Recently in Economics--Labor Market Category

James Pethokoukis writes:

There is an economic downside to trying to keep people in their homes, even if underwater. Mike Feroli over at JPMorgan points out the Osward Hypothesis: "Higher homeownership rates increase the natural unemployment rate because it reduces geographical mobility and the ability to move someplace else to find a job."

Ceteris paribus, perhaps.  But ceteris is seldom paribus, and it is certainly dangerous to paint with a broad brush.  What is the relevant market?  Both housing markets and labor markets are regional to a very large extent.  Different markets are linked together in complex ways.  Examples abound of areas where home ownership is high and the natural rate of unemployment is low and vice versa.

The real problem today, of course, is a lack of liquidity.  People can own homes and still be mobile if a house can be sold in a reasonable amount of time.  It's not the rate of home ownership that matters so much as the ability of individuals to enter and exit.

Right now, exit from the housing market is more costly than usual.  As a result, people are likely to be less mobile and the natural unemployment rate may very well rise.

However, Pethokoukis and Feroli are only telling half of the story.  Policies designed to keep people in their homes may increase the unemployment rate, but only to the extent that such policies make voluntary exit from the market more difficult or costly.  Policies designed to keep people who are still employed (and likely to remain employed) in their homes with a minor modification to the loans are not going to have much of an effect on the natural rate of unemployment.

On the flip side, policies that allow people to voluntarily exit without any cost or penalty at all do pose a moral hazard problem.  I can imagine some creative yet politically and practically unrealistic policies to try to address this and get the incentives right, but I'm not confident that what we are going to get is going to get the incentives right.

But the point is that this is not an insoluble problem.  Furthermore, don't fall for the claim that home ownership is the problem.

Barely. Here's an update of the graph I created last month. Payroll figures continue to be ugly. The labor market is now deteriorating at a faster pace than the 1981 recession.

Brad DeLong thinks we need a bigger stimulus. I'll credit him for consistency in that he has argued for a larger stimulus all along. However, I don't think that today's data should really change anyone's opinion on what is necessary or advisable. I think we all knew when we went to bed last night that the morning news would not be pleasant. Next month's news will not be pleasant either, but I've already built that into my expectations. I am expecting continued losses but perhaps not the 600,000 numbers that we've been seeing lately. I'm hoping for under 500,000. That would be encouraging, but I'm not supremely confident.

The BLS reported today that 598,000 nonfarm payroll jobs were lost in January and the unemployment rate climbed to 7.6%. No one is surprised by this.

For anecdotal evidence, as I move around the region listening to and reading the local news, I am hearing more stories of what seem to be classic demand-driven layoffs. The ripple effects of the housing slowdown and the global demand slump are being felt in more places than they were 6 months ago.

Nearly 600,000 jobs lost sounds like a lot. Certainly, we haven't seen those kind of absolute numbers for a very long time. However, the labor market is much larger than it was in past recessions. So let's go behind the numbers and look at the job losses as a percentage of total employment.

Let me be very clear about the methodology behind these charts. I am calculating the cumulative net job losses during all post-war recessions as a percentage of peak employment near the start of the recession. In some cases, employment dipped slightly and rose again before taking a larger downturn. In those cases, I considered the peak to be the month before the larger downturn.

Though all of these downturns are associated with NBER recessions, not all of the peak employment dates coincide exactly with NBER business cycle peaks. (For example, the NBER dates a business cycle peak in November 1973, but employment did not turn down until July 1974, so I used the July 1974 date for this chart.)

As an example, employment peaks in July 1953 at 50,536,000.  Thirteen months later, in August 1954, a total of 1,711,000 jobs were lost, which was about 3.39% of peak employment (dark green line). 



It is a fairly busy chart, but we can see the current recession (orange) is very similar to the 1981 recession (light green) in terms of job losses as a percentage of peak employment. But we have had sharper downturns in percentage terms.

If you believe that this recession is not fundamentally different from other demand-driven post-war recessions, then a forecast of job losses continuing for another 6 to 9 months would not be out of line. Furthermore, looking at past cycles, one would expect it will be at least a year (possibly more if the recovery looks more like that after the 2001 recession) before employment reaches the previous peak. Personally, my expectation is that it will take 18 to 24 months (from now) to get back to the previous peak.

Finally, here's a version of the chart with the last four recessions (including the current one) that is a little less cluttered.  I've also included the 1957 recession as an example of a sharper downturn in payrolls.

UPDATE:  Spencer at Angry Bear looks at household survey (CPS) data and concludes that the current downturn is the worst in the post-war period.  Actually, according to the household data the 1953 recession is worse, but that's a minor point.  But there is another issue to consider.  I'm not sure what to make of the fact that the household survey is subject to less variation (in both directions) than the payroll survey, especially during the '70s and '80s.  It is widely acknowledged that the household survey is a less reliable indicator of the labor market overall than the payroll survey.  So I am understandably nervous about the possibility of understating the severity of the employment declines in the '70s and '80s if we were to rely on that data.

Nonfarm payroll employment down 533,000 in November.  (Full text)

No nice way to spin that news.  The things I'm reading lead me to expect December to be nearly as bad.

I'm giving a seminar in a bit, so I don't have time for more now.  Hopefully more time this weekend.

UPDATE:  David Henderson at EconLog points out that the 533,000 is a smaller percentage of the labor force than it was in the 1970s.  True.  That's one of the aspects of the problem I intend to think about this weekend.  Also, see the comments to his post for more discussion.  I think it is safe to say that in percentage terms the job losses from this recession are going to exceed the last two recessions, and rival those of the '70s and '80s.  More later.

Here's the monthly report from the BLS.

Let's start with the bad. The unemployment rate was up 0.2%. While I was expecting a little more of an increase, it is still a negative signal. Also, 51,000 nonfarm payroll jobs were cut last month. It is now quite clear that employment peaked in late 2007 and has been trending down ever since. Finally, as Brad DeLong points out:

The U-6 measure of unemployment--reported unemployed plus part-time for economic reasons plus marginally-attached workers all divided by the labor force plus marginally-attached workers--has risen by 1.1 percentage points in the past three months to its current level of 10.3 percent. It now stands 2.2 percentage points above its mid-2000s low, and is just a hair below the maximum reached in the 2001-2003 episode. As you all know, I have been unhappy with the conventional unemployment rate this decade--it has not been telling the same story as the other labor market indicators. U-6 seems to be a better fit to the overall state of the economy.

And by my book, U-6 is now telling us that we are in a recession.

I share his unhappiness with the conventional unemployment rate. It's a rough guide at best. And while I do think that U-6 is a useful indicator, what DeLong doesn't point out is that even today U-6 is a good point-and-a-half below where it was in 1994 (earliest year for which data is available in that series). That was a couple years after the end of a historically shallow recession (granted, it was still a period of labor market weakness).

So yes, some aspects of the economic situation are about as bad as during and shortly after the 2001 recession. Some are worse, and some are not as bad. When you consider the fact that the weakness in manufacturing is part of a longer term structural change, it looks less like a traditional recession even though it may soon (if not already) meet the NBER business cycle dating committee's criteria. If you're in some (though not all) types of manufacturing, this has been one long recession for a decade.

Other sectors of the economy, most notably education and health services, are probably wondering what all the fuss is about. They didn't feel the recession in 2001 and probably won't here either.

So that's the bad news.

But there is a silver lining. First, the average seasonally adjusted mean duration of unemployment dropped from 17.5 to 17.1 weeks. Next, and I think this is a crucial point, the percentage of the unemployed who are reentrants or new entrants to the labor market both increased. In fact, since March, the percentage of the unemployed who are job losers dropped from 53.7% to 50.2%. Reentrants have increased from 27.4% of the unemployed to 30.8%. New entrants have increased from 8.8% to 9.2%. Here's a chart for reentrants. (See also Table A-8 in the report.)

It should be pointed out that this is a pretty noisy signal over the time horizon of a few months to a year. This isn't enough to conclusively determine that a recession is over--if there was one to begin with. However, it is something that bears watching over the next few months. Roughly half of the increase in the unemployment rate this month was due to reentrants into the labor market.

Here's my bottom line. If we have turned the corner and are at or near a the bottom of this business cycle, then you're going to see unemployment due to reentrants rise further, which may temporarily (i.e. between now and the election) push the headline rate higher. If we have not yet turned the corner, then this may have been noise and the labor force participation rate may fall a bit in coming months. I can't say which it is yet. I don't think anyone can.

But the labor market is a lagging indicator, so if the rise in reentrants is not just noise, then that is good news indeed. But the U-6 number is bad news any way you slice it. So yet again we are left longing for more information. The economy remains at a critical point--teetering on the edge of a recession. Maybe in one. Maybe pulling out of one. I do have a feeling that if a recession is declared, it will not be declared until we are actually out of it. Perhaps we already are. But the labor market weakness lasts a couple years after the "official" end of a recession, so keep that in mind going forward.

It's getting late as I post this, so I'm not going to do an in-depth post tonight. But I will be anxiously awaiting the job numbers in the morning. I have a rather ominous feeling about the unemployment rate given all that has transpired recently. But by the same token, I have to report that micro level data is mixed. Here in our county I've been hearing some good things about the number of jobs being created. If this is a recession, it's certainly not an ordinary one. More on that later.

***

I'll be getting back to a somewhat more regular posting schedule in the coming days. In the last 6 weeks, I taught an MBA course as well as tried to prepare some things for the fall and spend a lot of time with the family. But with the beginning of August comes a feeling of urgency to get back to work. And the blog, in a way, is tied to that as a reflection of what I read and think about.

So as I dig in and revise those macro principles notes before the start of the semester, I will try to pull myself away a bit more often. Hope your summer has been as enjoyable as mine has.

That's less than I expected. Here's a link to the current BLS report and the first paragraph summary.

Nonfarm payroll employment was little changed in April (-20,000), following job losses that totaled 240,000 in the first 3 months of the year, the Bureau of Labor Statistics of the U.S. Department of Labor reported today. The unemployment rate, at 5.0 percent, also was little changed in April. Employment continued to decline in construction, manufacturing, and retail trade, while jobs were added in health care and in professional and technical services.

While it was less than I expected, I'm certainly not jumping up and down. The economy is marking time until it works its way out of the credit market problems that the housing boom created. Today's data makes me marginally more confident that we can avoid an "official" recession. Yet at the same time, I don't think anyone should be under the illusion that this is over yet.

Felix Salmon had an interesting post recently that sums it up well.

I feel like we're at a fork in the road right now. There are two possible outcomes: either the crisis will remain contained within the housing and finance sectors, in which case we should be able to bounce back, or else it will feed through into the economy more generally, in which case defaults will rise, employment will fall, and a nasty recession, complete with negative official growth rates, will bite right in the middle of the presidential election campaign. The Fed has done what it can; the dice are rolled. All we can do now is watch to see what happens.

Thing is, we have been standing at that fork for a while now.

I am going to be out of the office for the rest of the day, but here are links to the usual sources:

Economics Unbound, Angry Bear, Big Picture, Calculated Risk

Also MSM pieces from Reuters and NY Times.

BLS press release

80,000 jobs lost and the unemployment rate went up to 5.1% from 4.8%.

My recession probability is not yet at 100%, but getting close.

A student sends me this piece by Dani Rodrik.

How does globalization reshape wealth and opportunity around the world? Is it mainly a force for good, enabling poor nations to lift themselves up from poverty by taking part in global markets? Or does it create vast opportunities only for a small minority?

To answer these questions, look no farther than soccer. Ever since European clubs loosened restrictions on the number of foreign players, the game has become truly global.

Read the whole thing.

The Automatic Data Processing (ADP) employment report shows a loss of 23,000 jobs in February. I would look for Friday's BLS payroll survey to be pretty flat--maybe 50K in either direction. We shall see.

Tomorrow, I'm in the Quad Cities for an economic outlook breakfast. William Strauss from the Chicago Fed is the keynote speaker. I may have more to say after that.

This was one of those things that you knew was going to happen sooner or later. The trend in the growth of nonfarm payroll employment has been heading down for the last couple years. Now we get the first actual decline since 2003. (Read the press release from the BLS.)

More worrying perhaps is that the annual revision knocked off a whopping 191,000 jobs from the original month-to-month changes accumulated during 2007.

Of course this will solidify the case that we are in a recession already or just about to slide into one. Economists are terrible at picking out these turning points in real time. The NBER always waits until a clear picture has been established before making the call, and they have not made that call yet. I wouldn't make the call yet that the recession has begun on the basis of the data we have available now. But the continuing downward trend in job creation is becoming harder to ignore, and the potential for a significant slowdown is very real.

The employment picture is most bleak for teenagers. Unemployement among 16 to 19 year olds has increased from 15% to 18% in the last year.

Among all workers, average and median duration of unemployment are both up. The seasonally adjusted mean duration rose from 16.5 to 17.5 weeks in 2007 and the seasonally adjusted median duration rose from 8.2 to 8.8 weeks.

All in all, not a good way to start the weekend.

Nonfarm payrolls increased by 93,000 last month. The average for the last 5 months is approximately 99,000. We would certainly like it a little better if these numbers were a little higher. If we need somewhere around 125,000 jobs per month to keep up with population growth, then we are falling behind--albeit slowly, but falling behind nonetheless. What is interesting is that the gains have been reasonably consistent. In the last 5 months, three data points were in the 90,000s, just two outliers--one at 170,000 and one at 44,000. So it's as if the economy is trying to keep growing, but just isn't getting the traction it was getting in 2005 and 2006. We're stuck in neutral, unable to get back into the higher gear.

But at the same time, the labor force participation rate climbed a bit last month, back up to 66.1%--where it was in July. The employment to population ratio is at 63%--also the same as its July level. These numbers had dipped a bit in recent months, so this could be a good sign that there still is some strength remaining under the surface.

Of course employment is a lagging indicator. I wouldn't recommend trying to forecast when a recession will begin by looking at the last month's job growth. But given the preponderance of the evidence this number is not going to be enough to cause the Fed to stop cutting interest rates yet. Not until there's a clearer sign that we've finally let out the clutch and are safely in high gear again.

Good, not great. That's how I'd sum up today's employment report.

Employment rose in September, and the unemployment rate was essentially unchanged at 4.7 percent, the Bureau of Labor Statistics of the U.S. Department of Labor reported today. Nonfarm payroll employment rose by 110,000 following increases of 93,000 in July and 89,000 in August (as revised). In September, health care, food services, and professional and technical services continued to add jobs, while employment trended down in manufacturing and construction. Average hourly earnings rose by 7 cents, or 0.4 percent.

The unemployment rate actually climbed just a bit, but a 0.1% move in either direction is within the statistical margin. That said, it was 4.5% this summer, so that gives fuel to the argument that things have marginally deteriorated. Spencer was correct with his hypothesis that August numbers would be revised upward due to a fluke of timing in the data collection period. The last 3 months have clocked in at just under 100,000 per month on average.

That's no great shakes, but it's not recessionary yet. Opinions differ as to just how much job growth is needed per month to keep up with population growth. For a long time 150,000 was the number tossed around. Demographic changes have likely lowered that number, though probably not under 100,000. In casual conversation these days, I settle on around 125,000 and admit to uncertainty. In that respect, the recent numbers are good, but not great.

Given the current uncertainty about whether we are on the cusp of a recession, however, the numbers take on greater importance. Everyone wants to know whether this will move the Fed. No doubt this is a data point in favor of a less aggressive move, or perhaps no move at all. The Chicago Board of Trade binary options indicate that the probability of some kind of cut in October dropped from 56% to 41%. It's still pretty much a coin toss, but the bias of the coin shifted just a tad.

Then there was this speech today by Don Kohn. (h/t Calculated Risk) Here are a couple of salient points.

Many people had expected the Federal Reserve to follow a gradual path of rate reductions in response to financial market developments--say, 25 basis points in September and another 25 basis points in October. Such a path would be in keeping with how we have often approached our policy choices, as it has the advantage of allowing us to calibrate our policy as we see how the economic situation is evolving and responding to earlier policy moves. However, given the circumstances at the time of the September FOMC meeting, there were strong arguments in favor of the larger action of a 50 basis point decrease in the federal funds rate. For one thing, it seemed that a decrease of that size could well be necessary to promote moderate growth. We had been holding the federal funds rate at 5-1/4 percent, well above the expected rate of inflation, in part to compensate for what had been very narrow yield spreads and readily available credit. We did not know how quickly markets would recover, the extent to which credit terms and standards would be tightened, or precisely how households and businesses would respond to recent or forthcoming financial developments. But, pending further evidence, a 50 basis point easing was not an unreasonable first approximation of what might be required to keep the economy on a sustainable growth path.

In addition, I thought that economic performance would be better served by the Federal Reserve taking its chances on responding too much, or too rapidly, to the turmoil in financial markets rather than acting too little, or too slowly. Sluggish or inadequate easing risked a weaker real economy that might cause lenders to pull back even more, leading to a deteriorating situation that could prove difficult to reverse. With the news on inflation relatively favorable of late and with inflation expectations seemingly well anchored, I believed that we would be able to offset the cut in the federal funds rate--if it turned out to be larger than needed--in time to preserve price stability.

And what will become the headline for this speech...

We will need to be nimble in adjusting policy to promote growth and price stability.

That would seem to suggest a Fed that was ready to move decisively with an opening gambit of 50 basis points just in case all of it was needed. That doesn't mean that the next move will be that aggressive. Furthermore, they're ready to take it back if prices jump.

But right now inflation still seems contained. The CPI is up just 2% in the last year (that's the headline number, not core) and the most recent reading was actually a slight decrease. If the economy softens, it will keep the pressure off. But if not... then it pays to be nimble.

There is still a very large amount of sentiment for continued rate cuts, and today's employment number, while not entirely dismal, will probably not diminish that sentiment among those who hold most tightly to it. For those on the margin, it may be enough to urge them to wait. If the rest of the month's data turns out to be consistent with the labor data, then they may put off a cut until December. It seems to me right now that a further cut in October is speculative--an insurance policy in case the housing problems spill over into the broader economy. The more insurance they take, the greater the likelihood that they find themselves needing to reverse course in 2008.

For now, for my money, it's still a tossup.

In related action, PGL discusses the labor force participation rate and employment/population ratio. He mentions our discussion from way back. I'll say it again. The long run trends are for lower LFPR and E/P as the baby boomers retire. But that's not what we're seeing here. I raised the point then and repeat it now as a caution to not necessarily expect the "optimal" ratios today (and in years to come) to equal those of the late '90s. However, the declines over the last year (like the increases in the previous year) are of a more high frequency nature. PGL is correct to raise the issue. The decline in these numbers since December is indicative of some potential problems below the surface that deserve more investigation. More on that later.

Not good. Nonfarm payrolls fell by 4000 workers--given that revisions can be substantially larger than this, we can just say that it is flat. And flat is not good at all.

Here's something out of the report that highlights the even softer underbelly of this already soft report:

The number of persons employed part time for economic reasons, at 4.5 million in August, was 359,000 higher than a year earlier. This category includes persons who indicated that they would like to work full time but were working part time because their hours had been cut back or because they were unable to find full-time jobs.

That's roughly an 8% jump in the number of people employed part time for economic reasons since last year. Hard to put a positive spin on that.

As usual, PGL at Angry Bear reports on the employment to population ratio and the labor force participation rate.

How does this affect the outlook for the FOMC meeting? Glad you asked. The CBOT binary options now put a 93% probability on at least a 25 basis point cut and now a 60% probability on at least a 50 basis point cut.

The market is now firmly, solidly, and perhaps irrevocably convinced.

And it is hard to argue with it. All along this path, I have argued that it would take genuine weakness in the economy to push the Fed off its trajectory. As we went through spring and summer many observers, myself included, became more convinced that the strong 2nd quarter GDP figures may be the last gasp before significantly slower growth in the 3rd and 4th quarter. Recession probabilities are higher. Given that the labor market is lags the overall economy slightly, it is possible that in hindsight we will look back and say that at this point a recession had already begun. While I still think it is premature to make that call, it would be wrong to ignore the possibility.

Just in case you don't have enough to worry about already, listen to what Michael Mandel has to say.

If you are worrying about the economy, don't start with the weak job report--start with yesterday's productivity report. Nonfarm productivity has only risen by 0.9% over the past year. The four-quarter average of nonfarm productivity has only risen by 0.5%, the smallest increase since 1995.

Productivity growth establishes the sustainable growth rate of the economy. The fact that productivity growth is so slow has two consequences. One, it means that the economy is much more vulnerable to shocks that can push it into recession. Second, it puts Bernanke into a bind...with productivity slow, he has to be much more worried about inflation.

It is the latter consequence, the inflation threat, which weighs over Chairman Bernanke's head as they go into the meeting. While it is unlikely that a single rate cut this month would significantly increase inflation, the Fed needs to be very careful about what they communicate about where they will go from here. In 1998, the Fed cut three times and didn't remove the accommodation for almost a year. Some say that contributed to the end game of the bubble and bust of the dot coms in 1999 and 2000. An analysis of that episode and its lessons for today could spawn many essays.

And so we go into this weekend under the expectation that a rate cut on the 18th is almost a foregone conclusion. And that means it's time to start thinking seriously about October, which is not a foregone conclusion yet. We'll see a lot of data between now and then. Lots of things to think about.

For the 18th, a 25 basis point cut with the risks weighted equally towards lower growth and higher inflation would acknowledge what is happening without locking them into anything for the next meeting. It is imperative that they not lock themselves into anything for October. I have less objection to a 25 basis point cut if it is understood that it is NOT the first in a sequence. If it is interpreted as the first in a sequence, that will only lead to problems later.

(BLS press release)

Approximately 92,000 (seasonally adjusted) jobs were added in July. The number that typically gets tossed around these days as the number required to keep up with population growth is about 125,000. That would suggest that today's figure is a little on the weak side. However, if you look a little more closely, you will see that private (i.e. non-government) employment was up 120,000 while government employment was down 28,000 to yield the 92,000 increase overall.

Nearly all of the decrease in government was at the state and local levels, and half was in local government--education.

Looking back at last month's press release, we find that it was almost precisely the opposite. Private payrolls added 92,000 and government added 40,000 for a total of 132,000. Those numbers were revised today to show private payrolls adding 107,000 and government adding 19,000 for a total of 126,000.

This would suggest to me that private employment is slightly on the weak side. Of course, as Michael Mandel points out, the labor market is becoming very dependent on the health care sector for growth. Health care added over 36,000 jobs in July. I don't think that trend will change any time soon.

Elsewhere in the blogosphere, James Hamilton was not impressed with today's report. Barry Ritholtz says,

The WSJ reported that "Stocks may slide early Friday after a weaker-than-expected July employment report, raising fears that troubles in the housing market may be spreading."

Retail employment was flat, so we're not out of the woods yet.

At this moment, the Dow is down about 30, and S&P down about 8. Hardly a sell-off...the news was probably about what they expected. The week ends with a whimper.

UPDATE: I spoke too soon. Reuters describes a late day surge. And then, a finish with the Dow down over 281 points. It wasn't the payroll report that did it. The Bear Stearns conference call seems to be what drove the late slide.

In case you were wondering, Fed binary options on the CBOT did not move at all today. So despite a lot of commentary to the contrary, those who put their money where their mouth is don't see today's events as impacting Fed policy very much. At the moment, I concur.

Last week, James Hamilton led off a post with the line,

Let's admit it-- the other shoe is not yet dropping.

Opinions differ on what it will sound like when it does. April employment numbers came out day (+88,000 non-farm payroll jobs), and PGL doesn't like what he heard.

The fall in the household survey reporting of employment was 468,000. The unemployment rate would have risen even more had it not been for the fall in the labor force participation rate (LFP). The decline in the employment to population ratio (EP) was disappointing. Maybe it’s time that the Federal Reserve lower interest rates.

Then Craig Newmark points us to a Bloomberg article that says,

April 30 (Bloomberg) -- Federal Reserve Chairman Ben S. Bernanke's assertion that interest rates may need to increase to curb inflation is wrong. That's what Goldman Sachs Group Inc., Merrill Lynch & Co. and UBS AG are saying.

While Bernanke warned last month that the odds of worsening inflation have increased, chief economists at the three firms say the worst housing slump in a decade may drive the U.S. economy into a recession and stifle consumer prices. Their chief economists say the Fed will cut its target for overnight loans between banks at least three times this year.

...

Bernanke is missing ``the linkage between residential housing investment and the broader economy,'' Jan Hatzius, chief U.S. economist at New York-based Goldman, the world's most profitable securities firm, said in an interview. ``The housing downturn is of the first order of importance.'' Hatzius says the Fed will cut rates three times this year, to 4.5 percent from 5.25 percent.

Kash says it's too early to tell if and when rates will fall.

While disappointing, it is hard to imagine today's data having any impact on the outcome of next week's FOMC meeting, at least in terms of the policy action. Whether it will cause policymakers to prepare to soften their stance in the coming months is another question. My guess is, not yet. The change in language in the last FOMC statement made a lot of people thing that rate cuts are now off the table. I don't think that is the case, but rather that a rate cut in 6 to 9 months is more likely than it was a few months ago.

After today's employment report, my subjective probability assessment for the funds rate in 6 to 9 months has edged even a bit more towards a rate cut.

So as we head into the week of the FOMC meeting, the question is whether and when they will make a more direct reference in the press release acknowledging that the risk of slower growth is greater than the risk of inflation. Predicting when and if that will come is like predicting when "measured pace" would disappear last year. Lots of people will call it before it happens, and a few will be surprised. But the Fed knows that they have to choose their words carefully, and in this instance it may mean that a rate cut will come without a lot of advance warning provoking speculation in the financial markets. The relevant reference for you is January 2001. If a rate cut comes, it could very well take us by surprise in terms of its timing. If, on the other hand, things improve this summer and a rate hike becomes necessary, the signs will be much more clear.

And so we sit, still waiting for the other shoe to drop.

From the Wall Street Journal:

WASHINGTON -- U.S. payroll growth accelerated during November, suggesting the economy isn't slowing too much, and worker wages grew slightly less than expected.

Nonfarm payrolls increased by 132,000 jobs, after rising by a revised 79,000 in October, the Labor Department said Friday. The jobless rate inched up as expected to 4.5% from 4.4%.

Wall Street expected payrolls to grow by only 110,000 jobs last month. Another surprise came with average hourly earnings, up by just 0.2% to $16.94 instead of the expected 0.3% increase.

Blog roundup:

Angry Bear reminds us that the employment to population ratio hasn't changed much. Big Picture is waiting for the inevitable revisions. Calculated Risk thinks that construction job losses will start spilling over into the retail sector soon. General Glut compares this November to the last couple of Novembers and is unimpressed. Global Trader's Diary points out that the dollar isn't impressed either. Greg Mankiw touts wage growth. James Hamilton calls it "solid."

All in all, it's a little better than I expected, but it doesn't change much in the way of my thoughts on the likely path of Fed policy. Employment is still not a good early warning indicator. (On that note, macroblog has good news to report on some leading employment indicators.) In the last cycle, the peak employment was in February 2001 when the Fed had already begun to cut rates. Still, the figure is encouraging. The broader economy continues to steam ahead even as housing and autos falter. It has kept up longer than some people thought it would, and I don't think it has run out of legs quite yet. Time will tell, and people will be watching the December and January numbers very closely.

New York Fed economists James Kahn and Robert Rich have a new model for estimating productivity growth in "real time." Real time in this case still implies a year or two lag. However since productivity growth is notoriously volatile, there will always be the possibility that a change in trend will remain hidden for several quarters. The results do look promising. They suggest that we are still in a high productivity growth regime. A big tip of the hat goes to Mark Thoma for the pointer. I'm posting the link in light of my comments yesterday in which I worried aloud that slower productivity growth could make it harder for the Fed to fight inflation [cf. the recent argument that an undetected productivity slowdown caused the Fed to follow a more expansionary policy than they should have in the 1970s]. Chalk this up as evidence that it may be premature to announce the end of this period of high productivity growth that began (or rather, resumed) in the 1990s. I certainly do hope they are correct.

Tracking Productivity in Real Time, by James A. Kahn and Robert W. Rich

UPDATE: Bill Conerly points to a post of his from about a month ago where he said this:

What we see in last quarter's data is the weak GDP growth more than an underlying trend of weak productivity. In fact, long-term trends are the only way to really see what's happening in productivity. Unfortunately, it's hard to catch the first signs of a change in trend productivity. However, the actual data are consistent with the same old underlying trend productivity growth, plus a weak quarter of GDP. That's a judgment call, not hard science, but I feel very comfortable with this call.

This time, it comes from Ed Leamer, who is worth listening to: (Reuters)

SAN FRANCISCO (Reuters) - The U.S. economy will expand at a weak pace next year, setting the stage for lower interest rates, according to a UCLA Anderson Forecast report released on Thursday.

The forecasting unit's latest report projected quarterly real gross domestic product growth no higher than 2.7 percent next year, reflecting the weak housing market.

...

As a result, the Federal Reserve will cut interest rates to stimulate business, said Edward Leamer, director of the UCLA Anderson Forecast.

"We think the Fed will shift from an inflation concern to a sluggishness concern so that we'll get some rate cuts," Leamer said, adding that he sees the Federal Funds rate falling to 4.5 percent by the fourth quarter of next year.

...

Manufacturing has already shed so many jobs it is in no position to produce the kind of massive layoffs that paired with a housing downturn would trigger recession, Leamer added.

"We've trimmed it to the bone," Leamer said, referring to factory work. "It's already lean and mean."

Additionally, the economy will avoid recession because credit is abundant and consumers will continue spending at a moderate pace, Leamer said.

Interesting. Their prediction of moderate growth (2.7%) is certainly less than average, but equally certainly not indicative of a recession. It is quite similar to the GDP growth in 1995 (a "soft landing" year). And while there were two rate cuts in 1995 (and one more in early 1996), those cuts were to bring the funds rate down to 5.25%. Ironically, that's where we are now. So, while I'm not ready to predict three rate cuts in 2007 to bring the funds rate down to 4.5%, I would say that the UCLA forecast is in the ballpark.

Given all that has transpired in recent days, I would regard a rate cut in the first six months of 2007 to be more likely than a rate increase in that same time frame. That said, I continue to hold to the view that a rate cut at this time would slow the return of core inflation to its comfort zone. The fact that productivity is not growing as fast as it was in the first half of the year and that Mr. Bernanke has suggested that potential output growth may be slowing only serve to reinforce that view. Unlike 1995 and 1996 when productivity was rising rather than falling, the Fed will not have the luxury of cutting rates while inflation trends down.

The part of me that wants to give a prediction that is right is turning to the view that there will be at least one rate cut in 2007.

The Cassandra in me is having a tough time with that.

UPDATE: Calculated Risk quotes the LA Times version of the story, which includes Leamer quotes such as:

"If you are a builder or a broker, it will feel like a deep depression," he said. "But the rest of us will hardly notice."

and...

His conclusion: "The models say 'recession'; the mind says 'no way.' I'm going with the mind."

UPDATE 2: Leamer isn't alone. At least some people's models agree with his mind.

NEW YORK (Reuters) - The economy will likely pick up in 2007 after output growth slows rapidly in late 2006, according to a survey conducted by the Philadelphia Federal Reserve Bank released on Thursday.

Economic growth for 2008, released for the first time in the survey, was forecast at 3.0 percent.

Economists lowered their forecasts for U.S. growth in the first half of 2007 to 2.8 percent from 3.0 percent when the previous survey was taken six months ago. They forecast growth at 3.1 percent for the second half of 2007.

This month's fedgazette from the Minneapolis Fed features poverty as its theme. There are a number of interesting articles that are worth your attention. In this post, I call your attention to one in particular on income mobility--the ability of individuals and households to move through the income distribution over time.

Income mobility is one of those things that we like to talk about and make claims about. We like to hear stories of "rags to riches", but how often does it really happen? It turns out that we understand relatively little about mobility from a statistical standpoint. Ronald Wirtz writes in the fedgazette article,

Bhashkar Mazumder, an economist at the Federal Reserve Bank of Chicago, has authored several mobility studies in recent years. He said, also via e-mail, that prevailing mobility research throws water on the common notion that U.S. income is highly mobile and more mobile than other countries. More recent studies, like his own, have used much richer longitudinal data that track income over longer periods of time, giving a more accurate reading of lifetime incomes in the United States. Research over the past decade and a half shows that “mobility is relatively low in the U.S. and lower than we thought,” said Mazumder.

...

Nathan Grawe has also done research on income mobility as an economics professor at Carleton College in Minnesota. In his estimation, the four best studies done to date on intergenerational mobility have both positive and negative findings, and most results were not statistically significant; in other words, the findings aren't particularly trustworthy. “All told,” Grawe said via e-mail, “I'd say we have no evidence of change.”

Part of the problem is that studies done before about 1990—which generally concluded that the United States had high mobility—are widely discredited today as faulty, mostly because they relied on very small windows of income data, often just a few years or less. In 1992, Solon published one of the first papers suggesting that U.S. mobility was not as high as everyone thought.

Mazumder's research comes to the same conclusion. But his most recent effort with Daniel Aaronson (also of the Chicago Fed) might have something of a silver lining. They found that current mobility might simply be returning to its historical trend line after experiencing an uptick in the 1970s. In other words, mobility might be worse compared to the 1970s, but it might well be in line with the country's historical average.

Obviously, a lack of longitudinal data going way back will remain a problem. Maybe in fifty years we'll be able to put the current period in historical perspective. Too long for some. You know what Keynes said about the long run. While the evidence here is conflicting and the conclusions not quite conclusive, the situation is better than that concerning the question of how much mobility is optimal.

The notion of perfect mobility—an equal chance for any outcome, regardless of where you start—has a hint of social and economic chaos, by virtue of the fact that it implies a lack of predictability in outcomes regardless of the very things that families and societies tend to value: effort, ability, education and other human capital investment, and parenting.

Economists believe incentives motivate behavior. Grawe, from Carleton College, noted that mobility research was often written “in ways which suggest more mobility is better.” But a society with no obvious determinants for income “would clearly have all sorts of incentive problems.”

For example, parents' attempts to offer certain advantages to their kids—reading to them, sending them to better schools, saving for college, transmitting certain values—might be for naught in a world where these things have no lasting economic effect. In a 2002 working paper on the notion of perfect mobility, sociologist Adam Swift of the University of Oxford wrote, “Even those that regard current mobility patterns as evidence of morally unacceptable unfairness should acknowledge that some mechanisms by which parents transmit advantage—or disadvantage—to their children are unobjectionable and would exist even in an altogether just society.”

In other words, there is no clear guidance at all on how much mobility is optimal or even what we mean by optimal mobility. One cannot escape the fact that mobility requires an appeal to long run incentives, but people do not always behave in accordance with those long run incentives. Hence, a divergence between opportunity and outcomes is assured. This is the world in which we live. And this is why the solution to the problem of income inequality is more difficult than many people realize.

Via Reuters:

The Labor Department said a seasonally adjusted 299,000 workers filed new claims for state unemployment insurance benefits in the week ended October 14, down from 309,000 claims a week earlier.

Economists polled by Reuters were expecting a slight increase in jobless claims to 312,000 from an original reading of 308,000 in the week ended October 7.

Separately, the Conference Board released its index of leading economic indicators today. From their website:

The Conference Board announced today that the U.S. leading index increased 0.1 percent, the coincident index remained unchanged and the lagging index increased 0.2 percent in September.

and...

The leading index has fallen 1.0 percent below its most recent high reached in January. At the same time, real GDP growth slowed to a 2.6 percent (annual) rate in the second quarter, following a 5.6 percent gain in the first quarter. The behavior of the leading index so far suggests that economic growth should continue at the slow rate in the near term.

According to the Wall Street Journal, analysts had expected a 0.3% increase in the leading economic indicators. So once again the news is mixed. Overall, it appears that the economy is slowing a bit. Growth for the remainder of the year will probably remain below average, but there is no indication yet of a full-blown recession.

Taken as a whole, this week's data releases leave us pretty much where we started. If there were only a couple pieces of conflicting evidence, it would be more puzzling. The preponderance of conflicting signals reinforces what most of us have been thinking for a while. For the past few months, indeed most of this year, the economy has been slowly inching toward a critical point where either growth will slow (perhaps briefly turning negative) or resume at a more normal pace. That's a good argument for not doing anything to rock the boat at the moment.

UPDATE: On Tuesday, I admitted that the mixed bag of data makes it impossible for me to be Harry Truman's "one-armed economist". Today, James Hamilton also cannot avoid saying "on the other hand."

I'm wondering though whether "no change" might be the least likely outcome at this point. If we start to see some serious financial repercussions develop in housing, I'd look for a rate cut, and wouldn't worry in that event about commodity prices, since I would expect to see commodities fall sharply on news of a big downturn in economic activity. On the other hand, if instead we have seen the bottom for housing and the core inflation numbers remain this high, I'd look for the Fed to tighten further.

That is the direction the data has been pushing me as well. Unlike Kash, who seems more convinced than I that we've reached a peak (though he does leave some room for doubt), I see the upside and downside risks as roughly equal.

It might be that the upcoming 3rd quarter GDP data could be the news that gives us a clue as to which way this will break. It probably won't be the overall growth rate (which is likely to be positive but below everage), but the different subcategories of consumption and investment that will tell the story. At least until that point, I'm prepared to use both hands when explaining where the economy seems to be going, and what direction interest rates might take in 2007.

UPDATE 2: David Altig finds himself in general agreement and is almost ready to take the next step--but not quite.

But, for reasons I'll detail in a later post, I'm beginning to wonder about the reach of developments in [the housing] sector. I'm not quite ready to take the anti-Roubini bet with the degree of confidence that Nouriel himself puts on his recession call. But I'm getting there.

Absent any additional negative shocks, I would agree. I'm not quite ready to call it a soft landing yet, but I too am getting there.

Busy weekend ahead, so I'll try to do justice to this next week. In the meantime, I refer you to Economonitor, SCSU Scholars, and Cafe Hayek for comment on the jobs data.

We may have been underestimating the number of jobs by over 800,000. That takes some of the sting out of this month's 51,000 jobs (analysts had expected more than twice that many). Barry Ritholtz again wins by going with the "under," but we may have to audit some of his previous predictions in light of what Felix Salmon calls the "mother of all revisions."

Enjoy the free lunch from the Wall Street Journal. Richard Epstein and Michael Reich debate the minimum wage, particularly as it relates to Mayor Daley's decision to veto a minimum wage that would have applied to big-box stores in Chicago.

I found the following exchange to be the highlight.

Reich:

I agree that higher minimum wages might lead to somewhat higher prices. But this might be a good tradeoff. To find out, again we must draw from careful empirical studies, not general statements, to quantify the effect. My San Francisco study found that a 26% increase in the minimum wage increased restaurant prices by about 2.5%, or 25 cents for an average $10 menu item. We now know, using Wal-Mart's own data, that if Wal-Mart's hourly pay and benefits scale increased to match those in its industry as a whole, and the costs were fully passed on to consumers, its prices would increase by only a penny on the dollar. Moreover, profit margins have been increasing in large retail companies, so there is room for pay increases that do not translate entirely into price increases. See "Wrestling with Wal-Mart: Tradeoffs between Profits, Wages and Prices."

On the issue of turnover costs, no one is arguing that low-wage firms would individually choose to increase their pay and lower turnover, as the savings would not be sufficient. If all firms are required to do so, however, employment can actually increase. In the field of labor economics, this is a standard argument used to understand minimum wage effects. You will find it in every major undergraduate textbook, including those by free-market-oriented economists such as George Borjas and David MacPherson. You will also find an emphasis on turnover issues in understanding labor markets in the 2006 Economic Report of the President.

Epstein:

On the Wal-Mart profit figures, the numbers that I have seen differ. The average profit per employee is around $2,000 per year. That hardly speaks of massive exploitation of workers. Rather it is consistent with the lower prices that it offers to consumers, often from the least advantaged areas, where prices are estimated at around 8% to 13% below what they would otherwise be. Finally, I am totally puzzled why any labor text would argue that high-wage-low-turnover strategies are only efficient if everyone in town adopts them. The brief explanation that Michael offers here is just not credible.

Why won't the savings be sufficient to induce the change? Indeed any change in position, however small, that improves output should be welcomed, period. There is no prisoner's dilemma game here. A firm that gets higher output from adopting superior strategies should be thrilled if its competitors lag behind. So absent the statute, there should be a really strong incentive to make changes in employment strategies that other firms cannot duplicate. Nor is there any reason in theory to expect non-covered firms to raise wages unless demand for labor increases as the cost increases. It is every bit as likely that non-protected workers will be more numerous and could easily receive lower wages, if they stay in the community at all.

Read it all.

I'm working on a longer more analytical post on this. For now, you can read this: (Reuters)

WASHINGTON (Reuters) - U.S. employers added 128,000 workers to their payrolls in August, evidence of a cooler -- but still solid -- pace of economic growth that could let the Federal Reserve hold interest rates steady.

...

"It looks an awful lot like a soft landing, at least for now," said Dana Johnson, chief economist at Comerica Bank in Detroit. "It's an economy that is not firing on all cylinders -- it's got a clear concentration of weakness in the housing sector -- but, otherwise, an economy that is moving ahead at something resembling a trend-like pace."

The prospects for a soft landing will be the topic for a post or two next week. Are the 128,000 jobs added this month a sign that the economy is still fundamentally on track or that recession looms? More to come.

One of the nice things about blogging is that you can go back and revisit ideas from the past because the blog archives all of those old words and pictures as if you wrote them yesterday. For a professor, this is nice because invariably you want to bring up things that you discussed in the past, but a new set of students needs to get up to speed. In economics, it is essential to have access to that background. This post is in that spirit--just in time for classes to start in a couple weeks. Just about exactly one year ago I wrote this post that looked at the core CPI inflation rate and payroll employment starting from the trough of the last two recessions. Go back and take a look.

Here are the corresponding charts updated to the lastest data (including the labor market data from Friday).

The payroll charts tell us what we already know, but put it into context with the previous recovery. With the exception of the last few months, job growth has hovered around the 150,000 per month widely recognized as necessary to keep up with the growth of the population. There has not been the kind of robust recovery of jobs that came along after the 1990-91 recession.

Note also that the point in the recovery from the 1990-91 recession that corresponds best to today is early 1995. The labor market did experience a bit of a slowdown in 1995, and the possibility of recession was openly discussed. At the same time, inflation, which had been high at the start of the recovery, was starting to come down. This was the beginning of one of the most successful non-recessionary disinflations that this country has seen in the modern era. And in Feburary 1995, the Fed raised rates one last time by 50 basis points to 6.0% where it would stay until July when rates began to fall slowly for the next couple years.

Remember, the 1994-95 episode was a success in terms of a soft landing--a cooling of GDP to a sustainable pace, fall in inflation, and no recession.

Is it any easier to predict what is coming on the basis of this information than it was last year? Not really. I do have to say one thing, however. The totality of the picture is less encouraging than it was a year ago. Martin Feldstein shows his concern in Monday's Wall St. Journal opinion pages.

The Fed governors and Reserve Bank presidents appear to believe this [soft landing] will happen. Their "central tendency" economic projections, summarized in the July Monetary Policy Report, state that the Fed's favored measure of inflation, the PCE price index excluding food and energy, will decline from the 2.9% rate in the most recent quarter to between 2% and 2.25% in 2007, presumably on its way to Ben Bernanke's "comfort zone" of 1% to 2% in 2008. They project this to occur with real GDP growing above 3% and the unemployment rate remaining under 5%. Indeed, not a single one of the 19 FOMC members projected growth of less than 2.5% in 2007 or an unemployment rate above 5.25%.

Although this optimistic outlook is possible, experience suggests that it is unlikely. A mild slowing of economic growth is generally not sufficient to reverse rising inflation. That generally requires a sustained period of excess capacity in product and labor markets, with GDP growth falling significantly or even turning negative.

The last recovery being an significant exception to that rule, but as we have discussed, that requires everything to go right, especially productivity. Feldstein continues...

The official estimates of productivity growth showed a gradual decline of productivity growth in the nonfarm business sector from 3.9% in 2003 to 3.4% in 2004 and 2.7% in 2005. The result of the slower productivity growth and rising compensation per hour (from a 4% rate in 2003 to 5.1% in 2005) caused the increase in unit labor costs to accelerate from 1.3% in 2003 to 2.1% in 2004 and 2.8% in 2005. Taking the new GDP estimates into account is likely to lower the calculated productivity growth rates and cause estimated unit labor costs to have risen faster than 3% in the most recent quarter. There is no reason to anticipate a favorable productivity surprise of the type that contained inflation in the 1990s.

...

While this risk provides a rationale for a pause at tomorrow's meeting, it would be wrong to focus just on this downside risk. The probability that inflation will rise above the FOMC forecast is at least as great. The unemployment rate of 4.8% still represents a tight labor market. Waiting for more data before deciding to raise rates is not costless. If the Fed does not act and core inflation continues to rise, expected inflation may rise further. Higher expected inflation would cause faster increases in wages and prices. If the core PCE inflation rate rises above 3% in 2007, it would take a very substantial slowdown and a large loss of GDP and employment to bring it back under 2%.

In assessing the current interest rate decision, the FOMC members should recall that during the Volcker and Greenspan years the Fed pushed the fed funds rate to 8% above the concurrent rate of CPI inflation in the early 1980s, to 4% in 1989 and to almost 3% in 2000. That measure of the real fed funds rate is now less than 1%.

The Federal Reserve has a difficult task ahead. It is understandable that it would like to achieve the soft landing of low inflation with continued solid growth. But that may not be possible. And if the Fed wants to convince the markets that inflation will be contained in the future, it must show that it is willing to take the risk of tightening too much.

Feldstein is entirely correct to point out the real fed funds rate is still a bit low. This is what I have railed about time and again. If it is not possible to have the soft landing as we did in the '90s, then this meeting is a test of inflation fighting resolve. One long term lesson this episode may provide future generations of policy makers is just how important productivity growth is for cushioning the economy in situations like this. Without productivity growth, and worse yet with further increases in oil prices, it forces the Fed to go further than it would have ideally wanted in order to get the real rate where it needs to be. I am sympathetic to the argument that they should have raised rates faster in the early going (2004) to head off the rise in inflation. But in light of the weak labor market data--a labor market that the above charts show just never got off the ground the way it did in the mid-'90s--I can't say that I would have done anything differently. To raise rates too quickly while productivity growth rates were falling might have choked off the recovery even sooner.

On that note, I conclude with this little blurb from the NY Times.

DO WE HEAR A PAUSE? The central economic news of the week will emerge from the meeting of the Federal Open Market Committee, which will decide whether to raise the benchmark short-term interest rate — now at 5.25 percent — yet another time. On the one hand, many economists predict that the Fed will leave things unchanged, which would be welcome news for investors, but then again, who knows? The dearth of one-handed economists is legendary. (Tuesday.)

Try as I might to be single-handed, I can't say for sure what the Fed will do. I can come up with a lot of reasons to raise them one more time. I do worry a bit about the 6 to 18 month picture for inflation if they pause now. I'd like the increase to be now and the pause to be for the remainder of the year. But the markets have made up their mind, and Bernanke may not want to stir them up too much. They could very well send a strongly worded statement that though they pause now an increase in October is the default option unless new data changes that stance. If such a statement comes off as credible, it would be an acceptable compromise. We shall see.

UPDATE: The September meeting escaped my mind. Change the above to read September or October.

UPDATE 2: It gets worse. Michael Mandel reports that productivity figures for 2004 and 2005 are likely to be revised downward.

UPDATE 3: Mandel was correct. Productivity was revised downward for the last couple years and the 2nd quarter of 2006 came in at 1.1 percent growth. The trend is downward. While Mandel tries to tell a positive story of the 10 year productivity growth rates, his chart shows exactly what is troubling in this situation. In the mid-'90s, the trend was reversing and going up making it easier to contain inflation. Today, the growth rate of productivity is falling.

Approximately 211,000 net new jobs (seasonally adjusted) were created last month. That makes over 2 million in the last year. If you exclude the months of September and October where were distorted by the effect of the hurricanes, that gives an average of over 200,000 jobs per month for the last year. Even if you include those two months of slower than average growth, the average for the last 12 months is still almost 175,000 per month. That's not bad. It's even enough to bring some cheer to an Angry Bear (PGL). He's happy because the employment to population ratio hit 63%. That's up 0.6% since last March. The trend is definitely good--which is important since that ratio comes from the Household Survey, which has a small sample size (meaning month-to-month changes can be noisy).

I've been taken to task by some for being too optimistic and by others for being too pessimistic (though the count is probably in favor of those who call me too optimistic). For the past several months (since late spring/early summer of last year), I've been of the opinion that the labor market is better than a lot of people are giving it credit for, but not destined to break any records of the late '90s. In those months, we've seen some ups and downs, but all in all it seems very consistent with a "soft landing" scenario.

So is it too early to call a "soft landing"? Probably. We should at least get the 1st quarter GDP and another month of job growth before making the call. Some (the ones who call me too optimistic) would say that we should wait even longer. How long? What will be the characteristics of the data that signal a soft landing? A year of 200,000/month job growth? (Excluding the hurricanes, we got it.) A return to 3-4% real GDP growth after a quarter of below average growth? Declining inflation pressure and a pause in interest rate hikes by the Fed? I'd go with all of the above, and it does sound like it is possible in the next few months. (Though some think the Fed might put off that pause a little too long for their tastes.)

That said, let's look a little deeper into the labor report and see where the action is.

Labor force participation:
Men--up
Women--down
Teenagers (both men and women)--up
Black men and women 20 and over--up
Black teenagers--down (with an increase in their unemployment rate from 30.8% to 33.1%)
No high school degree--down
High School (no college)--up
Some college--up
Bachelor's degree-down

In most cases, the employment to population ratio behaves similarly. In short, the overall picture looks good, but different demographics will see things differently. If you've been following along, this is nothing new to you. Obviously Black teenage unemployment is a concern. I have mentioned in the past that labor force participation for women in general may have peaked, or nearly so, in the late '90s. Overall their rate for women continues to fall while for Black women (who participate in the labor force at a greater rate than white women) the rate continues to rise.

Finally, let's look at the change in the distribution of unemployment duration. Since March of 2005, how has the percentage of people unemployed for various lengths of time changed?

Less than 5 weeks--increased from 32.8% to 38.1%
5 to 14 weeks--decreased from 30.5% to 28.6%
15 to 26 weeks--decreased from 15.2% to 14.9%
27 or more weeks--decreased from 21.5% to 18.4%

That, my esteemed readers, is an improvement indeed!

February nonfarm payroll employment increased by 243,000 in February according to the Bureau of Labor Statistics.

Link to report--note, this address is not the archived version

Labor force participation and the employment to population ratios were little changed (LFPR was up 0.1%), and the unemployment rate moved up from 4.7% to 4.8%.

Average hourly earnings were up 5 cents from $16.42 to $16.47 with the largest percentage gains going to the professional and business services sector. Manufacturing jobs and wages continue to decline. (See Mark Thoma for some more details on where the growth in jobs is or is not.)

That 243,000 jobs were created in February is good news. PGL is happy that even more entered the labor force (which explains why the unemployment rate ticked up slightly). Of course we are talking about data from two different sources--the household survey definitely being the more noisy of the two. So while I always hesitate to put too much emphasis on one month's data, the developing trend does have me cautiously optimistic. I feel better about the labor market than when I woke up this morning.

The bond market is taking it fairly well. One would expect that a positive job report would give more ammunition to those who want the Fed to take rates higher. Yields are a little higher today, but apparently it will take more than this to undo the inversion. As Barry Ritholtz reminds us, the Fed's decisions are going to be dependant on incoming data such as this. He's also watching the bond market very closely on this one. As of this moment, the 10 year yield is up 4 basis points. That's significant, but hardly an overreaction. I think the market is responding appropriately and recognizing that we are slowly, gently, edging towards a higher definition of "neutral" interest rates for monetary policy. That could be a good thing. If the target is edging higher, the risk of overshooting is lower. We could be close to a "soft landing."

The 243,000 jobs exceeded expectations a little, but not by too much so as to jolt the bond market. It is very good news.

Look at these two graphs, then look at my comment on the previous post (the 14th comment).

I don't think this is a full explanation. Some of the comments to the previous post make good points that could also explain part of what we're seeing. However, if manufacturing declines, the inflow into those occupations must slow over time. If 16-19 year olds are the inflow, then this is what you might see.

Thoughts?

Background post from May 2005. I showed these charts to give the long view starting before the 1990-91 recession.

Let's zoom in on just the last 10 years or so, including the time since I made these graphs. By the way... all of these are seasonally adjusted.

Men age 20 and over are employed at a rate about equal to that in late 2001 or early 2002. That's also about the same as late 1995. Since early last year this series has definitely been moving in the right direction.

For women the peak-to-trough distance was smaller and so have the gains since the trough. But look at where we are compared to where we were. For women 20 and over, e/p is at mid-1997 levels. If you eyeball that chart and think that 58% might be a reasonable place to be in the long run, it's hard to find too much fault with where we are.

Hmmm... I think we've found the problem. Surely some of this can be explained by schooling. Poor job prospects lead people to stay in school longer. But this is quite a persistent change from the last couple decades. Looking at the raw data going even further back in time we find that e/p fell to around 40% after the recession in the early '80s. We've been stuck between 36 and 37% for almost 3 years. I was puzzled by it last May, and I am still stumped. Why did this recession cause teenage employment to crash, and why hasn't it picked up like it did in the early '90s? By the way, the labor force participation looks the about the same for this group. There has truly been a shift in labor force participation by teenagers that is much larger and longer lasting than after previous recessions.

Does this mean that students believe that staying in school will improve their future job prospects or that it is an alternative to poor job prospects? Both? The timing with the recession would suggest the latter, but I suspect many of those students also believe the former. It's hard to disentangle them.

Snapshots of other demographics are not as bewildering. Labor force participation is very slowly and slightly shifting down for men and up for women in their twenties and thirties. There's bound to be some intra-household labor substitution that will help the overall LFPR stabilize.

But if you were to ask me what is the one question about the current labor market I would most want answered, I think you know what it is.

UPDATE: Conversation continues in the comments.

UPDATE: MaxSpeak has links to other commentaries on the labor market.

BLS News Release:

Nonfarm payroll employment increased by 193,000 in January, and the unemployment rate fell to 4.7 percent, the Bureau of Labor Statistics of the U.S. Department of Labor reported today. Job gains occurred in several industries, including construction, mining, food services and drinking places, health care, and financial activities.

Most forecasts were up in the 250,000 range. Again, the actual number failed to hit the mark. Why? Is the labor market really that bad? Or are we missing something?

Here's a really simplistic prediction of job growth in the long run. Assuming that people who enter the labor market do find a job in the long run, the growth rate of population plus the growth rate of the labor force participation rate (LFPR) should equal the growth rate of employment. It is a biased predictor because the best measure of job growth (the payroll survey) uses different data than the CPS which measures LFPR. The CPS counts more people as employed than the payroll data picks up. But if that bias is fairly stable, it won't affect our growth rate prediction that much.

So here it is, the simplistic prediction versus actual payroll growth.

For being so simple, it picks up the movement fairly well. Looks like a smoothed version of the series, which it should. The fast growth in the '80s and even in the '90s is due in part to population growth, which is reflected in the predicted value.

Lately, population growth has slowed, and you can see this in the simple prediction. Demographers predict population growth to continue to slow as the boomers age. Hence, the percentage change in payroll employment will slow unless LFPR continues to rise. Whether it is reasonable to expect LFPR to rise is another question. To say "opinions differ" is an understatement. The fact is that at this point, we don't know. I think all that it is safe to say is that LFPR can't continue to rise in the way that it did up through the mid 1990s. But suppose that we are currently right at the approximate long run LFPR--that for the next 50 years it will fluctuate above and below its current level. If that is the case, the growth rate of payroll employment will slow to less than 1% per year on average. As the chart above shows, years with job growth of 3% seem to be over. Years with job growth of 2% will become fewer.

Remember, this is a very simple prediction that is not meant for forecasting month-to-month changes, but long run trends. It's a reason to be skeptical of predictions of huge job growth over the next decade, for sure. On the LFPR (and the employment/population ratio), I've been reserving judgement. More on that later. But it's not just LFPR, population growth (which is harder to do anything about) is a big part of the story too.

BLS News Release:

Nonfarm payroll employment grew by 215,000 in November, and the unemployment rate was unchanged at 5.0 percent, the Bureau of Labor Statistics of the U.S. Department of Labor reported today. Over the month, job growth was widespread, with large gains in construction and food services.

Barry Ritholtz took the "under" and was just barely right.

Earlier this week, I said that if it didn't hit 200,000 we'd all be scratching our heads over the weekend. Now, if only we could string three or four of these in a row.

Off to class. More later.

UPDATE: Here's what the BLS had to say about Hurricane Katrina:

Beginning in October, questions were added to the household survey to identify persons who evacuated from their homes, even temporarily, due to Hurricane Katrina. Data collected through these questions do not represent all evacuees; persons living outside the scope of the survey—such as those living in hotels or shelters—are not included. The questions were asked of persons in the household survey sample throughout the country, since some evacuees relocated far from the storm-affected areas. The questions also determined whether evacuees had returned to their homes by the time of the survey. This additional information enabled analysis of the employment status of this subgroup of evacuees. (The total number of evacuees estimated from the household survey may change from month to month as people move in and out of the scope of the survey.)

Information gathered in November showed that about 900,000 persons age 16 and over had evacuated from where they were living in August due to Hurricane Katrina. These evacuees either had returned to their homes or were living in other residential units covered in the survey in November. Half of the evacuees had returned to their August 2005 residences. Of all evacuees identified, 55.2 percent were in the labor force in November. The employment-population ratio for these evacuees was 43.9 percent. The unemployment rate for persons identified as evacuees was 20.5 percent; it was much higher for those who had not returned home (27.8 percent) than for those who had returned (12.5 percent).

PGL notes that the household survey showed a drop in employment and that the employment/population ratio (e/p) fell from 62.9 to 62.8.

Drilling down a little further into the household data we see that e/p increased slightly for white women and white teens while e/p for African-Americans fell from 58.5 to 57.3.

So then we read the BLS's statement on page one of the report:

In November, the state population controls used for the household survey were adjusted to account for displacements due to Hurricane Katrina. These adjustments had a minimal impact on the national household survey estimates.

Perhaps minimal on the total number, but if the displaced residents are disproportionately African-American, it may affect those numbers. To be honest, I'm not sure if their population controls take race into account or how reliable those adjustments might be for this sample. I think it might be too early to say just how much of it is a statistical quirk and how much is truly the impact of Katrina on African-American unemployment. But it does suggest a sizeable impact on African-American unemployment that stands out in today's data.

African-American teenage unemployment jumped from 32.9% to 38.8% this month. It is now at the highest level since August 1995. For those who want the longer view, here is the chart I generated from the BLS website (which has one of the best web facilities for getting the data just the way you want it).

Most other aspects of the labor data appear to be holding their own or slowly improving. This picture is not, and not just because of the hurricanes either. That is a little disconcerting.

UPDATE 2: Andrew Samwick also weighs in.

According to the BLS, the economy added a net 56,000 jobs in October, well below the level needed to keep up with population growth. This only adds to the building puzzle over why job growth has been sluggish even as real GDP posts 10 consecutive quarters over 3%.

I'll add a graph later, but I'm trying to do about three things at once and wanted to get this story up with a couple of links right away.

See the NY Times story here. They note that:

The department revised its figures for August and September. It said that 148,000 jobs were created in August instead of 211,000 that it previously thought and that 8,000 jobs were lost in September instead of 35,000. As a result, the data shows 36,000 fewer jobs were created over the two months than the department previously estimated.

Kash (Angry Bear) notes that construction job growth has been strong while all other sectors have really fallen off in the last three months.

The household survey continues to show a different picture than the establishment survey. The unemployment rate fell to 5.0% and the labor force participation rate and employment to population ratios have been basically steady for the last three months or so.

The annual revisions in a few months might be interesting. Stay tuned.

UPDATE: PGL leaves a comment and has a post at Angry Bear

UPDATE 2: Here's the graph I've been meaning to make. It's not a pretty picture.

As Barry Ritholtz said today in a post entitled "Ignore the noise," meaning the noisy monthly payroll data.

...Considering that much of the noise comes from professional economists – who should know better – I find this to be disappointing.

...

The monthly jobs data is too noisy to focus on any one single point; That’s even more true post Katrina/Rita/Wilma. So why concoct all this happy talk?

And Ritholtz certainly doesn't want to focus on the household survey either. (I don't either--but I'm going to be interested in how the annual revisions shape up.)

Anyway, this graph eliminates some of the noise. Look at how the current downturn (the last three months have averaged about 65,000 jobs per month) in job growth looks similar to the drop in 1995, but the post-recession peak this time around was lower than in 1994. That observation is not discernable from the noisy monthly data.

Read here for Jared Bernstein's take on the labor market, posted by Max Sawicky. Bernstein shows three charts that illustrate the slack in the labor market even when controlling for changing trends in men's and women's labor force participation rates. The model takes data through 2001QI and forecasts through 2005QII. In his words,

These are not large gaps, but they’re not trivial either. In tandem with other evidence—the absence of wage acceleration (in fact, the most recent data show decelerating nominal wages and compensation)—they support a slack story more than a full employment one.

Based on the charts in that post, the above conclusion is quite reasonable.

Dave Altig is skeptical, as am I. Altig says,

But even excepting the legitimacy of identifying the cyclical part of a time-series representation with slack, there is the inevitable ambiguity in exactly how that representation ought to be constructed. I'm not sure exactly how STAMP works, but let me try another cycle/trend decomposition, based on a statistical tool known as the "Hodrick-Prescott filter".

I'm also more familiar with the H-P filter, but I'm not sure it does a whole lot better here. Filtering the monthly LFPR data with lambda=14,400 makes any deviations at business cycle frequencies pretty hard to see. The data is pretty noisy. Month to month variations are pretty small (a couple of tenths of a percent). In other words, I don't take too much solace from the fact that we're currently above trend by this measure. By Altig's chart, it looks like the same could be said of about half the months since the last recession. Indeed, you have to look very hard to see the recession in the deviations from trend.

Now, I don't think that Altig is actually arguing to look at the data this way. It does, however, illustrate what seems to be his real point that it matters how you construct the trend and extrapolate from it. Yep. 'Tis true.

Then he really humbles us (rightfully so)...

We have a way to go on that agenda. I am a macroeconomist by trade, and operate in the realm of the usual macro models that are, in the end, not much different than they type you would find in a typical intermediate-level macro-theory textbook.... The fundamental shortcoming of these models is that they treat the labor market as if it is, more or less, a spot market. They make no allowance for the long-run nature of employment relationships, the zero/one nature of labor force participation, and the like. Furthermore, these are models in which the key variable is hours -- there is no distinction to be made between changes in hours that result from the number of hours an average employee works and the number of people working (the latter being with what labor force participation and unemployment rates are solely concerned).

All of this may be OK if what we are focused on is forecasting GDP growth or inflation. It is obviously a disaster if we are trying to forecast unemployment or some measure of slack that depends on labor force participation rates. To put it another way, I think macroeconomists know even less about labor-market slack than we do about most of the other things that we nonetheless pontificate about at great length. And that is saying something.

True. But lest you abandon all hope, there is a bright side. This is how the state of knowledge advances. With all the attention that is being focused on this issue (more than, say, in 1995 as I recall) we are asking good questions. We are experimenting with different ways of looking at the data and subjecting our analysis to immediate comment and criticism. There's little doubt in my mind that this labor market recovery will be studied in great detail and much will be learned.

A million sounds like a lot. A million here, a million there,... you know the rest. I'm pretty sure that talk about millions of whatever just sounds to most people like "a lot." But economists know that you need to understand the significance of a million (or a billion, or whatever) in context--are we talking gross or net?

We begin with Paul Krugman's latest op-ed: (NY Times)

I used to live next door to a Russian émigré. One day he asked me to explain something that puzzled him about his new country. "This place seems very rich," he said, "but I never see anyone making anything. How does the country earn its money?"

The answer, these days, is that we make a living by selling each other houses. Since December 2000 employment in U.S. manufacturing has fallen 17 percent, but membership in the National Association of Realtors has risen 58 percent.

58 percent is a lot--that is true. In that same period, the BLS figures for the real estate sector show an increase of 150,000 jobs, or about 10 percent of the total jobs in the sector. Obviously, not every job in the real estate sector is accounted for by Realtors. However, the overall increase in the sector is not out of line with the growth in the 5 years preceeding that period. It is blatantly obvious that employment in rental and leasing is practically flat (actually down just a bit) since December 2000. It's probably reasonable to hypothesize that some people who worked in the rental business are now real estate agents. If the housing bubble deflates, I think we could expect that trend to reverse.

Given the interest many of my readers have in the housing picture, I thought you might like these charts. The BLS sure has a great web site.

Krugman continues:

The housing boom has created jobs in two ways. Many jobs have been created, directly and indirectly, by a surge in housing construction. And rising home values have fueled a simultaneous surge in consumer spending.

Let's start with home building. Between 1980 and 2000, which was before the housing boom, spending on the construction of new homes averaged 4.25 percent of G.D.P. In the most recent quarter, however, the figure was 5.98 percent. That difference is equivalent to about $200 billion a year in additional spending, generating roughly two million extra jobs.

He's correct about the $200 billion a year in additional spending. But two million extra jobs? Since he's talking about an increase in the fraction of GDP spent on housing construction, isn't he really talking about two million more jobs in the economy being supported by spending on construction as opposed to being supported by other spending? Fewer jobs are being created from increases in other types of spending. It's gross job creation vs. net job creation, and it includes both direct and indirect effects.

Then there's the jump in house prices. Over the past five years housing prices have grown much faster than the overall cost of living, adding about $5 trillion to the public's wealth. Typical estimates say that each additional dollar of housing wealth adds about 3 cents to annual consumer spending, as families reduce their savings and borrow against their newly valuable homes. So we're talking about an additional $150 billion in spending, and roughly 1.5 million more jobs.

Same story; same multiplier. We're talking about new wealth in the last 5 years contributing to additional consumption spending. Consumption spending kept right on rising during the recession and recovery of the last 5 years while investment and net exports fell. Again, Krugman is really talking about jobs that are now being supported by consumption spending (housing induced) as opposed to being supported by investment or exports. They may be new jobs, they may not be. We are, after all, talking about one sector growing while others are shrinking. Of course, to the extent that during the recession and shortly afterwards total jobs fell, it is probably the case the additional spending was taking up the slack and keeping net job destruction from looking worse. The same is true of construction. Whether these jobs are taking up the slack in the labor market now is less obvious. Again, it's gross vs. net.

I can tell a (really simplified) story that illustrates the point. Imagine a small town--we'll call it Keynesville. Suppose aggregate home values rise by $10,000 in Keynesville and people take out all that equity. They spend it all on consumption goods in town which, after a multiplier process, raises total income by $50,000 (assuming a simple Keynesian multiplier of 5). The increased demand causes a local firm to hire an additional worker for $50,000. A job is created. Of course, if net exports exogenously fell at the same time, a job might be destroyed. Gross job creation (the story Krugman is telling) from the increase in home values only tells half the story.

Does anything else in the U.S. economy rival housing as a source of job creation? Well, there's also the military buildup. The Economic Policy Institute estimates that increased military spending over the past four years has created 1.3 million private-sector jobs.

I looked up the EPI brief whence cometh the 1.3 million estimate. The author cites the administration (OMB), but doesn't give a link and I couldn't find it. If anyone has a link, let me know.

Here's what the EPI says:

Federal, state, and local government spending has created 2.1 million jobs in the last four years (see footnote). According to the administration's budget report, defense spending is generating 1.33 million more jobs in the private sector this fiscal year (for a total of 3.85 million jobs) than it did four years ago (2.52 million). At the same time, government jobs have increased by 760,000. In a fully employed economy, the additional 1.2 million jobs generated by government spending would have caused a reduction in jobs in other parts of the economy. But because the current labor market is so weak, the spending by people holding these 2.1 million new jobs has actually resulted in what is known as a "multiplier effect," leading to the creation of even more jobs in the labor market.

Footnote:

This estimate does not include private-sector jobs resulting from more spending for homeland security, health, education, and other purposes.

The report goes on:

Defense spending gave its largest single-year boost to private-sector jobs in the fiscal year that ended in September 2004. In that year alone, defense spending directly generated almost half a million jobs (495,000). The multiplier effect from that spending no doubt contributed to the other 434,000 jobs added in the private sector that year. By contrast, in this fiscal year, additional defense spending is supporting only 70,000 more jobs, a small fraction of the 2 million private-sector jobs being added this year.

Where do I begin? Estimiting the size of the multiplier here is pretty dicey business. But if you try to assume something reasonable for 2004, you end up concluding that the effect is pretty small this year. Pretty small indeed. This was a temporary effect.

I also question the extent to which the private sector defense jobs created many additional jobs that would not have been created without defense spending. This multiplier is greater than one perhaps, but I'd like to see evidence.

And what of the 706,000 jobs created by the government in the last 4 years? That's about right, in fact. But when you dissect that number, it's not that impressive. Here is the BLS chart:

If my memory serves me correctly, the spike in May 2000 would be temporary census workers. Notice that growth of government jobs has slowed recently. Federal government employment is actually falling.

At the end of all of this, we've used the word "million" a number of times but haven't really pinned very much down. Krugman is talking about housing supporting a few million more jobs out of 134 million, some new, some not. Couple this with the EPI attributing defense spending for 1.3 million private sector jobs, again some new and some not. Some taking up the slack in the labor market, some displacing growth that could come in other sectors. What do we make of all this?

Krugman is using these millions of (gross) jobs created because of changes in the composition of spending (with multipliers flying around everywhere you look) to make a statement about how the labor market is on shaky ground because of these shifts. The real story beneath the story here is that he's saying that the economy is more dependent on housing than before--and that much is true. All those millions of (gross) jobs created directly and indirectly by housing must mean millions of (gross) jobs directly and indirectly lost elsewhere, right?

But the indirect effects, that is, the (gross) jobs created through these multiplier effects (as elusive to track down as they may be) would have been created anyway in all likelihood. I find it a little troubling to talk about multiplier effects resulting from people moving from jobs in the rental sector to the real estate sector. Ditto for an increase in construction's share of GDP.

And where are the jobs that have been indirectly destroyed by the sectors of the economy that are shrinking? There must be millions of them. But you might never know which jobs they are. The restaurant worker whose job was created indirectly by factory expansions 30 years ago might now be earning a living serving real estate agents instead of factory workers. Aside from any social implications beyond the scope of this discussion, he probably doesn't care.

If you believe in any sort of resiliency in the labor market, I think you would want to put less emphasis on the jobs created through the indirect effects. When the pendulum swings in the other direction, some jobs will be directly created, some will be directly destroyed, and a lot of others will notice changes, but they'll see it through.

I have trouble seeing the purpose in using multiplier effects to think about gross jobs created by shifts in spending patterns unless you believe in a lot of labor market rigidity. If the housing boom ends and all those new real estate agents and mortgage brokers are out of work with no other prospects, then we could have problems. But I think the labor market is more resilient than that. It did, after all, survive the pendulum swinging in one direction. I'm confident the pendulum could swing back without breaking.