Projections suggest long-term labour market trends are set to continue. Examining the impact of these changes on income inequality and poverty levels, this study:
- Forecasts a rise in income inequality and poverty levels.
- Cites expected cuts in benefits and tax credits as a major cause of these rises.
- Suggests the impact on the overall level of relative poverty and inequality forecast for 2020/21 made by changes imposed on employment and pay structures will be small.
Conducted at a time when the role of skills development in a reformed welfare-to-work policy environment is high on the policy agenda, and when job prospects for less qualified workers look likely to remain poor, this research is a timely contribution to a growing debate.
- The projections in Working Futures 2010–2020 indicate that many long-term employment trends will continue, including shifts towards a knowledge- and service-based economy and increases in high-paid and low-paid jobs at the expense of those in the middle.
- These changes in employment structure will contribute to an increase in poverty rates by 2020, although it is the growing gap between benefits and wages that is the main driver of increasing relative poverty rates.
- Given the changes projected, and current tax and benefit policy, both the relative poverty rate (individuals in households with less than 60 per cent of median household income) and the absolute poverty rate (individuals with less than 60 per cent of 2010/11 median household income) will increase.
- The rise in relative poverty occurs because the anticipated changes in pay and employment increase median income, which raises the relative poverty line while having little impact on incomes lower down the distribution scale. The projected changes in employment structure have little impact on the real incomes of low-income households.
- The projected changes in employment structure also lead to increased inequality in net household incomes between 2010 and 2020, with income growth projected to be higher at the top than at the bottom.
- Alternative assumptions about changes in employment structure and pay are also considered, including reducing the gender pay gap and rebalanced regional growth. Generally, these have only small effects on income inequality and relative income poverty.
The research - by a team from the Institute for Employment Research at Warwick University and the Institute for Fiscal Studies
With constrained public spending, substantial progress towards the 2020 poverty targets would require large transformations in the distribution of labour-market income. This project examined how projected changes in employment structure (the way jobs are distributed by gender, occupation, qualifications, region and sector) are likely to affect household income distribution.
The research took place at a time of debate about the role of skills development in a reformed welfare-to-work policy environment, and when job prospects for less qualified workers look likely to remain poor. Evidence of continuing polarisation in UK and European labour markets is well established. While much employment growth over the past decade has been in high-skilled occupations, growth has also occurred in much less skilled areas (especially the less easily automated service sector). However, numbers of skilled manual and more routine white-collar jobs have continued to decline. Since earnings are closely related to skill level, this polarisation of employment has contributed to growing inequality.
This study analysed how further structural changes in employment might impact on poverty and inequality from 2010 to 2020. The main changes projected are: a continued shift towards a knowledge and service-based economy; increases in jobs at higher levels (managers, professional and associate professional) and in some less skilled areas of service provision; and significant increases in the qualification levels of those in employment.
Projections for poverty and inequality
The study combined employment projections from Working Futures with a micro-simulation model of household income distribution in 2020/21. The latter incorporates the employment projections and derives measures of relative poverty and income inequality. The household income distribution model is based on detailed data from the national Family Resources Survey (FRS), which includes the characteristics and income sources of about 25,000 households. The Working Futures analysis focuses on jobs and numbers of individuals in employment, while the micro-simulation model is used to generate the consequences of the employment changes for household income.
The baseline forecast for poverty and inequality is derived from the main Working Futures projections of changes in employment structure between 2010 and 2020. The assumptions about pay and employment structure were then altered to explore further implications for poverty and inequality in 2020/21.
Impact of projected employment structure on poverty and inequality in 2020/21
Figure 1 (see the PDF) illustrates changes in relative and absolute poverty rates for various household types given the projected employment levels for the financial year 2020/21; this is the 'baseline scenario'. In this baseline scenario, the employment rate and the overall employment level increase between 2009/10 and 2020/21 (and unemployment falls). Despite this, poverty and inequality measures are forecast to rise over the decade. Individuals are defined as being in relative poverty if they live in a household with a net equivalised income (adjusted for household size/composition) below 60 per cent of the median in that year. They are defined as being in absolute poverty if they live in a household with a net equivalised income below 60 per cent of the median in 2010/11.
More detailed comparisons for alternative definitions of relative and absolute poverty, for a range of inequality measures, and also for a wider range of different household types, are presented in the full report.
Why the rise in inequality and relative poverty? First, comparing the 2009/10 levels with the 2020/21 forecast where the employment structure does not change, inequality and poverty will rise primarily because earnings are forecast to grow in real terms over the decade, while benefit and tax credit rates typically will not (and some will fall in real terms through savings in social security spending). As earnings are a more important income source for those in the middle and top of the income distribution than those at the bottom, this causes inequality and relative poverty to rise.
Second, relative poverty will also rise because of the projected changes to employment structure. The rates are generally higher in the baseline scenario than in the hypothetical situation (where employment structure remains the same as in 2010). The projected changes in employment structure over the decade also increase pay inequality among workers. They increase median household income (and hence the relative poverty line) by about 2 per cent by 2020/21, but have a much smaller impact on incomes below the median.
Absolute poverty rates will rise primarily because tax and benefit changes over the next few years – principally the £18 billion of welfare cuts per year by the end of the current parliament – imply substantial reductions in household incomes. A key change is using the Consumer Price Index for benefits and tax credits (rather than the Retail Price Index as previously). Levels of absolute child poverty are slightly lower under the baseline scenario than in the hypothetical situation of an unchanged employment structure in 2020/21. The projected changes to employment structure considered alone have almost no impact on absolute levels of poverty, slightly increasing it among adults without children and slightly lowering it among children and their parents.
Varying the assumptions about employment and pay in 2020/21
The research also considered the implications for poverty and inequality of different assumptions about how employment and pay structures might change by 2020. It focused on possible employment patterns in 2020, using a fixed total level of employment indicated by the baseline scenario, but altering factors such as the gender pay gap, balance of employment by sector, and regional distribution of employment. A key finding was that the impact of plausible (but quite significant) changes in employment and pay structures on relative poverty and inequality is generally small.
- A general increase in workforce qualifications results in greater household income inequality, with only small effects on poverty measures.
- Increasing the female share of total employment and reducing the gender pay gap have very small impacts on poverty and inequality measures, largely because most adults (71 per cent) live in households containing both men and women. Such changes reduce gender inequality, but not household income inequality.
- Increased polarisation by occupation results in higher inequality, with little impact on poverty measures.
Implications for particular groups
A qualitative assessment considered the implications for potentially vulnerable groups. Some stand to lose out more than others. For instance, women, disabled people and people from minority ethnic groups (especially black or black British) are likely to be disproportionately impacted by an employment shift from the public to the private sector. This is important in terms of both employment and skills development, given higher levels of training in the public sector.
The consistently small impact on the overall level of relative poverty and inequality forecast for 2020/21 made by the considerable changes imposed on employment and pay structures emphasises the difficulty of reducing poverty and inequality using broad interventions. The three main problems in relying on the labour market to achieve sweeping changes to household income distribution (and to poverty and inequality) are:
- Many individuals in low-income households are out of work. Changing employment structure in isolation will not alter this (though it might affect which individuals are out of work).
- Changes to employment and pay impact on median incomes, and hence the level of the poverty line. In turn this affects the relative status of workless households as well as those on low incomes.
- Incomes at household level are key measures for poverty and inequality, but it is individuals who participate in the labour market. Hence household composition is crucial – for instance, some low-paid workers may be in higher income households. This limits how far general labour market changes can achieve specific poverty or inequality objectives. For labour market changes to make a significant impact, help for workers in low-income households is needed, not just for those who are low paid themselves.
The results should not be taken as saying that investment in an individual's own education and skills cannot help them get out of poverty. However, such investments need to be focussed on individuals from low-income households if they are to be effective in reducing poverty rates at a macro level.
The analysis focuses primarily on the effects of changing the structure of employment in 2020/21, holding the overall level of employment fixed. If employment rates in 2020 were to be higher or lower than implied by the baseline scenario, this would also affect poverty and inequality.
The implications of a higher employment rate would depend on the characteristics of the individuals employed (and of their households). For instance, if the overall employment rate were to increase as a result of people in workless households entering employment, this might help to reduce income inequality or relative poverty. But if adults from higher-income households take these jobs, it could actually increase inequality.
Policy needs to target groups or households most likely to be in poverty, rather than raising employment rates or skill levels across the board, which will simply raise the poverty threshold rather than helping those most in need. Any government wishing to achieve income distribution objectives alongside increases in employment levels would need to take this consideration seriously.