Policy paper

Chapter 2: Mobility outcomes

Published 12 September 2023

Highlights

The total occupational mobility rate – that is, the number of people in a different occupational class from their parents – has remained fairly constant for many decades. However, the surplus of upward over downward mobility is shrinking, as the growth in professional jobs has slowed.

In the most recent data, 52% of people from a lower working-class background have gone on to work in professional or intermediate jobs. For people from a higher working-class background, 61% have done so.

Adults in the UK with lower working-class parents are about 3 times as likely to be in a lower working-class occupation themselves (30%) as adults with higher professional parents (11%).

Women are less likely than men to experience upward occupational mobility. For example, only 8% of women moved from a lower working-class background to a higher professional job, compared with 14% of men.

There are very different mobility trends across ethnic groups. For example, Indian-British adults from a working-class background are about twice as likely to be in professional jobs (44%) as Bangladeshi-British adults from the same background (23%).

People who grew up in Outer London West and North West (46%) or Surrey and Sussex (40%) have the greatest likelihood of long-range upward mobility (from the working classes to the professional classes).^{[footnote 1]} People who grew up in Northern Ireland have the lowest (28%).

Absolute income mobility in the UK was good for people born in the mid-1970s, but has since declined.^{[footnote 2]} Just under 70% of those born in the mid-1980s have gone on to earn more than their parents did at a similar age.

Compared with almost 20 years ago, relative income mobility – the strength of the link between parents’ and children’s income – has been roughly stable.

People’s highest level of qualification is strongly related to the level of education that their parents achieved. Most people whose parents went to university go on to gain a degree (64%). Only 18% of those whose parents have no qualifications go on to gain a degree.

Upward educational mobility – measured as children obtaining degrees whose parents did not – is much more common among certain ethnic groups. For example, 64% of Chinese-British people whose parents had no degree went on to obtain a degree, compared with only 28% among White British people. It is also more common in certain regions: 39% in London, for example, and only 22% in the East Midlands.

Since the 1990s, there has been a considerable improvement in relative educational mobility. This means that people’s chances of obtaining a degree have become less related to whether their parents had a degree or not.

People whose parents owned their own home are much more likely to own their own home (71%) compared with those whose parents did not own their own home (46%).

Relative housing mobility – the link between parents’ home ownership and their children’s home ownership – has worsened consistently and significantly since 1991. This means that the link is now much stronger – parental home ownership is a better predictor of children’s home ownership. The link is also significantly stronger between women and their parents, than between men and their parents.

As we might expect, there is a link between parents’ and children’s wealth. A 10% increase in parents’ wealth is associated with around a 3% increase in their children’s wealth at a similar stage of life. Although data is limited, it seems plausible that relative wealth mobility is worsening along with relative housing mobility.

The mobility outcomes of people with a disability are consistently worse than the outcomes of those without, across occupation, education and housing.

Introduction

Types of social mobility measures

An individual experiences intergenerational social mobility when their life outcomes, such as their type of occupation, differ from their parents’. Change across generations, and the link between parents and children, are the core of social mobility. Change can be upwards or downwards.

Success in one outcome doesn’t guarantee success in another, and there is good evidence that shows that intergenerational links are strongest in the same types of outcome.^{[footnote 3]} For example, parents’ educational advantage or disadvantage seems most strongly linked to their children’s educational outcomes, and the same goes for occupation and income.

Additional research has shown that wealth is no exception to this rule, and it has suggested that education, occupation, income and wealth are the ‘big 4’ dimensions of intergenerational social mobility. This means that advantage, or disadvantage, seems to be transmitted across one of these ‘big 4 channels’.^{[footnote 4]}

Housing mobility is not included as one of the 4, as it is a major component of wealth for most people. However, we include it in this report. Data on wealth mobility is a problem for the UK, but we also make a first attempt to deal with it in this report.

Focusing on these 5 mobility outcomes – occupation, income, education, housing and wealth – gives us a more complete picture of individuals.

For occupation, income, education and housing mobility, we report on absolute and relative mobility rates, and also provide analysis by sex, ethnicity, disability and region, where permitted.

We should note that there has been little previous research in Britain on either education or housing mobility. We include them in this report for the first time.

Absolute and relative mobility

Absolute measures capture the number of people who have experienced mobility. They are usually expressed as percentages of the population. For example, the absolute occupational mobility rate is the percentage of people who are in a different occupational class from their parents. For income mobility, a common absolute measure is the percentage of people whose income is higher than their parents’ income was, at the same age. We can compare these rates across different regions of the UK.

Relative measures compare the chances that at least 2 groups have of reaching, versus avoiding, a particular outcome. It is this element of comparison that makes such measures relative. A relative mobility measure tells us that one group has better chances than another, rather than telling us the total number of socially mobile people. Low relative mobility means that those who start life in a particular position are more likely than others to be in the same position later in life. For that reason, low relative mobility can be thought of as ‘stickiness’, while high relative mobility can be thought of as ‘fluidity’.

Social mobility definitions

Upward social mobility means doing better than our parents, whether that’s in income, occupation, or other outcomes. How much better depends partly on effort and talent, and partly on starting point or background.

Higher absolute upward mobility means that more people are moving up in life. But our starting point still makes a difference – those who started further back are disadvantaged.

Higher relative mobility  means that our starting point matters less. The absolute upward mobility hasn’t changed, but people at the back are now relatively better off.

Odds ratios

An odds ratio can be interpreted as the outcome of a competition between people from 2 different origins to achieve a particular outcome and avoid the alternative outcome. It is a standard measure of relative mobility, as it is independent of changes in the distributions.

For example, in figure 2.26, we see that 71% of people whose parents were homeowners became owners themselves, compared with only 46% of people whose parents were renters. But what if we check the relative chances of becoming a renter? As expected, people whose parents were renters have a higher chance of being a renter themselves, but the numbers are different – 54% became renters themselves, compared with 29% of people whose parents were homeowners.

All 4 of these percentages could change in different directions over time. So we need a measure of the combined inequality in becoming an owner rather than a renter, taking into account all 4 percentages. The most commonly used method is to calculate the ratio of the odds, or the ‘odds ratio’.

Calculating the odds ratio is done as follows. If we think only about the children of homeowners, 71% of them became homeowners themselves, while 29% became renters. This gives odds of 71 to 29. For the children of renters, it was 46 to 54. We then divide the first set of odds by the second. 71:29 divided by 46:54 roughly equals 2.9. This figure of 2.9 is the odds ratio.

We can now see that all 4 percentages have been included in the calculation. For example, if the percentage of homeowners whose parents were homeowners (71%) were to increase, then the odds ratio of 2.9 would also have to increase, showing increased inequality. Or if the percentage of homeowners whose parents were renters (46%) were to increase, then the odds ratio would decrease, showing decreased inequality.

What are we doing this year?

Much research on social class mobility has been based on large-scale representative national surveys, such as the Labour Force Survey (LFS) and the UK Household Longitudinal Survey (UKHLS), also known as Understanding Society.

Respondents are typically asked to provide the information necessary to measure their own current occupational class position or educational qualifications. They are also asked to report the same kind of information about their parents’ occupational class and educational qualifications.

These datasets are rich, and provide large sample sizes that allow for some regional and intersectional analysis.^{[footnote 5]} However, the main social mobility questions, particularly those from the LFS, don’t go back far enough in time for us to have a large enough sample to provide a clear time series of mobility trends. We also need breakdowns by geography and protected characteristics.^{[footnote 6]}

In 2022, we focused on how mobility outcomes changed over time. We showed the intergenerational mobility patterns experienced by successive cohorts of people born throughout the 20th century. This year, we focus on comparing outcomes across regions and by groups of people according to gender or sex, ethnicity, and disability.

Summary of findings

Measuring mobility outcomes is challenging, especially in the case of income mobility where we have to rely on a small number of long-term panel studies for appropriate measures of parental income. While the inclusion of detailed questions on parental occupations in the LFS helps us to study occupational mobility and the Wealth and Assets Survey (WAS) enables the study of housing mobility, there are still serious data gaps for both education and wealth mobility.

There are also complex technical issues which limit how well we can use these data sources to understand social mobility. Some of these challenges include: recall bias, life-cycle bias, attrition and attenuation.^{[footnote 7]} These need to be addressed when analysing the available data, especially when looking at differences across ages. We hope to tackle this in future reports.

Nevertheless, the main findings are as follows. While there are a few shared patterns across the different aspects and dimensions of mobility, the difference in results is more striking. It is important not to over-simplify what is actually a complex set of results. This complexity is not surprising since we are comparing absolute and relative levels of mobility, different dimensions of mobility – occupational, educational, income, housing and wealth mobility – and we are investigating intersectionality with gender or sex, disability, ethnicity and region of the UK.

Broadly speaking:

• with respect to trends over time and across regions, housing mobility looks very different from the other main dimensions
• for absolute mobility, the most common pattern is to find more upward than downward mobility, although the contrast was more marked in previous decades than it is today
• for relative mobility, there continue to be quite high levels of intergenerational persistence in most domains, although the trends over time differ – in the case of housing mobility, there has been a marked increase in persistence in contrast to educational mobility where there has been a marked decrease in persistence
• differences between sexes vary from one type of mobility outcome to another
• on disability, we uniformly find that people with a long-term illness or disability are substantially disadvantaged in all domains (in absolute terms)
• in the case of ethnicity, patterns vary markedly from one group to another although absolute educational mobility tends to be significantly higher for many ethnic minorities compared with White British people, while absolute occupational mobility tends to be significantly lower for some minorities than for White British people
• when looking at geography, London and the South East appear to be particularly advantaged in terms of absolute upward educational and occupational mobility, but particularly disadvantaged in terms of absolute housing mobility – it is however quite difficult to detect significant differences between regions in terms of relative mobility (partly due to small sample sizes and high imprecision in the estimates), and most areas of the UK are quite similar in all domains

Occupational mobility

Occupational mobility has historically been the focus of social mobility research, for good reason. First, occupational mobility captures the link between parents’ occupational class and their children’s, providing a snapshot of generational outcomes in terms of type of jobs. Second, occupations are associated with a wide range of important life outcomes, including income, employment conditions and security, risks of unemployment, and health and wellbeing. This provides rich insights about mobility more generally.

Short-range occupational mobility means moving from one broad occupational category to an adjacent one. For example, moving from an intermediate origin to a working-class or professional occupation would be short-range mobility, as would moving from working class to intermediate, or professional to intermediate.

Long-range occupational mobility means moving either from a working-class origin to a professional occupation, or a professional origin to a working-class occupation.

Absolute occupational mobility

Absolute mobility concerns whether people have a higher or lower occupation level than their parents. Upward absolute occupational mobility can be measured by the proportion of people who have jobs in a higher occupational class than their parents did at a similar stage of life.

We define socio-economic background (SEB) as the occupation of the main earner in the respondent’s household when the respondent was aged 14 years. As mentioned earlier, we use 5 categories:

• higher professional and managerial
• lower professional and managerial
• intermediate
• higher working class
• lower working class

These are defined based on the National Statistics Socio-economic Classification (NS-SEC), which is the official socio-economic classification of the UK, as set by the Office for National Statistics (ONS).^{[footnote 8] [footnote 9]} This classification is used widely to understand the structure of socio-economic positions in society.

As we note in chapter 1, occupational class is not about salary. People in lower occupational classes can sometimes earn more than people in higher occupational classes.

In figures 2.0 to 2.3, we show long-term trends in rates of absolute occupational class mobility. We compare the experience of cohorts of men and women born in the decades from the 1910s to the 1990s.^{[footnote 10]} The oldest of these cohorts entered the labour market before World War 2 and the youngest entered the labour market 70 years later, in the 21st century. The most recent cohorts are still early in their working lives, and this could influence the results.

For both men and women, the percentage of upwardly mobile people has always been higher than downwardly mobile. This surplus of upward over downward mobility is a consequence of the changing shape of the occupational structure, with increasing room at the top, especially in the middle part of the period covered by figure 2.0.

In the case of men, the gap between the 2 curves – the surplus of upward over downward mobility – has been gradually shrinking, as upward mobility has declined and downward mobility increased. This shrinking could be due to the younger age of the latest birth cohorts, because people’s likelihood of being upwardly mobile increases as they get older, up until about age 40 years (see figure 3.56). But this is unlikely to fully explain it. More detailed research by Bukodi and Goldthorpe (2019) which studied recent trends across cohorts at the same age, confirms that upward mobility has been declining and downward mobility increasing.^{[footnote 11]}

One explanation for the closing of the gap is that there are now more people at risk of downward mobility, because of the expansion of the professional classes in their parents’ generation. Similarly, there are now fewer people from working-class backgrounds, so fewer people are in a position to move upwards. So what we are seeing is partly a consequence of the demographic changes arising from the great expansion of the professional classes in the 1960s and 1970s.

There is a third element that we need to take into account. For upward mobility to continue, we need more and more professional jobs to be created. The shape of the occupational structure is now changing at a slower pace than it was during the second half of the 20th century, and fewer professional jobs are being created. So, even though the chances of a man of working-class origin reaching a professional job improved greatly over the 20th century, we will still need to create more professional jobs for this improvement to continue.

Figure 2.0: There continues to be more upward than downward mobility for men, but the size of this surplus has been shrinking in recent decades and chances of long-range upward mobility have been declining

Percentages of men experiencing occupational mobility (upward, downward, and total), by birth cohort, UK, data collected from 1972 to 2022.

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

Source: The General Household Survey (1972 to 2005), British Household Panel Survey (1991 to 2009), Taking Part Survey (2005 to 2006), Understanding Society (2010 to 2019) and Labour Force Survey (2014 to 2022), male respondents aged 25 to 65 years.

Note: The figures for total mobility are the sum of the percentages upwardly and downwardly mobile. This represents the percentage of the sample as a whole who were in a different social class position from the one in which they were brought up (based on 5 social classes: professional, intermediate, own account, skilled manual, unskilled manual. Classes differ from those used elsewhere in this report, due to data availability).

Figure 2.1: The chances of a man of working-class origin reaching a professional job improved greatly over the 20th century. The cohorts born in the 1980s or 1990s have not yet reached occupational maturity, so these numbers will tend to be lower.

Percentages of men of working-class origin in professional jobs, and men of professional origin in working-class jobs, by birth cohort, UK, data collected from 1972 to 2022.

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

Source: The General Household Survey (1972 to 2005), British Household Panel Survey (1991 to 2009), Taking Part Survey (2005 to 2006), Understanding Society (2010 to 2019) and Labour Force Survey (2014 to 2022), male respondents aged 25 to 65 years.

Note: The figures for long-range upward mobility are percentages of those from a working-class background who went on to work in a professional occupation. The figures for long-range downward mobility are the opposite, that is, percentages of those from a professional background who went on to work in a working-class occupation. Analysis is based on 5 social classes: professional, intermediate, own account, skilled manual, unskilled manual. Classes differ from those used elsewhere in this report, due to data availability.

The picture is broadly similar among women, although the process began somewhat later. Among the oldest cohorts of women, in contrast to the picture for men, there was more downward than upward mobility. It was only among the cohort born in the 1930s, who would have been entering the labour market after the war, that overall upward mobility actually overtook downward mobility. This may be connected with the lack of employment opportunities for women in the inter-war period, particularly in higher-level professions and managerial work.^{[footnote 12]}

There appears to be a greater decline in the total rate of mobility among women than men after the 1950s. This is due to the downward mobility trends differing between men and women: men saw a bigger increase in downward mobility than women did over this period. This may be linked to the shift from an industrial to a post-industrial occupational system and a decline in gender segregation at work.

Figure 2.2: There continues to be more upward than downward mobility for women, but the size of this surplus has been shrinking in recent decades, as has the total percentage who are mobile.

Percentages of women experiencing occupational mobility (upward, downward, and total), by birth cohort, UK, data collected from 1972 to 2022.

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

Source: The General Household Survey (1972 to 2005), British Household Panel Survey (1991 to 2009), Taking Part Survey (2005 to 2006), Understanding Society (2010 to 2019) and Labour Force Survey (2014 to2022), female respondents aged 25 to 65 years.

Note: The figures for total mobility are the sum of the percentages upwardly and downwardly mobile. This represents the percentage of the sample as a whole who were in a different social class position from the one in which they were brought up (based on 5 social classes: professional, intermediate, own account, skilled manual, unskilled manual. Classes differ from those used elsewhere in this report, due to data availability).

Figure 2.3: The chances of a woman of working-class origin reaching a professional job improved greatly over the 20th century. The cohorts born in the 1980s or 1990s have not yet reached occupational maturity, so these numbers will tend to be lower.

Percentages of women of working-class origin in professional jobs, and women of professional origin in working-class jobs, by birth cohort, UK, data collected from 1972 to 2022.

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

Source: The General Household Survey (1972 to 2005), British Household Panel Survey (1991 to 2009), Taking Part Survey (2005 to 2006), Understanding Society (2010 to 2019) and Labour Force Survey (2014 to 2022), female respondents aged 25 to 65 years.

Note: The figures for long-range upward mobility are percentages of those from a working-class background who went on to work in a professional occupation. The figures for long-range downward mobility are the opposite, that is, percentages of those from a professional background who went on to work in a working-class occupation. Analysis is based on 5 social classes: professional, intermediate, own account, skilled manual, unskilled manual. Classes differ from those used elsewhere in this report, due to data availability.

Figure 2.4 shows the occupational class position of people aged 25 to 64 years in the UK by the occupational class of their parents. Overall we find substantial upward occupational mobility. Among those from a lower working-class background, around 70% experienced either short or long-range upward mobility, and 32% experienced long-range upward mobility into the professional classes. In contrast, only 18% (7% plus 11%) of people from higher-professional backgrounds experienced long-range downward mobility into the working classes. This surplus of upward over downward mobility reflects the great expansion of professional and managerial employment over the last decades of the 20th century.

Figure 2.4: People’s occupational class position is strongly related to their socio-economic background.

Occupational class position of respondents aged 25 to 64 years in the UK, 2022, by highest level of parental occupational class.

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

Source: Office for National Statistics, Labour Force Survey (LFS) 2022, respondents aged 25 to 64 years in the UK, data collected from July to September 2022.

Notes: Parental social class in the LFS is measured by asking respondents to recall the occupation of the main wage earner in their household when they were aged 14 years. A formal chi-square test shows that the relationship is statistically significant at the 0.001 level. Due to rounding errors, in some instances the totals may not add up to 100%

However, substantial inequalities continue in mobility outcomes. Respondents from higher professional backgrounds were by far the most likely to have higher professional positions: 34% were in the higher professional class, around 3 times as many as from the lower working class (11%).

This picture is closely in line with that from other recent research and previous State of the Nation reports.^{[footnote 13] [footnote 14] [footnote 15]}

Geographical analysis of absolute occupational mobility

By pooling all years of the LFS between 2018 and 2022, we have been able to look at absolute occupational mobility across regions. We compare the chances of people from different regions of the UK, but from the same SEB, of getting into the professional and managerial classes. Figure 2.5 shows that people who grew up in London and the south-east tend to have better chances of upward occupational mobility and those who grew up in the north and the south-west have the poorest . There are elements both of centre/periphery and north/south divisions although either division on its own is an oversimplification of a more complex reality.  This probably reflects the presence of several different causes of unfavourable mobility chances, as we describe in our section on the drivers of mobility.

Figure 2.5: Those who grew up in London and the south-east have the best upward occupational mobility rates, and those who grew up in the north and the south-west have the worst.

Chances of having a professional class position in the UK, 2022, by International Territorial Level 2 regions, controlling for socio-economic background (SEB).

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

Source: Office for National Statistics, pooled Labour Force Survey (LFS) 2018 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Note: The data used is weighted using the LFS probability weights. The data was analysed using a multilevel logistic regression model, controlling for SEB, in order to achieve more reliable estimates and avoid the risk of obtaining extreme outliers by chance.^{[footnote 16]} We have also checked the patterns by estimating rates of long-range upward mobility from working class origins to professional class destinations.

Intersectional analysis of absolute occupational mobility

We now look at how levels of absolute occupational mobility differ across sex, ethnicity, disability and place.^{[footnote 17]}

Differences between men and women

It is not surprising that we see important differences across sexes. Women are significantly less likely to be in the higher professional class than men from the same SEB.

We see this trend in figure 2.6. Among people from higher-professional backgrounds, 40% of men and only 27% of women are in the higher-professional class. Women from higher-professional backgrounds are also more likely than men to be in the lower-working class. 12% of women from higher-professional backgrounds are in lower working-class jobs, compared with only 9% of men from the same SEB (a statistically significant difference). This means that women are more likely to experience long-range downward mobility than men.

Figure 2.6: Women’s chances of being in the higher-professional class are poorer than the chances of men from the same socio-economic background.

Occupational class position of respondents aged 25 to 64 years in the UK, 2022, by highest level of parental occupational class and sex.

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

Source: Office for National Statistics, Labour Force Survey (LFS) 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: Differences between men and women in access to the higher professional class are statistically significant at the 1% level. 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%.

A range of explanations for this discrepancy have been suggested. Potentially important explanations are that women are more likely than men to take on caring responsibilities and to work part-time at some stage of their lives, and may therefore find it harder to make progress in their careers. Even for full-time workers, some high-level positions may not provide the time flexibility needed for coping with family emergencies (or may insist on long hours that do not fit with routine family responsibilities).^{[footnote 18]}

Differences between ethnic groups

We also find differences across ethnic groups. In figure 2.7, we find that people of Indian (44%) and Chinese (46%) backgrounds have significantly higher chances of long-range upward mobility than their White British peers, possibly reflecting their high rates of upward educational mobility. However, evidence shows that these patterns vary across generations. For example, it is common that first-generation migrants experience a notable social decline, while their children – second-generation migrants – experience advancement.^{[footnote 19]} This means that mobility rates, in turn, will be affected by the balance of first and second (or later) generation immigrants in that particular ethnicity.

Figure 2.7: People with Indian and Chinese ethnic backgrounds have a greater likelihood of long-range upward mobility from the working classes than other groups.

Proportion of respondents aged 25 to 64 years in the UK, from 2014 to 2022, experiencing long-range upward and downward mobility by ethnic group.

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

Source: Office for National Statistics, pooled Labour Force Survey 2014 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: The blue bars (long-range upward occupational mobility) represent the proportion of respondents from a working-class background who are now in a professional occupation. The red bars (long-range downward occupational mobility) represent the proportion of respondents from a professional background who are now in a working-class occupation. “Other” are people whose self-reported ethnicity is not 1 of the first 9 categories. The results for the Indian, Chinese, Bangladeshi and Pakistani groups are significantly different from the White British results at a 0.01 level of significance. The error bars show 95% confidence intervals.

We also see that there are some high rates of long-range downward mobility among ethnic minorities. For example, people from the Bangladeshi (45%) and Pakistani (40%) groups have the highest proportion of long-range downward mobility. This implies that some ethnic minority families may not be so able as White British families to protect their children from downward mobility.

Differences by disability status

Figure 2.8 shows that people from all SEB who have a limiting long-term illness or disability are much less likely than people without a disability to enter the professional classes. They are also more likely to have a lower working-class position. Extraordinarily high percentages of people with a disability hold lower working-class positions – from over 25% of those from higher-professional backgrounds to over 50% of those from lower working-class backgrounds. Long-range downward mobility rates are greater than upward mobility rates for people with a disability, in marked contrast to the general picture for occupational mobility. For example, only 16% of people with a disability were upwardly mobile into the professional classes compared with 38% from the higher-professional class who were downwardly mobile into the working classes.^{[footnote 20]}

Figure 2.8: People with a disability or long-term limiting condition have a much lower likelihood of long-range upward mobility and much higher chances of downward mobility than the population as a whole.

Proportion of respondents aged 25 to 64 years in the UK, from 2014 to 2022, experiencing long-range upward and downward mobility by limiting long-term conditions.

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

Source: Office for National Statistics, pooled Labour Force Survey (LFS) 2014 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

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 the LFS user guides volumes 3 and 4).^{[footnote 21]} The percentages of people with a disability who experienced long-range upward or downward mobility are all significantly different from the percentages without a disability at the 1% level. 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 regions

Where you grow up is also associated with how you get ahead in terms of occupational class. We see this in figure 2.9 where we look at differences in absolute occupational mobility by region, with a similar pattern to figure 2.5. London, and Surrey and Sussex (East and West), provide particularly high rates of long-range upward mobility. In contrast, Cornwall and Isles of Scilly, East Yorkshire and North Lincolnshire, and the Highlands and Islands of Scotland tend to have lower rates of long-range upward mobility or higher rates of long-range downward mobility. This suggests a distinction between the centre and the periphery. Areas of the central mainland tend to have higher rates than those that are coastal.

Figure 2.9: People who grew up in Outer London or Surrey and Sussex have the greatest likelihood of long-range upward mobility from the working classes.

Proportion of respondents aged 25 to 64 years in the UK, 2022, who experienced long-range upward and downward mobility, by area in which they grew up.

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

Source: Office for National Statistics, pooled Labour Force Survey 2018 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: The figure represents the proportion of people from a working-class background belonging to the professional classes (long-range upward mobility) and the proportion from a professional-class background belonging to the working classes (long-range downward mobility) across each region. The error bars show 95% confidence intervals.

Relative occupational mobility

Relative mobility compares the mobility chances of people from different social backgrounds. It focuses on the chances of someone from one SEB attaining a given occupational class compared with someone from a SEB. Measures of relative mobility can be thought of as describing the strength of the intrinsic link (or ‘stickiness’) between parents’ and adult children’s occupational class.

Figure 2.10 shows the extent to which your parent’s occupational class is related to your own occupational class. The numbers in the chart are all relative to 2014. A negative number implies that a person’s SEB is less related to their current occupational class than in 2014. In other words, relative mobility is higher than in 2014. Overall our figure shows that relative to 2014, the strength of the link between people’s occupational class and their parents’ occupational class is getting weaker, and significantly so from 2019 onwards. There are signs of social progress even though social inequalities are still marked in 2022.

Figure 2.10: Relative occupational mobility has been improving since 2014, with significant differences between mobility in 2019 to 2022 and the base year, 2014.

Relative occupational mobility in the UK from 2014 to 2022, uniform difference (UNIDIFF) parameter estimates compared with 2014. The UNIDIFF parameter shows whether odds ratios have grown or shrunk over time.

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

Source: Office for National Statistics, Labour Force Survey 2014 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: The UNIDIFF model assumes that all odds ratios differ by a common multiplier in comparison with a benchmark group (in this case, year 2014). This common percentage is expressed in log form, the log UNIDIFF parameter shown above. When the logged value is near 0, there is no change in the odds ratios – relative mobility is constant across all groups. But when it is negative, the link between origins and destinations is weaker – the odds ratios are lower, and relative mobility is higher. At very large negative values, there would be almost no link at all between origins and destinations.

Intersectional analysis of relative occupational mobility

We now look at how levels of relative occupational mobility differ across sex, ethnicity, disability and region.

Differences between sex and ethnic groups

We see in figure 2.11 some clear differences across groups in terms of relative occupational mobility.

Although the White Other group has a similar level of relative mobility to the White British majority group, several other minorities show significantly higher levels of fluidity (that is, a looser link between parents and children).

High levels of fluidity are often regarded as desirable, but higher fluidity can arise from downward mobility as well as upward. It could reflect the fact that many first-generation minorities experienced downward social mobility. This so-called perverse (or undesirable) fluidity is consistent with some of the findings for absolute occupational mobility – we see some of the highest rates of downward mobility for the same groups, particularly the Pakistani and Bangladeshi groups. Possible explanations might relate to a lack of fluency in the English language, foreign qualifications that are not easily transferable, or discrimination.

Deeper analysis is required, particularly distinguishing the first generation (the migrants) from the second generation (who were born in Britain and will typically be fluent in English and have British qualifications).

Figure 2.11: Men and women from Black, Pakistani, Bangladeshi and Chinese backgrounds experience higher levels of relative mobility than do White British or White Other groups.

Relative mobility by ethnic background and sex in the UK, from 2018 to 2022, uniform difference parameter estimates compared with the White British group.

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

Source: Office for National Statistics, pooled Labour Force Survey 2018 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: The UNIDIFF (uniform difference) model assumes that all odds ratios differ by a common multiplier in comparison with a benchmark group (in this case, White British). This common percentage is expressed in log form, the log UNIDIFF parameter shown above. When the logged value is near 0, there is no change in the odds ratios – relative mobility is constant across all groups. But when it is negative, the link between origins and destinations is weaker – the odds ratios are lower, and relative mobility is higher. At very large negative values, there would be almost no link at all between origins and destinations.

Differences by disability status

We can also use the UNIDIFF model to compare the level of relative mobility among people with and without a disability.

Figure 2.12 shows that among those with a disability there is a stronger association between social class origins and current social class position. This is in line with our findings in chapter 3 on intermediate outcomes which show that disability gaps tend to be larger among people from more disadvantaged backgrounds. One possible interpretation of this pattern is that people with a disability are more dependent on their parents for help than are people without a disability.

Figure 2.12: There is significantly greater intergenerational persistence of occupational class among people with a disability than among those without a disability.

Relative mobility by disability status in the UK, from 2018 to 2022, UNIDIFF parameter estimates for respondents with a disability compared with those without a disability.

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

Source: Office for National Statistics, pooled Labour Force Survey (LFS), from 2018 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

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. The UNIDIFF parameter estimate means that the association between people’s social class and their parents’ social class is significantly stronger among those with a disability than among those without a disability.

Differences between regions

Unlike the findings for absolute occupational mobility, we don’t see any clear regional distinctions. Instead, in figure 2.13, we see levels of relative mobility that are similar across most areas of the UK, including London.

These results are closely in line with those reported recently by Granström and Engzell (2023), who summarise their findings as follows:

“How do opportunities for intergenerational mobility depend on the place where you live? … there is a clear distinction between upward mobility, largely driven by structural change, and relative mobility which is thought to closer reflect (in)equality of opportunity. Capital regions are hubs of absolute upward, but not always relative, mobility. Absolute upward mobility is correlated with a range of human capital, labour market, demographic, and socio-spatial characteristics. By contrast, the only robust predictor of relative mobility is income differences between social classes. More inequality entails less mobility, and this relationship holds within countries.”^{[footnote 22]}

Figure 2.13: Levels of relative mobility are similar across most areas of the UK, including London.

Relative mobility by area in which people grew up, uniform difference (UNIDIFF) parameter estimates compared with Inner London – West.

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

Source: Office for National Statistics, pooled Labour Force Survey 2018 to 2022, respondents aged 25 to 64 years in the UK, data collected from July to September each year.

Notes: Areas are where respondents lived when they were aged 14 years. These do not refer to current geography, areas where respondents lived when they completed the survey. The UNIDIFF model assumes that all odds ratios differ by a common multiplier in comparison with a benchmark group (in this case,Inner London – West). This common percentage is expressed in log form, the log UNIDIFF parameter shown above. When the logged value is near 0, there is no change in the odds ratios – relative mobility is constant across all groups. But when it is negative, the link between origins and destinations is weaker – the odds ratios are lower, and relative mobility is higher. At very large negative values, there would be almost no link at all between origins and destinations.

Income mobility

The measurement of income mobility is more challenging than that of occupational mobility. While people might be expected to remember their parents’ occupation when they were growing up, it is unlikely that many people can accurately recall their parents’ income. This means that large-scale surveys like the LFS, which rely on recall data for parents’ occupations, are not generally used for income mobility. Instead, the preference is for panel studies, in which the same individuals are tracked over time.

In these panel studies, parents can be asked directly about their own incomes at the time when their children were still at home. In turn, their adult children can be asked about their incomes after they have entered the labour market 20 or more years later. However, panel studies of this sort are very expensive, and tend to have much smaller sample sizes than the LFS. They also suffer from attrition (people dropping out) over time, which reduces the precision of the estimates.

Another option, which has been used in North America and elsewhere, is to link parent and child tax records. This potentially provides better quality data and much larger numbers of observations. Unfortunately, we do not have this in the UK.

Analysis of panel surveys in Britain has therefore been the main source of estimates of income mobility. Published studies using these data sources have produced many valuable insights into the patterns of relative income mobility in Britain.

Absolute income mobility

Absolute income mobility concerns the proportion of people whose income is higher than that of their parents when they were the same age. Having a higher income is referred to as upward absolute income mobility, and is generally strongly influenced by the growth of real household income.

Last year we reported on absolute income mobility in the UK by comparing it to other countries. We did not do our own analysis due to data constraints, but summarised the results from Manduca and others (2020), which discuss absolute mobility for some birth cohorts between 1960 and 1987 for some countries.^{[footnote 23] [footnote 24]} This year, we have included the same figure. Overall, absolute income mobility in the UK was good, at well above 70%, for those born in the mid-1970s but has since declined to below 70%.

Figure 2.14: Absolute income mobility in the UK was good for those born in the mid-1970s, but has since declined.

Estimates of upward absolute income mobility by country and birth cohort from 1960 to 1987.

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

Source: Manduca and others (2020). Trends in absolute income mobility in North America and Europe.

Note: The upward absolute mobility rate is the percentage of children in each birth cohort whose pre-tax, post-transfer family income at age 30 years, adjusted for inflation, was higher than their parents’ family income at age 30 years. Incomes are measured using a combination of register and survey data in each country. The results presented here are the same as those shown in our State of the Nation 2022 report, figure 2.5.^{[footnote 25]}

Relative income mobility

There have only been a few panel surveys in Britain which have been suitable for measuring intergenerational income mobility. The earliest was by Atkinson and colleagues.^{[footnote 26]} The study covered the years 1975 to 1978, in which the researchers surveyed the adult sons of fathers who had been interviewed as part of Seebohm Rowntree’s 1950 survey of poverty in York.^{[footnote 27]}

Atkinson’s study was quite small-scale and was not nationally representative. Subsequent larger-scale nationally representative studies have been based on 2 birth cohort studies. The first of these was the National Child Development Study (NCDS) which interviewed parents of children born in 1 week of March 1958. The children were then followed up through school and then through their working careers. The second was the 1970 British Cohort Study (BCS) which followed the same process. Parents of children born in 1970 were interviewed and then their children have been followed up regularly during their schooling and adult lives. A third study along the same lines – the Millennium Cohort Study (MCS) – interviewed parents of children born in 2001 and is continuing to follow up the children. As of yet, the children are not old enough to make an effective study of income mobility, but in a few years’ time the MCS will be able to provide important new evidence on income mobility.

In addition to these birth cohort studies, the ongoing UKHLS (which started with an adult sample in 2009) helps to link data on parents’ and children’s incomes in early adulthood. This can be used to make some up-to-date estimates of income mobility among younger people.

Relative income mobility is most commonly estimated by taking logarithms of both parents’ and children’s income (since income tends to be quite skewed with a long ‘tail’ consisting of a few people earning much higher incomes). The degree of intergenerational persistence (the link between parents’ and children’s incomes) can then be estimated with a linear regression model.^{[footnote 28]} Another method of estimating relative income mobility is to use parents’ and adult children’s centile positions in the income distribution. This method tends to give better results, but was not always reported in earlier research.

In table 2.15 we show estimates of relative intergenerational income mobility from the published studies plus our own estimates from the UKHLS 2020. We must emphasise, however, that there are major differences between the data and methodology of the various studies, and considerable imprecision in the estimates. The results that can most safely be compared are those for 1991 and 2004. These show the well-known increase in intergenerational persistence (that is, a decline in relative mobility) in the last decade of the 20th century. In table 2.15, the estimate shows the percentage increase in a child’s income associated with a 1 percent increase in their parent’s income. For example, in 2020, a 1 percent increase in someone’s parent’s income was associated with a 0.29 percent increase in their income.

Table 2.15: Intergenerational income persistence increased between 1991 and 2004 and has probably remained at a similar level in the 21st century.

Estimates of the strength of intergenerational income mobility (intergenerational elasticity), from 1975 to 1978, and 2020.

	1975 to 1978	1991	2004	2009 to 2016	2020
Estimate	0.36	0.21	0.33	0.27	0.29
95% confidence interval	0.14–0.57	0.15–0.27	0.27–0.36	0.22–0.32	0.15–0.43

Sources: 1975 to 1978: Atkinson and others (1981) based on follow up of the 1950 social survey of York; 1991 and 2004: Blanden and Machin (2008) based on National Child Development Study and Birth Cohort Study 1970; 2009 to 2016: Rohenkohl (2019) based on linked British Household Panel Study and UK Household Longitudinal Study (UKHLS) data; 2020: own calculations based on the UKHLS 2009 and 2020.^{[footnote 29] [footnote 30]}

Note: The 1975 to 1978 study measured fathers’ and sons’ weekly earnings. The age of the sons was not specified. The 1991 study measured parental household income and their children’s earnings at age 33 years. The 2004 study measured parental household income and their children’s earnings at age 34 years. The 2009 to 2016 study measured parental household gross income and adult children’s gross household income at age 25 years or older. The 2020 study measured parental household gross income and sons’ (and daughters’) gross personal income at age 25 years or older (average age 30 years).

While we recommend extreme caution in comparing results from the 2 birth cohort studies with those from the UKHLS, the results from the latter data source are broadly in line with those from the BCS 1970 (Blanden and Machin 2008).^{[footnote 31]} In other words, we cannot reject the hypothesis that relative income mobility has been unchanged over the first 2 decades of the 21st century.

We also need to be aware that estimates of relative income mobility vary over the life cycle. Estimated intergenerational persistence is lowest when the respondents’ income is measured during their 20s but then increases in their 30s and 40s before dropping again at age 50 years (Gregg and others, 2017).^{[footnote 32]} The estimates in table 2.15 are likely to be underestimates of the likely strength of intergenerational persistence in Britain.

Taking the above into consideration, our provisional conclusion is that relative income mobility has remained at roughly the same level in recent years.^{[footnote 33]}

Educational mobility

Educational mobility is studied in exactly the same way as occupational mobility, using large-scale representative surveys. Respondents are asked to recall the educational levels that their parents had reached. Just as parents with professional jobs are more likely to have children who go on to do professional jobs, parents who achieve a higher educational level have children who are more likely to do the same.

Absolute educational mobility

As with occupational mobility, figure 2.16 shows that there has been substantial upward educational mobility. For example, among people whose parents had no qualifications at all, 70% have achieved some qualifications, and 18% have obtained university degrees. Overall, 39% of the sample were upwardly mobile educationally compared with 26% who were downwardly mobile. This reflects the great expansion of higher education (HE) and school qualifications over the last 4 decades of the 20th century.

Figure 2.16: Upwards educational mobility rates are high, reflecting the expansion of school and higher education qualifications since the 1960s.

Highest level of qualification of respondents aged 25 to 64 years in the UK, 2020, by highest level of parental qualification.

Explore and download data on absolute educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Notes: Parental education is measured by using whichever parent has the higher level of qualification. If there is data on only one parent, then only this data is used. The available measure of parental education in the UKHLS distinguishes those with university degrees, those with some post-school qualification, those with a school qualification, and those without any qualification. The respondents’ own highest level of qualification has been recoded into the same 4 categories. The data used is weighted using the UKHLS population weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Nevertheless, substantial inequalities persist in the outcomes for those at each end of the spectrum. Respondents whose parents had obtained a university degree were by far the most likely to have degrees themselves: 64% of these people obtained degrees. This is more than 3 times as many as were obtained by those whose parents had no formal qualifications (18%).^{[footnote 34]}

We can also explore trends over time in absolute educational mobility by comparing our results with those from economist Brian Bell and others (2022).^{[footnote 35]} Bell and his colleagues use the linked census data of the Longitudinal Study to compare the proportions of (adult) children from graduate and non-graduate homes who obtained a degree. They compare results for people aged 28 to 37 years in the 1991, 2001 and 2011 censuses. They found that in 1991, only 9% of people from non-graduate homes (where neither of the parents was a graduate) became graduates themselves. This increased to 19% in 2001 and 35% in 2011. From our additional calculations, following the same methodology, the UKHLS 2020 data shows a figure of 33% – little has changed since 2011.^{[footnote 36]}

Previous studies have all shown broadly similar patterns to ours, with considerable upward educational mobility alongside major inequalities in the outcomes of people from different backgrounds. The large increase in the rate of upward educational mobility that Bell found between 1991 and 2011 corresponds to the great expansion of tertiary education in Britain in the last decade of the 20th century and the first decade of the 21st century. This expansion has subsequently stayed the same, and so it is not surprising that upward educational mobility has also flatlined.

Intersectional analysis of absolute educational mobility

Next, we turn to differences in absolute educational mobility outcomes across sex, ethnicity and region.^{[footnote 37]}

Differences between men and women

Figure 2.17 shows absolute educational mobility for men and women. Overall, there are only small differences in men’s and women’s educational mobility. The main difference involves post-school qualifications. Women are significantly more likely than men to obtain these qualifications. It should be noted, these include a range of professional qualifications which may be below degree level such as nursing and teaching certificates.

Another notable result was the differences in obtaining university degrees. Among people whose parents had obtained degrees, women were significantly more likely than men to obtain degrees, whereas a similar gap was not seen among those whose parents had non-degree qualifications.

Figure 2.17: Men and women have very similar patterns of absolute educational mobility.

Highest level of qualification of respondents aged 25 to 64 years in the UK, 2020, by highest level of parental qualification and sex.

Explore and download data on absolute educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: Differences between men and women in the percentage gaining post-school qualifications are significant at the 5% level within each parental education level except post-school qualifications. The data used is weighted using the UKHLS population weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Differences between ethnic groups

Looking at differences between ethnic groups we see something different. As shown in figure 2.18, there are particularly high proportions of people from the Chinese, Other White, Indian and Pakistani groups who experienced upward educational mobility, coming from families where the parents did not have a degree but gaining degrees themselves. The other ethnic groups also show slightly higher proportions experiencing upward educational mobility than their White British peers, but these differences are not statistically significant.

Again, mobility rates are affected by the balance of first and second (or later) generation immigrants in that particular ethnicity. Higher rates of upward educational mobility than among White British people could reflect the fact that many of the migrants will have been educated in their countries of origin. In several of these countries average educational attainment is lower than in the UK so there is more scope for upward educational mobility.

Figure 2.18: There are particularly high levels of upward educational mobility among people of Chinese, Other White, Indian and Pakistani ethnicity.

Percentage obtaining degrees of those whose parents did not have degrees, respondents aged 25 to 64 years in the UK, 2020, by ethnic group.

Explore and download data on absolute educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: Sample restricted to people whose parents did not have a degree. The percentages in the figure can be interpreted as the proportion of those from a non-graduate family who are upwardly mobile educationally. Differences from the White British percentage are statistically significant at the 1% level for the Indian, Chinese, Other White, and Pakistani groups. The data used is weighted using the UKHLS population weights.

Differences by disability status

Figure 2.19 shows the highest level of educational attainment by both parental educational attainment and disability. In contrast to both sex and ethnicity, we find that people with a disability have poorer chances of obtaining a university degree than their peers from the same educational background. People with a disability are more likely to have no qualifications than their peers. This is a pattern of consistent disadvantage across the board.

Figure 2.19: People with a long-term illness or disability have poorer chances of upward educational mobility than their peers.

Highest level of qualification of respondents aged 25 to 64 years in the UK, 2020, by highest level of parental qualification and disability status.

Explore and download data on absolute educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: Respondents were asked: “Do you have any long-standing physical or mental impairment, illness or disability? ‘Long-standing’ means anything that has troubled you or is likely to trouble you over a period of at least 12 months.” This is a broader definition than that used in analyses of Labour Force Survey (LFS) data since the LFS specifies that the condition should be both long-term and limiting. Differences between those with and without a disability gaining a university degree are significantly different at the 5% level. The data used is weighted using the UKHLS population weights. Due to rounding errors, in some instances the totals may not add up to 100%.

Note that, although the question asks about a long-standing illness or disability, we do not know how long respondents have had the illness or disability. It is possible that education had already been completed before the illness or disability occurred. To investigate the causal impact of disability on educational mobility, we would need a panel study (or at the least retrospective data about the timing of onset of the disability). So these results are likely to underestimate the effects of disability.

Differences between regions

We show rates of educational upward mobility in the 12 regions at International Territorial Level (ITL1 level) in figure 2.20.^{[footnote 38]} This is because sample sizes are not large enough for us to get reliable estimates for more granular areas at the ITL2 level. Overall, in most regions of the UK, patterns of upward educational mobility are quite similar. However, some regions stand out – the East Midlands, Wales and London. In London a significantly higher percentage of people from non-graduate homes have degrees – 39% compared with the national figure of 29%. In contrast, there are significantly lower percentages than expected both in the East Midlands, Yorkshire and the Humber, West Midlands and Wales.

Figure 2.20: Upward educational mobility is lower in the East and West Midlands, Yorkshire and the Humber, and in Wales and is higher in London and the South East.

Percentage obtaining degrees of those whose parents did not have a degree, respondents aged 25 to 64 years in the UK, 2020, by area of current residence.

Explore and download data on absolute educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: Sample restricted to those whose parents did not have a degree. Region is that of current residence. Area of current residence was used, as the version of the UKHLS used for this report did not contain details of the area where respondents grew up. Results for the East and West Midlands, Yorkshire and the Humber and Wales are significantly lower than the overall level at the 5% level of significance while results for London and the South East are significantly higher. The data used is weighted using the UKHLS population weights.

However, it is important to note that the data relates to where respondents currently live, not where they grew up. So the high percentages in London may reflect the inflow of graduates from other areas, and the low percentages in Wales and the East Midlands could reflect the exit of graduates to other areas. Unfortunately it has not yet proved possible to obtain data on where these respondents grew up. It should be noted however that Bell and others (2022) were able to use data on the area where people grew up and also found a higher rate of upward degree mobility in parts of London.^{[footnote 39]}

Relative educational mobility

In order to track trends in relative educational mobility over time, we draw on the results of Bell and others (2023) for 1991, 2001 and 2011, and update these with 2020 data from the UKHLS.^{[footnote 40]}

Because of the limited measures of education available to Bell and colleagues in the censuses, for comparability we use a binary measure of education. This variable measures whether respondents and their parents had attained an undergraduate university degree or not. Bell’s analyses were restricted to adults aged 28 to 37 years in each census, and we therefore use the same age group in the UKHLS. This binary measure also means that we can’t take into account higher degrees, or the selectiveness of universities. These factors may be particularly important for relative mobility, and we go into more detail about this in intermediate outcome 2.3.

Figure 2.21 shows that relative educational mobility has improved very considerably from the very high level of ‘stickiness’ seen in 1991.^{[footnote 41]} However, there is still considerable inequality with an odds ratio of 4 in 2020. An odds ratio can be interpreted as the outcome of a competition between people from 2 different origins to achieve a particular outcome and avoid the alternative outcome. It is a standard measure of relative mobility as it is independent of changes in the distributions.

Figure 2.21: There has been an increase in relative educational mobility among young adults, 1991 to 2020.

Parent:adult children odds ratios relating to university degrees, UK, respondents aged 28 to 37 years.

Explore and download data on relative educational mobility on the State of the Nation data explorer.

Source: UK Household Longitudinal Survey (UKHLS), 1991 to 2020. Respondents aged 28 to 37 years in the UK.

Note: Education is a binary measure of attainment of an undergraduate degree qualification. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility.

Intersectional analysis of relative educational mobility

Differences between men and women

Due to the small sample sizes available for the analysis of educational mobility, it is generally difficult to see statistically significant differences between groups. This does not mean that such differences don’t exist. In our analyses, there are detectable differences in the relative educational mobility of men and women. Women’s chances of obtaining a degree seem to depend more strongly on their parents’ educational level than men’s chances. In other words, relative educational mobility is worse among women than men.

Figure 2.22: There is a higher level of relative educational mobility among men than among women.

Parent:adult children odds ratios relating to university degrees, UK, 2020, respondents aged 25 to 64 years by sex.

Explore and download data on relative educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: For this analysis, education is re-coded, for both parents and respondents, as a binary measure of attainment of an undergraduate degree qualification versus not. Formal modelling of the data with logistic regression confirms that the odds ratios for men and women are significantly different from each other. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility. The data used is weighted using the UKHLS population weights.

Differences between ethnic groups

Levels of relative educational mobility are broadly similar between different ethnic groups in the UK. But small sample sizes and large confidence intervals mean that we cannot detect significant differences between them.^{[footnote 42]} For example, in the case of the Chinese group we are unable to calculate a confidence interval because of the small sample size.

Figure 2.23: Relative educational mobility is similar across all the major ethnic groups in the UK.

Parent:adult children odds ratios relating to university degrees, UK, 2020, respondents aged 25 to 64 years.

Explore and download data on relative educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: For this analysis, education is re-coded, for both parents and respondents, as a binary measure of attainment of an undergraduate degree qualification versus not. Formal modelling of the data with logistic regression confirms that the odds ratios for the different ethnic groups are not significantly different from each other. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility. The data used is weighted using the UKHLS population weights. Odds ratios for the Chinese group could not be calculated due to the small sample size.

Differences by disability status

Relative educational mobility is broadly similar among people with and without a long-term illness or disability. However, the small sample size means that the estimates have wide confidence intervals and so are less precise, making it difficult to be sure.

Figure 2.24: Relative educational mobility is similar among people with a long-term illness or disability and those without.

Parent:adult children odds ratios relating to university degrees by disability status, UK, 2020, respondents aged 25 to 64 years.

Explore and download data on relative educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: For this analysis, education is re-coded, for both parents and respondents, as a binary measure of attainment of an undergraduate degree qualification versus not. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility. The data used is weighted using the UKHLS population weights.

Differences between regions

Figure 2.25 illustrates that Scotland has the highest level of relative educational mobility. However, we should note that Scottish educational institutions are different from those in England. In particular Scotland has different qualifications for university entry and a long tradition of 4-year university courses.

Figure 2.25: The level of relative educational mobility is greater in Scotland but does not vary significantly across the other parts of the UK.

Parent:adult child odds ratios relating to university degrees, UK, 2020, respondents aged 25 to 64 years by area of current residence.

Explore and download data on relative educational mobility on the State of the Nation data explorer.

Source: The UK Household Longitudinal Survey (UKHLS), 2020 calendar year, respondents aged 25 to 64 years in the UK.

Note: For this analysis, education is re-coded, for both parents and respondents, as a binary measure of attainment of an undergraduate degree qualification versus not. Formal modelling of the data with logistic regression confirms that the odds ratio for Scotland is significantly lower. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility. The data used is weighted using the UKHLS population weights.

Housing mobility

Housing is often used as a proxy for wealth, and like wealth can be passed directly from one generation to another. This can influence long-term living standards.^{[footnote 43]} In addition, housing is of considerable interest in its own right, as overcrowding and substandard housing is associated with poor health and poorer educational results for children. Homeowners often have better housing conditions than those who are renting. For further discussion see Heath and others, 2018, chapter 5 and Blanden and others 2021.^{[footnote 44] [footnote 45]}

Since house prices in the UK have risen faster than in many other countries, home ownership has become an important factor in wealth accumulation. This has created concerns about intergenerational fairness – younger people who are unable to buy a house won’t benefit from this accumulation.

Absolute housing mobility

One recent study of housing mobility in Britain using the WAS, NCDS and BCS has shown starkly different patterns of change over time from any of the other mobility trends.^{[footnote 46]} This study finds that regarding absolute mobility, there has been a substantial decline in upward housing mobility. Among people born in the late 1950s, 74% owned their own home even though their parents had not been homeowners. This fell to 49% of people born 20 years later in the late 1970s.

Figure 2.26 shows the home ownership status of people by the home ownership status of their parents, derived from the Wealth and Assets Survey (WAS). This shows that people whose parents owned their own house are themselves much more likely to own their own house (71%, compared with 46%).

Figure 2.26: There is substantial intergenerational persistence in home ownership.

Parental home ownership of respondents aged 25 to 64 years in the UK, from 2016 to 2020, by own home ownership.

Explore and download data on absolute housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020). Respondents aged 25 to 64 years in the UK.

Notes: This plot shows the current tenure by parental tenure. For example, 71% of those who own a house in adulthood had parents owning a house when a teenager. The error bars show 95% confidence intervals. The data used is weighted using the WAS individual weights.

Intersectional analysis of absolute housing mobility

Differences between men and women

Figure 2.27 shows the home ownership status of men and women by the home ownership status of their parents. Among those whose parents were homeowners, women (64%) are less likely than men (75%) to own their own home. Similarly, among those whose parents were not homeowners, only 35% of women compared with 55% of men owned their own homes. These sex differences are statistically significant.

Figure 2.27: Women are significantly less likely than men to own their homes.

Home ownership of respondents aged 25 to 64 years in the UK, from 2016 to 2020, by parental home ownership and sex.

Explore and download data on absolute housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020). Respondents aged 25 to 64 years in the UK.

Note: The error bars show the 95% confidence intervals for each estimate. The data used is weighted using the WAS individual weights.

Differences by disability status

Figure 2.28 shows that as with other outcomes, home ownership is significantly lower among people with a disability, whether or not their parents were also homeowners. Of those who are disabled and whose parents were homeowners, 61% are homeowners themselves, compared with only 34% of those who are disabled but did not have parents who are homeowners. This is in line with previous research on the financial situation of people with a disability.^{[footnote 47]}

Figure 2.28: Home ownership is much lower among people with a disability than among their peers.

Home ownership of respondents aged 25 to 64 years in the UK, from 2016 to 2020, by parental home ownership and disability.

Explore and download data on absolute housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020). Respondents aged 25 to 64 years in the UK.

Note: The error bars show the 95% confidence intervals for each estimate. The data used is weighted using the WAS individual weights.

Differences between regions

Figure 2.29 shows the regional breakdown in home ownership status of people by the home ownership status of their parents. Overall, Greater London stands out as a region where home-ownership is substantially lower than elsewhere in England, Scotland and Wales. This applies both to people whose parents were owners and to those whose parents were renters. Apart from Greater London, differences between different regions are small and generally not statistically significant, although upward housing mobility is somewhat higher in the South West and Wales. These patterns are likely to reflect levels of house prices, although we should note that there are likely to be substantial variations within regions.

Figure 2.29: Home ownership is less common in London regardless of parental housing status, whereas upward housing mobility is somewhat higher in the South West and Wales.

Home ownership of respondents aged 25 to 64 years in the UK, from 2016 to 2020, by parental home ownership and area of residence.

Explore and download data on absolute housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020). Respondents aged 25 to 64 years in the UK.

Notes: The area where the respondent grew up is not available in the WAS, and so this chart shows areas of current residence. The error bars show the 95% confidence intervals for each estimate. The data used is weighted using the WAS individual weights.

These results are similar to those of Bell and others (2019) based on linked census data up to 2011. Their research shows that absolute rates of upward housing mobility were significantly lower in parts of London and significantly higher in more rural areas such as West Wales.

Relative housing mobility

As with the intersectional analysis of relative educational mobility, we use odds ratios to measure relative housing mobility. A higher odds ratio indicates greater intergenerational persistence while a lower odds ratio indicates greater relative mobility.

Figure 2.30 shows this odds ratio over time. What we find is a steady decline in relative housing mobility from 1991. In other words, intergenerational persistence has increased. This is likely to be related to the increase in real house prices over time and the increasing need for existing family resources for people to buy their first property.

Figure 2.30: Relative housing mobility has declined steadily since 1991.

Odds ratios of the relationship between parental and respondent home ownership in the UK, from 1991 to 2016 and in 2020, among younger respondents.

Explore and download data on relative housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 and 7 (respondents aged 30 to 34 years) and Bell and others (2022, table 6, respondents aged 28 to 37 years) in the UK.^{[footnote 48]}

Note: The error bars show the 95% confidence intervals for each estimate. The odds ratio is a measure of relative mobility. It is the ratio of the odds (of owning a house or not) among those whose parents owned a house to the odds among those whose parents had not. The data used is weighted using the WAS individual weights.

Intersectional analysis of relative housing mobility

Differences between men and women

Figure 2.31 shows that there is significantly greater intergenerational persistence among women than among men with respect to home ownership. To our knowledge, this finding has not previously been reported. One possibility is that women are more likely to rely on help from their parents (perhaps because of gender pay gaps in employment) and less likely to be able to accumulate the financial resources needed for home ownership such as a large deposit. In other words, women may be more likely than men to need help from ‘the bank of mum and dad’ (a phrase now used for when parents of adult children give a large sum of money to help them buy a house). However, in-depth research is needed to test this interpretation.

Figure 2.31: There is significantly greater intergenerational persistence (‘stickiness’) in home ownership among women than men.

Odds ratios of the relationship between parental and respondent home ownership in the UK, from 2016 to 2020, by sex.

Explore and download data on relative housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020), respondents aged 25 to 64 years in the UK.

Note: The error bars show the 95% confidence intervals for each estimate. Logistic regression modelling confirms that there is a statistically significant difference in the relative housing mobility of men and women. The data used is weighted using the WAS individual weights.

Differences by disability status

Figure 2.32 shows the same odds ratio of home ownership as figure 2.31, but instead splits it by disability status. As with sex differences, we find significant differences in relative housing mobility between people with and without a disability. Intergenerational persistence is greater among those with a disability. This probably reflects the greater reliance on parental wealth necessary for those with a disability to be able to buy a property.

Figure 2.32: Relative housing mobility is significantly lower among those with a disability.

Odds ratios of the relationship between parental and respondent home ownership in the UK, 2016 to 2020, by disability.

Explore and download data on relative housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2017) and 7 (from 2018 to 2020), respondents aged 25 to 64 years in the UK.

Note: The error bars show the 95% confidence intervals for each estimate. Logistic regression modelling confirms that there is a statistically significant difference between the relative housing mobility of those with and without a disability.  The data used is weighted using the WAS individual weights.

Differences between regions

Figure 2.33 shows the odds ratio of home ownership by region in the UK. While there is an interesting pattern to the odds ratios, there is considerable imprecision in the estimates of relative housing mobility in the different regions of the UK in the WAS dataset. We can see from the confidence intervals that none of the estimates for the different regions are significantly different from the national average (2.85). So it may be that relative housing mobility is similar across the different regions of the UK.

Figure 2.33: Relative housing mobility does not differ significantly across regions.

Odds ratios of the relationship between parental and respondent home ownership in the UK, 2016 to 2020, by region.

Explore and download data on relative housing mobility on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (2016 to 2017) and 7 (2018 to 2020), respondents aged 25 to 64 years in the UK.

Note: The error bars show the 95% confidence intervals for each estimate. Logistic regression modelling confirms that there is no statistically significant difference in relative housing mobility between different regions. The data used is weighted using the WAS individual weights.

However, if we were able to use a more granular measure of geography (and a much larger sample), significant differences between areas might emerge. Bell and others’ analysis of linked censuses suggests that there was significantly greater intergenerational persistence (stickiness) in parts of London and significantly greater relative mobility in West Wales and some other rural areas in 2011.^{[footnote 49]}

Wealth mobility

There has been increasing interest in wealth mobility recently.^{[footnote 50]} Parents’ wealth can be important for their children’s living standards and for children’s mobility chances in other domains. For example, wealthy parents may be able to use the resources they have accumulated during their lifetime to help their children buy their first house or to make investments in their children’s education. Intergenerational wealth mobility is also likely to involve different processes from those involved in educational, occupational or income mobility as wealth can be transferred directly to later generations through gifting or inheritance.

Wealth can however take several different forms. The major ones are:

• net property wealth, such as the sum of all property values minus the value of all mortgages and amounts owed (for example equity release)^{[footnote 51]}
• physical wealth, such as the sum of the values of household contents, collectibles and valuables, and vehicles
• net financial wealth, such as the sum of the values of financial assets, plus the value of endowments purchased to repay mortgages, less the value of non-mortgage debt
• private pension wealth, namely the sum of the value of current occupational pension wealth, current personal pension wealth, additional voluntary contributions, plus the value of pensions expected from a former spouse or partner and value of pensions in payment

The main data source available in the UK is the regular ONS Wealth and Assets Survey, which only covers Great Britain. Drawing on the most recent rounds of this survey we show, in figure 2.34, the average levels of these 4 components and how they vary among people of different ages.

This doesn’t allow us to draw conclusions about wealth mobility. Instead, we interpret the increasing levels of wealth across age groups as the result of a life-cycle process with people on average accumulating wealth until retirement, and then using up their wealth to a greater or lesser extent during retirement. We should note however, that there will be considerable variation around this average, reflecting differences in people’s incomes, home ownership and occupational positions.^{[footnote 52]}

Figure 2.34: Wealth, and its main components, increases over people’s lives – people tend to accumulate wealth as they get older.

Financial wealth, pension wealth, physical wealth, property wealth and total wealth by age group in Great Britain, from 2016 to 2020.

Explore and download data on levels of wealth on the State of the Nation data explorer.

Source: Wealth and Assets Survey (WAS) waves 6 (from 2016 to 2018) and 7 (from 2018 to 2020), respondents aged 25 to 64 years.

Note: For further details see the Office for National Statistics (2022) household total wealth in Great Britain: April 2018 to March 2020.^{[footnote 53]} Data has been adjusted for inflation. The error bars show the 95% confidence intervals for each estimate. The data used is weighted using the WAS individual weights.

Relative wealth mobility

While the WAS is the primary source on the distribution of wealth in Britain, it does not include direct measures of parental wealth. It does, however, ask respondents to recall their parents’ housing tenure and educational attainment (as well as some other variables). We therefore follow Gregg and Kanabar (2022) who estimate parental wealth on the basis of the observed relationships between education, housing tenure and wealth. This enables us to produce a rough estimate of relative wealth mobility.^{[footnote 54]} We find that a 10% increase in a person’s parents’ wealth is associated with roughly a 3% increase in their own wealth. However, we must note this is not a causal estimate but shows the magnitude of the relationship between the wealth a parent accumulates and how this might predict their children’s future wealth.^{[footnote 55]}

This estimate is slightly higher than the one we found for income mobility (see above), but because of differences in methodology and data sources it would be safer to conclude that they are similar in magnitude. Our results are also similar to those found by Gregg and Kanabar (2022) who used an earlier round of WAS (namely 2010 to 2012).

Gregg and Kanabar also carried out some more detailed supplementary analysis comparing older and younger respondents. They concluded that it was likely that intergenerational wealth persistence was increasing over time. Given the finding (see above) that intergenerational housing persistence had increased between 1991 and 2011, it seems plausible that wealth persistence has been increasing too. This is an important issue that we plan to investigate in more detail in future work.

Intersectional analysis

We are able to look at sex differences in intergenerational wealth persistence. Overall we find that intergenerational wealth persistence is greater among women than men. For men, a 10% increase in their parent’s wealth is associated with a 2.9% increase in their own wealth, whereas for women it is 3.8%.^{[footnote 56]} As was shown earlier, intergenerational housing persistence was significantly greater among women than men, and so it is plausible that overall wealth persistence might be somewhat greater among women. On the other hand, our earlier results on relative occupational mobility suggested that there was slightly less intergenerational occupational persistence among women, which might be expected to reduce intergenerational pensions persistence. Since Gregg and Kanabar also found greater intergenerational wealth persistence in the 2010 to 2012 round of the WAS, it seems likely that intergenerational persistence is indeed greater among women. Further research is needed to explore why intergenerational wealth persistence varies between men and women.

Conclusion

The 5 types of mobility we have measured – occupational, income, educational, housing and wealth – show different patterns and trends over time.

Last year’s State of the Nation report showed that occupational mobility had remained broadly stable over a long time, with perhaps a slight improvement in relative occupational mobility (meaning that the strength of the association between parents’ and children’s jobs may have slightly weakened). This year, we have concentrated on breakdowns by geography and individual characteristics. Geographically, there is a clear centre-periphery divide. Those who grew up in the south-east have the best upward occupational mobility rates, and those who grew up in the north and the south-west have the worst. Women are more likely to experience long-range downward mobility than men, while people from certain ethnic groups, notably Chinese and Indian, have better chances of long-range upward mobility than their White British peers. People with a disability face, unfortunately, worse mobility prospects than those without.

Absolute income mobility has declined since the 1970s, but remains higher compared with countries such as the US and Canada. Relative income mobility has remained roughly the same in the past 20 years.

On educational mobility, as expected, we find that people with university-educated parents are more likely to obtain a degree. Women are more likely to obtain degrees than men, as are people from Chinese, Indian, Black African and White Other backgrounds. In fact, members of most other ethnic groups from non-graduate families are significantly more likely than White British people to gain a degree, apart from Black Caribbean and Bangladeshi groups.

Housing mobility is the one outcome that has very significantly worsened over time. The link between parents’ home ownership and children’s home ownership has strengthened since 1991. It also shows a strong regional pattern, with London being considerably worse than other areas of the country. This is likely to be due to rising house prices in London.

For wealth mobility, we also find that the wealth a person accumulates is linked to their parents’ wealth. This relationship appears to be stronger for women.

When measuring our mobility outcomes we face challenges with the data, as our findings are often only derived from a small number of birth-cohort studies.

Case studies

Chantelle Powers, age 30 years, from Solihull

I always remember feeling embarrassed to talk about my upbringing and where I was from. I grew up in a single-parent household in an area of deprivation within Birmingham. I often remember my mother giving me the last pennies out of her pocket just so that I wouldn’t miss out and could go out with friends. Now that I’m older I’ve realised how much she went without and I feel lucky I’m in the position I am now to make sure that doesn’t happen again.

Throughout school I developed an ambition to work in animal care, so I took the option to complete a level-3 BTEC [Business and Technology Education Council]. When I tried to apply for university to become a vet, I didn’t realise the barriers I’d face with not picking A Levels as I’d just assumed the BTEC was equivalent. With a number of declines, and offers only at universities too far away, I didn’t think I’d manage to get a degree. I remember my mother trying to calculate how she could support me but going to university, paying course fees and accommodation costs was something that just felt unattainable. I then decided to think about other career prospects. I headed to an interview for a Level 3 Business course at my local college, where I was surprised to be offered an alternative route – studying for my degree through the college, enabling me to pursue the possibility of graduating while remaining in my area!

While at college I developed an interest in HR [human resources]. I remember asking my tutor about the possibilities and how I could get into the sector, and I reached out to my local careers service as it felt they were the only people who might be able to support me. It just happened that a few weeks later an organisation reached out for a volunteer to support their HR team so I took the chance! Little did I know this opportunity would be the starting point to my career.

A college tutor then shared Jaguar Land Rover’s (JLR) undergraduate placement with me and I had to take the chance to apply. I’d had no prior experience of an assessment day or what to expect and no family member to ask to share their experience, but to my surprise I was offered the placement! I remember someone close to me saying that I’d just been selected on the basis of where I chose to study and the area that I was from, not recognising that I had the potential to succeed. JLR has been a great employer, enabling me to continue onto their graduate scheme and providing me with a number of opportunities and experiences.

I’m so proud of where I am today – the imposter syndrome often kicks in but I’m really lucky that I’ve returned to a role in early careers where I know I’ll be able to support students throughout their career choices. There’s so much more to do in this space and we’re just getting started.

Aysha Patel, age 28 years, from London

I grew up in Bolton, not far from Manchester. My parents both migrated from India, so for them this was a completely new country. They were starting from scratch, building from scratch, and bringing up a family in a new environment.

My dad did whatever jobs he could find. He worked in curtain factories, takeaways and as a shop assistant. It was inspiring to know that my parents came from literally nothing and worked their way up. I saw how hard my dad worked and so I saw him as a role model.

Because English wasn’t their first language, my parents couldn’t help me with homework. The curriculum was different from India and neither of them finished school. I had to be quite self-sufficient, but they always pushed me and my siblings to study and made sure we took advantage of opportunities that came our way.

From the start, I excelled at maths. My teacher was great. There was a point where I was the only girl in the class, but he was super helpful and always pushing me. He was one of the reasons I chose to study maths at university.

In the last year of my degree, I started applying for jobs. I got involved with a charity called Upreach, which supports individuals from low socio-economic backgrounds. They set up introductory sessions with different companies. Because of them, I was exposed to more options.

Initially I worked in consulting, then I got a scholarship to do an MBA [Master of Business Administration] at Manchester University. Over the summer, we had the opportunity to do an internship and I applied for one at Amazon. At the end of the 12 weeks, we had to make a presentation about the improvements we’d made and any recommendations we had. The environment was very new to me, as I wasn’t used to working with a high-vis jacket and safety shoes on in a warehouse type environment. But I made sure I put the hard work and effort in, and was offered a job.

My parents are super proud of me and quite excited to see what comes next. I’ve just got married and me and my husband often talk about the opportunities we would like to provide for our children; opportunities that maybe we didn’t have. I think there’s always going to be something that’s going to push you down in life, but keep trying. Keep putting the effort in and those efforts will pay off.

Now I am a senior programme manager. I implement and improve new technology. Because of how big Amazon is, some of the projects I work on are worldwide. Whenever I receive a package from Amazon I think, wow the amount of effort that’s gone into getting this one package to me at this level of speed is amazing!

1. See chapter 1, table 1.2 for an explanation of these region names. ↩

2. Absolute income mobility concerns the proportion of people whose income is higher than that of their parents, when they were of the same age. ↩

3. Max Thaning, ‘Multidimensional intergenerational inequality: resource specificity in education, occupation, and income’, 2019. Stockholm University. Published on SU.FIGSHARE.COM. ↩

4. Martin Hällsten and Max Thaning, ‘Wealth as one of the “big four” SES dimensions in intergenerational transmission’, 2021. Published on ACADEMIC.OUP.COM. ↩

5. Intersectional analysis means looking at more than one characteristic at once, to check how they might interact. For example, we might look at both SEB and sex, and the interaction between them. ↩

6. 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. ↩

7. “Recall bias” means that people’s memory can affect the accuracy of results. “Life-cycle bias” means that the outcomes we are looking at are correlated with people’s age, so comparing people of different ages is difficult. “Attrition” means people dropping out of surveys over time. “Attenuation” means that random measurement error can make it hard to spot genuine statistical relationships. ↩

8. The Office for National Statistics collects, analyses and shares statistics about the UK’s economy, society and population. ↩

9. The NS-SEC was developed from a classification known as the Goldthorpe Scheme. It emphasises aspects of occupation such as labour-market situation, relationship to the employer, job security and advancement, rather than salary. So people in lower occupational classes can sometimes earn more than people in higher occupational classes. There can also be great variation in earnings within a class. ↩

10. Cohorts are groups of people used in a study who show similar characteristics such as age. ↩

11. See figure 2.3 in Erzsébet Bukodi and John Goldthorpe, ‘Social mobility and education in Britain: research, politics and policy’, 2019. Published on CAMBRIDGE.ORG. ↩

12. A Halsey and Josephine Webb, ‘Twentieth-century British social trends’, 2000. Published on SEMANTICSCHOLAR.ORG. ↩

13. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. LSE.AC.UK; Jo Blanden and others,‘Trends in intergenerational home ownership and wealth transmission’, 2021. Published on CENTRE FOR ECONOMIC PERFORMANCE LSE.AC.UK. ↩

14. Franz Buscha and Patrick Sturgis, ‘Declining social mobility? Evidence from five linked censuses in England and Wales 1971-2011’, 2018. Published on EPRINTS.SOTON.AC.UK. ↩

15. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

16. Outliers are values or points that are extreme and different from most other parts of a data set. ↩

17. “Intersectional analysis” means looking at more than one characteristic at once, to check how they might interact. For example, we might look at both SEB and sex, and the interaction between them. ↩

18. Claudia Goldin and Lawrence Katz, ‘The cost of workplace flexibility for high-powered professionals’, 2011. Published on JOURNALS.SAGEPUB.COM. ↩

19. 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. ↩

20. For a general review of the economic situation of people with a disability see, ‘Being disabled in Britain: a journey less equal’, 2017. Published on EQUALITYHUMANRIGHTS.COM. ↩

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

22. Olivia Granström and Per Engzell, ‘The geography of intergenerational mobility in Europe’, 2023. Published OSF.IO.PREPRINTS. ↩

23. Countries included are: US, Sweden, Denmark, Norway, Canada, Finland and the Netherlands. ↩

24. Robert Manduca and others, ‘Trends in absolute income mobility in North America and Europe’, 2020. Published on IZA.ORG. ↩

25. See figure 2.5 of Social Mobility Commission, ‘State of the Nation 2022: A fresh approach to social mobility’, 2022. Published on GOV.UK. ↩

26. Anthony Atkinson, ‘On intergenerational income mobility in Britain’, 1981. Published on TANDFONLINE.COM. ↩

27. Anthony Atkinson, ‘On intergenerational income mobility in Britain’, 1981. Published on TANDFONLINE.COM. ↩

28. Economists typically fit a regression model of the form Y^Son = α + βY^parent + u (1), where Y^Son represents the son’s income (logged) and Y^parent represents the parents’ income (logged). α is the intercept, β is the regression coefficient representing the strength of association between parents’ and adult children’s income, and u is an error term. The regression coefficient (also known as intergenerational elasticity) has a natural interpretation. For example, a coefficient of 0.3 means that if 2 families have (log) incomes that differ by 10%, their sons’ (log) income will differ by about 3%. ↩

29. Jo Blanden and Stephen Machin, ‘Up and down the generational income ladder in Britain: past changes and future prospects’, 2008. Published on CAMBRIDGE.ORG. ↩

30. Bertha Rohenkohl, ‘Intergenerational income mobility in the UK: new evidence using the BHPS and Understanding Society’, 2019. Published on UNDERSTANDING SOCIETY.AC.UK. ↩

31. Jo Blanden and Stephen Machin, ‘Up and down the generational income ladder in Britain: past changes and future prospects’, 2008. Published on CAMBRIDGE.ORG. ↩

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

33. We should note that Rohenkohl’s estimates are based on multiple observations of parents and children’s income whereas we currently have only a single observation for each. Rohenkohl also controls for age and age square for both parents and children and for year of birth. In future work we will see whether these additions make any difference to our results. ↩

34. Significant at the 1% level. ↩

35. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

36. Additional analysis using only participants aged 28 to 37 years from the UK Household Longitudinal Survey (UKHLS) was conducted to allow for better comparability with Bell and others’ findings. However, it is important to note that the linked censuses that Bell uses and the UKHLS that we use, are not identical. The differences in these data sources mean that comparisons may not be wholly reliable. ↩

37. Intersectional analysis means looking at more than one characteristic at once, to check how they might interact. For example, we might look at both SEB and sex, and the interaction between them. ↩

38. 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. ↩

39. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

40. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

41. Low relative mobility can be thought of as ‘stickiness’, while high relative mobility mobility can be thought of as ‘fluidity’. ↩

42. A large confidence interval means that there’s a lot of uncertainty in the estimate. ↩

43. Ricky Kanabar and Paul Gregg, ‘Intergenerational wealth transmission and mobility in Great Britain’, 2022. Published on ONLINELIBRARY.WILEY.COM. ↩

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

45. Jo Blanden and others, ‘Trends in intergenerational home ownership and wealth transmission’, 2021. Published on CEP.LSE.AC.UK. ↩

46. Jo Blanden and others, ‘Trends in intergenerational home ownership and wealth transmission’, 2021. Published on CEP.LSE.AC.UK. ↩

47. Jamie Evans and Sharon Collard, ‘Facing Barriers: exploring the relationship between disability and financial well-being’, 2022. Published on BRISTOL.AC.UK. ↩

48. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

49. Brian Bell and others, ‘Where is the land of hope and glory? The geography of intergenerational mobility in England and Wales’, 2018. Published on CENTRE FOR ECONOMIC PERFORMANCE. ↩

50. See for example, Martin Hällsten and Max Thaning, ‘Wealth as one of the “big 4” SES dimensions in intergenerational transmission’, 2021. Published on ACADEMIC.OUP.COM. ↩

51. Equity release mortgages here would be a parent accessing more cash through their home. ↩

52. For a comparison of Wealth and Assets Survey results with those from other sources see, The Resolution Foundation, ‘The UK’s wealth distribution and characteristics of high-wealth households’, 2020. Published on RESOLUTIONFOUNDATION.ORG. ↩

53. Office for National Statistics, ‘Household total wealth in Great Britain: April 2018 to March 2020’, 2022. Published on ONS.GOV.UK. ↩

54. We use the same statistical method as in the case of relative income mobility (see previous section). That is, we take logarithms of both parents’ and adult children’s wealth (for those whose wealth was not 0). The degree of intergenerational persistence can then be estimated with a linear regression model as with income mobility. Alternative methods which take account of those with 0 wealth produced very similar estimates. ↩

55. To derive these estimates we use the Wealth and Assets Survey waves 6 (from 2016 to 2018) and 7 (from 2018 to 2020), adjusted for inflation, respondents aged 25 to 64 years, weighted data, 95% confidence intervals. Total wealth estimates for respondents are derived by adding up the value of different types of assets owned by households and subtracting any liabilities. Estimates of parental wealth are imputed using a 2-stage least squares method. ↩

56. This difference is statistically significant at the 0.01 level. Estimates derived from Wealth and Assets Survey waves 6 (from 2016 to 2018) and 7 (from 2018 to 2020), adjusted for inflation, respondents aged 25 to 64 years, weighted data, 95% confidence intervals. Note, the total wealth estimates for respondents are derived by adding up the value of different types of assets owned by households and subtracting any liabilities. Estimates of parental wealth are imputed using a 2-stage least squares method. ↩