Why are so many ex-prisoners being sent back to prison?

0:00Hello and welcome to more or less with me Tim Haferd. Think of us as the Bob Dylan of BBC Radio than ever present and enigmatic stalwart devoted to constantly reinventing ourselves all while being slightly in love with our own mythology. This week, has the UK's stock of millionaires been depleted by 7%. Are European wildfires that record lows? Are seagulls stealing £171 million of food each year? And how many roads must a man walk down before they call him a man? But first, the fact that prisons in England are nearly at full capacity is an issue at the top of the political agenda and so also at the top of our inbox. The new Prime Minister, Andy Burnham, revisited the issue almost immediately after taking office. But meanwhile, questions are being raised about the decision back in 2024 to begin releasing some prisoners early.

1:02What effect has that decision had? Lawyer Listener, Laura Morgan spotted one BBC news story online that surprised her. I've just read this top story on BBC News with the headline, Prisoners Freed Early are returning to jail at rapid rate, Governors tell BBC. The implication is that early release makes a return more likely, but I can't see anything in the article which says that, is the headline justified? The BBC article says that you can see evidence of an increase in the number of people who've been recalled back to prison. There were 51,419 recalls in the year ending March 2026, a number that was up by 28% from previous year. But what does it mean to be recalled to prison? I asked Cassia Rowland, senior researcher at the Institute for Government. That means that someone who has been released from prison and is being supervised by the probation service gets sent back into prison normally for a fixed 56 day period.

2:06But sometimes it can be for the whole west of their sentence. Has the number of recalls gone up in the last few years? It's gone up massively. So the number of recalls in 2025 was about double the average in the late 2010s and even more importantly the rate of recalls. So the number of recalls for every 100 people on the probation caseload after release from prison went from 33 in 2018 to 74 in 2025. So we've seen this really big increase in just a few years. Wow, so that is 74 recalls for every 100 people in that probation system. That's right. It's important to flag that doesn't mean 74% of people get recalled. So there could be a few people who are being recalled over and over again, but it's still incredibly high. Yeah, and more than, more than double what it used

3:10to be. Exactly. Are they being recalled because they have committed another crime? Generally not. So it's about a fifth to a quarter of recalls because someone's been charged with a further offense. And the total number of people we called for a further offense has stayed pretty flat over time. Instead, people are being recalled mostly for non-compliance, which is basically just any other breach of the conditions of their release from prison. So that could be things like contacting an ex or it could just be missing an appointment or failing to live in a particular place or get approval for moving to a different place or even becoming homeless. They're more likely to be sent back to prison for a relatively minor breach than they were a few years ago. That's partly because the probation service is really stretched at the moment. You've

4:14got a lot of often quite inexperienced probation officers who are struggling with large caseloads and they kind of don't have time to do as much in-depth risk assessment and things like that as ideally they would be doing. And I think also there's this really risk-averse culture in the probation service that has kind of amped up significantly over the last few years. And so if there's kind of any doubt at all about whether someone might pose a risk or anything like that, they just straight away get recalled to prison rather than exploring potential ways of managing that risk in the community. So fundamentally, this rise in the number of recalls that we're observing. Is it fair to say that that is a measure of stress on the probation service? I think it's a combination of stress on the probation service and the changes we've seen in terms of prison releases as well. So we don't have any evidence that early release makes you more likely to reoffend, but we do know that it

5:20makes you more likely to be recalled to prison. And that's often because you don't have as much time to prepare for your release and that makes it much harder to get things like housing, food bank vouchers, all those kinds of support that help people stay out of prison and get their lives back on track. When we look at serious offenses committed on probation, what are the trends there? There has been quite a jump in recent years in the number of people charged with a serious further offense whilst they were being supervised by probation or immediately after their supervision ended. So that's very serious. Offences including kind of murder, attempted murder, rape, those kinds of things. These are the COVID nightmare scenarios which end up on the front page of the newspapers. Exactly. Exactly what we don't want to happen. But even so, the numbers are still really quite small, but we are still talking about quite a small number of total offenses. So in 2024, 25,

6:26there were just over 800 people who were charged with one of those offenses. And that's up about 50% compared to 2022, 23. So there has been a real increase. And this is actually a case of where it looks like things are getting worse, but pervertly it might be a sign that things are getting better. So the reason behind that big increase is actually an increase specifically in people being charged with and convicted of rape. And that is largely because of really big efforts across the criminal justice system to increase prosecutions and convictions for rape. So whilst they've gone up among people being supervised by probation, they've also gone up across the population as a whole. And basically the two have risen at the same rate. So I think this probably isn't actually a sign that more people on probation are committing these very serious offenses. It's a sign that we're getting better

7:30at catching and prosecuting them when they do. Thanks to Cassia Rowland from the Institute for Government. The Summer The Telegraph published this headline. Number of UK millionaires shrinks to lowest level since financial crisis. High taxes and inflation blamed for rich people leaving Britain. Is it true that the lesser-spotted UK millionaire is fleeing our fair shores due to high taxes and inflation? And is there really someone with a clipboard ticking them off a list as they go? It is certainly true that the super rich do sometimes move, hedge fund billionaire Chris Rokos just moved from the UK to Greece. But is this kind of billionaire relocation what the Telegraph headline is talking about? To find out we spoke to Duncan Weldon, economist, friend of the program, and author of Blood and Treasure. This headline and others like it is based upon estimates produced

8:30by the Adam Smith Institute, a free market think tank who have for some years created a millionaires tracker. They estimate that there are now under 442,000 billionaires in the UK, and that's 7% fewer than their count last year. So is this a figure we can trust? It does sound like quite a small number. The honest truth is that the UK doesn't have any really good statistics on how many millionaires we have. Our tax system is generally based on income rather than wealth, so we've got good data on income because we tax it. This is the thing people often stumble on. If I asked someone to picture a millionaire that probably conjured an image of a very small moustache old and monocled man wearing a top hat and holding large sacks of cash, but that is very much not what the modern millionaire looks like. A lot of millionaires look like Dave down the road, someone whose house, which he bought in 1980 for £5 and a half-eaten packet of biscuits, is now valued at over a million

9:36pounds. I exaggerate, but you get the picture. Most of the UK's millionaires are those whose wealth is tied up in things such as property and pensions. When it comes to wealth, where reliant on things such as the wealth and assets survey from the office for national statistics, which isn't very timely and there's a lot of data quality issues with how it's collected. So that's the first thing to say about the headlines. There is no little black book of millionaires. What the Amesmith Institute have attempted to do with their millionaire tracker is to take two different data sets from the office for national statistics. Put them together. Apply what's called a Pareto distribution to it. The Pareto distribution is simply a curve, in this case, describing the estimated number of people with different amounts of wealth. So instead of literally counting every millionaire, the Adam Smith Institute took estimates of total household wealth, used data from the wealth and asset survey

10:40to try to work out the model of wealth distribution and then used that model to estimate how many people sat above one million pounds. A central driver of the estimate is the total amount of wealth in the UK. So we're trying to go from the total amount of wealth we think is in the UK to how many millionaires they are. They've adjusted that for changes in the price level over time and they've come up with these two numbers that there are 442,000 constant price sterling millionaires in the UK and worked out that this number is 7% lower than it was the year before. As we mentioned earlier, this wealth is not the same as annual income and it's not just cash in the bank but includes shares, pensions and housing wealth. Now something else the headlines have got a bit excited about is the claim that the number of millionaires in the UK is at its lowest level since the financial crisis. Now that sounds like there were more people who had a million pounds

11:45in 2007 and there are in 2025, right? Well that is not what the Adam Smith Institute counted. So they're back project. How many millionaires there would be in every year using 2025 prices? To be at 2025 millionaire, i.e. to have a net worth, that was the equivalent of more million pounds in 2025. If you go back to 2019, that's about 770,000 pounds. What we're saying there is that 770,000 pounds in 2019 is the equivalent of a million pounds in December 2025. And yes, I think that's fair, but it does give us the rather old quirk that in this chart the Adam Smith Institute have of millionaires over time, many of those people who are counted as millionaires now would not actually have had a million pounds at the time.

12:47Basically the Adam Smith Institute aren't trying to estimate the number of millionaires in each year, they're trying to estimate the number of people in each year who have wealth, equivalent to a million pounds today, which would be 770,000 pounds also in 2019, or even less going back to the financial crisis. If they looked at people who only had a million, plus I suspect the line would just be going up all of the time. I think that'd be very few years and which actually it actually falls. Looking at the data in the round, this looks to mainly be an inflation story, an interest rate story, a falling asset price story, rather than an excess of millionaires. As these numbers show, a very high percentage of these millionaires, their wealth is in the form of residential property and pensions, and these people are perhaps not as economically dynamic. I suspect it's towards the lower end, it's people who's, you know, assets have just been pushed below this quite arbitrary threshold by, you know, what's happened with inflation and

13:52national prices and house prices, rather than, you know, the ultra high net worth. So this is not a story about the surmoney bags millionaires of your mind's eye fleeing Britain because of high taxes. Maybe they are, maybe they aren't, but this headline really describes an estimate that is largely driven by the interaction of house prices and the rate of inflation. Sorry Dave, and thanks to Duncan Weldon. You'll listen to more or less. Many listeners will have followed with alarm the extensive coverage about the wildfires in France, Spain, and even the UK this year, but online a graph was circulating that seemed to imply this coverage was misleading. Entitled cumulative area burnt by wildfires by week, Europe, it seemed to show that 2026 was on track to be the quietest year for wildfires in Europe by some distance. There are many graphs circulating online and not all of them are to be believed,

14:56but this one bore the stamp of our world in data. As many loyal listeners will know, it's a nonprofit website that promotes data-driven understanding of some of the world's most important problems and it's a high quality and very respectable source. So what's going on? Are there devastating wildfires happening in Europe? Or are they at record lows? We went to Dr Hannah Richey. She's the deputy editor of our world in data and a senior researcher at the University of Oxford and of course a friend of the programme. I started by asking whether that graph was accurate. It is so on the graph for Europe 2026 at the stage of the year is the lowest on records since 2012. Would you like to, as the academics say, complicate that story for us? I would absolutely love to. So I think the really key caveat here is that in our regional group in for Europe and also in the underlying data sources,

15:57regional grouping for Europe, Russia is in there. Now Russia as part of Europe completely dominates the aggregate figure. It's pretty big, I think, famously. It's very big. And to put in perspective, so Russia's wildfire burn is responsible for around 83% of Europe's total. So basically whether it's high or low, well, far, year in Europe depends on whether it's a high or low year for Russia. And this year Russia is actually a record low since 2012. So if we look at a narrower definition of Europe, for example, the EU, what would the graph show? Yeah, so we also do the European Union, which again, I think when people are thinking about wildfires in Europe, they're thinking about Greece or Spain or Portugal, basically the EU. And on that aggregate, it's around the fifth highest in that time series. It is significantly higher than the median year over this quite short time frame,

16:58but it is kind of middle of the pack. Long question I have is what is a wildfire according to this data set? Yeah, so I think when people think about a wildfire, they are nearly always picturing a forest that's on fire. And actually, I am. That's only part of what we would define as a wildfire burn. So forest is part of it, but also you have the burning of grassland and savanna. And I think what's surprising in this data set or what surprises people is that agricultural burning is also included. So this data set does not differentiate between what you might imagine as a kind of truly wildfire. And actually agricultural land that has been purposely burnt. And again, that's just not what people think about when they think about wildfires. Well, I mean, wildfire, I tend to think of as something accidental, not a deliberate process that is part of the agricultural production system. Yeah, exactly. There is a way to slightly narrow it down. So our main data source is a global wildfire information system. The reason being we're

18:03are world in data. So we try to present the global picture. There is a specific European data set. So this is only looking at the European Union, which is the European forest fire information system, which I think gets closer to the definition of what people would think about when they think about a wildfire. It tries to exclude some of this agricultural burn. And the EFFIS, they are estimates for this year for Europe, has it running at the second highest year on record. And this time, this record goes back to 2006. So it's a slightly longer time series. And it's the second highest on record. So in this data set, excluding agricultural burns, Europe is actually having quite a high year. And how do we reconcile the data with the very alarming news stories from France and from Portugal and Spain earlier this year? Yeah, and I think this is the challenge that people often have, is that we

19:05are primed to focus in on this particular stories or the particular countries that are in the news. And they're in the news because they're having a very, very bad wildfire year. And there are some countries in Europe where it has been genuinely extremely bad. So France is a very good example. Belgium, which usually doesn't get very many wildfires, it has a very, very bad year. At the same time, there are also countries in Europe that have had a really low year. Like Greece will often have very extreme wildfires. This year has actually been quite low for Greece. And I think the challenge there is that it can be true that some countries in Europe have had a genuinely extremely bad wildfire year. And the average overall is kind of offset by this low wildfire burn in other countries as well. And I think the point is that both of these two things can be true at the same time. So I think we've done a good job of zooming in. Can we do the opposite and zoom out? Do we have any data on the global picture for wildfires? Yeah, so that again, the global picture, I think,

20:05will surprise people, which is that globally this year, again, is also tracking to be the lowest well for a burn on record. The caveat there being the record does not that long. It's just like a decade or two. But it is genuinely a low world for our year globally. And the main change here has been changes in land use. The global picture is very much determined by what's happening in Africa. So Africa is home to around 70% of the global burned area. And what we've actually seen over the last few decades is global wildfire burn trending down. So around 25% decline over the last few decades. And researchers mostly put this down to this conversion of savannas and grasslands in Africa to croplands. And it's true that the conditions for wildfire burn higher temperatures, less precipitation, drier conditions, this what we define as kind of fire hazard conditions have

21:07gone up in many regions. That's completely true. And you do genuinely see a trend in some regions where wildfire burn is trending upwards. So I think the key point there is that there are two opposing trajectories. There is this global decline. And that is in some sense masking some of this regional increase, which is moving in the opposite direction. Well thanks to Hannah Richie. She's the author of clearing the air, a hopeful guide to solving climate change. The seaside. What could be more British than strolling along the Esplanade, slightly soggy newspaper in hand failing to protect your fingers from the scolding heat? You find a bench, sit, unwrap and behold. Chips. But oh, oh no.

22:08We're sorry if that was triggering to any listeners. Yes, we're talking about sea goals. And the headline that Britain's lose 171 million pounds worth of food a year to sea goals research shows. In conceivable. To find out what is going on, we got our girl gal Lizzy McNeil on the case. Hi Tim. I hate the monocleuge just gave me. You're welcome. Now tell me about sea goals. Well, they're birds. Of course. Now we associate sea goals with the sea, but they also live in land and places like school playgrounds because they are opportunistic scavengers who will eat pretty much anything, including 171 million pounds worth of precious human food. Allegedly. Want me to unpack that stat? Yeah, like battered haddock from a newspaper. Sure. So the research was conducted by Compare the Market in what they called their great girl index. They asked 2000 UK adults whether they'd ever had food stolen or targeted by a sea goal. Compare the market, say those 2000 people were a nationally representative sample.

23:10Now you don't need to do any statistical trickery to find a nationally representative sample of sea goals because they're all dirty. Lizzy. Sorry. Now I noticed you said food stolen or targeted. Those are not quite the same things if we're looking to put a price on these events or indeed to eat our fish supper. Sharp as ever. Yeah. So they found that 71% of people had said they had food stolen or targeted by a sea goal at some point in their lives, not necessarily actively taken this year. That's a... well I'm not sure I'm surprised by that. I've definitely had sea goals eye up my ice cream at one point or another. They ask scavengers, but I don't think it cost me anything other than mild anxiety. Yes, that's what they're called getting Steven sea gold. As you said, actually getting food stolen is also definitely a thing. The way that compared the market got this eye watering figure was by taking the 71% of respondents who said they'd ever had food stolen or targeted by a sea goal and applying that to the UK adult

24:10population. Right. So 71% of over 55 million UK adults gives you around 39 million people. Yep. They then multiplied that by the average number of times the people who were attacked said they'd had food actually stolen by a sea goal in the last 12 months. This came out at 0.8 per person per year. So we've got 71% of the adult population multiply that by 0.8 thefts per person in the last 12 months and then times that by 5 pound 50, which is the average reported value of food involved. And voila, you get 171,889,518 pounds. And that's fine. It is not fine. So the first issue is that 5 pound 50 is the average reported value of food stolen or targeted, but targeted doesn't mean lost. The calculation implies that the whole value of targeted food was stolen. So the average costs could include a small cotton chips, whereas in reality,

25:11maybe just a couple of chips were stolen or maybe nothing was stolen. People were just being hassled by sea goals while eating. Yeah. So the average they've used for the cost of food probably doesn't reflect what was actually lost. Okay. So second issue. Well, look, the idea that 71% of the UK adult population have something actually stolen from them by a sea goal nearly once a year seems a bit high. That does seem high, yes. Actually very, very high now you come to mention it. Did they share the data? Yes. So for this question, 45% of people that 633 said they had not had food stolen in the last 12 months. 40% said they'd had food stolen once. Right. So nearly half of people didn't have food stolen. So how can the average be nearly once a year? Well, 13% had food stolen twice, 2%, that's 33 people said they'd had food stolen between 3 to 5 times and 6 people said they'd had food stolen over 6 times. Those people could be very unlucky or possibly children.

26:12Maybe unlucky, not children because the respondents of the survey were all over 18. My guess is that this was a quick survey that respondents might not have taken too seriously or put too much thought into it. So just clicked any random button. It could also have been bots as a few answers in an outlier category is just classic bot behavior. Or we could be wrong. Maybe sea goal theft is way more rampant than we think. Do we have any real world data of our own? Eh, I asked the BBC department that we were working whether they'd had food stolen in the last 12 months and just 10% of people had all in seaside locations and none of them had multiple thefts. So that's a lot lower than the 40% of respondents in the Compare the Market survey. And maybe they were all coast adjacent. Were the respondents for your office survey representative of the general population? BBC journalists. Absolutely not. So officially we think the price of food lost is way too high and the number of sea gull attacks is also bafflingly high. Although all that is just a subjective opinion. Thank you Lizzie. That's all we have time for but please send in your

27:17comments and questions as always to more aless at bbc.co.uk. The more aless podcast is still going strong. Search the internet for more aless behind the stats and you'll find us there with both our radio 4 episodes and our shorter Saturday shows. Until next time, goodbye. More aless was presented by me, Tim Harford. The producer was Tom Coles with Nathan Gower Phoebe Keane, Josh McMinn and Lizzie McNeil. The production coordinator is Brenda Brown. The programme was recorded and mixed by James Peard and our editor is Richard Varten. So, Alice Lockstone, I'm here for the history. Well, Ben Henderson, I like the formality and that's perfect because we have a lot of history to share. Why did T become such a British obsession? How did English turn into the language we speak today? And yes, why do women's clothes still not have pockets? Well, in our new podcast here for the history, we're investigating how

28:18stories from the past shape everyday life today. Basically, the things we've all noticed but never stopped to question. Listen on BBC Sounds or watch on YouTube. Just search for here for the history.