Labour were able to tread water against Ukip in Stoke but sank beneath the waves in Copeland, where the Conservatives’ Trudy Harrison won the seat.
In Stoke, a two-point swing away from Labour to the Tories and to Ukip, which if replicated across the country at a general election would mean 15 Conservative gains and would give Theresa May a parliamentary majority of 40.
And in Copeland, a 6.7 per cent swing for Labour to Tory that would see the Conservatives pick up 52 seats from Labour if replicated across the country, giving them a majority of 114.
As the usual trend is for the opposition to decline from its midterm position at a general election, these are not results that indicate Labour will be back in power after the next election.. That holds for Stoke as much as for Copeland.
The last time a governing party won a by-election was 1982 – the overture to a landslide victory. It’s the biggest by-election increase in the vote share of a governing party since 1966 – the prelude to an election in which Harold Wilson increased his majority from 4 to 96.
To put the length of Labour’s dominance in Copeland into perspective: the new Conservative MP was born in 1976. The last Conservative to sit for Copeland, William Nunn, was born in 1879.
It’s a chastening set of results for Ukip, too. The question for them: if they can’t win when Labour is in such difficulties, when will they?
It’s worth noting, too, that whereas in the last parliament, Labour consistently underperformed its poll rating in local elections and by-elections, indicating that the polls were wrong, so far, the results have been in keeping with what the polls suggest. They are understating the Liberal Democrats a little, which is what you’d expect at this stage in the parliament. So anyone looking for comfort in the idea that the polls will be wrong again is going to look a long time.
Instead, every election and every poll – including the two council elections last night – point in the same direction: the Conservatives have fixed their Ukip problem but acquired a Liberal Democrat one. Labour haven’t fixed their Ukip problem but they’ve acquired a Liberal Democrat one to match.
But that’s just the electoral reality. What about the struggle for political control inside the Labour party?
As I note in my column this week, the settled view of the bulk of Corbyn’s internal critics is that they need to keep quiet and carry on, to let Corbyn fail on its his own terms. That Labour won Stoke but lost Copeland means that consensus is likely to hold.
The group to watch are Labour MPs in what you might call “the 5000 club” – that is, MPs with majorities around the 5000 mark. An outbreak of panic in that group would mean that we were once again on course for a possible leadership bid.
But they will reassure themselves that this result suggests that their interests are better served by keeping quiet at Westminster and pointing at potholes in their constituencies. After all, Corbyn doesn’t have a long history of opposition to the major employer in their seats.
The other thing to watch from last night: the well-advertised difficulties of the local hospital in West Cumberland were an inadequate defence for Labour in Copeland. Distrust with Labour in the nuclear industry may mean a bigger turnout than we expect from workers in the nuclear industries in the battle to lead Unite, with all the consequences that has for Labour’s future direction.
If you are marking a date in your diary for another eruption of public in-fighting, don’t forget the suggestion from John McDonnell and Diane Abbott that the polls will have turned by the end of the year – because you can be certain that Corbyn’s critics haven’t. But if you are betting on any party leader to lose his job anytime soon, put it on Nuttall, not Corbyn.