Informal voting - don't blame the voters!

As the Joint Standing Committee on Electoral Matters begins its enquiry into the conduct of the 2004 Commonwealth election, one of its interests will be the continuing high rate of informal voting.

At the 2004 Commonwealth election, around half a million voters wasted their trip to the polling booth. These were voters who despite their best efforts to cast a vote for their candidates of choice, had their House of Representatives ballot papers rejected because they failed to meet the exacting formality requirements set down in the Commonwealth Electoral Act.

A total of 639,851 ballot papers were excluded from the House of Representatives count, 5.18 per cent of all votes cast, up 0.36 percentage points from the 2001 election. Yet the same voters had less difficulty with the far larger and more complex Senate ballot paper, 466,370 informal ballot papers, 3.75 per cent, down 0.14 percentage votes from 2001.

Research at past elections has shown that between a quarter and a third of informal ballot papers have no discernible first preference votes. Presumably these ballot papers represent the politically alienated, voters confused by voting procedures and voters simply unimpressed by the choice of candidates and parties. Many of these voters will have turned up simply to avoid being fined.

That leaves between two-thirds and three-quarters of informal votes with preference for one or several candidates on the ballot paper. These votes would have counted except for the other compulsion voters face, compulsory preferential voting. Despite a clear preference being evident, the complex rules for House of Representatives elections set down by the Parliament simply will not allow such votes to count towards electing representatives.

Clearly both forms of compulsion play their part in creating informal votes. The argument for compulsory voting is one of civics and minimum levels of political participation and the outcome is some spoilt ballot papers. But there is no civics argument in forcing voters who do turn up to vote, to then have to express a preference for every candidate. If compulsory voting sets a minimum level of civic participation, then compulsory preferential voting pushes the bar even higher.

Since informal voting became a political issue after a dramatic rise in its incidence at the 1984 election, most researchers and commentators have tended to address the issue of how voters can be made better informed on how to cast a formal vote.

To which my response is, why blame the voters? Why not blame the legislators, who preside over an electoral act which allows certain types of votes to be formal in the Senate, but the same vote in the House is informal? Why not blame the legislators for insisting on compulsory preferential voting, the cause of most informal votes?

For a formal House vote, a ballot paper must have a clear first preference, followed by a complete, unrepeated and unbroken sequential ordering of preferences for every other candidate on the ballot paper. The only “savings” provision is that if the last preference is blank or out of sequence, the final preference can be implied.

So in a field of 10 candidates, a vote will only be counted if it has valid preferences from 1 to 9, any missing or incorrect final preference being deemed to apply to the tenth square. A vote with less than nine preferences is informal. A vote with any duplicate preferences is informal, including any votes with two 9th preferences. Any elector who votes creatively, such as preferencing up from 1 using odd numbers, base prime numbers or factors of ten, would also have their vote declared informal. An elector who voted sequentially before marking the last square 100 would be casting a formal vote, but the same sequence with the last two squares marking 99 and 100 would be informal. Any use of ticks and crosses makes the vote informal, but a ballot paper completed with roman numerals is formal.

Yet at the same time, voters have less difficulty with the more complex Senate ballot paper. If voters cast a “below the line” vote, they must provide a valid sequence for every candidate, though a more generous set of savings provisions allows up to three sequencing errors. But most voters choose to vote “above the line”, casting a single vote in a group ticket voting square. In this case, a single “1” is formal, and the formality rules also allow a single tick or cross to be treated as a formal group ticket vote.

But any voter who uses this method on the House ballot paper is casting an informal vote. Past evidence has shown that between a third and half of all informal House ballot papers are of this type, which strongly suggests that voters are using the much simpler Senate voting system to mark their House ballot papers.

As ignorance is never a defence in law, perhaps it is all the fault of the voters for not reading the instructions? Yet even in these days of increased cynicism about politics, voters are entitled to believe that well informed legislators would not go conducting conjoint elections under rules where a formal vote on one ballot paper would be informal on the other. Can you imagine the howls if such strictures were applied to tax returns or BAS forms?

The impact of these rules on the 2004 election can be seen in the 14 candidate contest for the outer Sydney seat of Greenway. It recorded the country’s highest rate of informal voting at 11.83 per cent. In the Senate, on a forbidding Senate ballot paper nearly a metre long and including 78 candidates distributed across 30 columns, the informal rate was only 4.07 per cent, nearly a third of the rate in the House.

The inanity of compulsory preferential voting is that any vote for the Liberal or Labor candidates in Greenway needed a further 13 unnecessary preferences to be admitted to the count. An error with lower preferences would see the vote excluded from the count, even if those preferences would never need to be counted. Any Liberal or Labor primary vote could be excluded because at the 12th preference it did not have a clear distinction between candidates from the Fishing Party, the Citizens Electoral Council or the Independent campaigning for spelling reform. For a formal vote, all preferences must be correct for any preference to count.

The current formality rules and their insistence on absolutely perfect sequences of preferences demean the electoral process. Voters either carefully transcribe how-to-vote material on to the ballot paper, or are forced to randomly allocate preferences to unknown and unwanted candidates just so their ballot paper can pass the formality requirements and register a vote for the candidates they do want.

There are solutions to high informal voting, but it requires a deviation from the current fashion in Australian politics that all wisdom somehow emanates from Canberra. Different formality rules are applied at state and territory elections, and there are clear lessons that can be applied to Commonwealth elections.

First, jurisdictions using optional preferential voting have lower informal rates. Fully optional preferential voting applies in New South Wales, Queensland and the ACT, and all three jurisdictions currently have the lowest level of lower house informal voting.

Second, elections where voters receive an upper house ballot paper allowing single preference voting tend to have higher informal rates. So in each state at Commonwealth elections, and at Western Australian state elections, informal voting is higher, and it is also higher in the lower house compared to the upper house. In Victoria, which uses the same electoral system in both houses, the informal rate is lower, and it is lower in the lower rather than the upper house.

South Australia provides the model on how to lower the informal vote while maintaining compulsory preferential voting. All candidates in an electorate have the right to register one or two ticket votes. These registered tickets are displayed on voting screens in polling places as guides to voters, and are also used as a “savings” provision, allowing ballot papers with incomplete preferences to remain in the count. However, candidates are banned from advocating a vote that does not include preferences. The system is not designed to replicate the upper house group ticket voting system.

The effect of the provision can be seen in the informal rate at South Australian lower house elections.

Informal and Ticket Voting – South Australian Elections 1985-2002

Source: South Australian State Electoral Office. Informal vote as a percentage of total vote, ticket vote as a percentage of formal vote.

Ticket voting has halved the level of informal voting that would have applied under Commonwealth legislation. South Australian ticket voting saves ballot papers where the elector mistakenly uses the Legislative Council’s voting method in the House of Assembly. It also saves many ballot papers where an elector has made a mistake transcribing a party’s ticket of preferences.

What ticket voting does not save are ballot papers where an elector made a mistake creating their own sequence of preferences. Ticket voting is very kind to voters who have not understood the instructions, or who make a transcription error in copying preferences. It is less helpful for informed voters who make an honest mistake while voting to truly express their preferences.

Parliaments may insist on compulsory preferential voting, but it is doubtful whether voters have the same commitment. At the South Australian Constitutional Convention in 2002, the 323 attendees chosen as part of a deliberative poll nominated optional preferential voting as the most important reform that could be introduced to the political process in the state. (See summary of finding published by Issues Deliberation Australia.) This was despite the issue not being on the agenda of the convention.

Optional preferential voting always helps the candidate with the highest primary vote. The Labor Party has some history of supporting optional preferential voting, based on experience with the Democratic Labor Party and three-cornered contests prior to the 1980s. This was the thinking behind Labor’s introduction of optional preferences in NSW in 1980.

Yet more recent elections indicate that Labor is now the main beneficiary of compulsory preferences. Between 1955 and 1972, the Labor Party won only a single House of Representatives contest having trailed on the primary vote compared to 34 such victories for the Coalition. Since 1980, the Coalition has won only 5 contests where its combined vote trailed Labor compared to 41 such come from behind wins for Labor.

The introduction of optional preferential voting in Queensland before the 1992 election followed its recommendation by the Electoral and Administrative Review Commission. Initially both sides of politics chose to continue recommending preferences on how to vote material. In 2001, the Labor Party broke with practice and advocated “Just Vote 1” to its supporters. Supporters of other parties clearly took this slogan to heart, and the 2001 election saw a dramatic increase on ballot papers marked with a first preference only. According to surveys in a sample of districts, the number of ballot papers with only a single preference rose from 23 per cent in 1992 and 20 per cent in 1995 to 60 per cent in 2001.

The huge rate of exhausted preferences between competing conservative candidates saw Labor win seats in 2001 with as little as 35 per cent of the primary vote. Yet Labor has also found the system can work to its disadvantage. The failure of Green voters to direct preferences to Labor cost it the seat of Mulgrave at the 1995 state election, and also prevented Labor winning a clear victory in Mundingburra, Labor’s narrow victory eventually overturned by the courts and the Goss Government defeated after a subsequent re-election. Exhausted preferences have also thwarted attempts by Labor in NSW to run third and elect Independents in safe Coalition seats.

So victories achieved with less than half of the formal vote are raised as one of the failures of optional preferential voting. That Labor managed to encourage a higher rate of exhausted preferences with its “Just Vote 1” strategy is seen as a manipulation of the electoral system.

Yet why is this any more outrageous than manipulating compulsory preferential voting to engineer an outcome? There are numerous examples of major parties running dead in their opponent’s safe seats, slipping to third place and then using compulsory preferential voting to engineer victory for an Independent. An even more remarkable case was the Victorian state electorate of Warrnambool in the mid-1980s, when the Labor party ran third and elected a Liberal at a by-election. At the next state election, with the same three candidates, with the Liberal MP substantially increasing his vote since the by-election, Labor’s preferences this time elected a National MP. Yes, on both occasion, compulsory preferential voting delivered a majority vote after preferences. But if voters thought so little of their lower preferences that the simply followed a party’s recommendation, then what is the point of compulsory preferential voting? If the strategically directed preferences of the bronze medallist get to determine who wins gold and silver, it is surely a bigger manipulation of the electoral system than simply leaving the two leading candidates to fight it out between themselves.

Under optional preferential voting, as well as voters getting more choice, parties can avoid the pointless debates about who they are putting last on their how to vote cards. In both 2001 and 2004, the Queensland Labor Party decided not to choose between One Nation and the National Party. At the 2003 NSW election, the Liberal Party chose to exhaust preference in several seats rather than choose between Labor and the Greens. This latter example is surely a more sensible decision than at the 2004 Federal election, when after dire warnings about letting the Greens near the levers of power, the Howard Government chose to help elect the Greens in every contest where they had a legitimate chance of victory over Labor.

In the end, cross party concern on the rate of informal voting will encourage the Joint Standing Committee on Electoral Matters to recommend some form of change. The most likely solution is the South Australian ticket voting system. In the end, neither side is likely to offer voters what they would most want, which is the right not to have to vote for candidates they don’t know or don’t care about.