Hard Evidence: does door-to-door campaigning work?

Miliband is fighting a grassroots campaign - but will it push Labour over the line? Nick Ansell/PA Wire

With the 2015 General Election set for May 7, Britain’s political parties have lost no time in stepping up their campaign activities. The country faces four months of intense electioneering, as the parties jockey for position in what is likely to be the least predictable contest since February 1974.

In his opening campaign salvo of the year, Labour leader Ed Miliband announced his party’s intentions to run a strong grassroots campaign. Miliband said his party would hold town hall meetings and talk directly to four million voters on their doorsteps in the run-up to polling day.

The aim is to boost the party’s vote share. In a close-fought contest, anything which increases a party’s vote – even a small amount - could mean the difference between success and failure. But will Labour’s doorstep strategy work?

A considerable amount of academic research has been conducted down the years on the effectiveness of grassroots election campaigning, so we are in a good position to judge. Encouragingly for Labour, the consensus view is that constituency campaigning really does pay dividends. The harder parties campaign locally, the greater their share of the vote there, and the lower their rivals’ shares.

We can illustrate this by looking at the effect of local campaigning on the result of the 2010 general election. By law, every candidate standing for election in the UK must make a public declaration of how much they spent on their constituency campaign. Past research has shown this is a very good proxy for how hard the candidate’s campaign worked locally. So, collating those spending declarations for all candidates in all seats gives us a good indicator of how effective the local campaign was.

The graph below shows the average differences in vote share in 2010 between seats where each party spent nothing on its local campaign and seats where it spent up to the legal limit. Spending is restricted in the final months of the campaign, and this graphic looks specifically at campaigning in the period between January 2010 and polling day.

Charles Pattie, Author provided

To get a sense of what this means, think of two hypothetical constituencies which are identical in every way except in terms of how hard Labour campaigned there. Let’s say that in the first constituency, Labour did not campaign at all, while in the second it spent up to the legally permitted maximum on its campaign.

The Labour candidate in the second seat could expect a 2010 vote share almost 9 percentage points higher than the candidate in the first. The Conservative candidate would have done slightly worse (by around 0.6 percentage points), and the Liberal Democrat candidate noticeably worse (by almost six percentage points) in the second seat, compared to the first seat. Similar effects hold for the other parties’ 2010 constituency campaigns.

We can use the same models to estimate how the 2010 election would have turned out had the parties’ local campaigns been at different levels of intensity. Below, we compare three different scenarios: the number of seats won by each of the major parties in the actual election; the numbers they would have won had Labour not campaigned in any constituency; and the winners had Labour mounted a 100% campaign in every seat. In both of the latter scenarios, it is assumed that the other parties’ campaigns were fought at their actual intensity.

Charles Pattie, Author provided

Had Labour decided that its local campaigns were a waste of resources and had not made any effort in the constituencies, its defeat in 2010 would have been even worse than it actually was: about 40 more Labour seats would have been lost. But had the party been able to campaign to the hilt everywhere – and had its rivals not responded – it could have emerged as the largest party.

The research shows that meeting voters really does help. Face-to-face campaigning, where politicians and party members talk direct to the public, is shown to be more effective as a means of persuading people to vote than more remote forms of campaigning, including telephone calls, e-mails, social media interaction and leafleting.

So on the face of it, Ed Miliband’s plan is a sensible one. Extensive and active constituency campaigning does work, especially when it involves direct contact with voters. In a tight election, it could make all the difference between victory and defeat. That said, Miliband would be unwise to expect dramatic gains from his local campaign strategy, as there are some major obstacles to overcome.

A tough fight

The research on constituency campaigning suggests that its effect comes primarily from ensuring that those who already support a party actually do turn out and vote for it. It is much less effective as a means of making new converts to a party’s cause. Another consideration is that political parties do not campaign in a vacuum – if Labour boosts its local campaigning efforts, the other parties will respond in kind, partially cancelling out gains.

Labour’s plan to directly contact four million voters is quite ambitious. But political parties are not awash with resources. Constituency campaigns are mostly financed by local donations and fund raising, and most local parties are very strapped for cash indeed. While voluntary effort will help, the parties do still need to fund the local battle. Whether they can afford to do so at the level being suggested here is a moot point.

This year, Labour, the Conservatives and the Liberal Democrats face a much more uncertain environment than in previous elections. In previous contests, they could and did finesse their resource problems by focusing their local campaign efforts in a relatively few marginal seats. Each party’s campaigns tended to be low-key in seats where it either had no chance of winning, or was almost certain to retain the seat.

But in 2015, the big three could face serious threats from insurgent parties like UKIP and the SNP, even in seats they would normally expect to be safe. Campaign resources which would normally go to the marginals may well be reserved to avoid embarrassing defeats in seats usually considered safe. So even if Miliband does achieve his target of four million conversations, the question remains – will they be in the right places to help his party?