Does AV lead to more spoiled ballots?

7 April 2011

Changing the voting system we use to elect our MPs could increase the risk of voters failing to understand the system, and inadvertently returning invalid ballots, or so some of opponents of the change argue.

Were such a situation to arise this would lead to more people accidentally being denied a say on polling day. But is it guaranteed that such a rise in spoiled ballot papers would occur?

The No2AV campaign has released figures from Australia and certain Canadian Provinces suggesting that where AV is used the rate of rejected ballot papers is higher. These have been picked up in media reporting

For example, in Australia the rate is 5.6 per cent compared to under one per cent in the UK.

It seems a reasonable assumption that a change of system, particularly to one perceived as more complicated, may lead to increased confusion. But there is a debate over how applicable these two examples will prove to be to the UK experience.


In Australia there are several factors contributing to the rate of spoiled ballot papers that are not directly related to AV as it would be applied in the UK.

Compulsory voting, compulsory ranking in some elections, long candidates lists, complex ballots, different systems for different ballots, and the language proficiency of some voters.

A report by the Australian Electoral Commission into the House of Representatives election in Australia looked specifically at the issue of informal voting (spoiled ballots), reporting some important conclusions.

On the issue of informal ballots, it reports: "More than half of all informal ballots in 2010 had incomplete numbering or were totally blank... While it appears that most informal voting continues to be unintentional, there was a substantial increase in assumed intentional informal voting (in particular, blank ballots) at the 2010 House of Representatives election."

Looking at why there is such a high number of informal ballots, they conclude that while a number of issues are important, "English language proficiency and the number of candidates appear to be the strongest predictors of informality rates (or changes in informality rates) in 2010.

The example ballot paper below, taken from the British Academy report into AV, gives an example of a ballot paper, where 22 candidates must be ranked.

Another issue associated with the high numbers of spoiled ballots in Australia is compulsory voting, where some voters who have to vote will turn-up and spoil their ballot.

The British Academy report goes on to argue that it is not how complicated an electoral system is that determines the levels of spoiled ballots.

It says: "There is no good evidence that any of the systems we consider here is 'too complicated for voters to understand', as is sometimes claimed. Instructions to scrutineers are sometimes extremely complicated, notably for STV, but instructions to voters are not. However, when voters are asked to vote in two or more simultaneous elections using different electoral systems, the number of spoiled ballot increases. Notable examples of this were the Scottish Parliament and local government elections in 2007 (when ballot paper design exacerbated the problem), and also the 2004 London Mayoral and Assembly elections, which were held simultaneously with the European Parliament election."

The report also looks at Australia, where in some elections ranking is compulsory: "In elections to the Australian Senate, until 1983 if voters did not rank order all of the candidates their ballot paper was deemed invalid and not counted; after that date, if there were 10 or more candidates voters had to rank order only 90% of the candidates for their ballot papers to be deemed valid."

In the UK, neither voting nor ranking all candidates would be compulsory, taking away one issue that might lead to increased spoiling of ballots.

Further, how complicated ballot designs and instructions are, and whether voting education and support is adequate to overcome language issues are not determined by what electoral system is being used.

Canadian Provinces

In Canada, the provinces of Alberta, Manitoba and British Columbia used AV at various points between the 1920s and 1950s. Under AV in these states the number of spoiled ballots was higher than under their later plurality voting system (First Past the Post system).

A study published in the Canadian Journal of Political Science looked at the changes in the rate of rejected votes when these provinces changed away from the AV system.

The findings suggested that the rate of spoiled ballot papers was higher when AV was in use — but that there were a variety of factors driving the trends.

The study's author, Harold J. Jansen, argued the particular complexity of the voting process in British Columbia under AV also added to spoiled ballots, while there were differences in spoiled ballots between Alberta and Manitoba depending on whether a single X next to one candidate's name was accepted or rejected. The study also pointed to a general higher level of spoiled ballots in other non-AV systems during this period.

The relative inconclusiveness of the Canadian situation makes it difficult to assess how much impact the shift to AV had on the figures, and thus even harder to draw firm lessons for the UK.

However, a briefing paper from the Political Studies Association, a political science and current affairs organisation, concluded the following: "Evidence from the Canadian provinces that formerly used AV suggests that AV in the form proposed for the UK might increase the number of spoilt ballots. So long as ballots where an "X" was placed against one candidate were counted as valid, however, the difference was very small — certainly insufficient to draw any general conclusions."

The Electoral Commission confirmed to Full Fact that if a single X is marked alongside one candidates name, with no markings next to any other candidate, then this will treated in the same way as if a '1' had been marked, signifying a first preference under the AV system.


There is a legitimate concern that the change of voting system risks increases the number of voters who fail to complete their ballot papers in the right way.

However whether this will ultimately happen, and to what extent, cannot be gauged with any certainty by reference to international examples.

For example there are other elements of the Australian electoral system contributing to their higher rate, than simply the elements that will be replicated in the UK should voters back the change of system.


Full Fact fights bad information

Bad information ruins lives. It promotes hate, damages people’s health, and hurts democracy. You deserve better.