Analysis: $400m cost for big government

The financial disaster called the Office of Shared Services explodes the myth that bigger is better – a lesson for everyone, including health officials, who thinks centralising bureaucracy is anything but about consolidating power.

According to an Economic Regulation Authority report which recommends decommissioning the OSS, the net cost of the service was more than $400 million to date is not only five times more expensive than originally estimated in the first business case put up in 2003 but it has taken longer to deliver than expected and failed to deliver the efficiency that was promised.

Originally, it was meant to save $56 million per year from a $315 million corporate services budget across the WA government. It is delivering nothing like that.

And having cost $440 million, less about $40 million of actual positive gains, it is still estimated to cost hundreds of millions more to make it work. That is why decommissioning, thought to also cost hundreds of millions, is the cheapest option to government after years of trying to make this work.

The costs, of course, don’t include the amount of time consumed by the state’s leadership which could have been put to more meaningful tasks.

While no-one ought to knock the well-intentioned target of reducing the cost of corporate services such as payroll to government, those who championed this idea in particular ought to have done their research better before embarking on such a costly exercise.

State opposition leader Eric Ripper, as the WA Treasurer at the time, will have to cop his fair share of the blame in this – which is perhaps why Premier Colin Barnett has been happy to let the OSS run along for a few years and then dribble out bad news before killing it off.

In my view, this project was all about consolidating power back to state Treasury and, therefore, the key government leadership. If you think about it, this turned a myriad of minor deals at department level into a handful of major contracts where the bureaucracy had the theoretical opportunity to squeeze savings by teasing suppliers with the chance to cut out the competition.

Perhaps the vision obscured the reality of getting there.

Even back in 2003 there was plenty of history on the difficulties of shared services in state governments, not to mention plenty of well-publicised case studies on massive IT projects that had gone wrong.

Well before WA pressed the button on its model, NSW was already bogged down dealing with its version of this concept.

In the private sector, according to the ERA report, those looking into the concept found shared services had a generally positive track record, showing efficiency gains of 5-15 per cent. I dare say that the private sector has less reason to report on its failures in this respect, so it’s possible that figure was skewed to the positive in the first place.

Culturally, big private sector businesses tend have a lot more opportunity to gain from streamlining their back office than governments which operate across a vast array of fields.

So you’d have to think that the then-Labor government’s expectation of a 13.5 per cent gain was wildly optimistic.

It should also be noted that, according to the ERA report, the original forecast of $56 million per year in savings was whittled back due to changes to $49 million. Despite that, the government of the day determined the saving should be $55 million per year.

So the 13.5 per cent wasn’t just optimistic projection, it was a target.

There is some irony that much of the shared services failure was in the effort to consolidate payroll. The IT system which attempted to deal with this could not cope with the complex award system that decades’ of union activism had created.

The Labor government’s honourable effort to reduce waste in the public sector was undone by the same inefficient system of determining a worker’s pay as it’s federal counterpart later foisted back on the private sector in the form of modern awards as they killed off individual contracts.

The OSS experiment shows that centralising bureaucracy is rarely an exercise in efficiency.

I am a big believer in our federation as a system of government, where we can learn from the experiences of other states. There are things, such as defence, where national interest offers a more efficient structure than each state attempting to manage this.

But the federal government’s efforts last year to consolidate control of health into the hands of the Commonwealth is an example of the opposite.

It is hard to imagine what benefits there were in putting Canberra in charge of this area? Apart from money (of which even federal governments don’t have an endless supply), I don’t believe the boffins of the ACT could run a health system any better than the ones we have here. In many ways, I believe we’d be better off with more regionalisation than we have today.

The beauty of a federal system is the jurisdictions experiment with their own ideas, adapting them to their own unique circumstances. Hopefully, where services are comparable, inefficiency becomes obvious – at first to bureaucrats and ultimately to voters.

Centralised systems have their advantages but the one-size-fits-all approach does not work in every circumstance.

See Mark Beyer's news article: http://www.wabusinessnews.com.au/en-story.php?/1/90124/Govt-writes-off-440m-IT-project