When I did the chart this morning for the betting market on the year of John Howard’s departure, it was much the same as it has been for some time. However, an hour or two after I did the graph, it became clear John Howard was staying to fight the next election. Fourteen or so hours after this morning’s graph, the betting market has adjusted.
You may need to hit the refresh button on your browser to see the latest chart.
Following last week’s CPI figures, the stock market is expecting one and possibly two interest rate rises between now and the next election. As a consequence, I had expected the betting market to downgrade Howard’s chances of winning the 2007 Federal election. So far it has not happened. It appears that Costello’s tilt at the leadership was more debilitating for the Liberals than an interest rate rise. Go figure.
The average of the four bookmakers suggested a 56.7 per cent probability of a Coalition government (unchanged on last week), and a 43.3 per cent chance of a Labor government following the 2007 Federal election.
The usual graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.
There are three recognised approaches to predicting election outcomes: opinion polls, the betting market and economic models. Today it is time to look at economic models.
Leigh and Wolfers made the observation, “The logic of the economic models is simple: voters are more likely to re-elect incumbents who deliver a robust economy. This pattern can be motivated either as voters providing an incentive for politicians to deliver good outcomes, or as voters using available information to discern high-ability incumbents.” Leigh and Wolfers discussed a number of economic models that have been developed to predict electoral outcomes. The most statistically significant variables across the range of models are: the inflation rate and the change in the unemployment rate over the electoral cycle.
On July 7 this year, Mumble repeated a plain-English paraphrase for the language of mathematics that underpins the economic models. “If two or more of [the unemployment, inflation and interest rates] rise over a full, three-year electoral cycle, the government will lose. Conversely, if two or more fall the government will be returned.” Apparently, this formula, which Mumble called the Sawford Formula, has picked every Federal election since 1961 except 1974 and 1980.
Yesterday’s jump in the inflation rate has resulted in speculation of an interest rate rise next week from the Reserve Bank (which other bank economists see as between 75 and 90 per cent certain, and which the stock market appears to be punting close to 100 per cent certain). As a consequence, we are starting to see increased speculation that the 2007 election will see a change of government driven by a change in the economic fundamentals.
However, there are some factors that need to be considered before we jump to any conclusions.
First, while there has been a 250 per cent increase in the price of bananas since cyclone Larry, bananas are easily substitutable for other items of fruit in the household budget. The Bureau of Statistics assumption that Australians have continued to buy bananas, at the same rate as before the price hike, is simply wrong. I don’t recall seeing any bananas at the fruit market when I was there last Saturday. I believe the impact of bananas on inflation should be ignored.
Fuel costs are more problematic to assess. If we assume that household fuel demand is relatively inelastic. Then the rise in petrol prices has a comparable demand dampening impact as a rise in interest rates. Where it becomes difficult is the broad-based contribution of fuel costs to production and distribution costs, which risks continued inflationary pressures – the so-called second round inflation. With underlying inflation at almost 3 per cent, this may already be the case.
The difficulty with rising fuel prices is not new. High crude oil prices in 1973, 1981 and 1991 all resulted in recessions. It could be argued that those high prices were supply driven and today’s prices are demand driven, largely from the US and China. However, with the current Middle East conflict, and the lingering Iraq war, the contribution of demand and supply elements to pricing is difficult to assess.
Speaking of the recession threat, Megalogenis made the observation that an interest rate rise would be felt disproportionately in Sydney.
Sydney has the nation’s most-exposed home borrowers. They are the most likely to squeal, and close their wallets, if the bank pushes up interest rates on Wednesday, as almost every pundit expects.
But what happens if Sydney responds with a mini recession, as it did in the second half of 2000?
The most recent national accounts confirmed NSW as our slowest economy, with state final demand crawling at less than half the national average: 2.1 per cent compared with 5 per cent in the 12 months to the March quarter. Western Australia, by contrast, soared 10.6 per cent over the same period, Queensland was up 9.2 per cent and even Victoria was looking good with 4 per cent growth.
In my view the Reserve Bank has a hard call. We must be getting close to the tipping point where the next interest rate rise (or the one after) will significantly impact on aggregate demand. Too much foot on the brake would see the economy crash into recession. With high oil prices, not enough could see inflation spiral out of control. Both scenarios would have disastrous electoral outcomes for the Howard government.
Morgan has broken its fortnightly reporting cycle to bring the results from the first weekend of face to face polling. The headline, national two-party preferred vote prediction had the Coalition on 52 per cent (was 46.5 per cent in the previous poll), and Labor on 48 per cent (was 53.5 per cent). It was a 5.5 percentage point movement in a week.
The national primary vote predictions were: 45 per cent for the Coalition (up 6), 36.5 per cent for Labor (down 6), 1.5 per cent for the Australian Democrats (down 1), 7.5 per cent for the Greens (down 1), 2.5 per cent for family first (unchanged), 1.5 per cent for One Nation (up 0.5), and 5.5 per cent for other.
Let’s call a spade a spade. These results are an anomaly. Public opinion does not move that quickly or far in a week, without a bloody good reason. John Howard’s triumph over Peter Costello is not a compelling enough reason. Furthermore, the volatility in the Morgan series since September last year is breathtaking.
The usual opinion poll graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.
According to today’s Australian, the ALP is on track to win government in 2007. Well that was the headline. What actually happened is Australia’s foremost psephologist, Malcolm Mackerras has produced his electoral pendulum for the 2007 Federal election, following the recently proposed (but not finalised) boundary changes in Queensland and New South Wales.
In new figures that will send a shiver down the spines of Coalition strategists, veteran election analyst Malcolm Mackerras shows that the swing required for Labor to unseat the Howard Government at the 2007 federal election has dropped from 4.4 per cent to 3.3 per cent.
…
When Mackerras releases his “pendulum” for the 2007 election, it will show that the new “median seat” - the least marginal Coalition seat required to fall for a change of government - will be Eden-Monaro, in rural NSW.
Eden-Monaro is held by Liberal MP Gary Nairn by a margin of 2.2 per cent, but will firm to 3.3 per cent after the redistribution. On the new pendulum, Eden-Monaro becomes the 14th most marginal Coalition seat. A uniform swing of 3.3 per cent would thus wipe out the Coalition’s 28-seat majority in the House of Representatives.
On present electoral boundaries, the median seat is John Howard’s seat of Bennelong, in northern Sydney. His margin of 4.4 per cent is cut to 4.0 per cent by the redistribution.
The average swing towards Labor over the past four Newspolls has been 4.3 per cent - precisely within the band that, on the new boundaries, would be sufficient to achieve government.
My pendulum based on the current boundaries is here. The new pendulum will need to make some interesting assessments about the three independents (Andren, Katter and Windsor), all of whom will be affected by the boundary changes.
Update 26 July 2006: The new Mackerras pendulum is here.
Update 28 July 2006: The Poll Bludger has done the math and gets a slightly different outcome to Mackerras