• RSS Feed

The LinkUp Blog The Industry's Best-Kept Secret

October 4, 2011 / Toby Dayton

Despite Mild Gains on LinkUp In September, Jobs Picture Remains Bleak

For the 2nd consecutive month, LinkUp is reporting that both new and total job listings indexed from company websites rose from the prior month. In September, new job listings increased from August by 19,095 (5%), while total job listings rose by 36,388 (4%).

LinkUp is the largest, fastest growing job search engine that only indexes jobs found on corporate websites throughout the U.S. Updated daily, LinkUp’s job search engine includes approximately 850,000 job openings indexed from over 22,000 company websites. Because LinkUp does not allow anyone to post jobs directly to the site, the search engine does not include any garbage listings such as job scams, phishing posts, work-at-home-scams, or old listings. And because LinkUp only indexes jobs from a single source (the employer’s corporate website itself), there are no duplicate listings that pollute aggregator sites such as Indeed and Simplyhired.

In September, 33 states showed an increase in new job listings, while 44 states showed an increase in total job listings.

New and total jobs by category also rose in September, by 6% and 4% respectively. 20 of 31 job categories showed increases in new job listings, while 25 of 31 categories showed an increase in total job openings. Technology, healthcare, and retail categories reported the strongest gains during the month.

While the increases in new and total job listings are smaller than the gains reported in August, the rise in new and total job listings indicates that private sector demand for labor appears to be continuing its expansion. Unfortunately, forecasting the extent to which the increases in new and total job openings translate into actual job growth is far murkier and involves a series of assumptions that are becoming harder and harder these days to make with any degree of certainty.

First, we have added back into our calculations the 46,000 Verizon workers that were on strike last month and were subtracted from the BLS’ Establishment Survey data for August. Second, we believe that there is a good chance that the August numbers will be revised when the BLS report is issued on Friday, but we have not factored that into our forecast. And finally, when we average our paired sets of job data for August, the September numbers actually show a decrease in both new and total job listings. This final point is worth a more detailed explanation.

Because LinkUp is constantly adding new companies to our index, we have to ‘normalize’ the data in order to accurately compare one month to the next. To do this, we take the set of companies indexed by LinkUp on the first day of a given month, and compare the new and total job listings from those companies in that month and the next month. The following month, we do the same thing with a new set of companies that includes the employers that have subsequently been added to the index. As a result of this methodology, we get two sets of numbers for each month – the first when we compare a month (t) to the prior month (t-1), and the second when we compare the month (t) to the following month (t+1). In making our jobs forecast each month, we use the average of the numbers for the prior month, but are limited to only one data point in the current month. (With September, for example, we have to rely on one set of data because we have not yet compared September to October).

Taking into account all of these factors, we arrive at the following data table that includes our forecast for both September’s jobs report that will be released on October 7th and October’s jobs report that will be released on November 4th.

Based on the gains we saw in August, when new job listings rose by 6.9% and total job listings rose by 1.2%, we expect that the jobs report on Friday will be slightly better than August’s report (which, again, we have revised up to a positive 46,000 to account for the Verizon strike). We typically see a 30-60 day lag between our data and the BLS numbers which makes intuitive sense given that a job listings on a company website takes time to translate into an actual hire. Unfortunately, the decrease between the average of our two August data points and September’s LinkUp jobs numbers leads us to believe that only 15,000 jobs will be created in October.

Regardless of how close our forecast is to the actual BLS numbers, there is absolutely no doubt that job growth continues to be stuck in neutral, at best, and might actually be heading back into reverse. When looking at the raw data of the number of job openings on LinkUp in 2011, it becomes obvious that demand for labor is slowing down. For the first time ever, the number of job openings indexed from company websites has dropped below the 90-day rolling average.



If the charts above provide any indication of what’s in store for the next few months or quarters, not to mention the thousands of other gloomy forecasts, the mild anxiety in Washington should be replaced with absolute panic. The ‘2nd Great Contraction’ is having a horrendous impact on the economy, 25 million Americans remain out of work or are underemployed, and actual unemployment is north of 16% and probably headed higher. The President’s jobs bill, while better than nothing, is about half of what is required to foster meaningful job growth. Based on extensive analysis from Bridgewater, history has shown that unemployment cannot decline until GDP growth reaches 5%, and in a $15 trillion economy, we need about $750 billion in new spending to get there. The American Jobs Act, with its $447 billion in spending, gets us part of the way there, but it’s not enough. It’s simply not enough. And if Washington doesn’t act soon, and given the abundance of spinelessness and wildly flawed priorities there is no reason to believe it will, my guess is that 2012 is going to be another grim year for the job market.