Why are some boroughs missing out from London’s education success story?

London’s education success story is by now well known: over the past 15 years, educational attainment in London has improved dramatically, surpassing the rest of the country. However, not all boroughs are sharing in London’s success, with students’ performance resembling the rest of England more closely in some boroughs. This blog looks at two such boroughs – Havering and Hillingdon, and considers to what extent academic research on the ‘London effect’ can explain their performance.

More than four in ten (44 per cent) of 19 years olds in the London borough of Havering lacked level 3 qualifications in 2016, the same proportion as England as a whole. This is the second worst attainment in London and significantly below the London-wide proportion of 35 per cent. Hillingdon, at 39 per cent, performs better but is still in the bottom third for London boroughs.

Proportion of 19 year olds lacking level 3 qualifications

Map of

Sourece: 16 to 19 attaintment statistics, DfE. Data is for 2016.


How do these boroughs fit in with research that has sought to explain the ‘London effect’? We’ll consider the main theories in turn.  The first is that policy responses such as the London Challenge, Teach First and Academies (1) are responsible for the improvement. Research by Greaves et al has discounted these policies as the primary driver of the effect, by demonstrating that the high achieving teens of the mid

noughties had been showing the ‘London effect’ in attainment since primary school, before these policies were implemented (2).

The second theory is ethnic mix: London achieves more because London has more students from ethnic groups with high educational attainment, such as Chinese and Indian students, and lower proportions of White British students who tend to perform comparatively badly (3). This offers some explanation for Havering (83% white British, higher than the England average), but not Hillingdon (52% White British and twice the London average for Indian residents).

A third theory is the performance of disadvantaged students. Blanden et al (4) have looked specifically at the London effect for disadvantaged students. They found that disadvantaged students in London performed far better than disadvantaged students in the rest of England.  They also found a small amount of the ‘London effect’ was the reduction in London in the negative effect of having many disadvantaged students in a school. In other words, non-disadvantaged students had lower educational attainment in schools with a high share of disadvantaged students before 2002, but in London a decade later this was no longer the case. They have suggested that gradually improving school standards (especially in primary school) since the mid-1990s in London are the most likely explanation for this change. 

Students eligible for Free School Meals (FSM) (used a proxy for students in poverty) at the Inner and Outer London level, perform better than FSM students in the rest of England, and the attainment gap between FSM students and other students is smaller in London than across England (5) However, once again the London-wide rates obscure differences at the borough level.  

Havering had the worst attainment level at 19 of any London borough for FSM students in 2016, with 72 per cent not achieving a level 3 qualification, 8 per cent lower than FSM students across England, and 26 percentage points lower than the London-wide rate. 55 per cent of Hillingdon’s FSM students didn’t achieve this qualification at 19, also significantly below the London-wide level. Both boroughs also have attainment gaps between FSM and non-FSM students significantly bigger than the London-wide gap.           

Both boroughs have a median income similar to the London-wide median, and a similar level of income inequality between the poorest fifth and wealthiest fifth as the citywide ratio (6), so these boroughs don’t appear to be generally poorer or more unequal relative to other boroughs. Both boroughs have smaller shares of students eligible for FSM than the London-wide rate. 

These findings seem consistent with Blanden et al’s findings- while boroughs with high levels of deprivation have seen a large improvement, as both disadvantaged and non-disadvantaged students living in deprived areas have gained ground in attainment, Hillingdon and Havering, as not-very-deprived boroughs have not experienced the same gains.  

In 2008, when the cohort who were 19 in 2016 were finishing primary school, both Havering and Hillingdon were in the top half of London boroughs of KS2 attainment rates in English and maths (7)- the attainment gap seems to have opened up between ages 11 and 19. This raises more questions than it answers. Does this indicate that primary schools in Hillingdon and Havering have been improving in line with primary schools across London, meaning that the problem lies with the Borough’s secondary schools? If so, why have the primary schools improved, but not the secondary schools? Most importantly, what is needed to bring these secondary schools’ attainment in line with other London schools?

What impact does the lower educational attainment of students in these boroughs have on their life chances? The immediate consequence seems to be more young people going directly into employment in these boroughs than elsewhere in London, and fewer going on to further education. While 72 per cent of students in Outer London who took A-levels in 2015 were attending some kind of further education in 2016 (either university or further education college), only 67 per cent of former A-levels students did so in Havering and 68 per cent in Hillingdon. Havering also had the joint lowest proportion of A-levels students attending a university of any London Borough- only 44 per cent compared to 58 per cent across Outer London (Hillingdon did much better at 57 per cent).

Young people bypassing further education in these boroughs and going directly into the workforce is likely why these boroughs have higher youth employment rates than London as a whole (although these must be treated with caution at the local authority level). People with lower educational qualifications are more likely to be low-paid and to experience unemployment or underemployment over time(8), so these higher employment rates might mask longer-term negative effects of these boroughs’ low educational attainment.

The term ‘London effect’ suggests that something consistent across the entire city has driven up educational attainment.  Various possible drivers of this story have been discussed. Ethnic mix may be part of the story in Havering (which has an ethnic profile more like England than most of London), but seems unlikely in diverse Hillingdon. Both boroughs have a fairly low rate of low-income students, who are in aggregate behind much of the improvement. Lastly, it is also possible that schools (either primary and/or secondary) in these boroughs have not seen the gradual improvement that other boroughs have.  Rather than one ‘magic ingredient’, it’s likely that a complex combination of factors lie behind London’s success. Taking a closer look at differing educational attainment across boroughs may be the key to identifying how the ‘London effect’ can be brought to boroughs that are so far missing out.



(1) CfBT (2014) Lessons from London Schools: Investigating the Success. CfBT.
(2) Greaves et al (2014) Lessons from London schools for attainment gaps and social mobility, Social Mobility and Child Poverty Commission.
(3) Burgess, Simon (2014) Understanding the Success of London’s Schools, The Centre for Market and Public Organisation. 
(4) Blanden et al (2015) Social Policy in a Cold Climate Working Paper 21: Understanding the improved performance of disadvantaged pupils in London, CASE & LSE.  
(5) Ibid. 
(6) Calculated as ratio between weekly income for full-time employees in each at 20th percentile and 80th percentile in 2016 in each borough. Data is from ASHE via Nomis.     
(7) Analysis of Department of Education Data: ‘Percentage of pupils achieving Level 4 or above in the Key Stage 2 tests by Local Authority (LA), Government Office Region and gender’ 2007.
(8) Tinson et al (2016) Monitoring Poverty and Social Exclusion 2016, NPI.