The February jobs report comes out tomorrow, and hopes are high that the numbers will finally show that the U.S. economy is truly and firmly in recovery mode. As Ben Bernanke said recently in his semiannual testimony on Capitol Hill, the recovery that seems to be occurring in virtually every economic and financial metric except job growth cannot truly be regarded as a recovery until businesses start hiring again and the 13.9 million unemployed people in the U.S. are able to find work. (Keep in mind that the 13.9 million number does not include the 8.4 million people that are ‘underemployed’ or the 2.8 million people that are ‘marginally attached’ to the workforce). After the lackluster jobs report in January, which reported that the U.S. added only 36,000 jobs during the month, that chasm between the recovery that businesses, economists, and the media are reporting and what is actually taking place on ‘main street’ only widened. And while everyone remains hopeful that tomorrow’s numbers will finally start to narrow that gap and ADP’s numbers yesterday strengthen the bullish view, I don’t think anyone has any visibility into what the report will show. Unfortunately, LinkUp’s February jobs report seems to be equally as opaque.
LinkUp reported today that new job listings on company websites fell by 3% in February, while total job listings on company websites rose by 1%. Equally as confusing, a majority of states saw a decline in new job listings (27 to 22), but a majority of states also saw an increase in total job listings (32 to 15).
In terms of jobs by category, LinkUp’s February data showed the exact same numbers. New jobs by category fell 3% while total jobs by category rose 1%. Of the 32 job categories that we track, 18 showed an increase in new job listings and 23 showed an increase in total job listings.
LinkUp is the fastest growing job search engine on the web and indexes only jobs that are found on company websites. The search engine does not list any jobs from other job boards, nor does it allow anyone to post jobs directly to the site. Rather, LinkUp currently lists about 700,000 job openings indexed and updated daily from approximately 20,000 company websites throughout the U.S. As a result of this completely unique approach to job listings online, LinkUp’s search engine does not include any old jobs, scam jobs, phishing jobs, or duplicate listings. But while we may have the best data set of anyone in the industry, making sense of ours or anyone’s numbers remains a herculean task, to say the least.
The government numbers are notoriously unreliable and are constantly being revised. ADP can only accurately count their payroll numbers for their clients in the private sector, and Inuit only examines hiring for small business. The conference board counts classifieds from the daily newspapers (as small as that number may be these days) and online, pay-to-post job boards such as Monster and Careerbuilder, as well as aggregators such as Indeed and Simplyhired. Those numbers from online job sites, however, include old jobs, duplicate listings, job scams, fake listings, phishing posts, and other fraudulent and garbage listings, so their numbers are horrendously flawed. The Monster index itself isn’t any better for those same reasons, and the same goes for Indeed’s monthly numbers.
Even our own LinkUp numbers present challenges. We are constantly adding new companies to the index, which makes it difficult to compare monthly numbers beyond the previous month. We do normalize the company list when comparing February to January, for example, but that matched pair of months includes far more companies than, say, the September to October comparison. So while the chart below may provide a glimmer of hope that total job listings from company sites have increased dramatically between September of 2010 and February of 2011, the gains are largely due to the growth in the number of companies included in our search engine. But even as we’ve added thousands of new companies to our search engine since September, new job listings have dropped by 16%. This fact, perhaps more than any other, stands as the most conclusive data point we have confirming the anemic pace of job growth over the past few months.
Another fairly reliable data we can provide is the percentage increase or decrease from month to month in the number of new and total job listings on the 20,000 corporate websites indexed by LinkUp. But even with that data, 4 of the 5 months since October have shown a percentage decrease in new job listings, while all 5 have shown a percentage increase in total job listings.
The other complicating factor for us is trying to ascertain the extent to which the job listings on LinkUp, indexed from company websites, are a leading indicator of future hiring activity. If T-Mobile or Dell adds jobs to their company website, it’s a clear sign that they are hiring. But how quickly those listings translate into new hires cannot be known for certain. My best guess is that there is at least a 30 day lag between job listing activity on LinkUp and the jobs numbers reported by the Department of Labor (DoL). If that is truly the case (and that is a big if), given the big uptick we reported a month ago for January’s numbers, then tomorrow’s jobs numbers should be very strong. Unfortunately, our numbers for February would then indicate that we’re in for a weak jobs report for March.
So is job growth truly recovering? It’s as clear as mud.