In an election year, it was inevitable that both last week’s Budget and Kevin Rudd’s reply would be viewed more in political than economic terms. In particular, observers have been focused on the prospect of the government getting a popularity “bounce” from the week, and fairly salivating in anticipation of the next round of opinion polls.

On Saturday, Bryan Palmer at Ozpolitics poured a bucket of cold water on such expectations. It’s worth reading the whole piece, but his central message is that random variations make it impossible to interpret anything sensible from a single poll movement: “Whatever the papers say about a post-Budget bounce, don’t believe it.”

Palmer is right, but he leaves out a key point: in politics, perception can be more important than reality. The fact that the polls may be intrinsically meaningless does not make them unimportant. If commentators, albeit wrongly, find meaning in them, then political debate is going to take its cue from that rather than from the underlying mathematics.

With those caveats in place, we can now look at the first post-Budget poll to appear: Galaxy, in this morning’s News Ltd tabloids. It shows a complete absence of bounce. Labor leads (two-party-preferred) by 57% to 43%, a swing to the Government of just 1% since the previous poll.

A large majority, 67%, said the Budget would make no difference to how they voted. On industrial relations, Labor’s policy was favored clearly but not overwhelmingly, 52% to 35%.

Galaxy is the newcomer to the Australian polling scene, but its record so far has been excellent — the Courier Mail says that “Galaxy published the most accurate polling for the 2004 federal election and four state elections since”. We shouldn’t base judgements on one poll, but if we do, then Galaxy is as likely as any to be right.

As I’ve said before, the really striking feature of the last three months of opinion polls has been their consistency. If Labor can weather the post-Budget period –meaning not just one poll, but the next few weeks’ worth — then it will suggest very strongly that voters have made up their minds, and that Rudd’s huge lead may be unassailable.

The papers, with some justification, don’t really believe it. The Herald Sun remarks that if Galaxy’s figures were repeated at an election it would result in Labor “taking more than two dozen seats from the Coalition.”

The actual number is 43.