Anyone who’s ever encountered a level crossing in France may have seen a sign reading Un train peut en cacher un autre — one train can hide another. That warning also applies to the two by-election defeats for the Conservative party this week. So spectacular was the Liberal Democrats’ win in Tiverton and Honiton that it risks obscuring the significance of Labour’s supposedly so-so win in Wakefield. Yet, inasmuch as by-elections can be harbingers of a general election to come, the latter matters just as much — if not more — than the former.

That’s not, of course, to deny the import of what happened in Devon on Thursday — and not just because it joins the list of iconic victories by Britain’s third party, going all the way back to Orpington in 1962. Following on from Chesham and Amersham and North Shropshire, it completes a hat-trick of huge wins in traditionally true-blue seats for Ed Davey’s party over the past year.

True, if the experience of the 1980s and 1990s is anything to go by, then all three of those seats may well return to the Tories at the next general election — especially if the sheer scale of their defeat in Devon persuades Boris Johnson’s MPs to ditch him. The loss of Eastbourne to the Lib Dems in 1990 persuaded their predecessors to ditch Margaret Thatcher and John Major, her successor, held on in 1992.

But that should come as no comfort at all to the Conservatives. There are somewhere between 20 and 30 seats (recent boundary changes make it difficult to be more precise) that the Lib Dems could take from them on a 10-point swing at a general election. And the swing in Tiverton and Honiton was a truly terrifying 30 points.

Moreover, what was particularly encouraging for the opposition about the result there, and certainly for the Lib Dems, was the seeming willingness of Labour voters to vote tactically, setting aside any lingering resentment at the party going into coalition with the Conservatives in 2010. Yes, Richard Foord is the constituency’s new MP mainly because the Tory vote fell by 22 points while the Lib Dem vote rose by 38. But the fact that Labour’s vote-share collapsed from 19.5 per cent to just 3.7 per cent came in very handy too.

That collapse matters because there were 17 seats where the Lib Dems came second to the Tories in 2019 and where the number of votes won by Labour, the Lib Dems and the Greens put together exceeded not only the Tory majority but the combined total for the Tories and the Brexit party. If, at the next general election, the Lib Dems can squeeze the Labour (and Green) vote in anything like the way that they’ve managed to do in these recent by-elections, those seats are ripe for the taking.

Sadly, for Keir Starmer’s party, there are actually relatively few seats in which they finished a decent second last time round and where Lib Dem voters returning the favour and voting tactically for Labour is likely to make that much of a difference.

Obviously, were Liberal Democrats to win a fair number of Conservative seats at the next general election, it wouldn’t harm Starmer’s chances of becoming prime minister, albeit, perhaps, as the leader of the largest party rather than one with an overall majority. But his main task is persuading people who voted Tory in 2019 to switch directly over (and, in many cases, back) to Labour.

And that’s why we shouldn’t ignore Wakefield.

Admittedly, the result there was hardly a ringing endorsement of Starmer and the Labour party, relying more, perhaps, on the Conservative vote dropping by 17 points than Labour’s rising by 9. Labour might also be a tad concerned that some former Tory voters may have plumped for minor parties rather than coming over to them.

But a 13-point swing in a so-called Red Wall seat still shouldn’t be sniffed at. For Conservatives, the Lib Dems are clearly a worry. But, even after Thursday, they simply aren’t their main competitor. That’s Labour, and a swing on the scale of the one Labour achieved in Wakefield would be more than enough to kick the Tories out of office at the next election.

Originally published at https://www.ft.com/content/92685858-90cf-42b5-8f63-ae882114fad5