West Lancashire makes it three in a row for Labour

Three by-elections in three months with three big swings to the opposition

The third by-election in a Labour vs Conservative seat in three months took place last night in West Lancashire, where the MP since 1992 Rosie Cooper stood down to become Chair of the Mersey Care NHS Foundation Trust. It hands Labour the third retention on a big swing in a row in a ‘top two’ contest. The scores on the doors:

Labour: 62.3 (+10.2)

Conservatives: 25.4 (-10.9)

ReformUK: 4.4 (+0.1)

Liberal Democrats: 4.1 (-0.8)

Green: 2.9 (+0.5)

Monster Raving Loony: 0.9 (+0.9)

Conservative to Labour swing: 10.55 points

This was a very similar result to the one Labour recorded in Stretford and Urmston last December - the swings were almost identical swing (10.45 points then, 10.55 points now), in both elections Labour’s vote share rose to its highest level in more than 30 years, and in both contests the Conservative share fell below even the New Labour low ebbs. West Lancashire is a more competitive seat than Stretford though - it was Tory held in the 1980s and competitive as recently as 2010. It now looks very safe for the opposition.

Despite much backbench and frontpage anxiety about a renewed challenge on the Brexit right, ReformUK once again failed to live up to the hype, winning barely any more votes at the by-election than its Brexit Party predecessor managed in 2019. Consolidating the right is a popular strategy in parts of the Conservative party, but would do little to help in contests like this, where the Conservatives are losing the centre ground to Labour. A hard line on small boats won’t win over many seats when the ReformUK vote remains consistently below 5 per cent.

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

After three solid by-election performances in a row, Keir Starmer’s by-election record is starting to look a bit healthier in historical terms. The average swing to the opposition in ‘top two’ (Lab vs Con) seats is now up to 5.5 points, a figure which rises to 9.3 points if we take the period since December 2021, leaving out the by-elections conducted during the ‘vaccine honeymoon’ of summer 20201. Each of the last three five changes of government - 1970, 1974, 1979, 1997 and 2010 - has been preceded by an average double digit swing from government to opposition in top-two by-elections. Labour’s performances over the whole Parliament still fall a bit short, but their performances since ‘partygate’ broke are in line with this yardstick.

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)

We can also update Labour’s performance against two other yardsticks - average by-election swings to the opposition in Parliaments where the government goes on to win, and those where the incumbent loses. Another healthy swing hass improved Labour’s standing against both measures. The average swing to Labour over the whole Parliament - 5.5 points - is now ahead of the average swing in Parliaments where the incumbent wins another term (4.2 points) though still well behind the average swing in past Parliaments where the opposition won the next election (8.9 points). However, if we restrict ourselves to the period since December 2021, Labour’s average swing is now, at 9.3 points, ahead of the historical average for winning oppositions. West Lancashire is the latest in a run of performance which history suggests points to a victory to come.

“Big two” by-elections per year, Parliaments since 1945

We may not get many more chances to gauge Labour strength in “big-two” by elections. These are much less frequent now than they used to be. In the 1950s and early 1960s, politics junkies could looks forward to a steady diet of by-elections, with an average of 8-12 a year, nearly one a month in busy Parliaments. In the last few decades, this figure has seldom exceeded 3 a year. In the final New Labour government, there were just 4 Con vs Lab or Lab vs Con by-elections in the whole five year term. There were more in a single month several times in the 1950s.

One reason for the decline is the change in the political landscape. The vast majority of seats in the 1950s were straight Labour vs Conservative contests, so nearly all by-elections were “big two” fights. Far more seats in recent decades have involved a third party in one of the top two slots. The fragmentation of politics means less opportunities for government and opposition to test their strength in a straight fight.

This isn’t the only cause of the decline though. Far fewer MPs are vacating their seats than used to be the case. Three factors account for a lot of this: elevation, expiration and resignation. Many of the contests triggered in the early post-war decades were the result of elevation, with MPs vacating their seats in the House of Commons to take up a seat in the House of Lords. Conservative MPs in particular got bumped up to the Lords (or inherited a title) an awful lot in the middle of Parliaments in the early post-war decades. More recent governments have tended to avoid moving sitting MPs to the upper chamber at midterm, perhaps to avoid putting seats at risk, while Lords reform and a more middle class Conservative parliamentary party have put paid to inheritance driven by-elections.

Better health, younger MPs and rising life expecancy have all sharply cut expiration driven by-elections. Labour MPs were, and are, more likely to die in office mid-term, triggering a by-election, but death rates have fallen sharply. Thirteen MPs died in less than two years at the tail end of the 1974-79 Parliament, with the resulting vacancies complicating the task of overworked Labour whips struggling to maintain a Labour minority government. Vacancies due to the death of an incumbent have averaged less than one per year in more recent Parliaments.

Finally, MPs seem less prone now than before to simply resign, either in order to take up another job, due to poor health, or just because they no longer want to be politicians. Modern Parliaments have a higher share of career politicians, who seldom leave except to take up other political jobs. The parties may also have become better at persuading restive or disaffected incumbents to wait for a general election before moving on. Rosie Cooper is the first MP to depart for a non-political job since 2017, when two moderate Labour MPs unhappy with the leadership of Jeremy Corbyn - Tristram Hunt and Jamie Reed departed Parliament for jobs outside politics.

There is, however, one source of hope for by-election fans eager for a few more big party showdowns before election day: recall petitions. The Recall of MPs Act 2015 has already triggered two by-elections directly and contributed to a third.¹ A lengthy suspension from the House following a Standards Committee report is one of the triggers for a recall petition, and one prominent MP is currently under investigation by the Standards Committee for well documented and widely reported misdeeds during the COVID lockdown period. If the parliamentary sanctions applied to Boris Johnson are sufficient to trigger a recall petition, then there could be at least one more blockbuster by-election left in this Parliament, with a former Prime Minister fighting for his political life in a seat which, on current by-election form, Labour would be strong favourites to win.

1. Boris Johnson’s botched attempt to avert a recall petition for Owen Paterson ended up triggering the North Shropshire by-election it was seeking to prevent, when Paterson resigned his seat