Questions for Vincent? If not, let me start off by noting that recent commentary—not only the anecdotal discussions around this table but also the reports from our various Beige Book contacts, directors, and the like—have all pointed to an extraordinarily expansive economy and to the emergence of underlying inflationary pressures. On the other hand, when we look at the basic data for the recent period, we see a series of downward revisions. This kind of disparity usually is not found in the same set of information unless we seasonally adjust the anecdotes, and I’m not quite sure how one goes about doing that! With regard to the data, the latest revisions for GDP and new orders have been down, initial claims have definitely stopped declining, and this morning we learned that the latest data on purchase originations from the Mortgage Bankers Association fell fairly significantly. There is something very interesting going on, and in time we will figure out what it is. I am sure that our favorite ex-member of the Board of Governors, in addition to his three rules that were mentioned earlier, has a fourth rule that will suggest how that is done.
I think that the surprises, strangely enough, are not in measures of inflation. The big surprise is in profit margins. They have surged at a pace we haven’t seen in more than twenty years. The arithmetic of profit margins tells us something very interesting. At the level of about 12 percent for nonfinancial corporations on a consolidated basis, the arithmetic essentially indicates that a 1 percentage point increase in profit margins over a one-year time frame elevates the rate of inflation by 1 percentage point if unit costs are flat. And, indeed, if profit margins were to increase by 1 percentage point per quarter when overall unit costs are flat, we would end up with an additional 4 percentage points of inflation. In fact, starting in 2003, profit margins rose, on average, close to 1 percentage point per quarter; fortunately, this was partly offset by a drop in overall labor and nonlabor unit costs. Looking at the monthly patterns that are implicit in the income data for this year, it appears that profit margins have been essentially unchanged thus far through the month of May. What we are observing for the early part of this year is a little upward movement in unit costs and price increases that do not reflect further significant increases in margins.
Profit margins, as has been pointed out, are at the upper edge of the range beyond which historically they have not tended to rise. The reason is largely that, if they were to rise further, the result would be to contravene the long-term stability of the ratio of profits to national income. Indeed, what we are beginning to see, I suspect, is an effort to harvest these very large profit margins. In that regard, we are beginning to get a significant expansion in the utilization of labor and capital, the effect of which is to cause unit costs to move up. The way that apparently is occurring at this stage is largely through the effects on productivity, as we were discussing earlier.
The compensation of employees does not appear to have accelerated materially, although it is unclear what the last couple of months are showing. It is fairly evident that the real issue does not relate to structural productivity growth but to actual growth. What the data for April and May are suggesting is a very significant slowing in productivity and, indeed, a marked pickup in unit labor costs. Unit nonlabor costs actually, if anything, are reflecting a general decline in fixed costs as the economy’s expansion spreads fixed costs over a wider base. So, what we are observing through the first five months of the year may be the beginning of a shift from the falling overall unit costs observed last year to a pickup largely because actual productivity growth, which was growing at a pace well above that of structural productivity, has moderated.
We are moving from the behavior of inflation as wholly the consequence of rising productivity, and therefore rising profit margins to increasing evidence of stabilizing margins, with some evidence that the stabilization is being induced by some slowing in productivity growth. In turn, unit labor costs are beginning to move up. It now looks as though increases in unit labor costs are somewhat above 2 percent, maybe between 2.5 and 3 percent at an annual rate in the most recent month or two.
Incidentally, the underlying inflation rate, using the market-based core PCE measure, is at an annual rate of 2.1 percent for April and May, and for the latest three months it goes up to 2.4 percent. So core PCE seems to have moved into a channel that appears to be slightly above a 2 percent annual rate, but that is in the context of still very high profit margins, which could very readily begin to decline and bring the inflation rate down.
It is difficult to judge at this stage how to play this; but it is very obvious, looking at the data, that what we are interpreting as cost pass-through is essentially an increase in profit margins. The reason is that, if final prices remain unchanged and costs increase, that is another way of saying that the cost pass-through does not exist and profit margins contract. On the other hand, if there is a full cost pass-through, of necessity final prices must rise as profit margins remain intact. Therefore, profit margins are probably a good indicator on a consolidated basis of the degree of cost pass- through, which we have seen happening at a fairly pronounced rate. As I said before, we’ve had profit margins rising at the fastest pace in twenty years, and that’s another way of saying that we have moved from a price discounting environment to one of full pricing power. Remember, it’s competition that enforces the tendency of real wages eventually to parallel productivity growth. If that is the case, algebraically we are talking about a constant share of profits to national income in the long run.
Our major outlook problem at this stage gets to the issue of trying to estimate total unit costs and more specifically, on a consolidated basis, unit labor costs. The latter, as you know, compose two-thirds or more of consolidated costs. Here we have a very interesting set of numbers. We know, as some have commented, that average hourly earnings have really been quite contained and, indeed, are running below 2 percent at an annual rate. Wages and salaries per hour that are implicit in the personal income data are running significantly higher. And I have the data here if I can only find them in my usual well-organized filing system. [Pause] I have them now. The table with these data was sitting where it was supposed to be, and therefore I couldn't find it!
In the first quarter, wages and salaries per hour were rising at a 3⅓ percent annual rate and were increasing at a somewhat higher pace in the April–May period. What that tells us is that the average hourly earnings of supervisory workers, which are not published separately, must be rising fairly significantly or, alternatively, the way we usually term it is that the skilled–lesser skilled differential is continuing to open up in the wage market. Another way of putting this is that average hourly wages of college graduates are rising faster than those of lesser educated workers. Now, this obviously raises questions about the meaningfulness of the aggregate unemployment rate as it affects this total system. As far as I can see, there does not appear to be anything in the monthly data that suggests that overall wages and salaries per hour are accelerating in any material way. Obviously, our analysis is running into other labor income and supplements that are somewhat stronger, but the problem at this stage does not appear to be an acceleration in wages. The basic problem is an apparent slowdown in productivity that is creating the possibility of increased unit costs that will either affect profit margins or get passed through into higher prices.
All that said, the data as best I can judge them at this stage appear to be consistent with flat profit margins and units costs that are rising at an annual rate in the area of, say, 2½ percent, or alternatively with gradually declining profit margins, which is what one would expect in this environment. That would bring the core inflation rate down to a level of about 2 percent with no real evidence of acceleration going farther out.
Without getting into the details of the data we have constructed for nonfinancial, non-energy corporations, if we consider the impact of oil, obviously we can see fairly significant increases in unit costs stemming from higher energy costs. Incidentally, when we perform this type of consolidation, we also see that unit import costs are rising and then flattening out. In any event, gasoline prices are coming down quite considerably. One of the reasons is that margins both at refineries and at marketing operations are running well in excess of what one would ordinarily expect, given the refinery acquisition cost of crude oil. In fact, the data point to an excess margin of something like 20 cents a gallon, and if that comes down, it will have the effect of damping inflation psychology.
I might reiterate what has been pointed out previously, namely that measures of long-term inflation expectations are remarkably contained. Indeed, the measure that I find most useful in that regard is the forward rate on one-year maturities nine years out. In other words, it’s the yield associated with the last one-year tranche of a ten-year note. That yield has barely moved. Well, actually it has moved; it has moved down in recent days and is now definitely below its level in August 2003. In short, there is no evidence in these data that confirms some commentary to the effect that market participants think the Fed is behind the curve. If markets thought that, we would see some very significant acceleration in inflation expectations and increases in interest rates. But there are no real indications in the data that markets perceive us to be behind the curve. The question is what we should do to make sure that continues to be the case. For one thing, I think we have to remember that how we move forward is a function of how we evaluate balance sheets in the economy. If everybody held one-day paper on the asset side of their balance sheet and overnight maturities on the liability side, then whatever we wanted to do would have zero effect on the balance sheets and the income statements. That is a critical issue of the financial sector. But because such a maturity structure does not generally exist, we have to be careful not to move too fast and risk destabilizing the financial system by enforcing maturity mismatches and other idiosyncrasies in the financial system.
Remember, while the carry trades have come down and unwound to a significant extent, as Vincent has pointed out, so long as the markets perceive that we will move only on the dates of our scheduled meetings, we can be absolutely assured that tomorrow the carry trade will come back for six weeks because one leg of the carry is thought to be guaranteed. So, in effect there is no way we can create market expectations that put us in a position of having zero effect at the end of the day when we eventually move.
So where we are at this stage in my view is clearly that we are going to move, or I hope we’re going to move, 25 basis points today. We have to leave open the possibility that we may wish to move 50 basis points in August. Even so, I would suggest that our post-meeting statement retain the notion that our further removal of policy accommodation will be at a pace that “is likely to be measured.” But I also think it would be wise on our part to follow that sentence with a new sentence that essentially nullifies it but doesn’t eliminate it. That sentence would say, “Nonetheless, the Committee will respond to changes in economic prospects, as needed, to fulfill its obligations to maintain price stability.”
If we decide to go ahead in that way, we will leave open the possibility of dropping both sentences at our next meeting and then dealing strictly with the issue, as Vincent pointed out, of the balance of risks statements. I’m not saying that we need to, or should, but we open up the possibility. I would be very concerned were we to drop the word “measured” this time largely because the markets have adjusted to it. They are in fact behaving as though they know where we’re going today and are in the process of adjusting. If policy expectations continue to work out that way over time, the market effect will be virtually nonexistent. Now, that is not going to happen as we certainly know. Even so, I think we have to be careful not to take any action without significantly foreshadowing what we are going to do one way or another. We have the forthcoming semiannual testimony to calibrate how we want to be positioned for the August meeting. And if it turns out in the next three weeks that we have significant acceleration in inflation pressures, I would be prepared to suggest that 25 basis points is not necessarily what we had in mind. But we don’t need to say that today, and I think it would be a mistake to do so.
I conclude with a recommendation that we move up our target for the federal funds rate by 25 basis points and that we issue the press statement that Michelle will now circulate. It refers to the possibility that we may move at a faster pace, but we will do so only if necessary. We need to be cautious because, if we move at a faster pace, we may be taking risks that we do not need to take. [Pause] Is that statement acceptable to everybody?