Dealing with a junior coalition partner always has its difficulties. The Liberal Party (in its different incarnations) has more than 90 years’ experience of dealing with the Nationals (formerly the Country Party), and it’s still not all roses. The Labor/Greens relationship is much newer, so it’s no surprise that it’s a lot stormier.

Keep that background in mind when reading the story in this morning’s Age about a prospective preference deal between Labor and the Liberals. The idea is that the Liberals will preference Labor in the federal seat of Melbourne, held by the Greens’ Adam Bandt, and Labor in return will preference the Liberals in Mallee, held by retiring Nationals’ MP John Forrest. With any luck, each major party would therefore pick up a seat at the expense of its junior partner.

For the Victorian Liberals, preferencing against the Greens has rapidly become conventional wisdom since doing it in the 2010 state election, which they surprisingly won. The two things might, of course, be unrelated, but they have become conflated in the Liberals’ minds, so preferencing Labor is seen as a winning strategy. (Many Liberals also seem to regard it as obvious that the Greens are more ideologically opposed to them than Labor is; I find this view deeply mysterious, but that’s a topic for another day.)

But if it becomes clear that the Liberals are going to preference against Bandt anyway, then the party loses its value as a bargaining chip.

Better to keep options open in the hope that either Labor or the Greens will offer something significant in return.

That’s the attitude Labor has always taken to directing preferences as between Liberals and Nationals: it’s generally had no ideological reason to prefer one or the other, so it’s allocated preferences in such a way as to cause maximum disruption and ill-will between its opponents. That often means doing deals with the Nationals, and also a strong tendency to preference against sitting members regardless of party.

What the casual reader won’t realise from the Age story is that the opportunities for this sort of dealing are very limited. In the vast majority of seats Labor and Liberal preferences won’t be distributed, so where they’re directed has only symbolic value. Mallee and Melbourne are among the rare exceptions.

Bandt won Melbourne in 2010 with 56.0% of the vote, after receiving just on 80% of Liberal preferences. If those had gone 80% the other way, he would have lost it by a similar margin. With the advantage of incumbency this time and Labor more on the nose, he’ll certainly be in with a chance even without Liberal preferences, but the betting would favour Labor.

Mallee hasn’t had a three-cornered contest since 1993, when Forrest was first elected. On that occasion he had a lead of 5.2% over the Liberals on primary votes, but Labor preferences flowed almost 60-40 against him. In the end he won with just 50.5%, a lead of 736 votes. Labor preferences in rural seats don’t flow in a unified manner because the party generally doesn’t have a lot of people to hand out how-to-vote cards; a lot of Labor voters are evidently reduced to guessing.

But what’s really striking about Melbourne and Mallee is how they show up the different state of coalition (or quasi-coalition) relations. For the Liberals, contesting Mallee is controversial; even though their sitting member is retiring, the Nationals are outraged at having to face a Liberal opponent, and The Age reports, plausibly, that “Tony Abbott was also opposed to a Liberal challenge”.

No one, on the other hand, seems to think it’s improper for Labor to try to win back Melbourne, even though Bandt has loyally supported the government. On the contrary, many Labor people talk about it in a way that suggests they regard it as a top priority and think of the Greens as a more dangerous enemy than the Liberals.

The Liberals would like to win Mallee, but it will always be a side issue for them. There’s no chance that they’ll pour vast resources into it and neglect marginal Labor seats. But for Labor there’s a real chance that its obsession with Melbourne will divert attention from the seats it needs to win against the Liberals, as happened at the last state election.

It seems to me that each of the major parties should be able to learn something from the other. The Liberals need to develop more of a spine and move towards the sort of self-assertiveness towards the Nationals that Labor displays towards the Greens.

Labor, on the other hand, needs to somehow find its way towards a more harmonious relationship with the Greens — one that allows Labor and the Greens to have different policy emphases and contest each other’s seats without treating each other as mortal enemies. At its best (and it’s often been at less than its best), the Liberal-National relationship shows them how it’s done.