February 16, 2024
The introduction of the National Living Wage (NLW) in April 2016 marked the start of a series of increases in the level of the UK minimum wage which have brought it close to the international frontier. The ambitious level and coverage of the NLW have made it the flagship policy aimed at helping low-paid workers in the UK. In both the UK and internationally, there is broad consensus that minimum wages are effective at raising wages at the bottom of the distribution, as intended. But this compression of the wage distribution – commonly referred to as a reduction of ‘pay differentials’ between different jobs – may also create challenges for firms, above and beyond the direct cost implications of raising wages for the lowest paid. Here we briefly set out evidence on the extent of wage compression within firms caused by the UK’s NLW.
The compression of the wage distribution that has been observed across the UK labour market is clearly visible within firms, and primarily for those firms employing larger fractions of their workforce at low levels of pay. This can be seen in Panel A of Figure 1, which plots the average change in the share of firms’ workforce (vertical axis) paid at each wage level (horizontal axis), comparing the years before the NLW introduction (April 2012 to April 2015) with those after (April 2016 to April 2019). The chart is based on a sample of large firms that employed at least 20 per cent of their workforce below the 2016 NLW in April 2015, i.e. a group of firms that were heavily exposed to the introduction of the NLW. We restrict our analysis to large firms because in the UK we only have survey data with wage information on a subset of firms’ employees, meaning for small firms we cannot confidently measure their exposure. Still, our large firm sample covers 46 per cent of employees in the UK, with one fifth of these in firms that were heavily exposed to the NLW introduction (Panel A).
Figure 1. Change in within-firm employment share by wage bin, pre- and post-NLW (ppts)
Panel A: Heavily-exposed firms
Panel B: Less-exposed firms
Note: Sample of large firms (at least 20 observations in 2015 ASHE) corresponding to 46 per cent of all worker-year observations. Pre-NLW refers to 2012-15, post-NLW refers to 2016-19. Panel A is based on a sample of firms where NLW coverage in 2016 was at least 20 per cent (19 per cent of our sample), Panel B is based on a sample of firms with coverage below 20 per cent (81 per cent of sample). Error bars show 95 per cent confidence intervals. The x-axis measures distance to the 2016 NLW in 2016 prices (CPI).
Panel A of Figure 1 shows a sharp reduction in the share of employees paid below the 2016 NLW and a large increase in the share paid at the 2016 NLW itself. Interestingly, we can also see a smaller, though positive increase in the fraction of workers paid one and two pounds above that minimum, with no detectable change at higher wage levels. Panel B is constructed in the exact same way as Panel A, but uses a sample of large firms with limited exposure to the NLW introduction, namely firms that employed less than 20 percent of their workforce below the 2016 NLW in 2015. Here effects are much more muted throughout the wage distribution.
Three key observations emerge from Figure 1. First, the introduction of the NLW generated substantial within-firm wage compression at the bottom of the distribution. Second, this effect is more pronounced among heavily exposed firms. Third, following the introduction of the NLW, wage increases occurred not just at the level of the new minimum, but also up to two pounds above it. Thus, employers who employ significant numbers of minimum wage workers also raise the wages of some workers above the minimum. These “wage spillovers” mitigate the overall impact on pay differentials between jobs close to the bottom of the national wage distribution (though within low-wage firms these effects can stretch a long way up the within-firm wage distribution, as shown below). In additional analyses, we found that spillover effects tend to be larger among incumbent employees than among new hires, and larger in public or unionised firms. Some care should be taken inferring the exact size of the spillovers from these charts, since subsequent increases in the NLW between 2017 and 2019 mean the real distance between the current NLW and its 2016 level is positive rather than zero (although less than one). This means spillovers after 2016 may be overstated slightly.
We also investigate the effect of the introduction of the NLW on wage ratios within firms. Wage ratios are a measure of wage inequality between workers at different points in the firm’s wage distribution. In order to do this, we divide firms into four groups depending on the ‘coverage’ at the NLW introduction. Here, coverage measures the share of workers in a firm who were paid below the 2016 NLW in 2015. Table 1 shows the effect of the introduction of the NLW on the wage ratios of firms in each group, relative to the least exposed group of firms (those with less than 5 per cent coverage).
We see that the NLW led to a decline in the 90-10 and 50-10 ratios for less-exposed (5-20 per cent coverage) and moderately-exposed (20-40 per cent coverage) firms, but not in the highly-exposed firms (40-100 per cent coverage). This is because the majority of workers in the highly-exposed firms earn at or near to the minimum wage, and so there are limited differentials to be reduced. There is evidence of a fall in the 90-50 ratio for the moderately- and highly-exposed firms, showing that the effects can extend far up the wage distribution in firms with many minimum wage workers. It is an important reminder that, in those firms where many of the impacts of the NLW actually play out (i.e. highly-exposed firms), and whose reactions will matter most in shaping the impacts of the NLW, it is by no means merely a matter of bringing the very lowest-paid workers closer to the rest. These firms are typically responding to a shock much more substantial than that.
Table 1. Change in within-firm wage ratios relative to ‘least exposed’ firms (< 5 per cent coverage)
Note: Changes in levels of the ratio. Sample of large firms (at least 20 observations in 2015 ASHE) corresponding to 46 per cent of all worker-year observations in ASHE. Standard errors shown in parentheses. These stars indicate whether the results are statistically significant at different critical values (denoted by p). *p<0.10 **p<0.05 ***p<0.01
Employers worry about shrinking wage differentials since they can reduce workers’ incentives to do, or progress into, more demanding roles. Restoring wage differentials is however costly. Depending on their ability to adjust pay scales, firms might respond to wage compression in different ways. Those that can cope with wage bill increases might partly restore wage differentials by raising the pay of workers above the minimum wage. We see evidence of these wage spillovers in Figure 1, though we only observe substantial spillovers up to £2 above the incoming rate.
Survey evidence for the UK suggests that some firms have responded to pay compression by changing their organisational structures and cutting back on job ranks above the minimum to maintain pay differentials between staff levels (Low Pay Commission, 2020). To capture the extent of changes in occupational structure, we group occupations into 10 deciles based on the nationwide median wage in that occupation before the introduction of the NLW. We then use the share of employment in each of these occupation-wage groups as a proxy for the occupational structure of the firm. Figure 2 shows the association between the coverage of the 2016 NLW in 2015 and in the change in employment share from before to after the introduction of the NLW. We do not see significant differences, suggesting that the introduction of the NLW did not lead to any changes in firm structure by this measure. This does not rule out the possibility of more subtle changes – wage hierarchies are difficult to observe in survey data containing only a subset of employees, and the removal of grades on a pay scale may not show up in occupational changes in the data.
Figure 2. Association between NLW coverage and change in employment share by occupation wage decile
Note: Sample of large firms (at least 20 observations in 2015 ASHE) corresponding to 46 per cent of all worker-year observations in ASHE. Pre-NLW refers to 2012-15, post-NLW refers to 2016-19. Error bars show 95 per cent confidence intervals. Occupation wage decile based on pre-NLW median wage for each 3-digit SOC code.
The evidence presented in this piece shows how minimum wage increases can have a significant impact on highly exposed firms. There are various ways they might respond to the large cost shock and/or the change in their pay differentials – for example, by adjusting their pay setting, contractual arrangements or organisational structure. Those responses, which remain relatively under-studied, would potentially have their own important implications for the careers of low-paid workers.
References
Low Pay Commission. 2020. “National Minimum Wage Low Pay Commission Report.” Low Pay Commission.
This blog summarises a piece of research on the impact of the NLW on differentials commissioned by the Low Pay Commission. It is written by the researchers listed at the top of the article. We will publish the full research article in late 2024.
]]>
Since it was introduced in 2010, the Apprentice Rate has gone up faster than any of the other minimum wage rates. On 1 April 2024 it will increase again to £6.40.
The face of apprenticeships has changed over the lifetime of the rate, with an increasing share of older apprentices pursuing higher-level qualifications, including degree-level apprenticeships. The vast majority of apprentices are paid more than the Apprentice Rate (LPC Report, 2022). However, the rate remains the standard for some apprenticeships and sectors.
We want to make sure we hear evidence from all those affected by the minimum wage rates, including apprentices. So, as part of our ongoing research programme, we commissioned researchers from Sheffield Hallam University to carry out in-depth interviews with current and former apprentices. This complements our wider programme of evidence gathering, where we have heard from apprentice employers and organisations representing apprentices, as well as those working closely with apprentices and young people.
The resulting report is published today and gives a rich insight into the experiences of the apprentices interviewed. As the researchers highlight, the voices in the report reflect just some of the possible apprentice experiences – the research was not designed to be representative of all apprentices. Nevertheless it is an important contribution to our understanding of the workings and impacts of the Apprentice Rate.
For the remainder of this blog, we hand over to Charlynne Pullen, Bob Jeffery and Teri-Lisa Griffiths – the researchers who carried out the project – to tell us what they found.
***
Our research is based on 10 in-depth interviews either individually or paired, with current or former apprentices who were paid the apprenticeship minimum wage at some point during their apprenticeship. It was a significant challenge to find participants, and this group is only representative of certain kinds of apprenticeship minimum wage experiences. Nonetheless, they do offer some striking insights.
Some of these were unsurprising - none had been given a positive view of apprenticeships by their schools, and as a result most had found their apprenticeships through an online advert, directly through the apprenticeship employer, or for one young person, at a college open day. As a result, the idea that individuals systematically review a series of options and then select the most appropriate apprenticeship for them would appear to be misleading.
The most important thing for all the interviewees was that their apprenticeship would lead to a stable and secure job and for most a clear career path requiring additional training – “a foot in the door”. This concept of prospects meant many accepted the low wages at the beginning of the apprenticeship. However, for those living independently, typically those over 18 or in care, the reality of trying to pay rent and bills on the apprenticeship minimum wage meant claiming benefits where possible, asking for family support, and in one case, accessing food banks.
Most apprenticeship experiences with employers and training providers were positive, particularly with individual trainers and tutors. However, some apprentices experienced employers who exploited them, either failing to provide training, or by asking apprentices to undertake roles that were inappropriate, and training providers delivering poor training, or in one case closing during an apprenticeship. Places where apprentices could come together, either for in-person training for hairdressing or beauty therapy, or cohorts of apprentices meeting in larger employers, were important to create a community and keep apprentices engaged.
Progression was key for most apprentices, with some committed to a particular job role at the start – for example accountancy or hairdressing – while others chose Business Administration (or more recently Customer Service) as a way of learning about a range of different career options before specialising. All of the apprentices could see a future career pathway that involved at least a Level 3 apprenticeship, and some of the former apprentices were already on higher level apprenticeships, with one having completed a degree.
The clearest message from this research is that a Level 2 apprenticeship on the apprenticeship minimum wage, at its best, can be an effective stepping stone into a career pathway. It can suit individuals with a range of prior experiences, qualifications and ages, but pay needs to be set high enough to ensure that individuals can provide for themselves and have an incentive to stay the course.
[1] All apprentices aged 19+ must be paid the relevant minimum wage rate for their age from the second year of their apprenticeship onwards. Throughout an apprenticeship, the relevant minimum wage must be paid for all hours worked, including off-the-job training time. For further information, visit [https://www.gov.uk/become-apprentice/pay-and-conditions]
]]>If you want to participate, get in touch with us if you would like to meet our Commissioners. You can email at lpc@lowpay.gov.uk or call us on 07511 162 380.
Where we will be this year:
Date | Location |
20-21 March | Isle of Wight/ Southampton/ Portsmouth |
17-18 April | Dover |
15-16 May | Cardiff/ Newport |
12-13 June | North Ayrshire and Glasgow |
3-4 July | Bangor |
31 July- 1 August | Coventry/ Wolverhampton |
We select our locations based on their minimum wage coverage and unemployment levels. Our guiding aim is to listen to a diverse range of people from across the UK and visit locations we have not been to before or recently.
The LPC takes very seriously what we hear on all our visits, so it is a chance for anyone with an interest to have their voice heard.
Who do we want to hear from:
We are looking to meet with people affected by the minimum wage in all of these locations: workers, employers, local authorities, charities and anyone else with an interest. We are happy to come to you directly or equally to host a meeting ourselves in your area.
What do we want to hear about:
From workers -
From businesses -
Our annual written consultation will be launched shortly, detailing all the areas we want to gather evidence on.
Why we go on regional visits:
Since 1998 the Low Pay Commission (LPC) has made recommendations to the Government on the levels of the National Minimum Wage (NMW). It has also reported and produced recommendations on related policy matters including enforcement and compliance.
To inform these recommendations, we consult widely across the economy with businesses, workers, local authorities, charities, and other interested parties.
We produce evidence-based advice, and our regional visits are an essential part of our evidence gathering cycle. Through visiting different areas of the UK, we speak with workers on or close to the minimum and living wage as well as the employers who are directly affected by the change in rates.
The real-life experiences of those in low-paid work and of those employers in low-paying sectors are hard to understand from the statistics alone. By talking face to face with the people most affected by our recommendations, Commissioners can understand what is happening on the ground and the themes and issues which are emerging.
On our visits in 2023, we heard valuable evidence from workers on the cost of living, insecure employment and work intensification. Employers told us they had responded to the NLW increases by increasing prices, reducing profits and pay differentials and smaller firms reduced investment. These insights enable the LPC to better understand the impact of our work on both workers and employers.
Last year, we focused on compliance and enforcement and presented data on underpayment and explored the persistence of underpayment for individual workers. The evidence we heard formed our report on Compliance and enforcement of the National Minimum Wage.
If you want to participate, get in touch with us if you would like to meet our Commissioners. You can email at lpc@lowpay.gov.uk or call us on 07511 162 380.
]]>But what do we mean by low-paid labour market? We want to identify the businesses and workers who are most affected by the minimum wage. This doesn’t mean looking only at those paid at the minimum; changes in the minimum wage have spillover effects on the wages of those paid some way above it as well.
To help us identify these workers and businesses, we use two definitions: low-paying occupations relate to job roles that are often low-paid – for example, ‘sales assistants’; low-paying industries are based on the main activity of the employer – for example, ‘retail trade’.
These definitions were last updated in 2017, shortly after the introduction of the National Living Wage (NLW). A lot has changed since then: the level of the minimum wage has increased rapidly, potentially changing the types of workers and businesses affected by it. The ONS has also updated how it classifies occupations, moving to a new set of standard occupational codes (SOC 2020) in the datasets we use. This move was completed for the Annual Survey of Hours and Earnings (ASHE) – our main data source for hourly pay – in autumn 2022.
To make sure our work keeps up with these changes – and remains relevant once the NLW meets its target in 2024 – we have reviewed and updated our definitions of low-paying occupations and industries. This blog outlines the changes we’ve made. We have also published tables with full details of the new occupation and industry groups, available here.
There is considerable continuity between our old and new definitions. Many of the occupations and industries most affected by the minimum wage in 2017 continue to be amongst the most affected today. Retail and hospitality remain the occupation and industry groupings with the largest number of minimum wage workers, and the new definitions do not change the overall trends in employment in low-paying sectors (see Figure 1).
The most significant change is the addition of two new occupation groups covering low-paying roles in education and healthcare, such as teaching assistants and hospital porters. These groups include a large number of workers paid just above the current NLW. We expect these workers will continue to be paid above the minimum, but their rates of pay are still likely to be influenced by the NLW.
We have also added several new industry groups, including transport and non-food processing. The new industry groups are small and don’t significantly change the scale or characteristics of the low-paying industries when taken together. A full list of the revised occupation and industry groups can be found at the end of this blog.
As Figure 1 shows, more people are employed in low-paying sectors under the new definitions. Occupations see the largest change. The new education and healthcare groups explain around half of this increase, with most of the remaining increase explained by new job roles added to the ‘office work’ occupation group.
This doesn’t mean that there are more low-paid workers in the workforce – or that pay of workers in the new occupations has got closer to the minimum wage. In fact, many of the new groups were already considered ‘borderline’ in the last review in 2017. In some cases, the change to SOC 2020 has allowed us to split lower- and higher-paid groups that were previously classified together – such as teaching assistants and higher-level teaching assistants. But our new approach is also intentionally broader to reflect what we know about spillovers from the NLW.
This means we capture more minimum wage jobs: over 80 per cent of minimum wage jobs are now in a low-paying occupation (Table 1). On the other hand, we also capture more better-paid workers. This slightly reduces the share of jobs in the new low-paying occupations that are paid the minimum wage (‘coverage’). However, the majority of these better-paid jobs are still paid close enough to the minimum wage to be affected by it.
As the NLW has increased, it has pushed wages up – not only for those on the minimum wage, but also those paid some way above it. Research we commissioned from the IFS (Cribb et al, 2021) suggests that these spillovers affect wages up to £1.50 above the NLW. We took this into account when we defined the new low-paying sectors.
Figure 2 compares the share of jobs paid below this higher in each of the new low-paying occupations to the share covered by the minimum wage. The size of the bubbles illustrates the number of jobs paid less than £1.50 above the NLW. This shows how the two new occupation groups – education and healthcare – have low minimum wage coverage compared to the other occupation groups, but higher broad coverage (NLW + £1.50) than other established occupation groups, such as non-food processing, office work and call centres. In this they are similar to the existing social care group.
The differences between workers in low-paying and other occupations and industries are not limited to pay. Part-time work, zero-hours contracts and temporary contracts are more common for those working in low-paying occupations, while small firms are more likely to be in low-paying industries. This remains true under both old and new definitions.
The new definitions increase the share of women falling into low-paying occupations – where they were already over-represented. They also capture more public sector workers, although the private sector continues to predominate in both low-paying occupations and industries.
There is a lot of overlap between low-paying occupations and low-paying industries – for example, just under 60 per cent of jobs in low-paying occupations are in low-paying industries. But occupations and industries do not map directly onto one another: there are many workers in low-paying occupations within non-low-paying industries (such as a call centre worker at a utility company) and vice versa. By the nature of occupation and industry definitions, low-paying occupations tend to have a higher concentration of workers paid close to the minimum wage than low-paying industries.
We will use the new low-paying occupation and industry definitions in our work going forward. While we transition, we’ll continue to compare them to the old definitions and provide figures under both where necessary.
If you’d like to know more, you can find a full list of the occupations and industries included in the new groups on our website, alongside an overview of the methodology we used. We’ll also feature more on the new low-paying sectors in our forthcoming 2023 Report.
]]>
As part of our ongoing monitoring of the impacts of the National Minimum Wage (NMW) on different groups of workers, the Low Pay Commission has regularly reported on trends in minimum wage coverage rates (the share of workers paid at or below the minimum wage) and employment rates by ethnicity, disability, and gender.
However, the main data sources used in that reporting – the Annual Survey of Hours and Earnings (ASHE) and the Labour Force Survey – have limited the extent of the analysis that we conduct. ASHE has few personal characteristics (just age and gender) while the Quarterly Labour Force Survey is limited in sample size.
We know from our reporting that women, ethnic minorities, and those with disabilities have lower average rates of employment but when employed are more likely to be paid the minimum wage. That suggests that, although these groups may be most likely to gain from minimum wage increases, they may also be at greater risk of job loss as a result of those increases. McKnight (2022) analyses data from the decade prior to the pandemic (2009-2019) to explore changes in minimum wage coverage and employment rates. She also investigates the impact of the introduction of the National Living Wage (NLW) on employment retention and pay progression.
The study uses data from two worker surveys – the Annual Population Survey and Understanding Society – which both have booster samples to increase the number of responses from ethnic minorities. The author makes adjustments to the data to account for issues with worker-reported estimates of hourly pay.
The research confirmed previous findings that minimum wage coverage is higher among women (than men), those with disabilities (than those without) and among Black, Asian and Minority Ethnic (BAME) employees (than non-BAME employees).
Minimum wage coverage was considerably higher among women by disability and ethnicity. Indeed, minimum wage coverage was higher among non-disabled female employees than among disabled male employees. However, the coverage gap (the disparity in coverage with their non-BAME counterparts) was greater for BAME men than women as non-BAME men had very low minimum wage coverage.
Across different types of functional impairment and ethnicity, minimum wage coverage varied. Among those with disabilities, minimum wage coverage was lowest for those with hearing difficulties and highest among those who did not recognise when they were in physical danger, or who had communication or speech difficulties. Among ethnicities, Bangladeshi and Pakistani employees had the highest coverage with Indian employees having similar coverage to non-BAME employees.
In the period prior to the introduction of the NLW, as minimum wages had gradually increased, coverage rates had also generally increased. Increases were greater for those with disabilities than those without. There were also increases across ethnic groups. This lack of increased coverage after 2016, despite the large increases in the NLW, is in line with our findings reported in various reports since 2017.
Employment rates had also generally been increasing in the decade before the pandemic. Employment rates among men and women with disabilities followed this pattern, with increases continuing after the introduction of the NLW. There was some evidence that the disability employment gap had fallen. BAME employment rates are lower than non-BAME ones with the employment gap larger for women than men. BAME employment rates also increased over the last decade, particularly for women (and especially for Bangladeshi and Pakistani women) leading to a narrowing of the employment gap.
The study used econometric techniques to estimate the impact of the NLW on employment retention looking at variations by disability, ethnicity and gender. , the study found statistically significant positive effects on employment retention for workers with disabilities but these became insignificant when controlling for age and gender. The authors also found no significant negative effects in alternative specifications.
There were no negative effects associated with BAME workers in aggregate. This was also the case for Bangladeshi and Pakistani workers, who might have been expected to be the most affected by the large increases in the NLW as they have the highest coverage. However, there was a significant negative employment effect found for Indian employees, particularly men.
The research found that wage progression overall appeared to have been affected by the introduction of the NLW in 2016. That is, from one year to the next, there were significant falls in upward mobility (leaving a minimum wage job for a high-paid job) and an increase in the likelihood of remaining in a minimum wage job. This effect was stronger for men but did not appear to hold for BAME employees or those with disabilities.
In the main specification, the research found no statistically significant change in annual transitions out of minimum wage jobs (including following the introduction of the NLW) for BAME employees or those with disabilities. However, in an alternative specification there was some evidence of an increase in the likelihood of remaining in a minimum wage job for both groups.
The research concludes that increases in the minimum wage in the decade before the pandemic had largely been beneficial to those with disabilities and for ethnic minorities. Increases in the minimum wage have led to higher rates of pay but don’t appear to have harmed job prospects. It reiterates the importance at looking beyond broad categories of disability and ethnicity. However, it also highlights the data limitations of attempting to do so.
One potential way of addressing that concern is to link the ASHE data to the 2011 Census, which has been done by the Wages and Employment Dynamics (WED) research project. This potentially enables us to identify personal characteristics of workers in the ASHE sample. We are currently exploring these data, working with the WED team, and hope to provide additional information on the impact of the minimum wage on different groups of workers in time for our 2023 Report.
]]>Since 2016 the National Living Wage (NLW) has increased the wage floor further and faster than in preceding periods. The evidence suggests this hasn’t cost jobs overall, but the risks are not evenly distributed. We know that certain groups of workers are more likely to work in low-paid jobs and to be paid less than their counterparts. This is the case for women, for some ethnic minorities and for disabled workers. We track employment rates for these groups, but there is room for more sophisticated approaches to assess whether the minimum wage actively reduces inequality between groups.
We commissioned researchers at the London School of Economics to develop one such approach. They used data from a single (anonymous) employer to examine what happened when the firm varied its rates of pay in different parts of the UK. The employer delivers services for local authorities, some of whom require their contractors to pay the Living Wage Foundation’s rates, while others do not.
The researchers had access to the company’s complete HR records. This rich dataset enabled the researchers to study exposure to an effective minimum wage set at a level considerably higher than the NLW. This provides clues as to how firms may react to a higher minimum wage. About one in five of the company’s workers were classified as Black, Asian and Minority Ethnic (BAME), meaning there was room to examine in detail how different wage levels affected their treatment.
The study found that in the areas where the Living Wage Foundation’s rates were used, entry-level workers earned about 7 per cent more than those in establishments not subject to the higher rates. The introduction of the Living Wage increased average wage bills by 1.5 per cent but reduced the BAME wage gap within the organisation by around half, from 7 per cent to 3.5 per cent.
Despite the increase in the wage bill, the research found no evidence of negative employment effects, whether measured by the number of positions, headcount employment, hours and the BAME employment gap. Instead, the evidence suggested some positive effects in terms of number of positions and headcount employment. The researchers suggested this was consistent with monopsony, which is where employers have more wage-setting power than they would in a competitive labour market. That conclusion was further strengthened by establishments In a more competitive environment, the opposite would be expected - an increase in the relative wages of entry-level workers would be expected to lead to a reduction in their relative employment.
The introduction of the Living Wage also led to a reduction in the number of points on the pay scale, compressing the pay distribution. There was also evidence of spillover effects on to non-minimum wage jobs. However, there was no evidence of differences in the likelihood of promotion between the Living Wage establishments and others.
Overall, the results from this study are encouraging. They suggest that a higher minimum wage has aggregate benefits for BAME workers with minimal employment losses. However, it should also be noted that these results are estimated from data for a single firm and so may not be generally applicable.
For more findings, including how the firm responded in other ways, including progression, pay structures, skill composition and how these varied by ethnicity, see the full report (Datta and Machin) here.
]]>Our main job at the Low Pay Commission is to recommend future National Minimum Wage (NMW) rates. Under the NMW, there are different rates for different age groups and a separate rate for apprentices.
The National Living Wage (NLW) is the highest of the National Minimum Wage rates. It currently applies to all workers aged 23 and over (excluding first year apprentices) and from April 2024 is due to apply to all workers aged 21 and over.
Since 2016, the Government has asked us to make NLW recommendations in line with a target. In 2015, the Government asked us to raise the NLW to 60 per cent of median hourly earnings by 2020. When the NLW hit that target, the Government set a new one for the NLW to reach two-thirds of median hourly earnings by 2024.
Making recommendations in line with these Government targets is more complex than it might first sound. First, we need to establish what the target is exactly. Since the NLW was introduced, we have aimed for a target in October of each year (e.g., two thirds of median earnings by October 2024). October is halfway through the minimum wage year and so is the best proxy for a whole year average. Were we to aim for April instead, when the rates come in, we’d likely be below target for most of the year as the NLW stayed constant but average earnings continued to rise.
The key challenge is forecasting what median hourly wages will be in October 2024. We use three steps to get our on-course rate estimate. These are shown in Figure 1.
We take the most recent estimate of median hourly earnings excluding overtime from the Annual Survey of Hours and Earnings (ASHE), an employer survey of 1 per cent of employees in PAYE as our baseline. The latest (April 2022) estimate of median hourly pay for workers aged 21 and over was £14.90 (the navy dot on the chart below). By the time we recommend the rate in October 2023, we will have new ASHE data so our baseline will be median hourly pay in April 2023.
Next, we estimate pay growth from the baseline to October 2024 (the target date). A challenge with this stage is that there are no sources of hourly pay data that allow us to do this. So instead, we use two sources of weekly pay data as proxies. We project growth in line with Average Weekly Earnings data (12-month-on-12-month growth), where this is available (currently up to January 2023). This is the light blue line in Figure 1.
We then use forecast growth in average weekly earnings to project further out to the target date. We use the median of forecasts by HMT’s independent panel of forecasters, adding the OBR and Bank of England forecasts to the panel. This is the purple line in Figure 1.
This step gives us an estimate of median hourly pay in October 2024 (orange dot on chart opposite). To reflect uncertainties in the forecasts we also estimate a range where pay growth rates in the projection period are 1 percentage point a year lower or higher than our central estimate.
Finally, we project the rate of the NLW forward. Given we are in the final year of our target, we do this by simply taking two-thirds of our estimate of median hourly pay in October 2024. Our central estimate of the 2024 on-course rate using this approach is £11.16 (the yellow dot on the chart). Figure 2 (below) shows the range (£10.90-£11.43) around our central estimate, which reflects uncertainty over wage growth.
This year our job of projecting median hourly pay forward is particularly difficult. Pay growth has increased rapidly in the last two years and is now forecast to fall (see Figure 3). We have increased the range around our path estimates to account for increased uncertainty around wage forecasts.
The recent large changes in pay growth also make apparent two issues with our on-course rate calculation, which we discuss below. We have developed experimental adjustments for these issues (see Figure 2), which, if applied would reduce our on-course rate estimate by 12 pence.
Issue 1: We use the timeliest available pay forecasts available: the median of a panel of pay growth forecasts compiled by HMT. However, these forecasts are only available for the growth in calendar year average pay, not the October to October growth we need. We apply these monthly by assuming growth is even across the year (accounting for compounding). However, when pay growth is forecast to change rapidly, our approach effectively lags the impact of these rapid changes on our on-course rate estimates. Given that pay growth is forecast to fall, this issue currently is pushing our on-course rate estimate upwards. If we were to apply our experimental adjustment for this issue, it would reduce our on-course rate estimate by 12 pence.
Issue 2: Part of our path calculation uses a 12-month smoothed average of Average Weekly Earnings (AWE) total pay to account for the latest available data on pay. We smooth to take out volatility in the AWE. However, smoothing requires relying on a longer period of historic data. This can mean our smoothed average responds more slowly to recent changes in pay growth. Currently this issue has no effect on our headline path estimate. We will continue to monitor it between now and October when we make our recommendations.
These issues make a difficult job (forecasting future wages) even harder. While we’ve developed adjustments for these issues, they are experimental and add month-to-month volatility to our on-course rate estimates. Stability and predictability are important to not only help employers plan for future upratings but also to help Commissioners reach agreement on the appropriate rate to recommend.
There is no easy solution to the current issues with forecasting pay, it is a matter of a judgement. The Low Pay Commission will take a decision in October this year when we recommend next year’s rates. Until then, we will continue to monitor pay data and forecasts closely.
Finally, the Government has asked us to consider wider economic factors in our recommendations. We can advise the Government to “adjust the target” if the “economic conditions warrant it.” We will base our recommendations on evidence of the economic impacts of the UK’s minimum wages and the broader economic context. You can share your views with us in our consultation here, which is open till 9 June 2023.
For source notes see our uprating report here.
]]>We commissioned researchers from London Economics to look into whether this change had affected how 23 and 24 year olds fared in the jobs market (Lee, Manly, Patrignani and Conlon, 2022). They faced the difficult task of disentangling the effects of the pandemic and the end of the Brexit transition period from the impact of the change in entitlement to the NLW.
The researchers’ main finding was that the change in entitlement to the NLW did not affect how likely 23-24 year olds were to have a job. While employment remained strong, the researchers also found that – compared with their slightly older peers – 23-24 year olds in low-paying occupations worked fewer hours on average following the change.
How should we assess these findings? In this blog, we explain the method used by London Economics and how we can interpret the results. We also talk about the implications for further reducing the age of NLW entitlement to 21 in 2024. Overall, the findings support what we have heard from stakeholders – that 23 and 24 year olds have done well in the labour market since moving to the NLW.
This blog focusses on the main findings of the research. For a full account of the research findings and methodology, you can find a summary in Appendix 2 of our 2022 Report or the full research paper can be found on our website.
London Economics compared outcomes for 23 and 24 year olds before and after April 2021 with outcomes for 26 year olds over the same period. They picked 26 year olds because they had similar trends in key labour market outcomes (such as employment) in the years leading up to the change in age threshold. Although closer in age, 22 year olds and 25 year olds did not always follow the same patterns.
The research hinges on the fact that 26 year olds experienced all the events that 23 and 24 year olds experienced around April 2021, except for the large change in their minimum wage. By comparing two groups who usually follow a similar path, and who are affected equally by events in the wider economy, the researchers can isolate the effects of a policy that changes wages more for one group than the other. In reality, of course, the groups will have differences – for example, they may be more or less likely to work in sectors affected by lockdowns during the pandemic. The researchers take this into account, and ‘control for’ the potentially different impact of region, sector and a range of personal characteristics, such as education level. Importantly, this means that when we talk about changes for 23 and 24 year olds, these are changes relative to (similar) 26 year olds.
Most importantly, we should see hourly earnings for 23-24 year olds go up faster than for 26 year olds. We would expect the minimum wage to have a direct effect on wages. It is hard to imagine it has indirect effects on employment or hours, without first having a direct effect on wages.
However, the focus of the research is on some of the possible ‘side effects’ of the policy change. For example, following the change employers may have decided to employ fewer 23 and 24 year olds or employ 23 and 24 year olds for fewer hours. Similarly, some 23 and 24 year olds may have chosen to work more or less.
We would expect any effect to be small. Even before the change, close to 90 per cent of 23 and 24 year olds were paid at or above the NLW, and for these workers the policy change would not have had a direct effect on their wage.
There was no evidence of an ‘employment effect’. The researchers found that 23-24 year olds were no more or less likely to be employed in the year after they became entitled to the NLW. This finding was consistent across different variations of their model and held for most subgroups.
23 and 24 year olds with jobs in low-paying roles or sectors were more likely to be working part-time following the change. For 23-24 year olds as a whole, the researchers did not find any change in either hours worked or earnings. When they looked at those employed in job roles that are usually low-paid (and so are more likely to be affected by the minimum wage), they found that 23 and 24 year olds were more likely to be working part-time – and therefore fewer hours – than before the change. This primarily affected women.
The charts below show how many hours 23-24 year olds and 25-26 year olds worked per on average in the two most important groups of low-paid jobs for young people: retail and hospitality. We can see that average hours worked by 23-24 year olds remained below their pre-pandemic levels for longer than they did for 25-26 year olds. At the same time, 25-26 year olds employed in retail occupations began to work more hours than they had prior to the pandemic.
See below for source notes.
The team at London Economics carefully designed their methodology to try to identify whether the minimum wage change had caused changes for 23 and 24 year olds. However, there are some important caveats to their findings.
We don’t always find the earnings effect that we would expect if the minimum wage caused the change. For example, the strongest negative effect on hours worked is found for women in low-paid occupations (an 8 per cent reduction in hours worked on average), but the increase in hourly earnings for these women is not significantly different from zero. For their male counterparts, meanwhile, there is a significant increase in hourly earnings, but no significant reduction in hours worked. In other cases, reduced hours and increased earnings are found together. The fact that the relationship is not consistent makes it difficult to be sure that there is a genuine causal link between the two.
It is possible for two groups close in age to be differently affected by events – even after applying controls. The researchers compared 23 and 24 year olds to 26 year olds because 25 year olds already had different trends in employment and hours (even after accounting for different characteristics). This shows that closeness in age does not guarantee a similar experience, even without a policy change. This could be because there are differences between the two groups that we can’t measure in the available data.
In fact, when we look at the analysis of young people in low-paying job roles, we see that there was already a significant fall in the hours worked by 23 and 24 year olds relative to 26 year olds in the year before the change in the age threshold. This could be because employers had anticipated the change, but it could also mean that the fall was caused by other factors.
These considerations do not mean we should ignore the findings – there is strong evidence that there has been a shift in patterns of full-time work for different age groups and we should try to understand this. But we should be cautious in saying that the minimum wage caused this change.
The changes we see could reflect employers preferring to give hours to more experienced, older workers once 23-24 year olds lost the advantage of a lower minimum wage rate. However, despite the fall in their average hours, 23-24 year olds were not reporting that they wanted to work more hours: the share of workers reporting any kind of underemployment – that they wanted to work more than they currently do, assuming the same pay – was lower following the change than it had been before the pandemic.
We also know that employers were reporting acute labour shortages at this time. For the employers we spoke to, worker shortages were a much more important issue than the change in age threshold. Taken together with low underemployment, this suggests that the results could be capturing changes in the choices made by 23 and 24 year olds, rather than – or as well as – the choices of employers.
If this is the case, it may be a more positive story, but there are nuances to consider. Recent work by the Resolution Foundation (Murphy, 2022) notes that young workers in low-paid jobs often talk positively about working less, but that this choice is made in the face of constraints that make working full-time in low-paid sectors unfeasible, such as requirements to work on evenings or weekends. Murphy (2022) also notes that shifting from full-time to part-time work could affect young people’s future career .
We expect to go ahead with the further change in the age threshold for the NLW – to age 21 – in April 2024. The finding that 23 and 24 year olds’ employment remained strong after they became entitled to the NLW reassures us that it is possible to do this without harming employment prospects for 21 and 22 year olds, and our stakeholders – both employers and workers – continue to tell us that it is the right thing to do.
While 21 and 22 year olds are different to 23 and 24 year olds, most employers already pay 21-22 year olds at the adult rate: 85 per cent of 21-22 year olds were paid at or above the NLW in April 2022. The LPC has recommended a large increase in the 21-22 rate this year, to smooth the transition on to the NLW. This will leave the difference between the two rates at only 24 pence per hour.
Related links: The full paper by Lee, Manly, Patrignani and Conlon; the LPC’s 2022 Report.
Source notes for charts: LPC estimates using LFS microdata, weekly (13-week rolling average), not seasonally adjusted, 21-26 year old population, UK, April 2017 – June 2022. See Appendix 4 of our 2022 Report for a full list of SOC codes included in retail and hospitality occupations. We have included both 25 and 26 year olds to increase the sample size, even though London Economics’ research looked only at 26 year olds.
]]>What is the Accommodation Offset? How does it work?
The offset comes into play when a worker lives in accommodation provided by their employer, who charges them for that accommodation.
The offset sets a limit on what an employer can charge a minimum wage worker for accommodation. In January 2023, the National Living Wage (NLW), the minimum wage for workers aged 23 and over is £9.50. A worker earning the NLW, can only be charged £8.70 per day, or £60.90 per week for accommodation.
Simple so far. But the calculation can get more complicated: for example, when workers earn more than the NLW, when employers charge more than the offset amount, and when pay is calculated on a weekly or monthly basis rather than day by day. More detail, and worked examples, are available on GOV.UK, in HMRC’s manual or on page 155 of our annual report.
Another important point is that the offset should include essential utilities – if employers charge extra for energy or water, that could take workers below the NMW.
A review of the Accommodation Offset – why?
Our recent report is the culmination of a long review of the Accommodation Offset, where we tried to answer the question: what should we do with the offset in the future?
The Accommodation Offset has been part of the National Minimum Wage (NMW) since day one – but it’s different from other rates.
Our recommendations on other NMW rates are guided by a clear brief. For the NLW, this takes the form of a target based on median pay. For the ‘youth rates’ it’s an imperative to raise rates as high as possible without damaging employment.
But the offset is distinct – neither the LPC nor the Government has set out a clear principle to apply in determining the level. And, unlike the rates, there are few evidence sources which let us track the effects of our recommendations on employers’ or workers’ decisions.
It’s helpful to think of the offset as balancing benefits between workers and employers. When it rises faster than the minimum wage, it’s helpful for businesses in offsetting their costs. When it rises more slowly, workers pocket more of their pay increase.
What did we find?
A lot of the evidence we heard concerned the settings where low-paid workers are accommodated by their employers; the nature and quality of this accommodation; and what drives employers’ decisions.
Our inquiry focused on agriculture and, to a lesser extent, hotels. Employers told us about the importance of high-quality accommodation to attract and retain workers. Workers we spoke to nevertheless had been put up in cramped or dirty conditions, about which they often felt powerless to complain.
The level of the offset does not cover employers’ costs in full, except in marginal cases. But this is a feature of the policy rather than an accident; and the need to keep a happy workforce gives businesses a strong incentive to invest in accommodation. Crucially, there’s no guarantee that pushing the offset higher leads to better quality accommodation, because there’s no clear and universal set of standards for accommodation or rigorous means of enforcement.
What were your conclusions?
Our central conclusion was about how fast the offset should increase in the future.
A useful way of thinking about the level of the offset is to benchmark it against other rates. Between 2013 and 2020, we pushed the offset up to align it with the 21-24 Year Old Rate. This meant the offset increased faster than most other rates; in other words, what employers could charge minimum wage workers for their accommodation increased faster than their pay.
In this review, we concluded that, in the future, the offset should not increase faster than other rates unless two conditions are met.
Until these conditions are met, we will not make recommendations that significantly increase the value of the offset as a proportion of the NLW.
Our final recommendation is about one group of workers in particular: seafarers. We think seafarers should be exempt from the Accommodation Offset while on board ship. This group is in a unique situation while at sea, of being confined to their work premises which also happen to be their only available accommodation. Secondly, the seafarer’s presence on the vessel is required even while they are asleep – the worker is required to be present in case of an emergency.
If you’re interested in looking at our work in greater depth, read Chapter 8 of our 2022 Report. If you would like to provide evidence on the Accommodation Offset to help inform our 2023 recommendations to government, please get in touch at lpc@lowpay.gov.uk.
]]>If you want to participate, get in touch with us if you would like to meet our Commissioners. You can email at lpc@lowpay.gov.uk or call us on 07592 272382.
Where we will be this year:
Date | Location |
22-23 March | Dunfermline / Edinburgh |
19-20 April | Birmingham |
10-11 May | Waltham Forest / London Borough of Haringey |
14-15 June | Rhondda Cynon Taf |
5-6 July | Belfast |
2-3 August | Oldham / Tameside |
We select our locations based on their minimum wage coverage. Our guiding aim is to listen to a diverse range of people from across the UK and visit locations we have not been to before or recently.
The LPC takes very seriously what we hear on all our visits, so it is a chance for anyone with an interest to have their voice heard.
Who do we want to hear from:
We are looking to meet with people affected by the minimum wage in all of these locations: workers, employers, local authorities, charities and anyone else with an interest. We are happy to come to you directly or equally to host a meeting ourselves in your area.
What do we want to hear about:
These are the main things we want to hear about:
Our annual written consultation will be launched shortly, detailing all the areas we want to gather evidence on.
Why we go on regional visits:
Since 1998 the Low Pay Commission (LPC) has made recommendations to the Government on the levels of the National Minimum Wage (NMW), including the National Minimum Wage (NMW). It has also reported and produced recommendations on related policy matters including enforcement and compliance.
To inform these recommendations, we consult widely across the economy with businesses, workers, local authorities, charities, and other interested parties.
We produce evidence-based advice, and our regional visits are an essential part of our evidence gathering cycle. Through visiting different areas of the UK we speak with workers on or close to the minimum and living wage as well as the employers who are directly affected by the change in rates.
The real-life experiences of those in low-paid work and of those employers in low-paying sectors are hard to understand from the statistics alone. By talking face to face with the people most affected by our recommendations, Commissioners can understand what is happening on the ground and the themes and issues which are emerging.
On our visits last year, we heard valuable evidence from workers on the cost of living, insecure employment and work intensification. Employers told us they had responded to the NLW increases by increasing prices, reducing profits and pay differentials and smaller firms reduced investment. These insights enable the LPC to better understand the impact of our work on both workers and employers. Last year in Leicester, we concentrated specifically on compliance issues, speaking with workers, employers and interested parties in the garment industry. The evidence we heard formed our report on Compliance and enforcement of the National Minimum Wage: the case of the Leicester textiles sector.
]]>