The Audit - 2025

What did I get right and wrong?

Every year, in my last piece before Christmas, I do an audit of posts to see what I got right and wrong, with some lessons to take into next year’s writing (here are the ones for 2022, 2023, and 2024).

I do it for two reasons. First, it keeps me honest during the year as I know that if I make any “bold” predictions that aren’t justified I’ll have to own up to looking daft. Secondly, it forces me to confront mistakes and try to learn something from them. If you think I’ve been too kind to myself or have any other feedback please do stick it in the comments.

The audit starts with UK politics and policy, then the US and how I thought Trump 2.0 would play out. It concludes by assessing posts looking at broader trends and setting out lessons to take into 2026.

UK politics

Labour:

Looking back at posts from 2024, my assessment of the challenges the new government would face was largely accurate. I repeatedly expressed concern that Keir Starmer didn’t have a clear philosophy or sense of direction, which has now become widely accepted as one reason for the government’s struggles.

In a post written the day after the election last July I noted that Labour now held 90 constituencies with Reform in second place and another 40 where the Greens were runners-up. I thought this was going to lead to divisions over how to deal with values issues. There was an opportunity to govern on “competence” but that was going to be difficult given the fiscal situation:

“It remains hard to see how Labour will get through a full spending review without significant tax rises and/or changes to fiscal rules. Changes to the planning system and industrial strategy might, ultimately, generate more growth. But it will take time, and in any case, any benefits could be dwarfed by international ructions created by wider global instability, especially, as if now looks likely, Trump wins the Presidency again.”

In August 2024 I tried to estimate the size of the “real” gap between what the new government would want to spend and revenue available. I thought they’d need “well over £60 billion” more. In the two budgets since taxes have gone up by £68 billion.

In a post on the new Parliament I argued that Labour’s majority was shakier that it looked and the first big rebellion would probably be on welfare, which it was:

“MPs have pressures on them beyond those of party loyalty. Apart from their consciences, they have professional and personal networks whose support they wish to retain, and in which they want to improve their status…..Labour’s PLP….is dominated by people with professional backgrounds in services that have been hit by austerity, and where they have had more direct experience of poverty.”

I didn’t expect Starmer and Reeves to be quite as unpopular as they are now but it was always going to be extremely challenging, and the decision to run a very cautious campaign that wasn’t honest about the scale of fiscal problems was always going to come back to bite them.

By this summer it was clear that Starmer was in real trouble. I wrote one of the first pieces, back in July, saying that he could be challenged for the leadership after next May’s elections. Early last month I took a look at possible replacements – most of whom have been on fairly obvious manoeuvres since. We won’t know how accurate my thoughts on likely winners are until next year (if there is a contest).

One of my lessons from previous years was not to react too quickly to things like reshuffles and resignations, which can seem like big moments but often turn out to be unimportant. It’s easy to overestimate the significance of an event in the immediate aftermath. Generally I’ve stuck to this but I did do a piece after the September reshuffle as I was pretty confident about why it had happened and what it signified. The explanations have held up well, and, as far as I know, I was the first person to suggest that Lucy Powell might be a surprise contender for the Labour deputy leadership, which she went on to win.

The Right:

The decline of the Tory party has been an ongoing theme of this substack since the start. I’ve kept coming back to their core structural problem:

“Their challenge is finding a way to reconnect with younger and middle-aged people earning, or hoping to earn, decent salaries and own a home. It is these people who formed the core of the historic Tory vote and they have been comprehensively lost.”

Added to that they now have the unprecedented problem of an alternative right-wing party with significant support. At the start of the year I did a post on whether Reform would kill the Tory party. So far things have played out as predicted:

“It seems more plausible that the parties will settle into an uneasy equilibrium where neither is able to kill the other and both retain 15-30% support. As long as Reform can avoid imploding through internal tensions, they are not going to lose much of their support back to the Tories, as most of their voters strongly dislike Badenoch’s party. Equally enough Tories dislike Farage that even if they continue to underperform they will retain support. Right wing media will remain both Reform curious and institutionally attached to the Conservative party. There are enough rich old men who don’t like Rachel Reeves to sustain the finances of both parties.”

But I also noted that this isn’t a sustainable equilibrium in a first past the post electoral system. More recently I updated with another post on the potential for a Tory collapse – a lot is going to depend on whether key people defect or not. For now the polls have stabilised.

Predicting that Reform was going to struggle running councils was not a difficult call – and it has.

Elections:

As always I attempted to predict all the results in the May local elections. I got the four mayoral votes right and the Runcorn by-election, as well as 15 of 23 councils.

But I missed the extent of Reform’s victory, I thought they’d win three councils outright and they won ten. That’s because they got a significantly higher share of the vote than polls at the time suggested. This could have been predicted because their voters are much more enthusiastic than those for other parties, which matters in low turnout elections. The same thing happened with UKIP in the 2013 locals. We have a big set of elections coming up in 2026 and I’ll be taking enthusiasm levels into account this time (though I’d expect turnout to be higher in the contests for the Welsh and Scottish parliaments).

My other main error this year on UK politics was a sin of omission. I didn’t spot the importance of Zack Polanski’s election as leader of the Greens, or the immediate impact it’s had on their polling, which has changed the dynamic of politics in the “centre/left bloc”. Again this should have been predictable, given the party was pulling votes away from Labour during the election campaign, so were primed to do well. Shifting to a single leader, with a lot more charisma, and a clearer strategy has unlocked the potential that was always there. I also overestimated how much Polanski would alienate voters in more rural areas who have shifted Green, and have different values to urban progressives. That doesn’t seem to have happened (yet). So another lesson to take into 2026 is spending more time looking at the Green vote, and the party’s potential ceiling.

Policy:

Writing about politics lends itself to predictions more than policy. But I’ve focused quite a lot on the NHS this year, highlighting reasons why the government were not going to make much progress against their targets. In February I looked at how they were getting caught in the same trap as previous governments – stuck between wanting to improve immediate outcomes in hospitals while talking about long-term shifts to community and preventative care. I returned to the same theme when they published their “10 year plan” in July, frustrated that there was so little discussion about fixing the immediate capacity problem in hospitals. (Something I was a victim of later in the year).

Waiting lists have barely fallen in 2025 and A&E numbers are getting worse. Though, to be fair, productivity has improved more than I expected, it’s just leading to yet more appointments rather than people being removed from lists. The ongoing junior (‘resident’) doctor strikes have also had an impact.

Elsewhere I wrote about universities, special needs in schools and the appalling state of the criminal justice system. All issues where problems are getting worse and will dog the government next year.

US politics:

At the start of the year I wrote two posts looking at how Trump’s second term might play out. The first focused on what the set-up of his team told us. It correctly identified that his West Wing operation would be more stable, in terms of personnel, this time, and what his main areas of domestic focus would be. I was also right the Republicans would get into a muddle on healthcare savings and that DOGE would save no money, as well as the combustibility of the Trump/Musk relationship:

“Whether their shared enjoyment in annoying liberals will be enough to hold them together when their interests clash will be the big thing to watch over the coming months.”

But there were some mistakes in there too. I thought tariffs would end up being used in a similar way to term one, as leverage for negotiations. That has happened but there has also been a more widespread imposition of tariffs on everyone. I also thought the administration would struggle to reduce irregular migration as per their first term, but they have reduced it very significantly, primarily via use of the military on the southern border and the incredibly aggressive behaviour of ICE.

There are two lessons here. One is that, while I realised Trump appointing a more loyal team would mean less resistance to his more extreme ideas, I didn’t fully think through what that would mean. And secondly, I was being UK-brained when thinking about the challenges of stopping immigration. It is just easier to close a land border than to stop boats crossing a small channel, especially when you have a much larger military and no interest whatsoever in legal niceties.

The second post was about Trump and the law. Again there’s a lot in there that’s held up. He has, as predicted, used the justice department for retribution more aggressively than last time. He has struggled to pass much legislation through Congress, though Republican representatives have been even more craven than I expected.

I was overly generous to Supreme Court justices John Roberts and Amy Coney Barrett, expecting them to split from the other Conservatives more often than they have. But, while they have contributed to some very politically motivated judgments, extending executive power well beyond previous limits – most of the big tests are coming up. We have not yet had judgements on tariffs, on Trump’s ability to fire Fed governors without cause, or on birthright citizenship.

I think my conclusion stills hold: we are not going to see the Supreme Court uphold decisions that are unquestionably unconstitutional (including direct election interference) and that Trump’s victories will come from the more insidious use of threats: “his consistent modus operandi is to impose significant costs on criticising him. The more this works the easier it is to do as opposing voices become more isolated.” And, also, that ultimately what will undo him is competence. Something I wrote about more recently in a post about his collapsing approval ratings.

Probably my favourite post of 2025 was the one I did explaining the emergence of Musk, Peter Thiel and other silicon valley billionaires as a force on the political right. It didn’t really offer any predictions but did highlight tensions that would prevent them from influencing government in the way they’d hoped.

The big picture

Along with the analysis of politics and policy I’m trying to write more about longer-term historical trends that help explain where we are. I’ve written a lot this year about the rise of the radical right and the intellectual tensions in the movement (e.g. here and here). While these types of post don’t lend themselves to specific predictions, we are starting to see these tensions become more prominent in, for instance, the rows over Nick Fuentes in the US, or about whether the CDU should work with the AfD in Germany.

I’ve also written a few pieces about the decline of literacy and the changing way people consume information – like this one on the increasing importance of TikTok in politics (Starmer joined the platform a few weeks ago). These are themes I’m going to come back to. I also want to write more on critical moments in recent decades like this post on the “financial crisis theory of everything”.

One question I think about a lot is: what am I not writing about that’s going to turn out to be critical? Readers often wonder why I don’t do much on the radical left, and my answer is that I don’t think it’s anywhere near as important as the radical right. It’s further away from power in almost every country and is fundamentally constrained by an economic worldview that’s not compatible with running highly indebted economies dependent on market sentiment. But the rise of the Greens in the UK, as well as the relative success of Die Linke in Germany and the left in France does suggest something is going on worth looking at in more detail.

I’ve also been pushed by readers to write more about the effects of climate change, which I keeping avoiding because I’m worried about my lack of technical knowledge. The same applies to AI. But in both cases I can do more to look at the political consequences of changes that most experts agree are likely without having to become one myself.

Lessons for 2026

Looking back at my recommendations to myself from previous years I think I’ve applied them fairly well. I’ve mostly avoided commenting on day-to-day noise and done better at focusing on underlying issues, though I still want to do more of this. When I have made specific predictions they’ve had a good hit rate (and there are a lot of elections next year).

Every year I say I want to find more positive things to write about and then fail to do so, though I did manage a post on why Britain isn’t Broken over the summer. This is largely because there is a lot of bad news around but I do worry that pessimism bias means I’m going to miss more positive trends when they happen, particularly when it comes to the economy and technology. But then I’m often more optimistic than readers, particularly on the fragility of the radical right and wannabe autocrats like Trump.

My lessons for next year:

• When making predictions for low turnout elections look at how enthusiastic voters of parties are as well as poll ratings.

• Spend more time looking at the left, as well as the right, as politics continues to polarise. In the UK context that means more analysis of the Greens and their voter coalition.

• When writing about other countries remember that the context for policy problems is different from the UK and the outcomes won’t necessarily be the same. (In the US this applies to immigration but also e.g. debt management, which is very different for a country that runs the world’s reserve currency).

• Broaden out the bigger trends I look at including more on climate change, technology and European politics.

• Keep looking for things to be positive about…

Let me know what you think I should be doing differently in the comments (I’m not sensitive) and I hope all our subscribers have a great Christmas.