Policy paper

Chapter 3: Intermediate outcomes

Published 12 September 2023

Highlights

Densely populated urban areas have higher levels of economic insecurity for young people – unemployment, economic inactivity, and lower working-class jobs.

London and adjoining areas have more promising prospects for young people – higher qualifications, earnings, and occupational level. This means that London in particular has a high concentration of young people at both extremes.

Some areas, most notably South Yorkshire and Eastern Scotland, score poorly on both economic insecurity and employment prospects. There are high levels of the former and low levels of the latter.

People of some ethnic backgrounds, such as Chinese and Indian, have much better educational outcomes than others (intermediate outcomes 1 and 2). There is also evidence that socio-economic background (SEB) has less impact on young people from these groups.^{[footnote 1]}

All ethnic minorities apart from Black Caribbean are more likely to gain a degree than White British young people from the same SEB, although their degrees may come from less selective universities (intermediate outcome 2.3).

Yet these better educational outcomes don’t always yield better occupational outcomes. Several ethnic minority groups (Black Caribbean, Black African, Mixed, Pakistani and Indian) are more likely to be unemployed than White British young people from the same SEB (intermediate outcome 3.2).

Similarly, Pakistani, Bangladeshi and Black African ethnic groups have higher proportions of university graduates (intermediate outcome 2.3) than the White British group, but not higher proportions in the professional classes (intermediate outcome 3.3).

There are interactions between SEB and sex: the male-female gap in economic activity (intermediate outcome 3.1) among people aged 25 to 29 years is only 4 percentage points among those from a higher professional background, but almost 4 times larger, at 15 percentage points, for those of a lower working-class background.

People with a disability do significantly worse in every intermediate outcome. In some cases, the gap is even wider among those from a lower working-class background, suggesting that professional families are better able to mitigate the effects of disability on young people’s life chances.

Introduction

Intermediate outcomes compare people’s starting point with an endpoint in their teens, 20s, or early 30s, as they move through education and into the labour market. The skills, qualifications and experience of work that young people acquire will affect their social mobility. So we examine these earlier outcomes for people from a range of SEBs.

We call these ‘intermediate outcomes’ both because they are measured earlier in life than the mobility outcomes in chapter 2, and because they give an early sense of what the mobility outcomes might be later in life. We report on them annually, since the experiences of each cohort of people leaving school and entering the labour market may change from year to year – think of the effects of the pandemic, for example.

In this section, we examine the following measures:

• years of compulsory schooling – 5 to 16 years

• post-16 qualifications and progression into the workplace – 16 to 29 years
• work in early adulthood – 25 to 29 years
• career progression – 35 to 44 years

Why do we break the outcomes down by background?

For any analysis of social mobility, we need to know where a person starts (their background) and where they end up (their outcome). For example, to measure Angela’s occupational-class mobility, we need to know her parents’ occupational class (Angela’s background), and her own occupational class (Angelas’s outcome). This way, we can see whether Angela has moved up or down.

If we want our measure to describe the whole population, we need to be able to summarise everyone’s backgrounds, and outcomes, in a small set of numbers. And to begin to understand the effect of social background on outcomes – the essence of social mobility analysis – we need to look at the outcomes of everyone sharing a certain background.

This year we use a new, revised 5-class measure of SEB, in place of the 3-class measure used in the 2022 report. We divide SEB into the following 5 parts: higher professional, lower professional, intermediate, higher working class, and lower working class.

Indicators for this year

We use a number of indicators to predict later social mobility outcomes.

We begin with the years of compulsory education. These years are critical, because socio-economic differences in children’s skills are present even before they start school, and can increase throughout their development.^{[footnote 2]} There isn’t a particular age to pinpoint when these disparities emerge, so we consider all stages of childhood.^{[footnote 3]}

Post-compulsory schooling years are also important for making progress in the labour market and social mobility, so we look at routes to work, early career progression, and work in early adulthood. This period starts from when young people leave education and move into apprenticeships, work, training, employment, or economic activity.^{[footnote 4]}

For the indicators of career progression (intermediate outcome 4), we use a different methodology and data sources from the previous report. This new measure is more accessible to readers.

We have also changed our indicator relating to the class pay gap. Last year, we didn’t take level of education into account. That means the effects of both SEB and education level were included in the indicator, with no way to separate them. This year, we have added replacement measures of 2 types. Firstly, we report the link between education and earnings, for people of the same SEB. Secondly, we report the link between SEB and earnings, for people of the same level of education.

Geographical breakdowns

In a major improvement to last year’s annual report, we are also providing geographical breakdowns of intermediate outcomes. We can do this because of the large sample size of the Labour Force Survey (LFS). Although there is considerable imprecision in each individual indicator when broken down in this way, we can get more reliable results by combining related indicators together (see figure 3.0).

Unfortunately, we still have very limited ability to compare education-related intermediate outcomes across the UK. This is mainly because there is no consistent measure of SEB in educational administrative data. Worse, the measure that does exist, eligibility for free school meals (FSM), is not well suited to comparing different regions, because the characteristics of the non-FSM group will be very different across different regions (for example, pupils not on FSM in London may have different educational experience than those in rural Scotland). This remains a major data gap.

Where people live versus where they grew up

When we give estimates for intermediate and mobility outcomes broken down by region, we are referring to people’s region of origin, not where they currently live. This is sometimes referred to as ‘adolescent geography’ (where someone lived while growing up), in contrast to ‘current geography’ (where they live now). For example, we see in figure 3.0 that young people from London are more likely than average to be unemployed. This means young people who were living in London aged 14 years, no matter where in the UK they live now.

Figure 3.0: Examples of adolescent geography.

Angela grew up in west central Scotland and moved to Northern Ireland as an adult. She then became a lawyer. She counts in the statistics for higher-professional employment in west central Scotland, because she:

• is currently in a higher-professional job
• grew up in west central Scotland

She will appear in region 39 on the map as higher professional.

Bruno grew up in south London and moved to Kent as an adult. He happened to be unemployed at the point he responded to the Labour Force Survey. He counts in the unemployment statistics for south London, because he:

• is currently unemployed
• grew up in south London

He will appear in region 3 on the map as unemployed.

Breakdowns by protected characteristics

Just as the large sample size of the LFS allows us to do geographical breakdowns, it also allows us to break mobility levels down by protected characteristics, such as sex or ethnic background.^{[footnote 5]} This can reveal important new insights.

Not all breakdowns are possible. Sometimes, protected characteristics are not included in a dataset, so we can’t use those characteristics.^{[footnote 6]} In other cases, especially where we are not using the LFS, there is a relatively small sample size, and breaking this down by various characteristics reduces the sample size further. This means that the resulting estimates are not reliable enough to publish.

However, we should treat these breakdowns with caution. One reason for this is that ethnicity and region are correlated. For example, people of certain ethnic minorities may be more likely to live in certain regions such as London than others. As London may on average perform better on some outcomes than other regions, this may lead to some ethnic minority groups (those who are disproportionately more likely to live in London), to also perform better on such outcomes. Alternatively, the presence of ethnic minority groups in London may cause London to perform better (in educational outcomes, for example).

Geographical analysis: social mobility across the UK

Summary

By combining related measures we can get a more reliable picture of geographical mobility patterns.

The data reveals 2 sets of related intermediate outcomes, measured when people are in their 20s. Firstly, ‘promising prospects’, which are made up of attainment of degrees, professional occupation, and high hourly earnings. Secondly, ‘precarious situations’, made up of economic inactivity rates, unemployment, and lower working-class occupations.

Young people brought up in Greater London and some adjoining areas are doing well on promising prospects, while those brought up in more rural or more remote areas, and some former industrial areas, are doing less well.

In contrast, young people brought up in densely populated urban areas are more likely to be in precarious situations than those brought up in more rural areas. London has both high levels of promising prospects and high rates of young people in difficult economic circumstances.

A new approach to monitoring social mobility by region

Looking at single indicators of social mobility, like unemployment or highest qualification, could be misleading. This is because results have to be estimated from sample surveys, and sample sizes at a regional level can be small (as described in figure 3.0). To deal with this problem, we have constructed 2 summary measures (‘composite indices’) relating to intermediate outcomes. Each composite index summarises results from 3 indicators, giving a more reliable picture.

Table 3.1. Summary of composite indices for the intermediate outcomes.

Index	Indicator	LFS data used
Promising prospects	IN2.3 Highest qualification (university degree)	Net levels of a university degree among young people in each area after controlling for socio-economic background (SEB)
	IN3.3a Occupational level (professional occupation)	Net proportions of young people in higher professional-class jobs in each area after controlling for SEB
	IN3.4 Hourly earnings	Mean hourly earnings among young people in each area after controlling for SEB
Precarious situations	IN3.1 Economic inactivity	Net levels of inactivity among young people in each area after controlling for SEB
	IN3.2 Unemployment	Net levels of unemployment among young people in each area after controlling for SEB
	IN3.3b Occupational level (lower working-class occupation)	Net proportions of young people in lower working-class jobs in each area after controlling for SEB

Promising prospects

This index brings together 3 measures capturing promising prospects for young people, as measured by their levels of education, occupational positions and earnings. The index adjusts for SEB and so measures how well young people from similar backgrounds do in education and the labour market.

Figure 3.2: London and some adjoining areas stand out as offering promising prospects to young people.

Index of promising prospects.

Explore and download data on the index of promising prospects on the State of the Nation data explorer.

Source: Labour Force Survey pooled, from 2018 to 2021. Source data used from the following indicators: intermediate outcomes 2.3, 3.3a and 3.4.

Note: Areas are where respondents lived when they were aged 14 years. We follow the procedure used by Anand and Sen (1994) for constructing the UN’s Human Development Index. To ensure that all indicators are on a common metric, indicators are first rescaled, setting the best performing area’s score on the indicator to 1 and the least well-performing area’s score to 0. For more information on how each area was scored, please see the technical annex.

Greater London and some adjoining areas stand out as ones where young people do particularly well. At the other extreme, young people brought up in more rural or remote areas, together with those from some former industrial areas, tend to do less well. There is, however, likely to be a considerable variety of prospects within these broad areas, particularly in geographically larger areas such as North Yorkshire which contains both remote rural areas and thriving urban centres.

We must emphasise that these are descriptive results and do not necessarily establish that areas have a causal effect on young people’s outcomes. We should also emphasise that most areas are fairly similar, especially those in the middle of the distribution (with paler colouring in the map in figure 3.2).

Precarious situations

This index brings together 3 measures capturing young people in difficult economic circumstances: economic inactivity, unemployment, and lower working-class employment. The index adjusts for SEB and so measures how precarious are the situations of young people from similar social backgrounds.

Figure 3.3 shows that densely populated urban areas tend to do worse on this composite measure when compared with more rural areas. In particular, parts of London which may generally be considered to be particularly productive also have high rates of young people in difficult economic circumstances.

Figure 3.3: As with childhood poverty and disadvantage, metropolitan areas are characterised by higher levels of precarious situations.^{[footnote 7]}

Index of precarious situations.

Explore and download data on the index of precarious situations on the State of the Nation data explorer.

Source: Labour Force Survey pooled, from 2018 to 2021. Source data used from the following indicators: intermediate outcomes 3.1, 3.2 and 3.3b.

Note: Areas are where respondents lived when they were aged 14 years. We follow the procedure used by Anand and Sen (1994) for constructing the UN’s Human Development Index. To ensure that all indicators are on a common metric, indicators are first rescaled, setting the best performing area’s score on the indicator to 1 and the least well-performing area’s score to 0. For more information on how each area was scored, please see the technical annex.

Overall, the 2 composite indices suggest there may be a lot of polarisation and inequality in some areas. For example, there is a relatively high proportion of young people in parts of Greater London who are in precarious situations, which contrasts with the high proportions in the same areas with promising prospects. In other words, both extremes coexist in London.

However, there are also areas where young people do worse on both indices. For example, Eastern Scotland ranks low for both promising prospects and precarious situations, which may suggest that young people may find themselves with relatively fewer opportunities than those from similar social backgrounds in other regions in the UK.

Breakdowns by protected characteristics

Summary

Intersectional analysis shows consistent gender or sex differences, disability gaps, and ethnicity gaps among people from similar SEBs, in a range of intermediate outcomes.^{[footnote 8]}

Disability is the only protected characteristic where we find disadvantage across all outcomes.

Among ethnic groups, we find considerable diversity. Some groups, such as Chinese and Indian, are more likely than their peers to obtain university degrees, while other groups, such as Black Caribbean and Black African, are much more likely than their peers to be unemployed.

The Pakistani, Bangladeshi and Black African groups have higher proportions of university graduates than their White British peers, but do not have higher proportions in the professional classes.

There are signs of a complex interplay between SEB, ethnic group and economic outcomes. SEB plays a much smaller role among some minority groups, such as the Chinese, Bangladeshi and Pakistani groups, than among White people.

In the case of gender or sex, we find gaps in favour of young women in education but gaps in favour of young men in earnings.^{[footnote 9]}

Viewing through an intersectional lens, we see that gaps between sexes and disability status are often larger among people from lower working-class backgrounds.

Intersectional analysis by sex and socio-economic background

Education

SEB is positively related to young people continuing in full-time education and training (a difference of 15 points between young men from higher professional and lower working-class backgrounds). It is negatively related to those not in education, employment or training (known as ‘NEET’). For example, if someone is NEET they are likely to be from a lower SEB. In comparison with SEB, sex differences are quite modest but young women (aged 16 to 24 years) tend to be slightly more likely to be in education and training than young men, and the largest sex differences (reaching 4 percentage points) are among those from professional backgrounds. Conversely, young men are more likely to be NEET.

This picture is similar when we move on to an intersectional analysis of young people’s highest level of education aged 25 to 29 years, although the SEB gaps are much larger when we compare the full range of qualification levels. Here we find a very strong positive relationship between SEB and the highest level of qualification (a gap of 43 percentage points in those getting a university degree when comparing men from higher professional and lower working-class backgrounds respectively). There is a modest sex difference, with young women from all SEBs having higher qualification levels than their male peers (by 5 to 7 percentage points).

Economic activity

The picture then changes radically with respect to economic activity among young adults aged 25 to 29 years. Firstly, the role of SEB is much smaller than in the case of the highest level of qualification. The sex difference is reversed, with women from all SEBs more likely to be economically inactive, probably reflecting child-care responsibilities. The sex difference is at its largest among women from lower working-class backgrounds, reaching 15 percentage points, compared with only a 4 percentage point sex difference among women from higher professional backgrounds. This is the most striking example of an interplay between SEB and sex differences.

In contrast there is no sex difference with respect to unemployment among this age group.

Turning to occupational level, the relationship with SEB remains strong, almost as strong as in the case of education (with a gap of 37 percentage points between men from higher professional and lower working-class backgrounds). But the female advantage in university degrees disappears in the labour market, where there is a consistent female disadvantage in higher professional positions. From all SEBs alike, women are less likely than their male peers to be in higher professional occupations. They are also less likely to be in lower working-class jobs and instead are over-represented in lower professional and intermediate class positions.

There is also a significant sex difference with respect to hourly earnings at age 25 to 29 years, with women earnings around 90% as much as men.

In short, outcomes are better for women within education but worse in the labour market (other than unemployment). There is also a notable interplay between SEB and sex with respect to economic activity.

Intersectional analysis by ethnicity

We are looking at a large number of small groups, so it is more difficult to detect differences. We have therefore combined some categories of SEB and some outcome measures to improve the precision of the estimates. However, some inequalities are clear.

First, people from all ethnic minority groups (apart from Black Caribbean) are more likely to gain a degree than White British people from the same SEB. However, their university degrees may come from less selective universities.^{[footnote 10]}

Second, and in striking contrast, people from several ethnic minority groups (Black Caribbean, Black African, Mixed, Pakistani and Indian) are significantly more likely to be unemployed than White people from the same SEB. This is also a finding from previous research. It is possible that racial discrimination in the labour market is a factor in these high rates of ethnic minority unemployment.

Third, people from some groups are unable to obtain occupational levels in keeping with their educational success. In particular, the Pakistani, Bangladeshi and Black African groups have higher proportions of university graduates than the White British group, but do not have higher proportions in the professional classes. Indeed, the Pakistani group has a significantly lower proportion (48% against 59% among those of the highest SEB).

Fourth, economic activity rates tend to be lower among some ethnic minorities (Bangladeshi, Chinese and Pakistani) than among White British people from similar SEBs. More detailed research is needed to uncover the reasons for these disparities. Previous research suggests that, in the case of the Chinese group, it may reflect high rates of continuation in higher education while among the Pakistani and Bangladeshi groups it may reflect higher rates of economic inactivity among young women with caring responsibilities.

There are also signs of a complex interplay between SEB, ethnic group and intermediate outcomes in economic activity and occupational attainment. What we tend to find is that SEB differences play a much smaller role among some minority groups, such as the Chinese, Bangladeshi and Pakistani groups, than among the White ethnic group. There are a variety of possible reasons for this, such as the role of migration and the strength of ethnic capital.^{[footnote 11]} However, most ethnic groups are stratified by social background in much the same way as the majority group is.

We should note the complicating factor of migration status. People from some groups such as the Black Caribbean group, who began to arrive in Britain soon after World War 2, are now largely second or third generation – that is, they were born and educated in Britain. However, other groups such as the Chinese group include a larger proportion of first generation (that is migrants). Migration tends to be associated with downward mobility (for reasons such as lack of fluency in English and foreign qualifications) whereas the second and later generations will tend to have mobility patterns closer to those of the White ethnic group.

Intersectional analysis by disability status

There are disability gaps for every intermediate outcome that we investigate. In every case these gaps are larger than the sex differences. In the labour market, these gaps are cumulative. That is, people with a disability are more likely to be economically inactive. But among those who are active, unemployment rates for disabled people are significantly higher than for people without a disability. And among those who are in employment, hourly earnings are lower.

There are also notable examples of interplay between disability and SEB. So in the cases of NEET, employment (among 16 to 24 year olds) and economic activity (among 25 to 29 year olds), the disability gaps are significant for people from all SEBs but are even larger among those from lower working-class backgrounds. This raises the possibility that professional families may be able to use their resources to help young people with a disability, while those from lower working-class backgrounds may be more dependent on help from the state.

We should note that our measure of disability does not provide information about when the condition started (although the conditions are long-term in the sense of being reported to have been present for at least 12 months). This means that in the case of some outcomes, such as highest qualification, the disability might have come after the outcome rather than before it. It is therefore possible that the data underestimates the extent of the effects of disability on such outcomes. More detailed research using panel data is required to investigate this in depth.

For further information, please see the technical annex for a more detailed analysis and explanation of the differences across SEB, sex, ethnicity and disability for each intermediate outcome covered in the intersectional analysis.

Intermediate outcome 1: The years of compulsory schooling (aged 5 to 16 years)

Summary

The attainment gap between pupils eligible for free school meals (FSM) and those not eligible remains large.

Among children eligible for FSM, girls are much more likely to achieve well than boys.

FSM children from some ethnic backgrounds achieve very well. For example, FSM children of Chinese background perform better than the national average for non-FSM children.

The school years form a critical period in which children develop. These years build an important foundation for getting on in work and in life. Monitoring education and skills development is therefore important for understanding any early differences in outcomes by social background.

Our first set of intermediate outcomes cover the years of compulsory schooling. Early indicators include: level of development at age 5 years, attainment at age 11 years, and attainment at age 16 years.

Social background measures and accountability systems vary across the UK. Therefore we only present the measures for England, but hope to include UK-wide measures in 2024.^{[footnote 12]}

It is worth noting that we rely on administrative data collected by the Department for Education (DfE) for monitoring trends in achievement at ages 5, 11 and 16 years. There have however been a number of recent changes in the official assessment criteria, particularly at age 5 years, so it is difficult to draw any firm conclusions about trends over time. A further difficulty is that the measures of SEB in the administrative data are derived from eligibility for FSM, where there have also been some policy changes over time. Nevertheless, it is clear that there remains a substantial ‘disadvantage’ gap at all 3 stages of the school career, with disadvantaged children doing markedly worse than more advantaged children. The gaps also appear to have widened somewhat at ages 11 and 16 years following the disruption to learning due to the COVID-19 pandemic.

We also find a substantial sex difference at all 3 ages, with girls doing markedly better on the tests than boys with the same FSM status. There are also substantial differences between ethnic groups with Asian pupils, especially those from Indian or Chinese ethnic backgrounds, achieving substantially better than White pupils. Among those eligible for FSM, Black Caribbean pupils’ ratings are little different from White British pupils’ at age 11 years.

1.1 Level of development at age 5 years

Starting with the youngest pupils, we look at ‘good level of development’, as defined in the early years foundation stage (EYFS) profile. This measure shows the percentage of children who achieve a ‘good’ level of development at the age of 5 years – children achieving the expected level in the 3 main areas of learning, and in literacy and numeracy.

As with last year, due to the devolved nature of the education system, we can only monitor this measure for children in England. The only SEB measure available is eligibility for FSM. FSM captures roughly the poorest 15% of students; while not ideal, it is the only SEB measure available in schools data. In particular, due to the transitional protections covering FSM eligibility as we move from old-style multiple benefits to Universal Credit, there is a greatly increased number of children eligible for FSM. This also means that the average child on FSM today is probably not as disadvantaged as the average child on FSM 10 years ago. So this may contribute to closing the measured gap, even with no underlying change in the pattern of achievement.

Figure 3.4 shows that the proportion of children achieving a ‘good’ level of development at the age of 5 years increased in the 7 school years ending in July 2019. Overall, 52% of all children achieved a ‘good’ level of development at age 5 years in the 2012 to 2013 school year, and this increased to 72% for the 2018 to 2019 school year. It then dropped to 65% for the 2021 to 2022 school year. This trend is consistent across both FSM eligible and non-FSM eligible backgrounds, as we reported last year. However, a gap remains in the 2021 to 2022 school year as 69% of children not eligible for FSM achieved a ‘good’ level of development, compared with only 49% of children eligible for FSM. This gap of 20 percentage points is not directly comparable to the previous years due to a change in the EYFS profile and assessments. However, we emphasise that the size of the gap shows there is room for improvement.

Figure 3.4: The gap in the percentage of children achieving a ‘good’ level of development between those eligible for free school meals (FSM) and those not eligible remains large.

Percentage of students achieving a ‘good level of development’ at age 5 years by eligibility for FSM in England, from September 2012 to July 2022.

Explore and download data on level of development at age 5 on the State of the Nation data explorer.

Source: Department for Education. Early years foundation stage (EYFS) profile results from the 2021 to 2022 academic year, 2022.

Note: The grey line represents all children. The percentage ‘good level of development’ tracks development at age 5 years in England only. A child achieving at least the expected level in the early learning goals within the 3 main areas of learning and within literacy and numeracy is classed as having a ‘good level of development’. The EYFS was significantly revised in September 2021 which means we cannot directly compare the outcomes for 2021 to 2022 with earlier years. Data collection during the 2 school years ending in July 2021 was cancelled due to the COVID-19 pandemic. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 13]}

Intersectional analysis of level of development at age 5 years

Differences between boys and girls

Figure 3.5 shows the proportion of boys and girls who achieved a ‘good’ level of development in 2021 to 2022. Overall, girls (75%) are more likely to achieve a ‘good’ level of development compared with boys (62%). FSM eligible girls (57%) are also more likely than FSM eligible boys (42%) to achieve the measure. However, the gap between those eligible for FSM and those not eligible is almost similar for boys and girls with 20 and 18 percentage points respectively.

Figure 3.5: A higher proportion of girls achieve a ‘good’ level of development than boys and the gap between those eligible for free school meals (FSM) and those not eligible is smaller for girls.

Percentage of students achieving a ‘good level of development’ at age 5 years by eligibility for FSM and gender in England, in the academic year 2021 to 2022.

Explore and download data on level of development at age 5 on the State of the Nation data explorer.

Source: Department for Education. Early years foundation stage profile results from the 2021 to 2022 academic year, 2022.

Note: The percentage ‘good level of development’ tracks development at age 5 years in England only. A child achieving at least the expected level in the early learning goals within the 3 main areas of learning and within literacy and numeracy is classed as having a ‘good level of development’. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 14]}

Differences between ethnic groups and between sexes

Figure 3.6 shows the proportion of children who achieved a ‘good’ level of development in the 2021 to 2022 school year by ethnicity. We focus on children who were eligible for FSM. Here we see that, among those eligible, children from the Black, Asian and Mixed or multiple ethnic groups are more likely than White children to achieve a ‘good’ level of performance. The explanation for this finding is not entirely clear, but it could perhaps reflect the large numbers of ethnic minorities living in London where results tend to be better. It could also reflect the high aspirations of ethnic minority parents.

We should also note that there is considerable diversity within these very broad ethnic groups that the DfE uses here. Within the Black group, there are important differences between those with Black African and Black Caribbean backgrounds. And within the Asian group there are important differences between students with Bangladeshi, Chinese, Indian and Pakistani backgrounds (see the more detailed analysis in figure 3.7).

Figure 3.6: Among FSM eligible children, those with White or Other ethnicities have the lowest rates of achieving a ‘good’ level of development.

Percentage of FSM-eligible pupils achieving a ‘good level of development’ at age 5 years by ethnicity in England, from the academic year 2021 to 2022.

Explore and download data on level of development at age 5 on the State of the Nation data explorer.

Source: Department for Education (DfE). Early years foundation stage profile results from the 2021 to 2022 academic year, 2022.

Note: The percentage ‘good’ level of development tracks development at age 5 years in England only. A child achieving at least the expected level in the early learning goals within the 3 main areas of learning and within literacy and numeracy is classed as having a ‘good level of development’. For this outcome the DfE only publishes results for the 5 broad categories shown in the figure. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 15]}

Figure 3.7: Among FSM eligible children, girls are much more likely to achieve a ‘good’ level of development than boys from the same ethnic background.

Percentage of students FSM eligible achieving a ‘good’ level of development at age 5 years by ethnicity and gender in England, from the academic year 2021 to 2022.

Explore and download data on level of development at age 5 on the State of the Nation data explorer.

Source: Department for Education. Early years foundation stage profile results from the 2021 to 2022 academic year, 2022.

Note: The percentage ‘good’ level of development tracks development at age 5 years in England only. A child achieving at least the expected level in the early learning goals within the 3 main areas of learning and within literacy and numeracy is classed as having a ‘good’ level of development. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 16]}

Differences between regions

Since outcomes at school are based on administrative data, they are not subject to the sampling error that affects estimates from the LFS. This means that we can show regional results (for England only).

Figure 3.8 shows that, among FSM-eligible children aged 5 years in England, the highest proportions achieving a good level of development are in London, East Yorkshire, North Lincolnshire and the West Midlands. The lowest proportion is in Cumbria. The other areas of England (those in the 4 lower quintiles of the distribution) are all close to each other. The major story is that all the areas of London do well when considering this measure. In later figures, this ‘London effect’ reappears throughout the educational career.

Figure 3.8: FSM eligible pupils in London, East Yorkshire and North Lincolnshire and the West Midlands are the most likely to achieve a good level of development at age 5 years.

Percentage of FSM-eligible pupils reaching a good level of development at age 5 years by International Territorial Level 2 (ITL2) regions in England, from the academic year 2021 to 2022.^{[footnote 17]}

Explore and download data on level of development at age 5 on the State of the Nation data explorer.

Source: Department for Education (DfE). Early Years Foundation Stage result in 2022.

Note: The DfE shows results for each local authority (LA) in England. This data has been aggregated into ITL2 regions by weighting the LA results by the number of pupils in each authority.^{[footnote 18]}

1.2 Attainment at age 11 years

To monitor attainment at age 11 years we consider the proportion of pupils who achieve the expected standard in reading, writing and maths. This is important to help us understand how academic attainment at age 16 years and beyond may develop.

Figure 3.9 shows the proportion of all pupils who meet the expected standard in reading, writing and maths in the 7 school years to July 2022, by disadvantage status. Of all pupils, 53% achieved the expected standard in the 2015 to 2016 school year, and this increased to 65% in the 2018 to 2019 school year, but decreased to 59% in the 2021 to 2022 school year. Overall, 66% of non-disadvantaged pupils achieved the expected standard in the 2021 to 2022 school year, compared with 43% of disadvantaged pupils. This represents a decrease in attainment for both groups since the last results were published on the 2018 to 2019 cohort.^{[footnote 19]} The decline was 5 percentage points for non-disadvantaged children compared with a drop of 8 percentage points for disadvantaged children. This suggested that disadvantaged children may have been impacted more severely by the disruptions in learning due to the COVID-19 pandemic.

Figure 3.9: Children from disadvantaged backgrounds are less likely to reach the expected standard in reading, writing and maths at key stage 2 (KS2). This gap has widened since before the pandemic.

Percentage of students reaching the expected standard in reading, writing and maths at KS2 by disadvantage status in England, from September 2015 to July 2022. No data was collected for the 2 academic years starting in 2019 and 2020 due to the COVID-19 pandemic.

Explore and download data on attainment at age 11 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at key stage 2 in England, 2022.

Note: The grey line represents all children. Disadvantaged pupils are defined as those who were registered as eligible for free school meals at any point in the last 6 years, and children looked after by a local authority (LA) or who left LA care in England and Wales through adoption, a special guardianship order, a residence order or a child arrangements order. Figures for the 2021 to 2022 school year are based on revised data. Figures for other years are based on final data. Attainment in all of reading, writing and maths is not directly comparable to some earlier years (2016 and 2017) because of changes to teacher assessment frameworks in 2018. Between the academic years 2018 to 2019 and 2021 to 2022, there was a break in assessments due to the pandemic, though these last two data points are comparable.

Figure 3.10 shows the disadvantage gap index between the academic years of 2010 to 2011 and 2021 to 2022 in England. As reported last year, this is a relatively new measure used by DfE and is a positional measure based on rank rather than overall levels. It measures how pupils from ‘disadvantaged and non-disadvantaged backgrounds’ differ in their positions in rankings of performance. This makes the measure more robust to changes in assessments over time. A disadvantage gap score of 0 would indicate that pupils from disadvantaged backgrounds perform equally well as pupils from non-disadvantaged backgrounds. A disadvantage gap score of +10 would mean that every non-disadvantaged pupil did better than every disadvantaged pupil.

Figure 3.10 shows that the disadvantage gap increased by 11% between the school years 2018 to 2019 (2.91) and 2021 to 2022 (3.23). This is the highest level since 2012 and suggests a reversal of the progress made between 2011 and 2018 when the gap reduced every year. As acknowledged by DfE, this suggests the disruption to learning due to the COVID-19 pandemic has had a larger impact on pupils from disadvantaged backgrounds.

Figure 3.10: The disadvantage gap reduced between 2011 and 2019 but has increased since the pandemic to be at its highest level since 2012.

Disadvantage attainment gap index for England at key stage 2 (KS2), from 2011 to 2022.

Explore and download data on attainment at age 11 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at KS2 in England, 2022.

Note: Each year refers to the year in which the academic year ends, for example 2022 refers to the 2021 to 2022 academic year. Comparisons are made by ordering pupil scores in reading and maths assessments at the end of KS2 and assessing the difference in the average position of disadvantaged pupils and others. The mean rank of pupils in the disadvantaged and other pupil groups are subtracted from one another and multiplied by a factor of 20 to give a value between -10 and +10 (where 0 indicates an equal distribution of scores). Disadvantaged pupils are defined as those who were registered as eligible for free school meals at any point in the last 6 years, and children looked after by a local authority (LA) or who left LA care in England and Wales through adoption, a special guardianship order, a residence order or a child arrangements order.

Intersectional analysis of attainment at age 11 years

Differences between boys and girls

Figure 3.11 shows the proportion of children who achieve the expected standards in reading, writing and maths by gender. Overall, 70% of non-disadvantaged girls achieve the expected standard, compared with 61% of non-disadvantaged boys. For those who are disadvantaged, 47% of girls met the expected standard in the 2021 to 2022 school year, compared with 39% of disadvantaged boys. Although girls tend to do better at achieving the expected standard, the gap between those from disadvantaged and non-disadvantaged backgrounds is similar for girls and boys at around 22 to 23 percentage points.

Figure 3.11: In the 2021 to 2022 school year, girls were more likely than boys to reach the expected standard in reading, writing and maths.

Percentage of students reaching the expected standard in reading, writing and maths at key stage 2 (KS2) by disadvantage status and gender in England, in the academic year 2021 to 2022.

Explore and download data on attainment at age 11 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at KS2 in England, 2022.

Note: Disadvantaged pupils are defined as those who were registered as eligible for free school meals at any point in the last 6 years, and children looked after by a local authority (LA) or who left LA care in England and Wales through adoption, a special guardianship order, a residence order or a child arrangements order. Figures for 2022 are based on revised data.

Differences between ethnic groups

Figure 3.12 shows the proportions of pupils reaching the expected standard by FSM status and ethnicity. There are striking differences in overall achievement levels across different ethnicities. For example, 76% of FSM-eligible children of Chinese ethnicity reach the standard but only 12% of Gypsy or Roma ethnicity.

Figure 3.12: The percentage of free school meal (FSM) pupils reaching the expected standard by age 11 years varies greatly by ethnic background.

Percentage of FSM-eligible pupils reaching the expected standard in reading, writing and maths at key stage 2 (KS2) by ethnicity in England, in the academic year 2021 to 2022.

Explore and download data on attainment at age 11 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at KS2 in England, 2022.

Note: Figures for 2022 are based on revised data. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 20]}

Differences among regions

Figure 3.13 shows that, among FSM-eligible students in England, the highest proportions meeting the expected standard at age 11 years are in London, the West Midlands, and Tees Valley and Durham. In contrast, the lowest proportions are in southern and eastern areas of the country. We should however note that the percentages achieving the expected standard are fairly similar across the 3 lowest-performing quintiles, ranging from 35% for the lowest area to 40% for the middle quintile. In contrast there is a large range within the top quintile.

The pattern observed here has been found in previous research. It has often been termed ‘the London effect’, but previous research has also noted that a similar phenomenon of higher-than-expected achievement among FSM pupils is also found in other densely urban areas. The explanation for the London effect is not entirely clear. One major factor is undoubtedly the presence of large numbers of pupils with an ethnic minority background in these metropolitan areas. Large proportions of some minority groups are eligible for FSM. And as we showed in figure 3.7, FSM-eligible minority pupils outperform White British pupils. However, this is not the whole explanation.

Figure 3.13: Disadvantaged pupils in London, the West Midlands, and Tees Valley and Durham are the most likely to achieve the expected standard at key stage 2 (KS2).

Percentage of free school meal-eligible pupils reaching the expected standard in reading, writing and maths at KS2 by International Territorial Level 2 (ITL2) regions in England, in the academic year 2021 to 2022.

Explore and download data on attainment at age 11 on the State of the Nation data explorer.

Source: Department for Education (DfE). National curriculum assessments at key stage 2 in England, 2021 to 2022.

Note: DfE shows results for each local authority (LA) in England. This data has been aggregated into ITL2 regions by weighting the LA results by the number of pupils in each authority.^{[footnote 21]}

1.3 Attainment at age 16 years

The attainment of children at the end of their compulsory education is just as important as the beginning. A young person’s educational outcomes at age 16 years help shape their path onto higher or further education (HE or FE), training and employment. To look at how a person’s SEB influences this progression, we consider the overall levels of attainment for disadvantaged pupils and all other pupils. We also use the KS4 disadvantage gap index for schools in England. The disadvantage gap index summarises the relative attainment in GCSE English and maths between disadvantaged pupils and all other pupils.^{[footnote 22]}

Figure 3.14 shows the proportion of children who achieve a pass (grade 5 or above) in both GCSE English and maths, by disadvantage status. Overall, in the 2021 to 2022 school year, 49.8% of all pupils passed both GCSE English and maths. 30% of disadvantaged pupils achieved a grade 5 or above in both subjects, compared with 57% of all other pupils. This implies a gap of 27.4 percentage points, which is similar to the previous year when the gap was 27.5 percentage points.

Figure 3.14: In the 2021 to 2022 school year, there was a small drop in the proportion of pupils at key stage 4 (KS4) achieving a grade 5 or above in GCSE English and maths, and the gap between disadvantaged and other pupils was similar to previous years.

Percentage of students achieving a pass (grade 5 or above) in both GCSE English and maths by disadvantage status in England, from 2018 to 2022.

Explore and download data on attainment at age 16 on the State of the Nation data explorer.

Source: Department for Education (DfE). National curriculum assessments at key stage 4 in England, 2022.

Note: Pupils are defined as disadvantaged if they are known to have been eligible for free school meals at any point in the past 6 years (from year 6 to year 11), if they are recorded as having been looked after for at least one day or if they are recorded as having been adopted from care. Figures for the school years 2019 to 2020 and 2021 to 2022 are based on revised data. Figures for the 2018 to 2019 school year are based on final data. The 2021 to 2022 year assessment returned to the summer exam series, after they had been cancelled in 2020 and 2021 due to the impact of the COVID-19 pandemic. During this time alternative processes were set up to award grades (centre assessment grades, and teacher assessed grades).

Figure 3.15 shows the disadvantage gap index for KS4 in England in the 12 school years ending in July 2022. This shows the disadvantage gap index widened slightly between 2017 and 2019. In 2020, due to the disruptions to exams caused by the pandemic, centre assessed grades were used instead of exams. This resulted in a slight narrowing of the gap. However, in 2021 although exams were still cancelled, the gap widened. In 2022 as exams were re-introduced, the gap continued to widen and now stands at its highest level since 2021. The DfE states this widening may reflect the ‘difficult circumstances’ which many pupils experienced during the pandemic, resulting in more home learning and restricting attendance in school.^{[footnote 23]}

Figure 3.15: The disadvantaged gap index has widened compared with the 2020 to 2021 school year, and is the largest gap since the 2011 to 2012 school year.

The disadvantage attainment gap index for England at key stage 4 (KS4), from 2011 to 2022.

Explore and download data on attainment at age 16 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at KS4 in England, 2022.

Note: Each year refers to the year in which the academic year ends, for example 2022 refers to the 2021 to 2022 academic year. The disadvantage gap index summarises the relative attainment gap (based on the average grades achieved in English and maths GCSEs) between disadvantaged pupils and all other pupils. The index ranks all pupils in state-funded schools in England and asks whether disadvantaged pupils typically rank lower than non-disadvantaged pupils. A disadvantage gap of 0 would indicate that pupils from disadvantaged backgrounds perform as well as pupils from non-disadvantaged backgrounds. Pupils are defined as disadvantaged if they are known to have been eligible for free school meals at any point in the past 6 years (from year 6 to year 11), if they are recorded as having been looked after for at least one day or if they are recorded as having been adopted from care. Figures for the school years 2019 to 2020 and 2021 to 2022 are based on revised data. Figures for the school year 2018 to 2019 are based on final data. The 2021 to 2022 year assessment returned to the summer exam series, after they had been cancelled in 2020 and 2021 due to the impact of the COVID-19 pandemic. During this time alternative processes were set up to award grades (centre assessment grades and teacher assessed grades).

Intersectional analysis of attainment at age 16 years

Differences between boys and girls

Figure 3.16 shows the proportion of pupils achieving a pass in both GCSE English and maths by sex and disadvantage status in the 2021 to 2022 school year. Overall both non-disadvantaged and disadvantaged girls have higher rates of passing GCSE English and maths than boys. 60% of non-disadvantaged girls passed both subjects, compared with 54% for boys. Similarly, 32% of disadvantaged girls passed both subjects compared with 27% for boys. At 28 percentage points, the disadvantage gap for girls is very similar to that for boys, who have a gap of 27 percentage points.

Figure 3.16: In the 2021 to 2022 school year, girls were more likely than boys to achieve a pass in both GCSE English and maths regardless of their disadvantage status.

Percentage of pupils achieving a pass (grade 5 or above) in both GCSE English and maths by disadvantage status and gender in England, in the academic year 2021 to 2022.

Explore and download data on attainment at age 16 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at key stage 4 in England, 2022.

Note: Pupils are defined as disadvantaged if they are known to have been eligible for free school meals at any point in the past 6 years (from year 6 to year 11), if they are recorded as having been looked after for at least one day or if they are recorded as having been adopted from care. Figures for 2022 are based on revised data.

Differences between ethnic groups

Figure 3.17 shows the proportion of FSM-eligible pupils who achieve a pass in both GCSE English and maths. The figure shows substantial variation between the bottom ethnic group (Gypsy or Roma at 6%) and the top-performing ethnic group (Chinese at 70%). Overall, FSM-eligible pupils of South Asian ethnicities (such as Indian and Bangladeshi) have much higher rates of achieving a pass in both subjects compared with White British FSM-eligible pupils.

Figure 3.17: There is great variation across ethnicities in the attainment of pupils eligible for free school meals (FSM).

Percentage of pupils achieving a pass (grade 5 or above) in both GCSE English and maths for FSM-eligible pupils by ethnicity in England, in the academic year 2021 to 2022.

Explore and download data on attainment at age 16 on the State of the Nation data explorer.

Source: Department for Education. National curriculum assessments at key stage 4 in England, 2022.

Note: Figures for 2022 are based on revised data. FSM eligibility is defined as collected in the school census which states whether a child’s family have claimed eligibility. Parents are able to claim FSM if they receive certain benefits.^{[footnote 24]}

Differences among regions

Figure 3.18 shows a similar geographical pattern to figure 3.13. In particular we see a similar ‘London effect’, with other densely populated urban areas also showing good results. Similar to figure 3.13, we also see that the percentage achieving passes (grade 5 or higher) in English and maths are fairly similar across the 3 lowest-performing quintiles, the percentages ranging from 21% for the lowest area to 27% for the middle quintile. Once again there is a large range within the top quintile. The correlation (at the area level) between attainment at age 11 years with attainment at age 16 years is 0.44 (statistically significant at the 0.01 level).

Figure 3.18 also shows a fairly clear pattern for lower percentages of FSM-eligible pupils in rural areas of England such as Cornwall and Cumbria achieving passes (grade 5 or higher) in English and maths. However, these will be areas with relatively few ethnic minority students. A more detailed intersectional analysis is therefore required to disentangle these different effects.

Figure 3.18: Disadvantaged pupils in London, the West Midlands, and Surrey and Sussex are the most likely to achieve passes (grade 5 or above) in English and maths at GCSE.

Percentage of free school meal-eligible pupils achieving a pass (grade 5 or above) in both GCSE English and maths by International Territorial Level (ITL2) region in England, in the academic year 2021 to 2022.^{[footnote 25]}

Explore and download data on attainment at age 16 on the State of the Nation data explorer.

Source: Department for Education (DfE). National curriculum assessments at key stage 4 in England, 2022.

Note: DfE shows results for each local authority in England. This data has been aggregated into ITL2 regions by weighting the local authority results by the number of pupils in each authority.

Intermediate outcome 2: Post-16 qualifications and progression into the workplace

Summary

Young people (aged 16 to 24 years) of a lower working-class background are much more likely to be not in employment, education or training (‘NEET’) than those of any other background. Yet the SEB gradient in NEET rates is not a smooth one. Differences across higher working-class, intermediate, and professional backgrounds are comparatively small. Only those of lower working-class background stand out. This suggests that there is a relatively small group at the bottom in a precarious economic situation.

Entry to HE presents a different picture, with much lower rates among those of lower working-class background, and much higher rates among those of a higher professional background.

The class inequalities are particularly large when we look at higher degrees rather than first (bachelor’s) degrees. Nearly 4 times as many young people from higher professional backgrounds have a higher degree than those from lower working-class backgrounds, compared with around 2 times as many for first degrees.

People from Chinese, Indian, Black African, Mixed and Other ethnic groups are more likely to obtain degrees than White people from the same SEB.

Young people with a disability are less likely to have a university degree, and more likely to have low qualifications than their peers who do not have a disability.

When compulsory schooling ends at age 16 years, young people have a choice of which path to take. With an increase in young people continuing their education until the age of 18 years, this decision may come later. The number of young people staying in education or training until age 18 years rose steadily until 2020. Whether it is made at age 16 or 18 years, this decision can greatly impact their future careers.

The transition from ages 16 to 29 years from school to work is represented by the next set of intermediate outcomes. In other words, this represents the transition of school leavers to FE, HE, training or employment. We have already highlighted that socio-economic disparities start early in life, before a child starts compulsory schooling and continue during those years. This is also the case for a person’s career in the labour market.

Our indicators here include the rates of young people who are in education, employment or training and who are in neither (‘NEET’) or enrolled in HE, and the highest qualifications they have obtained. These are useful measures to give insight into the socio-economic differences we have already mentioned.

The new 5-class measure of social background reveals greater inequalities than those reported in State of the Nation 2022. Young people from lower working-class backgrounds are particularly disadvantaged (relative to the overall average), while those from higher professional backgrounds are particularly advantaged. Particularly stark inequalities can be seen for postgraduate qualifications and outcomes of those who are NEET.

One notable finding is that the lower working class (which includes those from workless family backgrounds) are well behind other social classes on a number of indicators such as NEET. The large proportions of young people from lower working-class backgrounds who have only low levels of school qualifications or are NEET is especially disturbing as low qualifications and limited labour market experience could severely impact their future prospects.

However, there are some improvements we hope to make in the future. We currently do not monitor progression into FE and apprenticeships. We are also interested in capturing progression from FE into apprenticeships and university, as this is another pathway that can help people progress into the labour market. We would also like to consider adult apprenticeships and capture other vocational training, such as professional and language qualifications for people with English as a second language.

2.1 Destinations following the end of compulsory full-time education

Figure 3.19 shows the proportion of young people aged 16 to 24 years who are in education and training, employment, or NEET. In 2022, young people from a higher professional background were the most likely to be in education and training (36%) and the least likely to be NEET (9%). In contrast, those from a lower working-class background were the most likely to be NEET (21%) and the least likely to be in employment (48%).

It is also notable that there are inverted “U-shaped” relationships between SEB and employment, and to a lesser extent with education and training as well. To some extent this reflects the fact that, among young people aged 16 to 24 years, those from more advantaged backgrounds will tend to remain in education longer and will delay their entry into the labour market. Conversely, those from lower working-class backgrounds may have greater difficulties in finding employment given their typically lower levels of qualification.

Note, for this indicator we were able to report for 2022 as at the time of analysis the 2022 LFS data had become available. However, we did not have sufficient time to update the analysis for all other indicators, but this is what we will do for a future update of our index.

Figure 3.19: Young people from higher professional backgrounds are more likely than their peers to be in education and training, while those from lower working-class backgrounds are more likely to be NEET.

Percentage of young people aged 16 to 24 years in the UK who were in education and training, employment or NEET, 2022, by socio-economic background (SEB).

Explore and download data on destinations following the end of compulsory full-time education on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) 2022, respondents aged 16 to 24 years in the UK, data collected from July to September 2022.

Notes: NEET is defined as ‘not in employment, education or training’ in the week before the survey. SEB refers to the main wage earner’s occupation when the respondent was aged 14 years. Where there was no earner in the family, SEB is included in the lower working class. The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

The 2022 State of the Nation report (figure 3.6) distinguished only 3 social class origins (combining the higher and lower working classes, combining the higher and lower professional categories, but excluding those from workless homes).^{[footnote 26]} The new analysis for this year demonstrates the importance of using a more detailed measure of SEB and brings out the distinctiveness of the lower working class. This was not visible in our 2022 report.

Other research has shown that young people with low or no qualifications, and those leaving care, are particularly vulnerable to being NEET.^{[footnote 27]} Parental worklessness (included in the lower working-class category) has also been shown to be associated with children’s worklessness (Macmillan 2014).^{[footnote 28]} For further discussion of the 2022 figures see the Office for National Statistics (2022) and for a detailed discussion of NEET and risk factors see House of Commons Library (2021).^{[footnote 29] [footnote 30]}

Intersectional analysis of destinations following the end of compulsory full-time education

Differences between men and women

Figure 3.20 shows that the likelihood of being in education and training, employment or NEET are broadly similar among young men and women from each socio-economic background.

The differences between women and men among young people from professional class backgrounds may well reflect the high proportions of women from these backgrounds who continue with their education after age 16 years. There are also hints in the data that the sex difference is reversed among young people from lower working-class backgrounds. But, in this case, the gap is not statistically significant.

Figure 3.20: Social class differences in the likelihood of being in education and training, employment or NEET are similar among young men and women.

Percentages in education and training, employment and NEET, 2014 to 2022 (combined), respondents aged 16 to 24 years in the UK, by socio-economic background (SEB) and sex.

Explore and download data on destinations following the end of compulsory full-time education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey 2014 to 2022, respondents aged 16 to 24 years in the UK, data collected from July to September each year.

Notes: NEET is defined as ‘not in employment, education or training’ in the week before the survey. SEB refers to the main wage earner’s occupation when the respondent was aged 14 years. Where there was no earner in the family, SEB is included in the lower working class. Due to rounding errors, in some instances the totals may not add up to 100%.

Differences between ethnic groups

We see from figure 3.21 that, when looking at young people from lower working-class backgrounds, there are considerable differences from the White British profile in the percentages in education and training, employment, and NEET. In particular, there are much higher proportions of people from ethnic minorities in education and training, and much lower proportions in employment.

Overall, the proportion of people from lower working-class backgrounds who are NEET is highest among the White and Black Caribbean ethnic groups (22%) and lowest among the Chinese and Indian ethnic groups (10%). Those of a Chinese ethnicity are most likely to be in education or training with 62%, compared with only 26% of White ethnic people – the least likely. This is reflected by 51% of White people being in employment – the highest proportion among all groups – compared with only 25% of Black African ethnic people – the lowest proportion.

The explanation for this pattern may be more controversial. One possibility is that it reflects minorities’ expectations of discrimination in the labour market, while another (not incompatible) explanation focuses on the high aspirations of young people from ethnic minorities (perhaps reflecting the ‘positive selection’ and high aspirations of their parents’ generation) and their more ambitious educational choices after age 16 years.^{[footnote 31]}

It is striking that the Black African proportions are more similar to the Chinese ethnic group than the Black Caribbean ethnic group. This may well reflect that many Black African parents were relatively well-educated in Africa but experienced downward mobility into the lower working class after migrating to Britain.^{[footnote 32]}

Figure 3.21: The likelihoods of being in education and training or employment are very different among people from ethnic minorities from lower working-class backgrounds in comparison with White people.

Estimated percentages in education and training, employment and NEET, 2014 to 2022 (combined), respondents from lower working-class backgrounds aged 16 to 24 years in the UK, by ethnic group.

Explore and download data on destinations following the end of compulsory full-time education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey from 2014 to 2022, respondents from lower working-class backgrounds aged 16 to 24 years in the UK, data collected from July to September each year.

Notes: The estimated percentages are derived from a logistic regression model by ethnic group, controlling for sex. The model assumes that class effects are the same within each ethnic group. The estimated percentages shown are those for men. We show percentages only for those with lower working-class backgrounds for illustrative purposes. Due to rounding errors, in some instances the totals may not add up to 100%.

In figure 3.22 we look at the proportion of young people from higher professional backgrounds who are either in education and training or employment, or NEET. Here we can see that there is a clear contrast between those from an ethnic minority background compared with White people. Overall, among these young people from higher professional backgrounds, the percentages in education or training are higher among all ethnic groups and the percentages in employment or NEET are correspondingly lower.

Figure 3.22: The likelihoods of being in education and training or employment are very different among people from ethnic minorities from higher professional backgrounds in comparison with White people.

Estimated percentages in education and training, employment and not in education, employment or training, 2014 to 2022 (combined), respondents from higher professional backgrounds aged 16 to 24 years in the UK, by ethnic group.

Explore and download data on destinations following the end of compulsory full-time education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2022, respondents from higher professional backgrounds aged 16 to 24 years in the UK, data collected from July to September each year.

Note: The estimated percentages shown are those for men. We show percentages only for those with higher professional backgrounds for illustrative purposes. The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Differences by disability status

From all socio-economic backgrounds alike, young people with a disability are much more likely to be NEET and much less likely to be in employment than those without a disability. Differences in the proportions in education are not, for most SEBs, statistically significant. The main finding is that young people with a disability are much more likely than their peers to be NEET rather than in employment.

Figure 3.23: From all socio-economic backgrounds (SEBs) alike, young people with a disability are much less likely to be in employment and much more likely to be NEET than those without a disability.

Percentages in education and training, employment, and NEET, 2014 to 2022 (combined), respondents aged 16 to 24 years in the UK, by SEB and disability.

Explore and download data on destinations following the end of compulsory full-time education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2022, respondents aged 16 to 24 years in the UK, data collected from July to September each year.

Notes: We use the LFS variable DISEA (disability status). This provides a measure of disability consistent with the Equality Act. It considers whether the respondent has a health condition or illness lasting 12 months or more (or both), and whether that condition reduces their ability to carry out day-to-day activities (for details see the LFS user guide volumes 3 and 4).^{[footnote 33]} The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

In our supplementary analysis (see our online tool) we find that the regional estimates for this indicator have a large margin of error. This means we should be careful not to rank areas. Instead, we should pay more attention to the overall pattern across several indicators. For this reason, we have produced the composite indices, above.

2.2 Entry of young people into higher education

We consider differences in entry to HE across SEBs, but note that there are many other routes someone can take following their school education. Monitoring SEB differences in entry to HE is important because many traditional professional class occupations have historically recruited university graduates. This means it is still important that people from all SEBs have the opportunity to proceed onto HE – should they wish to do so. Having the opportunity to study at university is particularly important as research by the Institute for Fiscal Studies (IFS) shows that the gap in earnings between those from the poorest and wealthiest backgrounds is half the size for graduates of HE than across the general population.^{[footnote 34]} This emphasises the important role HE can play in enabling social mobility.

Figure 3.24 shows the proportion of young people aged 18 to 20 years who began studying in HE by SEB in 2021. Overall, young people from a higher professional background (51%) had significantly better chances of participating in HE than people from other SEBs (including those from a lower professional background). And, people from a lower working-class background had significantly lower chances (21%) even when compared with those from a higher working-class background. We find a 30 percentage-point gap in HE participation between those from the higher professional and the lower working classes. This is one of the largest class inequalities that we report.

Figure 3.24: There are large differences across socio-economic backgrounds (SEB) in the proportion of young people entering higher education (HE).

Percentage of young people aged 18 to 20 years in the UK enrolled in HE, 2021, by SEB.

Explore and download data on entry to higher education on the State of the Nation data explorer.

Sources: Office for National Statistics, Labour Force Survey (LFS) 2021, respondents aged 18 to 20 years in the UK.

Notes: The data refers to participation rates of young people aged 18 to 20 years. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

The State of the Nation report in 2022 (figure 3.9) shows modest class differences in entry into HE.^{[footnote 35]} Using LFS data on young people aged 19 years, the report found that “the rates of young people from professional class backgrounds undertaking full-time first degrees has remained relatively stable, especially for women (men 44.6% in 2014 and 37.2% in 2021 versus women 45.8% in 2014 and 43.7% in 2021). But, the respective rates of men and women from working-class backgrounds have risen from 9.8% to 21.7% and 16.4% to 32% over time.”

Our new analysis demonstrates the importance of using a more detailed measure of SEB and brings out major differences within both the working classes and the professional classes.

It is worth noting that there are also important social class differences in entry to more prestigious universities.^{[footnote 36] [footnote 37]} We also published a report in February 2023 which finds that people from poorer backgrounds are less likely to attend more selective universities than wealthier people.^{[footnote 38]} This is particularly important because the report finds evidence from the IFS which suggests those from the poorest backgrounds may be able to overcome most of their earnings disadvantages by attending the most selective universities.^{[footnote 39]} We propose to investigate this in future work.

2.3 Highest qualification of young people

Next we consider the highest qualifications which people have achieved by their mid to late 20s. This is important because by this stage of someone’s life, the qualifications they have accumulated will likely shape their working career. This also helps us understand how our range of mobility outcomes might develop in the future.

Figure 3.25 shows the breakdown of the highest qualification achieved for 25 to 29 year olds by SEB in 2021. It shows a clear pattern of class differences. The more advantaged a young person’s background, the higher the chances that they will secure first or higher degrees. Over two-thirds (71%) of young adults from higher professional backgrounds secure a first or higher degree, compared with just over a quarter (27%) of those from the lower working class. An important new finding is that class inequalities are even higher in the case of postgraduate degrees than they are in the case of first degrees. When considering all degrees, there are 2.5 times as many students from higher professional than lower working-class backgrounds.

Figure 3.25: Socio-economic background (SEB) is strongly related to the qualification level that young people achieve.

Highest level of qualification achieved by young people aged 25 to 29 years in the UK, 2021, by SEB.

Explore and download data on highest qualification on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) 2021, respondents aged 25 to 29 years in the UK.

Note: Parental social class is measured by the main wage earner’s occupation when the respondent was aged 14 years. The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Last year we looked at trends from 2014 onwards, finding that professional men and women were more likely to have a degree than working-class men and women, but also that gaps between these groups had narrowed.^{[footnote 40]} However, this finding did not separate higher degrees from first degrees and used a 3-part class structure. This year’s figures show that, in 2021, people from higher professional backgrounds were 3.5 times more likely than those from lower working-class backgrounds to obtain a higher degree – a much greater level of inequality than that shown in last year’s report.

There is a long tradition of sociological research demonstrating class inequalities in access to higher levels of education (see for example Halsey, Heath and Ridge 1980).^{[footnote 41]} While there may have been some equalisation over time of class chances of achieving school-level qualifications, it appears that class inequalities may not have declined in HE. One account is that, as disadvantaged groups begin to catch up, the advantaged classes will strive to preserve their advantage by ‘raising the stakes’ and focusing on ever higher levels of education (Lucas 2001).^{[footnote 42]} The surprisingly large class inequalities with respect to postgraduate degrees is in line with this account (although more detailed over-time analysis is needed to be sure).

In newly-published work, In and Breen (2022) show that there is a tight link between postgraduate education and the type of undergraduate institution previously attended.^{[footnote 43]} The type of undergraduate institution attended appears to be a key factor and so we will need to look at institution type in future work.

Intersectional analysis of highest qualification of young people

Differences between men and women

Figure 3.26 shows that SEB is related to qualification level among young women in much the same way as among young men. However, within all SEB groups, women have a greater likelihood of attaining a first degree and are correspondingly less likely to have lower-level qualifications than men. The least qualified are young men from lower working-class backgrounds and the most qualified are young women from higher professional backgrounds.

Figure 3.26: Within all socio-economic backgrounds (SEB), higher proportions of young women than young men have a first degree.

Highest qualification, from 2014 to 2021 (combined), respondents aged 25 to 29 years in the UK, by SEB and sex.

Explore and download data on highest qualification on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Differences between ethnic groups

Figure 3.27 shows that while all ethnic groups are divided internally by SEB, people from Chinese, Indian, Black African, Mixed and Other ethnic groups are more likely to obtain degrees than White people from the same SEB. On the other hand, young Black Caribbean people have similar chances of attaining a university degree as young White people. This work also suggests there is some evidence that White British working-class young men experience similar disadvantages with respect to university education as young Black Caribbean men.

Figure 3.27: People from several (but not all) ethnic minorities do better than White people from similar socio-economic backgrounds (SEBs) in gaining a degree.

Estimated percentages obtaining a university degree, 2014 to 2021, respondents aged 25 to 29 years in the UK, by SEB and ethnic group.

Explore and download data on highest qualification on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Because of small sample sizes in the case of some ethnic groups, the outcome measure is simplified to whether the respondent has a university degree or not. The estimated percentages and confidence intervals are derived from a logistic regression model on the likelihood of attaining a degree by ethnic group and SEB, controlling for sex. The model assumes that class effects are the same within each ethnic group. Further tests indicate that this assumption cannot be rejected. The percentages shown are those for men. Percentages are shown only for those with lower working-class and higher professional-class backgrounds for illustrative purposes. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

We should note, however, some complicating factors that might well be present. First, ethnic groups’ educational achievement tends to vary according to whether they arrived as migrants (the first generation) or were born in Britain (the second generation). Second, especially among migrants, SEB may refer to parental occupations in the country of origin, which may not be comparable with those of young people born in Britain. Thirdly, people from ethnic minorities may be less able to gain access to high status universities than White people.^{[footnote 44] [footnote 45] [footnote 46] [footnote 47]}

Differences by disability status

Figure 3.28 shows a consistent ‘disability gap’ across all SEBs. Young people with a disability are less likely to have a university degree, and more likely to have low qualifications than those who do not have a disability. However, we should note that we do not know the precise age at which the illness or disability first occurred. So it could be that in some cases people had completed their education before the onset of the illness. These observed ‘disability’ gaps might therefore underestimate the effect of disability on educational attainment.

Figure 3.28: Within all socio-economic backgrounds (SEBs), lower proportions of young people with a disability have a university degree than other young people.

Highest qualification, from 2014 to 2021, respondents aged 25 to 29 years in the UK, by SEB and disability.

Explore and download data on highest qualification on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: We use the LFS variable DISEA (disability status). This provides a measure of disability consistent with the Equality Act. It considers whether the respondent has a health condition or illness lasting 12 months or more (or both), and whether that condition reduces their ability to carry out day-to-day activities (for details see LFS user guide volumes 3 and 4).^{[footnote 48]} The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

](https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/methodologies/labourforcesurveyuserguidance#labour-force-survey-lfs-user-guides)

Intermediate outcome 3: Work in early adulthood (aged 25 to 29 years)

Summary

Young people from a lower working-class background are significantly less likely to be economically active.

There is a fairly smooth relationship between SEB and young people’s earnings – the higher the background, the higher the earnings.

This earnings gap holds true even when comparing young people with the same educational level.

There are lower levels of economic activity among women, and among young people from Pakistani, Bangladeshi, Chinese and Other ethnic groups, compared with their peers from similar SEBs.

Young people from Pakistani and Black ethnic groups also have significantly higher risks of unemployment than White British young people.

However, considering only those in work, young people from Chinese and Indian backgrounds earn significantly more than White British young people.

From all SEBs, young women are less likely than young men to be in higher professional occupations, and earn less on average.

Geographical analysis of work in early adulthood shows strong correlations at the regional level between childhood poverty and young people’s unemployment, as well as between parents’ and children’s employment.

There is again a disability gap in these early work outcomes. The gap seems to be smaller among those of higher SEBs.

Early steps in a person’s career affect the subsequent years in the labour market. They can determine whether people end up in good or precarious jobs. Professional and managerial jobs are associated with higher earnings and greater security, while precarious jobs tend to be short-term contracts with low wages and little room for progression. Entry into these jobs is usually based on qualifications. However, that isn’t the only factor: social background also makes a difference.

We have included measures of unemployment, occupational level, and earnings among young people to ensure we cover early labour market experiences. The measures cover ages 25 to 29 to cover young people who have gone through HE. The measures also include economic activity.

The geographical distributions of these indicators are quite dissimilar, with rather low correlations at the area (ITL2) level.^{[footnote 49]} The strongest correlation is, unsurprisingly, between higher professional occupations and earnings of young people.^{[footnote 50]}

Turning to the relationship between the drivers and these labour market intermediate outcomes (which we also expect to anticipate eventual mobility outcomes), the most notable correlations are between:^{[footnote 51]}

• the distribution of childhood poverty (driver 1.2) and the distribution of young people’s unemployment (figure 3.30) (with a correlation coefficient of +0.47)
• the distribution of parental lower-working class employment (driver 3.3b) and young people’s economic activity (figure 3.29) (with a correlation coefficient of -0.40)
• the distribution of parental higher professional employment (driver 3.3a) and young people’s own net rate of higher professional employment (figure 3.31) (with a correlation coefficient of +0.66)
• the distribution of parental higher professional employment (driver 3.3a) and young people’s own net level of hourly earnings (figure 3.32) (with a correlation coefficient of +0.61)

We must emphasise that these are preliminary results, and should not be taken to represent causal claims – this means we do not imply a change in one of these indicators causes a change in another. They simply show patterns of association between the different geographical distributions. However, they do suggest that there may be a range of different underlying processes which account for the different distributions of intermediate outcomes. We plan to include further drivers and outcomes in future, and will also use more advanced modelling techniques to improve our understanding of how people’s characteristics and those of the area they are from relate to social mobility outcomes.

As with socio-economic inequalities in education and transition into work, young people from lower working-class backgrounds are particularly disadvantaged when it comes to economic activity and occupational level. However, we also need to recognise that there are some intermediate outcomes – specifically unemployment and earnings – where the position of the lower-working class is not significantly different from that of the higher-working class. The key point is that the pattern of class inequalities can vary across different outcomes. For example, we find different patterns of socio-economic inequalities when we look at economic activity than when looking at earnings.

3.1 Economic activity of young people

Next we focus on young people who are either in employment or seeking employment. Our focus here is not on the type of employment, but instead on whether or not young people are actively participating in the labour market. The official definition of ‘economically active’ is whether someone is in work, or available for and actively looking for work. People can be economically inactive for a range of reasons such as being in full-time education, looking after family, being prevented from work by disability or ill health, or being discouraged from looking for work as a result of discrimination or previous bad experiences.

Figure 3.29 shows the proportions of people aged 25 to 29 years who were economically active in 2021. People from a lower working-class background had the lowest proportion who are economically active (77%), significantly lower than the proportion from any other SEB. In contrast, the proportions of young people from all other groups were not significantly different from each other. These findings parallel those for rates of NEET among 16 to 24 year olds (Intermediate outcome 2.1 above).

Figure 3.29: Young people from a lower working-class background are significantly less likely to be economically active.

Percentage of young people aged 25 to 29 years in the UK who were economically active in 2021, by SEB.

Explore and download data on economic activity on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) 2021, respondents aged 25 to 29 in the UK.

Note: Economically active is defined as either being in work, or available for and actively looking for work. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

The State of the Nation 2022 report (figure 3.12) shows trends in economic activity by sex from 2014.^{[footnote 52]} The report concludes: “Men from working-class backgrounds are just as likely to be active in the labour market as those from professional backgrounds, and this has remained stable from 2014 to 2021.” Our new results raise some questions about this conclusion. In further work we will explore why young people from the most disadvantaged backgrounds have a lower level of economic activity. One possibility is that earlier negative experiences of being NEET or unemployed have had ‘scarring’ effects on those affected and led to them becoming discouraged workers. This means having been NEET at some point earlier in life may have had long-lasting consequences on future employment and earnings outcomes. For detailed studies of scarring see Gregg and Tominey (2005), Scottish Government (2015), Li and Heath (2018). See also Macmillan (2014) on intergenerational persistence of worklessness (that is, worklessness that persists across generations in one family).^{[footnote 53] [footnote 54] [footnote 55]}

The Scottish Government, ‘Consequences, risk factors and geography of young people not in education, employment or training (NEET), research findings’, 2015. Published on GOV.SCOT.

Intersectional analysis of economic activity among young people

Differences between men and women

In figure 3.30 we show the proportion of women and men who are economically active by SEB. We see that the gap in sex is reversed in comparison with the ones for education, with women from all SEBs more likely to be inactive. This could reflect women being more likely to take on child-caring responsibilities.

Second, the sex difference is at its largest between men and women from lower working-class backgrounds, at 15 percentage points, compared with only a 4 percentage point difference among men and women from higher professional backgrounds. This is the most striking example of an interplay between SEB and sex differences. We need to be careful because of floor and ceiling effects (lower or upper limits), but formal tests using logistic regression confirm that the sex differences are larger in more disadvantaged classes.

Figure 3.30: Young women are less likely to be economically active than young men from the same socio-economic background (SEB).

Percentage of young people aged 25 to 29 years in the UK who were economically active, 2014 to 2021, by SEB and sex.

Explore and download data on economic activity on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: The economically active are those who are either in work or who are available for and actively looking for work. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Differences between ethnic groups

Figure 3.31 shows economic activity rates across ethnic groups. Here we see that there are lower levels of economic activity among young people from Pakistani, Bangladeshi, Chinese and Other ethnic groups  compared with young people from similar SEBs. This could be partly due to higher rates of continuation in HE, but it is also possible that some of these are ‘discouraged workers’ who have withdrawn from the labour market as a result of difficulties in finding work. See Heath and Martin for an in-depth analysis of this.^{[footnote 56] [footnote 57]}

Figure 3.31: Young people in some ethnic minority groups are less likely to be economically active than their peers. This could be because they are ‘discouraged workers’.

Percentage of people who are economically active, from 2014 to 2021, respondents aged 25 to 29 years in the UK, by socio-economic background and ethnic group.

Explore and download data on economic activity on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: The estimated percentages and confidence intervals are derived from a logistic regression model on the likelihood of being economically active by ethnic group and SEB, controlling for sex. The model assumes that class effects are the same within each ethnic group. We will test this assumption in further work. The estimated percentages are those for men. Percentages are shown only for those with lower working-class and higher professional-class backgrounds for illustrative purposes. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Differences by disability status

As in the cases of NEET and employment (among 16 to 24 year olds), we see from figure 3.32 that the disability gaps in economic activity are significant for people from all SEBs. However, this gap is even larger among those from lower working-class backgrounds. Young people with a disability from a lower working-class background are the least likely to be economically active. The ‘disability gap’ is relatively small among those from a higher professional background.

The difference in these gaps shows the possibility that professional families can use their resources to help their young people with a disability, while those from lower working-class backgrounds may be more dependent on help from the state.

Figure 3.32: Young people with a disability from a lower working-class background are the least likely to be economically active. The ‘disability gap’ is relatively small among those from a higher professional background.

Percentage of young people aged 25 to 29 years in the UK who were economically active, by socio-economic background (SEB) and disability.

Explore and download data on economic activity on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: We use the LFS variable DISEA. This provides a measure of disability consistent with the Equality Act. It also takes account of whether the respondent has a health condition or illness lasting 12 months or more (or both). And whether that condition reduces ability to carry out day-to-day activities (for details see LFS user guide volumes 3 and 4).^{[footnote 58]} The ‘disability gap’ among those from higher professional backgrounds is significantly lower than among other SEBs. This results from a logistic regression model with interaction terms between SEB and disability. The data used is weighted using the LFS probability weights.

3.2 Unemployment among young people aged 25 to 29 years

Figure 3.33 shows that young people from a lower working-class background had the highest rate of unemployment (6%), while those from a higher professional class background had the lowest rate (3%). However, these differences in the unemployment rates between those from different backgrounds were only borderline significant for 2021. We should note that the rates of unemployment for people aged 25 to 29 years are substantially lower than those shown in figure 4.13 for young people aged 16 to 24 years. This reflects the strong association between age and risks of unemployment.

Figure 3.33: There were no significant socio-economic background (SEB) differences in unemployment among young people in 2021.

Percentage of young people aged 25 to 29 years in the UK who were unemployed in 2021, by SEB.

Explore and download data on unemployment among 25 to 29 year olds on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) 2021, respondents aged 25 to 29 years in the UK.

Note: The unemployed are defined as those who are not in work but available for and looking for work. This means that economically inactive people are excluded from the calculation. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

The State of the Nation report in 2022, figure 3.13, showed the trends by sex and SEB from 2014 to 2021 with no clear patterns of change over time.^{[footnote 59]} The report concluded: “Overall rates are fairly low by historical standards, and around their lowest level since their dramatic rise in the late 1970s and early to mid-1980s. […] However, we must continue to monitor these trends, particularly for those exposed to poverty or with poor social mobility prospects.” Heath and others (2018) showed that young people under 25 years have much higher risks of unemployment than those aged 25 years and over.^{[footnote 60]} They also suggest that social inequalities in unemployment may be ‘hypercyclical’. That is to say, when there is a slack labour market with high rates of unemployment, SEB differences will tend to be larger whereas when there is a tight labour market, SEB differences will tend to be suppressed. Bell and Blanchflower (2011) have also shown that young people were particularly hard hit by the Great Recession in 2008.^{[footnote 61]} So the worry is that the socio-economic inequalities may become magnified over the next year or two if there is another major recession.

Intersectional analysis of unemployment among young people

Differences between men and women

Figure 3.34 shows that, among economically active young men and women from the same SEB, there are no significant sex differences in risks of unemployment. However we note that, when we have the large numbers from pooling data across 2014 to 2021, the higher risk of being unemployed among young people from lower working-class backgrounds becomes clear (and highly significant).

Figure 3.34: Among young men and women from the same socio-economic backgrounds (SEBs) there are no significant sex differences in risks of unemployment.

Percentage of young people aged 25 to 29 years in the UK who were unemployed, from 2014 to 2021 (combined), by SEB and sex.

Explore and download data on unemployment among 25 to 29 year olds on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Differences between ethnic groups

Our findings in figure 3.35 shows that young people from Pakistani, Black Caribbean and Black African ethnic groups have significantly higher risks of unemployment than White British young people from the same SEB. Indeed, these groups have almost 3 times as high unemployment rates as the White British ethnic group. This finding is consistent with some previous research.^{[footnote 62]}

More detailed research taking account of levels of education, migration status and social background has confirmed this finding. Field experiments of discrimination have demonstrated that young people from these minority groups have to make nearly twice as many applications for jobs as White British young people to get a positive response from employers.^{[footnote 63]}

Figure 3.35: Young people from Pakistani and Black ethnic groups have significantly higher risks of unemployment than White British young people from the same socio-economic background (SEB).

Estimated percentages of being unemployed, from 2014 to 2021, respondents aged 25 to 29 years in the UK, by SEB and ethnic group.

Explore and download data on unemployment among 25 to 29 year olds on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: The estimated percentages and confidence intervals result from a logistic regression model on the likelihood of being unemployed by ethnic group and SEB, controlling for sex. The model assumes that class effects are the same within each ethnic group. A formal test confirms this assumption. The estimated percentages are those for men. Percentages are shown only for those with lower working-class and higher professional-class backgrounds for illustrative purposes. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Differences by disability status

In figure 3.36 we show that young people with a disability are also around 3 times as likely to be unemployed as people from the same SEB without a disability. We should also recall that this comes on top of the large disability gaps with respect to economic activity that were shown in figure 3.32. In other words, young people with a disability are less likely than their peers to be economically active, and on top of this those who are economically active are 3 times as likely to be unemployed as their peers. This suggests a cumulative pattern of disadvantage.

Figure 3.36: Across all socio-economic backgrounds (SEBs) disabled young people have around 3 times the likelihood of being unemployed as their peers without a disability.

Percentage of young people aged 25 to 29 years in the UK who were unemployed, from 2014 to 2021 (combined), by SEB and limiting long-term illness or disability (or both).

Explore and download data on unemployment among 25 to 29 year olds on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: We use the LFS variable DISEA (disability status). This provides a measure of disability consistent with the Equality Act. It considers whether the respondent has a health condition or illness lasting 12 months or more (or both), and whether that condition reduces their ability to carry out day-to-day activities (for details see LFS user guide volumes 3 and 4).^{[footnote 64]} The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

3.3 Occupational level of young people aged 25 to 29 years

Figure 3.37 shows clear SEB differences in the occupations taken by young people. Young adults from a higher professional-class background were nearly 3 times more likely to be in a higher professional occupation than those from a lower working-class background. For those from working-class backgrounds who do make it to a professional occupation, they are still twice as likely to be in a lower-professional occupation. These results closely parallel those found for class differences in highest qualification (figure 3.25). For some of the indicators in Intermediate outcome 4 (see below) we explore the effects of SEB after controlling for the highest level of education attained.

Figure 3.37: Socio-economic background (SEB) is strongly related to the occupational class which young people are in.

Percentage of young people aged 25 to 29 years in the UK in different social class positions, 2021, by SEB.

Explore and download data on the occupational level of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) 2021, respondents aged 25 to 29 years in the UK.

Note: The data used is weighted using the LFS probability weights.Due to rounding errors, in some instances the totals may not add up to 100%.

In the State of the Nation report 2022, figures 3.14 and 3.15, showed the trends over time (by sex) from 2014 to 2021, but used a 4-category classification of occupations.^{[footnote 65] [footnote 66]} Our results, using a 5-class grouping show greater class inequalities than in last year’s report, though the trends over time are likely to be similar.^{[footnote 67]} Some members of the group who have never worked may still be in HE, perhaps pursuing a higher degree, so should not be equated with a disadvantaged social position such as unemployment.

There has been extensive research on the relationship between SEB and early occupational class.^{[footnote 68]} One notable finding is that the link is particularly marked among young people with lower qualifications. In other words, there is an interaction between SEB, qualification level and occupation. This is consistent with other research, such as Bukodi and Goldthorpe’s, who also find especially large SEB differences among those with lower qualifications in the chance of gaining a professional-class job.^{[footnote 69]} The authors also find that SEB may help prevent someone from dropping down in occupational class more than act as a barrier to going to a higher class (in other words, be a “glass floor” rather than a “glass ceiling”).

Intersectional analysis of occupational level of young people

Differences between men and women

In figure 3.38 we see that young women from all SEBs are less likely to be in higher-professional occupations when compared with young men. Instead, they are more likely to be found in lower-professional occupations. This results in the proportion of young men and women in all professional jobs being quite similar. We can also see that young women are more likely than young men to be in intermediate-class jobs (which are typically clerical and service occupations). This is balanced by slight under-representation of young women in the 2 working classes. These patterns are long-standing, and historically the disparities between men’s and women’s occupational distributions have been declining.^{[footnote 70]}

Figure 3.38: Young women from all socio-economic backgrounds (SEBs) are less likely than young men to be in higher-professional occupations.

Percentage of young people aged 25 to 29 years in the UK in different social class positions, from 2014 to 2021 (combined), by SEB and sex.

Explore and download data on the occupational level of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Differences between ethnic groups

Figure 3.39 shows that young people of Indian and Chinese ethnic backgrounds have higher chances of entering professional occupations than young people from similar SEBs. However, young people of a Pakistani background have significantly poorer chances. The higher chances of those from Indian and Chinese backgrounds reflects their higher proportions gaining university degrees, but education alone cannot account for the lower-than-expected chances of young people of a Pakistani ethnicity. An alternative possibility is that the areas of the country where this group tends to live do not offer such good occupational opportunities. We shall explore this in further work.

Figure 3.39: While young people from an Indian and Chinese ethnic background have higher chances of entering professional occupations compared with other ethnic groups, young people of Pakistani background have significantly poorer chances.

Estimated percentages obtaining a professional occupation, from 2014 to 2021 (combined), respondents aged 25 to 29 years in the UK, by socio-economic background (SEB) and ethnic group.

Explore and download data on the occupational level of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Because of small sample sizes the outcome measure is whether the respondent has a professional occupation (either higher or lower professional). The estimated percentages and confidence intervals are derived from a logistic regression model on the likelihood of being in a professional occupation by  SEB and ethnic group, controlling for sex. The model assumes that class effects are the same within each ethnic group. A formal test shows that this assumption does not hold for the Chinese group. The estimated percentages are those for men. Percentages are shown only for those with lower working-class and higher professional-class backgrounds for illustrative purposes. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Differences by disability status

In figure 3.40 we see that the story of cumulative disadvantage for young people with a disability is also present in the case of occupational level. On top of the disadvantages around economic activity and employment, young people with a disability are about twice as likely as young people from the same SEB without a disability to be in lower working-class jobs. Balancing this over-representation in the lower-working class, we find under-representation spread across the other classes. The one exception concerns those from higher-professional backgrounds, who are slightly over-represented relative to their peers from the same background in lower-professional employment. Again, this suggests that these families can use their resources to help their young people with a disability in ways that are not possible for those from other backgrounds.

Figure 3.40: From all socio-economic backgrounds (SEB), young people with a disability have higher risks than those without a disability of being in a lower working-class occupation.

Percentage of young people aged 25 to 29 years in the UK in different occupational positions, from 2014 to 2021 (combined), by SEB and disability.

Explore and download data on the occupational level of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: We use the LFS variable DISEA. This provides a measure of disability consistent with the Equality Act. It considers whether the respondent has a health condition or illness lasting 12 months or more (or both). And whether that condition reduces ability to carry out day-to-day activities (for details see LFS user guide vols 3 and 4). We note that percentage point gaps may be misleading here because of floor and ceiling effects, but the odds ratios are also larger for those from working-class origins than higher-professional origins. The data used is weighted using the LFS probability weights. Due to rounding errors, in some instances the totals may not add up to 100%.

3.4 Earnings of young people aged 25 to 29 years

Figure 3.41 shows a clear class-based trend in the average earnings of young people. The more advantaged a young people’s SEB, the higher their average weekly earnings. So young people from a lower working-class background earn 70% of what those from a higher-professional background earn. While there is a steadily rising relationship between SEB and average earnings, those of young people from lower-working class, higher-working class, and intermediate backgrounds are not significantly different from each other. However, they are significantly lower than the average earnings of young people from higher- or lower-professional backgrounds. More detailed analysis shows that the results are the same for median earnings.

Figure 3.41: Socio-economic background (SEB) is strongly related to the level of young people’s earnings.

Mean hourly earnings of young people aged 25 to 29 years in the UK, 2021, by SEB.

Explore and download data on earnings of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) 2021, respondents aged 25 to 29 years in the UK.

Note: Self-employed respondents and those without earnings are excluded. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals. Earnings have been adjusted for inflation with a base year of 2021.

Economists have tended to study the relationship between parents’ earnings and their adult children’s earnings. Unfortunately, we cannot reliably estimate parents’ earnings from the LFS and have to use other sources, such as birth cohort studies. One important result is that the relationship between parents’ and children’s earnings varies across the life cycle, being weaker when the adult children are still in the early career stages and substantially stronger at later stages.^{[footnote 71]} We explore whether this applies to the relationship between SEB and earnings when we examine career progression below.

Intersectional analysis of earnings of young people aged 25 to 29 years

Differences between men and women

Figure 3.42 shows that young women’s hourly earnings are around 90% of the hourly earnings of young men from the same SEB. This may be due to the tendency of young women to work part-time, since hourly rates for part-time work are often lower than for full-time work.

Figure 3.42: There is an earnings gap between young men and women from all socio-economic backgrounds (SEB).

Mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by SEB and sex.

Explore and download data on earnings of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Self-employed respondents and those without earnings are excluded. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals. Earnings have been adjusted for inflation with a base year of 2021.

Among those from lower working-class backgrounds, young women earn 87% of young men’s earnings. This increases to 91% for those from higher-professional backgrounds. However these differences are not significantly different.

Differences between ethnic groups

Figure 3.43 shows that among those in work, young people from Chinese and Indian ethnic backgrounds earn significantly more than White British young people from the same SEB. This reflects their higher probability of being in a professional class. In contrast, young people from a Bangladeshi ethnic background earn significantly less than White British young people. It is not clear why this might be the case, but could perhaps reflect local labour market conditions where they live.

Figure 3.43: Among those in work, young people from Chinese and Indian backgrounds earn significantly more than White British young people.

Estimated mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by socio-economic background (SEB) and ethnic group.

Explore and download data on earnings of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: The estimated means and confidence intervals are derived from a linear regression model of log hourly earnings by SEB and ethnic group, controlling for sex. The model assumes that class effects are the same within each ethnic group. However, the assumption does not hold for the White Other group. The means shown are those for men. Means are shown only for those with lower working-class and higher professional-class backgrounds but other SEBs are included in the analysis. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals. Earnings have been adjusted for inflation with a base year of 2021.

We should also note that some (but not all) ethnic groups such as those from South Asia have high rates of self-employment. However, self-employment earnings could not be included in our analysis. There is therefore a risk that the true figures for the earnings of South Asian groups might be lower than those shown here.

Differences by disability status

Figure 3.44 shows that young people with a disability tend to earn significantly less than young people without a disability from the same SEB. This is what would be expected given their lower occupational positions. While the disability gap appears to be largest among those from higher-professional backgrounds, we should note the imprecision of the estimates.

Figure 3.44: Young people with a disability tend to earn significantly less than those without a disability.

Mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by socio-economic background (SEB) and disability.

Explore and download data on earnings of young people aged 25 to 29 years on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Self-employed respondents and those without earnings are excluded. Also note that among people with a disability, those in work are a more selective group (since their inactivity rate is higher). Because of the skewed distribution of earnings, we take the log of earnings when checking for interactions between disability and SEB. With this model specification we do not find a significant interaction. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals. Earnings have been adjusted for inflation with a base year of 2021.

3.5 Returns to education for young people: returns in earnings

Figure 3.45 shows the difference between what 2 different young people of the same SEB would earn on average, if one had the lowest level of qualifications and the other had a higher level. For example, figure 3.45 shows that, if we considered 2 young people from the same SEB, we would expect the one with a higher degree to earn 63% more than the one with no GCSEs, while those with a first degree (but not a higher degree) earn 54% more. Furthermore, those with qualifications at GCSE, A level or FE below degree level, earn approximately 10, 20 and 30% more than those with the lowest levels of education.

Figure 3.45 illustrates the link between education and earnings, not the link between SEB and earnings. It can usefully be compared with figure 3.50, which illustrates the link between SEB and earnings for people with the same level of education.

Figure 3.45: Young people with higher levels of education earn substantially more than those with lower levels of education.

Percentage differences in hourly earnings of young people aged 25 to 29 years in the UK, from 2019 to 2021 (combined), relative to those with lower level (below GCSE grade 1 or equivalent), controlling for socio-economic background (SEB), sex and age.

Explore and download data on income returns to education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2019 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Percentage differences were estimated from a linear regression model of log hourly earnings by educational level, controlling for SEB, sex and age. We pool the data for years 2019 to 2021 in order to obtain more accurate estimates. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021.

In addition, figure 3.46 shows that, among young people with similar educational levels, there are significant SEB pay gaps. So those from higher-professional backgrounds earn 18% more than those from a lower working-class background who have the same level of education.

More detailed economics research (for England, using the Longitudinal Education Outcomes linked dataset) has shown that, among those with degrees, returns vary according to the prestige of the university, the subject studied and the class of degree.^{[footnote 72] [footnote 73] [footnote 74] [footnote 75] [footnote 76]} There is also a large literature in sociology showing that HE brings occupational advantages.^{[footnote 77]}

highest educated in the United Kingdom’, 2022. Published on JOURNALS.SAGEPUB.COM.

Figure 3.46 shows that the earnings gaps between young people with different levels of education have remained more or less constant since 2014 to 2016, and from 2019 to 2021. However, it seems that the earnings gap between those with higher degrees and those with first degrees has narrowed somewhat. As can be seen from the confidence intervals, the earnings gap was significant in the earliest period but non-significant in the latest period.

Figure 3.46: The earnings gaps between levels of qualifications have remained roughly constant between 2014 to 2016, and 2019 to 2021.

Hourly earnings in pounds (£) of young people aged 25 to 29 years in the UK, three-year moving averages from 2014 to 2016 until 2019 to 2021, by highest qualification controlling for socio-economic background (SEB), sex and age.

Explore and download data on income returns to education on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) from 2014 to 2016 and from 2019 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Hourly earnings were estimated from a linear regression model of log hourly pay by educational level, controlling for SEB, sex and age. The estimates shown refer to the hourly earnings of  men who were from a lower working-class background. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

We now consider whether there are sex, disability and ethnicity gaps among young people who have the same levels of education and similar social backgrounds. Education is a major driver of earnings and could explain the gaps in part, but this also depends on whether there are equal opportunities for people with the same level of education.

Intersectional analysis of returns (in income) to education

Differences between men and women

What we find in figure 3.47 is that there are still significant sex differences in hourly earnings at most levels of education, though the gaps are somewhat smaller among the least and the most highly educated. One possible explanation for the small gap among those at the lower level of education may reflect the minimum wage, which effectively puts a ‘floor’ under women’s earnings.

Figure 3.47: Young women’s hourly earnings are significantly lower than those of young men with the same level of qualification and from the same socio-economic background (SEB).

Estimated mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by educational level and sex, controlling for SEB and age.

Explore and download data on income returns to education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Hourly earnings were estimated from a linear regression model of log hourly pay by educational level and sex, controlling for SEB and age. Estimates are shown for people aged 27 years from lower working-class backgrounds. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Differences between ethnic groups

In our earlier analysis (figure 3.43), we found that young people with Indian and Chinese ethnic backgrounds had significantly higher hourly earnings than White British young people from the same SEB, while those with a Bangladeshi background had significantly lower earnings. In this new analysis, which takes account of levels of education, in contrast, we find little in the way of significant earnings gaps.

In figure 3.48 we use a simplified measure of educational level, distinguishing degree-level qualifications from non-degree levels. We do this in order to have adequate sample sizes for intersectional analysis. As can be seen, hourly earnings are broadly similar between White British and ethnic minority young people, both for those with and without a degree. This strongly suggests that the high Indian and Chinese earnings that we saw earlier were due to the high levels of education of these 2 groups. In other words, the educational level largely explains the earnings gaps.

There are, however, hints in the figure that graduates of Black Caribbean and Bangladeshi ethnicities do not obtain as good returns to their education as the other groups do. In-depth analysis is needed to verify this finding and investigate why it occurs.

Figure 3.48: The hourly earnings of young people with an ethnic minority background are similar to those of White British young people with the same level of qualification.

Estimated mean hourly earnings in pounds (£) of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by ethnic group and educational level controlling for socio-economic background (SEB) and age.

Explore and download data on income returns to education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Hourly earnings were estimated from a linear regression model of log hourly pay by ethnic group and educational level, controlling for SEB and age. Estimates are shown for people aged 27 years from lower working-class backgrounds. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Differences by disability status

In figure 3.49 we see that the pattern of disability earnings gaps parallels the finding for sex differences. The gaps are reduced at the lowest and the highest levels of education, but at all the intermediate levels there are still significant and substantial gaps in the hourly earnings of those with and without a long-term health condition or disability. Among those with a first degree (but not a higher degree), young people with a disability earn only 84% of the hourly earnings of those without a disability. This is the same magnitude of gap as those that we saw earlier where we took account of SEB but not educational level.

Figure 3.49: The hourly earnings of young people with a disability are significantly lower than those of young people without a disability with the same level of qualification.

Estimated mean hourly earnings in pounds (£) of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by disability status controlling for socio-economic background (SEB) and age.

Explore and download data on income returns to education on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Hourly earnings were estimated from a linear regression model of log hourly pay by disability and educational level, controlling for SEB and age. Estimates are shown for people aged 27 years from lower working-class backgrounds. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

3.5 Returns to education for young people: Direct effect of social origins on hourly earnings

We can also look at hourly earnings from a different perspective and examine how earnings differ for people with the same educational level but different social origins. Whereas figure 3.45 looked at the direct effects of educational level on hourly earnings, figure 3.50 shows the direct effects of SEB on hourly earnings. We compare the earnings of young people from different SEBs but with similar educational levels. As we can see, there are significant SEB pay gaps. Those from higher-professional backgrounds earn 18% more than those from a lower working-class background with the same qualification level.

Figure 3.50: Young people from professional backgrounds earn significantly more than those from other backgrounds but with the same level of education.

Percentage differences in hourly earnings of young people aged 25 to 29 years in the UK, from 2019 to 2021 (combined), relative to those from lower working-class backgrounds, controlling for highest educational level, sex and age.

Explore and download data on the direct effect of social origin on earnings on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2019 to 2021, respondents aged 25 to 29 years in the UK.

Notes: Percentage differences were estimated from a linear regression model of log hourly pay by SEB, controlling for educational level, sex and age. The reference group is men who were from a lower working-class background and had lower-level qualifications (below CSE grade 1 or equivalent). We pool the data for years 2019 to 2021 in order to obtain more accurate estimates. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Intersectional analysis of direct effect of social origins on hourly earnings

In this section we show the results of intersectional analysis of direct effects of social origins by controlling for education and age. The findings are mostly parallel to those shown for the intersectional analyses of returns to education but they are presented from a different perspective.^{[footnote 78]}

Differences between men and women

As with the intersectional analysis of returns to education, we see that there are significant sex differences in hourly earnings among young men and women. This holds true for young people from all SEBs, with the gaps tending to be slightly smaller among those with lower levels of education.

Figure 3.51: Young women’s hourly earnings are significantly lower than those of young men with the same level of qualification and from the same socio-economic background (SEB).

Estimated mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by SEB and sex, controlling for educational level and age.

Explore and download data on the direct effect of social origin on earnings on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: Hourly earnings were estimated from a linear regression model of log hourly pay by SEB and sex, controlling for educational level and ageInteractions between sex and SEB were not significant and have therefore not been included. Estimates are shown for people with the lowest levels of education and aged 27 years. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Differences between ethnic groups

In our earlier analysis, (figure 3.43) we found that young people with Indian and Chinese backgrounds had significantly higher hourly earnings than White British young people from the same SEB, while those with a Bangladeshi background had significantly lower earnings. The new analysis takes account of levels of education and, in contrast, we find little in the way of significant earnings gaps between people from different ethnic groups. Hourly earnings are broadly similar between White British and ethnic minority young people from similar SEBs. Earnings are however somewhat lower among young people from the Pakistani and Bangladeshi ethnic groups (significantly so in the case of the latter group).

Figure 3.52: The hourly earnings of young people from an ethnic minority background are generally similar to those of White British young people from similar socio-economic backgrounds (SEB).

Estimated mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021, by ethnic group and SEB, controlling for educational level and age.

Explore and download data on the direct effect of social origin on earnings on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: Hourly earnings were estimated from a linear regression model of log hourly pay by ethnic group and SEB (2 categories only, namely professional and non-professional), controlling for educational level and age. Since interaction terms between ethnicity and SEB were of marginal significance, they are not included in the model. Estimates are shown for those with the lowest levels of education and aged 27 years. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Differences by disability status

In figure 3.53 we again see disability earnings gaps among young people from all SEBs even after controlling for education. These gaps are slightly smaller in percentage terms than the disability earnings gaps that we saw in figure 3.44 (which did not control for educational level). This suggests that education partly explains the gap shown in figure 3.44 but the gap remains very large and statistically significant: young people with a disability earn less than 90% of what young people without a disability earn.

Figure 3.53: The hourly earnings of young people with a disability are significantly lower than for young people from the same socio-economic background (SEB) and educational level but with no disability.

Estimated mean hourly earnings of young people aged 25 to 29 years in the UK, from 2014 to 2021 (combined), by disability status and SEB, controlling for educational level and age.

Explore and download data on the direct effect of social origin on earnings on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 29 years in the UK.

Note: Hourly earnings were estimated from a linear regression model of log hourly pay by disability and SEB, controlling for educational level and age. Estimates are shown for those with the lowest levels of education and aged 27 years. The data used is weighted using the LFS probability weights. Earnings have been adjusted for inflation with a base year of 2021. The error bars show 95% confidence intervals.

Intermediate outcome 4: Career progression (aged 35 to 44 years)

Summary

The proportion of people with university degrees increases between the ages of 25 and 32 years. In other words, many people are getting further qualifications between these ages. There is no clear difference across SEBs, a finding which contrasts with work carried out using birth-cohort studies.

Young people from more advantaged social backgrounds are more likely to progress in their careers.

In this section, our measures show the progress people make in their early-to-mid career (in their 20s to 30s). This helps us to compare a person’s origins to their destination and offers insight into possible future mobility outcomes. This is monitored through income (the amount someone earns) and occupational class (the types of job someone does).

We find significant SEB differences in the cases of occupational and income progression. In general, we find that young people from professional backgrounds pull away from other young people between the ages of 25 and 40 years in terms of income and occupational level, while young people from lower working-class backgrounds fall further behind.

Surprisingly, in the case of university degrees, the number of those gaining a degree increases by more or less similar amounts between the ages of 25 and 32 years. This is different to findings from some other research (using different methodologies) which have found that social background gaps when obtaining further qualifications increase over people’s working careers. One possibility is that young people from working-class backgrounds take longer to complete their degrees (for example because of the need to earn in order to finance their studies).

4.1 Further training and qualifications

The results show that the proportion of people with a degree increases by 10 percentage points or more between the ages of 25 and 32 years. At both ages there is a clear difference between people from different SEBs. Young people from professional backgrounds have the highest percentage with a degree and those from the working class have the lowest. However, all 3 groups increase their proportions with a degree over the life course.^{[footnote 79]}

So, among those from professional backgrounds, the proportion increases by 13 percentage points, among those from intermediate backgrounds, by 21 percentage points, and among those from working-class backgrounds by 14 percentage points.^{[footnote 80]}

Figure 3.54: The proportion of young people with university degrees increases between the ages of 25 and 32 years.

Percentages of young people born in 1989 who had obtained degrees at age 25 years (in 2014) and age 32 years (in 2021) in the UK, by socio-economic background.

Explore and download data on further training and qualifications on the State of the Nation data explorer.

Source: Office for National Statistics, Labour Force Survey (LFS) from 2014 and 2021, respondents born in 1989.

Notes: Age and cohort analysis.^{[footnote 81]} We combine the higher and lower professional classes, and the higher and lower working classes, in order to obtain more accurate estimates. This analysis compares the percentages of young people born in 1989 who had obtained university degrees at age 25 years (in 2014) and age 32 years (in 2021) respectively.^{[footnote 82]} It provides a comparison of independent samples in the 2 surveys, not comparisons of the same individuals at different time points (which would have required a panel study). The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Previous research has generally shown that people from higher SEBs gain more from ‘lifelong learning’ than those with lower SEBs.^{[footnote 83]} For example Bukodi (2017) uses the 1970 Birth Cohort Study (BCS) to examine acquisition of further qualifications over the course of panel members’ working lives (that is after the completion of full-time education and taking up their first ‘significant’ job).^{[footnote 84]} This data covers the period from around 1990 to 2008, so is considerably older data than the LFS data used here. Bukodi’s results differ considerably from ours: she finds that people with a professional background benefit most from FE. Bukodi also finds that those of lower SEB are more likely to obtain vocational qualifications, with significant differences between women and men, perhaps because vocational qualifications may help women to re-enter the labour market after a career break.^{[footnote 85]} Unfortunately, it is not currently possible to replicate Bukodi’s analysis with newer data, although this will become possible eventually when the Millennium Cohort Study participants have spent longer in the labour market.

David Raffe, ‘The ‘alternative route’ reconsidered: part-time further education and social mobility in England and Wales’, 1979. Published on JOURNALS.SAGEPUB.COM.

4.2 Occupational progression

Figures 3.55 and 3.56 show the patterns of upward career mobility into the professional classes (known as intra-generational mobility) among young men and women. The chances of gaining access to the professional classes increases steadily from the age of 25 to 35 years, although the increase is greater for young people from professional backgrounds. The rate of increase slows in people’s late 30s and early 40s, and even appears to decline for people coming from intermediate or working-class backgrounds, especially for women. One possible explanation for this decline is that women (and perhaps men) with childcare responsibilities take up lower-level part-time employment. For a more detailed analysis of men and women’s career paths, see Bukodi and others 2012.^{[footnote 86] [footnote 87]}

Figure 3.55: Young men from more advantaged socio-economic backgrounds (SEB) are more likely to progress occupationally.

Probability of access to the professional classes for men by SEB and age controlling for survey year in the UK, from 2014 to 2021 (combined).

Explore and download data on occupational progression on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 44 years in the UK in work at the time of the survey.

Notes: Estimates are the average marginal effects derived from a logistic regression model of access to the professional classes by SEB and age, controlling for survey year. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Figure 3.56: Young women from more advantaged socio-economic backgrounds (SEBs) are more likely to progress occupationally.

Probability of access to the professional classes for women by SEB and age controlling for survey year in the UK, from 2014 to 2021 (combined).

Explore and download data on occupational progression on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 44 years in the UK in work at the time of the survey.

Notes: Estimates are the average marginal effects derived from a logistic regression model of access to the professional classes by SEB and age controlling for survey year. The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

These findings do not come from a panel study in which the same respondents are re-interviewed yearly but from pooled annual surveys interviewing different respondents each year. They may not be as valid as those using a birth cohort design (although most birth cohort analyses suffer from major attrition, which will not be a problem in the case of our design). If we use the same birth cohort approach as for figures 3.55 and 3.56, looking at the occupational achievements of those born in 1989 between the ages of 25 and 35, we obtain a similar story of improvement in the chances of entering professional work over the course of  early careers. Interestingly, this cohort design does not show a significant widening gap between the classes, although it has less statistical power (because of its much smaller sample size) than the design used for figures 3.55 and 3.56.

These results are broadly similar to those found by Bukodi and Goldthorpe (2011) using BCS data.^{[footnote 88]}

4.3 Income progression

Figures 3.57 and 3.58 show the pattern of earnings progression among young men and women, using the same methodology as for figures 3.55 and 3.56. Average incomes increase steadily from the age of 25 to 35 years, and tend to flatten out after. The average increase is greater for young people from professional backgrounds than for those from intermediate or working-class backgrounds, for whom flattening out in mid-career is more evident. This widening gap could be a consequence of the greater income progression gained by those with higher levels of education.

Women’s average earnings are lower, and do not increase as rapidly as men’s. Research shows this is likely related to women’s greater likelihood of part-time working mid-career. An important factor is that women spend less time in paid work, and more time working part-time, than men. As a result, they miss out on earnings growth associated with more experience (Dias and others 2018).^{[footnote 89]}

Figure 3.57: On average, annual incomes of young men from advantaged socio-economic backgrounds (SEBs) increase more.

Income progression of men in the UK, from 2014 to 2021, by SEB and age.

Explore and download data on income progression on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021 (pooled), respondents aged 25 to 44 years in the UK in paid employment.

Notes: Estimates are derived from a linear regression of annual income by SEB and age controlling for survey year, and number of dependent children.^{[footnote 90]} The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

Figure 3.58: On average, annual incomes of young women from advantaged socio-economic backgrounds (SEBs) increase more.

Income progression of women in the UK, from 2014 to 2021, by SEB.

Explore and download data on income progression on the State of the Nation data explorer.

Source: Office for National Statistics, pooled Labour Force Survey (LFS) from 2014 to 2021, respondents aged 25 to 44 years in the UK in paid employment.

Notes: Estimates are derived from a linear regression of annual income controlling forage squared (to account for the changing importance of age as people get older), survey year and number of dependent children.^{[footnote 91]} The data used is weighted using the LFS probability weights. The error bars show 95% confidence intervals.

These results are very similar to those found by Dias and colleagues, who used a similar methodology to analyse LFS data from 1993 to 2017.^{[footnote 92]} They broke down income progression by qualification level rather than by social background, but as we have seen SEB and educational level are highly correlated.

Conclusion

Our more detailed analysis of outcomes earlier in life has revealed striking patterns both across geography and individual characteristics.

Two individuals from the same SEB are likely to have different outcomes depending on where in the UK they grew up. Someone growing up in London and adjoining areas is more likely to attain higher qualifications, earnings and occupational level than someone from the same SEB growing up in a more rural or remote area. Yet for the same 2 individuals, the risk of unemployment, inactivity, and lower working-class employment is also higher in London. This contrast shows the importance of looking within areas, as well as between areas.

Breakdowns by individual characteristics show that different groups can have very different outcomes early in life. Unfortunately, those with a disability tend to do worse on all of the outcomes that we measure, including educational attainment, income, and employment. People with a disability are more likely to be economically inactive, unemployed, and earn lower wages. The gap between those with a disability and those without is also larger in lower SEB groups.

Ethnic background presents a complex pattern. People from all ethnic minorities (apart from Black Caribbeans) are more likely to gain a degree than White British people from the same SEB, although their university degrees may come from less selective universities. Yet this educational success does not translate into greater work success. The Pakistani, Bangladeshi and Black African groups, despite having higher proportions of university graduates than the White British group, do not have higher proportions in the professional classes. The Black Caribbean, Black African, Mixed, Pakistani and Indian groups are also significantly more likely to be unemployed. Interestingly, SEB differences play a much smaller role among some minority groups, such as the Chinese, Bangladeshi and Pakistani groups, than among the White ethnic group.

Young women (aged 16 to 24 years) tend to be slightly more likely to be in education and training than young men, and the largest sex differences (reaching 4 percentage points) are among those from professional backgrounds. The picture for those not in education, employment or training is the mirror image.

This picture is similar when we look at sex differences. Young women from all SEBs have slightly higher qualification levels than their male peers, and are less likely to be NEET. Yet there is a consistent female disadvantage with respect to occupational position and earnings.

Case studies

Simon, age 16 years, from Norwich

When I was at high school, I found maths hard. Our class sizes were quite big so the teacher couldn’t tell when I was falling behind. I feel like I didn’t always get things the same way as others in the class.

My school had a sixth form, but I decided to go to City College, Norwich, where they do T levels. I’ve always liked computers, so the T level in Digital Production Design and Development looked great. But, it turned out that without GCSE maths, I couldn’t go on to do the level 3. Maths made me feel kind of nervous and I didn’t like the idea of trying the GCSE all over again, but my school referred me for some free one-to-one tutoring with Get Further. I was a bit worried about having tutoring because I hadn’t done anything like that before.

When I had the first call, the tutor told me he was a proper teacher and that we’d have calls every Thursday with either just me or one or 2 others. He asked me how I liked to learn. We worked out that I get it best by going through the course work together and then going away and practising on my own after the sessions. It was the same tutor every time which was good because it meant he knew my level and how I learn, so it suited me.

Now I’m at college. I do my pre-T-level course 4 days a week, and on my off day I do maths. I can’t progress on the T level without maths. But, now I’ve already done half my maths GCSE and I’m on for a pass. This means I should be able to do level 3 next year.

I can see how much my work has improved quality-wise, not just in maths, since getting the tutor. I feel much more motivated to do well at college now that I know I can.

When I finish my T level I’ll have learnt all about coding, spreadsheets and cybersecurity and I could work in the field one day, I might even go on to uni to learn more about infrastructure cybersecurity support or I might just go straight into work if I like my placement.

I’d definitely recommend anyone else finding maths hard [to] ask their school about tutoring because I know there are others like me that need maths to do what they want at college. Sometimes when you don’t get something in a big group setting it just means you need to learn it in a different way, there’s no reason you shouldn’t try again a second time.

Lily Bakewell, age 18 years, from Walsall

I was around 7 when my parents split up. I switched a lot between the 2 houses. My dad’s a plumber, my stepmum cleans in a hotel and my mum’s an accountant. When my parents split, we were quite badly off with money, but my mum would do anything she could to make it feel like we weren’t struggling.

At school, I found it hard to make friends. I was shy, and found it difficult to answer questions in class. But in year 9, I had a lovely health and social teacher, who said I had talent and was really passionate about persuading me to go into it as a career. She was an angel sent to earth. It gave me direction.

Although I had been planning to do A levels, I decided to do a T level in health and social care at Walsall College. A lot of my family have suffered from mental health problems. I’ve been exposed to it from a young age. There is a lot of inspiration around me, because I’ve seen what a difference good help can make.

Obviously when you’ve got to make choices about your future, you’re still quite young. I did a lot of weighing up the pros and cons. During the whole application process into college, there was a lot of uncertainty about what the course would lead to. But I decided that to get into my dream career it would make more sense to do the T level.

In the first year, we learned about legislation and policy. You’re doing research, planning, presenting and we did a placement one day a week. I worked at Walsall hospital on an elderly care ward. Being 16 and being thrown into such a hard workplace was a big learning curve. You pretty much take on the role of a student nurse at university. Within the first 3 months, I had witnessed someone die. It’s a lot of responsibility.

When I started, I suffered from social anxiety. But being a nurse, there’s no getting out of communication. You have to communicate with other professionals, patients and family members. It helped me get out of my comfort zone and embrace the professional role.

I’ve just got my offers for university to study mental health nursing. I’m the first person in my entire family to go to university, so it’s a big deal. I wasn’t sure how many universities would accept such a new qualification, but it’s not been a problem. Most of us had 3 or 4 offers. I am over the moon.

Maria Diaz, age 27 years, from London

I decided I wanted to be a lawyer before my GCSEs, when I was about 13 or 14. I didn’t really have a reason - I didn’t know any lawyers and no-one in my family had been to university - but I was doing well in school and I just wanted to do something professional that my family would be proud of.

My parents came to the UK from Colombia when there was a lot of civil unrest there. My mum was an orphan who had to look after her younger sisters - she never had the chance to go to university or get an education, so she always wanted that for me. My mum always tells the story of my Year 3 teacher at Parents’ Evening saying ‘You have to send Maria to university!’ and I’m really proud to have done that.

My school was always really good at letting us know about opportunities that might be useful to us, and when I was in Year 12 they told me about a London law firm that offered a scholarship every year. I got it, and the firm gave me £5,000 for every year I was at university! I studied at the London School of Economics, and a lot of people there had the luxury of not having to work as they were being funded by their family. Not having that made it quite difficult, and that scholarship was so important in helping me get through university. I don’t think I would have managed it otherwise because I would just have had to work all the time!

A lot of the other students also had experience in networking, so they started off knowing which events to attend and how to apply for the right sort of vacation schemes. I wasn’t sure when to apply, or even what schemes to apply for - I had no idea what a good law firm was! I was just sitting down doing a lot of research by myself, so I felt quite behind my peers in that respect, especially when it meant lots of my friends left uni with training contracts and job offers already in place.

Law has really specific timings - you can’t just apply whenever you want - and if you fall outside the application cycle you have to wait until the next year. It would have cost me £16,000 to do the Legal Practice Course without sponsorship from a job, so I ended up working as a paralegal in a high street firm while applying. The Legal Practice Course opens up a lot of opportunities for you, so I knew people who just paid for it, even without a job, but that just wasn’t an option for me.

I ended up doing two different vacation schemes and getting job offers from both. I suffered from a lot of imposter syndrome at the beginning, because I was surrounded by people whose parents are lawyers, who have really grown up in a professional environment. When I started my job in 2020, having to work from home didn’t help - I couldn’t meet people in person so it felt much harder to network and develop good relationships.

I chose to work at Bryan Cave Leighton Paisner law firm because I really felt like they were focused on making the work environment as inclusive as possible. I really thrive when I see people that I can relate to at work who have similar life experiences to me, so I’m really proud to have inspired other people too - my younger sister has also now graduated, and there are other members of my family who want to go to university now because they’ve seen me do it!

Deborah Oxley, age 40 years, from Newcastle

I grew up in Newcastle with my mum, dad and 3 sisters. My dad drove and fixed lorries for a living. It was a working-class environment. Money was a bit tight.

School was very hard because with my dad’s job we moved around a lot. I went to 15 different schools. I was never a very good speller. I struggled quite a lot. I decided to be a silly teenager, messing around, truanting and getting in with the wrong crowd.

Because of that, I left school in year 11 with no education. I didn’t sit any GCSEs or anything like that. I went to work for a telesales company, selling kitchens and double glazing and working in cafes. I had my daughter at 19, then went on to have 2 more children with my now ex-husband. After that, I stayed at home to raise my family with the odd job here and there. The financial side of things was difficult.

Then my oldest daughter was like: “I’m going to college, do you want to come?” I wasn’t working or anything like that so I thought, “yeah I really would actually!” I did my English level 2 at Newcastle College. Then I thought I would give level 2 Healthcare a go and loved it. I’d never done an essay before. I’d never referenced anything. I didn’t even know what that was. Now I can reference anything!

I started working as a care worker last year alongside getting my level 3. I had up to 4 shifts a week and college was Monday to Wednesday, so some weeks I didn’t get a day off. It was hard but I did it. Early on, the manager called me into her office and said,“I’d like you to think about becoming a senior care worker”. And I was like, “really?” I’d never worked in care before. I thought: “I’ve found something I’m really good at!”

I work on a unit that currently has 23 residents. I have residents who have dementia or are end-of-life. What I do is personal care. I’ll get them washed, dressed, showered, bathed, talk to them, talk to the family. I do a lot of assisted feeding. There are some residents that don’t have anyone and I’ll go in and talk to them on my days off.

Originally, I wanted to be a midwife, but now I want to be a palliative care nurse. I’d never experienced death before I worked here, and now I feel like I deal really well in that situation. I want to make peoples’ last moments as comfortable as I can. I just have a feeling that’s what I’m supposed to be doing.

Financially, it’s made a massive difference. Now I have a full-time income and I’m hoping we can have our first holiday abroad. These are the kinds of things that make memories for my children. When I have a day off I’m able to say, “come on we’re going out! Let’s go ice skating! Let’s go bowling!” I’m just aiming higher and higher with every level I get.

1. Socio-economic background means the socio-economic situation of their parents. For example, this might be the parents’ occupational class, income or education. So for instance, when we talk about someone with a “higher professional background”, we mean that at least one of their parents had a higher professional occupation when this person was a child. ↩

2. Alice Sullivan and others, ‘Pathways from origins to destinations: stability and change in the roles of cognition, private schools and educational attainment’, 2020. Published on ONLINELIBRARY.WILEY.COM; Matt Dickson and others, ‘Early, late or never? When does parental education impact child outcomes?’, 2016. Published on ACADEMIC.OUP.COM. ↩

3. Raj Chetty and Nathaniel Hendren, ‘The impacts of neighbourhoods on intergenerational mobility II: county-level estimates’, 2016. Published on NBER.ORG. ↩

4. In England continuing participation in education, at least on a part-time basis, is now legally required until age 18 years. ↩

5. According to the Equality Act 2010, protected characteristics are age, disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, religion or belief, sex, sexual orientation, and race (including colour, nationality, and ethnic or national origin). It is against the law to discriminate directly against someone with any of these characteristics. ↩

6. See technical annex for an explanation of which characteristics are covered in each dataset. ↩

7. A metropolitan area is a highly populated urban area that often shares common infrastructure, industries and commercial centres. It often includes multiple large cities, such as Wolverhampton or Birmingham. For example, the West Midlands or Greater Manchester. ↩

8. Some datasets are broken down by gender, while others are broken down by sex. See in the technical annex how each dataset captures sex or gender. ↩

9. Depending on the data source used, some such as the Department for Education use gender whereas others such as the Labour Force Survey use sex. ↩

10. Universities with lower entry requirements. ↩

11. Ethnic capital is a sociological term meaning the trusting relationships which exist among those belonging to a particular ethnic or cultural group. ↩

12. The lack of harmonised education statistics across England, Wales, Scotland and Northern Ireland means that the only option at present is to have separate (non-comparable) measures for each of the 4 nations. If harmonised measures are not possible, we hope to present data for the separate nations in future years. However, the devolved nations do have similar examinations. Wales does GCSEs. Northern Ireland has the Nationals 4 and 5 and Scotland has National 3, 4 and 5, and also has Highers. ↩

13. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

14. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

15. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

16. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

17. International Territorial Levels are a code used to subdivide the UK geographically for statistical purposes. Office for National Statistics, ‘Territorial levels UK, international territorial levels’, 2021. Published on ONS.GOV.UK. ↩

18. International Territorial Level is a code used to subdivide the UK geographically for statistical purposes. Office for National Statistics, ‘Territorial levels UK, international territorial levels’, 2021. Published on ONS.GOV.UK. ↩

19. Due to the COVID-19 pandemic, the key-stage 2 assessments were cancelled in 2019 to 2020 and 2020 to 2021. ↩

20. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

21. International Territorial Level is a code used to subdivide the UK geographically for statistical purposes. Office for National Statistics, ‘Territorial levels UK, international territorial levels’, 2021. Published on ONS.GOV.UK. ↩

22. Pupils are defined as disadvantaged if they are known to have been eligible for free school meals at any point in the past 6 years (from year 6 to year 11), if they are recorded as having been looked after for at least 1 day or if they are recorded as having been adopted from care. ↩

23. Department for Education, ‘Key stage 4 performance revised’, 2023. Published on GOV.UK. ↩

24. See Department for Education guidance for more information on free school meal eligibility, ‘Early years foundation stage profile results’, 2022. Published on GOV.UK. ↩

25. International Territorial Level is a code used to subdivide the UK geographically for statistical purposes. Office for National Statistics, ‘Territorial levels UK, international territorial levels’, 2021. Published on ONS.GOV.UK. ↩

26. Social Mobility Commission, ‘State of the Nation 2022: A fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

27. Anthony Heath and others, ‘Social progress in Britain’, 2018. Published on GLOBAL.OUP.COM. ↩

28. Lindsey Macmillan, ‘Intergenerational worklessness in the UK and the role of local labour markets’, 2014. Published on ACADEMIC.OUP.COM ↩

29. Office for National Statistics, ‘Young people not in education, employment or training (NEET), UK: May 2022’, 2022. Published on ONS.GOV.UK. ↩

30. House of Commons Library, ‘NEET: young people not in education, employment or training’, 2021. Published on COMMONSLIBRARY.PARLIAMENT.UK. ↩

31. Anthony Heath and others, ‘Unequal attainments: ethnic educational inequalities in ten western countries’, 2014. Published on ACADEMIC.OUP.COM. ↩

32. Patricia Daley, ‘Black-African: students who stayed’, in Ceri Peach (editor) ‘Ethnicity in the 1991 census, volume 2’,1996. Published by OFFICE FOR NATIONAL STATISTICS. ↩

33. Labour Force Survey, ‘User guides, volumes 3 and 4’, 2023. Published on ONS.GOV.UK. ↩

34. Jack Britton and others, ‘Which university degrees are best for intergenerational mobility?’, 2021. Published on IFS.ORG.UK. ↩

35. Social Mobility Commission, ‘State of the Nation 2022: a fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

36. Vikki Boliver, ‘How fair is access to more prestigious UK universities?’, 2013. Published on ONLINELIBRARY.WILEY.COM. ↩

37. Vikki Boliver, ‘How fair is access to more prestigious UK universities?’, 2013. Published on ONLINELIBRARY.WILEY.COM. ↩

38. Social Mobility Commission, ‘Labour market value of higher and further education qualifications: a summary report’, 2023. Published on GOV.UK. ↩

39. Jack Britton and others, ‘Which university degrees are best for intergenerational mobility?’, 2021. Published on IFS.ORG.UK. ↩

40. See figure 3.10 in Social Mobility Commission, ‘State of the Nation 2022: A fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

41. Albert Halsey and others, ‘Origins and destinations: family, class and education in modern Britain’, 1980. Published on CAMBRIDGE.ORG. ↩

42. Samuel Lucas, ‘Effectively maintained inequality: education transitions, track mobility, and social background effects,’ 2001. Published on JOURNALS.UCHICAGO.EDU. ↩

43. Jung In and Richard Breen, ‘Social origins and access to top occupations among the highest educated in the United Kingdom’, 2022. Published on JOURNALS.SAGEPUB.COM. ↩

44. Vikki Boliver, ‘How fair is access to more prestigious UK universities?’, 2013. Published on ONLINELIBRARY.WILEY.COM. ↩

45. Anthony Heath and others, ‘Unequal attainments: ethnic educational inequalities in ten western countries’, 2014. Published on ACADEMIC.OUP.COM. ↩

46. Yaojun Li and Anthony Heath, ‘Class matters: a study of minority and majority social mobility in Britain, 1982–2011’, 2016. Published on JOURNALS.UCHICAGO.EDU. ↩

47. Mary Waters and others, ‘Second-generation attainment and inequality: primary and secondary effects on educational outcomes in Britain and the US’, 2013. In Richard Alba and Jennifer Holdaway, ‘The children of immigrants at school: a comparative look at education in the United States and Western Europe’, 2013. Published on NYUPRESS.ORG. ↩

48. Labour Force Survey, ‘[User guides, volumes 3 and 4’, 2023. Published on ONS.GOV.UK. ↩

49. Correlation is a measure of how much one variable moves with another. A positive correlation means as one variable moves another tends to move in the same direction. A negative correlation implies the variables tend to move in opposite directions. A correlation of 0 or close to 0 means that as one variable moves another does not tend to move. ↩

50. With a correlation coefficient of 0.56 which is significantly different from 0 at the 0.1% level. ↩

51. All the correlation coefficients listed here are significantly different from 0 at the 5%. ↩

52. Social Mobility Commission, ‘State of the Nation 2022: A fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

53. Paul Gregg and Emma Tominey, ‘The wage scar from male youth unemployment’, 2005. Published on RESEARCHPORTAL.BATH.AC.UK. ↩

54. Yaojun Li and Anthony Heath, ’Persisting disadvantages: a study of labour market dynamics of ethnic unemployment and earnings in the UK (2009-2015)’, 2018. Published on TANDFONLINE.COM. ↩

55. Lindsey Macmillan, ‘Intergenerational worklessness in the UK and the role of local labour markets’, 2014. Published on ACADEMIC.OUP.COM; ↩

56. Anthony Heath and Jean Martin, ‘Can religious affiliation explain ethnic inequalities in the labour market?’, 2010. Published on TANDFONLINE.COM. ↩

57. Nabil Khattab and Tariq Modood, ‘Both ethnic and religious: explaining employment penalties across 14 ethno‐religious groups in the United Kingdom’, 2015. Published on ONLINELIBRARY.WILEY.COM. ↩

58. Labour Force Survey, ‘User guides, volumes 3 and 4’, 2023. Published on ONS.GOV.UK. ↩

59. Social Mobility Commission, ‘State of the Nation 2022: a fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

60. Anthony Heath and others, ‘Social progress in Britain’, 2018. Published on GLOBAL.OUP.COM. ↩

61. David Bell and David Blanchflower, ‘Young people and the Great Recession’, 2011. Published on ACADEMIC.OUP.COM. ↩

62. Yaojun Li and Anthony Heath, ‘Class matters: a study of minority and majority social mobility in Britain, 1982–2011’, 2016. Published on JOURNALS.UCHICAGO.EDU. ↩

63. Anthony Heath and Valentina Di Stasio, ‘Racial discrimination in Britain, 1969–2017: a meta-analysis of field experiments on racial discrimination in the British labour market’, 2019. Published on ONLINELIBRARY.WILEY.COM. ↩

64. Labour Force Survey, ‘User guides, volumes 3 and 4’, 2023. Published on ONS.GOV.UK. ↩

65. The 4 categories are: never worked or unemployed, working class, intermediate and professional. ↩

66. Social Mobility Commission, ‘State of the Nation 2022: a fresh approach to social mobility’, 2022. Published on GOV.UK. See figures 3.14 and 3.15. ↩

67. We did not include a category for ‘never worked or unemployed’ as this is covered by intermediate outcome 3.1 and 3.2. ↩

68. For a recent analysis see Erzsébet Bukodi and John Goldthorpe, ‘Social mobility and education in Britain: research, politics and policy’, 2018. Published on CAMBRIDGE.ORG. ↩

69. Erzsébet Bukodi and John Goldthorpe, ‘Social mobility and education in Britain: research, politics and policy’, 2018. Published on CAMBRIDGE.ORG. ↩

70. Albert Halsey, ‘Twentieth-century British social trends’, 2000. Published on LINK.SPRINGER.COM. ↩

71. Paul Gregg and others, ‘Moving towards estimating sons’ lifetime intergenerational economic mobility in the UK’, 2016. Published on ONLINELIBRARY.WILEY.COM. ↩

72. The Longitudinal Education Outcomes linked dataset “connects individuals’ education data with their employment, benefits and earnings data to create a de-identified person level administrative dataset.” Taken from ‘About the LEO standard extract’. Published on GOV.UK. ↩

73. Chris Belfield and others, ‘The impact of undergraduate degrees on early-career earnings’, 2018. Published on IFS.ORG.UK. ↩

74. Jack Britton and others, ‘How much does it pay to get good grades at university?’, 2022. Published on IFS.ORG.UK. ↩

75. Ian Walker and Yu Zhu, ‘Differences by degree: Evidence of the net financial rates of return to undergraduate study for England and Wales’, 2011. Published on SCIENCEDIRECT.COM. ↩

76. Ian Walker and Yu Zhu, ‘University selectivity and the relative returns to higher education: evidence from the UK’, 2018. Published on SCIENCEDIRECT.COM. ↩

77. Jung In and Richard Breen, ‘Social origin and access to top occupations among the ↩

78. The results come from basically the same regression models as in the case of returns to education. The main difference is that in the former analyses we formally tested for interactions between each protected characteristic and educational level, whereas in the current analyses we test for interactions between protected characteristics and social background. However, there were very few significant interactions between protected characteristics and social background and we therefore show here the estimates for models without interactions. ↩

79. Due to small sample sizes, this analysis used 3 socio-economic background (SEBs) classes to obtain more accurate estimates. ↩

80. A formal test using log linear modelling shows that the changes for the 3 different SEBs are not significantly different in size from each other. ↩

81. The Institute for Social and Economic Research was unable to produce analyses of young people’s acquisition of additional educational qualifications using the UK Household Longitudinal Survey for State of the Nation 2022. We now conduct age, period, cohort analysis with the Labour Force Survey to obtain some results. ↩

82. We choose the age of 25 years as the starting point as by this age the great majority of young people will have completed full-time education and entered the labour market. 2014 is the earliest date at which the Labour Force Survey includes a measure of parental background, while 2021 is the most recent. So by choosing ages 25 and 32 years we maximise the length of career that can be covered with this analysis. ↩

83. Muriel Egerton, ‘Mature graduates I: occupational attainment and the effects of labour market duration’, 2001. Published on TANDFONLINE.COM; ↩

84. Erzsébet Bukodi, ‘Cumulative inequalities over the life-course: life-long learning and social mobility in Britain’, 2016. Published on CAMBRIDGE.ORG. ↩

85. Erzsébet Bukodi, ‘Cumulative inequalities over the life-course: life-long learning and social mobility in Britain’, 2016. Published on CAMBRIDGE.ORG. ↩

86. Erzsébet Bukodi and others, ‘Changing career trajectories of women and men across time’, 2012. Published on ELGARONLINE.COM. In Jacqueline Scott and others, (eds) ‘Gendered Lives: Gender inequalities in Production and Reproduction’, 2013. Published on E-ELGAR.COM. ↩

87. Another possible explanation is that older respondents will have entered the labour market in earlier years, when there will have been fewer openings in the professional classes. This could explain the lower achievements of older respondents if our early career position impacts our later position. In other words, expanding opportunities in the professional classes might benefit younger more than older entrants into the labour market. ↩

88. Erzsébet Bukodi and John Goldthorpe, ‘Class origins, education and occupational attainment in Britain: secular trends of cohort-specific effects?’, 2011. Published on TANDFONLINE.COM. ↩

89. Monica Costa Dias and others, ‘Wage progression and the gender wage gap: the causal impact of hours of work’, 2018. Published on IFS.ORG.UK. ↩

90. These results are not derived from a panel study in which the same respondents are re-interviewed yearly but from pooled annual surveys interviewing different respondents each year. If we use the same birth cohort approach as for intermediate outcome 4.2 looking at the earnings of those born in 1989 between the ages of 25 and 44 years, we obtain a similar story of widening earnings gaps between young people from professional and working-class backgrounds. ↩

91. These results are not derived from a panel study in which the same respondents are re-interviewed each year but from pooled annual surveys interviewing different respondents each year. If we use the same birth cohort approach as for intermediate outcome 4.2, looking at the earnings of those born in 1989 between the ages of 25 and 32 years, we obtain a similar story of widening earnings gaps between young people from professional and working-class backgrounds. ↩

92. Monica Costa Dias and others, ‘Wage progression and the gender wage gap: the causal impact of hours of work’, 2018. Published on IFS.ORG.UK. ↩