WSJ Factual Error

From the top story in today’s Wall Street Journal:

At one point during the day, investors were willing to pay more for one-month Treasurys than they could expect to get back when the bonds matured....That’s never happened before.

Actually it has happened before, not in the easily available data, but it has happened – in 1938 (and apparently several other times between 1935 and 1941).

My only source is a talk by Paul Samuelson, for which I cannot even point to a transcript, but I’m confident that primary sources will bear me out. I’m too lazy to go check old copies of The Wall Street Journal on microfilm, but take my word for it.

It’s actually pretty obvious if you look at the monthly data from the Fed. For example, in February 1941, the average yield on 3-month T-bills was 0.03 percent. Considering how the yield fluctuates from day to day and from hour to hour, it’s impossible to believe that it was not negative at certain points during that month. (Technically the Journal was referring to one-month bills, but it’s a safe assumption that, if 3-month bills were selling above maturity value, so were one-month bills for at least part of the time.)


UPDATE: Paul Krugman makes the same claim (hat tip: anonymous commenter)....and I continue to believe it is wrong. I'm not sure his claim is independent: he may have gotten his information from the Journal, or they may have gotten it from the same source, which I hope they will cite so we can follow it up and judge its reliability.

UPDATE2: Reuters and the AP, both citing Los Angeles-based Global Financial Data, report that the last time the 3-month T-bill was at or below zero was January 1940. (Could it merely have been "at" zero? It seems unlikely that the bid would have stopped at exactly zero.) Another AP report says that demand sent "the yield on the 3-month Treasury bill briefly into negative territory for the first time since 1940." Friedman and Jacobson, in A Monetary History of the United States, 1867-1960, say in a footnote that "yields on Treasury bills were occasionally negative in 1940." (Apparently my "obvious" conclusion about 1941 was not correct, buy my main point stands.)

Good News about US Inflation

Everybody’s getting so freaked out about the latest CPI report, and I don’t understand why. The supposed bad news – gargantuan increases in energy prices, substantial increases in food prices and in import prices – is last month’s news. You could have gotten most of this information by looking at commodity markets and foreign exchange markets a month (or two or three) ago, and, as Dean Baker points out, you could have gotten it with even more precision by looking at the latest report on import prices. I don’t see any real bad news in this report.

Over the past 3 months, energy prices have risen at a 33.8% annual rate. That’s dreadful, but it’s not news. Food prices have risen at a 4.3% annual rate. Kind of ugly, but also not news. In fact, the 0.3% November increase in food and beverage prices is kind of tame, considering. Transportation prices rose at a 14.4% annual rate over 3 months – that’s pretty much redundant information, since I already mentioned energy prices. Medical care prices rose at a 5.2% annual rate, but that’s not unusual.

The one thing that is both unusual and unexpected is the 0.8% monthly (4.1% annualized over 3 months) increase in apparel prices. But if you look at the 12-month change in apparel prices, it’s still down (-0.4%). While apparel prices may not fall as quickly in the future as they have in the recent past, I for one do not believe that we have suddenly entered a new regime during which apparel prices will be rising by 0.8% – or even 0.3% – every month. The fact that there was a blip in apparel prices in November – and not even enough to get the core inflation rate for November above 0.3% – is hardly a significant piece of bad news.

The last thing in the report that people may have found troubling is the 0.4% increase (3.6% for 3 months annualized) in housing prices (meaning mostly rent and owners’ equivalent rent). But are you really worried about housing costs – with huge inventories of unsold houses in most parts of the country? There is probably a temporary problem in the rental market, because people are getting foreclosed on and pushed into the rental market, and the properties they vacate are remaining vacant for a while. This problem may continue in coming months, but one can’t reasonably describe this as fundamental upward pressure on housing prices.

What’s left? The “other goods and services” category did rise by 0.3% in November – more than one would have hoped – but the other two categories, “education and communication” and “recreation,” each rose by only 0.1% – less than one would have expected. All in all, not a troubling report, unless you haven’t been paying attention to the news over the past few months.

If you’re obsessed with the aggregate inflation rate, you’re welcome to be horrified that “US inflation jumps to 4.3%” – as the headline on the weekend Financial Times declared, and you might also be unhappy with the 3-month core inflation rate (2.6% annualized). But when you look at the 12-month core rate of 2.3%, don’t be upset (like the Financial Times) that it is “higher than the Fed’s upper limit of 2 per cent.” That upper limit applies to the core personal consumption deflator, which is a different index and typically runs about 50 basis points (or anywhere from 0 to 100 basis points, depending on whom you ask) below the CPI. 2.3% core CPI inflation is not a problem.

Not Hitting the Fan Yet

The dollar is now at a record low against the Euro, down more than 20 percent from its peak in 2002, down so low it’s about equal to a Canadian dollar.

So begins Robert Reich's blog post to which I referred on Friday. By my arithmetic the situation is even more extreme, with the dollar having lost about 40% of its value in Euros since the 2002 peak. But should the drop in the dollar between January 2002 and September 2007 really be a cause for concern in the US?

I don't see why. Most of the drop happened in 2002 and 2003, and the remainder has happened slowly over the subsequent four years, with a slight acceleration over the past few weeks. That's water under the bridge. If that drop in the dollar were going to cause inflation in the US, or to cause a drop in US real incomes, it would already have happened.

But incomes (on average) have continued to rise, and as for inflation, I think the figures in the August Personal Income and Outlays report (see Tables 9 and 11, along with historical data available here) should put to rest any immediate concern. Looking at my favorite inflation indicator, the market-based personal consumption deflator excluding food and energy, we have the following annualized (logarithmic) growth rates:

1 month    1.15%
2 months   1.34%
3 months   1.39%
6 months   1.18%
9 months   1.67%
12 months  1.61%
18 months  1.89%
2 years    1.87%
3 years    1.80%
4 years    1.70%
5 years    1.57%
6 years    1.56%

None of this suggests that inflation has yet become a problem at all. Given the presumed target of 1.5%, the recent data would even make a better case for worrying about deflation than about inflation. (Remember that food and energy prices can be quite volatile, so one shouldn't ignore the risk that, for example, a reversal in the oil market could send the full deflator quickly into negative territory.)

Though I have worried in some recent posts about the potential for a drop in the dollar to force the Fed into a stagflation regime, it needs to be emphasized that this worry is about the future and not about the present. If the s___ is going to hit the fan, this is actually a pretty good time for it to hit.

No Help

Today the Conference Board announced that in August, its Help Wanted Advertising Index fell for the eighth straight month, by a larger-than-expected 8% (using the integer numbers that the Board reports), to its first new all-time low since 1958. ("All-time" in this case means since 1951, when the series began.) This means that major newspapers in the US carried fewer help wanted ads than they did at the depth of the 1958 recession. This is despite the fact that the scale of the US economy, by any measure, has more than doubled since 1958. (In particular, there were about 51 million employees on nonagricultural payrolls in mid-1958, compared to 138 million today.) It's also despite the fact that many of the smaller local newspapers, which used to compete with the surveyed major newspapers for advertising, have gone out of business or been absorbed by the major newspapers during that time.

Part of the explanation is the emergence, over the past decade, of online recruiting as an alternative to newspaper-based recruiting. This is only a partial explanation, though, and it only helps explain the general downward trend in newspaper help wanted advertising over the past decade, not the specific decline this year. Online help wanted advertising, according to the Monster Employment Index, has been (uncharacteristically) roughly flat since March (over which time the Conference Board index* has fallen by 23%), though the Monster index has doubled between October 2003 and today. But in 2003, online help wanted advertising had only 19% of the market share vs. newspapers, so even the doubling doesn't begin to make up for the 36% drop in newspaper help wanted advertising over the same period.

Also this morning, ironically perhaps, we got the news that new unemployment claims fell below 300,000 in the week ended September 22, for the first time since May. There has been no pattern of rising unemployment claims this year while help wanted advertising has been steadily falling. Apparently, the US is experiencing a new kind of economic weakness -- one where the trailing edge of the labor market keeps steady but the leading edge stops advancing.


*UPDATE: Just to clarify, I'm referring to the index of newspaper help wanted advertising discussed in the previous paragraph. The Conference Board also has an index of online help wanted advertising, but I haven't started to follow it yet.

Business Contractions by Month-of-Decade

I went through the last 15 decades (actually starting in 1855, when the NBER dating starts, and ending in 2004 to give an integer number of total decades) and calculated the frequencies of business cycle contractions (using the NBER dates) for each month of the decade. For example, January of the “zero” year is the first month, February of the “zero” year is the second month,…,January of the “one” year is the 13th month,…,January of the “two” year is the 25th month, and so on. (You can click on the picture to see a larger version.) These results are purely empirical and may not mean anything, and I’m not sure how to do statistical tests on them, because I find it hard to make a rigorous distinction about what is a sensible hypothesis to test.

But there is at least one thing that seems hard to attribute to chance: the highest frequency of contractions is just after the beginning of the decade, and the lowest frequency of contractions is just before the end of the decade. Since 1855, there has been only one contraction that included the second quarter of a "nine" year (1949). Whereas in 10 out of 15 of the "zero" years, at least November has been a recession month. And at least 3 of the remaining 5 are not very impressive: November 1940 wasn’t a contraction, but the nation was still recovering from the Great Depression, so it wasn’t exactly business as usual either; November 1980 technically wasn’t a contraction, but it was part of a very short expansion sandwiched between two contractions; November 2000 wasn’t a contraction just yet, but as I recall, the winds of recession were in the air. (Maybe somebody who knows the history better than I do will have stories about 1880 and 1950.)

If we take this chart seriously as a prediction for the future, it has an interesting implication for the present time. There is a peak late in the “seven” year, after which the chance of contraction drops rapidly. That suggests that right now is a critical time to avoid recession, and if we can get through the next few quarters without one, we can expect smooth sailing for the remainder of the decade.

The chart may also be depressing for Democrats and heartening for Republicans. During the first few months of President Obama’s term, it will look like he inherited a strong economy. Then once the unimpeded Democratic initiatives are implemented, it will appear that they are destroying the economy. The trick, though, for a Democratic president, will be to reduce initial expectations and have people expect a longer-run payoff. That seemed to work for Reagan, anyhow. Alternatively, one could hope that the recession will be as shallow as possible and the recovery as strong as possible, which worked for Nixon. My advice to the next president is to figure out now a way to blame 2010 on Bush and then save your real economic ammunition for 2011 and 2012. With any luck, it could be the second Clinton boom. (As per tradition, Clinton will be facing a Republican congress.)

Productivity Slowdown?

Today’s US productivity data were rather unpleasant (see Mark Thoma’s post and the links therefrom), but I’m far from ready to throw in the towel on this one. By my reckoning, the slowing of productivity growth still falls in the category of cyclical variation, and the long-term trend of productivity growth remains at least as high as it was during the late 1990s.

With the latest data, I’ll repeat the analysis from the last time I posted on this topic. (Click on the “productivity” label link at the bottom of this post.) The premise is that productivity goes in cycles, related to, but not necessarily coincident with, the business cycle. (For example, the 1990s business cycle contained two distinct productivity cycles.) I’m measuring cycles from peak to peak, and I’m making the conservative assumption that we have not yet reached the peak of the current cycle. (Obviously we are long past the peak.) Here are the average annual growth rates of business sector productivity:

1.4%  1986q1 to 1990q2
1.6%  1990q2 to 1994q4
2.5%  1994q4 to 2000q2
2.6%  2000q2 to 2006q4

An alternative, less conservative, approach, is to measure from trough to trough, with the assumption that we are currently at a trough:

1.4%  1987q1 to 1991q1
1.6%  1991q1 to 1995q3
2.6%  1995q3 to 2001q1
2.8%  2001q1 to 2006q4

Taken together, these results suggest that the current trend is approximately the same as it was in the previous cycle.

Help Wanted Advertising Stabilizes

US help wanted advertising, which dropped precipitously during the spring and summer of 2006, appears to have stabilized during the fall. The Conference Board Help Wanted Index was essentially flat from August through November (the latest report), at a level about 20% lower than where it was a year earlier. (Taking online advertising into account would mitigate the decline, but given the normal growth in help wanted advertising over time due to a growing economy, the 20% figure gives a pretty good indication of the amount of deterioration.)

What this means exactly I’m not sure, but in my mind it tends to reduce the chance of a recession. On the other hand, it also tends to confirm the expectation of an economy too weak to please Main Street, and perhaps a little weaker than Wall Street would prefer.

OK, put the trumpet back.

Today's PPI wasn't so nice. But keep the trumpet handy anyhow. The PPI is not a very reliable inflation indicator, and the divergence from the CPI would seem to indicate that labor-intensive content is becoming less expensive, not more expensive. The data are quite consistent with the view that inflationary forces within the US are not a problem. I'm nowhere near ready to concede that the herald-on-call has his job security back.

Blow the trumpet!

Price stability is here.

The faithful knew it would come eventually, but the core inflation number in today's CPI report was a bit of a shock. (As to the day and the hour, only the Father knows...)
Sent via BlackBerry from T-Mobile

Did I read this right?

From Wednesday’s Wall Street Journal (p. A6, “Labor Costs…”):

The Labor Department reported yesterday that … unit labor costs fell at a 2.4% rate during the second quarter, rather than growing at the originally estimated 5.4% pace.

Quiescence

Lately I’ve been having trouble focusing attention on this blog long enough to do a full post, because…well, you know, there’s a lot going on, etc…. To keep the blog alive, I’m going to try more quick posts, which will hopefully still be good food for thought but not so eloquent or well-reasoned.

Here’s a thought for today: the current condition of the US labor market should not be described as “weak” or “strong” but rather “quiescent.” There aren’t many layoffs going on, and there isn’t much hiring going on. There aren’t many people looking for jobs, and there aren’t many businesses looking for employees. I think most statistics bear out this view of the labor market. I’m not sure what the implications are. I do have some thoughts, which I may address in a later post.

There is no productivity problem.

You’ve probably heard about the awful US productivity numbers for the 3rd quarter (productivity roughly flat, instead of rising as it usually does, and this on top of unimpressive productivity growth in the 3 preceding quarters). If you’re a regular reader of my blog, you shouldn’t be surprised, and if you agree with the argument I made here, you shouldn’t be upset. The 3rd quarter is part of a cyclical downturn in productivity growth. It’s perfectly normal, and in this case, even predictable, and it does not mean that the era of rapid productivity growth is coming to an end.

Let’s take a look at the numbers for the last 4 productivity cycles. And let’s stack the deck against the current cycle by measuring from peak to peak and assuming that we have not yet reached the peak. (Obviously, we are past the peak, and if we wanted to measure accurately from peak to peak, we would have to leave off the weaker numbers for the last 2 quarters.) OK. From 1986Q1 to 1990Q2, business sector productivity rose at an average annual rate of 1.4%. From 1990Q2 to 1994Q4, the rate was 1.6%. From 1994Q4 to 2000Q2 it was 2.5%. From 2000Q2 to 2006Q3 it has been 2.8%. Revisions to employment statistics might shave a tenth of a percentage point off that average and give us 2.7%. Not a rate that I would complain about.

It’s always possible that the longer-term trend of productivity growth has also slowed, just as it’s possible that the US is already in a recession. So far, though, we don’t have evidence that either of these propositions is true.

Do labor shortages indicate a strong labor market?

Sorry I haven’t blogged in so long. Lack of ideas, mostly. Plus I’m busy not having ideas on my day job.

Today I have an idea. A lot of anecdotal evidence seems to suggest that labor shortages are a problem in the US today. Yet statistics – note, particularly, help wanted advertising, which is near a 50-year low – show little evidence of strong labor demand. Is this inconsistent?

I don’t think so. If different kinds of labor are complements in production, then shortages of certain kinds of labor could actually make the overall labor market weaker. For example, if individuals with a lot of different skills are required to make a product, then a shortage of one of those skills will depress demand for all the other skills. If chefs are in short supply, the demand for waiters will go down. Is this the kind of thing that’s happening in the US today? It seems plausible to me that shortages of certain technical and managerial skills could be depressing overall labor demand. Just an idea.

Bad Wages

Dean Baker notes that the preliminary benchmark revision to US payroll employment data will mean lower productivity growth. Another implication, provided that there are no compensatory revisions in average hours worked or total compensation, is that average hourly compensation will be lower. For those who have been complaining about anemic wage growth, it looks like the situation may be even worse than they thought.

Another Problem with the Anti-Core Argument

Following up on my previous post, I want to make an empirical point about Caroline Baum’s argument. She argues that a long lag cannot explain the failure of non-oil prices to adjust downward in the face of rising oil prices:

You might counter by saying that the price of something else will fall with a lag, not simultaneously; that when gas prices go up the consumer doesn't immediately cut back on his non-oil purchases.

Let's go to the video tape. The consumer price index was running at about 2 percent year-over-year during the deflation scare in the middle of 2003. Crude oil prices were hovering near $30 a barrel.

Three years later, with crude oil prices hitting a record $78.40 in July, the CPI was rising 4.1 percent. In all that time, the price of something else should have fallen to offset the higher oil prices. The fact that it didn't means our friendly central bank was accommodating the oil-price increase, printing enough money to prevent that from happening.

But there’s a problem here (beyond the theoretical problems mentioned in my previous post). First of all, nobody disputes that the Fed was accommodating the oil price increases in 2003 and 2004. So it shouldn’t surprise anyone that non-oil prices failed to adjust during 2005 and the first half of 2006. At the time (2003 and 2004), the Fed had legitimate concerns about a soft economy, but the economy subsequently proved to be stronger than the Fed had feared. It turned out that the intentional easy money policy of 2002 and 2003 had been successful.

As 2005 went on, however, the Fed stopped accommodating the oil price increases, and in 2006, the Fed has even started pushing the other way. This is where the lag comes in. Because there really is a lag in the behavior of consumers, we won’t know until 2007 (or maybe even 2008) whether the oil price increases (which continued through 2005 and the first half of 2006) have reduced demand in the rest of the economy, as Ms. Baum would expect if the Fed were behaving itself. My guess is that, when all the votes are in, we will find that inflation has lost and that the Fed was right to ignore the direct effect of oil prices.

Bernanke vs. Baum

Fed Chairman Ben Bernanke argues that targeting core inflation is appropriate because, otherwise, the Fed would have to “force down wages and other prices quite dramatically to keep the overall price level from rising” in the face of rising energy prices. Bloomberg columnist Caroline Baum (hat tip: Mark Thoma) objects:

In order to offset the immediate effect of a rise in energy prices, the Fed wouldn't have to do anything. The price of something else would fall, all things equal, as consumers adapt to the constraints on their budgets (paying more for gas means less money for other goods).

I have (at least) four problems with Ms. Baum’s argument.

First, I have a basic philosophical problem with clauses like “the Fed wouldn't have to do anything.” What exactly is meant by not doing anything? I suppose they could hang “Gone Fishing” signs on the doors of all the Federal Reserve Banks and tell all the employees to go home and be with their families for a few months. No doubt this would succeed in getting prices down, but I don’t think it’s what Ms. Baum has in mind. It’s really quite arbitrary what course of Fed action is defined as being passive. Keep the monetary base constant? Keep the interest rate constant? Increase the monetary base at a 2% annual rate? At a 10% annual rate? Follow a Taylor rule? Ms. Baum’s whole argument is premised on the idea that there is some “default” course for the Fed. But who gets to define that default? No doubt she can define the default in a way that will produce, on average, the results she imagines, but someone else will define the default differently.

Second, let’s accept the default that she perhaps has in mind, maybe some ideal constant money growth rule, and let’s assume away all the uncertainties about velocity of money and potential output. Increases in the average price level are caused, according to the cliché, by “too much money chasing too few goods.” Ms. Baum assumes that “too much money” is the problem. In the case of energy, however, it is more likely “too few goods.” In other words, if the Fed were following a passive money supply rule, an adverse oil shock would cause the average price level to increase anyhow. If oil becomes scarcer but money remains equally plentiful, the value of money – in terms of some basket that includes oil and products made with oil – will go down.

Third, even if my first two objections didn’t apply, let’s just imagine that Bernanke is using imprecise language by casting the Fed’s behavior as active rather than passive. Let’s say it is not the Fed but the market that has to “force down wages and other prices quite dramatically to keep the overall price level from rising.” That’s still a bad thing, because most wages and prices don’t go down without a fight, and the fight usually takes the form of a recession. The Fed is willing – actively or passively, depending on your economic theory and your semantic preferences – to accept temporary increases in the headline inflation rate in order to avoid unnecessary recessions. That strikes me as not only good judgment, but also the clear duty of the Fed under its current mandate.

Finally, let me make a basic point about the nature of oil as a commodity: oil is storable. If the price of is expected to rise, there will be strong incentive (provided that storage costs and interest rates are not too high) to hold large inventories in anticipation of higher prices. In practice, we never observe huge inventories, because producers, facing expected price increases, choose to delay production, which causes prices to increase immediately. In any case, the nature of the oil market is such that, normally, the most knowledgeable people will never, on average, be expecting large price increases. Otherwise, those price increases would already have happened. If the Fed’s objective is to stabilize prices going forward, then the Fed is operating well within reason to assume that oil prices will not rise dramatically. It is taking the market’s judgment, effectively outsourcing the task of forecasting oil prices. Past increases in oil prices are irrelevant, and it is rational for the Fed to ignore them by using core inflation as its measure of past price growth.

Q2 Labor Costs, Revised

The big increase in US hourly compensation in the second quarter didn’t make sense when it was first reported. Now that it has been revised upward, it makes even less sense. The best explanation I’ve heard comes from Dean Baker, who suggests, based on the NIPA statistical discrepancy, that some capital gains (obtained, for example, via exercise of employee stock options) might be misclassified as compensation. (Conceptually, in the case of stock options, the compensation took place when the options were granted, not when they were exercised. Anything that happened to the value in the intervening time was a capital change rather than income, but the value of the options doesn’t show up on the income side of the national accounts until they are exercised.)

Employment: No News is Good News

Now that Calculated Risk has pointed me out as one who follows such things, I feel obliged to comment on the labor indicators that have come out over the past 2 days. Overall, they were a bit stronger than I expected, but a bit weaker than I think the market expected.

Yesterday, we had both the July Help Wanted Index from the August Monster Employment Index. The picture is mixed. The Monster index bounced back from the distressing drop in July, and my first impulse was to put up a post under the title “Friendly Monster” to contrast with my earlier “Scary Monster” post. On closer analysis, though, the bounceback was not as impressive as it first appeared. The index rose from 165 to 173. In percentage terms, that increase is a little bit smaller than last year’s July-to-August bounce, but a little bit larger than the one in 2004. Since last year was a year of substantial job growth, I would say this year’s bounce still certainly goes in the “good news” column, but it is not large enough to make up for the unexpectedly large drop in July.

The Help Wanted Index dropped to 32 from an upwardly revised 34 in June. This leaves it a point lower than the original June report, which was already coming on the heels of a dramatic drop from February to May. The index is getting closer to its record low of 24 set in April 1958, and remember that the labor force has more than doubled in size since then. Relative to the level of payroll employment, the July figure is actually the fourth successive record low. If we combine the July Help Wanted Index with the August Monster index (using a procedure described briefly here) and normalize by payroll employment, the result is just 1.6% above the record low set in May 2003, and 15.5% below the recent peak in February.

Then there is the Employment Situation report that came out this morning. It adds up essentially to no news, which is news to me, because I expected it to be weaker than expected. Payroll growth at 128,000 is right in the middle of the range of values that are typically given as being sufficient to absorb the expected population growth. If you think the long-term trend in labor force participation is accelerating downward, then 100,000 might be enough, but if you attribute the decline since 2000 to a weak job market, then something like 155,000 are necessary. I tend to lean toward the high end of that debate, meaning I think that the August growth was a bit weak, but there is room for disagreement. Interesting that construction employment was up. Given what builder confidence indicators are saying, I expect that will change in coming months. Other than that I haven’t yet found anything interesting in this report.

Big Drop in Perceived Job Availability

In the Conference Board’s monthly Confidence Survey, they ask respondents if jobs are “plentiful,” “not so plentiful,” or “hard to get.” I keep track of the percentages of respondents (as reported in the press release), particularly for the “plentiful” category. In August, it dropped by 4.2 percentage points, the biggest one-month drop since October 2001. In the past (since 1978, when the survey began to be conducted monthly), such large one-month drops have always been associated with recessions.

I wonder if this has anything to do with the declines in help wanted advertising that I observed here and here. (Do you think?) It is a bit of a puzzle, though, why perceived job availability held up so well until July.

Q2 Labor Costs

By most indications US labor costs rose rapidly (by recent standards) in the second quarter. But something is wrong with this picture. Ordinarily one would like to attribute an increase in the price of something (labor, in this case) to either a leftward shift in the supply curve (decrease in supply) or a rightward shift in the demand curve (increase in demand). In the case of labor, there is no evidence that either of these things happened; indeed, the evidence suggests that the demand for labor declined.

I have noted before the dramatic decline in help wanted advertising. It is not matched by the official data on job openings; the latter are roughly flat but certainly show no sign of increasing. New hires, however, did decline in the official data. And aggregate payroll employment growth slowed to an average of 112,000 per month (not enough to keep up with recent average labor force growth) from 176,000 per month in the previous quarter (which is comparable to the average for the prior two years).

This apparent decline in the quantity demanded might suggest that the labor supply curve shifted. That is, there were fewer workers available at any given wage, so firms both raised wages and reduced hiring. But the direct evidence does not support this hypothesis. Labor force participation actually increased throughout the quarter (and also in July, so one cannot reasonably attribute the increases in May and June to sampling error). There was no increase in the rate of unemployment due to voluntary quits. There was a slight decrease in the overall unemployment rate, but the decrease was reversed in July.

Another explanation for rising wages might be an increase in inflation expectations, or a correction for price increases that had already occurred. The argument, I suppose, would be that employers deliberately raised wages because they realized that, given higher prices or higher expected inflation, they would need to do so to retain their desired workforce. Absent an actual increase in quits, however, I find this explanation implausible. If employers were so scared of quits that didn’t actually happen, why did they simultaneously reduce their help wanted advertising? If they decided to spend less on recruitment, why would they be willing to spend more on wages?

My guess is that the increase in labor costs will turn out to be a statistical illusion. Aggregate compensation per hour is reported to have risen at a 5.1% annualized rate for the business sector, but this number is subject to considerable revision. The employment cost index, which corrects for shifts across industry and occupation, shows an annualized increase of 3.6%, still high by recent standards but not quite as troubling. When incentive-paid occupations are excluded, the increase drops to 2.8%, which is roughly the trend growth rate of productivity.