Wealthy Greens the new DLP

The Greens are siphoning the votes of angry Labor voters to the Liberals via preferences.

In all the millions of dollars the Prime Minister and Treasurer spent on market research and advertising plans for their new resource super-profits tax, they don't seem to have been told the richest voters in Australia are not Liberals but Greens.

I don't expect the Greens senators to realise this; there is, after all, a good reason why the Greens rarely win House of Representatives seats. But you would expect the self-appointed campaign gurus of the modern ALP to realise it, especially after spending nearly $40 million on research and marketing spin instead of simply releasing the Henry report as a green paper at Christmas and listening to the community, fixing the bugs and moving on.

In terms of dumb political economics, Labor's share dividend tax on Greens voters is up there with then shadow treasurer Wayne Swan opposing Peter Costello's tax cuts in 2005, which caused the final humiliation for Kim Beazley.

When we look at the per capita sources of income for the different voter groups, the ones with the strongest profile for investment income, which includes investment homes, bank deposits and share dividends, are the Greens. This was the case in 2007 and would be the case now.

So the political group with the most to lose from a super tax on mining - ultimately funded by looting mining share dividends - is the Greens rather than the Liberals.

And almost half the ALP's present marginal seats are won on preferences from minor parties such as the Greens, a party that is also very close to breaking through in some of Labor's traditional safe inner-city seats.

In addition to investments, the Greens income comes from high per capita wages and salaries, then small businesses, then superannuation. The only income source relatively untapped by Greens at the 2007 election was welfare or transfer payments, the main income source for the ALP base vote.

So political modelling for Labor votes doesn't work for Greens.

Greens votes in 2007 were defined by what they studied at university: arts, society and culture, architecture and education. Professionally they tended to be consultants, or worked in the media, health or education.

Greens 07 were very well-paid inner-urban renters who made extensive use of public transport and had few religious convictions. They tend not to have children until their late 30s, if at all, which makes them even richer and gives them lots of spare time to organise local political activities and annoy the rest of us.

Some of them still haunt uni campuses, churning out more Greens arts graduates, but increasingly now the Greens comprise a well-heeled professional group. Most are inner-urban dwellers in their 20s and 30s, sending their one rather indulged child to a private school. In their 40s and 50s, they adjourn to nice spots such as Noosa or a pretty little tax-deductible farming property, dabbling in winemaking or exotic fruit, while they send their children to the same inner-city private school, but as five-day boarders.

In their dotage they move back to the outer suburbs, with worm farms and backyard chook pens, tending their raised vegetable beds according to the Peter Cundall weekly gardening guide.

In retirement, their income from superannuation is second only to that of the average Liberal. They're still rich compared with Labor voters and they're still dependent on income from shares through superannuation. They also start buying more shares.

In all phases of life, Greens are distinct from the typical Labor, Liberal or Nationals demographic but remain supportive of Labor for social rather than economic reasons. Not unlike middle-class Catholics in the 1950s and 60s.

But, as in the 50s and 60s, the evidence from 2007 shows the Greens bloc is shifting away from Labor in terms of voting for its economic, rather than its social, interests.

Politically, in 2007 the Greens were strongly linked to all the main groups swinging to the Coalition after preferences. A big Greens vote in the inner-city professional seats meant a swing to the Coalition after preferences, because the Greens took more primary votes from Labor candidates than they gave back in second preferences, even though their preference drift to Labor averaged almost 80 per cent.

While they were still Greens at heart, these Greens who voted for John Howard after preferences tended to be of the lower income, more mainstream variety. That was because, in the main, they had more children.

But the key to understanding the political effect of the Greens is where their votes came from, rather than where they went.

This is the context in which the sudden jump in Newspoll's Greens primary vote can be viewed. If we compare the last Newspoll with the 2007 election, the Greens primary vote has risen by 8.2 per cent. The ALP primary vote has fallen by 8.3 per cent in the same period.

Even if 80 per cent of these extra Greens votes of about 8 per cent preference back to Labor, this change in the Green-Labor vote since 2007 still loses Labor an extra 1.6 per cent of the preferred vote nationally if repeated this year.

The point for the broader Labor movement to bear in mind is that Labor is expecting to win four out of five preferences from the richest and best educated group of voters in the community after imposing a tax on their main source of income. Putting this as politely as one can, it looks unlikely.

It's hard to see Labor winning from here if the slump in Labor's primary vote isn't repaired at the expense of the Greens.

For starters, Melbourne and Sydney seats would be lost to the Greens, and perhaps Denison in Hobart.

Seats with high concentrations of miners - especially of the fly-in, fly-out kind, such as Swan, Hasluck, Solomon, Herbert and Macarthur - look terminal already and Western Australia, with its booming mining economy, is starting to look like a wasteland for Labor, as it was in 1974-75.

This is food for thought for the union officials presumably trying to win new mining industry members by supporting a tax on mining jobs. There is indeed a place for these blokes on Labor's present front bench.

When we look at the investment and superannuation income by federal seats as they now stand after boundary changes, we see a broad range of vulnerable seats such as Melbourne Ports, Brisbane, Bennelong, Gilmore, Corangamite, Deakin, Robertson and Hindmarsh. Then there are the ultra-marginal ALP seats that have margins of less than 1 per cent, such as Macquarie, Dickson and Bass.

And if the 80 per cent preference flow figure falls below 70 per cent, Labor loses an extra percentage point of its national preferred vote.

Governments are won or lost on these sorts of margins.

The Greens vote is now moving former Labor voters across to the Liberals in the same way that the old Democratic Labor Party moved Catholics into the mainstream Liberal ranks 50 years ago.

The recent Newspoll indicates the same trend is afoot, although it is reasonable to expect that a higher regard by Greens for the Liberal leadership would accelerate this trend. Bear in mind, however, that this trend was already evident in 2007 when the Coalition was led by Howard, not really a totemic Greens figure for academic lefties, professional consultants or retired worm farmers.

So I'd expect it to continue to the election this year. It could improve considerably for the Coalition if a rolled-Green, inner-urban, professional republican such as Malcolm Turnbull were invited back into the tent as part of the Liberal leadership team. This is a pretty good test of how fair dinkum Abbott is about winning.

But you would be a braver man than me to predict what motivates Greens voters. At the moment, they comprise one in six Australian voters, totally disenchanted with the ALP leadership over the budget and the emissions trading scheme, but they don't particularly like Abbott either.

The point for all of us to remember is the Greens tend not to have studied maths or economics, but they didn't get rich by being stupid.

Look at some sample seats.

Melbourne: Labor was on 54.7 per cent in the category of two-candidate-preferred against the Greens, and an 8 per cent swing from Labor to the Greens, as we've seen in Newspoll, gives the seat to the Greens on about 53 per cent.

Sydney: Labor drops to 40 per cent, the Greens rise to nearly 30 per cent and the Liberals remain on about 26 per cent. With 80 per cent-plus of the Liberals' second preferences, the Greens end up on about 52 per cent.

Denison may be a problem if the swing to the Greens is more than 10 per cent and the Liberal vote drops slightly. Grayndler is still OK for ALP, as is Batman.

Labor could also lose three of its ultra-marginals, on 51 per cent or below.

For example, Macquarie is on 50.1 per cent. Suppose the Greens vote goes up 8 per cent at the direct expense of the ALP primary vote but Labor wins back only 6.4 per cent, at an 80 per cent preference drift. Labor suffers a net loss of 1.6 per cent of the preferred vote. So all three are lost, and Labor loses government.

If the drift drops back to 70 per cent because of the super tax, then the swing against Labor is 2.4 per cent and Labor loses the whole shooting match by a big margin.