Benalla shouldn’t be a bogey for Howard or Beazley

Not too much can be read into the Benalla by-election. By-elections soon after general elections tend to be driven by momentum and are not a reliable guide to what is likely to happen in the next general election. There are a number of reasons for this.

In the first place the by-election is being held in the knowledge of the general election result. The government, particularly a first term government, is in its honeymoon phase. I don’t know of any qualitative research into electoral honeymoons but my guess is that they are unlike real honeymoons. They do not involve elation, but relief - relief that a decision has been made and that electors can again concentrate on the basics of their lives. A government’s popularity rises because electors avoid thought of political issues by psychologically endorsing the most recent collective decision.

By-elections just after a general election are particularly unwelcome, and accentuate this tendency. Electors are also likely to punish the party causing the disturbance. In the case of Benalla both of these currents were running in the same direction, re-inforcing each other.

Benalla also appears to have been won on the basis of a campaign for a tactical vote which used the knowledge of the general election result as a crucial ingredient. This is an ingredient that will be missing in a Federal election. Voters were asked to vote for Bracks because it would make Benalla a marginal government seat giving it more leverage than a marginal opposition seat. Marry this to a perception that they had been neglected by the outgoing Kennett government and abandoned by their former Member and a 7.4 per cent swing was not out of the question.

From a Federal perspective it is also in a different sphere of government. While the bush is understandably unsettled, in Victoria it is the Kennett government that it can directly blame for a loss of services. At a Federal level it is the Keating government that put in place similar reforms. The next Federal election will be five to six years after the defeat of the last Keating government. This is just within the time horizon where the electorate will blame a previous government for creating problems rather than the current one for not fixing them. The Coalition will project the shadow of Paul Keating onto the electoral canvass during the next Federal Election campaign considerably darkening Beazley’s credibility on services to the bush.

So, while the cumulative swing of the general election and the by-election was 15.3 per cent to, this seat is not in the same category as a Bass or a Canberra by-election - both bellwethers warning of a change in government - even though it is significant.

However, there are lessons to be learned. As the Prime Minister has observed, the habitual vote for the major parties has declined markedly. The corollary of this is not just that party loyalties have fractured, but that group loyalties have also splintered. Instead of one general election there are in reality up to146 separate tussles. In this context Benalla should not be uncritically seen as representing the type of the bush seat, just one example of a seat that has swung.

The distribution of the Hanson vote in the last Federal and Queensland State election illustrates the point. In the Eastern States, the Hanson vote peaked in Blair, dwindling as it moved away.  It took in rural and coalmining areas and was almost non-existent in Victoria.

 Benalla is being represented as a Victorian version of the same pressures. But in other parts of the country those pressures are just as likely to show up in seats like Oxley, or Newcastle (both held comfortably by the ALP) , as well as some marginal National Party seats like Wide Bay and Richmond, or marginal Liberal seats like Longman. However, numbers of these seats are not really bush seats at all - a rural seat an hour's drive from the capital is not bush, just rural. They are on the coast, and in the case of Richmond and Wide Bay marginal courtesy of the creep of urban development into erstwhile rural heartland. It would therefore be a mistake to think that the bush issues are the ones to play to win these seats, or that rebadging a wing of the party as country anything is going to be attractive to voters who don’t see themselves as rural.

Part of the Benalla theory relies on the possibility that a swing of 15 per cent might occur in a number of rural seats and sweep them across to Labor before the marginal city seats where most commentators have so far seen the next election being fought. Swings of that magnitude would wipe the Liberal and National Parties out entirely in the bush, leaving them with a ring of marginal urban electorates. Mike Steketee in The Australian picked Gwydir as a possible Labor win from the National Party, but Gwydir, on a margin of 13.6 per cent is the Coalition’s 12th safest seat. Surely these sorts of swings would show up in the publicly published polls which currently have the Coalition in front of the ALP at a point in the electoral cycle where it is more likely to be behind?

Indeed, when one looks at the swings on a seat by seat basis from the last election it is apparent that much of the swing back to Labor probably occurred during the last election. In a pointer to the Victorian State Election Gippsland actually swung 9.8 per cent to Labor. Wide Bay and Longman (two more of Steketee’s picks) swung 15.3 per cent and 9.0 per cent respectively. Liberal Campaign Strategist Nick Minchin reportedly told businessmen that the key to the next election is what happens to the One Nation Vote. But that vote was sourced from both Labor and Coalition, and under our preferential voting system most of it returned whence it came. It is not a block waiting to be picked up holus bolus. If the vote runs at the national average of 8.43 per cent and you win 60 per cent of it, then you gain an advantage of 20% over your opponent, or 1.7 per cent. That will do damage in a marginal seat, but not in a safe seat.

The problem for both Howard and Beazley is how to cobble together a sufficiently large coalition of groups to elect them and set their parties up for the future. In a fractured electorate this involves some fine calculations. There are electorates in the country that need to be courted, but there are a lot more in the cities. It is a larger problem for Howard than for Beazley because Labor has only 10 seats that it would lose on a uniform swing of 3 per cent, while the Coalition has 21. Added to that only a couple of the seats Labor has to defend could be described as bush, suggesting Howard’s bush strategy a defensive one. But on the electoral pendulum only 6 of Howard’s 21 are genuinely bush seats either. Short of some catastrophe not on the radar screens of the publicly available polls urban seats will be the key to the next election. Of course this makes sense. We are a highly urbanised society. Not many live in the Bush.