WELCOME TO THE LOW PAY COMMISSION

2.1 In this chapter we assess our most recent recommendations in light of the changing economic climate. We then look at the impact of the latest upratings on earnings. Having established that the minimum wage continues to exert a significant influence on wages at the bottom end of the earnings distribution, we turn our attention to assess whether this has, in turn, affected employment, hours, prices, profits, and productivity. We confine our main considerations in this chapter to the impact on the economy as a whole. More detailed analysis on specific workers, sectors, and on small firms follows in subsequent chapters.

2.2 Although we devote much of our attention in this chapter to the impact of the minimum wage since our last report, we are also concerned with the impact since its introduction and note such impacts where appropriate. We have commissioned a comprehensive programme of research for this report. Using the latest data available to researchers, these studies generally focus on the impact of the substantial upratings between 2003 and 2006.

2.3 Our recommendations for the October 2008 upratings were made in January 2008 in light of analyses of the impact to that date, the evidence collated and the consensus of independent economic forecasts. Weighing these together, and taking particular note of the uncertain economic outlook but the general robustness of the labour market, we recommended that the adult minimum wage should increase by 3.8 per cent (see LPC 2008 Report, paragraphs 5.34–5.42 for more detail). Similar percentage increases were also recommended for both youth rates (3.7 per cent for 18–20 year olds and 3.8 per cent for 16–17 year olds). These upratings were expected to be a little below average earnings growth but higher than the predicted increase in prices and marginally higher than the average level of pay settlements. The Government accepted these recommended rates but again rejected our arguments that the adult rate should encompass 21 year olds.

Table 2.1 Actual Outturn and Revised Forecasts Compared with 2008 Report Forecasts, UK, 2008–2009a

Source: HM Treasury (November 2007, January 2008 and March 2009) and ONS, GDP growth (ABMI); total employment as measured by Workforce Jobs (DYDC); claimant unemployment (BCJD); AEI including bonuses (LNNC), seasonally adjusted; RPI (CZBH); RPIX (CDKQ); CPI (D7G7), not seasonally adjusted, UK (GB for AEI), 2008–2009.

g. Forecast for 2009 not available in January 2008 so forecast from February 2008 used for this table.

2.4 At the time we wrote our last report, as shown in Table 2.1, the consensus forecasts showed the economy slowing down from above trend growth of 3.0 per cent in 2007 to growth below trend, at around 1.8 per cent in 2008, before picking up slightly to 2.3 per cent in 2009. These forecasts turned out to be wrong. Problems in US sub-prime credit markets in the summer of 2007 and the financial crisis that ensued has fed into increasing global economic uncertainty and a deterioration in output growth throughout the world.

2.5 Output growth in the UK was considerably below the 1.8 per cent that was anticipated by the consensus of forecasts in January 2008. Average earnings growth, at 3.2 per cent, was more subdued than anticipated and while RPI was close to forecast, the other measures of inflation (CPI and RPIX) were above forecast. The lower output growth was associated with higher unemployment.

2.6 Output in the UK grew by just 0.7 per cent in 2008. Further, output fell by 0.7 per cent in the third quarter of 2008 and provisionally by 1.5 per cent in the fourth quarter, the largest quarterly fall since 1980. As a result, forecasts for 2009 have been revised significantly downwards to -3.1 per cent. Thus, instead of output growing by 4.1 per cent over the two years (2008 and 2009) that embrace the 2008 minimum wage upratings, it is now expected to have fallen by over 2 per cent across those two years.

2.7 Real disposable household income fell over the last two years as taxes increased and average wage increases failed to keep pace with price rises. Despite this, Figure 2.1 shows that consumer expenditure was remarkably buoyant up to the first quarter of 2008, as households used savings and housing equity withdrawal (borrowing) to fund their spending. Indeed, the savings ratio became negative in the first quarter of 2008; British consumers were spending more than their incomes. Such a situation is not sustainable in the long-run and consumers will require to cut back in order to rebuild savings. The credit crunch has reduced access to funding and, as a result, consumer spending fell back in the final three quarters of 2008, but over the year it was still 1.6 per cent higher than in 2007.

2.8 Further, inflation on some measures has been higher than anticipated, with price rises concentrated on essential items such as food, energy and fuel. This has given consumers less scope to spend on durable and non-essential items. This, in turn, has affected those sectors, such as non-food retail, hospitality and hairdressing, that rely on discretionary consumer spending. Prior to the third quarter of 2008, when annualised real household spending growth slowed to 0.7 per cent, it had not been below 1.6 per cent since the introduction of the minimum wage. In the last quarter of 2008, real household spending fell by 0.2 per cent compared with the last quarter of 2007.

Source: LPC estimates based on ONS data, household final consumption expenditure (ABJR) and GDP (ABMI), quarterly, seasonally adjusted, UK, 1998–2008.

2.9 The deterioration in output has had a knock-on effect on the labour market. Employment growth had been expected to slow to 0.4 per cent in 2008, half its rate in 2007, with claimant unemployment forecast to rise to around 900,000 in 2008 and 950,000 in 2009. Employment, whether measured by the number of jobs or the number in employment, held up in the first half of 2008, growing faster than expected. A sharp decline occurred in the second half of 2008, however, with total employment falling by 0.5 per cent and the number of workforce jobs declining by more than 1.0 per cent in the six months since June.

2.10 Figure 2.2 shows that growth in the number of people in employment has fallen back sharply since February 2008. Total employment has actually declined since September 2008. Claimant unemployment and headline unemployment have risen sharply since the turn of the year. The 1.16 million registered as claimant unemployed in December 2008 is already higher than the original forecasts for the end of 2009. As we will discuss in more detail in Chapter 8, the forecasts for employment growth have been revised significantly downwards, suggesting that employment might fall by more than 0.8 million, with a corresponding rise in unemployment, by the end of 2009.

Figure 2.2 Employment Growth and Claimant Unemployment Levels, Thousands, UK, 1998–2009

Source: LPC estimates based on ONS data, total employment (MGRZ) and claimant count unemployment (BCJD), monthly, seasonally adjusted, UK, 1998–2009.

Note: Employment is measured for the three months to the end of the month shown.

2.11 When considering the 2008 upratings, inflation, as measured by the CPI, was forecast to remain around the Government’s target level of 2 per cent. But the RPI and RPIX measures of inflation had been forecast to fall back from elevated levels at the end of 2007 to a rate consistent with the CPI target. Commodity and import prices rose sharply throughout most of 2008, however, forcing all three measures to increase rapidly: CPI peaking in September 2008 at 5.2 per cent and the RPI peaking at 5.0 per cent in the same month. It was the first time since April 2002 that CPI had been greater than RPI and it has remained higher into 2009. As shown in Figure 2.3, since September 2008 both measures of inflation have fallen sharply. RPI stood at 0.1 per cent and CPI at 3.0 per cent in January 2009. These falls are expected to continue over the next twelve months with recent reductions to VAT and interest rates, along with falling energy prices, likely to lead to a period of negative inflation, as measured by RPI, during much of 2009.

Source: ONS, AEI including bonuses (LNNC), CPI (D7G7), RPI (CZBH), monthly, seasonally adjusted (AEI only), UK (GB for AEI), 1998–2009.

a. The AEI growth rates shown are three-month average percentage changes compared with the same period a year earlier.

b. The RPI and CPI growth rates are percentage changes over a year earlier. These figures are not seasonally adjusted.

2.12 Average earnings had been forecast to grow, in line with the growth observed in 2007, at around 4 per cent in both 2008 and 2009. Despite the squeeze on real disposable income as prices rose sharpest for essential items, such as food, petrol and energy, pressure on average earnings growth remained subdued. Average earnings including bonuses fell sharply to 1.8 per cent in January 2009 while average earnings growth excluding bonuses remained similar to median pay settlements, growing at 3.5 per cent a year. Indeed, in marked contrast to the previous ten years, Figure 2.3 shows that RPI had been consistently higher than average earnings growth between the end of 2006 and the end of 2008. Furthermore, from June 2008 to January 2009, average earnings including bonuses have grown more slowly than CPI, the first time that has happened since these series began in 1991.

2.13 This situation is forecast to reverse. RPI fell below average earnings growth at the end of 2008 and is expected to continue falling throughout 2009. CPI is likely to follow. We thus expect to return to a period of real wage growth (wage growth exceeding price rises) in 2009.

2.14 The recommended increases in the three National Minimum Wage rates of around 3.8 per cent in October 2008 were expected to be just under average earnings growth but far greater than price increases. As it turned out, the minimum wage looks to have increased faster than average earnings and, having been significantly below on 1 October, will be higher than price increases averaged over the year. This implies that the value of the minimum wage will increase in real terms and that its relativity, in relation to average earnings, will also increase.

2.15 As in previous years, the data available to analyse the impact of the most recent uprating (in this case October 2008) are limited. Therefore, in looking at the impact of the minimum wage, our focus will be on the 2007 upratings, although we will note any impact of previous upratings and any data available to investigate the impact of the 2008 upratings.

2.16 In this section, we look at the types of workers who are low-paid and examine the sectors in which they work. We define a minimum wage job here to be one that, in April 2008, paid less than the equivalent of the forthcoming October 2008 minimum wage downrated by the growth in average earnings between April 2008, the date of the latest available data, and October 2008. On that calculation, we define a minimum wage job as one that is held by an adult aged 22 and over paying less than £5.63; by a young person aged 18−21 paying less than £4.69; and by a 16–17 year old paying less than £3.47. In April 2008, there were about 1.13 million minimum wage jobs defined in this way, roughly 4.3 per cent of all the jobs in the UK labour market. Figure 2.4 shows that the majority (58.4 per cent) of minimum wage jobs are part-time and that just under two-thirds (64.3 per cent) are held by women. These figures are similar to those in previous years.

2.17 Using the Annual Survey of Hours and Earnings (ASHE), we also find that about 12 per cent of minimum wage jobs are temporary compared with only 6.4 per cent of all jobs in the whole economy. About 8 per cent of temporary jobs pay the minimum wage compared with around 4 per cent of permanent jobs. Around 40 per cent of workers in minimum wage jobs have been in their job for less than one year. This is nearly double the proportion of jobs of that duration in the economy (22.6 per cent). Therefore, three-fifths of workers have been doing the same minimum wage job for over twelve months.

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

Note: Minimum wage jobs defined as those held by adults (aged 22 and over) paying less than £5.63, by youths (aged 18–21) paying less than £4.69 and by 16–17 year olds paying less than £3.47 in April 2008.

2.18 We can see from Figure 2.5 that the distribution of coverage by age is generally U-shaped. That is, there is a higher proportion of young workers and older workers on the minimum wage than workers of middle age. About 6 to 7 per cent of jobs held by workers aged 16 to 21 are minimum wage jobs compared with around 3 to 3.5 per cent of jobs held by those aged 35 to retirement age. This percentage then rises again to about 9 per cent for those over the state pension age. If we considered only the adult minimum wage, the proportions of young people covered would be far greater.

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

2.19 A related grouping are the least-skilled, defined for our purposes as not having any formal qualifications. Many workers in this group will also be young. Using the Labour Force Survey (LFS),[1] we estimate that in the second quarter of 2008 the least-skilled workers were more than three times as likely to hold minimum wage jobs (17.5 per cent) as their more qualified counterparts (about 5 per cent).

2.20 Coverage among employees from ethnic minorities is greater than among white employees. According to the LFS, in the second quarter of 2008 just under 8 per cent of jobs held by employees from ethnic minorities were minimum wage jobs compared with fewer than 6 per cent of jobs held by white employees. Using LFS, we find similar differences for those with and without disabilities. Over 8 per cent of jobs held by employees with disabilities paid less than the downrated forthcoming minimum wage compared with 5.6 per cent of jobs held by those without any disability.

2.21 Figure 2.6 shows the proportion of minimum wage jobs in April 2008 by region, country and gender. The preponderance of women is again apparent. Unsurprisingly, the proportion of minimum wage jobs in the South East of England and London is lower than the national average. The areas with the highest proportions of minimum wage jobs are in Wales, the North East of England, and Northern Ireland. Similar patterns are found using residence-based regions. The geographical distribution of minimum wage jobs has changed little since 1999.

Figure 2.6 Minimum Wage Jobs,a by Gender for Work-based Country and Region,b UK, 2008

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

Notes:
a. Minimum wage jobs defined as those held by adults (aged 22 and over) paying less than £5.63, by youths (aged 18–21) paying less than £4.69 and by 16–17 year olds paying less than £3.47 in April 2008.

2.22 More than half, about 55 per cent, of minimum wage jobs are in large firms (those with 250 or more employees) although large firms employ two-thirds of all workers. Micro firms (those with 1 to 9 employees) employ fewer than 8 per cent of the total workforce but provide more than 15 per cent of all minimum wage jobs. Fewer than a fifth of all jobs are in small firms (those with fewer than 50 employees), but nearly a third (32 per cent) of all minimum wage jobs are in small firms.

2.23 Figure 2.7 shows that the proportion of minimum wage jobs declines with size of firm. Around 9 per cent of jobs in micro firms and 6 per cent in other small firms (those with 10 to 49 employees) were estimated to be minimum wage jobs. Fewer than 4 per cent of employees in large firms were in minimum wage jobs. The gender pattern is similar across all firm sizes, with women more likely to be in minimum wage jobs than men.

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

2.24 We look next at minimum wage jobs by industry and then by occupation. In our 2007 Report we reviewed and revised our definitions of low-paying industries and occupations.[2] We examine these in more detail in Chapter 3. Figure 2.8 shows that around two-thirds of minimum wage jobs are in the low-paying industries, as the Low Pay Commission defines them. About a third are spread throughout the rest of the economy.

Figure 2.8 Number and Percentage of Minimum Wage Jobs, by Low-paying Industry,ab UK, 2008

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

a. Minimum wage jobs defined as those held by adults (aged 22 and over) paying less than £5.63, by youths (aged 18–21) paying less than £4.69 and by 16–17 year olds paying less than £3.47 in April 2008.

2.25 The largest two industries, retail and hospitality, account for about 44 per cent of all minimum wage jobs. The next largest industries, social care and cleaning, each account for only 6 to 7 per cent. The remaining low-paying industries are quite small, between them providing fewer than 9 per cent of all minimum wage jobs.

2.26 The industries with the largest proportions of jobs covered by the minimum wage were cleaning, hospitality, and hairdressing. In all three sectors, over 20 per cent of employees were in minimum wage jobs in April 2008. Hospitality had about 234,000 workers who were in minimum wage jobs. Compared with hospitality, the absolute numbers in hairdressing and cleaning were very small. But the retail sector has more minimum wage jobs than any other sector: just under 260,000 retail jobs (representing about 8 per cent of jobs in the sector) were estimated to be minimum wage jobs.

2.27 The picture is similar when looking at low-paying occupations. That is removing managerial and supervisory posts and focusing on the lowest paid jobs in the sector (for example, in retail our definition includes checkout assistants, shelf fillers and trolley collectors). Hairdressing (barbers and stylists), hospitality (including bar and restaurant staff, hotel porters and catering assistants) and cleaning (cleaners) are the occupations that are most likely to be minimum wage jobs.

2.28 Figure 2.9 shows that minimum wage jobs again account for over 20 per cent of all the low-paying jobs in the hairdressing, hospitality and cleaning occupations. Using this occupational definition, however, retail still has the largest number of minimum wage jobs (over 235,000) with hospitality the second largest (about 217,000).

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

2.29 Minimum wage jobs are most prevalent in the private sector. There are few minimum wage jobs in the public sector. According to ASHE around two-thirds of all jobs are in the private sector but over 86 per cent of minimum wage jobs are in the private sector. Whereas a quarter of all jobs are in the public sector, just 8.5 per cent of minimum wage jobs are found in this sector. The voluntary sector accounts for less than 8 per cent of all jobs but only 5 per cent of minimum wage jobs.

2.30 In conclusion, minimum wage jobs are more likely to be held by women, young workers, those of retirement age, ethnic minorities, those with a disability, and those with no qualifications. They are also more likely than better paid jobs to be part-time and temporary. A higher incidence of minimum wage jobs are found in small firms, in the private sector, in particular areas of the UK, and in certain industries and occupations.

2.31 We have carefully monitored the impact of the National Minimum Wage since its introduction in 1999 to determine its effects on the economy in general and on the labour market in particular. A minimum wage increase raises the wage costs faced by employers, and this may have consequences across a range of economic outcomes. We start our assessment by looking at the impact on employee earnings. After establishing an impact here, we go on to look at the consequences for other economic variables. We first examine the aggregate economy before proceeding to look at the impact of the minimum wage on the low-paying sectors and low-wage groups of workers in greater detail in the chapters that follow.

2.32 As previously explained, our focus in this section is on the upratings in 2007 and, where data are available, 2008. The main earnings data set analysed in this section is the ASHE, which is conducted in April of each year. The latest data are from April 2008, when the adult minimum wage was £5.52 per hour, an increase of 3.2 per cent on the previous year. As employers often anticipate minimum wage upratings, though, the data may already reflect the October 2008 upratings to some extent. We also comment on any noticeable trends (or deviations from trends) over the lifetime of the minimum wage, noting from our initial analysis above that we might expect the impact to have been maintained since October 2006, when its value relative to earnings and prices rose to a peak. This value has remained high and, in October 2008, rose above the previous (October 2006) peak. Our analysis in this chapter will concentrate on those aged 22 and over. Chapter 5 will look in depth at the impact of the minimum wage on younger workers.

2.33 Much of the data that have become available since our last report (LPC, 2008) relate to the minimum wage upratings in October 2007, which, as previously shown in Table 1.1, all increased by around 3.2 per cent to £5.52 for adults, £4.60 for 18–21 year olds and £3.40 for 16–17 year olds. This increase compared with average earnings growth of 3.9 per cent, and price inflation ranging from 2.1 per cent (CPI) to 4.2 per cent (RPI). Towards the end of 2007, output and employment were growing strongly with unemployment falling. The impact of the global financial crisis, initiating in the US in the spring of 2007, did not affect the real economy in the UK until 2008.

2.34 In April 2008, as shown in Figure 2.10, about 0.9 per cent of employees aged 22 and over (around 224,000) were paid below the adult minimum wage.[3] This is slightly fewer than in 2007 but, as shown in Table 2.2, in line with numbers observed since 2000. There was a slight increase in the number (592,000) and proportion (2.5 per cent) of employees who were paid at the minimum wage rate in April 2008,[4] up from 577,000 (or 2.4 per cent) in April 2007.

Figure 2.10 Hourly Earnings Distribution for Employees Aged 22 and Over, UK, 2008

Source: LPC estimates based on ASHE 2007 methodology, low-pay weights, UK, April 2008.

Note: * Five pence bands except where stated otherwise (bands labelled by minimum pay amount).

2.35 In April 2008, there were around 1.26 million jobs that paid less than £5.73 per hour, the then forthcoming October 2008 uprating (now implemented) that had been announced in March when our 2008 Report (LPC, 2008) had been published. Not surprisingly, as the increase in October 2008 (3.8 per cent) was greater than the October 2007 increase (3.2 per cent), this represented an increase of nearly 50,000 in the numbers paid below the forthcoming uprating. We might, therefore, expect that the impact of the October 2008 uprating would have been greater than that of the October 2007 uprating. Table 2.2 shows that the number and percentage of those paid below the forthcoming rate are higher than 2007 but they are lower than those covered by the 2001 or 2006 upratings.

2.36 Looking at ten pence pay bands, for consistency with data published by ONS prior to 2004, Table 2.2 shows that ONS recorded the largest number of employees paid at the minimum wage in April 2008, and it was the highest proportion of employees affected since it was introduced in 1999. This supports the findings by Incomes Data Services (IDS, 2006b and 2007) and Swaffield (2008) that the minimum wage is having a greater impact on more and more firms and workers.

Table 2.2 Jobs Held by Adults (Aged 22 and Over) Paying Below the Existing National Minimum Wage and the Forthcoming National Minimum Wage, UK, 1999–2008

Source: ONS central estimates using ASHE without supplementary information and LFS, UK, 1999–2004; LPC estimates using ASHE with supplementary information, low-pay weights, UK, 2004–2006; and ASHE 2007 methodology, low-pay weights, UK, April 2006–2008.

Note: Prior to 2004, all our analyses were conducted in ten pence pay bands using the ONS central estimate methodology. In contrast to elsewhere in this report, where five pence pay bands are used, we use ten pence pay bands in this table.

2.37 Not all those paid below £5.73 in April 2008 will have directly benefited from the minimum wage uprating. As explained in our previous reports (LPC, 2007 and 2008), we would expect some of these workers to have received pay rises that would have taken their pay above £5.73 before October. Estimating the coverage of the minimum wage requires us to make assumptions about how wages would adjust in the absence of a minimum wage. One assumption is that wages would have increased in line with average earnings. An alternative, though less plausible assumption in the light of recent evidence, is that they would have increased in line with prices. Assuming the former, we estimate that about 4.1 per cent of jobs (0.99 million) held by adults will have been covered by the 2008 uprating.

2.38 Assuming that the wages of the lowest paid would have instead risen in line with prices, we estimate coverage between 0.86 and 0.99 million jobs (3.6–4.1 per cent), depending on whether CPI or RPI is used. These estimates for the 2008 uprating can be compared with our estimates that, using the wages or RPI price assumption, around 3.7 per cent of jobs held by adults (about 0.88 million jobs) were covered by the 2007 minimum wage uprating. Not surprisingly, we find that these estimates of coverage are greater for the 2008 uprating than the 2007 uprating. In contrast, using the CPI assumption, coverage falls from 4.2 per cent in 2007 to 3.6 per cent in 2008. But these estimates are much lower than our 2007 Report estimates of coverage for the October 2006 uprating, around 5 per cent of employees aged 22 and over (about 1.1 million such employees).

2.39 We define the bite of the minimum wage as its value relative to a specific point on the earnings distribution such as the mean, median or lowest decile. We would expect the bite on any of these measures to have fallen in April 2008 compared with April 2007 because the increase in the minimum wage relative to average earnings growth was greater in October 2006 than in October 2007. Table 2.3 confirms that is the case when the minimum wage is judged against the median or the mean of the hourly earnings distribution for employees aged 22 and over. The bite of the minimum wage does appear to have peaked at 51.0 per cent of the median and 39.6 per cent of the mean in April 2007, after the 5.9 per cent minimum wage increase in October 2006. The bite in April 2008 fell to 50.7 per cent of the median and 39.4 per cent of the mean.

Table 2.3 National Minimum Wage as a Percentage of Various Points on the Earnings Distribution, Employees Aged 22 and Over, UK, 1999–2008ab

2.40 Surprisingly, we find that the uprating in October 2007 that was expected to be modest (an increase of 3.2 per cent compared with average earnings growth of 3.9 per cent) has not affected the bite against the lowest quartile and, in fact, the bite has actually increased relative to the lowest decile. This provides further support for the research findings that the minimum wage is having a greater impact on wages at the bottom end of the distribution. We might expect this effect to have increased as a result of the 2008 upratings (up 3.8 per cent) given that average earnings growth has been much slower (up 3.2 per cent in the year to the fourth quarter of 2008).

2.41 An alternative way of looking at this is to examine how the value of the minimum wage has changed over time relative to both prices and average earnings. Using £3.60 in April 1999 as the base, we calculate what the minimum wage would have been had it been uprated in line with price inflation or average wage growth. We then take the value of the minimum wage and compare it with what it would have been following either of these formulaic approaches. Figure 2.11 shows this difference between the actual minimum wage and what it would have been had it been uprated by the growth in CPI, RPI or average earnings. Figure 2.11 therefore shows that the National Minimum Wage has advanced more rapidly than prices and average earnings over the period as a whole. In October 2008, the adult rate at £5.73 was 49 pence higher than it would have been, had it been indexed to earnings growth (£5.24), 99 pence greater than if it had been indexed to RPI (£4.74) and £1.44 greater than if indexed to CPI (£4.29). On these measures the value of the adult minimum wage was greater in October 2008 than it had previously been. Since that time the average earnings index and the price indices have all fallen resulting in a further increase in the value of the minimum wage relative to both wages and prices.

Figure 2.11 Increases in the Real and Relative Value of the Adult National Minimum Wage, 1999−2009

Source: LPC estimates based on ONS data, AEI including bonuses (LNMQ), RPIX (CHMK), RPI (CHAW) and CPI (D7BT), monthly, seasonally adjusted (not seasonally adjusted for RPI and CPI), UK (GB for AEI), 1999–2009.

2.42 Our analysis above confirms that the October 2007 upratings maintained the bite of the minimum wage and had a significant impact on the earnings distribution. We now look in more detail at the impact on wage differentials.

2.43 Dividing employees into 100 equally sized groups (percentiles) and ranking these groups by their hourly earnings from the lowest paid on the left to the highest paid on the right, Figure 2.12 shows how the earnings at each of these percentiles have changed on average each year compared with those at the median (earners in the middle of the distribution at the 50th percentile) for three periods: 1992–1997 (prior to the introduction of the minimum wage), 1998–2004 (covering the introduction of the minimum wage) and 2004–2008 (covering the more recent minimum wage upratings). The earnings at the median are normalised to zero. For example, over the period prior to the introduction of the minimum wage, 1992–1997 (depicted by the red line), the lowest decile (those employees at the tenth percentile) had wage increases that were, on average, 0.6 per cent lower per annum than the wage increases for those at the median. In the period from 1998 to 2004 (depicted by the dark blue line), however, the earnings of the lowest decile increased on average by 0.8 per cent more each year than for those in the middle of the distribution, while between 2004 and 2008 (represented by the light blue line), the annual increase was just 0.04 per cent higher on average than the increase at the median.

Figure 2.12 Annual Increase in Hourly Earnings Minus the Increase in Median Earnings, by Percentile for Employees Aged 22 and Over, UK, 1992–2008ab

Source: LPC estimates based on New Earnings Survey (NES), unweighted, UK, April 1992–1997; ASHE without supplementary information, standard weights, UK, April 1998–2004; ASHE with supplementary information, standard weights, UK, April 2004; and ASHE 2007 methodology, standard weights, UK, April 2008.

a. Comparisons have been made here for illustrative purposes only as no consistent earnings time series data are available from 1992 to 2008. This analysis uses ASHE with supplementary information for 2004 and ASHE new 2007 methodology for 2008. These two series are not strictly comparable although the data for 2006 from the separate series are similar.

b. Those jobs where pay was affected by absence in the reference period were removed before the percentiles were calculated.

2.44 Over the period prior to the introduction of the minimum wage, 1992–1997, the wages of the lowest paid increased by less than those at the median, whose wages in turn increased by less than the wages of those at the top of the earnings distribution. Following the introduction of the minimum wage, over the period 1998–2004, those at the bottom of the earnings distribution received higher pay rises than those at the middle of the distribution. Since 2004, the increases in the minimum wage appear to have had a smaller effect than the increases in the earlier period (1998–2004). But those at the bottom of the earnings distribution still received higher pay rises than those at the middle of the distribution.

2.45 Moreover, the minimum wage increases appear to have had a knock-on effect up the earnings distribution to about the 30th percentile. The size of the impact declines as we move away from minimum wage earners (in the bottom 5th percentile), however, falling to zero by the 30th percentile. In other words, those earning just above the minimum wage have seen their earnings rise faster than those at the median but not as fast as the earnings of those on the minimum wage. The more recent minimum wage increases (2004–2008) appear to have had a much smaller impact. This provides some evidence of spill-over effects of the minimum wage on the earnings distribution. It also suggests that differentials just above the minimum wage may have been squeezed, but beyond the lowest decile (the tenth percentile), there has been little further squeezing between 2004 and 2008. Despite this, the cumulative effect of the large pay rises for the low paid that squeezed differentials may remain.

2.46 As in previous years, we commissioned work to assess the impact of the latest upratings and employers’ ongoing responses to the minimum wage. Previous research, for example IDS (2007), had found that the minimum wage, particularly the 2003 to 2006 upratings, had had an impact on relative pay levels in many low-paying sectors, either through narrowed differentials, or by precipitating changes to pay structures in response to the narrowed differentials. In contrast to these findings, IDS (2009) found that the more modest 2007 National Minimum Wage uprating led to a partial restoration of differentials in several sectors, most notably retail.

2.47 We next look at average earnings growth and pay settlements in the economy. Generally, pay settlements deal with consolidated increases in basic pay while average earnings growth more fully captures increases in all aspects of pay, including promotion and bonuses. Wage inflation has failed to keep pace with the recent rise in price inflation. Indeed, as shown in Figure 2.13, the official ONS earnings measures – the Average Earnings Index (AEI) including or excluding bonuses – suggest that average earnings growth has been stable at around 3.6 per cent (the excluding bonus series) or has actually been slowing down from 4.1 per cent in September 2007 to 3.2 per cent in December 2008 and to 1.8 per cent in January 2009 (the including bonus series). In contrast, as also shown in Figure 2.13, pay settlement data from independent private sector sources (IDS, Industrial Relations Services (IRS), the Labour Research Department (LRD) and EEF, the manufacturers’ organisation) generally tracked the increase in RPI through the second and third quarters of 2008, rising from 3–3.5 per cent to 3.5–4 per cent, still below the RPI which peaked at 5.0 per cent in September 2008. RPI has since fallen sharply, and there are signs that both pay settlements and wage growth have also started falling.

Figure 2.13 Comparison of Growth in Average Earningsa (GB) with Median Pay Settlementsbc and Price Inflationd (UK), 1998–2009

Source: ONS, AEI including bonuses (LNNC), RPI (CZBH), and IRS, IDS, LRD, and EEF pay databank records, monthly, seasonally adjusted, UK (GB for AEI), 1998–2009.

a. The AEI growth rates shown are three-month average percentage changes on a year ago.

d. The RPI growth rates are percentage changes over a year earlier. These figures are not seasonally adjusted.

2.48 In this chapter so far, we have focused on the impact of the National Minimum Wage on an individual’s gross earnings. In this section, we give a brief account of how the minimum wage interacts with the tax and benefit system and show how the distribution of household income might be affected by the minimum wage.

2.49 An individual’s gross pay is subject to deductions for tax and National Insurance Contributions (NICs). Individuals may, however, be entitled to in-work benefits, such as Working Tax Credit and Child Tax Credit, and a range of other state benefits, such as Housing Benefit, Child Benefit, and Council Tax Benefit. Most of these benefits are means tested based on household income. Therefore, an individual’s income will depend on their own earnings, the earnings of others in the household and on household circumstances, such as the number and age of children, childcare costs, and whether any household member is entitled to disability payments. It is thus not easy to generalise about the impact of the minimum wage on the net income of individuals and households.

2.50 The Chancellor announced in the 2007 Budget that the basic rate of tax would be reduced from 22 pence to 20 pence in the financial year 2008/09. In addition, the 10 pence tax band would be abolished; tax credits and child benefit made more generous; and personal allowances for pensioners raised. These changes were confirmed in the 2008 Budget. Following much public debate, the Chancellor announced a further increase in personal tax allowances on 13 May 2008. Without changes to these personal tax thresholds, the income tax changes would have led to losses for all those earning between £5,225 and £18,605 a year. That is, many minimum wage workers (roughly those working more than 18 hours a week) would have lost out. The biggest loss would have been £223 a year for those earning exactly £7,455 a year.

2.51 But changes were made to personal tax allowances. The initial increase from £5,225 to £5,435 was equivalent to giving back £42 a year to basic rate taxpayers. The additional £600 increase announced on 13 May gave back a further £120 a year to basic rate taxpayers. This reduced the range of those worse off to people earning between £6,845 and £10,505 a year. The largest loss would then be £61 a year for those earning £7,455 a year.

2.52 Changes were also made to National Insurance thresholds. Taking account of these, only those earning between £7,082 and £9,317 a year were worse off. Overall, therefore, the losses peak at £37 a year for those earning £7,455 a year. We estimate that up to 12 per cent of minimum wage employees (275,000) would have been adversely affected by these tax changes. The impact might be mitigated, however, depending on household circumstances because Working Tax Credits and child benefits were made more generous. We estimate, however, that around 105,000 of these losers would have been under 25 years old and not eligible for tax credits.

2.53 The increase in the minimum wage was effectively greater than the loss in earnings as a result of the tax changes, but many minimum wage workers would not have received any gain until October 2008, midway through the tax year. The Government has since raised the personal allowances for the 2009/2010 tax year so that there should no longer be any losers.

2.54 When the adult minimum wage was £5.52 in September 2008, gross weekly income would have been £193.20 for a 35 hour week. Using HM Treasury estimates, this gross income would have been equivalent to a net income of around £189.66 for a single person working full-time with no children (a net wage of £5.42 an hour for a 35 hour week). The corresponding amount for a couple with one child (one partner working and the other not) was £292.95 (equivalent to a wage of £8.37 an hour for a 35 hour week). Again assuming a 35 hour week, gross weekly income would have increased to £200.55 following the minimum wage increase to £5.73. The net weekly income for a single person would have risen by around £2.22 in October 2008 and then by a further £4.69 to reach £196.57 in April 2009 taking into account the new tax regime. This increase is just below the £7.35 increase in gross weekly income. For the one child family, net income would rise by around £24.22 to £317.17 a week (equivalent to an hourly wage of £9.06). The effective hourly rate for the single person would be £5.62. In conclusion, the minimum wage increased by 3.8 per cent but changes in taxation led, on average, to an increase of 3.6 per cent in net take-home pay for a single person and an increase of 8.3 per cent for a couple with one child.

2.55 In line with research findings from Bryan and Taylor (2004) and the Institute for Fiscal Studies (2003), analysis by HM Treasury, as shown in Figure 2.14, shows that the minimum wage is not particularly well-targeted at the poorest households in the UK. But the poorest households (which include pensioners and benefit recipients) do not generally have anyone in work. The minimum wage is only able to help those in work. Once the analysis is restricted to those households where at least one member is in work, it appears that the minimum wage is targeted most at the lowest household earnings deciles. The analysis used in Figure 2.14 was generated to demonstrate the effects of a 10 pence rise in the minimum wage, however, similar results are gained for the actual minimum wage increase of 21 pence.

Figure 2.14 Distributional Impact of a 10 Pence Increase in the Minimum Wage, UK, 2008

Source: HM Treasury estimates based on Family Resources Survey (FRS), UK, 2006/07 uprated to 2008/09.

Note: These figures take account of changes in tax credits, benefits, taxes and National Insurance but do not take any account of likely behavioural change caused by a rise in hourly pay, such as changed levels of employment or hours worked. They also do not include the effect of the £25,000 disregard in tax credits, which allows income to rise between one year and the next by up to £25,000 before tax credits begin to be withdrawn. This means that the reductions in tax credits would in practice be significantly smaller, at least in the initial tax year.

2.56 Despite pay settlements and average earnings growth being relatively subdued throughout 2007 and 2008, Figure 2.15 shows that unit wage costs for the whole economy have increased since the middle of 2007. October 2007 saw an increase in both annual leave entitlement and the minimum wage. The increase in the minimum wage in October 2007, less than average earnings growth and lower than the level of most pay settlements does not, however, appear to have been a significant factor in driving up wage costs. From the end of 2007, non-wage labour costs have risen at a slightly slower pace suggesting that firms have not been overly affected by the change in statutory leave entitlement from 4 weeks (20 days) to 4.8 weeks (24 days) in October 2007. Research from IDS (2009) suggested that the annual leave changes had generally had minimal impact, although some sectors (such as fast food, pubs and restaurants; and the care sectors) were more affected than others. Evidence from the leisure and hospitality sectors supports the finding that firms had been affected but the additional cost had been less than 1 per cent of the wage bill. The periods in which unit labour costs were greater than unit wage costs, 2002–2004 and 2005, were times when either National Insurance Contributions had increased or when employers were putting extra resources in to their pension schemes.

Source: ONS, unit wage costs (LOJE) and unit labour costs (DMWN), quarterly, seasonally adjusted,
UK, 2001–2008.

2.57 Research from Forth, Harris, Rincon-Aznar and Robinson (2009) confirmed that the initial introduction of the minimum wage had the largest impact on the whole economy wage bill, an increase of 0.22 per cent, as the pay in a number of very low-paid jobs was raised to at least match the new minimum. Figure 2.16 shows that the small upratings in 2000 and 2002 generally had a muted effect on the overall wage bill, while the larger upratings increased the wage bill by about 0.1 per cent a year, approximately half the impact of the initial introduction.

Figure 2.16 Impact of the Minimum Wage on the Wage Bill for the Whole Economy, UK, 1999–2007

2.58 We can assess the cumulative impact of the minimum wage by calculating the value of the different adult minimum wage rates in April 1999 prices and looking at each of these rates as if it were the first ever minimum wage rate. Figure 2.16 depicts these cumulative impacts. For example, Forth, Harris, Rincon-Aznar and Robinson (2009) calculated that the 2007 minimum wage of £5.52 per hour would have been equivalent to £4.36 in April 1999 prices. On this basis, they estimated that coverage had increased from around 4.5 per cent in April 1999 to over 7 per cent in 2007 and that there was a notable change in the proportion affected between 2002 and 2004. As a result, the cumulative impact on the wage bill nearly doubles between 1999 and 2007, from just over 0.2 per cent in 1999 to just under 0.4 per cent in 2007. This provides further evidence that the minimum wage is having a greater impact on firms over time.

2.59 The above discussion and analysis have demonstrated that the minimum wage has had a significant impact on the bottom of the earnings distribution. We can now investigate how firms have coped with the resulting impact on labour costs. In the face of an increase to the National Minimum Wage, employers have a number of options to limit the impact on their wage bill. We have seen that some firms have attempted to absorb these costs by cutting other aspects of the remuneration package such as pension provision, unsocial hours payments, overtime and shift premia, bonuses and non-wage considerations such as perks and staff discounts. Employers can also, for example, adjust the numbers employed, the number of hours worked, or seek to increase the productivity of the workforce through various means. If they fail to limit the impact, they might squeeze profit margins or increase prices. The magnitude of these adjustments will determine the extent of any adverse impact from the increase to the minimum wage. It is in these areas, and on minimum wage earners in particular, that we focus our attention in analysing the impact of the minimum wage.

2.60 Over the period of the minimum wage, the labour market had fared well, with employment growing from 27.04 million in April 1999 to 29.50 million in April 2008, an annual rate of growth of about 1 per cent. But recently things have changed. The labour market has been affected by the current credit crunch and the subsequent downturn in the economy. The total number of people in employment reached 29.54 million in May 2008, the highest on record, but has since fallen by 0.5 per cent to 29.38 million in January 2009.

2.61 There were around 29.30 million people in employment when the minimum wage increased by 3.2 per cent in October 2007. By the time that the minimum wage was next raised (by 3.8 per cent) in October 2008, employment, affected by the economic situation, had grown by only 135,000 (0.5 per cent). Table 2.4 shows that this is the slowest growth in employment experienced between minimum wage upratings. In stark contrast to the 239,000 increase in the number of employee jobs experienced between September 2006 and September 2007, there was a fall of 22,000 between September 2007 and September 2008. Growth in the total number of employees (0.6 per cent), however, was marginally higher than in the two previous years. The labour market looks to have deteriorated further since the October 2008 upratings were implemented.

2.62 Since the October 2007 upratings, there has also been a sharp increase in unemployment. On both measures, the claimant count and headline ILO unemployment, the increase, as shown in Table 2.4, had been faster than at almost any time since the introduction of the minimum wage in 1999. The exception being the sharpe rise in ILO unemployment between September 2005 and September 2006. There was an increase of over 10 per cent in both the claimant count and the ILO measure between September 2007 and September 2008. This increase in unemployment has continued at a faster pace since October 2008; the ILO measure rising above 2 million to hit 2.03 million in January 2009 and the claimant count reaching 1.39 million in February 2009.

Table 2.4 Change in Employment, Jobs and Unemployment in Each National Minimum Wage Period, UK, 1999–2008

Source: ONS, total employment (MGRZ), LFS employees (MGRN), employee jobs (BCAJ), working age ILO unemployment (YBSH), and claimant count (BCJD), UK, seasonally adjusted, 1999–2008.

a. To be comparable with other periods, growth has been adjusted to represent twelve months.

2.63 But Figure 2.17 suggests that the low-paying sectors have been less affected by the recession than the economy as a whole. Job growth in the whole economy outpaced that in the low-paying sectors in 2007 and the beginning of 2008; but between March and September the number of employee jobs in the low-paying sectors grew faster than in the economy as a whole. The number of employee jobs fell in both the low-paying sectors and the whole economy in the year to December 2008, although the job loss in the low-paying sectors was proportionately less.

Figure 2.17 Annual Change in Employee Jobs in the Whole Economy and the Low-paying Sectors, GB, 2006–2008

Source: LPC estimates based on ONS employee jobs series, three-monthly, not seasonally adjusted, GB, 2006–2008.

2.64 We showed earlier that the minimum wage affected the private sector more than the public sector, although there does not appear to be any relationship between the size of the changes in the minimum wage and private sector employment in aggregate. The strong growth in employment since the beginning of 2006 has been an entirely private sector phenomenon. Indeed, compared with the year earlier, public sector employment fell in every quarter between the second quarters of 2006 and 2008. Private sector employment appeared at first resilient to the initial impact of the credit crunch that originated in the US sub-prime markets in the summer of 2007; however, private sector employment growth slowed rapidly in the third quarter of 2008 before falling in the fourth quarter. At the same time, public sector employment actually increased, having fallen for nine consecutive quarters.

Figure 2.18 Change in Levels of Private and Public Sector Employment, Thousands, UK, 2000–2008

Source: LPC estimates based on ONS data, public sector employees (G7AU) and private sector employees (G7K5), quarterly, seasonally adjusted, UK, 2000–2008.

Note: The chart shows the difference in the number of jobs in the quarter shown compared with the same period of the previous year.

2.65 As would be expected from economic theory, the recession appears to have affected hours before employment. Firms might be expected to adjust hours before they start to make people redundant. The total number of hours worked per week peaked at 948 million in March 2008, an increase of 17 million hours on March 2007. Since that time, as shown in Figure 2.19, the annual growth in hours worked has slowed. The figure also shows that hours have been more sensitive to the recession than employment. The October upratings of the minimum wage coincided with a fall in the number of hours, compared with a year ago. This fall has continued. In December 2008, only 934 million hours were worked, the lowest since April 2007.

Figure 2.19 Change in Actual Hours Worked, Millions, and Total Employment, Thousands, UK, 1993–2009

Source: LPC estimates based on ONS data, total hours worked (YBUS) and total employment (MGRZ), monthly, seasonally adjusted, UK, 1993–2008.

2.66 When the minimum wage was introduced, we were concerned that self-employment might be used as a ruse to evade the regulations. We have thus regularly monitored self-employment to see whether there were any causes for our concern. The evidence to date does not suggest that this has been the case. Since the introduction of the minimum wage, self-employment in the economy has grown by 13.0 per cent while it has fallen by 1.0 per cent in the low-paying sectors. Over the last year (between the third quarters of 2007 and 2008), however, self-employment in the low-paying sectors has increased by 29,000 (about 2.9 per cent), faster than the growth in self-employment for the whole economy (0.9 per cent).

2.67 When the October 2007 upratings came into force, the stock of claimant unemployment was falling as claimant outflow was greater than inflow. By March 2008, however, this had reversed with those becoming unemployed outnumbering those leaving the claimant count. Since then, as shown in Figure 2.20, the monthly inflow has increased from 205,000 to 360,000 in February 2009. Over the same period, the outflow from the claimant count has also increased, but at a slower rate, from 202,000 to 250,000.

Figure 2.20 Claimant Unemployment Inflow and Outflow, Thousands, UK, 1989–2009

Source: ONS, claimant count inflows (DPRD) and outflows (DPRE), all aged 18 and over unemployed, monthly, seasonally adjusted, UK, 1989–2009.

2.68 Vacancies and redundancies can also be used to gauge the health of the labour market. Figure 2.21 shows how the recession has had an impact on both redundancies and vacancies. Like hours, the number of vacancies in the economy peaked in March 2008. At the same time, the number of redundancies started to rise, creating a mirror image.

Source: ONS, redundancies (BEAO) and vacancies (AP2Y), monthly, seasonally adjusted, UK, 2001–2009.

2.69 The research undertaken for this report is, for the most part, based on data up to the end of 2007. The downturn in the labour market only becomes apparent in the data from the spring of 2008. From this point, employment growth slowed, vacancies fell and redundancies rose steeply, leading to sharp rises in unemployment and, by the autumn, employment and the number of employee jobs had fallen. Although the research summarised below provides an informative addition to our knowledge of the impact of the minimum wage, it should be noted that it does not take account of the most recent downturn.

2.70 Dickens, Riley and Wilkinson (2009) examined the employment effects of the 2003–2006 upratings to the minimum wage. The researchers built on previous work by Stewart (2002, 2003, 2004a and 2004b) and Dickens and Draca (2005) to investigate the impact of the recent large upratings on individual employment transitions. Their analysis of job retention finds mixed evidence, for both men and women, that the minimum wage had affected the probability of remaining employed. They do, however, find a generally negative impact of the minimum wage on job entry for women, but it is not robust and is not statistically significant. The research from the local area analysis also fails to find strong evidence of an adverse employment effect, but they found some weak evidence that increases in the minimum wage may have led to increases in unemployment.

2.71 Dickens, Riley and Wilkinson (2009) also investigated the impact of the minimum wage on hours, developing the work by Connolly and Gregory (2002) and Stewart and Swaffield (2004). They too found evidence that hours have been reduced in some years and in some econometric specifications but, unlike the findings of Stewart and Swaffield, the latest research findings were not robust. In conclusion, Dickens, Riley and Wilkinson (2009) could find no compelling evidence that the recent large minimum wage upratings had led to an adverse impact on employment, but some effects were detected on hours.

2.72 Dolton and Wadsworth (2009) also built on earlier work by Stewart (2002), which pointed out how the minimum wage reaches further up the wage distribution in certain parts of the country than in others and used these variations to investigate the impact of the minimum wage on employment growth. They found that areas where the minimum wage has greater effect were associated with a significant fall in wage inequality in the bottom half of the wage distribution. Although they found little impact of the minimum wage on employment over its entire period of operation, examination of yearly effects suggests a small but significant positive effect of the minimum wage since 2003. Like Dickens, Riley and Wilkinson (2009), however, they also found that areas where the minimum wage bite is greatest experienced higher unemployment but noted that unemployment rates fell more in these areas in the latter part of the period under study.

2.73 Experian (2009) examined the impact of the introduction of the minimum wage, and subsequent increases in its level, on staff turnover, retention and recruitment. They found some evidence that the introduction of the minimum wage and the early upratings may have been associated with reduced job search activity and reduced pay-related search activity among minimum wage workers, but they could find no evidence of any effects from the subsequent large upratings. Their econometric analysis of hard-to-fill vacancy data suggested a number of statistically significant changes in reported recruitment problems coinciding with minimum wage upratings but no clear pattern was established. They concluded that there was no evidence that the minimum wage has had any significant effect on job-to-job moves among low-paid workers, or on recruitment difficulties among their employers.

2.74 In conclusion, as the research conducted for this report does not cover the current economic downturn, further research will be required to see if its benign conclusions continue in this period of economic uncertainty.

2.75 As noted previously, firms affected by the minimum wage might increase prices as a way of dealing with increased costs. Figure 2.22 shows that inflation generally rose throughout 2006, fell back in the early part of 2007 before rising again as food, energy and oil prices rose sharply. This sharp rise peaked at the end of the third quarter of 2008 before falling steeply as oil prices fell from nearly $150 a barrel in July to less than $40 a barrel at the end of 2008. Heavy price discounting from retailers, particularly in clothes and household goods, the temporary reduction in VAT from 17.5 per cent to 15.0 per cent, and the reduction in interest rates led retail price inflation to fall back from a peak of 5.0 per cent in September 2008 to just 0.1 per cent in January 2009. The fall in the CPI was not as sharp as the fall in RPI because housing costs are excluded from the CPI. In general, it does not appear that firms have been able to pass on the large increases in fuel, energy and food costs to their business customers. The Services Producer Price Index (SPPI) has followed the upward trend since 2006, but it has remained below retail price inflation, at a rate around 3.0 per cent. Like CPI and RPI, it also fell in the fourth quarter of 2008, to 2.9 per cent.

Figure 2.22 Consumer, Retail and Services Producer Price Inflation, UK, 1998–2008

Source: ONS, CPI (D7G7), RPI (CZBH) and SPPI (DZZ8), quarterly, not seasonally adjusted, UK, 1998–2008.

2.76 For many of the low-paying sectors affected by the downturn in the economy, service producer inflation has been more subdued, with prices for hotel services falling by 1.4 per cent over the year in 2008. Prices for cleaning services, security services, employment agencies, and hospitality all rose by less than the general rise (3.2 per cent) of prices in business services in 2008.

2.77 Although we did not commission any specific research on prices for this report, we can draw on two recent studies that give an insight into whether employers have passed on the costs of the minimum wage to their customers in the form of higher prices. Wadsworth (2007) found some tentative evidence that the introduction and subsequent uprating of the minimum wage may have led certain industries and services to raise prices. In a subsequent study, Wadsworth (2008) found no evidence of a significant change in prices in the month in which the minimum wage changed, but he concluded that prices in several minimum wage industries (those in which the workforce contained a high proportion of minimum wage workers) appeared to have risen relatively faster than prices in non-minimum wage sectors in the period after the minimum wage was introduced.

2.78 As discussed above, firms subject to an increase in labour costs resulting from an increase in the minimum wage may be forced to accept a squeeze on profits. There is little evidence that the minimum wage has had an impact on profits at the whole economy level, although analysis in the 2007 Report showed that the non-oil profit share had been below its long-run average (the average since 1980) since the introduction of the minimum wage in 1999.

2.79 One measure of profitability is the rate of return on capital employed. Another is the profit share, defined here as corporate surplus as a percentage of GDP. Figure 2.23 shows that these two indicators of profitability have tended to move in a similar direction since 1989. Their paths diverged at the end of 2006, however, and they no longer appear to tell the same story. The net rate of return on capital rose from 10.5 per cent in the first quarter of 2001 to peak at 13.4 per cent in the third quarter of 2007, before falling back to 11.6 per cent a year later. The rate of return to non-oil and non-financial services (17.1 per cent on average) has consistently been greater than to manufacturing (8.7 per cent) since the introduction of the National Minimum Wage. But the current rates of return in both sectors are below these averages. In the third quarter of 2008, the net rate of return to manufacturing was just 4.4 per cent compared with 15.5 per cent in non-oil and non-financial services. In contrast to the rate of return measure, profit share peaked at 24 per cent in the last quarter of 2006 and has remained at around 23 per cent in 2008.

Figure 2.23 Profit Share and Rate of Return on Capital Employed, UK, 1989–2008

Source: ONS, net rate of return on capital employed for non-UK Continental Shelf private non-financial corporations (LRXP) and corporate surplus as percentage of GDP (IHXM), current prices, quarterly, seasonally adjusted, UK, 1989–2008.

2.80 Draca, Machin and Van Reenen (2005 and 2006) found some significant and robust evidence that profits had been reduced in the care home sector as a result of minimum wage increases, although Georgiadis (2006) found no such relationship in his follow-up study. Experian (2007) in their analysis of the impact of the minimum wage on profits at industry level found no statistically significant effect for any industries.

2.81 In contrast, Forth, Harris, Rincon-Aznar and Robinson (2009) found a significant negative impact of the minimum wage on the rate of return on capital employed for firms in the ‘most exposed sectors’, defined by the researchers as those sectors with the highest proportions of workers affected by the minimum wage. Their findings on the price-cost margin, although negative, were not as robust and were not statistically significant. Together, these results suggest support for the earlier findings of Draca, Machin and Van Reenen but they are not sufficiently robust to allow us to draw any conclusions with confidence.

2.82 Official data from ONS, as depicted in Figure 2.24, show that productivity growth in the economy as a whole increased sharply in the second half of 2005 before flattening out at just over 2 per cent throughout 2006 and most of 2007. It then peaked at 2.6 per cent in the fourth quarter of 2007. Since then, productivity growth has slowed sharply, with productivity actually declining by 0.5 per cent in the third quarter of 2008. The economic downturn has led to a decline in output that has been faster than the fall in employment. The figure also shows that since 2005, productivity in the service sector has followed a similar trend to the economy as a whole, albeit with productivity growth slightly higher. This series is also available for broad industrial sectors, such as the distribution, hotels and catering sector, of which retail and hospitality are large components. We can see that productivity growth in the distribution sector rose even faster than in the service sector throughout 2006 and 2007, but it has declined much more sharply throughout 2008.

Figure 2.24 Growth in Productivity for the Whole Economy, Total Services, and Distribution (including Retail and Hospitality), UK, 1998–2008

Source: ONS, output per job for the whole economy (LNNP) and experimental series for total services and distribution, quarterly, seasonally adjusted, UK, 1998–2008.

2.83 Productivity data from the ONS Annual Business Inquiry (ABI) are available at a more disaggregated level but is less timely.[5] The latest data are for 2007, a period when productivity was increasing in the economy as a whole. Using gross value added data, and adjusting for employment and hours, the figures suggest that productivity growth, whether measured in terms of per worker employed or per hour, was sluggish in 2005 but rose sharply in 2006 and 2007 in both retail and hospitality.

2.84 As part of the research project on competitiveness, Forth, Harris, Rincon-Aznar and Robinson (2009) investigated the impact of the National Minimum Wage on productivity at both the industry and plant level. The analysis at industry level built on earlier work by Forth and O’Mahony (2003) that looked at the evidence up to 2002 but focused on the impact of the introduction of the minimum wage. Although noting some studies that had found that the minimum wage had led to some improvement in productivity, Forth and O’Mahony’s sectoral analysis found no systematic evidence of any productivity effects. The plant-level investigation follows in a similar vein to work, previously commissioned by us, conducted by Galindo-Rueda and Pereira (2004). They found some evidence of a positive impact of the minimum wage on labour productivity, particularly in service industries, but their results were sensitive to the econometric specification employed.

2.85 Using information on firms from the FAME (Financial Analysis Made Easy) database, Draca, Machin and Van Reenen (2005) also found a positive association between productivity growth and the introduction of the minimum wage but it was not statistically significant. A positive but insignificant relationship was found by Machin, Manning and Rahman (2003) in their study of care homes. In a follow-up study of these care homes, Georgiadis (2006) found no significant relationship between the minimum wage and productivity.

2.86 In their industry-level analysis, using data from the ABI, Forth, Harris, Rincon-Aznar and Robinson (2009) found a negative association between average labour productivity and the minimum wage for all sectors but this becomes positive when the sample is restricted to low-paying sectors or their ‘most exposed sectors’. But none of these findings are statistically significant. They also found some very weak evidence of a negative total factor productivity effect but conclude that there is little or no association between the minimum wage and productivity. Inconclusive results were also found using plant-level data from the Annual Respondents Database (ARD).

2.87 In summary, research has found little impact, adverse or otherwise, of the minimum wage on either labour productivity or total factor productivity in the economy as a whole or in the low-paying sectors in particular.

2.88 Minimum wages might also have an impact on business investment as any squeeze on profits is likely to impair the ability of businesses to invest. Alternatively, firms may increase investment as they try to substitute capital for labour or use investment in new technology to increase productivity. We investigated this issue by comparing business investment in the economy as a whole with that in an important low-paying sector, hospitality, for which relevant data are available. We can see from Figure 2.25 that business investment in the whole economy and in hospitality was stronger in 2007 than it was in 2008. Investment in hospitality held up better in the first half of 2008 than in the economy as a whole; however, business investment in hospitality has been similar to the whole economy averaged over the second half of 2008.

Figure 2.25 Annual Change in Business Investment for the Whole Economy and the Hospitality Sector, UK, 1997–2008

Source: ONS, business investment by industry, chained volume measures, quarterly, seasonally adjusted, UK, 1997–2008.

2.89 The number of firms that enter and exit the market might be affected by the National Minimum Wage. For those contemplating starting a business, an increase in the minimum wage will add to labour costs and might reduce the attractiveness of setting up on one’s own. Increases in labour costs will also affect the profitability of existing businesses. If the increase sufficiently reduces profits, the likelihood of a business failure will increase. Further, a reduction in profits may also deter business start-ups. But the most marginal businesses will be the ones most affected by these considerations. In such cases, an increase in the minimum wage might lead to an increase in productivity as the least productive and most marginal firms close or refrain from entry.

2.90 We start this section by looking at how the number of business start-ups and failures across the whole economy and in the low-paying sectors, particularly retail and hospitality, have changed over time. Although the data on VAT registrations and de-registrations are reasonably reliable, they are not very timely: the latest data available relate to 2007 and so pre-date the start of the current recession. We then look in more detail at a more timely series, company insolvencies. Unfortunately, this series does not provide any information on business creation.

2.91 As shown in Figure 2.26, there has been a net increase in the number of VAT-registered businesses in the UK since 1995. In 2007 around 206,000 businesses were created (up 24,000 on the previous year) with about 148,000 failing (up 5,000 on 2006). The stock of businesses in the economy increased by about 58,000 firms. The stock of VAT-registered firms in the low-paying sectors fell prior to the turn of the Millennium, but since then there has been an expansion in the number of firms. This expansion has been led by hospitality, where the relative number of businesses has grown faster than in the whole economy for every year since 1999 with exception of the latest year, 2007. The number of businesses in retail has also increased since 2003 and the rate of growth has outpaced the rest of the low-paying sectors, although it lags both hospitality and the economy as a whole.

Figure 2.26 Net Annual Change in VAT-registered Enterprises as a Proportion of the Stock, UK, 1994–2007

Source: BERR Enterprise Directorate Analytical Unit, VAT registrations and de-registrations, annual, UK, 1994–2007.

2.92 The number of insolvencies in the economy provides an alternative and more timely indicator of business closures. This is particularly pertinent in the current period of economic uncertainty. There were 4,607 company insolvencies in England and Wales in the fourth quarter of 2008, an increase of 52 per cent on a year ago. Of these insolvencies, company liquidations grew by 34 per cent to 1,562 and Creditors’ Voluntary Liquidations rose 62 per cent to 3,045. There have also been sharp rises in the number of companies in receivership and those in administration in England and Wales. The data record similar changes over the year in Scotland and Northern Ireland.

2.93 Unfortunately, the data by industry lag the whole economy data by a quarter, but they show that there has been an increase in the number of company insolvencies in the third quarter of 2008 in retail and hospitality as well as across the low-paying sectors as a whole. The increase has been in line with that for the whole economy.

2.94 Using the Annual Respondents Database, Galindo-Rueda and Pereira (2004) found evidence to suggest that the introduction of the minimum wage led to business creation being slower in the lowest-paying geographical areas. They were unable to replicate this finding using VAT registration data; however, Experian (2007), using the same source, did find evidence that business creation was lower in those regions most affected by the minimum wage (where pay was lowest). In their industry-level analysis, Forth, Harris, Rincon-Aznar and Robinson (2009) found some evidence that the minimum wage had increased the exit rates of firms.

2.95 In summary, our investigation of the impact of the minimum wage on the economy has found that it continues to exert a significant influence on wages at the bottom of the earnings distribution. The minimum wage has continued to rise relative to both wages and prices. We commissioned a comprehensive research programme that investigated how firms had coped with these additional wage costs, focusing on the period of the relatively large upratings between 2003 and 2006. Firms appear to have adapted to these increases in wage costs by changing pay structures, removing wage premia, and reducing non-wage costs. The research found little evidence to suggest that the increases in the minimum wage had led to reductions in employment or hours worked. There was also no evidence that the minimum wage had led to changes in productivity, but some evidence suggested that profits had been squeezed. In all, we conclude that the minimum wage continues to exert a benign influence on the economy

2.96 These findings have generally been drawn from data up to the middle of 2008, though, and the economic climate has changed dramatically since then. Employment and vacancies are falling, and unemployment and redundancies are rising sharply. It is the first time that year-on-year aggregate employment has fallen since the introduction of the minimum wage. As we show in subsequent chapters, these adverse outcomes are observed across the whole economy, with nearly every sector and most groups of workers affected.

2.97 We next go on to investigate the impact of the minimum wage on the low-paying sectors in Chapter 3, groups of workers in Chapter 4, and young people in Chapter 5. We discuss the recession and its implications in greater detail in Chapter 8.

[1]ONS regards ASHE as a more reliable data set than LFS for estimating earnings. Information on ethnicity, disability and educational qualifications are not available in ASHE, however, thus we need to use the LFS for these characteristics. The estimates from LFS suggest that 6.0 per cent of all jobs are minimum wage jobs, which is slightly higher than the 4.3 per cent estimated using ASHE.

[2]See Appendix 5 in our 2007 Report for a detailed description of this review.

[3]This is not to be regarded as a measure of non-compliance. There are legitimate reasons, such as apprenticeships and piece rates, that enable employers to pay less than the minimum wage.

[4]We define the minimum wage rate in Figure 2.10 as the five pence band that lies from £5.52 per hour to strictly less than £5.57 per hour. In Table 2.2 we will refer to the ten pence band that lies from £5.52 per hour to strictly less than £5.62 per hour.

[5]The Annual Business Inquiry is only available annually; the latest available is September 2007.