How John Curtice was wrong about Lib Dem MPs
Speaking in a Q+A session at a British Polling Council event near the start of the election, John Curtice was pretty down on Lib Dem prospects. Some of this was the view that’s been debated during the last Parliament about whether the party was, in his view, making a strategic mistake in whose support it was pitching for and the policies being put forward to achieve that.
But he also made a new point that even if the Lib Dems went on to get a good election result, it would be on the basis of MPs with low vote shares, and so leave the party – to use his analogy – up a creek without a paddle. We’ve seen a version of this argument already deployed over Labour – sure you won a huge landslide but look at your vote share… and I’ve also seen a few people knock the Lib Dem performance as being wins gifted to the party by Reform splitting the Conservative vote.
So what does the data now tell us about Lib Dem MPs? Handily Patrick Dunleavy has crunched these numbers for a different purpose already:
![Patrick Dunleavy data on size of MP majorities at 2024 general election v2](https://www.markpack.org.uk/files/2024/07/Patrick-Dunleavy-data-on-size-of-MP-majorities-at-2024-general-election-v2.png)
What those numbers show is that the record-breaking number of Lib Dem MPs wasn’t secured through low vote shares and unusually helpfully split opposition votes.
In fact, of the three largest parties, the Lib Dems have the highest proportion of MPs elected with over 50% of the vote and the lowest proportion elected with under 40%. (It’s also worth noting how well Labour does on these measures too, which is Patrick’s main point.)
It was the high levels of Lib Dem support that propelled the party to such success. Or to return to John’s analogy, the Lib Dem store room is overflowing with paddles. (Or something like that…)
Note: the first version of this post included a table with a small error in the number of Lib Dem MPs with a small plurality. The percentage figures, however, and hence the conclusions drawn were not affected by the error and Patrick has kindly provided an updated table that is now included above.
Sign up to get the latest news and analysis
"*" indicates required fields