Local elections: what is a “good” result?

When the Oldham by-election result was announced, commentators who had spent days predicting a UKIP win and failure for Labour were suddenly scrabbling around for alternative explanations. Pundits soon flipped their story from Labour losing because of Jeremy Corbyn to Labour winning despite him.

You can guarantee we’ll see something similar after May’s local elections. We will see spin from Corbynites worthy of New Labour, and of course on the other side there will be criticism from people who will perceive the exact same results as failure.

So I thought it would be helpful if I tried to clarify what might constitute “good” ahead of the game, rather than waiting until May 6th to pass judgment. And on that Friday morning, I will try and view the results with reference to the below. If you’re lucky, I may even write a follow-up piece.

I’ve written the below solely about the English local elections in May, although some of the historical data includes election results from across the UK. On we go, then…

Is a loss of seats acceptable?

This is the messaging coming out, with Corbyn’s team allegedly attempting to play down Labour’s chances. Nothing new in any of this; every party tries to give themselves a low bar to jump over on polling day. Losses of up to 200 seats have been suggested. I’m going to argue that this would be a disaster.

Here is a chart showing gains made by opposition parties in local elections, excluding general election years, going back to 1988.

In every single instance, opposition leaders gained councillors. The last time this didn’t happen, in 1987, Margaret Thatcher called a snap general election, and crushed Labour weeks later. Losing seats is not generally an indicator that things are going well.

A claim we’ll hear a lot is that 2012 represented a “high water mark” for Ed Miliband, coming as it did in the aftermath of the “omnishambles” budget. It is true that Miliband did very well, gaining 534 English council seats. However, these gains were from a low base following a poor showing in 2008; Miliband won 49% of the English seats on offer, up from 29% previously.

In 2014, Miliband actually won a higher proportion of English councillors with 50%, before losing badly in the following year’s general election. By this measure, a loss of seats in May would be worse than both 2012 and 2014.

To accept a fall in council seats would be to either admit that Jeremy Corbyn can’t match Ed Miliband, or to say that we can only beat the Conservatives when they are weak, as with the omnishambles. It could be argued that, with the Google tax fiasco, splits over the EU, and disputes with junior doctors, the government should be weak now too; and that perhaps Ed Miliband exploited Tory weaknesses more effectively than the current incumbent.

Away from the numbers, there are other reasons to not accept a fall in councillors. These are elected Labour officials, and form our base in various cities across the country. There is no way we should present the loss of Labour representation as an acceptable result.

But I suspect that Labour will not lose seats, and that talk of losses is a strategy to dampen expectations. In which case…

How do we measure success?

This is harder. Council seats are only one measure and we really need to consider three different aspects; the number of gains, the projected vote share, and where the gains occur. In terms of seats gained, here is a chart showing each opposition leader’s first set of council elections:

So every single one of the last six opposition leaders achieved at least 200 gains in their first council elections; and only two of those became Prime Minister.

So, what is good? Here, it becomes a little arbitrary. Martin Rogers here argues Labour should be looking to gain 100–150 seats. Elsewhere, Tristram Hunt is quoting poll data that suggest 400 seats should be the target; the data shows that, going back to 1974, 434 gains is the average for opposition parties, and 515 the average for new opposition leaders.

As a target, 400 gains is fine, but it’s an ambitious one; the number of seats contested vary across different elections, and Tony Blair’s phenomenal performance in 1995 warps the averages a bit. A par score from here feels like 250 gains; 400 would certainly be a good result.

But, as mentioned, seats aren’t the only thing. We also need to consider vote share. Below is a chart showing the difference between Tory and Labour projected vote shares in council elections since 1988 (projected vote shares are necessarily used as not all regions vote every year). The differential is worked out by subtracting the government’s vote share from the opposition’s:

Here, a 10% lead in vote share looks like the marker for success. The only opposition leaders to break through this barrier more than once are David Cameron and Tony Blair, who both achieved it in every council election they fought. The two big peaks are 1995–96 and 2008–09, both of which presaged a change of government.

Even a 10% lead doesn’t guarantee success; Michael Howard managed it twelve months before losing the 2005 election. By contrast, in the last parliament, Ed Miliband never outperformed the Tories by more than 7% in any local elections. If we are serious about winning, that’s where we need to be; 10% ahead of the Tories.

As for where we make gains, this is still harder to pin down, but we certainly need to be winning away from our heartlands. It’s no use piling up votes in safe seats; if we aspire to win the next general election, we need to reach beyond our base and win across the country. As Martin Rogers notes, we need to be winning in swing areas such as Southampton, Derby and Milton Keynes, and in southern seats such as Thurrock and Watford. The distribution of our support and our overall vote share will tell us a lot.

A final thing to watch will be turnout. One viewpoint held by supporters of Jeremy Corbyn is that he will expand the electorate by appealing to millions of non-voters. Turnout was 31% in 2012, and 35.7% in 2014. If Jeremy Corbyn really is energising non-voters, we should expect a turnout above 40%.


The English council elections are influenced by many local factors and setting expectations is tough. And there is the usual caveat that anything can happen between now and polling day. But I’ll be brave and stick my neck out.

Gains of 200–250 seats and a 5% lead over the Conservatives in projected vote share is acceptable. Not brilliant, but a platform on which to build, and would probably be enough to reduce the pressure on Jeremy Corbyn. Gains of 500 seats or more, and a 10% lead over the Conservatives — I’d start to reconsider my view of Jeremy Corbyn’s unelectability. For what it’s worth, my expectations are gains of 50–100 seats and a 2–3% poll lead.

No doubt any Corbynites reading will be screaming at this point, that a 10% poll lead is impossibly unrealistic, that we’ll never make 500 gains, and so forth. My response is, simply: a leader who is on course to win the general election in 2020 would. Tony Blair gained 1,807 seats in 1995 and had a 22% lead. As for David Cameron, in 2006 he picked up 316 seats and won by 13%.

We’ve heard a lot about how Jeremy Corbyn is our best hope, how he can appeal to millions. Surely, if this is the case, he can outperform Ed Miliband? Thursday May 5th is the first real chance for all Corbyn supporters out there to prove the rest of us wrong. And if being wrong means Labour smashing the Tories at the polls, I’ll happily accept that.

Two months to go.

Like what you read? Give Rob Francis a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.