The argument over Coalition costings has soured for Labor.

Prime Minister Kevin Rudd claimed yesterday there was a $10 billion hole in the Coalition’s savings estimates, basing this on confidential briefing papers from Treasury, Finance and the Parliamentary Budget Office. Later in the day the heads of the Treasury and Finance departments put out a media release stating they had not costed any Coalition policies. They pointed out that costing depends entirely on what you choose as the starting point: “Different costing assumptions, such as the start date of a policy, take up assumptions, indexation and the coverage that applies, will inevitably generate different financial outcomes.”

Shadow treasurer Joe Hockey pounced, saying it proved neither the Treasurer nor Prime Minister could be trusted. The tit-for-tat went on.

Disputes over policy costs are as old as politics. Candidates for the Roman Senate in 200BC were probably heckled with “where’s the money coming from?” Waffly political answers were probably as alienating for voters back then. So for some clarification …

What’s been done to improve policy costings?

The Charter of Budget Honesty Act passed in 1998 was meant to improve the situation. It allowed parties to submit policies to Treasury and Finance for costing during the election period. The act provided for release of a pre-election budget update, intended to end the practice of a newly elected government discovering a “black hole” in the budget that would allow it to abandon its promises. This second objective was achieved, and the Pre-Election Fiscal and Economic Outlook is now an important starting point for all parties in elections.

But the costings arrangements quickly broke down. The government realised it could have all of its policies costed confidentially before the election, enabling it to refine them and have no unpleasant surprises when they were released. Oppositions had no such luxury and faced an enormous political risk if they submitted policies: it might turn out they were unaffordable. As a result they took to either not submitting policies for costing or doing so at the last minute. The system was skewed in favour of government.

What’s the Parliamentary Budget Office?

To fix this problem, a Parliamentary Budget Office was created in 2012. It was meant to allow non-government parties to have their policies costed outside the hurly-burly of the campaign, so well-considered policies could be developed.

The Coalition claims all its policies have been costed in this way, but so far have refused to release the official documents. So they are in part at fault in last night’s brouhaha. There is merit in the argument that debate would be better informed if the PBO costings were in the public domain. But by the same token none of the Labor policies have been officially costed under either the Charter of Budget Honesty or PBO rules: Labor had them all done beforehand.

In fact, the only party that so far has used the PBO in the way it was intended has been the Greens, which has had 53 policy costings posted on the PBO website. If the Greens continue to have the balance of power, the PBO will have an assured future; but if not, the usefulness of it surely has to be questioned. The PBO has not prevented unedifying disputes over costings which in the end only come down to different assumptions.

How can parties game the rules?

There are all sorts of tricks used in presenting numbers during campaigns. If, as Labor did yesterday, you want to make an alleged black hole appear bigger, you add up the figures for the three forward estimates years as well the budget year. So a gap of $2.5 billion a year becomes a much more scary $10 billion. If you want to make a big promise but not affect the budget, you announce funding starting in year four or five. If you want to make the cost of a new policy seem small, you announce only the first year.

Where to from here?

What we have today is better than the situation before the Charter of Budget Honesty, when parties tended to make up numbers with very little science or methodology and there was no definitive way of testing them. The policies concerned were often abandoned, and voters’ expectations were dashed.

However, the current arrangement is only a little better. It would be possible to go further with a genuinely independent Budget Integrity Office that could cost policy announcements and enter into the public debate. Canada’s Parliamentary Budget Officer has greater freedom than ours, although institutionally Canada has more feisty and independent parliamentary debate. As a consequence their PBO is always getting into fights with the government.

It is also worth noting that while independent and transparent fiscal oversight agencies, as in Canada, the UK and the US, allow a better informed debate, there is little evidence that the quality of budget decision-making is better. It may be that there is no ideal solution; the best option depends on how much one thinks the cost of policies matters to the democratic choice that voters make.