3 Impact on Incomes

In Volume 1 we assessed the impact of the minimum wage on the earnings of workers in different groups and sectors of the economy. We showed that the earnings of the low paid have been rising faster than the earnings of those higher up the distribution, pointing to the impact of the minimum wage in narrowing the earnings distribution, and in narrowing the gender pay gap.

In this chapter we examine whether increased earnings for the low paid translate into increased income for low-income families. We look at the position in the income distribution of those affected by the minimum wage, and discuss the impact of the tax and benefits system and employment effects. We present new research on the impact of the minimum wage on poverty rates. Later in the chapter we consider the impact of the minimum wage in providing an incentive to work, and discuss the importance of the minimum wage as part of a package of in-work support to promote the decision to work.

In this section we look at the interaction of the minimum wage with the income distribution and numbers in poverty. Some critics have argued that minimum wages are ineffective in tackling poverty since the poor are mainly pensioners or those without work (see, for example, Gosling, 1996). Others have pointed to disemployment effects of minimum wages leading to increased numbers on low income. As well as presenting new research on these issues, we examine the role of the tax and benefits system in determining the gains from a minimum wage.

Looking at the whole population, it is clear from the first two columns of Table 3.1 that the minimum wage affects the bottom of the income distribution: 12.7 per cent of the bottom decile are affected, and 12.6 per cent of those affected are in the bottom decile. However, individuals in the bottom decile of the income distribution are not the group most affected, because the bottom decile contains many pensioners and individuals in workless households. Although two-thirds of those affected are in the bottom half of the income distribution, most of those affected by the minimum wage are in deciles two to five. The next two columns of Table 3.1 present the same information for individuals in working-age and employed households. The figures show that within this group the minimum wage is much better targeted. The highest incidence of those affected is in the bottom decile and 80 per cent of those affected are in the bottom half of the income distribution.

Table 3.1 shows that there are some people gaining from the minimum wage at the top of the income distribution. In work which we commissioned, Sutherland (2001) found that the most important reason households further up the income distribution benefit from the minimum wage is that they are in receipt of significant other sources of income or contain more than one worker.

Across all households, the effect of the minimum wage is to increase incomes by 0.9 per cent. The biggest gain occurs at the bottom of the income distribution: a 2.3 per cent increase for those in the bottom decile. So, although there are fewer individuals affected by the minimum wage in the bottom decile, the proportionate gain in income is the most important for this group. Leaving out pensioner and workless households means that the proportionate gain becomes even stronger – nearly 30 per cent of all gains going to the bottom decile. Gains from the minimum wage will therefore be disproportionately felt at the bottom of the income distribution, and we conclude that the minimum wage has the potential to reduce income inequality, in particular among working households. This effect could be reduced through the workings of the tax and benefits system or by negative employment effects, both of which are addressed later in this chapter.

We also looked at the effect of the minimum wage on household income by type of area. We commissioned Phimister et al. (2001) to examine the impact of the introduction of the minimum wage on poverty in rural, semi-rural and non-rural areas. The results of their analysis are presented in Table 3.3, which gives the proportion of households in each income quintile by type of area and the proportion of sub-minimum wage workers. Their figures use pre-minimum wage data, and numbers of sub-minimum wage workers are not therefore an estimate of non-compliance.

Phimister et al. (2001) conclude that, in the absence of any negative employment effects, the minimum wage can be expected to reduce income inequality to a small extent, as the proportion of people affected by the minimum wage is greater at the lower end of the income distribution. They state that 'the results suggest that the minimum wage will have positive effects on the household income distribution. However, depending on hours worked ... the potential effect would seem to be greater for the remote rural sample as 57% of sub-minimum wage workers fell into the lowest two household quintiles.'

So far we have looked at the distribution of minimum wage workers in the income distribution and the distribution of gross gains. In order to determine the effect of the minimum wage on net incomes, we need to assess the way in which the minimum wage interacts with the tax and benefits system. Appendix 3 describes this interaction and shows how income from benefits and tax credits falls as earnings increase.

The extent to which net income increases as gross earnings rise is measured by marginal deduction rates (MDRs), which show how much of every extra £1 that a worker earns is lost as they pay more tax and National Insurance and have their benefits reduced. As shown in Appendix 3, a family in receipt of Working Families' Tax Credit (WFTC), Housing Benefit (HB) and Council Tax Benefit (CTB) could face a MDR as high as 94 per cent. This means that for every additional £1 earned in gross pay, the family would see an increase in net income of just 6 pence. HB and WFTC have the highest tapers – 65 per cent and 55 per cent respectively – so workers in receipt of these, or a combination of two or more in-work benefits, tend to be the ones with the highest MDRs. The situation in which workers face high MDRs, and therefore have little incentive to increase their earnings, is called the 'poverty trap'.

The lower the MDR faced by workers the higher the gain in net income as earnings rise. It has not been possible to determine precisely the MDRs faced by minimum wage workers, but modelled estimates from the Family Expenditure Survey (FES) of numbers benefiting from the minimum wage in 2000/01 show that around 165,000 benefit units are eligible for WFTC and contain at least one worker on the minimum wage. This is around 13 per cent of all benefit units that contain a worker on the minimum wage, so a high proportion – 87 per cent – contain a minimum wage worker not in receipt of WFTC. Some of these will be in receipt of HB or CTB, but this figure suggests that there is a significant proportion of benefit units affected by the minimum wage which are not subject to the highest MDRs.

Note: Results from POLIMOD, based on 1994/95 and 1995/96 Family Expenditure Surveys, updated to 2000/01 prices and incomes, based on minimum wage, tax and benefit policy announced by April 2000.

In Volume 1 we reported on research we commissioned from Stewart (2001) to examine the employment effect of individuals affected by the minimum wage; the study found no evidence of an adverse employment effect. Stewart (2001) found that the minimum wage had a broadly neutral impact on employment for adult men and women and young people. Lack of a disemployment effect suggests that the results from our previous analysis – that the minimum wage should have reduced income inequality – hold good.

So far this chapter has established that the minimum wage helps those with low household or benefit unit income, especially the poorest working households. In this section we examine whether the minimum wage has furthered the Government's objective to reduce overall poverty, which for the purpose of this chapter is defined as 60 per cent of household median income.

There are three ways in which the minimum wage can have an impact on poverty. One is direct: by raising the income of those in work. The other two are indirect: by providing a wage floor onto which tax credits and in-work benefits can be built, and by increasing the returns from work and so encouraging people to enter work. In this section we examine how far the National Minimum Wage has been successful in the first two ways; the section on work incentives, later in this chapter, assesses the third.

In 1998/99 15 per cent of working age adults, and around a quarter of children, lived in households below 60 per cent of median income before allowing for housing costs (DSS, 2000). Sutherland (2001) found that of those classified as poor, only 5.1 per cent are affected by the minimum wage. This figure is lower than the percentage of individuals in households in the lowest decile affected by the minimum wage, which was estimated by Richard Dickens (see Table 3.1). The reasons for the difference are that: Sutherland's work was based on a £3.70/£3.20 per hour minimum wage while the work by Richard Dickens uses a £4.10/£3.50 per hour minimum wage; Sutherland's analysis is on a benefit unit basis and Dickens uses individuals in households; Sutherland's analysis is based on a model of the latest tax and benefits changes while Dickens uses the raw FES data; and finally the two studies use data sets in different years.

Sutherland (2001) estimated the impact on poverty rates of the introduction of the National Minimum Wage in combination with other recent changes to the tax and benefits system announced between April 1997 and April 2000. The analysis shows that while these tax and benefits changes reduce the number of people who are poor by 23 per cent (a decline in the poverty rate from 18.6 to 14.4 per cent), only 1.2 percentage points of this reduction are attributable to the impact of the minimum wage. The reason for the small impact is that poverty is concentrated in households with no one in work or, where they are in work, with high MDRs.

Sutherland estimated that pre-minimum wage, 16 per cent of those affected by the minimum wage were below poverty levels, and that the introduction of the minimum wage reduced this to 12 per cent. This shows that, although receipt of the minimum wage is not concentrated in the poorest households, of the poor who do benefit, around a quarter are brought above the poverty line. Sutherland suggests some reasons why the remaining three-quarters remain below the poverty line: high MDRs which erode the increase in income due to the minimum wage; families not receiving in-work benefits (either because they are not entitled to them, or because they fail to take them up); and working too few hours.

The National Minimum Wage also has an indirect effect on poverty by providing a foundation of earnings onto which tax credits like the WFTC can be built. There is some concern that firms rather than workers will be the beneficiaries of subsidies (such as tax credits) to workers. The introduction of an in-work benefit/tax credit could result in a fall in wages that firms pay rather than an increase in the employment and net income of low-paid workers. Our second report noted that the National Minimum Wage limited the scope for employers to depress wages and rely on tax credits to substitute for pay. Sutherland (2001) argues that where the introduction of the WFTC lifts families with children out of poverty, part of the 'credit' for this should go to the minimum wage, because of its under-pinning of the in-work benefits and tax credits structure. Indeed, she concludes that this under-pinning is 'the main contribution made by the NMW to poverty reduction'.

Central to the Government's approach to tackling poverty is making work pay. Budget 2001 states that 'the Government is committed to tackling the unemployment and poverty traps by making sure that work pays more than welfare, and providing people with incentives to move up the earnings ladder'. In Volume 1 of this report we note that the minimum wage did not appear to have had an adverse impact on the level of labour market participation. Indeed, there is evidence from Stewart and Swaffield (2001) that the minimum wage has increased the range of jobs that unemployed people would be prepared to take. Here we consider incomes in work from the minimum wage, tax credits and benefits, and show how the National Minimum Wage, in combination with recent tax and benefits changes, is making work pay more than out-of-work benefits, thereby encouraging workers to try to earn more.

Source: Based on numbers in Appendix 3. Two-parent family with one worker on the minimum wage working 35 hours a week

Sutherland (2001) looked at the importance of in-work benefits to work-entrant families, estimating the proportion of in-work income that is attributable to in-work benefits. She estimated that benefit units in receipt of in-work benefits and working 16 hours a week at the minimum wage will receive on average 45 per cent of their in-work income in the form of in-work benefits or tax credits. If hours worked are increased to 38 hours a week, this proportion falls to around 27 per cent.

Households without children are not eligible for WFTC, but the Government announced in the March 2000 Budget that the principle of in-work support through tax credits will be extended to people without children through a new Employment Tax Credit. In its evidence to us, the Government pointed out that 'in order to address comprehensively poverty and work incentives, the Government strategy is to combine the NMW with a system of in-work tax credits targeted on low-income households and welfare to work policies, notably the New Deal, aimed at addressing poverty by helping people into work'.

Budget 2001 stated that more details of this strategy would be published in the Summer. We welcome the Government's intention to extend in-work support through tax credits to people without children and we will watch the development of the new tax credit, and of the Integrated Child Credit announced in the 2000 Budget, with interest.

The financial incentives to work depend on the scope and nature of the tax and benefits system as well as entry wages for those returning to work. The problems inherent in the tax and benefits system can be described by two concepts. The first is the poverty trap, where workers who increase their earnings or hours find themselves little or no better off; the second is the unemployment trap, where income in work is little more than benefit income out of work.

We discussed the minimum wage and the poverty trap in the section on tax and benefits and showed that some, but not all, benefit units affected by the minimum wage face high MDRs. The effect of the minimum wage on the poverty trap is slight. Increasing the minimum wage reduces the numbers on means-tested benefits and hence reduces numbers facing high MDRs. The increase in the minimum wage to £4.10 per hour is estimated to float off HB and CTB up to 20,000 benefit units, and around 5,000 will float off WFTC. Thus the minimum wage will reduce the poverty trap for these benefit units.

People seeking work are described as being in the unemployment trap when their income in work is less, or not much more, than income out of work. The extent of the unemployment trap is measured by the replacement ratio (RR) which measures the ratio of out-of-work to in-work income. The higher the RR, the lower the difference between in-work and out-of-work income and the lower the incentive to move from benefits to work. Research by the Policy Studies Institute in 1998 suggests that people aim to be around £40 per week better off in work than out of work. New research by Marsh et al. (2001) among low-income families in Britain found that over half of couples with children and lone parents were, on average, £40 and £36 per week respectively better off in work and claiming FC (the predecessor of WFTC) than living on Income Support; and a majority of them thought that 'work, even uncongenial work, was always to be preferred to unemployment'.

Looking at the level of RRs, Sutherland found that out-of-work income remains a substantial proportion of in-work income. She estimated that the average RR is 77 per cent for benefit units with a worker on 16 hours a week, falling to an average of 62 per cent for benefit units with a worker on 38 hours a week. The RR is higher for couples than for singles, with the highest RRs being for couples with children. Even though the cash returns to work for couples are higher, the relative difference between in-work and out-of-work income is smaller than for benefit units with only one adult.

Sutherland also showed that within family types the range of RRs was considerable. Across single person benefit units working 38 hours a week, 14 per cent faced RRs of more than 80 per cent and 69 per cent faced RRs of less than 60 per cent. Of couples with children, 51 per cent faced RRs of more than 80 per cent, and 7 per cent faced RRs of less than 60 per cent.

Over time, the National Minimum Wage, in combination with tax and benefits changes, has reduced RRs. The latest figures from the 2001 Social Security Departmental Report show that the proportion of the full-time workforce with RRs of 80 per cent or over fell from around 3 per cent in 1997/98 to 1.6 per cent in 2000/01. In Budget 2001 the Government estimated that 'by October 2001, a couple with two children will have to earn just over £150 per week to be £40 a week better off in work than on benefits, compared with nearly £260 a week in 1997'.

Disincentives to work are not just financial. Research into why people work suggests that uncertainty about the transition into work and lack of knowledge of the benefits system are also barriers to work. This issue was raised in our visit to an In-work Benefits Project in Newcastle, where we were told that many unemployed people did not know what tax credits and benefits they could receive in work, and that a substantial proportion of those who were entitled did not claim them. We were also told that, outside the New Deal programme, unemployed people were not routinely advised what they might receive in work. The minimum wage, by providing clarity to the expected level of earnings on return to work, may serve to reduce some of this uncertainty.

The new Working Age Agency to be established later in 2001 will bring together Employment Service and Benefits Agency services for working age people. We hope that this will provide an opportunity to ensure that information and advice on tax credits and in-work benefits are given automatically. We very much welcome the recent advertising campaign by the Government about WFTC. Publicity will be similarly important in promoting high take-up of the proposed Employment Tax Credit and Integrated Child Credit.

In Volume 1 of this report we showed how the minimum wage led to a narrowing of the earnings distribution. The analysis presented here has shown that beneficiaries and gains from the minimum wage are concentrated in the bottom half of the income distribution. Since the minimum wage cannot improve the incomes of pensioner or workless households, minimum wage beneficiaries are better targeted when we look at the income distribution of working age households with someone in work.

Looking only at the income distribution does not take account of the interaction of the minimum wage with the tax and benefits system, and the effect of the minimum wage on net incomes. Taking account of these factors, we found that the minimum wage increases the net incomes of the poorest households, and it reduces the numbers below poverty levels to a small extent. In addition, the minimum wage provides a floor below which earnings cannot fall, thus ensuring that the gains from in-work benefits and tax credits go to workers rather than employers. Our evidence suggests that the minimum wage increases incentives to work and, along with other tax-benefit reforms, helps to reduce some of the problems preventing individuals from moving into work.

The impact of the National Minimum Wage on the income distribution and poverty depends on widespread compliance because, without it, the minimum wage cannot improve the incomes of low-paid workers. Compliance and enforcement are the subject of the next chapter.