Regcast training : Hyper-V 3.0, VM high availability and disaster recovery

They say you can't manage what you don't measure. But what do you do when you don't trust what you are measuring or when you are measuring the wrong thing?

This morning, the US Labor Department's Bureau of Labor Statistics announced its jobs report, as it does early every month to provide data on the preceding month. Like most people here in the States, I never paid much attention to the jobs report beyond reading about the unemployment rate and processing the number of jobs lost with a flavour for industry breakdown. But since I came aboard El Reg last September, I have been drilling down into the report, trying to assess the lay offs among the IT vendor community.

According to the bureau's sampling and calculations done on that sampled data, America shed 598,000 non-farm jobs in January, worse than expected and boosting the unemployment rate to 7.6 per cent. Much has been made in the press about this being the highest number of jobs lost in 34 years. But no one in the press points out that to make such a comparison fairly you need to adjust the number of jobs lost against the size of the labour pool, which is much larger in the States than it was 34 years ago.

Similarly, business news organizations are always talking about how much the Dow Jones Industrial Average or the Financial Times Stock Exchange index drops by X or Y points, the most since Z, and they do not adjust that drop against the size of the index at the time, I guess to make a 300-point drop sound more dramatic than it really is. As an index or a labour pool grows, you would expect swings to be bigger in number. The point is this: It isn't as bad as it sounds, even though it is bad.

Since the recession began in December 2007 - and isn't it funny that America only officially admitted that it went into recession a few months ago - the US has lost 3.6m jobs, and about half of them have gone in the past three months. The bureau revised its data for job losses in November and December 2008, and now says 597,000 jobs were cut in November (up from 584,000) and 577,000 jobs were lost in December (up from 524,000). It is reasonable to expect that January's data will be revised too, probably not in a direction that makes people feel better.

According to today's report, manufacturers shed 207,000 jobs in January, which the bureau said was the biggest one-month decline since October 1982, during the Reagan administration's recession. The construction industry lost 111,000 jobs in January, which is no surprise given the collapse in the housing market and funding for commercial real estate projects. Since January 2007, more than 1m construction jobs have evaporated.

The financial services industry lost 42,000 jobs in January and has shed 388,000 jobs since peaking in December 2006. The health care industry added 19,000 jobs in January, which ironically might show just how sick the US healthcare system is. But let's not get into that right now.

The bureau does not track the IT industry specifically in its monthly jobs report and you have to tease the effect job cuts have out of the data that the government supplies. Uncle Sam does not classify workers by job and then count them as they are hired and fired, but rather counts workers and ex-workers by the industry sector in which they work (and with very coarse industries) or by other demographics, such as sex and age.

The Labor Department does track manufacturers by the stuff they make, including those that make IT gear, as well as employers who participate in what it calls the "information industry", which includes publishing, moving, broadcasting, telecom, data processing, hosting and other information services. The bureau also counts jobs in a category called computer systems design and related services, a subset of the professional services category. You can look at the raw data for January right here (PDF).

In the January jobs report, computer and peripheral equipment makers in the States added a net 700 jobs (this is seasonally adjusted data and heaven only knows what witchcraft goes into that number). Communications equipment makers lost 500 jobs, and semiconductor and electronic component makers shed 5,900 jobs. All told, these three sub-sectors account for 722,300 jobs, according to the bureau, with more than half of them in the semi and component manufacturing biz.

In the information industry classification's data processing and hosting sub-sector, 200 jobs were added, boosting employment to 256,700 jobs. In the computer system design and related services area, 3,500 jobs were cut in January, leaving a pool of 1.46m workers in this field.

What would be more enlightening, particularly should you be trying to plan or maintain a career, would be a fine-grained count of jobs by title, regardless of industry. Here's why. Employees who work as programmers who happen to get their paychecks from a manufacturer are counted as manufacturing employees when they get laid off. But they are really IT employees, not workers on some assembly line.

Ditto for IT project managers, system administrators, chief information offers, database designers and so on. It is hard to be certain of the magnitude, but clearly there are more IT employees working in companies outside of the IT vendor community than inside. At least, that is, until we all move to cloud infrastructure running in the Antarctic supported by crack IT employees working from deep within the Amazon jungle. ®

Requirements Checklist for Choosing a Cloud Backup and Recovery Service Provider