April 07, 2013
Is the recovery dying?
The Bureau of Labor Statistics reported on Friday that the number of Americans with jobs only increased by 88,000 in March on a seasonally adjusted basis. That's one of the weakest months in the last two years. Although it's clearly a disappointment, I would caution against reading too much into the latest number.
Nonfarm payroll employment is one of the most important and widely followed indicators. Notwithstanding, there's a lot of noise in the initial estimates. The graph below plots the job growth numbers as they were initially reported (in yellow), what they were reported by the BLS to be one year later (fuchsia), and what the BLS is claiming today was the appropriate figure for that month's job growth. The average absolute change between the first two numbers is 52,000. In other words, it would be quite typical if one year from now the BLS tells us that March job growth was really 130,000, or was only 36,000.
For example, last month the BLS had estimated that nonfarm payrolls grew by 119,000 in January and 236,000 in February. On Friday they revised these estimates to 148,000 for January and 268,000 for February. And even once all the revisions are finished, the number is still only an estimate, from which the true value could differ significantly.
Fortunately, there's a lot of persistence in the underlying true employment situation, and a little better information comes from averaging the last 3 months, as shown in the figure below. The BLS currently estimates that the U.S. added 168,000 jobs each month in the first quarter of 2013. That's down from the 209,000 for 2012:Q4 but better than Q3 or Q2 last year. It's a little better than the average of 148,000 that we saw over 2010-2012.
I think that the right take-away from the latest report is that U.S. economic growth is continuing at about the pace it has been-- sluggish growth and high unemployment continue to be with us.
Posted by James Hamilton at April 7, 2013 07:13 AMdigg this | reddit
The data probably capture some deterministic factors beyond the noise component. We did have a fairly stiff increase in the payroll tax and according to a number of accounts that increase was (somehow!) seen as unexpected by the public. So we might be seeing consumers undershooting spending in response to the payroll tax shock. Perhaps it's because the payroll tax cut was deliberately implemented in such a way that people wouldn't notice it when their paychecks increased (and thereby led to higher spending); but the downside of that strategy is that restoring the payroll tax apparently caught those same consumers by surprise. And the March numbers are likely reflecting the anticipated effects of the sequester. For example, in February DoD started canceling FY2013Q3 and FY2013Q4 planned inductions into overhaul programs at its depots. That started a reaction down the supply chain as replenishment orders for repair parts were cancelled. And over at USDA some of the meat inspectors operating under temporary contracts were released in March, which caused some cutbacks in meat and poultry plants. And I don't think the unusually bad weather in March helped any either.
Posted by: 2slugbaits at April 7, 2013 08:06 AM
Let's not forget that these are based on surveys and are therefore subject to sampling error. Right now the BLS estimates the 90% confidence interval for the change in nonfarm employment is about plus-or-minus 90,000.
Posted by: Brian at April 7, 2013 08:34 AM
Still trying to find the same ready for use economic data and graphs within the ECB or peripheral public organisations based in Europe. Still one may use the Fred data as a proxy when keeping in mind that growth in USA is offering a positive differential with the European zone.
Money velocity M2
Whilst there is considerable literature with reference to the money velocity and employment, nothing decisive comes in mind. Samuelson "you can take a horse to the river but cannot force him to drink"
Some stickiness in prices may be the best read.
Posted by: ppcm at April 7, 2013 10:01 AM
I do see the recovery is being limited now by the dynamics of Effective demand. Since you put the post above in terms of employment I will offer this link...
To put the troubles of the economy into even a broader perspective, I wrote this post today...
My research pinpoints limits to the recovery... and we are reaching those limits. If you don't use my theory, you in effect feel for the limits of the recovery like a blind person walking in the dark. I can see the limits and I can see how close we are to those limits. We are getting close. I did some calculations last night using 1st quarter values for capacity utilization and unemployment, and one thing is for sure, we have gotten closer to the limit of the recovery even if the labor share number increases from 4Q-2012. The likelihood though is that it will come out less. Bottom line... The recovery got closer to the limit of Effective demand in 1Q-2013.
Posted by: Edward Lambert at April 7, 2013 10:49 AM
Gallup's self-reported hiring activity is holding at around a 13-month high but at the level of the onset of the Great Recession and before the financial markets implosion at the onset of the debt-deflationary regime in '08.
CBO reports a surge of 9% yoy in social insurance receipts, which is attributable to higher federal taxes.
Growth of state and local gov't income tax receipts of 4% yoy against reported yoy wage and salary growth of 2.7% is consistent with the reported 1.4% yoy growth of non-farm employment, Gallup's self-reports, and 4% yoy nominal GDP.
However, with no or negative federal gov't spending and hiring (cuts to war spending and unemployment payments) vs. 5-6% long-term growth, the loss of gov't spending implies a net 1% reduction in the post-'00 trend rate of ~3.6% and post-'07 trend of ~2%.
At the trend rate of wage and salary growth and reported labor productivity, non-farm employment is growing above capacity on a secular trend basis and thus is poised to decelerate and contract sometime in '13 with a recession. Tax policy (tax increases, Obamacare, etc.) is reducing capacity whereas housing policy is pulling in consumption from out years, precipitating yet another price bubble that will have to correct with less growth in the intermediate term.
Keynesians will blame the recession on the cut (reduction in growth) in gov't spending, and the supply-siders will blame it on tax increases, neither of which will help; but beyond that the persistent structural causes are too much public and private debt/GDP, Boomer demographic drag effects, and energy cost constraints.
Posted by: Bruce Carman at April 7, 2013 11:05 AM
Mein liber Gott! I actually agree with Bruce Carman. There is no easy fix. Gotta suck it up.
Posted by: c thomson at April 7, 2013 03:32 PM
I generally agree, but there's two points worth adding:
A large part of the difficulty of making early estimates is that US employment is very seasonal, and seasonality patterns seem to have fluctuated enough in recent years that the BLS can't accurately adjust for seasonality. For three years running we've been told every winter that the job market is improving, and every summer that it's getting worse. With the benefit of hindsight, it's obvious the seasonal adjustments were just wrong and the trend was actually quite stable. I'm pretty sure the same thing is happening this year.
In such situations you have to fall back on year-on-year comparisons. You compare this March to last March or this February to last February, instead of this March to this February or this February to this January.
The year-on-year trend peaked in Feb-Mar 2012 with over 2.4m jobs added during the trailing 12 months. That number has been winding slightly down ever since and with this March 2013 report, we are looking, very preliminarily, at less than 2m jobs added over the trailing 12 months.
In other words, this March report is indeed very preliminary, and should indeed be taken with a big grain of salt. It is very unlikely that anything important changed between February and March. But the news you've been reading about another strong winter is nothing more than poorly seasonally adjusted data. The job market has actually been very gradually softening since early 2012.
Posted by: Tom at April 7, 2013 06:26 PM
Why should 'job growth' be any stronger though? If the the economy grew to much between 1996-2007, then it is right on target.
That is the point when looking at typical private sector patterns. If the upwardly revisions say there was 200,000 jobs in the first quarter, eh so what?
Posted by: The Rage at April 7, 2013 06:48 PM
"At the trend rate of wage and salary growth and reported labor productivity, non-farm employment is growing above capacity on a secular trend basis and thus is poised to decelerate and contract sometime in '13 with a recession. Tax policy (tax increases, Obamacare, etc.) is reducing capacity whereas housing policy is pulling in consumption from out years, precipitating yet another price bubble that will have to correct with less growth in the intermediate term"
That whole statement made no sense. Your trying to hard and crossing your self up.
There is no reduction in capacity nor any price bubble in RE happening hence there is no consumption to pull foward.
Credit by nature pulls consumption foward and that has little to do with Real Estate except when it is targeted for expansion. Auto's have become the big target and booming areas of the US will drive RE. But I see little push of mortgage products, indeed, they are still in recession.
My guess 300,000 jobs probably were created in February. My guess 120,000 jobs were created in the Easter slackened March. Couple with most of the layoffs occurring in March for the stupid sequester, you get a weaker than 'expected' number.
Big deal. The layoffs will pass, already seeing a surge in job ads in April, what you expect after the Easter impacted March.
In otherwards, move on, nothing to see here.
Posted by: The Rage at April 7, 2013 06:57 PM
@The Rage, real yoy house prices in many of the formerly bubbly areas are again back to 15-20% bubble-like rates as occurred in the late '80s and mid-'00s. Gov't-sponsored no-doc refis/workouts, banks deferring foreclosures and holding properties off the markets, and big-money all-cash buyers to rent or flip properties are creating an artificial floor for house prices above the levels at which wage and employment growth will support.
The difference of yoy change of full- and part-time employment weakened significantly in Mar., whereas the yoy change of full-time employment is at a recessionary rate historically.
Gov't hiring constraints and Obammacare taxes risk pushing full-time employment growth negative yoy in the months ahead.
Thus the US economy is indeed at or near capacity constraints for further employment growth, exspecially private full-time employment.
Posted by: Bruce Carman at April 7, 2013 08:45 PM
The bad news was fewer than needed new jobs. The good news was the unemployment rate went down. The bad news was that fewer people were looking for jobs so the unemployment rate went down. The good news is the Obama is focused on gun control.
Unemployment Rate (1)
7.9 7.8 7.8 7.9 7.7 7.6
Change in Payroll Employment (2)
160 247 219 148 (P) 268 (P) 88
We all have our priorities.
Posted by: Bruce Hall at April 8, 2013 07:00 AM
"Keynesians will blame the recession on the cut (reduction in growth) in gov't spending, and the supply-siders will blame it on tax increases..."
Not all tax cuts are supply-side oriented. In fact, the payroll tax was a Keynesian (demand side) tax cut. It was implemented to help alleviate our AD problem, not to increase AS.
Granted one can argue that the increase in capital gains will have supply side ramifications, but I doubt they have too much drag on the current economy.
Posted by: A Common Anomaly at April 8, 2013 09:27 AM
February JOLTS data comes out tomorrow(4/9) at 10AM.
Through January -
Chart 2 is interesting.
Note how job openings and hires lead employment.
Note how job openings and hires are starting to flat line.
Hires remain above separations so employment continues to grow. The fact that the rate of hiring is slowing is another bit of evidence that strong employment growth will not return in the near future.
Chart 4 is also interesting. The outward shift in the Beveridge Curve since 2009 (the purple section) indicates a structural change in the labor market relative to the 2000 - 2009 period.
Posted by: tj at April 8, 2013 09:39 AM
@Bruce Hall, it's better than you say, brother. We're about 10-11 years behind Japan, but we're turning more Japanese with each passing year:
Be patient, it just takes time.
Full-time private employment is a whopping 27% of the population, which is where we were in 1987:
The number of employed men age 16-24 as a share of population is at a 54-year low at about half the level of peak Boomers (really bullish for housing!!!):
Age 16-34 is 66% of the peak Boomer level:
Better yet, total US credit market debt owed per full-time private employee is only $660,000:
The imputed compounding interest costs to that debt/employee to perpetual term is only about an equivalent of 18 times an employed American's avg. annual wage/salary.
At the rate of growth of total credit market debt to full-time private employee, we'll be at $1 million in total debt/employee by decade's end! And that will be just 28 times today's American's avg. annual/wage salary today.
This is yet more evidence of just how wildly "productive" US full-time private employees are to be able to support such "rich" debt levels.
Buy sunglasses futures! The future is bright for such a wealthy nation as the US.
Posted by: Bruce Carman at April 8, 2013 10:22 AM
@tj and all,
Hires to total nonfarm:
Annual change and U rate:
Quits to nonfarm:
Annual change rate:
Hires less quits to nonfarm:
Openings to hires + quits:
Openings less quits to nonfarm:
Annual change of part-time employment and the U rate:
Posted by: Bruce Carman at April 8, 2013 11:36 AM
The subprime US economy shifts into high gear again, this time with autos and student loans:
Consider what auto sales would be without subprime auto loans.
Think about what post-secondary enrollment would look like without the gov't sponsoring lending to the working class to borrow tens of thousands of dollars they can't afford to pay back to study for credentials that do not prepare anyone for paid employment.
Perhaps we can train millions of young university-educated people to be commissioned sunglasses salespersons to sell glasses at $1 million a pair (extra for anti-glare and scratch-resistant lenses) to the top 0.1-1% whose view is blinded by their bright prospects at the expense of everyone else. The federal gov't could tax sunglasses at $950,000/pair and pay the salesperson $45,000/year and $1,000/pair in commission. Imagine the sunglasses productivity we could achieve! After a couple of lifetimes, the commissioned sunglasses salesperson's great, great grandchildren might even be able to afford a pair for themselves.
Posted by: Bruce Carman at April 8, 2013 03:27 PM
It's also worth reminding that weekly initial jobless claims look worrisome so far in March:
In addition, the ISM Manufacturing Index for March posted a substantial slowdown:
There's still room for doubt about the overall trend for March (Jan & Feb by comparison look quite encouraging). We'll know more, perhaps much more, in a few weeks. But it's premature at this point to dismiss the payrolls slowdown, although Professor Hamiliton's points about not rushing to judgment are still spot on... for now.
Posted by: James Picerno at April 8, 2013 05:11 PM
@James, no worries, dude; it's all good. We don't need yoy employment growth with the major stock market indices at or near all-time highs, right?
The "wealth effect" will save us. Just lever up your retirement account and bet the underwater suburban house, marriage, and your kids' overpriced university credential on Russell 2000 call options. Bubble Ben Shalom Zimbabwe has got your back, Jack.
Get your ticket on the Bernanke-Russell crazy love train to permanent-plateau prosperity. Millions of people are dying (financially) to get onboard.
Yeah, you go, Bubble Ben! We're all betting that you will print bank profits to permit levered dark pool bank prop trading call options on equity index futures in perpetuity. Bubbles be damned!
I cain't get no satisfaction without Bubble Ben Bernanke's permanent Fed member banks' ZIRP balance sheet liquidation.
Posted by: Bruce Carman at April 8, 2013 06:07 PM
Small business optimism gets smaller:
Self-employment and as a share of non-farm employment:
New small business growth is the source of net new employment, and business starts and self-employment continue in a secular decline, suggesting that there is no structural impetus for sustained growth of employment in the US hereafter; quite the contrary.
Posted by: Bruce Carman at April 9, 2013 06:37 AM
Interesting how Peak Oil is repeatedly documented as a main factor or even the probable cause of recession/depression, then repeatedly forgotten as such, all on the same site.
Posted by: Robert H at April 9, 2013 02:05 PM
Headline U is but one measure; the number working part-time (1-34 hours) for economic reasons fell by 350,000 and since the number "normally working" part-time fell, I infer that these individuals returned to full-time work. That said, even correcting for this shift, employment only just kept up with population growth (by my reckoning, after correcting for age composition, by 52K, which is as the comments stressed statistically insignificant).
Posted by: mike smitka at April 10, 2013 06:02 AM
Stuck in Lodi again:
Posted by: Bruce Carman at April 10, 2013 03:24 PM