Someone suggested on the last thread that I should use the possibility of a December or January election as an excuse to write a blogpost about "Unseasonal Elections And Their Effects". I think he was probably trolling me, but I'm going to do it anyway. Never let it be said that I'm not accommodating.
Most general elections in recent decades have taken place in either May or June. There were a couple of elections outside "peak months" in April 1992 and October 1974, but the last truly "unseasonal" election was in February 1974, and just like the one that's about to come, it took place in the midst of a national crisis. Edward Heath's Tory government had a perfectly sustainable majority that could have seen him through all the way until mid-1975, but buoyed by favourable opinion polls, he instead took the fateful decision to seek a fresh mandate that would supposedly send a message that it was the elected government that governs, and not the unions. Polling day was 28th February - officially the last day of winter, although as we all know, early March often feels like an extension of winter in much the same way that early September often feels like an extension of summer. As polls closed the expectation was still that Heath's gamble would just about pay off, even though Labour had managed to stall his momentum somewhat during the campaign. But early results showed a surprisingly decent swing to Labour, and although the Tories did narrowly win the popular vote, that translated into a very slight lead for Labour in terms of seats. After a short delay of a few days, Labour leader Harold Wilson was invited by the Queen to form what was effectively a caretaker government until a new election could be held later in the year. Crucial to the outcome was the fact that every Ulster Unionist that was elected was opposed to the Sunningdale Agreement, and therefore no longer took the Conservative whip. If the UUP had still been inside the Tory fold, Heath would almost certainly have clung onto power, albeit at the head of a minority government. It was also, of course, a big breakthrough election for the SNP - they jumped from two seats to a new all-time high of seven. And there was a Liberal surge that didn't really produce any meaningful rewards as far as seats were concerned.
Does this tell us that Labour, the SNP and the Lib Dems can expect to do well in "unseasonal" elections? Probably not. I think the main thing it tells us is that winter elections are likely to only come about as the result of a major crisis, and that the outcome of the election will be determined largely by voters' reaction to that crisis, not by the temperature outside. Although oddly enough, the only other post-war winter election wasn't (as far as I'm aware, anyway) triggered by an immediate crisis - Labour PM Clement Attlee went to the polls in February 1950, a few months earlier than he needed to, and paid the penalty. His huge majority from 1945 was all but wiped out, and although he clung on to power for another year and a half, guerilla tactics in the Commons led his exhausted (literally physically exhausted) government to feel they had no choice but to call a snap election in late 1951, which they narrowly lost to Winston Churchill's resurgent Tories.
If you watch election results programmes from the distant past, you'll find the theory always used to be that a "high poll" (ie. a big turnout) favoured Labour, which might lead us to conclude that bad weather in winter that deters people from voting could be good news for the Tories. But arguably the 1992 result gives the lie to that - there was a bumper turnout of 78% (which hasn't been repeated in any general election since), but Labour did much worse than anticipated.
For my money, the biggest issue with a winter election is the slight danger of freak weather conditions such as the Beast From The East that would make it impossible for many people to vote, and to the best of my knowledge there are no legal provisions to postpone a vote at the last minute because of the weather. If, in a parallel universe, the Liberal Democrats had gone into coalition with the SNP in 2007 and had agreed to Alex Salmond's preferred date of St Andrew's Day 2010 for an independence referendum, there would have been major disruption because of heavy snow. There probably would have been controversy for years afterwards about whether the outcome of the vote was really legitimate.
Final thought: if the EU extend Article 50 until 31st January and we need an election before that date to break the deadlock, surely it'll have to be just before Christmas? I know it's getting very tight if that's going to happen, but the alternative would be either a mid-January election that would require campaigning to take place over the festive period, or a late January election that would be right up against the cliff-edge.