ultimately it is unit labor costs which matter. And here the data are probably better than any of the measures of average hourly labor costs. And here we're getting, from the non-financial corporate area, which I think is the best thing to use -- I have problems with the non-corporate data system that's implicit in the national incoming product account. So in order to get a sense of what's really going on in various relationships, I find that dropping out the non-corporate sector is usually a far better way of knowing what's going on.
In any event, unit labor costs in the last year in the non-financial corporate area are up about 1.2 percent. But that is, to a very significant extent, being offset by unit non-labor costs; capital consumption allowances, indirect business taxes and interest. Indirect business taxes are largely reflected in the tax cuts in the state and local governments, which is not an immaterial issue in the corporate consolidated income statement.
Nonetheless, because prices have been essentially flat in this area -- and it's true both for manufacturing and non-manufacturing -- I'm sorry, manufacturing and non-manufacturing, non-financial corporate business -- the price levels have just been steady. And these are the implicit data that you get out of the GDP accounts. And as a consequence of that, we have seen some very modest pressure on profit margins, which have been going down over the last two or three quarters, really.
This has been a very unusual phenomenon, because it's been my experience over the years that when we are dealing with falling or moving profit margins, moving down in the context of still fairly good business activity, that price pressures tend to become difficult to deal with. It's not happening now, and the major reason, at least from a strict statistical balancing, is that the productivity data are behaving far better than I think any of us expected.
And remember, one of the problems that we have in measuring productivity is whether or not you take it from the income side -- that is, gross domestic income per hour or gross domestic product. And the major difference, of course, is its very dramatically widening statistical discrepancy.
I've always been of the opinion that if you really want to know what's happening with productivity, you have to do it from the income side because the data on profits are good, the data on prices are reasonably good, and all the unit costs are a better measure to get the fallout; in other words, to get -- if you have unit labor costs and some varying judgments of what's going on in compensation per hour, obviously productivity comes out as a residual. And it is a far more consistent story about what's going in the world to reconcile the fact that despite increasing compensation per hour, accelerated compensation per hour, prices have done nothing. The only way in the full accounting system that that can happen is that productivity is rising -- I should say productivity growth has accelerated in recent years, and indeed, all the collateral evidence does suggest that that is indeed the case. And even though we don't have data for non-financial corporations for the third quarter, preliminary stuff that we put together for the manufacturing part of that shows that the third-quarter productivity numbers are still moving at a fairly good pace.
So what we have is a very unusual phenomenon in which we are seeing price increases, or I should say pricing power on the part of American business virtually non-existent in the context of a fairly substantial set of pressures where you would have expected long ago prices to have accelerated. I don't recall moving this far into a business cycle expansion with the price level, inflation rate, staying down, and by some measures even continuing to fall. I do not remember this type of inflation pattern existing for this long into a business cycle pattern.
Thus, having spoken a good deal longer than I intended to, just let me sum up by saying the situation in the American economy has become very fluid for all the reasons I've been trying to outline. To date, to repeat, the economy has remained in reasonable shape, with good growth and low if not declining inflation as a crucial element with respect to the existing state of affairs.
But we are clearly facing a set of forces that should be dampening demand going forward to an unknown extent. In particular, a marked shift in investor psychology away from risk and toward liquidity and safety has exacerbated the problems in foreign markets, where deflationary forces remain virulent and have spread to the financial markets in the United States.
We do not know how far it will go or how much it will affect consumer and business spending here at home. This is a time for monetary policy to be especially alert.
I thank you very much. It's been a pleasure being with you this morning.