# Proportional representation

The Australian Senate is the least democratic part of Australia's democracy. There are two major problems. The first is the above-the-line voting means that the final Senate seat in each state in each election is decided by preference deals. These are technically available to the public, but even if people were to see them, their interpretation is not always easy. The second problem is that each state gets equal representation. This is appalling, since NSW has a much larger population than Tasmania.

I think that a better method would be either the D'Hondt method or the Sainte-Laguë method. Both work on the same principles. The latter method tends to give more seats to the minor parties than the former.

The voting system is simple: you tick the box of your preferred party. In this proposed system, there are no state boundaries. Because I don't want parties with tiny proportions of the vote potentially getting a seat (and because it saves me digging through all the "Others" in the data), I've assumed half-Senate elections, as they are now.

Using the first-preference figures from the 2001 and 2004 Senate elections, we get the following distribution of seats (2001 + 2004 = total):
 Party % Votes Current method D'Hondt Sainte-Laguë Current with no states Lib/Nat 41.8, 45.0 18 + 21 = 39 17 + 20 = 37 17 + 18 = 35 17 + 18 = 35 ALP 34.3, 35.0 12 + 16 = 28 14 + 15 = 29 14 + 14 = 28 14 + 14 = 28 Dem 7.3, 1.8 4 + 0 = 4 3 + 0 = 3 3 + 1 = 4 3 + 1 = 4 Grn 4.9, 7.7 2 + 2 = 4 2 + 3 = 5 2 + 3 = 5 2 + 3 = 5 ON 5.5, 1.7 0 + 0 = 0 2 + 0 = 2 2 + 1 = 3 2 + 1 = 3 FF 0, 2.1 0 + 1 = 1 0 + 0 = 0 0 + 1 = 1 0 + 1 = 1

The last column of the table requires some guesswork. I've assumed the current system, but allocated 38 seats to the country as a whole, not split them into states. Of course, I don't know how preferences would have gone in such a case, so I made reasonable guesses. It is interesting that it appears to give the same results as Sainte-Laguë, but of course we don't know if it really would have done so.

I think the D'Hondt system should be used.

Edit from much later (31/8/07): If any of these systems were used, I would like a 2% threshold requirement, to keep the really small parties out of parliament.
Tags:
• Post a new comment

#### Error

default userpic
When you submit the form an invisible reCAPTCHA check will be performed.