Scene: Somewhere in the Riverlands, south of The Twins.
Tywin Lannister: “When the battle commences, you and your wildlings will be in the vanguard.”
Tyrion Lannister: “The vanguard? Me and the tribesman, on the front lines?”
After a brief recounting of the savagery of the Stone Crows and the Moon Brothers the night prior, Tyrion exclaims, “Surely there are ways to have me killed that are less detrimental to the war effort.”
Tywin: “There will be no more discussion on the matter.”
Tyrion, leaving the table: “It appears I’m not hungry after all. Excuse me my Lords.”
For certain, the vanguard is no place for the weak or the cowardly given that it’s where the most intense action takes place. In today’s global capital markets, the vanguard is arguably the U.S. unemployment rate. While the nation’s labor market has served as a critical marker for the country’s anemic recovery from the Great Recession, Bernanke’s decision to peg the duration of QE III to the unemployment rate adds even more weight to the figure. Leaving aside for a minute the soundness of the Fed Chairman’s decision (it’s entirely appropriate and congress should be adding stimulus), an already critical data point has become a colossus. The largest economy in the world ($15.7 trillion v. China’s $8.3 trillion) is receiving an IV drip of pure adrenaline until such time as the patient is sufficiently healthy to rise from his deathbed. And the only reading on the EKG is whether or not U.S. unemployment has dropped below 6.5%. Needless to say, the Bureau of Labor Statistics has never had such a massive global audience.
And what of Bernanke’s decision to peg QE III to the unemployment rate? The answer has 3 parts that can be summarized as follows:
1) It is, absolutely without question, the right prescription for the deathly ill U.S. economy (see Krugman)
2) The only drawback is that unemployment is notoriously difficult to measure and BLS data is not completely reliable.
3) QE III is working (despite what Jim Grant might argue to the contrary)
Krugman’s columns and his outstanding book (End This Depression Now) are more than ample evidence to support Bernanke’s case on the need for stimulus. In fact, the doctor should be administering even more aggressive stimulus, and were it not for the Kevorkian Congress, we’d be in even better shape today.
In regard to point 2, measuring true unemployment remains as elusive as finding grumkins and snarks. The unemployment rate released monthly by the Bureau of Labor Statistics is a horribly bastardized figure (see blog post here), and the job growth figures are equally as unreliable. And while I completely disagree with Jim Grant’s abhorrence of Bernanke’s easing regimen, I do agree with his skepticism in pegging central bank programs to suspect economic figures.
In the February 8th Grant’s Interest Rate Observer, Grant writes, in criticizing the decision of the Bank of England’s Mark Carney to shift his focus to GDP instead of inflation,
“If Britain’s macroeconomic data are inferior even to America’s, you wonder how Carney can make good on the policy change he is supposed to have under his hat. The move at which he’s hinted is a shift to targeting nominal GDP instead of the rate of inflation, at which the bank has been shooting since 1992. But if GDP is a lagging indicator, and if it is hard to measure, and if those measurements are prone to revision – certainly, all this is true in the United States – how on Earth could a GDP-targeting central bank expect to hit what it is aiming at?”
Indeed, what is true of Carney’s challenge in targeting Britain’s GDP is true of Bernanke’s challenge in targeting U.S. unemployment and U.S. job growth.
As evidence of Bernanke’s dilemma, and because our current LinkUp 30-day forecast includes not only estimates for job growth in February and March, but also revisions to job growth figures for December and January, I’ll highlight the frequency and magnitude of the revisions that the BLS makes to its own data. The 2 charts below highlight in both gross numbers and percentages the 30-day and 60-day revisions made by the Department of Labor since January of 2011.
The average 30-day revision is 13%, with 13 upward revisions and 8 downward. The average 60-day revision, in relation to the initial data release, is 25%, with 18 upward revisions and 5 downward. The average change for the 60-day revision, in relation to the 30-day data point, is 17%, with 15 upward revisions and 7 downward. And lastly, there were 9 instances in which both the 30-day and the 60-day figures were revised upwards, 4 went up/down, 3 went down/down, and 4 went down/up. Needless to say, the BLS job growth numbers are quite the moving target. No doubt Bernanke’s a good shot, but the BLS data would test even the finest marksman.
So in any event, is the stimulus working? It is impossible to prove where we’d be without it (and I shudder to even think about it), so all we can do is look at where we are with it – and that’s not a horrible place. To be sure, the recovery has been miserably slow and incredibly painful, and things could and should be a whole lot better. I also have no doubt that they would be without the criminal dysfunction in Washington, particularly on the part of the House Republicans who would euthanize the economy rather than see any improvements and related credit accrue to Obama. But job growth is slowly creeping up, and the February numbers from LinkUp provide some optimism that job growth will accelerate in the months ahead. Which brings me back to the Vanguard.
Rather than looking backwards at historical data, particularly when it’s flawed or unreliable, I would suggest that the vanguard in assessing the strength of future job growth in the U.S. lies in obtaining insights from the job listings that employers use to fill job openings. The best available indicator of a future hire is a job listing that a company posts to attract job applicants to fill that opening. And the best source of those job listings is the company’s own corporate website itself because it’s free for the company to post jobs on its own site, there are no old, outdated, or already-filled listings, there are no duplicate job listings, and there is no garbage such as fraudulent listings from lead-gen marketing companies, work-at-home scams, or identity theft posts. Nearly as important, there are no listings from 3rd party intermediaries such as recruiters, staffing companies, temp firms, headhunters, or search firms.
The challenge in all this is that there are tens of thousands of corporate websites and indexing all of the job openings on all of them every day is unimaginably difficult. Fortunately, LinkUp does exactly that, and the 1.4 million jobs that we index from 25,000 corporate websites throughout the U.S. every day give us unrivaled ability to predict future job growth in the country. And what we see ahead of us looks quite good.
In February, new job listings in LinkUp’s job search engine rose 14% to 566,386, and total job listings rose 9% to 1,297,959. 46 states showed gains in new job listings while 48 states showed gains in total job openings.
In terms of job openings by category, new and total job listings both rose by 5%, with 17 of 31 categories showing gains in new job openings and 21 of 31 categories reporting increases in total job listings.
Based on LinkUp’s job listings data from November and December, we are forecasting that the BLS will, in Friday’s Employment Situation Report, revise December’s jobs numbers up to a positive 260,000 and January’s numbers up to a positive 275,000. January’s slight downtick in new and total job listings on LinkUp leads to a forecast of only 260,000 jobs being added to the U.S. economy in February.
But most encouragingly, strong data from LinkUp’s job search engine in February points to a net gain of 410,000 jobs in March.
It’s an ambitious forecast, to be sure, but then again, the vanguard is no place for the faint of heart.
Oh, and by the way……Winter is coming.