The global IT industry is so abundantly served by information sources and opinion givers that it is sometimes difficult for the conscientious editor to know who or what to give credence to. So when the pundit Robert X Cringely wrote in his blog (I, Cringely) that IBM is preparing to lay off 150,000 people – yes, 150,000 – most of us in the professional press didn’t report the speculation.

That was not because Cringely lacks credibility. The problem was that the story didn’t make much sense: Cringely, relying on multiple but unnamed IBM sources, said most of the layoffs would come in the US, and in particular within the IBM Global Services (IGS) division. But IBM only employs 130,000 people in the US; and its profitable $48 billion IGS business only employs 190,000 worldwide.

Even so, the story had some angles that gave it credence. The gist is that IGS, by far the world’s biggest IT services company, has disappointed IBM financially, and that it needs to become much more profitable. The way to do this is by adopting what it calls a “lean” methodology, reducing staff and replacing highly paid and highly pensioned (mostly US) workers with much cheaper employees in places like India and Brazil. IGS executives were on a 10-city planning tour as Cringely wrote.

The story is otherwise short on facts, but the forecast of huge job losses was enough to promote a lot of responses in the US, not least from IBM, which said the story was creating “needless activity” within the company. It issued a garbled statement that fell far short of a full denial, saying that while it had “recently implemented a focused resource reduction in the US…. these actions are well within the scope of our ongoing workforce rebalancing efforts.”

Given that, Cringely may yet be proven to be at least partially right. IBM’s IGS division has made small scale layoffs, the unit is certainly growing too slowly for investors’ liking (around 2% in 2006), and its margins are lower than IBM as a whole. Even so, realistically, nothing dramatic is really likely.

But there is a big story unfolding in the IT services industry – one that is much less widely reported because it is happening more slowly and more widely: a price war is underway, or looming, in services, and it is not cyclical, but structural and long-term.

Why? Over 40 years, the cost of IT has always consisted of three parts: hardware, software (even if delivered as a service); and services – hiring the humans to make it all work.

Thanks to global expansion, Moore’s law and global high-tech manufacturing, hardware prices have fallen year by year. Software remains expensive, because it embeds services, but a combination of packaging, automation, open source and high volume sales in programming are bringing that cost down.

Services, however, have continued to be extremely expensive. As every CIO knows, services eat up the bulk of the IT budget, whether it is for maintenance or new development or operations management.

At the lower tiers, the surge in offshore has helped to curtail the cost of these services, as has the large number of suppliers that now compete for business. But the demand is huge, so the prices have largely held up. Sun’s chairman Scott McNealy was half right when he once said that services is the place where hardware companies go to die; what has really happened is that all IT suppliers go there when their revenues and margins elsewhere die.

But what if services margins start to stumble? In order to reduce the cost of services, both organisations and suppliers have three tactics in their armory. First, they can go offshore, where labour is cheap; second, they can attempt to tackle process inefficiency, especially by productising and formalising services. And third, they can automate services altogether, eliminating human intervention wherever possible.

Most IT users and services companies are really only doing the first of these – and they are treading cautiously. In terms of process efficiency and automation, it has scarcely begun.

This is where Cringely has got it partly right about IBM Global Services. It is moving as many projects to low-cost economies as it can – as are most services companies. But it plans to go further: it is also introducing the principles of lean manufacturing, pioneered by Toyota in Japan, into services. That means making everything not only repeatable and formulaic, but also highly resource-efficient. These repeatable services make it easier to export work to developing countries, and easier, ultimately, to automate altogether. If it gets it right, IBM’s margins will go up, and if rivals do the same or better, its prices may even come down.