Friday, April 6, 2012

Jobs growth didn't slow down in March

Did private non farm employment grow in March by only 121K (per the establishment survey), or by a very impressive 318K (per the household survey)? Take your pick: no one really knows which one is right, even though they paint dramatically different pictures of how strong the economy currently is.

In the 30+ years during which I have been following these two monthly surveys, there have been many times when they have diverged. Over time they tell the same story, but over shorter periods of a few months or even a few years, they can tell different stories. One shorthand way of resolving the problem of a divergence is to simply split the difference, since the truth is likely to be somewhere in the middle. Doing that for March gives you a private sector payroll gain of 225K, which is right around where expectations were, and which is also consistent with what we've seen in the past few months. I don't see any reason to think that the unexpected slowdown in jobs growth that surfaced in the establishment number reflects any actual slowdown in the economy; the economy never turns on a dime without there being a number of indicators suggesting that something big is going on.

If markets are going to post a big reaction to this number, then they will be mistaken. The truth is most likely that there wasn't any jobs slowdown in March at all.

Looking at the chart above suggests another way of resolving this divergence problem. Note that the growth of private sector jobs from early 2010 through just a few months ago according to the household survey (red line) was slower than the growth according to the establishment survey (blue line). In the past two months the household survey has made up for that lagging performance by posting growth of 740K jobs. Since early 2010, the household survey now reports the addition of 4.4 million jobs (164K per month on average), while the payroll survey reports just under 4.1 million (150K per month on average). That's pretty close, and it's certainly good enough for government work.

I note also that this recovery has been unusual in the sense that, until recently, the household survey was showing weaker jobs growth than the establishment survey, when the typical pattern has been that the household survey tends to report faster growth coming out of a recession. That happens because the household survey is better at picking up new jobs that are being created by new, smaller companies that are not included in the establishment survey. This recovery was an anomaly, but now we seem to be back on track since the household survey is once again in the lead.

Much ado about nothing: the economy is most likely still growing at a 3-4% pace.


Jake said...

You're using the wrong figures. Seasonally adjusted Household figures show a decline in March (not a slowing, a decline)

Benjamin Cole said...

“We see a lack of sustainability in terms of strong job growth,” Tony Crescenzi, a strategist at Pacific Investment Management Co. in Newport Beach, California, said in a radio interview on “Bloomberg Surveillance” with Tom Keene and Ken Prewitt. “This is still not strong enough to create escape velocity, which is to say an economy strong enough to make it on its own without additional monetary stimulus from the Federal Reserve.”

I sense GDP is growing more in the 2 percent range, not in the 4 percent range.

QE3 we need, and not a timid display of it.

Things have changed since the 1970s. A lower exchange rate for the dollar helps our economy. Yet with global trade and a non-unionized private-sector work force, inflationary pressures are muted.

We have had a real estate bust, and property values need reflating. Japan did not reflate after their bust, and they never really recovered, becoming a backwater nation eclipsed by China.

Friedman advised Japan to monetize their debt through QE, print money. They didn't listen. But they did have 15 percent deflation with plenty of liquidity and zero interest rates.

William said...
This comment has been removed by the author.
Hans said...

Those that have stopped seeking employment has reached an all time high...

William said...

Gross Domestic Income (GDI) perhaps sheds some light on this divergence.

Analysis: Alternate growth gauge points to more U.S. job gains

"GDP expanded at a 1.8 percent annual rate in the third quarter and by 3.0 percent in the fourth quarter. In contrast the alternate measure, Gross Domestic Income (GDI), which measures the income side of the growth ledger, expanded at a much faster pace of 2.6 percent and 4.4 percent over the same periods. While they are both measures of economic activity, they are calculated from different data sources.

"Economists, including those at Goldman Sachs and at least one at the Fed, believe that GDI is more useful for trying to gauge the economy's underlying momentum. And it would point to sustained job growth. In a paper published in 2010....Fed economist Jeremy Nalewaik highlights the importance of GDI as a growth measure....GDI is published with the third revision of GDP, which falls two months after the initial estimate. It is calculated using corporate profits, compensation and proprietors income data, and updated based on tax returns."

What do think Scott?

Benjamin Cole said...

The market has been soaring; will an earnings slowdown inject a dose of reality?
Washington Post - ‎1 hour ago‎
For the stock market, it was a triumphant first quarter. But for earnings growth, the past three months were just ho-hum. Analysts are expecting earnings for companies in the Standard & Poor's 500 index to decline 0.1 percent compared to a year ago,....


The WaPo is right, earnings are going flat y-o-y.

And employment is soggy---sheesh, I can remember when we generated 200k+ jobs every month, matter of course (1990s). Now we say 100k is okay. While mired in a slo-go mushy economy.

It may be time to consider QE as a conventional tool, not exotica. With no earnings growth, the stock market goes flat, and property already flat.

Then investors and consumers start pulling in their horns.

The world has changed in the last 40 years. Fighting inflation may not be enough for central bank policy. They may also have to start fighting for growth.

McKibbinUSA said...

The US employment to population ratio rose in March 2012 on both a monthly and annual basis -- this is the first dual rise in the employment ratio in years -- the rise in the employment to population ratio is good news for workers!

McKibbinUSA said...

PS: More at:

Scott Grannis said...

Jake: you are looking at the total, I'm looking at private sector only. Government jobs shrunk by a lot in march.

Anonymous said...

I see Scott's point, but I'm not getting the same number.

The Series ID I'm looking at (from the household survey) is LNS12032189, "(Seas) Employment Level - Nonagriculture, Private Industries Wage and Salary Workers".

February's number was 110,778.
March's number was 111,114.

That's a gain of 336K, not 318K. And doing a quick graph of the data on an Excel spreadsheet does look like it's made a big jump the past few months.

Scott Grannis said...

Re: alternate growth gauge (GDI). I see this as one of the reasons to suspect that the economy is not weakening like the payroll number suggests.

unknown said...

What is the BLS or Bloomberg code for household employment private? thanks!

Scott Grannis said...

I have traditionally taken my data from BLS Table A-8. I simply subtract "Government" from "Nonagricultural industries" (sa)

unknown said...

2 mln private sector jobs created in Q1!!!

Squire said...

I do my own seasonal adjustments by comparing year to year, then month to month change of that year to year figure, then compare that to the previous year pattern and determine how much better or worse this year is than last.

The NSA Household and Establishment is so discrepant I am staggered. If my methodology is good, it means that small business is hiring at an explosive rate and established business which fill out forms each month for the government are actually cutting back.

It is nonsensical to me. I must disregard employment in its entirety.

marcusbalbus said...

don't you ever stop being panglossian?

William said...

Panglossian - from French, Dr. Panglosse, name of the optimistic philosopher and tutor in Voltaire's Candide. A person who views a situation with unwarranted optimism.

But who is the you in the "don't you"? Dr William J McKibbin, unknown, William,...??

McKibbinUSA said...

Here is another view of the US employment to population ratio since 1948 in the form of a heat chart:

In truth, the US employment to population has still not dropped to the levels experienced in the 1950's and 1960's -- far from it...

William said...


April 6, 2012: Weekly Leading Index (WLI) Rises

"A measure of future U.S. economic growth rose in the latest week, as did the growth rate on an annualized basis, a research group said on Friday.

"The Economic Cycle Research Institute said its Weekly Leading Index rose to 126.5 in the week ended March 30 from 125.8 the previous week.

"The index's annualized growth rate rose to 1.0 percent from 0.0 percent a week earlier. It was the highest level for the index since August 5, 2011 and the highest level for the growth rate since August 12, 2011 according to ECRI.

BamBamFunkhouser said...

What employment survey are you looking at? The BLS Household survey showed a LOSS of 31k employed persons in March.

BamBamFunkhouser said...

""t may be time to consider QE as a conventional tool, not exotica. With no earnings growth, the stock market goes flat, and property already flat. ""

That's a recipe for galloping inflation. You have 100% confidence that the Fed can reverse QE at the very right moment and in the very right amount so as to avoid either out of control inflation or a deep recession.

Hans said...

Could someone please explain what established and household survey means?

Thank you kindly!

Hans said...

"Jobs Report at a Glance

Here is an overview of today's release.

US Payrolls +120,000 - Establishment Survey
US Unemployment Rate dropped .01 to 8.2% - Household Survey
Average workweek for all employees on private nonfarm payrolls fell .1 to 34.5 hours
The average workweek for production and nonsupervisory employees on private nonfarm payrolls was unchanged at 33.8 hours.
Average hourly earnings for all employees in the private sector rose by 5 cents.

Recall that the unemployment rate varies in accordance with the Household Survey not the reported headline jobs number, and not in accordance with the weekly claims data."