Preferences That Decide Nothing and Everything
In a federal electorate, first-preference results are: Candidate A (major party) 38%, Candidate B (major party) 34%, Candidate C (Greens) 12%, Candidate D (independent) 9%, Candidate E (minor right party) 7%. Voting is compulsory and full preferential. On the two-candidate-preferred count, A leads B 52 to 48. Candidate D's how-to-vote card recommends D first, then E, then B. Candidate E recommends E first, then D, then A. In the actual count, 60% of D's voters follow D's card and 55% of E's voters follow E's card; the rest distribute roughly evenly between A and B.
1. Explain why the two-candidate-preferred result is the meaningful measure of contestation here, and what it reveals that first preferences do not.
Read the worked answer
First preferences show A ahead of B by 4 points, which understates how close the contest is once minor candidates are eliminated. Under full preferential voting, the winner is the candidate who reaches 50% plus one after exclusions, so the relevant question is where the 28% of votes for C, D and E flow.
The two-candidate-preferred count of 52 to 48 shows A winning but with a margin of only 4 points after preferences - a swing of about 2 points would change the outcome. First preferences alone would suggest a comfortable 4-point lead; the TCP figure shows the same 4 points but measured on a different, more decisive base.
The exam-relevant judgement is that first preferences measure first-choice support, while TCP measures the ability to attract preferences and thus to win the seat. A candidate can lead on first preferences and lose on preferences, and a party's primary vote can fall while its seat count holds because of favourable flows.
2. Using the flow assumptions, estimate whether D's card or E's card has more effect on the outcome, and explain the reasoning.
Read the worked answer
D's voters number 9% of the electorate; E's number 7%. D's card sends preferences to E, then B - so if followed, D's votes do not reach A until E is eliminated, and ultimately favour B. E's card sends preferences to D, then A - so if followed, E's votes ultimately favour A.
Estimate D's flows: 60% of 9% = 5.4 points follow the card toward E/B; the remaining 3.6 points split evenly, giving 1.8 to each of A and B. Estimate E's flows: 55% of 7% = 3.85 points follow the card toward D/A; the remaining 3.15 points split evenly, giving about 1.58 to each.
So D's card is the larger single influence, because D has more votes and a higher follow rate. But the two cards pull in opposite directions: D's card helps B, E's card helps A, and the effects partly cancel. The net advantage to A from E's card (about 3.85 plus 1.58) is smaller than the net advantage to B from D's card (about 5.4 plus 1.8). That is consistent with a narrow A win only if A's underlying position is strong enough to absorb it - a useful reminder that preference flows are marginal, not decisive, when the primary gap is large.
3. A proposal is made to replace full preferential voting with optional preferential voting for this electorate. Assess the likely effect on Candidate A.
Read the worked answer
Under optional preferential voting, voters may number only one box, and ballots that exhaust are set aside rather than transferred. Exhaustion tends to reduce the total pool of preferences available and to favour the candidate with the highest first-preference vote, because their opponents need transfers that may not exist.
A holds 38%, the highest primary. Under optional preferential voting, some D and E voters who currently must express a full sequence might stop at their first choice, so the 5.4 and 3.85 point flows shrink. The remaining preferences are likely to be the more committed ones, which may be more ideologically consistent - D's committed voters lean B, E's lean A.
The net effect on A is ambiguous but probably favourable: A already leads the primary and wins on TCP, and exhaustion reduces the chance that a large transfer of minor-party preferences overturns the lead. The judgement is to reason from mechanism (exhaustion, compulsory numbering) rather than to assert that optional preferential voting 'helps' a particular side; the direction depends on where the exhausted votes sit.