Gerrymandering is basically the way physical electoral boundaries influence the outcome of an election. The following diagram from a Washington Post article last year (adapted from Stephen Nass) illustrates how the drawing of electoral boundaries can seriously change and distort who wins an election:
The above diagram is a simplified example which shows how moving electoral boundaries in a state with 5 electorates and 50 voters can change the outcome of an election. In the example above, 60% of the people typically vote blue and 40% of them typically vote red. In an ideal world, a voting system which accurately represented the people of this state would elect 2 candidates from the red party and 3 candidates from the blue party to parliament.
The first split shown – labelled ‘Perfect representation’ – illustrates that the only safe way to achieve an accurate representation of the political perspective of these voters, is for all voters of the same political persuasion to live right next to each other. It’s the equivalent of saying – all Labor voters must live in one suburb and all Liberal voters in another. Clearly that’s not practical.
In the second split shown above – labelled ‘Compact but unfair’ – the electorate boundary lines mean that each of the five electorates includes an equal number of supporters from both the red and blue parties. On the surface this sounds fine. However because each electorate only votes in one candidate, it results in only blue candidates being elected, and those who support red candidates bring unrepresented in parliament. This is not an accurate representation of voters’ wishes and is what often happens to the Green vote in Australia. This is because Green supporters are distributed across all Australian electorates – meaning there is rarely enough Green voters in a single electorate to get a candidate elected.
In the third split – labelled ‘Neither compact nor fair’ – the concentration of red voters in a small number of electorates means that the state ends up with three red party representatives in parliament and the blue party only with two. Again this is a distortion of the intention of the voters in that state. This is arguably what happens with the National vote in Australia. They get less than half the votes of the Green Party but have seven times as many representatives in the House of Reps. Why? Because National voters are concentrated in only a few electorates – instead of distributed across the country – so they end up with more representatives than their primary vote suggests they should have.
The bottom line is that the way boundaries are drawn between different electorates – or groups of voters – will determine how many representatives from each party end up in parliament without necessarily any regard for what people’s preferences are about this. It also demonstrates how difficult it is to get an accurate representative model when you are using physical electoral boundaries alone – as our current election model in the House of Representatives does – to determine who should represent us in parliament.
The reason the Nationals get more seven times as many seats in our House of Representatives than the Greens is not because the Australian Electoral Commission has failed in its job to allocate electoral boundaries fairly, or that someone is rigging the system – it’s because the electoral system itself is flawed.
The good news is that there are alternative voting models which factor in both voters’ location and their political perspective – delivering a more accurate political result for voters. (See here for more information.)