"Governments have mainly responded to this by stoking supply. (For example through Help to Buy.)"
Do you mean "stoking demand"? Stoking *supply* would actually be helpful.
"And then, of course, we have the small boats. The numbers involved are relatively small as a proportion of overall migration"
Not sure what counts as "relatively small", but it's already hit 10k and we're not even halfway through the year, so that's on track to be about a quarter of total migration.
How do you make it a quarter of total migration? Even it it gets to 50k this year (which is totally possible), net migration is forecast to be around 300k and gross immigration considerably higher than that.
I don't disagree that in absolute terms the numbers are non-trivial. But in recent years small boats have accounted for fewer than 5% of total immigration, so they are not driving e.g. the pressure on house prices (though bring other costs).
Would be interested in a deeper dive on out of work benefit numbers. Clearly they are way too high, but the Tories spent a decade plus trying to get them down and Labour don't exactly like the rise either. You allude to criteria for numbers being too wide, but I'd be curious how that happened. It's not like the Tories were eager to widen criteria from 2010 onwards when a lot of the increase happened. And Labour haven't (yet) made many steps in the area.
I'm not certain I have the deep expertise to do that. But the very different way our society now approaches mental health compared to even a decade ago certainly seems to be part of it.
I guess one thing I would say, is that it does seem that unless this is an area where the government is actively bearing down on, the numbers always start to creep up again. Maybe that's because people learn the system, or because it was designed for one set of challenges and then new ones arise, or a combination of both those things, but it does seem to be a trend.
The need to increase renewables/decrease our reliance on imported gas is a defence issue as much as anything. Plus the high cost of energy in the UK has less to do with renewables and more to do with the fact that the price of electricity is tagged to gas.
Yes, which is why investing in renewables is smart but shutting down our own gas supplies (and consistently failing to build nuclear) before our renewables are ready to take the strain is less so.
The 'Heaven knows we're miserable' graph is really eye opening on the perspective driving votes to Reform and Lib Dems!
Appreciate your insight onto the weakness of Reeve's fiscal rules (and revealing that all the commentary is nonetheless from people wanting to loosen them, rather than people wanting to tighten them).
footnote 2: jhope is hoping no one runs spellcheck.
I don't understand footnote 7. Labour have not yet improved public services, but some are trying. Some what? Some public services? Some people in Labour?
I've personally seen a lot more grass roots championing of not building enough homes (people putting signs in their gardens in 3 communities that I sometimes drive through) than I have for university expansion (one segement by a stand up comedian on Radio 4).
Love the line "Despite the protestations of some Anywheres, the laws of supply and demand do apply to labour economics."
I think you're absolutely on the money in the effects of immigration on housing, but the graphs you present for the effects of immigration on GDP look seriously flawed:
The GDP per capita and net migration - 10 year moving average. I realise you didn't put this graph together, just copied it from Neil O'Brien but
1. A 10 year moving average (in GDP? In migration? In both?) is wildly inappropriate for correlation analysis. It is also bizarre to present it quarterly. However there may be an escape - both data sets are much too spikey to be 10 year moving averages, so I think this is quarterly data, and the issue is that Neil O'Brien has no idea what data he's presenting.
2. The primary y-axis is labelled GDP per capita, but is in %. I'm going to assume it's GDP per capita growth (but again, the person who made this graph has heard data is cool, but has no idea what it is).
3. Correlations in time series data are discerable, but perilous. Two unconnected data series which are largely stable apart from one level shift in each will superficially look correlated. Therefore you have to be careful. GDP per capita growth has a level shift in 2007, presumably caused by the financial crash. Net migration goes through a transition phase between 1997 and 2007, caused by some New Labour/Geopolitics mixture. These explain the entire anti-correlation between these lines. In fact once you remove this effect all the micro-shifts are positively correlated (2012, 2020 - note that the line goes below zero but the scale chops off the data. Did I mention that the person who built this graph is a complete ignoramous with data? 2022-3). This graph says GDP/capita growth does modestly positively correlate with immigration, but this is swamped by other effects.
You then present a graph with what the OBR says about the contribution of immigrants, which shows that the average migrant is net positive till age 92 (?!). Now maybe the OBR have completely bungled the data and somehow not counted dependents as also immigrants, but if so you should present a graph of valid data that supports your point. As it is you have a graph that contradicts your point, and then you discuss it as if it agreed with you. (I get that it's the average that contradicts your point, and you're trying to discuss the low earning subset, but the average of the whole is more impactful than the average of a subset!)
Excellent points on dependents and some universities selling immigration instead of education.
I'm unsure whether "moderate out activities" was supposed to be "moderate our activities".
Footnote 7: some Labour Ministers are trying to reform public services - for example, Wes Streeting is doing quite a lot with the NHS. But this takes time to see results that the public will notice.
Homes/Universities: Good point! When I said 'people' I actually meant 'politicians' (a clear category error!).
On the two graphs, yes, agree the first is a bit dubious but I put it in because it's a visual demonstration of what is obviously true: that we had formerly had high GDP growth per capita and low immigration and now the reverse. Possibly I shouldn't have done.
I don't understand your beef with the second:
- I've trusted the OBR on the maths, but it seems plausible that an average wage migrant could have a cumulative net positive fiscal impact to 92 (it's not saying that 92 year olds pay more in tax than they get out that year; it's saying over their life they have).
- The graph literally says we have a whole chunk of immigrants (about 1/3) who have a net negative fiscal impact (even before accounting for pressure on housing, integration issues) - who we could just not take. This is relevant, not the average, because we are not choosing immigrants at random from a pool, we literally control who we let in via salary thresholds, and could get higher wage migrants by the very simple mechanism of raising it.
- In the source document I linked to, as Sol has said, the OBR explicitly says they have not included the impacts of dependents. It seems pretty obvious that dependents (mainly kids) will be fiscally negative (at least in the short term).
Real shocker that the OBR just left out dependents. I haven't read their justification, but it feels like experts deliberately gaslighting the population.
Obviously we would rather use a data source that wasn't so distorted, but I guess bloggers can't be choosers. My objection was:
1. Every real group can be split into subgroups that show opposite things.
2. In the discussion you ignored the average line, picked the low wage line, compared it to the average UK resident (is a low wage UK resident relevant?) and then jumped to "we should only let in the high wage ones". And the average ones who appear to be contributers?
I realise this is all undercut by the OBR distorting the data.
(Also I had missed the cumulative point on the OBR graph, which is my bad. It's literally written on the axis, and the graph shape is way too smooth for individual data)
It's really interesting seeing your reaction to the OBR chart. I guess that, objectively, I agree with you, but at the same time when that chart came out it felt a massive step forward: it was one of the first times that a body like that had explicitly published data that showed that not all migration is economically positive.
On your point 1, I'd push back that it's inevitable the lines (representing the 25th, 50th and 75th percentile of working immigrant salaries) are where they are, or that the 25th would be negative. We have a lot of control over this, in theory, and had the minimum salary thresholds not been £26,200, but something a lot higher, one could easily see all three lines being net positive.
On 2, this is the home vs hotel lens. Britain (just like any other nation) is a home. There are people here who are not net contributors, but it is their home, and we need to support them. But Britain is not a hotel: we have a choice over who we admit, and we can and should choose to admit only those who will positively contribute.
I accept you could make an argument for the average wage ones too; my own view is that the pressure on house prices and integration outweighs the benefit of doing this (plus the dependents issue), but I totally get you could argue this either way.
I agree with most of this, however the literal claim I made in point 1 "Every real group can be split into subgroups that show opposite things." is true.
You're right that we could change the conditions such that *these* subgroups matched, but there would be a further cut (e.g. people who suffered/revealed a debilitating condition shortly after arrival) that again gave conflicting answers.
There are plenty of arguments of the form "immigrants are totally awesome [if you focus on this carefully chosen subgroup]". The correct conclusion is not "immigrants are definitely awesome" but "we should shoot anyone who bases their conclusion about a group on the data of their chosen subgroup, instead of on the data of the whole group." Consider this me shooting you.
I find it surprising that Labour voters are the least negative about the how the direction of the country has been under a decade of Conservative rule.
It may be because Labour is currently in power. There are some amazing charts from the US that show Republican and Democrat perceptions of the economy changing almost literally overnight between presidential administrations.
Sure - but the survey question is "do you think the country has gotten better over the last decade?", not "do you think the country is getting better now?" Can people not read? Maybe we get the politicians we deserve...
Mental health as the primary driver doesn't rule out long covid - if you are tired all the time and there's no very obvious physical cause, which is a common experience of long covid, it's much easier to get a depression diagnosis than anything else.
And much (most?) of long covid looks psychosomatic, making it literally a mental health issue (though one where thinking of it in mental health terms is probably more productive than thinking of it as caused by covid).
Read the room mate. Your very first point is wrong, in my opinion, and therefore you lost me at that point. No, no, no to the new homes. We really don't need them. The solution most people want, reflected in the recent by-elections, is fewer immigrants. We want stricter planning, not laxer planning. We want our countryside green and unspoilt, not open season for spivvy developers to trash the countryside.
I also notice a lack of anything to do with Net Zero and the fake "green" energy agenda - step one in most rural communities would be a total moratorium on onshore wind and solar farms.
Because your first suggestion is diametrically opposed to what Reform voters actually want, it makes it hard to get beyond that and see if any of your other ideas stack up.
Really good and well-argued post.
"Governments have mainly responded to this by stoking supply. (For example through Help to Buy.)"
Do you mean "stoking demand"? Stoking *supply* would actually be helpful.
"And then, of course, we have the small boats. The numbers involved are relatively small as a proportion of overall migration"
Not sure what counts as "relatively small", but it's already hit 10k and we're not even halfway through the year, so that's on track to be about a quarter of total migration.
https://www.standard.co.uk/news/politics/people-smuggling-gangs-migrant-channel-small-boats-crossings-record-b1224931.html
How do you make it a quarter of total migration? Even it it gets to 50k this year (which is totally possible), net migration is forecast to be around 300k and gross immigration considerably higher than that.
I don't disagree that in absolute terms the numbers are non-trivial. But in recent years small boats have accounted for fewer than 5% of total immigration, so they are not driving e.g. the pressure on house prices (though bring other costs).
Uh... because my hasty mental arithmetic lost a zero somewhere. Sorry.
I was worried mine had!
Hope you can edit and then delete this comment, but just before Footnote 12 you say 'stoking supply' when you mean 'stoking demand'.
Thank you! Amended.
Would be interested in a deeper dive on out of work benefit numbers. Clearly they are way too high, but the Tories spent a decade plus trying to get them down and Labour don't exactly like the rise either. You allude to criteria for numbers being too wide, but I'd be curious how that happened. It's not like the Tories were eager to widen criteria from 2010 onwards when a lot of the increase happened. And Labour haven't (yet) made many steps in the area.
I'm not certain I have the deep expertise to do that. But the very different way our society now approaches mental health compared to even a decade ago certainly seems to be part of it.
I guess one thing I would say, is that it does seem that unless this is an area where the government is actively bearing down on, the numbers always start to creep up again. Maybe that's because people learn the system, or because it was designed for one set of challenges and then new ones arise, or a combination of both those things, but it does seem to be a trend.
The need to increase renewables/decrease our reliance on imported gas is a defence issue as much as anything. Plus the high cost of energy in the UK has less to do with renewables and more to do with the fact that the price of electricity is tagged to gas.
Yes, which is why investing in renewables is smart but shutting down our own gas supplies (and consistently failing to build nuclear) before our renewables are ready to take the strain is less so.
The 'Heaven knows we're miserable' graph is really eye opening on the perspective driving votes to Reform and Lib Dems!
Appreciate your insight onto the weakness of Reeve's fiscal rules (and revealing that all the commentary is nonetheless from people wanting to loosen them, rather than people wanting to tighten them).
footnote 2: jhope is hoping no one runs spellcheck.
I don't understand footnote 7. Labour have not yet improved public services, but some are trying. Some what? Some public services? Some people in Labour?
I've personally seen a lot more grass roots championing of not building enough homes (people putting signs in their gardens in 3 communities that I sometimes drive through) than I have for university expansion (one segement by a stand up comedian on Radio 4).
Love the line "Despite the protestations of some Anywheres, the laws of supply and demand do apply to labour economics."
I think you're absolutely on the money in the effects of immigration on housing, but the graphs you present for the effects of immigration on GDP look seriously flawed:
The GDP per capita and net migration - 10 year moving average. I realise you didn't put this graph together, just copied it from Neil O'Brien but
1. A 10 year moving average (in GDP? In migration? In both?) is wildly inappropriate for correlation analysis. It is also bizarre to present it quarterly. However there may be an escape - both data sets are much too spikey to be 10 year moving averages, so I think this is quarterly data, and the issue is that Neil O'Brien has no idea what data he's presenting.
2. The primary y-axis is labelled GDP per capita, but is in %. I'm going to assume it's GDP per capita growth (but again, the person who made this graph has heard data is cool, but has no idea what it is).
3. Correlations in time series data are discerable, but perilous. Two unconnected data series which are largely stable apart from one level shift in each will superficially look correlated. Therefore you have to be careful. GDP per capita growth has a level shift in 2007, presumably caused by the financial crash. Net migration goes through a transition phase between 1997 and 2007, caused by some New Labour/Geopolitics mixture. These explain the entire anti-correlation between these lines. In fact once you remove this effect all the micro-shifts are positively correlated (2012, 2020 - note that the line goes below zero but the scale chops off the data. Did I mention that the person who built this graph is a complete ignoramous with data? 2022-3). This graph says GDP/capita growth does modestly positively correlate with immigration, but this is swamped by other effects.
You then present a graph with what the OBR says about the contribution of immigrants, which shows that the average migrant is net positive till age 92 (?!). Now maybe the OBR have completely bungled the data and somehow not counted dependents as also immigrants, but if so you should present a graph of valid data that supports your point. As it is you have a graph that contradicts your point, and then you discuss it as if it agreed with you. (I get that it's the average that contradicts your point, and you're trying to discuss the low earning subset, but the average of the whole is more impactful than the average of a subset!)
Excellent points on dependents and some universities selling immigration instead of education.
I'm unsure whether "moderate out activities" was supposed to be "moderate our activities".
The OBR deliberately did not include dependents as per this link: https://obr.uk/docs/dlm_uploads/FRS-migration-supplementary-forecast-information-release-Mar-2025.pdf
They also modelled spending and public services use as the same as the average UK resident.
Footnote 7: some Labour Ministers are trying to reform public services - for example, Wes Streeting is doing quite a lot with the NHS. But this takes time to see results that the public will notice.
Homes/Universities: Good point! When I said 'people' I actually meant 'politicians' (a clear category error!).
On the two graphs, yes, agree the first is a bit dubious but I put it in because it's a visual demonstration of what is obviously true: that we had formerly had high GDP growth per capita and low immigration and now the reverse. Possibly I shouldn't have done.
I don't understand your beef with the second:
- I've trusted the OBR on the maths, but it seems plausible that an average wage migrant could have a cumulative net positive fiscal impact to 92 (it's not saying that 92 year olds pay more in tax than they get out that year; it's saying over their life they have).
- The graph literally says we have a whole chunk of immigrants (about 1/3) who have a net negative fiscal impact (even before accounting for pressure on housing, integration issues) - who we could just not take. This is relevant, not the average, because we are not choosing immigrants at random from a pool, we literally control who we let in via salary thresholds, and could get higher wage migrants by the very simple mechanism of raising it.
- In the source document I linked to, as Sol has said, the OBR explicitly says they have not included the impacts of dependents. It seems pretty obvious that dependents (mainly kids) will be fiscally negative (at least in the short term).
Real shocker that the OBR just left out dependents. I haven't read their justification, but it feels like experts deliberately gaslighting the population.
Obviously we would rather use a data source that wasn't so distorted, but I guess bloggers can't be choosers. My objection was:
1. Every real group can be split into subgroups that show opposite things.
2. In the discussion you ignored the average line, picked the low wage line, compared it to the average UK resident (is a low wage UK resident relevant?) and then jumped to "we should only let in the high wage ones". And the average ones who appear to be contributers?
I realise this is all undercut by the OBR distorting the data.
(Also I had missed the cumulative point on the OBR graph, which is my bad. It's literally written on the axis, and the graph shape is way too smooth for individual data)
It's really interesting seeing your reaction to the OBR chart. I guess that, objectively, I agree with you, but at the same time when that chart came out it felt a massive step forward: it was one of the first times that a body like that had explicitly published data that showed that not all migration is economically positive.
On your point 1, I'd push back that it's inevitable the lines (representing the 25th, 50th and 75th percentile of working immigrant salaries) are where they are, or that the 25th would be negative. We have a lot of control over this, in theory, and had the minimum salary thresholds not been £26,200, but something a lot higher, one could easily see all three lines being net positive.
On 2, this is the home vs hotel lens. Britain (just like any other nation) is a home. There are people here who are not net contributors, but it is their home, and we need to support them. But Britain is not a hotel: we have a choice over who we admit, and we can and should choose to admit only those who will positively contribute.
I accept you could make an argument for the average wage ones too; my own view is that the pressure on house prices and integration outweighs the benefit of doing this (plus the dependents issue), but I totally get you could argue this either way.
I agree with most of this, however the literal claim I made in point 1 "Every real group can be split into subgroups that show opposite things." is true.
You're right that we could change the conditions such that *these* subgroups matched, but there would be a further cut (e.g. people who suffered/revealed a debilitating condition shortly after arrival) that again gave conflicting answers.
There are plenty of arguments of the form "immigrants are totally awesome [if you focus on this carefully chosen subgroup]". The correct conclusion is not "immigrants are definitely awesome" but "we should shoot anyone who bases their conclusion about a group on the data of their chosen subgroup, instead of on the data of the whole group." Consider this me shooting you.
Definitely prefer several shorter posts over single longer one 🙂
I find it surprising that Labour voters are the least negative about the how the direction of the country has been under a decade of Conservative rule.
It may be because Labour is currently in power. There are some amazing charts from the US that show Republican and Democrat perceptions of the economy changing almost literally overnight between presidential administrations.
Sure - but the survey question is "do you think the country has gotten better over the last decade?", not "do you think the country is getting better now?" Can people not read? Maybe we get the politicians we deserve...
Mental health as the primary driver doesn't rule out long covid - if you are tired all the time and there's no very obvious physical cause, which is a common experience of long covid, it's much easier to get a depression diagnosis than anything else.
Fair point.
And much (most?) of long covid looks psychosomatic, making it literally a mental health issue (though one where thinking of it in mental health terms is probably more productive than thinking of it as caused by covid).
Read the room mate. Your very first point is wrong, in my opinion, and therefore you lost me at that point. No, no, no to the new homes. We really don't need them. The solution most people want, reflected in the recent by-elections, is fewer immigrants. We want stricter planning, not laxer planning. We want our countryside green and unspoilt, not open season for spivvy developers to trash the countryside.
I also notice a lack of anything to do with Net Zero and the fake "green" energy agenda - step one in most rural communities would be a total moratorium on onshore wind and solar farms.
Because your first suggestion is diametrically opposed to what Reform voters actually want, it makes it hard to get beyond that and see if any of your other ideas stack up.
I’m told that the water situation has even more of an impact than energy costs. Regulation of the energy market (especially basing on spot price of gas) has been sub optimal and led to high prices but the system does at least work. On water people feel that catastrophic failure has been rewarded https://dieterhelm.co.uk/publications/digging-an-ever-deeper-hole-the-next-chapter-in-the-thames-water-saga/