The labour market has changed dramatically across most developed economies over the last four decades. These changes are characterised by widening earnings inequality, a rise in female participation in the paid labour force, an increase in education levels, an increase in alternative work arrangements, an increase in immigration and an increase in both domestic and offshore outsourcing. Britain has seen some of the largest of these changes, although the same underlying trends have been experienced in many economies in North America and Europe. Over the 40-year period until 2019, Britain moved from being in the middle of the earnings inequality pack of OECD countries, with a 90:10 percentile ratio of gross earnings for men of 2.7 in 1980, to 3.5 in 2019 and second only to the US. There were similar but less dramatic rises among women. Even though these changes are large, the rise in the 90th percentile of earnings does not capture the huge increases in earnings experienced at the very top, resulting in a highly unequal distribution of labour market earnings, with median CEO pay to the median UK full-time worker pay standing at 119:1 in 2019.
A panellist’s introduction, by Richard Blundell
These overall trends disguise three quite distinct inequality episodes over this 40-year period in Britain: the inequality ‘boom’ of the 1980s and early 1990s; the broader growth period from the mid 1990s to the financial recession; and the productivity stagnation of the last decade. The first episode, the inequality boom, saw large increases in the wages of skilled workers, little real growth in earnings for the low-skilled and a rise in workless households. This set the scene for many of the ensuing trends in labour market inequality and subsequent policy responses. The late 1990s and early 2000s showed a continuing increase in earnings inequality for men but at a more modest rate, except among top earners where earnings continued to race away. There were though increases in real earnings across the distribution for men and for women, and a steady reduction in the number of workless households. Indeed, earnings inequality between women, and between all individuals, was beginning to fall back. It was also a period when the large expansion in degree-level education began to flow through to the labour market. There were concerted policy responses too to deal with low earnings, with the introduction of the minimum wage and the expansion of in-work transfers, mostly through the tax credit system. These policies had a major impact on supporting wages and incomes at the bottom.
The third and final episode, the past decade, has seen a dramatic fall in productivity together with poor growth in earnings across the distribution and a switch away from benefits toward an increasing reliance on the minimum wage to support earnings at the bottom. Minimum wages appear to have been an effective tool to support low wages, with evidence of positive wage spillovers up to at least around the 20th percentile of the wage distribution and little impact on employment. Household earnings inequalities have reversed direction, hours of work for low-wage men stopped falling and hourly wage growth was higher at the bottom of the distribution for both men and women. Nonetheless, these changes did little to dent the overall increases in earnings inequality stemming from the 1980s and 1990s. The UK remains a very unequal economy in terms of labour market outcomes. The stability in household disposable income inequality which was a feature of the British economy from the mid 1990s up until the financial crash has also evaporated in part due to cuts to working-age income-related transfers.
The UK labour market is now in a very challenging position, with low productivity growth, stagnant real earnings and high levels of inequality, by both historical and international standards. The COVID-19 crisis has also left a generation of lower-educated workers with large gaps in their work experience and training. Understanding what is driving labour market inequalities and what can best be done to address the concerns raised by them is at the heart of the contributions for the IFS Deaton Review. Although many of the underlying trends in the UK are shared by other developed economies, there are distinct differences between countries too. Employment rates in Britain have been maintained, unlike in the US and in France. Minimum wages are approaching the levels of some of the high-minimum-wage economies such as France but contractual arrangements are much more flexible and Britain has seen an increasing proportion of those in work not covered by the minimum wage, with very strong growth in solo self-employment. These, and other, differences across countries can provide key insights into what is going on in the UK. They explain why, in the Review, we have sought to draw on the wider experiences of other countries.
The set of contributions to the Review that are focused squarely on labour markets, which we discuss here, examine the key drivers and concerns. But it is worth highlighting that labour market inequalities show up in many other dimensions of inequality covered in other major areas of the Review, including inequalities by gender, race, ethnicity and geography. There are other potential drivers of inequality too such as globalisation, immigration, technology and the behaviour of firms, which are also covered elsewhere in the Review, as are questions about tax and welfare policy. Finally, there are inequalities in family background, educational investments and health which interact in major ways with labour market inequalities. All of these have their own separate contributions in the Review but must ultimately be considered together to get a complete picture of inequality and what should be done about it.
The chapter, by Giulia Giupponi and Stephen Machin, frames the inequality rise of the past four decades in the context of the profound changes that have occurred in the labour market – in particular, the rapid growth in employment in occupations at the bottom and at the top of the skill distribution, with a hollowing-out of occupations in the middle. The authors argue that changes in the world of work are central to the changes in labour market inequalities and that the changing nature of the labour market has resulted in a world of work in Britain that is much more unequal and divisive. They highlight dramatic changes in the structure of work, with more women in work, a strong growth in self-employment and in zero-hours contracts, a fall in manufacturing employment and a growth in services, a growth in immigration, a decline in union coverage and a growth in the minimum wage. The huge decline of unionisation and the diffusion of contractual arrangements at the boundary between dependent employment and self-employment are both likely to have weakened the bargaining power of traditional employees, in favour of employers and top executives within the firm. Union membership rates have decreased across the board from the early 1980s to 2019. They argue that the failure of unions to organise workers in new establishments has been the principal cause of such rapid union decline. In a rather startling graphic, the authors characterise Britain as moving from the north-west quadrant of high-union-coverage / low-minimum-wage countries to the south-east quadrant of low-union-coverage / high-minimum-wage countries over these years, making it clear that labour market institutions have shifted dramatically in Britain.
Despite these institutional changes, the authors attribute the main long-run structural trend of rising earnings inequality to technical change that has benefited skilled workers doing more complex tasks. They note the important role of the education expansion in Britain and that the wage premium for university-educated workers rose sharply from 1.49 in 1980 to 1.60 by 1990 but then remained quite stable for 20 years until 2010. They argue that this initial rise is in line with the idea that education has become more highly valued in the labour market and that this is one of the key features of rising earnings inequality. However, in the face of very big increases in the number of graduates, the graduate earnings premium stabilised and subsequently declined in the 2010s. The authors are at pains to point out, however, that there is no single explanation for the changes in earnings inequality and that institutional changes, particularly through the decline in unions and the growth in the minimum wage, have played an important role alongside technology and skills.
The authors emphasise the episodic nature of the evolution of earnings inequality in Britain and separate out each of the last four decades. The 1980s were clearly the years of fastest growth in earnings inequality, with earnings gaps widening in both the upper and lower parts of the distribution. The 1990s appear to be a toned-down version of the 1980s, with slower but still positive increases in earnings inequality. Up to the 1990s, earnings inequality evolved similarly for men and women. The 2000s saw upper-tail inequality rise significantly but lower-tail inequality grow less. The growth in male earnings was much greater at the top of the distribution than at the bottom, with upper-tail inequality continuing to rise among men, reaching its peak in 2011. Conversely, growth in hourly wages was relatively smoother along the distribution, with faster growth at the bottom and at the top. These dynamics imply that, for men, growth in earnings inequality over this period is largely explained by drops in hours at the bottom of the wage distribution. The authors identify the final decade as a period of slowdown and stagnation for earnings growth, with a little change in upper-tail earnings inequality and a more pronounced reduction of lower-tail earnings inequality.
The authors put the trends in employment and earnings over the last decade together to argue that the slowdown of real earnings growth is not explained by a low employment rate or a high unemployment rate. Indeed, the aggregate employment rate in Britain has outstripped its pre-global-financial-crisis level since 2014 and the (ILO) unemployment rate was, by 2019, at its lowest level for 40 years. Instead, they point to the changes in the world of work as central to understanding the changes in labour market inequalities, especially the growth of alternative work arrangements and job polarisation. The idea is that traditional work patterns have become less delineated in the face of big compositional shifts in who works and the definition and structure of modern jobs. Two related significant shifts have been the rise in solo self-employment and the emergence of new informality via the gig economy. These have radically changed the jobs landscape and the way people work.
The new forms of work the authors highlight include platform-mediated or gig-economy work, temporary contracts of very short duration, contracts with no guaranteed hours of work (on-call and zero-hours contracts), and more generally contracts falling under the umbrella of solo self-employment, such as freelance and contract work. These work activities are also mostly outside the scope of regulation applying to traditional employment contracts and are usually not covered by certain employment rights, such as holiday pay, unfair dismissal protection, minimum wage regulation and social insurance provision (e.g. unemployment insurance, health insurance, retirement benefits, paid sick leave and paid family leave). Solo self-employment is given particular prominence. The share of the lowest-earning quintile who are self-employed has increased by 54%, or 8 percentage points, from 15% to 23%, between 1999 and 2019. This rise has been entirely driven by growth in ‘solo self-employment’ – own-account workers without employees – which now accounts for 12% of all employment. The vast majority (85% in 2019) of the solo self-employed are sole traders, that is unincorporated businesses with a single owner and no employees. This growth has left the UK with one of the highest self-employment rates and with the highest proportion of solo self-employed among all self-employed individuals.
The authors note that the flexibility of solo self-employment and the gig economy can provide workers with ‘insurance’ against adverse shocks, but they also point to research that suggests that the majority of workers prefer the attributes of traditional employment relationships, and that this holds even among those in alternative work arrangements. They conclude that these results provide little support for the hypothesis that the rise of alternative work arrangements is merely the result of a shift in workers’ preferences, or an increase in the availability of work that they would always have wanted. Indeed, they point to research that finds short-term work in the gig economy delays re-entrance into traditional employment, with negative consequences for long-run earnings. The authors argue that, alongside the change in technologies that has created new occupations at the lower end of the distribution, part of the growth in alternative work arrangements is driven instead by the distorted incentives generated by employment and tax regulations, which encourage firms to hide de facto dependent employment positions under self-employment conditions. The solo self-employed are characterised by high rates of underemployment and the authors argue that the growth in solo self-employment, especially of the gig-economy type, is partly a symptom of poor opportunities in the traditional employer–employee labour market. Among the solo self-employed, only 30% perceive good opportunities for career advancement, the lowest share across all groups of workers. Consequently, the authors argue, the growth of solo self-employment may be capturing slack in the labour market.
The authors conclude that rising earnings inequality has come about because of relative demand shifts that have been driven by technical change. But this is not the whole story, especially over all four decades since 1980. There are at least two other important developments, the first being changes in labour market competition that have weakened the bargaining power of workers and the second being the role played by labour market institutions, in particular unions and minimum wages. A picture emerges of high employment levels, stagnant real earnings and increased earnings inequality, coupled with sizeable shifts in the composition of work. The authors argue that to halt or reverse the adverse inequality trends in the labour market requires significant attention be devoted to ways to restore and reinvigorate real earnings growth and to generate decent jobs with good career opportunities in an inclusive way. This aligns with the resurgence of interest in policies to create ‘good jobs’ with career progression for the low-educated and not to rely exclusively on minimum wage and in-work benefit support policies.
The degree to which employers have market power over the wages of their employees and can drive down the pay of their workers is a key consideration in understanding wage inequality and the design of appropriate policies to address low wages. In their paper, Alan Manning and Barbara Petrongolo develop a new empirical framework to study monopsony in the British labour market. They note that the market power employers have over their workers may derive from search frictions and/or heterogeneous tastes among workers for non-wage amenities offered by different firms. The important challenge for empirical measurement is finding the correct definition of labour markets, namely the set of workers who are likely affected by changes in monopsony power in a certain local labour market segment. Labour markets are typically defined according to geographic vicinity and/or industry and occupation similarity. The authors argue that many standard definitions of labour markets lead to mismeasured concentration indices due to workers considering options beyond their narrowly defined local labour market. The resulting empirical relationship between labour market competition and wages is therefore biased. Looking at the local labour markets in England and Wales, they characterise labour market segments based on 8,848 Census wards, and estimate mobility patterns across wards based on worker commuting flows from the 2011 Census of Population. They find that labour market concentration would be much overstated if one did not take into account worker mobility across wards, and that the concentration index obtained on overlapping local labour markets is negatively and significantly correlated to local wages. They find that even moderate local mobility – with a median commute of about 6 kilometres – has quantitatively important effects on the resulting concentration index. They find both indicators are declining over time, except for a moderate blip during the Great Recession. Their results point to a level of employment concentration in local labour markets that was slightly falling in England and Wales during 2000–19. They conclude that, although the level of market power in some local labour markets remained high, the evidence in their paper suggests that changes in monopsony power alone may not be an important factor behind changes in wages over the past two decades.
In his contribution to the Review, Orley Ashenfelter develops the theme of market power and takes a deep dive into the laws and law enforcement that have been associated with failures of competition in US labour markets. He argues that the rebirth of interest in competition in labour markets, in the US at least, has been stimulated by the divergence of wage growth from productivity growth. In the three decades following World War II, productivity growth and real wage growth in the US were similar. But in the four decades since that time, productivity growth has been four to five times as high as wage growth. The gap between wage growth and productivity growth implies that the share of aggregate real income paid to workers has declined. Of course, there are technology explanations for this but a decline in labour market competition is one obvious explanation. The author points to four key areas of failure of competition in the labour market: enforcement actions and antitrust lawsuits regarding explicit conspiracies to suppress competition in labour markets; forced abolition of franchise contracts that include ‘no-poaching’ clauses; mergers that affect labour market competition; and regulation affecting non-compete and non-solicit clauses in employment contracts. He points to some very visible examples of explicit collusion in labour markets, and argues that the decline in the strength and size of the trade union movement has meant that few natural checks on competitive failures exist outside of government actions. He provides some key examples from the tech sector relating to the ‘no-poaching’ and ‘no-solicitation’ agreements reached by Silicon Valley executives, which affect software engineers and animation coders. The case was brought against Adobe, Apple, Google, Intel, Intuit and Pixar alleging that these companies engaged in a series of bilateral ‘no cold call’ agreements. For low-wage workers, the author points to explicit contractual ‘no-poaching’ clauses that have existed in many franchise agreements until recently. These agreements typically prohibit a franchisee from hiring another franchisee’s employees for some pre-specified period of time after an employee’s departure. Since employees typically apply online, the application form simply asks if the applicant has worked for another franchisee. He notes that these clauses existed in about half of franchise agreements in the US, but were not universal. He concludes that it is difficult to know the extent of labour market anticompetitive behaviour today and how it has changed over time. As a result, it is difficult to provide any reckoning of its role in wage stagnation or increased inequality. What is clear is that the tools needed to investigate labour market competition do not exist in many countries. More transparency in employment contracts is a sensible starting point for studying these issues.
In their paper, David Howell and Arne Kalleberg explain labour market outcomes as a function of the balance of bargaining power between employers and workers, a reflection of both collective effects of institutional protections and market forces (aggregate demand as well as relative demands for skills). In this view, the extent to which technology, education and other workplace-relevant skills matter for long-term changes in the distribution of wages and income is also itself determined by prevailing institutions, policies and structures. If bargaining power is central to job-quality outcomes, there are important roles for product market regulations designed to increase competitive market forces by reducing employer monopsony power over suppliers as well as for protective labour regulations over wage-setting that provide workers with countervailing power and employment and income protections that reduce the cost of job loss. The authors argue that increasing relative demands for high non-routine cognitive skills do not offer a credible explanation for patterns of wage inequality across the rich world. Instead, they put forward a political economy perspective and explain labour market outcomes as a function of the balance of bargaining power between employers and workers. They argue that increasing the share of college degrees is not the remedy for turning around the last four decades of unshared growth in the US. Nonetheless, they argue for improvements in the quality of education systems and an increase in access to higher education, to match future demand and because it is essential for good citizenship in democracies. They argue that access to educational opportunities needs to be extended to institutional alternatives to colleges and universities, such as community colleges and other institutions that specialise in preparing workers for the future workplace. They conclude that institutional and policy arrangements are possible that can generate both higher wages and lower inequality while maintaining high levels of job opportunities.
Pierre Cahuc provides an important comparison with France. Like several continental European countries (in particular, Greece, Italy, Portugal and Spain), the French labour market is structured in a dual manner: on the one hand, holders of permanent contracts, protected by numerous and overly codified rules; and on the other hand, temporary contracts, whose terms are known in advance. He describes France as a high minimum wage country with a rigid scale of centrally or sectorally negotiated minimum wages, and legislation to ensure job stability for employees on permanent contracts. This, he argues, is very different from the approach adopted in the UK, where there is more flexibility in wage formation and employment contracts, and it has enabled France to contain wage inequality. The downside of the French strategy though is a low employment rate and a high unemployment rate with significant inequalities in access to employment. The upshot is that the poverty rate before taxes and transfers is particularly high in France compared with the UK and all OECD countries. However, the poverty rate after taxes and transfers is much lower in France than in the UK, since France is the country that reduces the poverty rate most through taxation and social transfers. The French policy has been to reduce employer social security contributions on low wages in order to limit the impact on employment of high labour costs. These policies have grown in scale and the reductions in social security contributions on low wages, below 2.5 times the minimum wage, reached about 60 billion euros in 2019, or more than 2.5% of GDP. These reductions in employers’ contributions have been shown to significantly reduce the cost of low-skilled labour. France is characterised by a very high rate of coverage of collective agreements which is not attributable to a high rate of unionisation (which is among the lowest in the OECD countries) but to sectoral collective agreements negotiated by trade unions and employers’ organisations and extended almost systematically to all employees. This labour market regulation significantly limits wage inequalities insofar as wage agreements set wage floors according to occupation, diplomas, seniority in the firm, work experience and the tasks the employee is capable of performing. The author argues that the high pre-tax-and-transfer poverty rate induced by labour market regulation implies that the relatively low post-tax-and-transfer poverty rate in France is achieved at a significant cost to public finances.
Daron Acemoglu and Pascual Restrepo, in their article, take a more technologically induced approach to explaining the pattern of wage inequality. However, rather than the standard skill-biased technological growth framework, they develop a task-based approach in which technical innovation can result in certain tasks being replaced while other new tasks can be created. The idea is that most technologies improve the productivity of a factor in some specific tasks or reallocate some tasks from one factor to another but typically also create new tasks. In this framework, the negative effects of automation, and its potentially adverse consequences for labour demand, can be counterbalanced by the creation of new labour-intensive tasks, which can reinstate labour into the production process. Problems occur when there is an imbalance between displacement and reinstatement. The authors’ evidence suggests that there has been a marked shift in the nature of technological progress over the last three decades towards much greater automation and much less reinstatement. They argue that this may have resulted in excessive automation. Added to which, tax policy in the US has created an inefficient bias towards automation, taxing labour much more than capital, especially capital involved in automation, such as equipment and software. The authors argue that the reason why young low-educated men have experienced a 15% real wage decline since 1980 in the US is that they were concentrated in various routine occupations in manufacturing, mining, retail and wholesale industries. They argue that the introduction of new tasks and the associated reinstatement of labour have slowed down in the US and they estimate that automation technologies and the task displacement they generate account for a large part of all changes in US wage structure.
To round things up, the paper by Jonathan Cribb, Robert Joyce and Thomas Wernham takes a deep-dive look at individual earnings inequality and household disposable income inequality over the last 25 years in Britain. The authors point out that many of the trends seen in Britain have been experienced elsewhere in the world with the US as an outlier but, within Europe, Britain has seen some of the most dramatic shifts in labour market inequality. They begin by noting that the proportion of working-age households with no one in paid work has been falling for most of the period, reducing inequalities in household labour income across the working-age population. Between the mid 1990s and the Great Recession, however, the gap in earnings between low-earning working households and higher-earning working households was rising, due in part to an increasing tendency for low-wage men to work part-time. But increasing fiscal redistribution kept the gap in disposable income between those same households roughly constant, while also closing the gap between the incomes of workless households and the rest. Together with the falls in worklessness, this was sufficient to achieve some decline in income inequality across the middle 90% of the distribution. In the past decade, key trends turned around. The reliance on redistribution through in-work benefits such as working tax credit switched dramatically toward the minimum wage. Household earnings inequalities reversed direction, as hours of work for low-wage men stopped falling and hourly wage growth was strongly progressive for both men and women – in part due to the rising minimum wage. Yet household disposable income inequalities also reversed, in the opposite direction, due to large cuts to working-age income-related transfers.