Oz Politics Blog

Today’s SMH outlines the potential problem for Labor in the Werriwa by-election.

With no official Liberal candidate running in the by-election, Labor’s Chris Hayes finds himself besieged by voters angry over state issues such as the police response to the riots, the M5 toll, lack of help for small business, and the inadequacy of roads and facilities in the new suburbs of the electorate. A host of off-beat challengers may spoil his succession to the seat that was once held by Gough Whitlam.

…

He is facing a line-up of 15 other candidates who include a DJ promising cheap beer, a former Labor branch-stacker who hates the party, and a 19-year-old economics student. But a preference-swapping campaign against Labor by the minor parties and independents could rob Mr Hayes of the seat if his primary vote falls below 50 per cent.

Newspapers thrive on this sort of speculation. As we all know, the small risk of novel change sells more papers than a steady-as-she-goes-more-of-the-same story. So it is worthwhile testing the proposition against facts.

We will begin by considering the primary vote for Labor in Werriwa in recent elections. This is set out in the table below.

From this table it appears possible that Labor’s primary vote for Werriwa could fall below the 50 per cent mark. This brings us to our next consideration. Let’s contemplate a few recent by-elections for Labor seats to test whether it is likely to fall, and if so, how far it might fall below the 50 percent mark.

Of note, these figures suggest that it is at least even money that the Labor primary vote increases in a by-election following a general election. In part, I suspect this was a consequence of the absence of a Coalition candidate in each of these by-elections. All other things being equal, it suggests that Hayes could well do better than Latham in 2004.

But Hayes doing better than Latham is not guaranteed. The 1996 Blaxland by-election has a number of parallels with the 2005 Werriwa by-election — a retiring Labor leader immediately after a Labor loss. Yet Keating’s primary vote in Blaxland in the 1996 general election was down more than ten percentage points on 1993; in contrast Latham’s 2004 vote was up a little over two points on 2001. This suggests that unlike the Blaxland by-election, there could be some room for Labor to move into negative territory in Werriwa in 2005.

Cunningham 2002 is the disaster scenario Labor fears in Werriwa 2005. In that by-election, Labor’s primary vote fell by more than 6 per cent. But there are a two critical differences between Cunningham and Werriwa.

First, Labor’s core primary vote is much higher in Werriwa. If, as in the Cunnigham scenario, it dropped six percentage points in Werriwa, it would still be over 46.5 per cent. For Labor to lose with a primary vote of 46.5 per cent, it would require a very tight preference deal, with every minor party and independent either putting an agreed candidate second or at least putting Labor last.

Here is where the rabble of 15 other candidates should play into Labor’s hands. Labor would only need to be ahead of the second-place-getter preference for 7 out of every one hundred voters that did not primary vote for Labor. In Cunningham, all of the candidates gave their preferences to the Greens ahead of Labor. Given there are candidates far to the left and far to the right of Labor in the Werriwa by-election, I find it hard to believe they would all agree a single candidate to second preference. On the day, I imagine there would be Green voters who would put Labor before (say) Janey Woodger from Australians Against Further Immigration (AAFI) and the right of centre Joe Bryant. Similarly, I imagine there is a small number of right of centre voters who would put Labor before (say) the Greens and Mike Head’s Socialist Equality Party. This should be enough to see Labor over the line — even in this worst case scenario of a six percentage point slump.

Second, I suspect that Labor’s slippage (if any) would not exceed 3 percentage points, and a primary vote of 49 per cent would almost certainly see Hayes elected. The factors that drove the slippage in Cunningham do not appear to be present in Werriwa. There is not an alternative union supported candidate like Peter Wilson in Cunningham who would steal primary votes from the Labor candidate. Latham’s resignation was for understandable health reasons; it is unlikely to be perceived as self-indulgent like Stephen Martin’s, which led to the Cunningham by-election. And notwithstanding the efforts of Alan Ramsey, Chris Hayes’ preselection for Werriwa does not appear to be as on the nose as Sharon Bird’s was for the 2002 Cunningham by-election.

All in all, while it is technically possible that Labor’s Chris Hayes could lose Werriwa, the odds should favour Hayes for the win in next Saturday’s by-election.