IT is a broad field, and it is hard to generalize about it.

Let's take what you did for a start. I believe you were a system admin, administering a network with pretty much the standard Microsoft desktop stack. (Windows with Office and Outlook running against Exchange with some appropriate networking, firewalls, and anti-virus in the mix.)

No matter how important this may be to the company, it is a potential target for outsourcing. Why? Because you're administering the same bog-standard stack of software that everyone else is. An outsourced provider will add an extra layer of management, but they also get to add the fact that they do the same task at a lot of places. So they have thought up tricks, have efficient backup and installation scripts, etc. This makes them potentially more efficient at doing your job.

In short, standardization leads to commoditization. In IT as well as anything else.

Now I'm also in IT, but my job is not particularly amenable to outsourcing. My job description says "reporting", which is pretty vague. But in reality it means that I need to understand our business processes, our application, where data is captured, how it is aggregated, what custom reports have been created, etc. Given that we have an unusual business model (how many companies give away money for a profit?), this expertise is very company specific. An outsourced provider would not be able to leverage their experience with other clients in any significant way to do my job better than I do it.

In short, attempted commodization is a bad idea where standardization has not lead. Think about this before you outsource.

Given that Skip works with AS 400s, and AS 400s are not a particularly commonly chosen piece of infrastructure, I'd guess that Imric's job would be a bad one to outsource. (Unless you were planning to also kill the AS 400.)

There is always a "But..." though. And the big one here is that while it would be a bad idea to outsource me, big decisions like "outsource IT" tend to come down from people far up in the food chain. When the CEO/CIO makes a decision like that, they are not going to dive down to my level in the org chart to find out that it is actually a bad idea to outsource me or Skip. (And everyone who is losing their jobs will they that they shouldn't...) So the decision will come down and be enforced both where it makes sense and where it doesn't. (This kind of management insensitivity is, ironically, a symptom of the internal inefficiency that can make outsourcing a good idea.)

Which means that Skip gets a front row seat for the demonstration of how bad an idea it can be at the same time that Peter gets a demonstration of how replaceable he was. And someone like Bill gets a more balanced view. And in that more balanced view, it could go either way.

Incidentally I'm actually reasonably safe from outsourcing for two reasons. The first is that I work for a small suborganization that is managed separately, so an outsourcing decision is likely to be made more carefully than in a large one. And secondly I'm on the org chart in software development, which for us is a core competency. Therefore we'd be unwise to outsource it.

(Of course there is no real protection from idiotic management...)

Cheers,
Ben