How many seats are 'safe' and how many votes 'count' under First Past the Post?

5 April 2011

The YES campaign have criticised the First Past the Post electoral system, saying that it results in only a minority of voters having an impact on the national outcome of elections, with most people's votes effectively 'not counting'.

Greg Dyke: "First Past the Post is broken. The last election was decided by fewer than 460,000 of voters — just 1.6% of the electorate. No wonder people feel so remote from politics and feel it doesn't address their everyday concerns. It is time for change and that is why I am backing a Yes vote."

Tim Farron: "Research has shown that the result of the last election was decided by fewer than 500,000 votes in a handful of constituencies that, by mathematical accident, happened to be marginal. That's out of nearly 40 million eligible voters. That means that only one in every 80 voters actually mattered last year."

'Yes to fairer votes': "The system has also meant that half of the seats in the UK are effectively 'safe' and are unlikely to ever change hands, effectively giving their MPs jobs for life. This easily leads to complacency and simply taking voters for granted. And we all saw the results of that in the MPs' expenses scandal."


A study by the Institute for Public Policy Research, a UK think tank, say that in the 2010 general election, 9 million voters were in marginal seats and 20.5 million people were in safe seats. They say that people who voted in safe seats were very unlikely to have an impact on the election result.

However, saying the votes of the people who live in 'safe' seats did not count is unfounded, as it would suggest your vote does not count simply because a majority of people you live with in your constituency support the same party as you.

It would instead be more accurate to say that with First Past the Post, some people's votes have more impact on the national outcome of a general election, because they live in marginal seats.

Looking at the issue of 'safe' seats within the UK under First Past the Post, the IPPR report cites data by Mark Pack which shows that: "...since 1945, one third of seats have consistently been held by the same party, a figure which rises to half of all seats since 1970."

Because this issue involves deciding what a vote that 'counts' is, there is inevitably a level of subjectivity involved in any research.

The study by the Institute for Public Policy Research (IPPR), looking at marginal seats from the 2010 general election, has sought to look at how many votes 'counted'.

The report argues that a seat with a majority of 10 per cent or less can be seen as marginal. This would work out as 31 per cent of overall voters, around 9 million people, living in marginal seats. These, the IPPR report argues, are seats where votes effectively 'count'.

The IPPR also looks at the number of voters in the majority of the winning candidates in the 2010 general election in seats that changed party hands (swing seats); and so were decisive in determining the outcome of the election. From this, they are able to calculate that around 460,000 voters, or 1.6 per cent of the electorate determined the outcome of the election.

Therefore, Greg Dyke and Tim Farron are both using accurate figures in estimating the number of voters who were decisive in determining the outcome of the election.

However, it would be a stretch of the meaning of this calculation to go onto say that the other 98.4 per cent of votes case in the general election did not count.

It is important to note that under an AV system, votes in marginal seats would still have more weight in determining the outcome of national elections. Only a true system of proportional representation would end this bias. 

Full Fact fights bad information

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