{"id":243995,"date":"2026-03-31T16:55:58","date_gmt":"2026-03-31T15:55:58","guid":{"rendered":"https:\/\/schoolsweek.co.uk\/?p=243995"},"modified":"2026-04-08T22:42:19","modified_gmt":"2026-04-08T21:42:19","slug":"earning-the-right-to-deprivation-funding","status":"publish","type":"post","link":"https:\/\/schoolsweek.co.uk\/earning-the-right-to-deprivation-funding\/","title":{"rendered":"Earning the right to deprivation funding"},"content":{"rendered":"\r\n

For years, the blunt measure of free schools meals eligibility has been used to allocate schools\u2019 deprivation funding. <\/strong><\/p>\r\n\r\n\r\n\r\n

But an alternative is being developed based on parental income that could achieve ministerial aims of more effectively spending money where it\u2019s needed.<\/a><\/strong><\/p>\r\n\r\n\r\n

\r\n\r\n\r\n

A government dataset matching parental income with children\u2019s education records could prove the key to ministers\u2019 plans to shake up how deprivation funding is allocated. The trouble is, it\u2019s not ready yet, an expert working on the dataset has warned.<\/p>\r\n\r\n\r\n\r\n

Last month\u2019s schools white paper<\/a> confirmed the Department for Education\u2019s plans to use family income data rather than free school meals eligibility to distribute school deprivation funding.<\/p>\r\n\r\n\r\n\r\n

Instead of the \u201cbinary\u201d FSM measure, the document suggested that income data could enable a \u201cstepped\u201d model to reflect different levels of disadvantage in school cohorts.<\/p>\r\n\r\n\r\n\r\n

\u201cThe model could take into account how low family income is, and for how long this has been the case,\u201d the white paper said.<\/p>\r\n\r\n\r\n\r\n

\"Carl\r\n
Carl Cullinane<\/figcaption>\r\n<\/figure>\r\n\r\n\r\n\r\n

According to Carl Cullinane, director of research and policy at education charity the Sutton Trust, \u201cthere are much greater levels of educational disadvantage in those who’ve been eligible [for FSM] for longer.<\/p>\r\n\r\n\r\n\r\n

\u201cSo I think that they’re moving in the direction of recognising the impact of greater poverty.\u201d<\/p>\r\n\r\n\r\n\r\n

But it remains unclear what precise measure the government wants to move towards, or what distributional effect it might have.<\/p>\r\n\r\n\r\n\r\n

Parent pupil matched data<\/h3>\r\n\r\n\r\n\r\n

Jon Andrews, interim chief executive of the Education Policy Institute (EPI) think tank, believes the government may use parent pupil matched data (PPMD), which shows income for households in England.\u00a0<\/p>\r\n\r\n\r\n\r\n

\u201cThis links parental income with individual pupils in schools,\u201d he told Schools Week.<\/p>\r\n\r\n\r\n\r\n

\u201cSo that’s income through earnings and through benefits. And that’s been in the Department [for Education] probably the best part of a decade now.\u201d<\/p>\r\n\r\n\r\n\r\n

The Cabinet Office and DfE commissioned two Durham University professors, Stephen Gorard and Nadia Siddiqui, to study the PPMD dataset, linked to the national pupil database.<\/p>\r\n\r\n\r\n\r\n

They were asked to look at the impact of the pupil premium on attainment, and to see how easy it was to use the dataset. Their first research report was published last year.<\/p>\r\n\r\n\r\n\r\n

\u201cWe were offered this dataset a few years ago now, maybe three years ago, and this was the first time it had been assembled and made available,\u201d says Professor Gorard.<\/p>\r\n\r\n\r\n\r\n

He thinks the government will switch from FSM to either the PPMD data or \u201csomething very similar\u201d for deprivation funding.<\/p>\r\n\r\n\r\n\r\n

Trying to correlate household income data to individual academic attainment proved difficult as there was so much volatility in the dataset. This included low-income children who do well and rich pupils who struggle.<\/p>\r\n\r\n\r\n\r\n

But when the researchers grouped the data into 20 income bands, a close relationship with attainment levels emerged.<\/p>\r\n\r\n\r\n\r\n

\u201cWe can make a really good estimate of how well a child would do at school just from knowing, on average, their income band\u201d, Gorard said.<\/p>\r\n\r\n\r\n\r\n

He and Siddiqui have now made the income data more granular by splitting it into 100 bands, and while the research won\u2019t be published for months, he says \u201cthe patterns still hold\u201d.<\/p>\r\n\r\n\r\n\r\n

But with the government planning to consult on its plans this summer, he warns the PPMD dataset isn\u2019t ready.<\/p>\r\n\r\n\r\n\r\n

\u201cWe’re really excited to have the data and to work on it. But the datasets we’ve seen I think are not ready, because there’s so many cases we don’t know about, we don’t know their family structure or the income of other adults.\u201d<\/p>\r\n\r\n\r\n\r\n

Data weaknesses<\/h3>\r\n\r\n\r\n\r\n

The dataset as it stands has several weaknesses.<\/p>\r\n\r\n\r\n\r\n

Data on families who have registered for benefits is thorough \u2013 \u201cwe tend to know if there’s another adult [in the household], and if they’re earning and how much they’re earning\u201d \u2013 but weaker for the rest of the population.<\/p>\r\n\r\n\r\n\r\n

Gorard said that if each age cohort of pupils in the dataset has around 640,000 children, there might be around 200,000 where full information only exists for one adult. In those cases, it\u2019s not clear if there\u2019s a second adult in the household or what their income is.<\/p>\r\n\r\n\r\n\r\n

While most low-income households probably claim benefits \u2013 meaning their full data is in the PPMD dataset \u2013 others may choose not to, or might fall just over the income thresholds. This means the PPMD figures don\u2019t fully cover the extent of income deprivation.<\/p>\r\n\r\n\r\n\r\n

\u201cWhen we started saying, \u2018there’s problems, there’s missing data and there’s unlinked families\u2019, they said, \u2018we’ll produce a better version\u2019,\u201d Gorard says.<\/p>\r\n\r\n\r\n\r\n

So far no better version has been forthcoming.<\/p>\r\n\r\n\r\n\r\n

\u201cThey said we could be updated if there was a better dataset. So either they’ve decided they’re not going to give it to us or, more likely, there isn’t yet a better dataset.\u201d<\/p>\r\n\r\n\r\n\r\n

Gorard adds that the better dataset, if and when it arrives, will likely use the same source data, but with the weaknesses ironed out as far as possible.<\/p>\r\n\r\n\r\n\r\n

London\u2019s housing costs<\/h3>\r\n\r\n\r\n\r\n

One potentially important feature of the PPMD dataset is that it takes no account of housing costs. Analysis of government data by Schools Week shows the potential impact of excluding housing costs from deprivation funding calculations.<\/p>\r\n\r\n\r\n\r\n

If deprivation funding was switched to the number of children in relative low income households \u2013 measured by the Department for Work and Pensions without accounting for housing costs \u2013 councils in London would lose out.<\/p>\r\n\r\n\r\n\r\n

For example, Kensington and Chelsea\u2019s share of the England-wide total of children in relative low income is half its share of the children drawing the FSM-driven pupil premium.<\/p>\r\n\r\n\r\n\r\n

In Islington, Lambeth and Southwark, among others, the share would fall by a third or more.<\/p>\r\n\r\n\r\n\r\n

But if deprivation funding was switched from FSM to the income deprivation affecting children index (IDACI), which was changed last year to reflect housing costs and is already used to allocate some deprivation funding in the national funding formula, the opposite happens.<\/p>\r\n\r\n\r\n\r\n

\r\n


Any measure must account for income after housing costs<\/p>\r\n<\/blockquote>\r\n\r\n\r\n\r\n

London councils\u2019 share of children living in the poorest areas under the IDACI measure rises sharply compared to their share of the England-wide pupil premium population. In Brent and Harrow it more than doubles.<\/p>\r\n\r\n\r\n\r\n

Notably, the white paper said the government was \u201calso considering whether to target funding based on the place a child lives, as well as their individual family economic circumstances\u201d. That could open the door for more use of the place-based IDACI in future.<\/p>\r\n\r\n\r\n\r\n

In both scenarios analysed, county councils lost out, particularly in southern England and the home counties, whereas more deprived urban councils in the north and Midlands gained.<\/p>\r\n\r\n\r\n\r\n

Cullinane argued a balanced approach is needed. Ignoring housing costs means overlooking the actual disposable income available to families. But \u201cwhile there’s lots of poverty in London, the pupils in London are affected less by that poverty when it comes to their educational outcomes\u201d.<\/p>\r\n\r\n\r\n\r\n

Ian Edwards, executive member for children and young people at representative body London Councils, says: \u201cLondon has a high concentration of deprived children and families \u2013 and with housing costs so high, any measure of deprivation must account for income after housing costs.\u201d<\/p>\r\n\r\n\r\n\r\n

FSM6 is \u2018crazy\u2019<\/h3>\r\n\r\n\r\n\r\n

The Durham researchers have conducted a regional analysis of the PPMD data. It hasn\u2019t yet been published, but unsurprisingly it shows average incomes in Greater London are higher than in the north and parts of the south west.<\/p>\r\n\r\n\r\n\r\n

\u201cIf the secretary of state wants to move money from London to (say) the north, then putting more money through low prior attainment would seem sensible,\u201d says Dr Tim Leunig, who helped devise the original national funding formula.<\/p>\r\n\r\n\r\n\r\n

The current approach of using the FSM6 measure \u2013 which includes children who were eligible for FSM at any point in the last six years \u2013 to allocate deprivation funding is \u201ccrazy\u201d, says Gorard<\/p>\r\n\r\n\r\n\r\n

At present, for example, Middlesbrough and Guildford proportionally receive similar levels of pupil premium.<\/p>\r\n\r\n\r\n\r\n

Middlesbrough has far more long-term FSM children, but Guildford has more children who only qualified for it briefly.<\/p>\r\n\r\n\r\n\r\n

A switch to income measures would allow the government to measure the depth of poverty, says Gorard.<\/p>\r\n\r\n\r\n\r\n

PPMD data could also improve measurement of duration of deprivation when children start school. They won\u2019t have built up years of FSM eligibility, but years of household income data should exist.<\/p>\r\n\r\n\r\n\r\n

Measuring both the depth and duration of deprivation could help better allocate different levels of deprivation funding.<\/p>\r\n\r\n\r\n\r\n

\u201cIt looks as though the pupil premium has lifted the attainment more easily of the people who have some earned income but are still below the [FSM] threshold,\u201d Gorard adds.<\/p>\r\n\r\n\r\n\r\n

\u201cBut the group who are earning nothing, have no earned income, are much harder to shift. In terms of attainment, I think that would require much more investment.\u201d<\/p>\r\n\r\n\r\n\r\n

A challenge will be how to transition to a new system. Funding floors and minimum funding levels in the national funding formula should protect schools from rapid funding cuts, but these protections don\u2019t exist in the pupil premium system.<\/p>\r\n\r\n\r\n\r\n

\u201cDfE will want to think about whether they phase something in,\u201d says Andrews. \u201cYou should probably think about protecting schools against big shocks from one year to the next.\u201d<\/p>\r\n","protected":false},"excerpt":{"rendered":"

How might ministers better target deprivation funding based on family income? <\/p>\n","protected":false},"author":315324,"featured_media":243813,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"om_disable_all_campaigns":false,"footnotes":""},"categories":[5],"tags":[324],"post-theme":[2425],"class_list":["post-243995","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-profiles","tag-funding","post-theme-school-funding"],"acf":[],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t