Statistical literacy is something that I am incredibly passionate about and feel that as a society we should be looking to develop. The release of the interim report of the Milburn Review has inspired me to get back to writing, both to look at the substance of the report itself (to come later), but also to look a bit deeper at a selection of statistics that have been widely shared following the release of the report, and to help make sense of them, and how they aren’t always as they first seem.

Addressing some Misconceptions around the Milburn Review

This week saw the publication of the interim report from the Milburn Review around Young People and Work, investigating the rising youth economic activity and NEET rates. The interim report is dense, with over 600 footnotes and 700 paragraphs of information, and this before any recommendations that are going to come in the final report. A number of takes have already taken hold in the media and media that strips away significant nuance from the report. From labelling young people as workshy, or making arguments about lowering minimum wage, I’ve also seen responses from the education sector of defensiveness and anger, when realistically speaking, the actions of schools in the report come off comparatively well. The problem is that this issue is incredibly nuanced, and is an intersection of various complex phenomena, which doesn’t lend itself to snappy soundbites or 280 character limits.

I am planning on making a second post around the substance of the report itself, but beforehand I want to address some of the comments that have been permeating social and legacy media, either as interpretations of the report, or statistics released alongside it. A number of these are not technically factually incorrect, but without the nuance the way they are interpreted is not always true.

1) The NEET population will cost the UK £125billion

Even before the report released, this headline note was already circulating, receiving a mixture of responses from absolute shock horror to mocking of the calculations. Both of these viewpoints miss what is meant by “cost” in a macroeconomic sense. Only an incredibly minute proportion of this represents real money payments from the taxpayer coffers. In chapter 8 of the report, and the summary tables, a breakdown can be seen.

Only the £3.4 billion of this is direct real money payments. Very different from what most people would interpret from the headline future. Now that’s not to say the £125 billion is a false figure. It involves projections of lost tax income, GDP reduction, lost years of personal wage growth and pension contributions etc (which are economic scarring, the long-term scars that remain for these people and the economy due to this situation). It’s a measure, but it’s not showing the actual direct cash cost.

I am going to use an imperfect football analogy here, it’s clearly not like-for-like, but it’s an easy way of getting your head around some of these ideas. Newcastle United recently sold Anthony Gordon to Barcelona for approximately £69 million. Let’s say, for some legal reason the transfer had to be cancelled. Newcastle therefore had to pay Barcelona £5m due to the legal breakdown. This would be a £5m cash cost, the remaining wages that Newcastle would pay the player would also be a cash cost, there would relative costs to rehab, medical facilities, travel that would be ringfenced for this player. However, the £69million would also be a cost. It’s money that Newcastle missed out on. However, it’s not direct cash they have to pay anyone or anywhere.

So hypothetically;

This is even before thinking about ‘scarring’. Which in this context could be the prevention of Newcastle from finding a new young prospect to sign and develop and ultimately sell at an inflated valuation in a couple of years-time. Not making this sale, delays that potential scouting, signing and development, which means the long-term ‘scarring’ is another £45m from a hypothetical sale to another major team that can be registered as a cost.

As stated, this is not a perfect analogy, and it’s not to say these costs aren’t issues economically, just that the interpretation of them is generally that of a taxpayer bill of transfer to the NEET young people, which is not what it actually represents.

2) The numbers are increasing and the situation is worsening

This is technically true, the numbers are rising, and there are reasons to feel that the current labour market is potentially making things worse. However, this is often being framed through a lens of it being a problem unique to this moment in time. These are long-term systemic issues, and through the report Milburn stresses over and over again, that these are not cyclical, it not this generation. Yes, there are things currently happening that could exacerbate this, but the interventions of the last 20-years, have not been successful long-term. The figure today sits roughly where it was 25-years ago.

In the above graph, there are two-sections of decrease highlighted, either showing the lowest percentage of NEET 16-24 year olds over a particular time period, or the sustained period of longest decrease.

Decrease from the Peak: This time-period represents the implementation of the Education and Skills Act of 2008, which effectively raised the education leaving age from 16  to 18 over the period of time of this decrease. This took 16 to 18 year olds outside of unemployment figures, and additional education time acted as a buffer to reduce these NEET figures, picking up the slack of the labour market – which Milburn references throughout the report. Unfortunately, this buffer is now starting to fail, which I will discuss in more detail when I make a second post looking more at the substance of the review.

The Lowest value: It’s hard to tell due to disruption of data collection during the pandemic, but this floor relates to a time again where the pandemic, online delivery, furlough etc, all led to an artificial ability to greater retain young people in training, ‘employment’/furlough or education. As soon as this finished, this dip was not sustained.

So yes, it is worsening, there is an upward trajectory returning, and this is after the adjustments made that brought it down at an earlier point are still in place. As pointed out by Milburn, there are some unique circumstances in society currently, however, the worsening is too easy to interpret as “it was going well and now isn’t”. Realistically, some things have been done that bring some numbers down, but the actual systemic problems have not been addressed and are ingrained significantly over the 25-year period. It’s that ingraining that makes it worse, this is a chronic issue.

This is what Milburn has tried so hard to emphasise throughout the report, the long-term issues, which is why it is upsetting to see media genuinely reporting under the likes of “Starmer’s snowflake generation: ‘They just abuse welfare!”.

3) There are 27 non-EU under 25’s hired for every young brit since 2020

This is not actually from the Milburn report at all. This is data released by HMRC (UK payrolled employments by nationality, region, industry, age and sex, from July 2014 to December 2025 – GOV.UK), which has been distributed by the centre right think tank, the Centre for Social Justice. In the days since the report, this has been run by the likes of the Express, Daily Mail, which at superficial level makes it look like it all comes from the same source, and has quickly been used to pivot the problem towards immigration, and again frame this is a unique issue for this point in time and not a long-term systemic.

How the figure is shared, you would be right to envisage that if there were 28 jobs, then 27 of them would go to people of non-EU backgrounds. The graph below is from the HMRC data, that shows the number of people in each age range in employment, from UK, EU and non-EU background.

The red bars for under 18 and 18 to 24 are clearly not 27-times smaller than the Non-EU bars. So, where does this figure come from?

The raw data used by the HMRC is not available, but the methodology that CSJ have used is. What the 27-to-1 figure represents is basically this.

The number for non-EU in 2025 minus The number for non-EU in 2020 = 290,000

The number for UK in 2025 minus The number for non-EU in 2020 = 11,000

290000:11000 is approximately 27:1.

We are missing the EU related change in that time, but even still there is no way that the statement as it stands (and has been released by the CSJ and repeated through the media) can be true, as it is worded.

For ease, I provide a hypothetical example involving employment of people between 16-24 of certain names.

At time 1, Jack is a very popular name, Arya, is not. When time 2 comes around, Arya is a much more popular name, whereas Jack holds roughly the same. Due to the increased numbers, more Arya’s get hired in comparison to the first measurement point. Realistically though, Arya’s represent an incredibly small number of all of the available names.

Furthermore, the only way that the conclusion can even begin to be true is if none of the original 26,000 Jack’s, or 200 Arya’s age out of the range in the time-period. This is impossible. An 8-year period will completely refresh the 16-24 age range. The statement as it is produced implies the 3,841,500 16-24 year olds employed in the year 2020, are part of the 3,852,300 in 2025.They are not, they are basing an interpretation on net change as if they were raw number changes.

In his report, Milburn stressed that immigration does not seem to be a significant contributing factor to the NEET issue. Again, this seems to be an attempt to label this as a short-term, unique to now issue, as opposed to a long-term systemic issues.

4) ‘the School system fails to prepare pupils for work’ (the report is negative on schools)

“…schools and teachers are doing often heroic work to optimise the life chances of their students, especially those from disadvantaged backgrounds.”

The above sits within paragraph 336 of the report, and is one of many times schools are discussed in highly positive terms. There is also a significant amount of praise focussed onto the Further Education sector and its disproportionate support for those at risk of becoming NEET, with acknowledge of the significant underfunding and challenges the sector has faced over the last 15 years, including this except from paragraph 371.

“Further education should be the main sector delivering vocational education and preparation for the workplace. It is also the primary recovery layer for those who leave school without strong academic results. Young people with lower prior attainment and from less advantaged socio-economic backgrounds are more likely to enrol in FE colleges at age 16……FE is not a marginal part of the post-16 landscape. For young people at risk of becoming NEET, it is the landscape.”

Again, there are more anecdotes and statistics that are arguably more difficult to swallow, but the positive outweigh the negative of what those involved in the day-to-day of education are doing. Despite this I’ve already seen this report being interpreted as an attack on schools in some quarters, and even some of people crafting defences through the lens of high expectations, knowledge rich curriculum and uniform policies being under attack. That’s not to say how implementation could be made down the line, but that really isn’t what is being suggested in the report. The judgement here is less about the performance of individual schools, but how curriculum, qualifications and accountability have led to emphasis of certain elements, which aren’t always helpful and supportive for the learners at most risk of becoming NEET.

There are survey responses from teachers and feedback in chapter 2 from NEET young people themselves, that all paint a picture of the system letting down the young people most at risk. It is therefore correct that more needs to be done to support these students, however, it’s stressed throughout it’s not just supply-side issue.

People may look to criticise schools for an emphasis on A Levels and academic pathways, but the data shows, that even though this is not as strong of a safeguard as it once was, young people with A Level and Undergraduate qualifications are much more resilient to NEET status. Therefore, for the ‘average student’, that is arguably good blanket advice, when looking at the metrics!

Furthermore, the vocational and technical pathways have been in absolute disarray through continuous reform and threats of defunding. People actually delivering these programmes have been unsure about their future, so how those down the schooling system expected to give reasonable advice around them? The last wave of the T-Levels, a bevy of qualifications first announced in 2017 (and first ran in 2020), still haven’t launched. In that time we’ve also had a brief flirtation with the Advanced British Standard, Alternative Academic Qualifications and now V Levels appearing on the horizon.

Similarly, as I’ve complained before, we’ve had a decoupled set of reforms, where vocational, artistic and technical education was significantly devalued within traditional secondary schooling. This clashed with a greater emphasis being pushed at post-16, with flagship T Level reforms. The problem is then perception. By default, the curriculum that young people experience shapes their viewpoint. They have spent 5-years being told implicitly that these vocational programmes are not worthwhile to study, by the actual curriculum set at the top of the hierarchy, most are not going to turn around and think it’s suddenly a worthwhile thing to do. It doesn’t need to be explicitly said by schools, the implication is there, and it comes from the very top. There was a lack of joined up thinking in this.

There’s then anecdotes of generic careers guidance, of young people not being shown how to apply for apprenticeships, or what is expected in an apprenticeship. In paragraph 354, there’s a note that only 18% of learners received apprenticeship advice. However, the demand side has collapsed. The apprenticeship levy has led to businesses upskilling pre-existing and experienced staff, and entry level apprenticeships have crumbled. I work with young people and I talk to them about apprenticeships, I show them where to find them, where to apply. But a lot of the time, they aren’t there. I have students saying to me “I want an apprenticeship in…” and I ask “Do you know if it actually exists?”. I go through and show these things aren’t always available, they aren’t realistic pathways. Would that be categorised as receiving apprenticeship advice?

This doesn’t mean education should shrug our shoulders and think we have everything right. In paragraph 352, Milburn suggests there’s evidence that institutions that put greater emphasis on the Gatsby Benchmarks for careers guidance have better outcomes from youngsters at risk of becoming NEET. Similarly, in paragraph 361, there’s a note that centres with on-site post-16 provision have better protections against NEET status in their students – which links to an academic paper from earlier this year (rsos.252154.pdf). This means interventions or structures can make a difference.

Ultimately though, education is an area where the state can apply instruments to support with this transition between education and work, which they can’t directly do so with private sector employers. I appreciate the frustrations of schools and colleges of the worry of doing more, education establishments becoming more and more responsible for things that fall outside of the remit of education, but this doesn’t. Part of what Milburn argues for is a more joined up approach to data. We have progression information, destinations of young people, but it’s not directly linked back to institutions for accountability.

The issues around link-up can also be seen in the timing of this report itself. There’s been a great analysis by education journalist, Warwick Mansell of this already. But there significant things here that really needed to be part of the ofsted inspection reform consultation, or needed to be part of the Becky Francis  curriculum review. Again, the systems are not talking to each other! For example, a keen and easy quick fix around accountability, FE institutions and Sixth Form Colleges have had judgements made on meeting regional and national skills needs for half a decade – why look at similar measures within secondary education as a whole? It may not be feasible, but it would be something that forces a level of accountability around employer relations, knowing the local area and economy etc.

This has not been a deep dive into the report itself. I am planning on doing that in the next week or so, primarily focussing on the first four chapters of the report, where Milburn discusses schools, colleges, the mental health of young people and data around the labour market. I need a bit more time to fully gestate my ideas on what has been produced, which kind of proves the problem of consuming the type of dense and nuanced information that sits within this report. People don’t consume at source, they aggregate. Whether it’s mainstream news, social media accounts or industry specific news sources, that is where people will generally consume what is in this report. All of these are required to make sure their reactions and analysis are timely, the news cycle moves in. This means superficial selection, or selection of elements that relate to specific viewpoints, or releasing pre-prepared statements, or in the case of some Education areas, simply stripping out the chapter on education, losing the nuance and cumulative aspects of everything else that sits around it.

I know most people don’t enjoy reading this type of thing and playing around with data. I get that. It is a burden on time, and if I wasn’t enjoying playing in Excel and R whilst writing this, looking at additional data around disadvantage, higher education access, Ofsted, bootcamps, trying to jump deeper into things referenced (and surprisingly not referenced) in the reports for pt.2, I wouldn’t be doing it. It’s what I like doing even when I haven’t written about it (almost 2 years since my last post!).

So stay tuned for part 2 coming soon. The current hope is to get bi-weekly updates going from now through the summer and to the end of the year! Time will tell!