Not at all unreasonable

The Ombudsman's decision to force the government to release the initial Treasury costings of its student loan policy is clearly a victory for open government - and, to the extent that those costings are higher than those given by Labour, for the National Party. But there seems to be an effort by some parties to paint Cullen's decision to withhold them as being a meritless attempt to suppress embarassing information. IMHO, this is wrong. While Labour clearly didn't want the costings released before the election (because then it would have to explain why it was disagreeing with its own officials), the initial decision to withhold them was not at all unreasonable, and complied with longstanding convention.

For those who don't spend their spare time prodding the government for embarassing details, access to government information is covered by the Official Information Act 1982. Section 5 of the Act establishes a general principle that

information shall be made available unless there is good reason for withholding it.

Some reasons are conclusive, others may be outweighed by a public interest in access to the information concerned. One of the latter is

Maintain[ing] the effective conduct of public affairs through...the free and frank expression of opinions [...] to Ministers of the Crown

This is expressly designed to protect policy advice to Ministers. However, it isn't conclusive, and access to such advice is the very point of the OIA. The convention seems to be that policy advice may be withheld while the policy is still under consideration, and released only when the Minister or Cabinet have made up their mind(s). This is relevant to the student loan costings because they were prepared as part of the government's Long Term Review of Tertiary Education Expenditure, which is still ongoing. They would thus normally be withheld, and Cullen didn't do anything unusual in refusing the National Party's request. John Key didn't do anything unusual in appealing the matter to the Ombudsman, either, and the end result of that process was a decision that the public interest in seeing the cost of a policy before the election outweighed the normal interest in protecting "free and frank advice". While I think this is a Good Thing, and I look forward to using the precedent in future election campaigns, it's not at all obvious from the law that that is where the balance of public interest lies, and Cullen's initial decision doesn't look at all unreasonable or devious.

As for the costings themselves, it's pretty much what you'd expect: Treasury is staffed by economists, and so they naturally assume things that have been disproven by the empirical evidence. Which is why Cullen based his publicised figures on a second set of costings based on more reasonable assumptions about uptake. OTOH, having read Keith Ng's latest, the other assumptions seem less reasonable; and while Labour is already attempting to justify them as reflecting wider access to allowances and capped fees, they should make such assumptions explicit in their model. Otherwise, it looks suspiciously like a strapped chicken...