Written evidence from Professor Melanie Jones and Professor Victoria Wass, Cardiff University (DEG0002)
We are two academic researchers who have a longstanding shared interest in disability inequality in the labour market and, as members of Disability@Work, a collaboration of four academics, we seek to make this research available to policymakers.[1] Our expertise is particularly in relation to the quantitative analysis of disability-related gaps in the UK. This submission is motivated by our concerns relating to the reliability of the indicator currently used to measure and monitor government policy, the disability employment gap (DEG), and our disappointment at the lack of progress in response to the recommendations made by the previous Department of Work and Pensions Disability Employment Gap Inquiry (May, 2016).[2] Our submission offers a critical reflection on previous and likely short-term progress on the DEG and makes several recommendations to ensure disability inequality in the labour market is more accurately monitored in the future by both government and employers so that the impact of policy and practice can be more accurately assessed.
Executive Summary
Our submission addresses four main points in response to the following two themes in the call for evidence.
- Trends in the Disability Employment Gap
- We offer a cautious and critical reflection on trends in the DEG and government evidence of narrowing (#1).
- We provide early evidence of a widening DEG post COVID-19 and predict a particularly difficult short-term labour market for disabled people (#2).
- Recommendations for the Government’s National Strategy for Disabled People
- We recommend a movement away from the exclusive focus on the DEG to multiple measures of disability prevalence and disability inequality at work as part of a more comprehensive national monitoring of disability labour market inequality in the UK (#3).
- We argue that in order to narrow the DEG, measurement and monitoring of disability inequality at a national level needs to be complemented by organisational (employer) monitoring (#4).
Context
The gap in employment rates between disabled and non-disabled people of working-age, the disability employment gap (DEG), is large and enduring in the UK. The Government pledged ‘to halve the disability employment gap [and] transform policy, practice and public attitudes so that hundreds of thousands more disabled people who can and want to be in work find employment’ (Conservative Party Manifesto, 2015). However, in light of the failure to meet this target, it downgraded its ambition in 2017 to a more manageable target at that time, to increase the number of disabled people in work by 1 million by 2027. We have been very critical of this change and the focus on an absolute measure, which depends on the economic cycle as well as the number of disabled people in the UK and, at the time, was likely to be met on the basis of prior trends alone.[3] We are therefore pleased to see the inquiry return to focus on the DEG as a relative and more accurate measure of disability inequality. Nevertheless, that many of the other recommendations from the 2016 Department of Work and Pensions Inquiry have not been acted upon is hugely disappointing and questions the purpose and effectiveness of these inquiries in addressing disability inequality at work.
Employment rates by disability status are typically estimated from the Labour Force Survey (LFS), a nationally representative household survey undertaken by Government. The data are subject to a number of discontinuities over time, which are critical to interpretation given the sensitivity of disability measurement to even small survey and question changes. These were discussed in our 2016 submission and we are pleased to see the LFS measure of disability has remained constant since then. We continue to emphasise the importance of continuity in the measurement of disability when tracking the DEG.
As shown in Figure 1, the trend in the DEG between 2013 and 2020 (the longest period over which we have consistent data) has been downwards, with a narrowing of less than 1 percentage point per year prior to COVID-19, meaning it would take more than 14 years to halve the current gap. This narrowing has been the focus of government attention and claims of policy effectiveness.
Figure 1: DEG and Smoothed DEG 2013-2020, LFS
Notes: Authors calculations based on the ONS economic activity of people aged 16-64. UK LFS 2013-2020. Equality Act disabled. The smoothed DEG is a rolling 4 quarter average to remove seasonal variation. The dashed green line denotes Oct-Dec 2019, the final quarter prior to COVID-19.
However, as researchers who have studied the DEG over a considerable period of time, we have serious concerns with the exclusive focus of the UK Government on a single indicator to measure disability inequality.
1. Disability measurement
Disability is typically measured using an activity-limiting definition of disability that is designed to be consistent with legislation. As a self-reported measure it is sensitive to changes in legal and social norms.
The fragility of self-reported activity-limiting disability measures was first illustrated by Baumberg et al. (2015) who demonstrated that narrowing trends in the DEG in the LFS 1998-2012 were not evident in two other government surveys (the General Household Survey (GHS) and the Health Survey for England (HSE)). This is illustrated in Figure 2.
Figure 2: Disability Employment Gaps in LFS, GHS and HSE 1998-2012
Source: Baumberg et al. (2015) Figure 2.
The most plausible explanation for this is the rise in disability prevalence, that is, the proportion of working-age people reporting disability has been rising steadily in the LFS, but not the other two surveys. There is a strong association between the level of disability and the disability employment rate because the people who move across the boundary between reporting and not reporting disability are likely to be less severely disabled (and hence have a higher probability of employment) than people with a high probability of reporting a disability.
We therefore recommend that the monitoring of disability-related employment inequality widens to include a set of indicators, which not only contain information on the prevalence of disability and comparisons across a range of surveys, but also functional limitations-based measures of disability to benchmark the Equality Act measure (see #3 below).
The total impact of disability on employment in the UK has not declined.
An alternative measure of the employment impact of disability is given by the combination of disability prevalence and the DEG. Disability prevalence in the LFS (as currently measured by the proportion of the working-age population classified as disabled according to the 2010 Equality Act) has been on a continuous upward trend from 16.4% in July-September 2013 to 19.9% in July-September 2020. As a result, even when using LFS data over a period of narrowing of the DEG, the combined measure, the prevalence adjusted DEG, does not show such clear evidence of narrowing in the UK (see Figure 3).[4] The total employment deficit as a result of disability has not therefore diminished. This suggests that the rise in disability prevalence contributes to the narrowing DEG in Figure 1 and questions the extent to which the relative employment prospects of the disabled population in 2013 would be superior in 2020 as is often incorrectly inferred by a narrowing DEG. In the context of COVID-19 the measure of prevalence x DEG has widened (see #2 below for discussion).
One possible explanation for the rise in disability prevalence is the broadened social interpretation of disability as awareness and acceptability of disability has grown. The increase in disability prevalence may therefore not reflect changes in underlying health and functional impairment and/or their impacts on activity-limitation, but be the result of changes in recognition, acceptance, law, policy and practice. This potentially confounds the ability of the DEG to measure progress on disability equality or the outcome of policy innovations. As such, we recommend further data collection in order to measure disability using multiple definitions, including measures based on functional limitations which capture more medical (objective) components of disability that can be used to understand why (Equality Act) disability prevalence has changed. This would fit best practice internationally. For example, the UK could use the questions recommended by the Washington Group on Disability Statistics.[5] It would also be desirable to look at medium-term trends by repeating a functional limitations module in the HSE, last administrated in 2001.[6]
Figure 3: DEG x Prevalence 2013-2020, LFS
Notes: Authors calculations based on ONS economic activity of people aged 16-64. UK LFS 2013-2020. Prevalence is measured using Equality Act disabled. The smoothed DEG x Prevalence is a rolling 4 quarter average to remove seasonal variation. The dashed green line denoted Oct-Dec 2019, the final quarter prior to COVID-19.
We therefore recommend that national monitoring of disability-related employment disadvantage in the UK includes a set of indicators, which not only include information on the prevalence of disability and comparisons using different surveys, but also include comparisons based on functional limitations-based measures.
2. COVID-19
The early evidence is consistent with COVID-19 having a disproportionate impact on the employment of disabled people in the UK and widening the DEG.
The vertical green lines in Figures 1 and 3 indicate the quarter of the LFS prior to COVID-19 (October-December 2019). Although it is very early to attempt to evaluate the differential labour market impact by disability the evidence suggests a reversal of the narrowing trend in the DEG in Figure 1 and an increase in the relatively stable recent DEG x prevalence rate in Figure 3. Therefore, even over a period of intense government employment support (the final quarter of the LFS available at the time of writing is July-September 2020) both these indicators have been on an upward trend over these three quarters. This is consistent with the argument that economic downturns have a disproportionate negative impact on the employment of disabled people, with US evidence from previous recessions finding that disabled people are ‘last hired, first fired’ (Kruse and Schur, 2003) and may suggest this is the start of a more pronounced trend. The different nature of the COVID-19 recession with its associated direct adverse impact on health makes the potential impact on the prevalence x DEG measure even more pronounced, increasing the importance of understanding the relationship between disability prevalence and the DEG (see #1 for discussion).
Our previous evidence based on nationally representative data showed that disability gaps in in-work indicators widened in the UK during the Great Recession (Jones et al., 2020).[7] Figure 4 reports the percentage of disabled and non-disabled employees affected by different recession-induced changes to their jobs. Disabled employees were significantly more likely than non-disabled employees to report increased workloads, work reorganisation, a wage freeze or cut, and restricted access to paid overtime and training. As such, the disproportionate employment disadvantage experienced by disabled people during a COVID-19 recession is likely to be reinforced by a disproportionate impact of recession-induced in-work changes for disabled people who remain in work, widening existing but neglected disability gaps in job quality (see #3 for further evidence).
Figure 4. Disabled Employees Experience of the Great Recession in Great Britain
Note: Data are from the Workplace Employment Relations Survey 2011. The darker bars indicate a statistically significant difference between disabled and non-disabled employees.
Our evidence suggests a disproportionate emerging and likely future labour market impact of COVID-19 on disabled people. Addressing this before it magnifies the long-standing disadvantage experienced by disabled people requires rapid intervention. In what follows we make two core recommendations to strengthen the National Strategy for Disabled People.
3. Future national monitoring
Collecting and reporting on disability and disability-related outcomes at the national level allows scrutiny of progress in relation to disability equality in the UK. It is therefore essential that such information is collected consistently and accurately, and that the statistics generated are interpreted meaningfully.
In addition to more robustly measuring disability and the disability-related employment impact (see #1) by undertaking comparisons of the DEG across surveys, monitoring and adjusting for disability prevalence and new data collection in relation to functional measures of disability from which to benchmark and understand the rising prevalence of disability, we recommend the government extend its focus beyond employment. First, we recommend that the government monitor employment flows, that is, rates of entry and exit from employment to better understand the underlying dynamic patterns that give rise to changes in the stock of employment. This is particularly important in the short-run given the impact of COVID-19 will be most immediately on flows. However, we also recommend the government extend its monitoring to in-work indicators consistent with a growing body of evidence on the range of dimensions of disability inequality in the labour market which are neglected by government policy. In particular, there is a sizeable disability gap in hourly earnings (Jones et al. 2006) and wellbeing at work as measured, for example, by job satisfaction and perceptions of fairness of managers (Jones 2016, Hoque et al. 2017). Not only are these measures important in their own right but they contribute to the recruitment and retention of disabled people and hence the DEG.
Given the availability of these measures in existing large and representative surveys, we recommend that a basket of indicators around disability and work are monitored, which cover disability prevalence and the experience of work, to provide a more comprehensive picture of disability inequality in the labour market and provide a broader foundation for policy.[8]
4. Organisational monitoring and reporting
Employers play a critical role in determining labour market equality for disabled people, a role historically and currently insufficiently recognised in government policy. In the majority of organisations disability is not measured or monitored which results in employers being unable to assess the impact of their policies and practices on disability inequality, or meet their obligations under the Equality Act. This will include the impact of substantial changes as a result of COVID-19, such as the government furlough scheme and homeworking.
The introduction of the Voluntary Reporting on Disability, Mental Health and Wellbeing framework, which recommends monitoring disability prevalence among the workforce provided a useful starting point, but requires significant amendment and more widespread utilisation if it is to be effective.[9] The most important and immediate change that is required is the integration of organisational and national measurement by ensuring a consistent Equality Act definition of disability is applied by organisations.[10] This not only ensures consistency between national and employer monitoring but in applying a common definition of disability it facilitates comparability between organisations. The latter is key if the government is to use such information, for example, as part of strengthening Disability Confident to include hard outcomes or integrating disability equality into Public Procurement in line with the recent consultation on the Social Value Act.[11] In doing so, the government would also then be in a position to implement mandatory reporting on disability prevalence for large employers, in a similar manner to Gender Pay Gap Reporting and the recent consultation on ethnicity.[12] Indeed, accurate and consistent measurement of disability is a prerequisite for the development of organisational measures of disability inequality e.g. in relation to recruitment and retention, or pay and job satisfaction gaps, from which to track progress. The government as an employer also has an important role in illustrating best practice in this regard.
We therefore recommend the government provides clear guidance to employers on measuring and reporting disability among their workforce on the basis of the 2010 Equality Act, that they adopt best practice as an employer and embed these measures in government policy to encourage widespread monitoring and reporting of disability equality by employers.
References:
Baumberg, B., Jones, M. and Wass, V. (2015). Disability and disability-related employment gaps in the UK 1998-2012: different trends in different surveys? Social Science & Medicine 141, 72-81.
Hoque, K., Bacon, N., Jones, M. and Wass, V. (2017). Are high performance work practices enabling or disabling? Exploring the relationship between selected HPWPs and work-related disability. Human Resource Management, 57(2): 499-513.
Jones, M. (2016). Disability and perceptions of work and management, British Journal of Industrial Relations, 54(1), 83-113.
Jones, M., Hoque, K., Wass, V. and Bacon, N. (2020). Inequality and the economic cycle: disabled employees’ experience of work during the Great Recession in Britain, British Journal of Industrial Relations, forthcoming.
Jones, M., Latreille, P. and Sloane, P. (2006). Disability, gender and the British labour market, Oxford Economic Papers, 58(3), 407-459.
Kruse, D. and Schur, L. (2003) Employment of people with disabilities following the ADA. Industrial Relations, 42(1), 31-66.
November 2020