Lies, Damned Lies, and Statistics

I begin to understand why social scientists working with govspeak get paid.

My trek into public documentation began with Michael Steele's remarkable assertion that working for the government isn't a job just because it's essentially temporary, while jobs in the private sector are by definition permanent. Even Andrew Sullivan was aghast at this statement.

I've been a contract labourer for much of my working career in one form or another. I've worked for contract staffing agencies, and I've worked on extended assignment from consulting firms to their clients. I can definitely say that a temporary job is still a job, and that to date being so employed has helped my career more than hurt it. Of course, I work in information technology, so the stigma associated with non-permanent employment is negated by the field in which I operate.

Assessing my experience as a facet of the entirety of the employment experience in the US, however, is hardly an easy proposition, and public resources are not exactly helpful.

The American Staffing Agency, for example, reported in 2006 that "staffing employees... represent about 2% of the actual work force." For perspective, the actual work force, according to the Department of Labor, was 71 million in 2007, which would make the ASA's number for staffing employees somewhere around 1.5 million. The same ASA survey showed that about 80% of these employees were considered full-time. The Bureal of Labor Statistics, as of about 1999, cited a slightly larger percentage and reported 3.4 million temporary workers in 2004, of whom it said "about half would prefer to have a permanent job."

Contract labor, contrary to Mr. Steele's assumptions, is very attractive to the private sector. Why? Quite simply, business perceives the temporary worker as an economy over the Full Time Employee, or FTE. Supply Chain Management Review states that using "contract labor substitution" - hiring temporary workers to supplement or replace FTEs - can "squeeze 10 to 50 percent in cost savings." With that kind of guidance it's hardly surprising that business would hire on contract rather than permanently employ any given professional: this makes Steele's assertion that private sector employment is permanent and stable by nature demonstrably false.

Beyond that, any search I tried in the Department of Labor, BLS, US Census Bureau or other governmental online database turned up anything EXCEPT recent statistics on the percentage of employees in the US working under contract rather than hired. I could find breakdowns of employment by industry by region, by race, by gender, by age group - all the usual markers for determining non-descriminitation in the workplace, by the way - by employment sector, by public v, private employment, and by occupation. Searches including the phrase "contract labor," in contrast, brought up such useful items as generic contract language for government grants or for individual employment; any statistics I found seemed to come from Germany. Searches including "temprorary worker" produced papers on the now-moribund "guest worker" program touted these past years as a partial solution to immigration or articles on how the slump is hurting that market - in Japan. One search (I forget which) couldn't give me information more recent than 1994 and tried to pass off statistics from 1987 as relevant - even though I had specifically included "2008" as one of my search criteria.

Google and the other search engines have been no more productive on this point. I presume this is largely due to the same issues I had with my own searches. If any of my readers have any tips for finding better statistics please leave them in comments.

As for Mr. Steele's commentary, following his logic, the unemployment rate (since temporary labor is obviously not "employment" in his dictionary) would need to be adjusted upward 3-5% at a minimum through factoring in all those temporary workers who he insists don't have a "job." We'd also have to filter out the employees of Blackwater, Halliburton and the other major players working government contracts that outsourced the kind of work the US Army and the Corps of Engineers used to do, which will likely add another 2-3% to the tally. Most of those, according to nearly every document I could find, are already in the private sector (while I can find no hard statistics, the tone of the articles I did find indicate that the US government has essentially eliminated contract staffing over the last decade for its individual departments: I can only assume from the contractor business that some entire departments have been outsourced instead). And since elected office is essentially temporary employment (renewed every 2-6 years depending on the office), we'd have to discount the President, Congress, all 50 governors and everyone in the legislatures of all 50 states, every county and city and various oversight boards. By my math that puts "unemployment" somewhere around 30%.

I will freely admit that I'm not an economist by profession. And after the experience of researching the various statistics for this post I'm not altogether displeased with that. But If I can take a couple hours and find this much to refute Steele's comments, I expect that many more qualified minds could come up with something much more concrete on the subject.

In the meantime, though, this is the second time I've been faced with the opacity of the US government. First there was the CPSC; now I've seen just how far from transparent the DoL and its various arms actually are. Small wonder that determining how bad the slump is has proven so difficult for Washington when even their own agencies tasked with oversight and recording can't readily produce useful information.