Making globalization work for the European middle class – How to combat job polarization

Import competition, offshoring and automation are threatening the middle classes in Europe. New jobs have been created mainly for the lower and the upper segments of the labour market. But not for the middle. And because most of those jobs are in the non-tradable sectors with low productivity growth, this trend also threatens long-term economic growth and increases inequality. Can jobs be created for the middle classes without embracing protective policies?

The struggle to create more and better paid jobs in the middle segments of the labour market in advanced economies is a consequence of what is referred to in economic literature as ‘job polarization’. Pioneering research during the first decade of the 2000s found that the share of employment in occupations that are most associated with the middle class and that pay around the average wage level, like office and customer service clerks, plant and machine operators, and assemblers, has declined rapidly in both Europe and the US. ¹ At the same time the share of employment at the upper end of the occupational skill distribution scale – involving jobs associated with critical thinking, judgment and decision-making, complex problem-solving, interacting with computers and thinking creatively – and at the lower end – low-skilled jobs associated with, for example, selling or assisting, serving and caring for others – have increased substantially in the past two decades.

The declining share of the middle segments of the labour market cannot therefore be attributed to the economic downturn of the past seven years, as data shows that employment structures in Western European countries, Japan and the US had been polarizing long before the financial and economic crisis started. ² And where the ‘squeeze’ of the middle class has often been referred to as an American phenomenon, the IMF’s 2011 World Economic Outlook shows that there is increasing evidence that all advanced economies face “a shift away from middle-income jobs”. ³

Figure 1: IMF, World Economic Outlook, Slowing growth, rising risks , September 2011.

Although there is consensus that typical middle-class jobs are in decline, it is debatable whether job polarization is occurring in all European countries. According to the IMF, which takes three employment shares (highest-paying, middle-paying, lowest paying thirds) based on wages, the middle class in Austria lost out the most between 1993 and 2006. However, during the same period, the country created high-paid jobs in particular and faced a small decline in low-paid jobs. The same trend of upward job mobility happened in France, Luxemburg and Denmark. However, in most other European countries, like Finland, the Netherlands, the United Kingdom, Sweden and Germany, the decline of the middle class indeed went hand-in-hand with a rise in both low- and high-skilled jobs. Interestingly, the total loss of middle-paid jobs was actually greater in the European Union than in the US (see tables). The economic malaise after 2007 accelerated the trend towards job polarization. During the great recession in Europe, it was especially low-paid service-sector jobs that continued to grow in terms of labour share, while manufacturing-sector and public-sector jobs declined at faster annual rates. ⁴

Long-term European job polarization explained

According to Eurofound’s economist Enrique Fernández-Macías, however, who uses relative change in employment by wage and skill quintiles, the picture is more diffuse, but explainable. ⁵Again, without doubt, typical middle-class jobs lost out between 1995 and 2008, but this was not always accompanied by a significant rise in lower-skilled jobs (see also Enrique Fernández-Macías’ expert opinion). Fernández-Macías found that continental European countries like the Netherlands, Germany, France and Belgium are most prone to job polarization. His explanation is that, at the same time, these countries experienced a very significant process of de-standardization of their lowest paid jobs which was absent elsewhere in Europe. The 1990s debate on the ‘eurosclerosis’ of continental Europe and the subsequent attempts by these countries to deregulate employment to create more jobs at the low end of the labour market.

In other parts of Europe, this did not take place at the same pace. Three Nordic countries Denmark, Finland, and Sweden, for example, experienced a similar process of unambiguous upgrading in their employment structures, however, with no growth in the bottom two quintiles and a very significant expansion of the top and second quintiles. This is explained by the fact that these countries have the strongest unions and the most compressed wage structures in Europe, which makes low-paid jobs comparatively expensive and tends to shift production towards higher value-added activities. And also significantly, these countries did not experience the same deregulation process as continental Europe, at least in the period before 2008.

Southern Europe is associated with a pattern of upgrading typified by a significant expansion of the middle layers of employment. This is probably the result of a different kind of process: the very fast pace of employment expansion that these countries experienced after the creation of the European Monetary Union and their access to unprecedentedly lax financial conditions. The resulting bubble in the construction sector led to an enormous expansion of jobs in the middle quintiles, especially in Spain, but also in Greece and Italy. Finally, the mild polarization with a very strong upgrading of the United Kingdom and Ireland seems consistent with the flexibility of their labour markets and the very large expansion of their financial sectors throughout the 1990s and early 2000s.

The drivers behind the decline of middle-class jobs

Generally speaking economists come up with two explanations for changing employment and occupation structures: technological change and economic globalization. In the 1980s and 1990s, the dominant view among labour economists was that technology was more important than trade as a driving force behind changes in the structure of employment.⁶ Change was biased in favour of skilled workers, leading to the hypothesis of ‘skill-biased technological change’ in which jobs that are created by new technology need higher skills, which lead not only to higher productivity but also to more and better (paid) jobs. ⁷  This means that in time, thanks to technology and by adapting to better skills, workers will improve their economic situation.

More recently, views have been shifting somewhat. First, there are different views on the impact of technological change on demand for different types of labour. Some believe that technology mainly replaces human labour in routine tasks – tasks that can be expressed in step-by-step procedures or rules – but cannot as yet replace human labour in non-routine tasks. This is known as ‘routinization-biased technological change’ in which the important point is that ‘routine’ does not map simply into a one-dimensional definition of skill.⁸ Although low-skill production-line jobs in manufacturing can be characterized as ‘routine’, so can many more skilled craft jobs and many clerical jobs that have never been the lowest paid jobs in the economy. By contrast, many of the worst-paying jobs, for example in housekeeping, hotel & catering and personal care, are non-routine in nature and have therefore been relatively unaffected by technological change. The result is job polarization. However, this more nuanced view has recently been criticized. There is a concern that new technologies will lead to the loss of many jobs that were considered non-routine, such as bus drivers, jobs in the financial sector, teaching and caring tasks as computerization advances much faster than previously assumed ⁹. (See much more on the debate about automation in article ‘Better skills will not save middle-class jobs from automation‘ by Jo Michell.)

The second shift in the economic debate is a growing awareness that the impact of economic globalization on employment in advanced economies has been underestimated for too long.¹⁰Concerns relating to trade openness, offshoring and outsourcing in the 1980s and 1990s were largely about the displacement of low-skilled manufacturing jobs to lower-wage countries. ¹¹Mainstream economists did not see a problem because manufacturing jobs would be replaced by high-end jobs in innovation and design and by a flourishing service sector. The major shift in economists’ views is the realization that goods trade with low-wage countries has contributed to the demise of manufacturing employment in the US and many European countries without sufficient upper-end jobs being created by way of compensation.¹²

The rapid economic growth of countries like India, Turkey, Brazil and China in recent years has made many feel that globalization is having a more powerful effect on the structure of employment now than in the 1980s. The focus of concern now is the prediction that service-sector jobs will relocate to developing countries, followed quickly by jobs relating to certain parts of the production process, like technology, innovation and design. ¹³ This article will look more closely at these arguments and what policies are needed.

There is another driver that has not been mentioned much by economists in the debate about job polarization: the importance of social policy and the institutional change. As shown above, job polarization between 1995 and 2008 in continental Europe could be related to the process of deregulation of employment relationships, declining trade union power and their traditional policy of wage compression (see more on the impact of social protection on the middle class in article ‘A divided middle‘ by Ursula Dallinger).

The link with productivity

New is the realization that the loss of middle-class jobs in advanced economies is related to a shift away from industries with high productivity levels and high productivity growth. For example, the construction sector and the financial services sector (which includes finance, insurance and real estate) experienced large employment increases during the housing boom of the pre-crisis period, but these sectors are not associated with high productivity levels. At the same time, there has been a significant decline in jobs in the manufacturing sector, which is well-known for its high productivity levels and productivity growth, also in countries like Japan and Germany where the sector is large and powerful. ¹⁴

This trend goes against the way countries have developed historically: reallocating labour from low- to high-productivity sectors increased national income and provided better job opportunities. ¹⁵ With the opposite now happening, jobs mainly being created in below-average productivity growth sectors (in both the low-skilled or high-skilled segments of the labour market), there is a common belief among economists that this will hamper economic output growth, continue to reduce the labour share in national income in favour of capital (see article ‘Profits without labour benefits‘ by Rolph van der Hoeven), increase inequality, and obstruct the creation of more and better jobs that are so urgently needed.

For many years, mainstream economists held the view that economic globalization benefits the economy as a whole, generating competitive markets in which highly-productive sectors benefit the most via comparative advantage. Less productive markets that need state intervention to survive disrupt the market mechanism at the expense of the productive sectors. In an open market the labour force will move away from sectors that are not performing to healthy sectors that can compete at international level. The winning sectors become the cornerstone of economic growth and the economy should thrive because of the increasing competitiveness, creating more and better jobs. ¹⁶

But the notion that creating jobs has been associated with low productivity gains assumes that something different is going on: workers are not adjusting to move towards more internationally competitive jobs. There are different views on how the labour force adjusts to trade openness. Some argue that blue-collar workers in particular have limited mobility to move to places with better job opportunities. Others point to the age of displaced workers to account for their sluggish adjustment. The older laid-off workers are, the lower their incentive to invest in retraining for a new job or relocating. ¹⁷

Jobs created

So what jobs are actually created and for whom as a result of economic globalization? And does economic globalization destroy more jobs than it creates? These are fundamental questions that can explain the decline of traditional middle-class jobs. Much of the research into economic globalization and its impact on middle-class jobs is focused on the US. Economists David Autor of the Massachusetts Institute of Technology, David Dorn of the University of Zurich, and Gordon Hanson of the University of California in San Diego estimate that rising Chinese import competition explains 25% of the decline in US manufacturing employment between 1991 and 2000, and 41% of the decline between 2000 and 2007. ¹⁸ They find that import exposure in the short and medium term leads to such a shift in the labour market and increases unemployment that it spurs a substantial increase in transfer payments to individuals and households in the form of unemployment insurance benefits, disability benefits, income support payments, and in-kind medical benefits. Their estimates imply that losses in economic efficiency from trade-induced increases in the usage of public benefits are, in the medium run, of the same order of magnitude as US consumer gains from trade with China (lower prices).

There is however a substantial body of evidence showing that foreign competition, especially from low-wage economies, can displace low- and medium-skilled workers in all advanced economies. ¹⁹ But there is more. Research also shows that many jobs in advanced economies have been created in the non-tradable sectors.  On the whole, tradable sectors like technology, consulting and finance have performed well but have not created a significant number of jobs. The retail, construction and hotel & catering sectors were among the major creators of jobs, but these were non-tradable. This shows that there is a relation between import competition and offshoring and the reduction in tradable sector jobs, mainly in the productive sector, which has resulted in the decline of typical middle-class jobs. ²⁰

What about the higher-skilled workers in the middle class? Research shows that offshoring firms employ more people in R&D and design, introduce new products and invest in advanced process technologies more frequently than non-offshoring firms. ²¹ This means more opportunities for high-skilled workers. Furthermore it seems that, at least for European and American firms, when firms start employing high-skilled workers offshore, they also increase the numbers of high-skilled workers employed at home. ²²

This means that offshoring is not, as yet, having a negative impact on jobs created in the higher-skilled segments of the labour market. That explains the trend of job polarization and the upward job mobility in many European countries. However, the evidence is focused on the micro-level of large corporations while, for an economy as a whole, there is no guarantee that the higher skills will not be affected by economic globalization. Especially because most jobs – including high-skilled jobs that have been created by economic globalization (which is more than just offshoring) – are in sectors with low-productivity growth, are likely to be concentrated more in the non-tradable sectors, while also threatened by automation. This will in time have an impact on higher-skilled jobs as well.

The impact of globalization on wages

Till now the evidence focused mainly on what jobs have been created. But what impact are offshoring and trade openness having on wages? There is evidence that higher levels of social protection and investments in training to reduce skill gaps could adjust to the shocks caused by trade openness. In the US it seems that wage movements for specific occupations can be linked directly with economic globalization. ^23. The evidence is much less clear for European countries. Figures from Germany and the UK suggest that relative occupational wage movements in Europe are not strongly correlated with offshoring variables. Although this differs from the evidence for the US, it is not necessarily inconsistent with it since manyEuropean countries have institutions (e.g. minimum wages and collective bargaining) that mute or stop a wage response, especially across middle- and lower-paying occupations.²⁴ (The article ‘A divided middle‘ by Ursula Dallinger in this dossier focuses on the influence of social protection on the European middle class.) However, there is an increasingly widespread view that, also in European countries, offshoring increases the wages of high-skilled workers and decreases those of low- and medium-skilled workers, thus contributing to rising wage inequality. ²⁵

The middle therefore face not only their jobs being replaced, but also a strong decrease in their relative wages. Trade liberalization lowers the long-run wage in the import-competing sector relative to the export sector. ²⁶ Training and reducing relocation costs can significantly help reduce the adjustment costs for workers, especially in the long term, as they increase the number of better-skilled workers. This means that, in the short term, as a consequence of trade openness, wage inequality could be explained by inter-sectoral wage inequality between the winning and losing industries, while in the long run it is driven by an increase in the skill gap. ²⁷ For middle-class workers, especially in import-competing sectors, it is important to achieve the right conditions to adjust to trade shocks, because that would be the difference between them gaining from them or losing out, with the threat of having to find work in lower paid non-tradable service sectors.

Smoothing structural change

There is an increasing awareness among economists and politicians that middle-class workers must be enabled to sustain their families and offer better opportunities for their children while they are at risk of being forced down the income ladder by economic globalization. As mentioned above, one solution that most experts put forward for the hollowing out of middle-income jobs lies in retraining, better education and increased productivity in non-manufacturing sectors. Given the ongoing structural changes and the need to adapt, education and skills have to be improved over time and governments have to promote and subsidize lifelong learning. But action is also needed to cushion some of the human costs of structural change, as workers are becoming prone to falling wages and to changes in the risk level and duration of unemployment due to economic globalization.

As economist Michael Spence argues, redistribution and social protection measures must be part of the policy response: the potential benefits include increased social cohesion and continued support for globalization. He cautions that if the employment challenges confronting the advanced economies are not tackled, countries may resort to “protectionist measures on a broad front [and] the global economy will be undermined”. Unemployment benefits and employer subsidies can raise employment and economic efficiency and mitigate distributional consequences. Employment benefits can support short-run adjustment because they help individuals undertake the move to new activities despite an initial wage cut. However, workers move slowly between activities not just because there are search costs, but largely because sector-specific skills and experience can be lost in transitions. Therefore employment subsidies are a more effective tool to address the loss of sector-specific skills because employment subsidies raise workers’ incentives to invest in new skills.³⁴

However, this does not change the structural causes of the decline in middle-class jobs. To combat that, work must again be linked with productivity growth. And as economic globalization is only one of many different drivers of the trend towards less productive jobs for the middle class (others being, for example, automation and financial globalization) a simplistic solution based solely on skills and education will not change a lot for the middle. Change is needed. And that means thinking about the whole structure of the global economy: starting with the ownership of capital, revaluing labour rather than capital, and investing in productive sectors that are more inclusive and sustainable for long-term economic growth. The good news is that now the pressure of economic globalization is being felt by the middle class, the political cornerstone of democratic states, European politicians are more aware of the consequences and are more willing to act than in the past decade.

The way they will react will be decisive for the future of Europe – economically and politically. This does not mean that Europe must be afraid for doing business on the global level. On the contrary, it should develop an alternative economic globalization agenda in cooperation with other global players that will reverse the short-termism adopted from financialization, reverse the global race to the bottom, and redress the current bias towards the supply side of the economy by recognizing the importance of demand-side policies (wage-led growth). Only then can the European middle class benefit more equally from economic globalization, and the middle will not see the rise of emerging countries only as a threat but as an opportunity on which to build a new future.

Footnotes