Women, ethnic minority and disabled workers are disproportionately represented among the beneficiaries of the minimum wage. Two-thirds of the beneficiaries of the 2004 upratings are estimated to be women workers.

We found clear evidence suggesting that the minimum wage has had a major impact in narrowing the gap between the pay of women workers and that of men at the lower end of the earnings distribution. Although still significant, the gender pay gap has also been narrowing slowly for some time in the middle range of the pay distribution independently of any influence of the minimum wage. The minimum wage has now had such a marked effect at the bottom of the distribution that only a very large uprating in relation to average earnings would have much further effect.

Upratings of the minimum wage have also helped to raise the wages of disadvantaged workers, particularly those with work-limiting disabilities, without harming their employment opportunities. In common with the rest of the labour market, the employment situation of disadvantaged groups has improved in recent years. Indeed, in many respects, those groups whose position in the labour market was weakest have demonstrated the greatest gains in employment.

The pattern of pay by ethnic group is complex. Although some minority ethnic groups suffer a pay penalty compared with the white majority, others do not. Indeed, some men from minority ethnic groups tend to earn more than their white counterparts and black women earn more on average than white women. However, some minority ethnic groups (such as those originating from Bangladesh and Pakistan) continue to earn considerably less than the average. The minimum wage has improved the position of these disadvantaged ethnic groups without adverse employment effects. Indeed employment rates of disadvantaged ethnic groups, though lower than those of white workers, are increasing.

The analysis in this Chapter also reveals the poor position of part-time workers relative to full-time workers. Part-time work is more commonly available in low-paying occupations. While female full-time workers have made significant progress in closing the earnings gap with male full-time workers, the same is not as true of female part-time workers. However, there is no gap between the earnings of female part-time workers and male part-time workers over most of the distribution. This suggests that while the labour market for full-time workers has strengthened, the pay and conditions of both male and female part-time workers has trailed behind.

Homeworkers are a largely hidden workforce. It is estimated that there may be up to one million of them, of whom 90 per cent are probably women, many of them from minority ethnic groups. These workers face particular difficulties in enforcing their rights to the minimum wage.
We believe that the new system of fair piece rates, which replaced the fair estimate agreement system, will help improve the situation of homeworkers. However, we continue to believe that this is an important area for enforcement activity. We would welcome any steps taken by the Government to heighten awareness of the minimum wage among these vulnerable workers.

We continue to believe that people with disabilities should be entitled
to the minimum wage and recognise that the Government's revised guidance since the publication of our fourth report (2003) has helped resolve many of the uncertainties about entitlement to the minimum wage. But we wish to highlight the fact that minimum wage upratings can result in some disabled people reducing the number of hours they spend in paid employment to avoid breaching the £20.00 per week earnings disregard. We would encourage the Government to take this into account as part of its efforts to increase work incentives for those with disabilities.

Some voluntary sector organisations have pointed to a lack of clarity about the position of volunteers. We recommend that the Government should review and draw together existing guidance into a single source to provide clear and accessible advice to the voluntary sector.

Back to top

Introduction

4.1 We were asked in our terms of reference to take account of the impact of the minimum wage on different groups of workers, and to be mindful when making new recommendations of the likely impact on these groups, especially on those most disadvantaged in the labour market. This Chapter focuses on outcomes in terms of pay and employment by gender, disability, and ethnicity, and on the interaction of these outcomes with the minimum wage. We also examine the position of volunteers, homeworkers and therapeutic workers. Another vulnerable group of workers - young people - is considered separately in Chapter 5.

4.2 We were also asked to look at the impact of the minimum wage on the gender pay gap and the earnings of different ethnic groups and the disabled. The quality of data available to study the gender pay gap is superior to that available for studying the impact on the disabled and ethnic groups. Nevertheless, we are able to make some general comments on the pay and employment of minority ethnic groups and the disabled.

Back to top

Employment, Unemployment and Inactivity

4.3 When the minimum wage was first introduced there was concern that it would lead to an increase in the level and rate of unemployment, particularly for those groups most likely to be low-paid. These fears have not materialised. The unemployment rate for all these groups - women, those with work-limiting disabilities and minority ethnic groups - has fallen since the introduction of the minimum wage in Spring 1999 as shown in Figure 4.1.

Figure 4.1

Working Age Unemployment Rates of Various Groups, 1998-2004

Source: Labour Force Survey (LFS), 1998-2004.

Notes:

1. The definition of ethnic groups in the LFS changed in Spring 2001 to be consistent with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.4 Because the unemployment rate has been so low, the focus more recently has turned to employment rather than unemployment. Working age employment rates of various groups are shown in Figure 4.2. It also paints an encouraging picture for those groups most affected by the minimum wage. There is no evidence to suggest a harmful effect of the minimum wage on the employment rates of these groups.

Figure 4.2

Working Age Employment Rates of Various Groups, 1998-2004

Source: LFS, 1998-2004.

Notes:

1. The definition of ethnic groups in the LFS changed in Spring 2001 to be consistent with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.5 Important features of the labour market in the last 20 to 30 years have been the increased participation of women and an increase in the inactivity rate for men. Table 4.1 shows inactivity rates and confirms the increasing labour market attachment of women and non-white minority ethnic groups since the introduction of the minimum wage. However, the falls in inactivity rates in the late 1990s and into the early 2000s have levelled off and there are now initial signs that inactivity rates may have started to rise.

Table 4.1 Working Age Inactivity Rates, 1998-2004

Source: LFS, 1998-2004.

Note: The definition of ethnic groups in the LFS changed in Spring 2001 to be consistent with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

Back to top

Labour Market Outcomes by Gender

Employment

4.6 Although women have traditionally had lower unemployment rates than men, they have also had lower labour force participation rates. Since the introduction of the minimum wage, the female unemployment rate has been consistently lower than that for males, as shown in Figure 4.1. Prior to the introduction of the minimum wage in 1999, the unemployment rate for men in 1998 was 6.8 per cent, but by the Summer of 2004 it had fallen to 5.3 per cent. By contrast, the unemployment rate for women in Summer 2004 was 4.6 per cent versus a rate of 5.4 per cent just prior to the introduction of the National Minimum Wage.

4.7 Further evidence of a strengthening position of women in the labour market can be gleaned from Figure 4.2 which shows that the employment rate of women increased by around two percentage points from Spring 1998 to just under 70 per cent in Summer 2000 and has remained at this higher level since then. This contrasts with an employment rate for working age men in the Autumn of 2004 of 79.3 per cent.

4.8 Nevertheless, a high percentage of women continue to work part-time at some stage of their working lives. In any given year almost half of women workers work part-time (43 per cent in Autumn 2004). The consequence of this for the pay and earnings of women is explored below.

Gender Pay Gap

4.9 While the minimum wage appears not to have harmed the job prospects of women, it has had a major beneficial influence on the aggregate earnings of women. As estimated in Chapter 2, two-thirds of the estimated beneficiaries of the 2004 upratings of the minimum wage are women workers. Of these beneficiaries, a large number are part-time workers. Indeed, around half of the beneficiaries of the minimum wage are women part-time workers.

4.10 While women continue to earn less than men on almost all relevant measures, the gender pay gap has been closing gradually for some time, even prior to the advent of the National Minimum Wage. In a report for the Women and Equality Unit, Anderson et al. (2001) found that the mean full-time hourly earnings of women rose from 64 per cent of men's in 1973 to 82 per cent in 2000. However, the pace of improvement over this period has been far from steady. The gap narrowed and the ratio increased quite dramatically in the early years, from 66 per cent in 1974 to 74 per cent in 1976, a move that was probably precipitated by the passage of the Equal Pay Act of 1970. It then stabilised before rising, albeit at a slower rate, from 1987 onwards.

'The minimum wage has been an effective instrument for ensuring equality at the low paid end of the labour market'

TUC evidence

4.11 The issue to be considered here, therefore, is whether the introduction and upratings of the minimum wage have produced any additional narrowing of the pay gap, above and beyond the rate implied by the long-term trend.

4.12 The size of the gender pay gap and the conclusions drawn about it depend greatly on how it is measured. Men spend more hours in (paid) employment per annum than women, so the pay gap derived from a measure of annual earnings is larger than that for monthly earnings, which in turn is larger than that for hourly pay. For instance, the April 2004 Annual Survey of Hours and Earnings (ASHE) survey showed median annual earnings of women (all employee jobs) as amounting to only 57 per cent of men's equivalent median annual earnings.

4.13 Researchers instead usually focus on the hourly wage rate but there remain complexities to be considered. The pay gap between the hourly earnings (excluding overtime pay) of all men workers and all women workers as derived from the Annual Survey of Hours and Earnings (ASHE) data for April 2004 is 25 per cent, i.e. the ratio of the median earnings of all women to the median earnings of all men was 75 per cent. However, as discussed below, the circumstances of part-time and full-time workers vary quite substantially, and it is probably not optimal for the purposes of analysis to group them all together. We also separate out workers aged 18-21 (those entitled to the youth Development Rate) from adult workers entitled to at least the full National Minimum Wage. Hence, for the rest of this section on gender differences in pay, part-time and full-time workers are considered separately, as well as those aged 18-21 and those aged 22 and above.

4.14 Differences in hourly pay can be used to illustrate the pay gap between the hourly pay of male full-time and female full-time workers, as well as the gap between female part-time workers and male part-time workers. As there are relatively few men working part-time - men make up 19 per cent of part-time workers and only 9 per cent of men in employment work part-time - most researchers in this area focus on the gap between the pay of female part-time workers and male full-time workers. The data in Tables 4.2 and 4.3 show differences along all these lines based on average hourly pay for these various categories of worker.

Table 4.2 Hourly Earnings of Men Working Full-time and Women Working Full-time and the Gender Pay Ratio, 1998-2004

Source: ASHE, April 1998-2004, without supplementary information.

4.15 Two summary statistics of the earnings distributions, the median and the mean, are presented in Table 4.2. Recent commentary on the pay gap has been based on the median, not the mean, because the median is viewed as the more representative and less volatile measure. The earnings distribution has a long tail at the upper end of the distribution, and the mean is more sensitive to these very high earnings, including annual bonuses, than the median. This makes it more difficult to discern long-term trends. Because of this resulting sensitivity, and because men are much more prevalent among very high earners than women, using the mean suggests a larger gender pay gap.

4.16 According to ASHE 2004 without supplementary information, the hourly pay gap for full-time workers in 2004 as measured by the median, was 14.3 per cent - i.e. the median gross hourly wage rate of women was 85.7 per cent that of men. Using the mean as the summary statistic rather than the median shows the pay gap to be 18.2 per cent.

4.17 From Table 4.2 it can also be seen that the pay gap for full-time workers has reduced since 1998, the year before the minimum wage was introduced, when the ratio of women's pay to men's pay stood at 78.8 per cent as measured by the mean and 82.6 per cent as measured by the median. Figure 4.3 also bears this out and the likely influence of the minimum wage can be detected in the visible reduction of the pay gap in the first decile. The comparison refers only to monetary remuneration. Non-cash forms of remuneration, such as childcare vouchers, private medical insurance and pension benefits, are excluded because of the difficulty of collecting and valuing this information. It is thought that the inclusion of these other forms of payment would have the effect of widening the pay gap, with men reaping more of these benefits than women.

Figure 4.3

Full-time Gender Pay Ratio by Percentile for Employees Aged 18 and Over, 1998-2004

Source: ASHE, April 1998-2004, without supplementary information.

4.18 In contrast to the narrowing of the gender pay gap visible in Table 4.2, Table 4.3 shows that female part-time workers have made little progress in closing the gap with male full-time workers, with the ratio improving by only 0.9 percentage points. However, women part-time workers are paid slightly more than men who work part-time. During the period over which the minimum wage has been in existence, the position of women part-time workers has slipped by 1.5 percentage points compared with women working full-time. They earn an hourly rate worth only two-thirds of that earned by women full-time workers and 56.8 per cent of the rate earned by men full-time workers. It is likely that the wage floor established by the minimum wage has prevented the ratio from deteriorating any further.

Table 4.3 Median Gross Hourly Earnings Excluding Overtime by Gender and Hours of Work, 1998-2004

Source: ASHE, April 1998-2004, without supplementary information.

4.19 In order to focus on whether the minimum wage has had any influence in closing the gender pay gap, we need to study changes in earnings at the bottom end of the distribution. Other things being equal, if more women are low-paid than men, and more women benefit from upratings of the minimum wage than men, upratings would result in a narrowing in the gap of the mean pay of women relative to the mean pay of men. As Table 4.2 shows, the pay gap as measured by the median narrowed by 3.1 percentage points since the introduction of the minimum wage, and that as measured by the mean by 3.0 percentage points.

4.20 The changes in the ratio of women's to men's earnings across the entire distribution can be seen in Figures 4.4-4.6 below. Examining the position for adult workers first (those aged 22 and above), Figure 4.4 strongly suggests that the minimum wage has had a major impact, substantially reducing the pay gap at the very bottom of the earnings distribution, and helping to close the gap further up the distribution through the impact on differentials. The pay gap has also appeared to continue to close, independently of minimum wage upratings, in the middle of the distribution. The only area where there has been no progress in reducing disparities is from the ninetieth percentile and above. At the highest levels of pay women remain considerably disadvantaged with respect to men, but this is clearly not a disadvantage on which the minimum wage can have any influence.

Figure 4.4

Women's Hourly Pay as a Percentage of Men's by Percentile for Employees Aged 22 and Over, 1998 Compared with 2004

Source: ASHE, April 1998-2004, without supplementary information.

4.21 Among young people aged 18-21 the minimum wage has had similarly powerful effects on the gender pay gap, with women in the lowest five percentiles now paid roughly the same as men, versus about 15 per cent below in 1998 (see Figure 4.5).

Figure 4.5

Women's Hourly Pay as a Percentage of Men's by Percentile for Employees Aged 18-21, 1998 Compared with 2004

Source: ASHE, April 1998-2004, without supplementary information.

4.22 Figure 4.6 shows changes in the ratio of women's earnings to men's by decile for adult workers. The pay gap is now considerably smaller for the lowest decile than the others, clearly illustrating the part played by the minimum wage in closing the gap.

Figure 4.6

Women's Hourly Pay as a Percentage of Men's by Decile for Employees Aged 22 and Over, 1998-2004

Source: ASHE, April 1998-2004, without supplementary information.

'Women experience a pay penalty for working part-time and the prevalence of women working part-time in the UK economy is a key factor affecting the gender pay gap. The clustering of women into low paid part-time work, and the associated under-utilisation of women's skills are key issues to be tackled if the gender pay gap is to be eliminated. '

Equal Opportunities Commission evidence

4.23 A key factor underlying the remaining gender pay gap is that women are more likely to be in part-time work than men and part-time workers are paid less than full-time workers. According to the LFS, in Autumn 2004 43 per cent of female employees were working part-time, as opposed to 10 per cent of male employees. As Figure 4.7 demonstrates, there is no significant gender pay gap among part-time workers except at the very top of the earnings tables. However, in the ASHE data for April 2004, the gap between the median pay of women working full-time and women working part-time was as high as 33.7 per cent. In a study for the Women and Equality Unit, Manning and Petrongolo (2004) found that women working part-time tend to be younger, to have less education, and to have more children than women working full-time. They also found that almost 25 per cent of women part-time workers worked in low-paid occupations as care assistants, cleaners, or shop assistants.

Figure 4.7

Part-time Gender Pay Ratio by Percentile for Employees Aged 18 and Over, 1998-2004

Source: ASHE, April 1998-2004, without supplementary information.

4.24 The longer-term trend indicates that the gender pay gap was closing gradually for several decades, long before the minimum wage was introduced. Overall, the data examined strongly suggest that the minimum wage has exerted a major influence in narrowing the pay gap at the lower end of the earnings distribution. To fully disentangle these effects and establish what portion of the narrowing of the gap is down to the minimum wage, and what portion is due to other influences, would be difficult and may be the subject of future research.

Back to top

Disability

Employment

4.25 Unsurprisingly, employment rates for those with a work-limiting disability are much lower than for those without. However, the labour market prospects of those with a work-limiting disability have been improving. One of the sharpest declines in unemployment rates in recent years has taken place among those with work-limiting disabilities, having fallen by two percentage points in the year to Summer 2004. Since the announcement of the introduction of the minimum wage in Spring 1998, the employment rate of those with a work-limiting disability has risen from 36.5 per cent to nearly 41 per cent in Summer 2004. It seems clear that this group has not been further disadvantaged in terms of employment and unemployment since the introduction of the minimum wage, nor as a result of any of the subsequent upratings.

4.26 Figure 4.8 shows that men with work-limiting disabilities have higher employment rates than comparable women. The increase in the employment rate is common to both men and women.

Figure 4.8

Working Age Employment Rates of Those with Work-limiting Disabilities by Gender, 1998-2004

Source: LFS, 1998-2004.

4.27 Figure 4.9 shows that men with work-limiting disabilities also have higher unemployment rates than comparable women. However, in the six years to Spring 2004, the decline in the male unemployment rate has been nearly six percentage points compared with about three for women.

Figure 4.9

Working Age Unemployment Rates of Those with Work-limiting Disabilities by Gender, 1998-2004

Source: LFS, 1998-2004.

Pay

4.28 The earnings data in this section is drawn from the LFS, while the discussion in the earnings section on gender was grounded in data from the ASHE. The ASHE is a superior source of data for the analysis of earnings. Thus, where possible we have used ASHE data, for example in arriving at our best estimate of beneficiaries of the minimum wage. However, data for the earnings of those with a work-limiting disability are only available from the LFS. Although there are problems with the earnings variables in the LFS, particularly the derived hourly earnings variable, it is plausible that LFS data do at least reflect relativities, for example between the pay of those with work-limiting disabilities and those without.

'A national survey of 100 employers and 100 disability employment advisers found general agreement across both samples that the National Minimum Wage had benefited disabled people by making low paid jobs better paid.'

Disability Rights Commission evidence

4.29 The employment prospects of those with a work-limiting disability have not been harmed by the introduction of the minimum wage and we calculate from the derived hourly pay variable that 12.5 per cent of all jobs held by those with work-limiting disabilities should benefit from the 2004 upratings of the National Minimum Wage. This contrasts with 9.6 per cent of all jobs, as calculated from the LFS. (The numbers of beneficiaries calculated from the ASHE as a percentage of all jobs is significantly lower.) The effect on the pay of female workers with work-limiting disabilities is particularly evident. As shown in Figure 4.10 (which also appears in Chapter 2), nearly 16 per cent of jobs held by women with a work-limiting disability stand to benefit from the 2004 upratings, compared with 13 per cent of all jobs held by women without a work-limiting disability.

Figure 4.10

Percentage of Employees Benefiting from the 2004 Upratings of the Minimum Wage by Disability

Source: LFS, Spring 2004. LPC estimates based on ONS low pay methodology.

4.30 As Table 4.4 demonstrates, the larger impact of the 2004 upratings of the minimum wage on the pay of those with a work-limiting disability follows from the fact that the typical hourly wages for those with a work-limiting disability are lower than those without such a disability.

Table 4.4 Median Gross Hourly Pay Excluding Overtime by Work-limiting Disability for Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004. LPC estimates based on ONS low pay methodology.

4.31 Unlike the gender pay gap, there has not been a consistent fall in the gap between the earnings of those with a work-limiting disability and the pay of other workers since the introduction of the minimum wage. Average (mean) hourly pay for those with a work-limiting disability was just under 85 per cent of those without a work-limiting disability in Spring 1998, this rose to nearly 89 per cent in 2002, before falling back to 85 per cent in Spring 2004. Similarly, using the median as the measure of the pay gap, it is almost identical in 1998 and 2004, with the gap standing at 12.8 per cent.

4.32 Of course, these trends refer to the central tendency, or midpoint, of the earnings distribution of those with work-limiting disabilities. It is possible, as in the case of gender, that the minimum wage may have had a role to play in narrowing the pay gap at the bottom end of the earnings distribution. However, in this case the data do not support such a suggestion. Figure 4.11 shows that for the bottom two deciles, there is no sign of the gap having narrowed between 1998 and 2004.

Figure 4.11

Disability Pay Ratio for Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004.

Back to top

Ethnicity

Employment

4.33 Even in the context of the overall strength of the labour market, there has been a particularly sharp increase in the employment rate of minority ethnic groups. Taken as a whole, the employment rate for minority ethnic groups has risen by over three percentage points since Spring 2000 to just less than 60 per cent in Summer 2004. There has been a corresponding decline in unemployment rates, which have fallen by two percentage points in the year to Summer 2004. In our discussion of this topic we have followed the classification of ethnicity adopted by the 2001 Census. In conducting this analysis, however, we recognise the difficulties inherent in any attempt to categorise people in terms of ethnicity, and the fact that no such system of classification can be entirely free of arbitrary delineations.

4.34 Using the self-reported classification in the LFS, there are significant differences in employment rates among the different minority ethnic groups. Figure 4.12 shows that whites have the highest male employment rates. Indians have the next highest. Blacks have similar rates to minority ethnic groups in general, while Pakistanis and Bangladeshis have the lowest. The male employment rate has increased from Spring 1998 to Summer 2004 for all the groups. The largest increase in these employment rates in this period has been among those with the lowest employment rates, namely Pakistanis and Bangladeshis.

Figure 4.12

Working Age Male Employment Rates for Ethnic Groups, 1998-2004

Source: LFS, 1998-2004.

Note: The definition of ethnic groups in the LFS changed in Spring 2001 to be in line with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.35 Figure 4.13 shows a similar pattern to men for women, although the employment rates are lower for all ethnic groups. Black women, however, have very similar employment rates to Indians. Pakistani and Bangladeshi women have very low employment rates.

Figure 4.13

Working Age Female Employment Rates for Ethnic Groups, 1998-2004

Source: LFS, 1998-2004.

Note: The definition of ethnic groups in the LFS changed in Spring 2001 to be in line with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.36 A similar picture is suggested when we look at unemployment rates for men (Figure 4.14) which have fallen for all ethnic groups. Pakistani, Bangladeshi and black men, however, have significantly higher unemployment rates than other ethnic groups.

Figure 4.14

Working Age Male Unemployment Rates for Ethnic Groups, 1998-2004

Source: LFS, 1998-2004.

Note: The definition of ethnic groups in the LFS changed in Spring 2001 to be in line with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.37 Similar unemployment rate patterns for women are shown in Figure 4.15 although it should be noted that black women have higher unemployment rates than Indian women, even though their employment rates are similar.

Figure 4.15

Working Age Female Unemployment Rates for Ethnic Groups, 1998-2004

Source: LFS, 1998-2004.

Note: The definition of ethnic groups in the LFS changed in Spring 2001 to be in line with the 2001 Census classifications; thus direct comparisons between the periods before and after should not be made.

4.38 Overall, one important message to emerge from all available data is that there is no evidence to suggest that the introduction of the minimum wage has harmed the employment prospects of these vulnerable groups in the labour market. On the contrary, the employment levels of those with the lowest rates of employment have strengthened the most. It is also possible that the introduction of the minimum wage has created incentives which led to higher levels of labour market participation among some of these groups.

Pay

4.39 Minority ethnic groups number disproportionately among the beneficiaries of the 2004 upratings of the minimum wage. While about 6 per cent of the adult working population belong to minority ethnic groups, we estimated that some 8 per cent of the beneficiaries of the 2004 upratings belonged to a minority ethnic group. The disaggregated situation with respect to the different groups is shown in Figure 4.16, which is also shown in Chapter 2. It is clear that, in terms of minority ethnic population groups, the largest percentage of workers from a particular population group to benefit from the 2004 upratings of the minimum wage were workers of Pakistani and Bangladeshi origin.

Figure 4.16

Percentage of Employees Benefiting from the 2004 Upratings of the Minimum Wage by Ethnicity

Source: LFS, Spring 2004. LPC estimates based on ONS low pay methodology.

4.40 We now turn to look at the ethnic pay gap. Our preferred source of data, (ASHE), does not have breakdowns by ethnicity. Therefore, as with the disability pay gap, we turn to the LFS as the most authoritative alternative source available.

4.41 Table 4.5 shows that overall, white workers get paid more per hour than non-white workers, but this conceals much diversity among minority ethnic groups overall and by gender. The lowest paid workers are of Pakistani and Bangladeshi descent. In contrast, workers from an Indian background tend to be paid the same if not more than white workers. Although white male workers earn more on average than black male workers, black women earn more than white women. This may be due to the fact that a larger proportion of white women work part-time and, as we have seen, part-time work is not as well paid as full-time work.

Table 4.5 Median Hourly Pay by Ethnicity for Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004. LPC estimates based on ONS low pay methodology.

4.42 Figure 4.17 suggests that the ethnic pay gap is much smaller than the corresponding pay gaps for both gender and disability. However, in calculating the ethnic pay gap we had to treat all non-whites as a single group because of small sample sizes, and this will have concealed some significant differences between groups. Even so, the pay gap shown in Figure 4.17 increases as we move up the earnings distribution to the median, before decreasing at the upper end. As the ethnic pay gap was small to begin with, and as it has not changed significantly since the introduction of the minimum wage, we can conclude on the basis of available data that the minimum wage has not had much overall impact on the ethnic pay gap.

Figure 4.17

Ethnic Pay Ratio for Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004. LPC estimates based on ONS low pay methodology.

4.43 Table 4.6 shows that, as a whole, white male workers are paid higher hourly wages than non-whites. In Spring 2004, the difference was about £1.80 per hour. Pakistani and Bangladeshi men, however, were paid on average about £3.00 per hour less than white male workers.
In contrast, it appears that men of Indian descent were paid much the same as white male workers.

Table 4.6 Median Hourly Pay by Ethnicity for Male Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004. LPC estimates based on ONS low pay methodology.

4.44 The pattern for women from minority ethnic backgrounds is very different, as illustrated in Table 4.7. Non-white female workers as a whole get paid about 70 pence an hour more than white women. Only Pakistani and Bangladeshi women are paid lower hourly rates than white women. Black, Indian and mixed and other women are paid much higher hourly rates than white women. This surprising finding may be a result of the fact that white women are far more likely to work part-time than women workers from other ethnic groups.

Table 4.7 Median Hourly Pay by Ethnicity for Female Employees Aged 18 and Over, 1998-2004

Source: LFS, 1998-2004. LPC estimates based on ONS low pay methodology.

4.45 When comparing the ethnic distribution of the earnings of full-time and part-time women workers, marked differences emerge. However, there is no ethnic pay gap among full-time women workers. As in the case of men, it is impossible to conclude from the available data that the National Minimum Wage has had any significant impact in lowering pay disparities among women from different ethnic groups.

4.46 We turn next to the impact of the minimum wage on three specific groups of workers: homeworkers, volunteers, and those undertaking therapeutic activity. We also consider some specific enforcement issues arising from the treatment of these groups under the National Minimum Wage.

Back to top

Homeworkers

4.47 Our previous reports have highlighted the problems which some homeworkers have encountered in the operation of the National Minimum Wage. While we are aware that there are homeworkers, particularly in skilled occupations, who earn well above the minimum wage, our focus continues to be on those carrying out low-paid, unskilled manual work at home. This might include assembly, finishing or packing of such goods as clothing, textiles and electrical items. It is difficult to assess accurately the size and characteristics of this largely hidden workforce, but a 2004 Oxfam Briefing Paper, building on earlier work by the National Group on Homeworking, suggested that there might be up to one million homeworkers undertaking this type of manual work, of whom 90 per cent were likely to be women and up to 50 per cent were likely to be from a minority ethnic group.

'Between 2000 and 2003, NGH was contacted by a total of 17 homeworkers engaged in [Christmas] cracker assembly and packing; their rate of pay varied from 73p
to £1.40 an hour.'

National Group on Homeworking, Oxfam and TUC, 2004.
Made at Home: British Homeworkers in Global Supply Chains

4.48 Firms employing homeworkers benefit from a flexible workforce which is able to respond very quickly to changes in demand. For the worker, undertaking these tasks at home may provide a valuable source of income which can be fitted around caring or other responsibilities. Homeworkers may, however, face particular difficulties in enforcing their rights to the minimum wage. This is often because they work in isolation, their employment status is unclear and they fear that if they complain the work will simply be transferred to someone else.

4.49 It has been common practice for homeworkers to be paid according to the amount of work they complete (often known as piecework), rather than being paid according to the number of hours spent on a task. Following the introduction of the minimum wage, employers were required either to pay homeworkers for every hour worked, or to come to a fair estimate agreement of the number of hours to be worked. The Regulations required that a fair estimate of the time the worker was likely to spend on output work could not be less than four-fifths of the time which an average worker would take to complete the task. In practice this meant that those undertaking output work might legally receive only four-fifths of the minimum wage. For previous reports we received evidence that fair estimate agreements were rarely used and, where they were, that they had often either been misunderstood or abused. We recommended in our third report that the Government consult on the case for a change to the Regulations.

'We welcome the arrangements for a new fair piece rate system for output workers ... and we will seek to monitor its application, particularly in relation to homeworkers, and report to the DTI on any problems with implementation.'

GMB evidence

4.50 In February 2003 the Government consulted on introducing a system
of fair piece rates to replace the fair estimate agreement system.
We supported the proposals in our response to the Government's consultation and welcome the new provisions which came into force on 1 October 2004 together with the accompanying guidance for workers and employers. Employers are now required to pay the minimum wage either for every hour worked or according to a fair piece rate, derived from the time that a person working at average speed would take to produce the piece in question. They are also required to carry out tests on at least a sample of workers to determine the average output rate. From April 2005, employers will have to increase the fair piece rate by a factor of 1.2 so that the majority of workers (not just those whose rate of work is above average) will receive at least the minimum wage.

4.51 In our fourth report (2003), we commented that the new rules for output work should be accompanied by effective, proactive, targeted enforcement to improve compliance, particularly in relation to homeworking. While we are aware that the Inland Revenue has established a specialist homeworkers team, which is able to investigate complaints and give advice to other enforcement teams, we continue to believe that this is an important area for enforcement activity. As noted in Chapter 6, we are also concerned that awareness of the minimum wage is lower in particular groups of vulnerable workers, including homeworkers. The Government is considering the introduction of a rolling programme of publicity and enforcement which will target a different low-paying sector each year. This is an initiative which we strongly support and we consider it important that homeworking is included among the sectors targeted early on in the programme.

Back to top

Volunteers

4.52 In preparing our earlier reports, we received evidence indicating uncertainty within the voluntary sector about the circumstances in which voluntary workers may be entitled to the minimum wage. Genuine volunteers are not entitled to the minimum wage since they are not covered by the definition of the term 'worker' under the National Minimum Wage Act 1998. Paid employees working in the voluntary sector are entitled to the minimum wage like any other worker, but the Act also exempts voluntary workers employed by charities, voluntary organisations and similar bodies, provided they receive only reasonable subsistence, accommodation or expenses, but no other monetary payments or benefits-in-kind. In addition, a genuine 'honorarium', or one-off gift with no expectation or obligation and of a reasonable amount, should not normally imply an entitlement to the minimum wage. The subject of voluntary workers was not raised in our fourth report (2003) and we received evidence on this topic from only one organisation in preparing this report. This may indicate that voluntary sector organisations are now more familiar with the operation of the National Minimum Wage, but that some problems remain.

4.53 The Central Council of Physical Recreation (CCPR) reported that there may be circumstances in which a sports club genuinely believes that an individual member is acting as a volunteer (and the member also believes this is the case), but an entitlement to the minimum wage could be inferred because a small non-cash benefit is offered, or where an 'honorarium' has become customary for holders of a particular office. The CCPR gave a number of other examples, including a scenario where a coach is paid for his or her time, but freely volunteers to give extra lessons at the same club with no expectation of payment for the additional hours. In such circumstances, the club might find it difficult to demonstrate that separate arrangements exist. It was also concerned that the requirement to keep records of hours worked could place a burden on organisations which operate on a not-for-profit basis and are run largely by volunteers. Some of the issues highlighted in the CCPR's evidence are discussed in existing guidance and we note that there are ongoing discussions between the Inland Revenue, the Department of Trade and Industry and the CCPR. We hope that these will help to resolve some of the outstanding concerns identified in the CCPR's evidence.

4.54 While those working in the voluntary sector who are entitled to the minimum wage must not be disadvantaged, we recognise that many sports clubs and other community activities depend on people who gladly give their time for scant financial reward. We would not wish the operation of the minimum wage to act as a disincentive to those who wish to participate in an activity on a genuinely voluntary basis. The Government's evidence to the Commission indicated that there might be areas where Inland Revenue compliance officers could spend less time, in order to free up time for tackling important breaches in low-paying sectors. We would support the application of this kind of proportionate enforcement approach to the voluntary sector, provided that genuine cases of non-compliance are not overlooked.

4.55 Nevertheless it remains important that voluntary sector organisations are clear about their responsibilities under the minimum wage legislation. Although some useful guidance for the voluntary sector exists, the booklet in question - National Minimum Wage, Guidance for the Voluntary Sector - was produced several years ago and may not be fully up to date. In addition, some voluntary organisations may be unaware of relevant information contained in the DTI publication, The Minimum Wage and 'Therapeutic Work.' We recommend that the Government should review and draw together existing guidance into a single source (adding any new points of clarification arising from the DTI and Inland Revenue's discussions with the CCPR and any other relevant organisations) to provide clear and accessible advice to the voluntary sector.

Back to top

Therapeutic Activity

4.56 In preparing our fourth report (2003) a considerable amount of evidence was received concerning the application of the minimum wage to people with a disability or a mental health problem. We have always maintained that people with disabilities should be covered by the minimum wage, and accordingly the legislation makes no distinction between the disabled and non-disabled, nor does it make any reference to a worker's productivity. But a range of activities are provided for people with disabilities or mental health problems and, as we noted in our previous reports, it is not always easy to distinguish between an activity focused on providing therapeutic benefits for the individual, and activities that should be regarded as work and attract the minimum wage.

'A further concern is that some day centres who previously paid expenses and gave those undertaking work-like therapy free meals have withdrawn these provisions for fear of falling foul on the minimum wage rules.'

TUC evidence

4.57 The Government produced a guidance note on the minimum wage and therapeutic activity in December 2000, following a recommendation in our second report. Uncertainty remained, however, and we subsequently discussed with a range of organisations the question of where the boundary should lie between work (attracting the minimum wage) and non-work activity. As a result we suggested in our fourth report that an activity should not be regarded as work if the following four conditions all applied:

• The activity is demonstrably focused on meeting the needs of the individual involved rather than the needs of the organisation;
• The individual is referred to the activity - and monitored and supported on an ongoing basis - by a health or social care organisation;
• The arrangement has been agreed with the individual and not made over her/his head; and
• The tempo of the activity, and of any output or delivery target, reflects the needs of the individual rather than those of the organisation.

4.58 Our fourth report shortly preceded the publication of revised Government guidance. We recommended that the Government should monitor its impact very closely and ensure that uncertainty about the minimum wage did not adversely affect people doing therapeutic activity. The Government subsequently wrote to interested organisations in August 2003 and July 2004 inviting feedback on the revised guidance. The Government's evidence stated that 'limited feedback was received but the anecdotal evidence is that the
guidance has reduced the uncertainty about therapeutic work
and the minimum wage'.

'Some organisations employing people with learning disabilities were either not aware of their duties under the NMW, confused about NMW application, simply avoiding the issue or scaling back the employment provision they were offering.'

Department of Health and Department for Work and Pensions Working Group on Learning Disabilities and Employment, report forthcoming

4.59 We welcome the Government's revised guidance and believe it has helped resolve most of the uncertainty about whether an activity would be classed as work for minimum wage purposes. Certainly few respondents raised this issue in their evidence. But there was some concern that the guidance has not been widely disseminated and a few examples of the minimum wage adversely affecting people performing therapeutic activity were also highlighted.

4.60 Disability organisations told us that some providers previously paying disabled workers £10-£15 per week had withdrawn payments to avoid being liable for the minimum wage. The Disability Rights Commission (DRC) had limited experience of the complaints procedure but felt that more efforts could be made to ensure that these were accessible to the range of disabled people. The DRC was concerned that the minimum wage guidance on therapeutic activity might not prevent exploitation, and that the emphasis should be on supporting people in open employment. The DRC saw value in an Inland Revenue investigation of pay and conditions in the day care and related sectors, and believed that doing a real job for real pay was integral to progressive employment policies.

4.61 The joint Department of Health and Department for Work and Pensions Working Group on Learning Disabilities and Employment (report forthcoming) considered the impact of the National Minimum Wage. The Group wants to maximise the opportunities for people with learning disabilities to undertake meaningful or supported employment - receiving at least the minimum wage. However, members of the Working Group were aware of a number of learning disability projects (including Local Authority operated schemes) failing to pay staff in line with the minimum wage. The Group wants key interested parties to come together and consider how to ensure all organisations that employ people with learning disabilities recognise their obligations under the minimum wage. It wishes to avoid employment projects being closed either due to ignorance of the rules or from fear of being liable for back payment of wages. The Group believes that an effective adjustment strategy to highlight and improve awareness of employers' responsibilities under the minimum wage is critical, and that employment projects should be able to seek additional advice in order to ensure their provisions comply with the minimum wage.

4.62 On a Low Pay Commission visit to South-East England we saw how one social enterprise (MCCH Society Ltd) had successfully managed to transform its services in response to the minimum wage. The organisation has divided its services to people with learning disabilities into three parts: the Training and Work Preparation Scheme which is unpaid and voluntary; Internal Supported Employment Placements where service users are paid the minimum wage; and External Supported Employment Placements which are voluntary for the first six weeks but with the intention of employment becoming permanent. MCCH Society Ltd reported that it viewed the minimum wage 'as a positive step to reducing stigma, discrimination and workplace exploitation. In our organisation the introduction of the minimum wage has acted as a catalyst to change and modernise the services with far reaching and beneficial results both to our practice and to the lives of service users.'

4.63 We continue to believe that people with disabilities should be entitled to the minimum wage and recognise that the Government's revised guidance has helped resolve many of the uncertainties about entitlement to the minimum wage. But some problems remain. We understand that a number of employment projects are failing to pay staff in line with the minimum wage. In other cases, payments have been withdrawn, with organisations perhaps unaware that reasonable expenses can be paid to those undertaking therapeutic activity without creating an entitlement to the minimum wage. Awareness of the revised guidance needs to be improved and social enterprises should also be helped to adapt their practices to comply with the minimum wage. We discuss awareness and enforcement of the minimum wage in Chapter 6.

4.64 A further concern raised by respondents relates to the interaction between the minimum wage and benefits paid to people because of illness or disability. The Permitted Work Lower Limit enables people to earn up to £20.00 per week without requiring their benefits to be reassessed - equivalent to about four hours work on the minimum wage. Some concern continues to be expressed that the earnings disregard is not increased in line with minimum wage upratings, resulting in workers with disabilities or mental health problems reducing the number of hours they work per week. We are also aware that some organisations believe that rather than simply raising the earnings disregard, other changes to the benefits system are required.

4.65 It is beyond our remit and expertise to advise on the design of the benefits system but we wish to highlight the fact that minimum wage upratings are leading some disabled people to reduce the number of hours spent in paid employment. The minimum wage could therefore be having unintended consequences for disabled people who are capable of working more than four hours per week, but who are reluctant (or whose carers are reluctant) to do so for fear of losing Income Support and associated benefits. This may also discourage these workers from seeking to gradually increase their working hours over a period of time. As part of its efforts to increase work incentives for those with disabilities, we would encourage the Government to take into account the fact that minimum wage upratings are reducing the number of hours that can be worked before reaching the £20.00 per week earnings disregard.

Back to top

Conclusion

4.66 The overall message of this Chapter is an encouraging one. The minimum wage has not led to a reduction in employment opportunities for any of the groups considered, with the possible exception of workers undertaking therapeutic work. It is also encouraging that most members of the groups examined here, including those disadvantaged in various ways, earn wages significantly above the level of the minimum wage.

4.67 On average, women still earn less than men, although the gap has been closing steadily. The effect of the minimum wage in narrowing the gender pay gap at the bottom of the earnings distribution is particularly striking in the data presented.

4.68 As between different groups of minority ethnic workers, the minimum wage has not been a clear driver of narrowing pay gaps, but that largely reflects the fact that pay gaps by ethnic groups are much less pronounced than by gender, and that the pattern is complex with some minority ethnic groups doing better than whites. Nor has the minimum wage closed pay gaps between workers with a work-limiting disability and others not similarly affected.

4.69 But across all the groups considered - women, minority ethnic groups and workers with work-limiting disabilities - the minimum wage has clearly benefited a large number of low-paid workers, and workers in these groups are disproportionately represented among the beneficiaries.

4.70 There is some evidence that compliance needs to be tightened and awareness of the minimum wage increased in the case of homeworkers. This is discussed in more detail in Chapter 6. In the next Chapter we look at two other groups of workers who may be particularly affected by the minimum wage, namely young people and trainees.