Labour's by-election hot streak continues

Big swings in Stretford and Urmston

The voters of Stretford and Urmston, just down the road from Swingometer HQ, braved the bitter cold yesterday to cast their ballots on a replacement for Kate Green. Green, the constituency’s Labour MP since 2010, is leaving Westminster after 12 years to take up the post of Deputy Mayor of Greater Manchester. This has triggered a second by-election in a Labour held North West seat in two weeks. The result is much like the first: a comfortable Labour hold on a big swing.

The scores on the doors:

Labour: 69.6% (+9.3)

Conservatives: 15.9 (-11.6)

Green: 4.3 (+1.6)

Liberal Democrats: 3.6 (-2.3)

ReformUK: 3.5 (+3.5)

Rejoin EU: 1.3 (+1.3)

Independent, Freedom Alliance, SDP: 1.8 (+1.8)

Conservative to Labour swing: 10.45 points

Another big win for Labour, with easily their highest share of the vote since the current seat was created in 1997. The Conservative vote share, which remained above 30% even in the Blair landslide years, has now slumped into the mid teens. The prospect of another Conservative MP in an area which returned Winston Churchill (grandson of the other Winston Churchill) as its MP for over 25 years now looks very remote indeed. Trafford, once a rare urban blue bastion, looks to be going the way of central urban areas across England - red and getting redder. The Conservatives, who controlled the local council as recently as 2018, now hold just 13 of the councils 63 seats.

This is the third double digit swing in a row from Conservative to Labour in a seat where the “big two” took the top two slots in the preceding general election. The last time Labour managed a hat-trick of this kind was in June 1994, when four by-elections held on the same day, in the middle of the contest to replace John Smith following his sudden death in May, delivered an average swing to Labour of 18.9 points. Tony Blair, though he secured some whopping swings in his years as opposition leader (including a post-war record Conservative to Labour swing of 29 points in Dudley West), never strung together three double digit swings in a row as opposition leader. In fact, I can find no historical precedent for three double digit swings in a row from a Conservative government to a Labour opposition in “big two” contests across the whole post 1945 period.

A strong portent, then, for Labour. But how strong? In the chart below I show the average swings from government to oppposition in “big two” by-elections - seats where the Conservatives and Labour were first and second in the previous general election - in each Parliament since 1945.¹ Five of the last six changes of government have been preceded by average swings of over 10 points in "big two" by-elections - 10.9 points ahead of the 2010 election, 14.4 points ahead of the 1997 election, 10 points ahead of the 1979 election, and 12.1 points ahead of the 1970 election. The only exception is the 1974 election, and even here one can argue that weak by-election swings (4.3 points on average) were an early omen of the messy outcome to come.

Average government to opposition swings, “big two” by-elections, Parliaments since 1945

Blue: Conservative opposition; Red: Labour opposition. Darker shades indicate Parliaments where the opposition went on to win the next general election

Starmer’s overall and recent performances as Labour opposition leader are shown on the right hand side of the chart. Labour’s overall performance across the whole Parliament, shown in grey, is still mediocre - an average 3.8 point swing. This is much better than Jeremy Corbyn managed in either of his spells in opposition, but well behind the swings achieved by Ed Miliband or Neil Kinnock in Parliaments where Labour went on to lose. But Labour’s performance in 2022 looks much more promising - an average swing of over 9 points. This is not far off the opposition performances achieved by Mrs Thatcher and David Cameron in the years before they took office, but also not much ahead of Michael Howard in 2001-5 and Neil Kinnock in 1983-87.

We can compress the lesson from by-election history even further, by comparing the average performance in “big two” by-elections since 1945 in Parliaments when the opposition went on to win with performance when the incumbent ended up being re-elected. There is a big difference, highlighting that consistently strong by-election performances for the opposition in “big two” contests are a meaningful leading indicator of victory to come. The opposition manages an average swing of 4 points in Parliaments where the incumbent goes on to win again, this rises to 9 points in Parliaments where the opposition goes on to take control.

Labour’s performance against this yardstick currently depends very strongly on the time horizon we use. Taking the Parliament as a whole, history suggests Labour’s average swing of 4 points points to another defeat next time. But if we take only the performances since last December, when Labour’s current hot streak got going, the average swing is 9 points, exactly in line with the historical average of past oppositions heading to victory. By-elections and opinion polls are all telling the same story right now - Labour are on course for victory. But there is a long way to go, and as the first two years of this Parliament remind us, the political mood can change very quickly.

Average swings to the opposition since 1945 in Parliaments where the incumbent wins (grey) and when the incumbent loses (black), and Labour’s performances since 2019 (light red) and since December 2021 (dark red)

1. The February to October 1974 Parliament is missing as there were no “big two” by-elections in this Parliament.