Which is Wrong: Data on Growth or Employment?
The WSJ's Greg Ip is rapidly becoming a proverbial two-handed economist. Over the past week, he has written not one but two articles on the ambiguous nature of the macro-data that has been coming forward over the past few quarters.
The first article was last week's Fitting Growth Data Into the Unemployment Puzzle and is timely, with prelim GDP data out tomorrow. This is fascinating issue I have been watching for many quarters now: When two (allegedly) reliable historical data points are in direct conflict, which one is correct? What should get more emphasis, and which should gets less?
Consider:
"When the Commerce Department publishes first-quarter growth estimates Friday, it will accentuate a puzzle with implications for interest rates, growth and U.S. living standards: With the economy growing so slowly, why is unemployment falling?
Explanations range from measurement problems to the peculiar behavior of construction jobs to the possibility that productivity growth is easing.
Wall Street estimates the economy grew 1.9% at an annual rate in the first quarter, the slowest in more than a year. That would bring growth for the past four quarters down to a sluggish 2.2%, below most economists' estimates of the economy's potential growth rate, which hovers around 3%. Potential is the rate at which the economy can grow over the long term, given growth in the work force and worker productivity.
A year of below-potential growth ordinarily would push up unemployment, because the economy isn't growing fast enough to employ all the new entrants to the work force. Instead, the unemployment rate has fallen to 4.4% in March from 4.7% a year earlier.
The answer to this puzzle has huge implications for inflation and interest rates. For the Federal Reserve, the most troubling explanation is that potential growth is closer to 2% than 3%. That may be because the work force is growing more slowly for demographic reasons, or because productivity growth has slowed, perhaps because companies are investing less in labor-saving technology."
There are several potential answers to this puzzle:
1) Growth is not slowing, GDP measures ares all wrong.
2) Unemployment is not falling, BLS measures are incorrect.
3) The relationship between GDP and Unemployment have decoupled.
We know that growth overseas has been nothing short of robust -- and perhaps globalization is seeing some spillover into the US economy in ways GDP doesn't measure. That's one possibility.
While the Unemployment Rate has notched down, new unemployment claims have moved higher; At the same time, the exhaustion rate -- the measure of those people who have used up all their unemployment insurance benefits but have not found a job -- is also higher. Another explanation is that the labor participation rate is falling as some unemployed leave the labor market entirely.
Regardless of the explanation, these two data series seem to be in conflict. Traditionally, there should be a greater correlation between the two, and this implies something is off: Either Growth is better than reported, or Employment is worse than reported.
~~~
On a related (two-handed) note, Ip follows up this morning with another interesting quandry: Why Market Strength And Economic Growth Don't Always Line Up. He pits Merrill Lynch's David Rosenberg against ISI's Ed Hyman -- thats a heavy wight battle for ya -- on what this means to the market:
"Economists differ. David Rosenberg, chief North American economist at Merrill Lynch & Co., thinks the market is misleadingly optimistic. He says there is a "disconnect between how the economy is doing and the way the equity market is doing." The U.S. economy has just completed four quarters of annualized growth below 3%, which, he says, has never happened in 60 years without being followed by recession. While he doesn't forecast one, he puts the risks higher than generally realized.
But Ed Hyman, chief economist at ISI Group, a New York investment dealer, says the market has it right. Just as in 1985 and 1995, the Federal Reserve has raised interest rates enough to slow the economy and bring inflation under control. That reassures investors that even higher rates won't be needed later that could tip the economy into recession."
The full piece is worth a read . . .
UPDATE: April 26, 2007 10:06am
A perfect example of both the Employment/GDP issue as well as the Global/Domestic divergment is Monster Worldwide -- they reported their quarterly profit fell -- their sales rose 28%, in large part due to a 64% jump in international careers business. Overseas sales is now over a third of their revenues . . .
>
Source:
Fitting Growth Data Into the Unemployment Puzzle (free WSJ)
Greg Ip
WSJ, April 21, 2007; Page A2
http://online.wsj.com/article/SB117710828774477368.html
Why Market Strength And Economic Growth Don't Always Line Up
GREG IP
WSJ April 26, 2007; Page C1
http://online.wsj.com/article/SB117755454840982990.html
Thursday, April 26, 2007 | 07:15 AM | Permalink
| Comments (47)
| TrackBack (0)
add to de.li.cious | 
digg this! | 
add to technorati | 
email this post
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c52a953ef00d8346e5e6453ef
Listed below are links to weblogs that reference Which is Wrong: Data on Growth or Employment?:
Comments
BR said: "Another explanation is that the labor participation rate is falling as some unemployed leave the labor market entirely."
This is right on the money. While I am not convinced that a Wealth Effect exists that drives consumer spending, I do think it impacts the workforce. The stock market goes up, people take early retirement. We get a three year bear market and many of those folks look to go back to work. As the markets go back up, people leave the work force again (maybe they had two jobs and now only have one, or one spouse in a family stays home).
I suspect we are seeing a lot of withdrawal from the workforce right now. Pessimists see this as a jobless recovery where folks simply give up looking for a good job. Optimists see it as an improvement in living standards.
Posted by: REW | Apr 26, 2007 7:31:00 AM
In a country with 11 million illegal aliens at work, it is dubious to claim that there is mass unemployment that is hidden by those nefarious bureaucrats at BLS. It is very difficult to find help in my neck of the woods, which is not so unlike most areas of the country.
I'm troubled, but not surprised, that some people can more easily believe in conspiracy theories about economic statistics than they can in full employment. After all, 40% of people believe in alien beings flying UFOs over Nevada.
Posted by: Nova Law | Apr 26, 2007 7:48:17 AM
Nova Law, did you know "free checking accounts" add to GDP? What about owner's imputed rent?
I mean, if you own your home, EVERY year it gets computed into GDP how much you'd have to pay to rent that same home. Ugly statistics, aren't they? So if more people own their homes, miracle, GDP is permanently higher since they are paying all those rents to themselves.
Man, Casey Serin is pushing up GDP strongly, as a matter of fact. At one point he was paying himself rent on 8 homes!
Posted by: Incognitus | Apr 26, 2007 7:58:50 AM
REW has a good point, but I think the basic reason the statistics seem out of whack is because we haven't had realistic, accurate statistical reporting from the government in years. All of these numbers (like inflation ex-inflation) are computed using politically massaged methods. On top of that, revisions have become larger and large in relation to the first-reported number. This should be a major red flag to anyone who thinks there is any accuracy or scientific value involved. (Whoops, that 75 degree temperature we reported this morning? It was actually 62 degrees. Sorry.)
If all able, working age people who aren't working are included (see U6 in the quote below, just to start), I think the actual rate in this job-free 'recovery' is probably above 10%, and you can tack on another couple of percentage points if you count unemployed illegal aliens. If they're still here and not working, that might be a better measurement of the country's real workforce.
To quote John Williams:
"Suggesting that the household [unemployment rate] survey is more accurate than the payroll survey, however, does not mean the household survey accurately depicts unemployment. While its measures have definable statistical accuracy, the accuracy is related only to the underlying questions surveyed and to the universe of people surveyed.
"The popularly followed unemployment rate was 5.5% in July 2004, seasonally adjusted. That is known as U-3, one of six unemployment rates published by the BLS. The broadest U-6 measure was 9.5%, including discouraged and marginally attached workers.
"Up until the Clinton administration, a discouraged worker was one who was willing, able and ready to work but had given up looking because there were no jobs to be had. The Clinton administration dismissed to the non-reporting netherworld about five million discouraged workers who had been so categorized for more than a year. As of July 2004, the less-than-a-year discouraged workers total 504,000. Adding in the netherworld takes the unemployment rate up to about 12.5%."
Also, keep in mind the nature of the jobs being created. Job 'growth' is always heavily weighted in government and service businesses. Without government bureaucracy and its continuing bloat, even the reported numbers would look a lot worse.
Posted by: dark1p | Apr 26, 2007 8:04:53 AM
BR,
You state:
Traditionally, there should be a greater correlation between the two, and this implies something is off: Either Growth is better than reported, or Employment is worse than reported.
But a few chapters above you hypothesize that one explanation is that GDP and employment have decoupled.
The numbers you are looking at are US GDP and US employment. I would propose that we consider the GLOBAL economy, which has been becoming larger in relationship to the US.
Perhaps the title of your next post should be called "The Decoupling"... or maybe just leave that to Tom Freidman.
Posted by: grodge | Apr 26, 2007 8:13:18 AM
Ocham to the rescue?
Perhaps Ip is demanding too much from the data. Perhaps if we look just a bit harder for evidence that things aren't too out ofu whack, we'll find that things aren't too out of whack. Ip's job, after all, requires a good story, so Ip may have a bias toward seeing things out of whack when they aren't.
There is that other "O" guy (what's his name?) who has a rule about the link between GDP around trend and changes in the jobless rate. That is at the heart of Ip's point. The latest jobless reading of 4.4% was first posted in this cycle 6 months ago, and the 3-month average has been quite stable for 5 months. So while a look at the jobless rate chart can give the impression the rate is still falling, in the period of interest to us - the 3 below trend quarters for growth - the jobless rate has flattened out. We know that most labor market stats are coincident to lagging. If the jobless rate bottomed starting in the second of 3 quarters of below trend growth, in what way do we have a big disconnect? We can also see a modest slowing in the average rate of hiring. That may be due to tight labor supply, but perhaps also to cooling demand. Won't know about that until we see compensation growth some quarters hence.
So I have to go with Nova. Even if the official data are bad, one must argue that they are getting worse, in a systematic way during the period in question, in order to account for a disconnect between growth and labor stats. There is a disconnect between compensation and productivity, but I'm not convinced that the disconnect Ip has pointed is real.
Posted by: kharris | Apr 26, 2007 8:18:48 AM
Both the CPI and GDP measures have been changing over time.
CPI has had changes that lower it, GDP has had changes that make it higher.
This means you can't compare today's CPI with the CPI in the past. And you can't compare today's GDP with the GDP in the past.
And you can't make international comparisons either. Most countries don't use hedonic deflators, owner imputed rent, free checking accounts, etc.
With a lower CPI for the same price action as in the past, and a higher GDP for the same economic activity as in the past, that means it's much harder to have a recession "in the numbers" today, than it was in the past.
Posted by: Incognitus | Apr 26, 2007 8:21:49 AM
Jacques Barzun defines a 'decadent' institution as one that is no longer capable of doing the job that people expect it to do. Sad to say, but many parts of the US government are there today. The can no longer be relied upon; they are politicized, which in turn means 'bought' or they have grown their bureaucracies to a point of collapsing under their own weight. You'd hope that things improve over time - that is what most Americans traditionally believe. But it is not true.
Posted by: wally | Apr 26, 2007 8:48:42 AM
From Pisani we learn that 48% of SP500 profits come from overseas. Also, US markets have a many foreign companies listed. It should not come as a surprise then that the US economy and US markets are not lining up.
Posted by: Lauriston | Apr 26, 2007 9:04:31 AM
"Jacques Barzun defines a 'decadent' institution as one that is no longer capable of doing the job that people expect it to do. Sad to say, but many parts of the US government are there today. The can no longer be relied upon; they are politicized, which in turn means 'bought' or they have grown their bureaucracies to a point of collapsing under their own weight...."
My never-gonna-happen solution to fixing the government is a total ban on political advertising on broadcast media. Politicians take money (bribes) to pay for TV time, and have to reward the crooks that made big campaign donations. If politicians could not buy TV time, it would make them correspondingly less corrupt. How much can you get from a TV soundbite, anyway? How about position papers on the web and some number of non-commercial debates on TV among the candidates?
Posted by: Tom B | Apr 26, 2007 9:08:46 AM
wonder how many readers of this blog who put most of their investments in cash are kicking themselves right now for listening to the big picture too literally and missing out on sizable gains. silly rabbit.
Posted by: Richard | Apr 26, 2007 9:09:13 AM
I agree with Lauriston. Barry, maybe your "big picture" isn't big enough?
Sure, the numbers can and will be massaged for political benefit, but also the numbers are way more complex in a global economy. Conclusions based on home prices in Kalamazoo Michigan, one Toyota quarter or capex spending trends in the US can be misleading in the really "big picture."
So far this quarter, S&P 500 companies are reporting earnings 6.2% above estimates with only a few companies missing their estimate.
Unless all the S&P is in on the ruse, it looks like someone is earning money. Unlike the 2000 tech meltdown, PE's today are reasonable, are they not?
Is DOW 13,000 and SPX 1500 a bubble or a reflection of earnings based on a global economy?
Get bigger.
Posted by: grodge | Apr 26, 2007 9:15:30 AM
Richard, the same could be said of anyone being left outside any Ponzi scheme, before it imploded.
Posted by: Incognitus | Apr 26, 2007 9:18:03 AM
"So far this quarter, S&P 500 companies are reporting earnings 6.2% above estimates with only a few companies missing their estimate."
grodge, most companies beat expectations even in the worst of times. You ought to ask yourself "why".
"Unless all the S&P is in on the ruse, it looks like someone is earning money. Unlike the 2000 tech meltdown, PE's today are reasonable, are they not?"
No, PE's are not reasonable for the top of an economic/debt cycle. At the top of a cycle multiples are usually very low. Just check any cyclical industry, for instance steel now, or homebuilders back in 2005 (every one of them had a PE below 10 back then).
Posted by: Incognitus | Apr 26, 2007 9:21:35 AM
"So far this quarter, S&P 500 companies are reporting earnings 6.2% above estimates with only a few companies missing their estimate."
Traditionally the stock market looks ahead, not back. Why is it different today?
Posted by: wally | Apr 26, 2007 9:23:04 AM
Barry...are we, in your opinion, still in a secular bear market? (as index after index break out). I am curious at what point you change that view.
Posted by: Fred | Apr 26, 2007 9:24:37 AM
Taking care of some buildings I work with 3 contractors - roofing, construction and painting. Starting with roofing 6 years ago and finally this year painting, none of the business owners have any employees anymore.
They all either hire a "Crew" from a company or "Rent" their workers from a rental company.
The people I meet who are doing the work are pretty much the same people as they were in years past.
Why? I ask.
No unemployment or workmen's comp insurance cost or risk. The workers are all "hired for the day" even though the project make take months.
Sure - the people-rental fees include government fees but some fees would change by the history (layoffs or injuries) of the employer which could be a big swing for a small employer - that does not happen anymore.
The workers are now self-employed contractors who are fired daily thus never really laid off.
No immigration risk. The company that rents out the staff does all the employment paperwork and takes the hit for "illegals".
NET: none of these people will be unemployed when not working.
++++
How big a drop in employment?
From an article about money flows through central banks in the WSJ this past Monday:
"Monthly remittances from the U.S. to Mexico have dropped every month since their peak of $2.6 billion in May 2006 -- shortly before new- home construction in the U.S. plunged. In February 2007, the latest month for which data are available, remittances to Mexico had slowed to $1.7 billion."
Posted by: JRip | Apr 26, 2007 9:27:44 AM
I think we should be looking at income statistics (including benifits) as a measure since unemployment is so low. Meaningful employment is probably in the crapper.
Posted by: LAEF2 | Apr 26, 2007 9:33:25 AM
I think BLS excluding persons on active duty in the Armed Forces can explain low unemployment and excluding "all inmates of institutions" from the labor force pool -- both have spiked lately.
"Who is not in the labor force?
Excluded are persons under 16 years of age, all inmates of institutions and persons on active duty in the Armed Forces."
http://stats.bls.gov/cps/cps_htgm.htm
In addition, according to BLS the data is unreliable "at economic turning points or during periods when there are sudden changes in trend". Currently, we are at "economic turning point"; therefore, BLS estimates are incorrect (according to BLS).
"The most significant potential drawback to this or any model-based approach is that time series modeling assumes a predictable continuation of historical patterns and relationships and therefore is likely to have some difficulty producing reliable estimates at economic turning points or during periods when there are sudden changes in trend."
Posted by: V L | Apr 26, 2007 9:40:18 AM
Incognitus,
Thank you for your comment. I'm not an economist-- as you could likely figure.
Are the cyclical indicators of US housing and US employment and even US GDP appropriate when looking at the global econmy, or should they be tempered by other numbers, say, the Shanghai housing index or the Pac Rim employment data?
Granted, many of these numbers are not available and much of the conjecture is based on unknown worldwide data.
For example, Domestic home building may be in a bear market, but worldwide steel may be in the midst of a bull market, in other words, they are decoupled.
The US consumer may be at "the top of an economic/debt cycle", but where is the rest of the world? US companies are servicing and supplying the entire world and looking at, say, US consumer confidence, does not necessarily predict how much steel US Steel will export or how much tobacco MO will export.
Could that be the reason for the disconnect?
Posted by: grodge | Apr 26, 2007 9:54:26 AM
grodge, China is the largest steel producer by a VERY large margin, and China is now a net steel exporter.
As for the ciclicality around the world, the debt/housing bubble was far from being a US-centric event, it happened in a lot of other places as well: Australia, France, Spain, Holland, UK, Ireland, etc. It didn't get the "stated income" features of the US bubble but close.
These debt/Housing bubbles are all nearing their end at the same time.
On the positive side, much of the developing world is indeed developing. But economies are mostly local, the effects of the debt/housing bubbles are much stronger than the effects of the developing world.
For instance, new debt in the US comes to about 20-25% of the GDP, if new debt slows the GDP will implode (also, there's around 5-20% of the GDP which is mostly made up, but that's another story).
Posted by: Incognitus | Apr 26, 2007 10:00:46 AM
nogo-
enough with the illegal aliens...how many times are YOU going to use that.
THis is the same person who puts words in other peoples mouths and then derides them for any position other than his own.
Are you now blaming illegal aliens for everything?? You are for the BLS #'s
BTW if we did'nt have all these "illegal aliens" I hope you and you're precious attitude enjoy paying over $7 for a head of lettuce.
GIVE IT A REST....
Blowhard
Ciao
MS
Posted by: michael schumacher | Apr 26, 2007 10:20:43 AM
Incog,
Thanks. So the conclusion is....
The DJIA and SPX are bubbly?
By "the GDP will implode", what do you mean... revised from 3% to 2.5%? If that happens, won't interest rates just decline to spur more borrowing, spending and investment?
If inflation is as big an issue as implied by the data, then why is there so much excess liquidity flooding the stock market? Is it all the fed?
I've got cash sitting and I've been waiting for this implosion for a year that doesn't seem to be happening.
Just trying to understand.
Posted by: grodge | Apr 26, 2007 10:26:50 AM
The bubble this time is different, it's not that the SPX, DJIA, RUT, DAX, etc are bubbly. It's that the bubble is inflating the fundamentals themselves.
Stocks are very expensive only because they trade at a rather normal multiple of overvalued fundamentals.
By "the GDP would implode" I can't advance any numbers. The situation right now, debtwise, is worse that before the Great Depression in 1929. But I guess anything that happens will be less drastic.
Posted by: Incognitus | Apr 26, 2007 10:30:56 AM
Point of clarification for Grodge and Incognitus:
Earnings for S&P 500 companies reporting are up 6.2% year-over-year, not 6.2% over estimates.
http://tinyurl.com/yvpgmd
This is not an issue of under-predicted earnings beating the estimates, but an actual 6.2% gain in earnings since the 2006 quarter.
Let me save you bears your time:
"6.2% growth? Full of accounting lies!"
"ONLY 6.2%?"
"6.2%, but if you back out all the profitable companies, it's -25.5% for earnings."
"6.2% earnings growth is a sure sign of a recession, since last year earnings were up more than 10% over 2005."
"How many gold coins can I buy with 6.2% growth?"
Flying saucer alert!
Posted by: Nova Law | Apr 26, 2007 10:31:40 AM
