Chapter 4. How skills policies can make the most of global value chains1

This chapter discusses how countries can make the most of global value chains through effective and well-co-ordinated skills policies. These policies have to develop the skills countries need to participate and specialise in global value chains; use their skills pools effectively and anticipate changing skills needs; enhance international co‐operation in education, training and innovation; and deal with the risks and implications of offshoring. A “whole of government” approach is needed to co-ordinate education and training policies with policies such as employment protection legislation, non-compete clauses and migration policies.


Skills play a crucial role in ensuring that participation in global value chains (GVCs) translates into better economic and social outcomes (Chapter 2). They are also vital to enable countries to perform and specialise in the most technologically advanced segments of GVCs (Chapter 3). A broad range of policies can affect the development and use of each country’s skills. Education and training policies can develop strong and relevant skills. Labour market policies influence the allocation of skills to industries and occupations. Management policies contribute to the effective use of skills within firms. Policies that attract foreign students and skilled migrants not only increase a country’s pool of skills but also tap into their knowledge of other cultures and networks. And policies that connect countries’ skills to international innovation networks can help to internationalise the production process.

This chapter shows how countries can design their skills policies to:

  • develop the skills they need to participate and specialise in GVCs

  • use their skills pools effectively and anticipate changing skills needs

  • enhance international co-operation in education, training and innovation

  • deal with the risks and implications of offshoring.

Many types of policies outside the skills domain also influence countries’ capacity to seize the benefits of GVCs, including trade, competition and infrastructure policies. It is beyond the scope of this publication to discuss such policies in detail, but this chapter gives some insights on the importance of aligning skills policies with other types of policies to make the most of GVCs. A “whole of government” approach is needed to enable such alignment.

The main findings of this chapter are:

  • Improving the design and quality of skills policies not only helps countries seize the benefits of GVCs, but also helps countries address youth unemployment challenges. In contrast, policies that target specific industries or aim to attract foreign direct investment are more risky and do not lead to this double dividend.

  • To seize the benefits of GVCs, many different kinds of policy need to be aligned. A “whole of government” approach is needed to co-ordinate education and training policies with policies such as employment protection legislation, non-compete clauses and migration policies. For instance, investment in education and training policies may not be able to boost performance in global value chains if migration policies prevent countries from attracting foreign students.

  • As tertiary education develops in emerging and developing economies, OECD countries compete within GVCs through the skills of their populations rather than through their level of education. This means each country needs to enhance the quality of its education and training systems. First, as well as focusing strongly on developing cognitive skills, countries need innovative teaching strategies and flexibility in the curriculum choice in tertiary education so that students can also learn social and emotional skills and multidisciplinarity can develop. Second, a uniform quality of education across schools and programmes is crucial. In many countries, learning outcomes are strongly tied to social background, not only limiting the skills pool but also blurring the signals for employers of the actual skills workers may have.

  • Countries’ comparative advantages in trade emerge from the interaction between skills characteristics and industry requirements, so countries need to improve co-operation between education and training systems and the private sector. Such co-operation can include vocational education and training with a strong work-based learning component, local initiatives to link education institutions to the private sector, and specific policies to foster interactions between the private sector, universities and research institutions.

  • It is vital to ensure that skills are matched to industries as well as possible. This can be achieved by designing employment protection legislation that provides flexibility to firms and security to workers, monitoring the development of non-standard forms of employment, and improving knowledge on the incidence and implication of non-compete clauses that can impede skills mobility.

  • Management policies can be a source of comparative advantage in GVCs. Entrepreneurship education can foster awareness and knowledge of best practices for employers and workers.

  • GVCs make it harder for countries to recoup their investment in education. To overcome this hurdle, countries that belong to the same value chain can co-operate in designing education programmes and possibly in the financing of education to achieve solutions that distribute equally the benefits of GVCs. Co-operation in designing education programmes is a way to ensure quality, maintain knowledge of the development of skills that have been offshored but could be brought back later, and raise the skills in developing economies.

  • Countries have to improve their recognition of skills acquired abroad or informally to facilitate internationalisation of the production process and benefit from it. Recognising skills acquired abroad would attract foreign students and foreign workers who can contribute to research, innovation and performance in an international context. Recognising skills acquired informally would help workers exposed to the risks of offshoring to gain further qualifications and adapt their careers to changing needs.

  • To limit the risks and costs of offshoring, countries need to find the right balance between short-term training and labour market programmes for displaced workers, and long-term policies that facilitate the development of skills at various phases of life. Removing the obstacles to adult education means working on various fronts: improving the tax system to provide stronger learning incentives, easing access to formal education for adults, and working with trade partners to enhance flexibility in the sharing of time between work and training.

Developing the skills for participation and specialisation in global value chains

Well-designed policies can develop strong skills, by providing high-quality education at all levels and flexible pathways to success, and by preventing students from dropping out of school (OECD, 2015a). This section examines how policies can develop the skills characteristics that countries need to specialise in the most technologically advanced industries (as evidenced in Chapter 3) and ensure that participation in GVCs benefits the whole economy by including many firms (as discussed in Chapter 2).

Education and training policies to develop strong skills mixes


Firms and workers need a wide range of skills to participate and perform successfully in GVCs. Cognitive skills play a leading role, but for countries to specialise in particular industries they need the right mix of different skills, including social and emotional aspects of skills as well as cognitive aspects (as evidenced in Chapter 3).

Emerging countries are enlarging access to education, improving the quality of their education and training institutions, and developing technical education programmes. In this new international context, tertiary education or science, technology, engineering and mathematics (STEM) skills become less and less a source of comparative advantage. However, when STEM skills are combined with other fields of knowledge, and other types of skills such as communication, teamwork and problem solving, they can be a source of comparative advantage. For instance, a population well equipped with data analysis skills has enabled Singapore to raise its participation in GVCs, but its data specialists generally lack good problem-solving skills, preventing the country from specialising in the most technologically advanced industries (UKCES, 2015).

The task-based skills indicators (Chapter 3, Box 3.1) show that occupations vary in the intensity with which skills are used on the job, which suggests that workers also vary in their ability in these skills depending on their occupation. The frequency of the performance of various tasks decreases progressively from high-skill to low-skill occupations (Figure 4.1). For all skills, there is a declining trend with age, but the trend is more marked for cognitive skills, reflecting the effect of the increase in educational attainment on the cognitive skills of workers (Figure 4.2). Skills involving personality traits, such as marketing and communicating, show a more stable profile with age, suggesting that these skills may be developed first through initial education and then on the job.

Figure 4.1. Task-based skills indicators varying by occupation
2012 or 2015

Note: Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders and data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD calculations based on the Survey of Adult Skills (PIAAC) (2012, 2015),

Figure 4.2. Skills indicators varying by age
2012 or 2015

Note: Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders and data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD calculations based on the Survey of Adult Skills (PIAAC) (2012, 2015),


OECD countries have recognised the importance of social and emotional skills for performance at work, and are taking action to develop these skills at all levels of education. Social and emotional skills, such as self-confidence, also affect the capacity to develop and use cognitive skills. In the large majority of countries and economies that participate in the OECD Programme for International Student Assessment (PISA), among high performing students, girls do worse than boys in mathematics (OECD, 2015b). However, when comparing boys and girls who reported similar levels of self-confidence in mathematics or of anxiety about mathematics, the gender gap in performance disappears.

Some intervention programmes, particularly in early childhood, have been consistently successful in improving social and emotional skills (Heckman and Kautz, 2013). These programmes typically include pre-school activities and meetings between parents and teachers. For adolescents and young adults, successful intervention programmes generally have a mentoring component as part of a work-based activity. Early interventions can also aim to develop a combination of social and emotional skills and cognitive skills (Box 4.1).

Box 4.1. Improving STEM skills in combination with social and emotional skills from early childhood

In recent years, several OECD countries have introduced reforms to improve instruction of science, technology, engineering and mathematics (STEM) programmes at different school levels. They have made the goal of developing STEM skills along with social and emotional skills explicit by introducing new curricula, standards or assessments, as well as new teacher education and training programmes. Some countries have introduced initiatives in the crucial phase of early childhood, like the Little Scientists’ House programme in Germany.

The Little Scientists’ House initiative was launched to create enthusiasm about science and technology among very young children, following the 2006 Programme for International Student Assessment (PISA) survey which showed that German pupils did not have sufficient skills and knowledge in natural sciences and technology. PISA results also showed that gender differences in attitudes to science were prominent, with boys reporting more positive characteristics than girls on at least five aspects of attitude.

The project is based on three principles:

i) Curiosity and enthusiasm are the gateway to natural sciences and technology for both children and adults. Most children have an inherent interest in natural science subjects, which is driven by curiosity.

ii) Children and teachers design the learning process together (co-construction). Children learn by working with others, but also through individual exploration and shared reflection. The aims of co‐constructive education processes are to develop new contents together; solve problems together; exchange ideas; and get to know different perspectives.

iii) Promotion of basic competences: in addition to stimulating curiosity and enthusiasm about natural sciences and technological phenomena, the project also provides children with basic competences such as learning, language and social skills as well as fine motor skills.

The main activities developed by the Little Scientist’s House initiative are workshops (involving teachers and children) and promotion days (such as Day of the Little Researchers). Teachers and early childhood educators interested in natural sciences, as well as natural scientists/engineers, have been trained by the Little Scientist’s House foundation to teach children and conduct workshops.

The Little Scientist’s House initiative has also set up local networks co-ordinated by a wide variety of partners, such as day care supporters, communities, museums and universities of natural sciences. These networks offer activities and educational programmes for day-care organisations to encourage children to get closer to natural sciences.

During the pilot phase in 2006, the initiative was implemented in 50 preschools in Berlin. Since 2007, it has been extended to the whole of Germany. So far, more than 20 000 preschools, day-care centres and primary schools have been involved. The project plans to expand activities to 47 000 institutions. All educators receive training, information materials and ideas through its local networks.

Source: European Union (2009), “ The little scientist’s house”, Compilation of good practice on fostering creativity and innovation in the fields of learning and cultural awareness,; Siemens Stiftung (n.d.), “Little scientists’ house: Discovering the world with scientific passion”, Siemens Stiftung website,

Promoting social and emotional skills such as self-confidence, self-management and relationship skills demands innovative teaching methods. Strategies that work with students’ feelings and relationships, like role-playing, collaborative-based pedagogies, gaming, case-study and social problem-solving pedagogies, are especially important for promoting communication and emotional skills, as well as pro-social attitudes (Le Donné, Fraser and Bousquet, 2016).

Multidisciplinarity and skills mixes can be developed in higher education by promoting students’ participation in the design of their degrees (Vincent-Lancrin, 2016). In many OECD countries, students can choose some of their courses, usually in addition to mandatory ones. Some countries have gone further by allowing students to take courses in an institution other than the one in which they are enrolled. A modular credit system and partnerships between institutions (for instance, in France and the United Kingdom) allow students to cumulate credits towards a relatively open degree. This openness teaches students to manage themselves and take responsibility by allowing them to study what corresponds best to their professional needs and personal aspirations, while offering them the flexibility to change their study path.

Being able to speak foreign languages is essential to ensure that individuals can move throughout the world and work freely in multi-cultural and multi-lingual business environments. Multilingualism also promotes countries’ competitiveness and ensures social cohesion and intercultural dialogue. A population fluent in foreign languages can attract foreign capital and encourage local companies to join multinationals’ supplier networks. This can be a source of comparative advantage for trade in GVCs, particularly for emerging and transition economies.

Schools and higher educational institutions provide the main opportunities to learn languages. In most EU countries, for instance, a clear majority of pupils learn English from primary level upwards (Figure 4.3). In other countries, teaching of foreign languages begins at secondary level. Sometimes there is inconsistency between different education levels, with the share of pupils learning English as a foreign language dropping remarkably at the lower secondary level (in Bulgaria and Croatia, for instance). The Erasmus Programme of the European Union, which enables pupils and young adults to live and study in another EU country, has been instilling multi-cultural understanding and developing foreign language skills for the last three decades. In addition to improving foreign language skills, Erasmus students increase their soft skills through the programme (European Commission, 2014).

Figure 4.3. Share of pupils in the European Union learning English as a foreign language
Across different education levels, 2015

Note: Data on the number of pupils studying foreign languages are related to the corresponding numbers of students enrolled in each country at each school level.

Source: Eurostat Database,

Achieving more equity in learning outcomes and more transparency in the population’s skills


To specialise in technologically advanced industries, countries need to have pools of workers who perform at the expected level, as these industries have long production chains and poor performance at any single stage greatly reduces the value of output (Chapter 3). Pools of workers who perform at the expected level can emerge in countries in which the skills dispersion across individuals with similar characteristics – the unobservable skills distribution – is narrow.

A large skills dispersion across individuals with similar characteristics may result from a variety of factors:

  • Individuals with similar levels of education do not always have the same level of skills within countries, as the acquisition of skills depends on the quality of schools and education programmes.

  • Work-based learning and training programmes are critical to strengthening the links between the education system and industries, and raising the skills of workers, but the quality of these programmes also varies.

  • There are differences in cognitive skills for the same level of educational outcomes particularly between vocational education and training (VET) programmes and general programmes. According to the Survey of Adult Skills, a much larger share of VET students are low-skilled than of academic students who have spent the same number of years in school but have pursued a general programme (OECD, 2015a).

  • Students from some social and migration backgrounds do not have equal access to opportunities. This also generate differences in skills between individuals with similar education level, for instance between natives with an immigrant background and other natives.

Disparities in learning outcomes start early in the education process and accumulate over life. Educational institutions tend to reinforce existing socio-economic advantage rather than share learning opportunities more equitably (OECD, 2013a; 2016a). PISA results show that 15-year-old students from socio-economically disadvantaged households tend to underperform in the different skills domains surveyed. Differences persist and often increase for young adults, as measured by the Survey of Adult Skills (Figure 4.4). Students and young adults, neither of whose parents has attained a tertiary degree, score in literacy proficiency below their peers with tertiary educated parents.

Figure 4.4. Disparities in literacy between individuals with and without tertiary educated parents at the age of 15 and for 26-28 year-olds
PISA 2000 (15-year-olds) and PIAAC 2012 or 2015 (26-28 year-olds)

Note: The standardised gap refers to the difference in the mean scores of individuals with at least one parent educated at the tertiary level and individuals without tertiary educated level parents divided by the average standard deviation of countries participating in the study. Countries are ranked in descending order of the gap in PISA. Diamonds highlighted in a darker shade, as in Spain, represent groups for which the gap is statistically insignificant at the 5% level. For Greece, New Zealand and Turkey the year of reference is 2015 for the 26‐28 year-olds and 2003 for the 15-year-olds.

Source: Borgonovi et al. (2017), “Youth in transition: How do some of the cohorts participating in PISA fare in PIAAC?”, OECD Education Working Paper, No. 155,

Disparities in learning outcomes resulting from socio-economic disadvantage are closely linked with countries’ unobservable skills dispersion (Figure 4.5). Countries with the smallest score differences in literacy proficiency between 15-year-olds from contrasting socio-economic backgrounds, such as Estonia, Finland, Japan, Korea and Norway (bottom left quadrant), have also the most narrow unobservable skills dispersion. In comparison, countries in the upper right quadrant (including Chile and France, among others), which are characterised by a strong relationship between reading performance and socio-economic status, show an above average skills dispersion for the adult population with similar characteristics. Providing equal learning opportunities to youth disadvantaged by their educational, migration and minority background would ensure that they leave education with stronger skills.

Figure 4.5. Correlation between the unobservable dispersion of literacy proficiency of adults and the impact of socio-economic status on reading performance of 15-year-olds

Notes: Countries on the y-axis are ranked by their unobservable skills dispersion of literacy proficiency as measured by the Survey of Adult Skills in an ascending order. The unobservable dispersion is plotted against the percentage of explained variance of reading performance of 15-year-olds by socio-economic status, which measures the strength of the relationship between reading performance and the PISA Index of social, economic and cultural status as estimated in a single-level bivariate regression model.

Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Adults’ data for Belgium refer only to Flanders and adults’ data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD calculations based on the Survey of Adult Skills (PIAAC) (2012 and 2015),, and PISA database (2012),

Some countries with large socio-economic disparities among 15-year-olds do not have wide unobservable skills dispersions, notably Austria, Germany, New Zealand and the Slovak Republic (in the bottom right quadrant of Figure 4.5). This means other factors are in play. High-quality VET programmes and multiple pathways within the education system, as in Austria and Germany, may provide greater opportunities for success at school and beyond, and thus neutralise the relationship between parental educational background and skills dispersion at earlier stages of education. Having a large share of students enrolled in VET programmes, however, cannot be automatically associated with a narrow unobservable skills dispersion (Figure 4.6); the quality of these programmes and particularly of their work-based learning components needs to be high.

Figure 4.6. Correlation between the unobservable skills dispersion and the share of students in vocational education and training programmes

Notes: Countries on the y-axis are ranked by their unobservable skills dispersion in an ascending order. The unobservable dispersion is plotted against the share of students in vocational education and training programmes.

Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference for adults is 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders and data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD compilation based on the Survey of Adult Skills (PIAAC) (2012, 2015),


In countries where social background has a stronger influence on student performance, differences in performance between schools are larger (OECD, 2016a). Policies should aim to reduce inequalities of opportunity among schools. One option is to try to lessen the concentration of disadvantaged and low-performing students in particular schools. This can require policies outside the skills domain, such as housing policies. Another option is to allocate more resources, including better teachers, to schools with larger concentrations of low-performing students and to disadvantaged schools. The design of the school funding system can not only reduce inequalities among schools but also enhance the quality of education (OECD, forthcoming a).

To ensure equity in learning outcomes, it is also vital to achieve more uniform quality across VET programmes by:

  • Ensuring that the provision of VET responds to labour market needs. This requires good information on labour market needs, mechanisms to link provision more directly to those needs, a diversity of offerings and pathways through education and training systems, and engagement of employers and unions in designing the system.

  • Increasing the quality of VET programmes, by: i) providing comprehensive skills development for employability; ii) integrating high-quality, work-based learning into all programmes; iii) ensuring that there are sufficient teachers and trainers, and that they have both good pedagogical skills and up-to-date technical expertise; and iv) providing adequate quality assurance and monitoring of the labour market outcomes of education and training providers.

  • Providing more support to work-based learning, including i) active and ongoing engagement of employers and unions at all levels; ii) robust, easy-to-understand, competency-based qualifications; iii) more information on costs and benefits; iv) well-aligned policies beyond work-based learning programmes (e.g. employment regulation); and v) well-targeted financial support.

In higher education, the funding system can play a key role in achieving homogeneity in the quality of programmes. Providing sufficient and stable resources to tertiary education, while ensuring equal access and strong outcomes, are central objectives in all countries. To achieve these objectives, direct public transfers to higher education institutions can be linked to their performance using broad performance indicators (Dougherty and Reddy, 2011). In addition, to admit students from disadvantaged backgrounds, performance indicators may be used to account for the characteristics of students. If tuition fees are in place or introduced, they need to be accompanied by measures that remove financial barriers to undertaking higher education in the first place (Johnstone, 2004; Johnstone and Marcucci, 2010).

Degrees and qualifications have to provide clear signals to the labour market about graduates’ skills. With higher education expanding massively, these signals are needed more than ever, but qualifications’ reliability as a guide to graduates’ learning outcomes has been questioned. Attainment does not always translate into skill proficiency: the Survey of Adult Skills found that around 20% of OECD young higher education graduates have numeracy proficiency below level 2 (Figure 4.7).

Figure 4.7. Numeracy skills of young university graduates
Percentage of young graduates scoring at each proficiency level, Tertiary-type A only, 20-34 years-old

Notes: Countries are ranked in ascending order of the percentage of graduates with literacy or numeracy at or below level 2. In Austria, the Czech Republic, Estonia, Finland, Flanders (Belgium), Germany, Japan, Korea, Northern Ireland, the Netherlands, Sweden and the United States, the estimated percentage of graduates performing at level 1 or below on literacy is not different from zero. Adults who obtained their highest qualification outside the host country – those with foreign qualifications and first generation migrants, who obtined their highest qualification prior to entering the host country – are excluded.

Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference for adults is 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders.

Source: OECD calculations based on the Survey of Adult Skills (2012, 2015),; Kuczera, Field and Windisch (2016), Building Skills for All: A Review of England,

Equipping graduates of vocational and university education programmes with reliable, competency-based qualifications requires the engagement of government, employers, unions, students and all other relevant stakeholders (OECD, 2014a). They need to: i) develop qualifications together to reflect labour market needs; ii) provide diversification of qualifications without too much fragmentation; iii) incorporate high-quality assessments into qualifications, including competency-based approaches; and iv) better link competency-based qualifications to the labour market.

Entrepreneurship education and start-up support in higher education institutions


In recent years, higher education institutions have been increasingly supporting entrepreneurship among their students. So far, the emphasis has been on activities aimed at developing attitudes, knowledge and skills that allow students to identify opportunities and turn them into successful ventures. Higher education institutions can also become environments conducive for nascent entrepreneurs. Over the last two decades, some have been introducing key complementary support services such as mentoring, active involvement of students in research activities, co-working spaces and incubation facilities. Institutions have also been helping students to manage intellectual property rights and to access public and private financing. Often, the demand for introducing these services comes directly from students and staff.

According to a recent international student survey on entrepreneurship, entering global markets quickly is a key growth objective for many university-based start-ups.2 The number of international students signing up for entrepreneurship education activities is rising, but far fewer are taking the next step of starting a business (OECD, forthcoming b). In some countries this gap has been noticed first by regional and local governments that want to see an increase in outward-looking and growth-oriented start-ups in their jurisdictions.

Multinational corporations and large established firms often expect to boost their innovation activities by collaborating with start-ups from higher education institutions, particularly in niche areas that require high levels of flexibility and creativity. Identifying start-up teams and their business ideas early is thus a strategic priority when large innovator firms collaborate with higher education institutions.


Governments in most OECD countries see higher education institutions as having a key role in promoting entrepreneurship. Policy support frameworks, however, vary substantially. When a sustainable support system is established, with long-term funding for strategic initiatives, training programmes and higher education networks, institutions are more likely to expand and sustain their entrepreneurship support. In parallel, graduate entrepreneurship education can be included in the performance contracts between education ministries and higher education institutions. To provide quality support for nascent entrepreneurs, it is vital that these institutions rely on partnerships with the business support organisations. Government can encourage and facilitate this co‐operation.

As higher education becomes more international, some countries have introduced support mechanisms to retain students with entrepreneurial intentions. In the Netherlands, for example, a residence permit scheme for start-ups gives ambitious entrepreneurs one year after graduation to launch an innovative business. Mentoring by an experienced entrepreneur or researcher based in the Netherlands is a criterion for eligibility. Many higher education institutions have seized this new opportunity to launch and support international start-ups.

Strengthening skills in small and medium-sized enterprises (SMEs)


Local small and medium-sized enterprises (SMEs) that are connected to inward foreign direct investment (FDI) and to GVCs as domestic suppliers of exporters can benefit from knowledge spillovers and transfer of technologies from multinationals, which are generally more innovating and productive than other firms (Chapter 2; OECD/World Bank Group, 2015). Links to supply chains can enable SMEs not only to enhance their access to international markets but also to improve their skills, for example through direct support from multinationals to meet requirements for quality, efficiency and speed of delivery. Improvements can spread further down the supply chain as first tier suppliers seek improvements in their own supplier base. SMEs’ integration into GVCs is a key condition for the transfer to the whole economy of the benefits of GVCs in terms of knowledge development, productivity and growth.

However, FDI and links to GVCs do not automatically enable SMEs to improve their skills, as investing multinationals may prefer to use skills from their home countries or from third countries. Countries’ skills have an impact on the ability of SMEs to compete and to exploit the opportunities represented by FDIs. Host countries can benefit from a GVC-skills virtuous circle: countries’ skills foster participation in GVCs and attracting FDI contributes to skills development, which results in higher value-added FDI and better skills. SMEs and entrepreneurs largely rely on local labour markets and institutions to find and develop skills. However, local labour markets vary widely in how well they put skills to good use. Many local economies are stuck in a low-skill trap and may lack opportunities to connect their companies to GVCs. Where the demand for skills among employers is low, and people’s skills are not fully utilised, productivity is undermined. Such a situation can also reduce the quality of local jobs in terms of salaries, job security and the possibility for career progression.

SMEs tend to have lower skills than larger firms in manufacturing and services industries, especially for cognitive skills (Figure 4.8). Smaller firms also make lower use of information and communications technologies (ICT), managing and communicating, and STEM skills than larger firms, suggesting that these skills might be less developed for workers in these firms.

Figure 4.8. Skills by industry and firm size
Average across countries, 2012

Source: OECD calculations based on the Survey of Adult Skills (PIAAC) (2012),


Targeted management training for SMEs that are involved in GVCs, or have the potential to establish connections, can help improve their understanding of procedures used by international companies (specifically tendering and associated processes) and ways of approaching international business, while developing some of the skills that companies need to join GVCs.

To enable multinationals to select local suppliers, it is vital that VET systems produce certifiable skills that correspond to international levels and the requirements of foreign investors. Language skills are especially important. In emerging economies and developing countries, the lack of English-speaking SME owners and employees with appropriate management and technical skills often prevents multinationals from involving local small businesses (OECD/World Bank Group, 2015).

Local leaders need infrastructures and networks that help them develop innovative employment and skills strategies to connect companies to GVCs. Partnerships are being used across the OECD to link local leaders to other firms and to multinationals. They generally involve a wide range of partners, including employment and training agencies, local colleges and universities, and economic development agencies (Box 4.2) (OECD, 2014b). Vocational education institutions can play an important role in such local development (OECD, 2014a).

Box 4.2. In the United States and Canada, innovative initiatives connect the private sector with education and training institutions

In the United States, federal economic development policy has fostered “industry clusters” as a way to drive regional economic growth (OECD, 2014b). The Jobs and Innovation Accelerator Challenge supported numerous regional efforts to strengthen and develop regionally based industry clusters, drawing in funding and technical assistance from several agencies.

In 2011, the United States launched the National Network for Manufacturing Innovation, also known as Manufacturing USA, a network of research institutes that support innovation, education and collaboration in manufacturing. These institutes are public-private partnerships, each with a distinct technology focus area, that work towards this common goal. Industry, academia and government partners are taking advantage of existing resources, collaborating and co-investing to nurture manufacturing innovation and accelerate commercialisation.

The institutes aim at connecting people, ideas and technology to solve industry-relevant advanced manufacturing challenges, restore and enhance industrial competitiveness, increase economic growth and strengthen U.S. national security. They focus on transforming promising, early-stage research into proven capabilities that are ready for adoption by US manufacturers. The institutes provide members with access to state-of-the-art facilities and equipment, as well as workforce training and skills development customised to support new technology areas.

Michigan’s InnoState project seeks to promote new product manufacturing capability within existing contract manufacturing firms, enabling them to better connect with GVCs. Working through the Pure Michigan Business Connect Site, the Michigan team provides marketing and other tools for outreach and branding.

The project is comprised of several regional partners, including:

  • the Workforce Intelligence Network, which provides labour market information and serves as a resource hub for local businesses and other stakeholders;

  • the Detroit Regional Chamber, which has several economic development programmes aimed at encouraging local businesses to diversify their customer base;

  • the Michigan Manufacturing Technology Center, which helps manufacturing companies with training, consulting and management mentoring;

  • the National Centre for Manufacturing Sciences, which seeks to drive innovation in manufacturing businesses; and

  • the Southeast Michigan Community Alliance, one of seven Michigan Works Agencies in Southeast Michigan that provide a wide range of workforce development services. Michigan Works Agencies help job seekers look for jobs and develop their employability and technical skills, and help employers recruit and retain qualified employees. The agencies also provide access to funding to train new hires through their On-the-Job Training programmes, and train employees through their Incumbent Worker Training and Customized Training programmes.

In Ontario, Canada, local community colleges and universities help employers to develop product market strategies locally and to better connect to GVCs (OECD, 2014b). Niagara College, for example, has not only geared its curriculum towards meeting local industrial demands in horticulture and wine making, but has also set up an applied research unit that helps local firms to upgrade their products and business strategies. It collaborates with firms in areas such as product and process applied research, engineering design, technology development, product testing, proof of concept, piloting and problem solving.


OECD (forthcoming b), Engaging employers in putting talents to better use in the workplace.

OECD (2014b), Employment and Skills Strategies in the United States,

Closing the gap between education institutions and the private sector


A close match between countries’ skills and industry requirements leads to international comparative advantages in these industries (Chapter 3). Closer links between education institutions and the private sector are needed to achieve such a match and ensure that education and training systems are capable of adapting to changes in the demand for skills.

Fostering co-operation between education and training providers, employers and other stakeholders can enhance the quality and market relevance of VET and higher education programmes. Such co-operation can help individuals develop the strong mixes of skills that are necessary in industries with higher potential for value creation. Many occupation-specific skills, as well as skills needed in business, such as teamwork and communication, can be best learned in the workplace. Work experience can thus signal the demand for different types and combinations of skills, smooth the education to work transition and enhance recruitment.

Governments and education systems can work with companies to understand their skills needs, so that more multinationals rely on domestic skills. This is particularly important for emerging economies, which often try to build whole industries to serve the local operations of a large foreign company. Work-based learning experience in the supplier network of such a company, in the company itself or abroad would offer multiple benefits, such as providing exposure to up-to-date equipment and expertise, and enhancing social and emotional skills necessary to succeed on the job.


Public-private partnerships between higher education institutions, research centres and the business sector can help further close the gap between the education system and the labour market by transforming research into innovation through collaboration (Box 4.2).

Exposure to the business world can link skills development to labour market needs and help companies create value. However, work-based learning is still under-utilised in many countries – even for traditional VET programmes – and a range of obstacles stand in the way of its broader use. A work-based learning component could be integrated into different levels and types of education and could be compulsory for VET programmes. Introducing such a requirement could streamline many programmes, as those which are of little interest to employers may not be able to fulfil the requirement. In some cases, financial support might be needed.

Making the best use of the skills pool and anticipating changing skills needs

The interaction between a country’s skills characteristics and industries’ requirements contribute to countries’ performance in GVCs in many ways: the allocation of skills to industries, the match between workers’ skills and the tasks performed on the job, the way skills are used on the job, and the capacity of education and training systems to adjust to changing needs.

On average, around one-quarter of workers report a mismatch between their existing skills and those required for their job, with significant differences between countries (OECD, 2013b). Being overskilled for the job is more widespread than being underskilled. A skills mismatch is associated with lower labour productivity as it reflects less efficient resource allocation (Adalet McGowan and Andrews, 2015). However, a pool of workers with skills that are not fully used can offer a potential for firms to introduce new technologies or working methods, and to become more productive in the long run.

Management policies can make the best use of skills


Management policies, which make a key contribution to performance and productivity in general, are a powerful tool for making the best use of skills assets, adjusting them to new needs, and thereby giving a country a comparative advantage in GVCs (Box 4.3).

Box 4.3. How management policies can give countries a comparative advantage in global value chains

A study in Denmark points to organisational performance as a source of economic and trade performance (Danish Agency for Science Technology and Innovation, 2013). The study compares “hidden champions” – firms in the manufacturing sector that are export leaders in their respective export markets – with other firms in the same sectors (core firms).

One difference regards formal education qualifications: the export leaders were found to have a higher share than core firms of workers with a general tertiary qualification or a tertiary vocational qualification. The major difference, however, comes from work organisation practices. Of the hidden champions, 58% have work organisation practices characterised by employee involvement to a very great extent or a great extent, against 37% for core firms. Almost all of the hidden champions have a management philosophy based on employee involvement and the delegation of responsibilities.

In India, an experiment was conducted on large multi-plant textile firms to assess the impact of management practices on firms’ productivity (Bloom et al., 2013). A group of firms (the treatment group, randomly chosen) received five months of extensive management consulting from a large international consulting firm. In the first month, the consulting firm identified opportunities for improvement in a set of management practices and then supported firms in the following four months in the implementation of recommendations. Another group of firms (the control group) only received the one month of diagnostic consulting.

The development of good management practices increased productivity by 17% in the five following years. The firms having received the treatment grew faster, with evidence suggesting that better management allowed them to delegate more and open more production plants in the three years following the start of the experiment. Firms had not previously adopted these methods because they were unaware of the benefits of practices such as daily factory meetings, standardised operating procedures and inventory control norms.

Source: Bloom, N. et al. (2013), “Does management matter? Evidence from India”, Quarterly Journal of Economics, Vol. 128/1.

Danish Agency for Science Technology and Innovation (2013), The Hidden Champions – Danish Industrial Export Successes, Copenhagen.

Global value chains increase the development of culturally diverse teams, making it even more important for firms to develop good management policies. Cultural diversity enhances productivity and innovation by increasing the level of relevant knowledge and the ability to solve problems, thanks to the diversity of experiences and perspectives (Horwitz and Horwitz, 2007). However, business performance is higher for teams with higher ethnic diversity only up to a certain threshold, beyond which the costs of communication, co‐ordination and cohesion become too high (Hoogendoorn and van Praag, 2014). Good management policies can push forward this threshold.

The World Management Survey gives information on the management practices used by firms in a large range of countries (Bloom and Van Reenen, 2007). Best management practices are defined as those that organise and explain the production process, monitor and analyse individual performance, set challenging and interlinked short and long-run targets, reward high performers develop workers’ skills and retain talents.

The use and development of management practices by firms varies widely between countries, with the United States, Germany, Japan, Sweden and Canada making the best use of them and several emerging economies using them much less (Figure 4.9). In the top-performing countries, dispersions across firms in the use of best management practices are not large. Where such differences are large, this suggests there is scope to expand best management practices to firms lagging behind in their use. Countries with the lowest average management practices scores are also those with the largest variation across firms in the use of these practices.

Figure 4.9. The use of best management practices by countries

Note: Mean and coefficient of variation of the management practices scores across firms. A higher score (mean) indicates better management practices. A higher coefficient of variation indicates a larger dispersion across firms in the scores of management practices.

Source: OECD calculations based on Bloom et al. (2012), “Management practices across firms and countries”, NBER Working Papers, No. 17850.


Various types of policies affect the use of management practices. The level of education of both managers and non-managers is strongly linked to better management practices (Bloom et al., 2012). Hence, policies to develop skills in general but also entrepreneurship and management skills can help to diffuse best management practices. Flexible employment protection legislation generally makes it easier to implement best management practices. Policies not related to skills also play an important role. Those that impede product market competition, including tariffs, are linked with poor management practices, as are policies that favour family ownership of firms, such as tax incentives for these types of firms.

These findings illustrate how achieving strong performance in GVCs requires avoiding misalignments between policies. Education polices to develop a set of strong skills could stimulate the adoption of good management practices, which could in turn be a source of trade comparative advantage. But this investment would be partly lost if other measures prevented competition, possibly with the objective of favouring domestic firms, as they would create an environment that in the end does not stimulate the adoption of best management practices.

In emerging and developing economies, growth in the industrial sector is often regarded as a critical step for broad-scale development and long-term poverty reduction. However, these economies could enhance the transformative power of these labour market opportunities by guaranteeing sufficient job quality. In some of these countries, low wages combined with health and safety hazards in industrial jobs deter skilled workers, who prefer to turn to self-employment in the informal sector (Blattman and Dercon, 2016).

Multinational firms could play a key role in improving working conditions and employment patterns, especially in developing and emerging economies where social regulations are often less developed. Raising the skills of domestic workers and developing knowledge about best management practices can lead to virtuous cycles in which multinationals improve job quality in the country. This requires co-operation between multinationals, their governments and emerging countries. One example is the Bangladesh Accord on Fire and Building Safety, chaired and facilitated by the International Labour Organisation, a unique agreement between lead firms and trade unions to promote respect for labour rights in the Bangladeshi clothing industry.

Employment protection legislation can ease reallocations and adaptation


Employment protection legislation affects countries’ capacity to perform in GVCs and adapt to their consequences. Flexible legislation can ease structural adjustment and enhance countries’ capacity to change their positioning in GVCs. Overly rigid labour market policies can prevent firms from adopting risky technologies and realising high productivity gains because such policies make it more costly to shed workers (Bartelsman, Pieter and De Wind, 2010).

At the same time, employment protection raises workers’ incentives to acquire firm-specific skills, which can be a source of trade comparative advantage (Tang, 2012). Countries with more protective labour laws tend to export more in firm-specific, skill-intensive sectors. Legislation must balance employers’ need for flexibility in a competitive environment with workers’ needs for possibilities and incentives to develop firm-specific skills.

Similarly, rigid legislation might reduce or delay the risk of unemployment of workers exposed to offshoring but be an obstacle to finding a new job. Overall, there is a broad agreement that employment protection legislation must balance workers’ needs for income, skills development and job security with employers’ need to adjust the workforce in an increasingly changing world of work (ILO, 2015; OECD, 2014c).

In recent decades, non-standard employment – which does not benefit from the same degree of protection against contract termination that permanent employees enjoy – has developed, even if permanent employment remains the most prevalent form of contract (OECD, 2014c). Globalisation and technological change, which increase the need for flexibility, are considered the main reasons for the development of non-regular employment (Eurofound, 2010; ILO, 2015). Non-standard employment includes fixed-term contracts, hiring through temporary employment agencies, and part-time work, but also more atypical contracts such as very short fixed-term and part-time work contracts, freelance contracts, non-contract work, zero-hours contracts and on-call work.

There is no international data on atypical contracts. Data covering major high-income countries show that the share of workers under a permanent contract has declined, while a category has emerged of employees engaged without a contract, which stood at 1.1 per cent in 2012, alongside a modest increase in the share of own-account workers (ILO, 2015). These trends point to a decline in employment security across the workforce in high-income economies between the pre- and post-crisis periods. Zero-hours contracts, which do not offer any guarantee to the employee about how much work is entailed, have developed in some countries. In the United Kingdom they have reached 2.9% of total employment (UK Office for National Statistics, 2016). However, they are mainly found in sectors not exposed to GVCs, such as accommodation, food, health and social work.

As the group of workers under non-standard forms of employment has become more heterogeneous, it is difficult to grasp the overall status of employment protection for these workers. Temporary and part-time contracts do not give access to the same level of rights and protections as standard contracts, but these rights and protections have increased over time and are well established. Nonetheless, pay increases, participation in the representative process, unemployment subsidies, pension rights and vocational training systems are linked to job tenure (ILO, 2015). Most reforms of employment protection legislation in developed and EU countries between 2008 and 2014 have consisted in increasing regulations on atypical forms of work so as to rebalance them with regulations on standard forms of work (Adascalitei and Pignatti Morano, 2016).


Participation and performance in GVCs requires employment protection legislation that provides flexibility to firms and security to workers. This can be achieved by attaching employment protection and social insurance to workers rather than to jobs, and by delinking these entitlements from job tenure, either by linking them to work experience or making them independent from these criteria. Some countries are considering the introduction of a single contract and of a guaranteed minimum income.

A better understanding of the incidence of atypical forms of work, and of social protection entitlements of workers under atypical forms of work, would help countries to design better employment protection legislation and ensure that duality, with a group of workers facing the cost of offshoring, is not increasing. Workers in all types of employment need adequate protection. In several countries, atypical forms of work are not the main employment but a way to get additional income (Eurofound, 2010).3 It is important to better understand this issue to allocate the funding of social protection correspondingly to various employers.

Non-compete clauses that do not hinder skills mobility


Non-compete clauses, under which employees agree not to use information learned during employment in subsequent jobs for a set period of time, are intended to protect employers’ intangible investment. By restricting workers’ mobility, they enable employers to benefit from their investment in training and R&D. At the same time, they limit workers’ opportunities to find new jobs after voluntary or involuntary job termination. They can be an obstacle to structural adjustments, and can be used to prevent competition. Non-compete clauses particularly affect employees with technology-specific skills, who have difficulties finding new jobs beyond firms that are direct competitors (Garber, 2013). These workers are less likely to change jobs when subject to non-compete clauses (Marx, Strumsky, and Fleming, 2009). Non-compete clauses can also limit the diffusion of knowledge and hinder innovation (Samila and Sorenson, 2009).

Differences in the legislation and enforcement of non-compete clauses may influence countries’ capacity to attract FDI and firms’ choice of location for offshoring (Garber, 2013). India differs from China and Brazil, for example, in that there is no enforcement of non-compete clauses. This could be a barrier to foreign investment from firms with high-value trade secrets. Nonetheless, India has attracted significant foreign R&D investment; multinationals might benefit from the lack of enforcement of non-compete clauses, as it allows them to tap into a broader set of workers with firm-specific knowledge, and other factors also matter in offshoring decisions.


It is important to have a better understanding of the incidence of non-compete clauses in OECD countries to assess more thoroughly their impact. There is no broad assessment of the prevalence of non-compete clauses. A study in the United States found that 38% of the workers had been submitted to non-compete clauses during their working experience and that 18% were under such an agreement in 2014 (Figure 4.10; Starr, Bishara and Prescott, 2016).

Figure 4.10. Workers under a non-compete clause in the United States, by characteristics
Percentage, 2014

Note: Total numbers are expressed as a percentage of the US labour force. Other numbers are expressed as a percentage of workers in the respective group.

Source: Starr, Bishara and Prescott (2016), “Noncompetes in the U.S. labor force”,

Policies can anticipate changes in skills needs


Policy intervention can help to address skills mismatches and skill shortages in the labour market. Doing so successfully, however, relies on having good information on current and future skills needs. Systems and tools for assessing and anticipating skills needs exist in all OECD countries, but approaches vary significantly in terms of how skills are defined, the time span, frequency, methodology and scope (OECD, 2016b).

In terms of methodology, most exercises rely on more than one source of information, thus reducing potential biases. In most cases these are quantitative sources of information (e.g. employer surveys, surveys of workers or graduates, quantitative forecasting models, sector studies and labour market information) but a few countries systematically incorporate qualitative information as well. Many countries carry out multiple exercises. For instance, Canada combines analyses of existing skill shortages with medium to long-run forecasts of skill needs. This enables the government to tailor immediate policy interventions, such as those targeted at new immigrants with needed skills or short-term training schemes for displaced workers, while also guiding longer-term policy development, for instance building new course curricula or apprenticeships (Box 4.4).

Box 4.4. How countries use skills needs assessment to steer policy

Governments of OECD countries use skills assessment and anticipation exercises to guide policy development in employment, education and migration. In terms of employment policy, skills needs information is commonly used to update occupational standards, and to design apprenticeship, retraining and on-the-job training programmes. For example, in Australia, Belgium and New Zealand, skills needs information feeds into the National Occupation Standards, to facilitate the rapid development of standards in new occupations or in occupations with changing skill requirements. Turkey uses skills exercises to design apprenticeships in occupations and industries where shortages are identified.

In France, Hungary, Ireland and Italy, skills needs information is used to help in the transition to a greener and digital economy.

Skills needs information is also commonly used in education policy to inform curriculum development and set the number of student places at all levels of education, including technical and vocational education and training programmes. It is also made available to prospective students and career guidance counsellors to inform decisions about which education level and field of study to pursue. For instance, Finland recently launched a web-based tool, ForeAmmatti, which allows users to browse information on vacancies, regional supply and demand of labour, and skills needed for particular occupations.

In migration policy, skills needs information is used to fast-track entry of immigrants with skills that are in high demand. For example, in Australia, job vacancy data and interviews with employers inform lists of current and future skills shortages, which facilitate migration of workers with relevant skills. Similarly, the United Kingdom’s Migration Advisory Committee uses general labour market information to identify occupations with shortages and to advise the government on immediate skills needs.

Source: OECD (2016b), Getting Skills Right: Assessing and Anticipating Changing Skill Needs,


To be successful, skills assessment and anticipation exercises must be well aligned with potential policy uses. For instance, skills needs should be defined so that they map easily to policy-making variables, like field of study or occupation. To be accessible and useful to policy makers, output should not be too technical, and should be sufficiently disaggregated at the regional, sub-regional or industry levels.

The engagement of key stakeholders in the design of skills assessment and anticipation exercises can ensure that they understand the outputs and use them for policy making. For instance, the strength of Norway’s skills assessment and anticipation system lies in the joint involvement of employment and education authorities in the design and development of forecasts carried out by Statistics Norway. This type of stakeholder engagement requires strong co-ordination. A variety of mechanisms have proven successful in promoting consensus among stakeholders, including working groups (e.g. the inter-ministerial skills working groups in the United States), round tables (e.g. the round tables in the Netherlands enhance collaboration among regional/sub-regional administrative levels), sector skills councils (e.g. in Canada, the Czech Republic and the United Kingdom) and independent national skills advisory groups (e.g. in Denmark, Finland and Germany).

Finally, it is generally seen as good practice to adopt a holistic approach to the assessment of skills needs that incorporates quantitative and qualitative information. For instance, the United Kingdom’s sector-specific holistic approach to forecasting relies on econometric models, surveys of employers’ opinions, skills audits, Delphi methods,4 case studies, focus groups, scenario development, and consultation with experts and employers (CEDEFOP, 2008; UKCES, 2010).

International co-operation for education, training and innovation

Investment, education, research and innovation have globalised along with the production process. Rather than embark in a race for talent and investment, countries can co‐operate on policies that affect the capacity to innovate, the mobility of students and workers and the funding of education (Box 4.5 and Figure 4.11). This co-operation is an important step to ensure that GVCs are not delinked from the local context and that governments can influence the impact of GVCs on their countries (Taglioni and Winkler, 2016).

Box 4.5. Capturing countries’ potential to be part of global education, innovation and research networks

The OECD has developed several indicators to capture various aspects of countries’ connections to global education, innovation and research networks (OECD, 2014d; OECD, 2015c). It is possible to summarise this information to get a single indicator that captures three dimensions of co-operation, each of them measured by a number of indicators:

1) International co-operation for research and innovation, measured by international co‐authorship, international co-inventions, and the international mobility of scientific authors.

2) Foreign/international students and high-skilled workers, measured by the share of international and foreign students enrolled in tertiary education and the share of foreign-born doctorate holders.

3) Funding incentives for international co-operation, measured by government expenditure on R&D, the share of government R&D expenditure financed from abroad, and the share of business R&D expenditure financed from abroad. The overall budget for R&D, as a signal of countries’ overall support for R&D activities, can trigger investment from abroad or attract multinationals. The share of this investment coming from abroad, either from the foreign private sector or from foreign governments, reflects a country’s capacity to attract R&D investment from abroad. Receiving funding from abroad encourages research organisations to work with other countries.

All indicators are normalised and a summary indicator is calculated for each dimension as the average of the normalised indicators (Figure 4.11). The indicator reflects the extent to which countries have framework conditions, policies and characteristics that can foster participation in global education, innovation and research networks.

Source: OECD Patent Database, OECD Main Science, Technology and Industry Indicators Database, OECD Education at a Glance Database; OECD (2015c), OECD Science, Technology and Industry Scoreboard 2015: Innovation for growth and society,

Figure 4.11. Countries’ potential to be part of global education, innovation and research network, a synthetic indicator

Source: OECD calculations based on the OECD Patent Database,; OECD Main Science, Technology and Industry Indicators Database,; OECD Education at a Glance Database,; OECD (2015b), OECD Science, Technology and Industry Scoreboard 2015: Innovation for growth and society.

Participating in global education, research and innovation networks


Innovation activities have become increasingly internationalised. Although the majority of R&D investments are still concentrated in companies’ home country, near headquarters, companies have also started to offshore R&D activities (de Backer and Destefano, forthcoming). Offshoring of R&D was initially intended to adapt products and processes to local market demands but companies then sought to use it to tap into foreign knowledge, technology and human capital.

In addition, co-operation for innovation has developed, through collaborative arrangements with external partners and suppliers, as can be seen from the rise in patents with co-inventors from different countries. Firms, universities, research institutions and government agencies are connected in global innovation and research networks.

Countries have to belong to global innovation networks to ensure that their investments in skills pay off within the context of GVCs. Having a pool of good researchers in a specific area may not be enough; their research needs to be relevant to the international market. Although innovation co-operation between countries, measured by patents involving inventors from different countries, is less developed than trade with GVCs, there is a significant geographic overlap between global innovation networks and GVCs (de Backer and Destefano, forthcoming).

The links between GVCs and global innovation networks are reflected in the leading role of multinationals in patenting. More than 60% of all patent applications and two-thirds of co‐inventions are related to multinationals’ activities (Figure 4.12). Of co-inventions, more than 50% concern co-inventors in different countries but with the same multinational as applicant (i.e. headquarters and/or affiliates).

Figure 4.12. The importance of multinational enterprises (MNEs) in patenting and international co-invention

Note: The left-hand chart shows the shares of patent applications that can be allocated to one multinational (same MNE), several multinationals (multi-MNE), multinationals and non-multinationals (MNE-non-MNE), non-multinationals (non-MNE) and those for which no information is available. The right-hand chart measures international co-invention by focusing on patents with multiple inventors who reside in different countries. It shows the share of co-inventions that can be attributed to same multinational (headquarters and/or affiliates), several multinationals, multinationals and non-multinationals, non-multinationals (non-MNE) and those for which no information is available.

Source: de Backer and Destefano (forthcoming), “The links between global value chains and global innovation network: An exploration”, OECD Science, Technology and Industry Working Papers.

Collaboration on scientific research among institutions in different countries, for instance through international co-authorship, improves networking and enhances the likelihood that research will be recognised or adopted by the private sector (OECD, 2015c). Scientific collaboration can enable small countries to reach higher value-added segments of GVCs and attract foreign direct investment. Estimates suggest a positive relationship between measures of scientific research collaboration and citation impact, especially for in countries with lower levels of scientific production (OECD, 2015c). Knowledge may also spread more rapidly when interpersonal links through joint work and co-operation create opportunities for learning that go beyond the exchange of codified information (OECD, 2016c).

Global education networks are just as vital as innovation and research links. Competition to attract international students is rising. Some countries have less success in attracting international students and researchers, which hinders their competitiveness in the search for talent and the economic impact of their higher education systems. It also diminishes the exposure of home students to international students, and thus their capacity to operate in global environments later on. Students in higher education who study and spend time in a foreign tertiary institution build links with others and gain experience that will be beneficial for their professional careers. In a group of countries, more than 30% of students at a doctoral level come from abroad (Figure 4.13).

Figure 4.13. Share of international or foreign students by level of tertiary education

Source: OECD (2016d), Education at a Glance 2016: OECD Indicators,


Countries increasingly compete to attract international investment through policies outside the skills domain (OECD, 2014d). These policies, including subsidies and tax breaks, can be costly and do not ensure that spillovers from multinationals to SMEs will occur. Even if they succeed in attracting investment, the benefits might be volatile and transitory as such incentives generally divert investments from one country to another within a geographic region.

Policies to develop strong skills through high-quality education and training systems also help to foster international co-operation for innovation and research. High-quality education systems, especially at the tertiary level, attract students from abroad who might be more involved in research co-operation. Specific policies to foster interactions between the private sector, universities and research institutions, such as clusters and hubs (Box 4.2), can facilitate the transfer of knowledge but this requires that members have the skills to benefit from these interactions, develop real exchanges of information, and work together.

In addition, some specific policies can foster the internationalisation of education and research systems and encourage research co-operation. The provision of education and training programmes in English can help to attract students from abroad and enable other countries to compete with English-speaking countries. They are unevenly developed in EU countries (Figure 4.14).

Figure 4.14. Proportion of tertiary education programmes provided in English

Note: At ISCED levels 5 or 6.

Source: Wächter and Maiworm (2014), “English-taught programmes in European higher education: The state of play in 2014”.

Other policies in tertiary education that can ease students’ mobility include adopting common standards in higher education, recognising foreign diplomas and skills acquired abroad, and enabling students to transfer credits acquired abroad. In Europe, as part of the Bologna process, countries have made efforts to harmonise their qualification frameworks and to make them more transparent. Most of the OECD EU countries have fulfilled all the steps of the European qualification framework (European Commission/EACEA/Eurydice, 2015a). Nevertheless, in most EU countries, higher education institutions make the final decision on recognition of foreign qualifications, while recognition of credits gained abroad is fully in the hands of higher education institutions. Improving international recognition of skills acquired abroad would help students move abroad at different points in their academic development and thereby acquire a diverse set of experiences and connections.

Funding skills development when the social returns are increasingly global


Countries benefit in several ways from the internationalisation of education. Some countries attract a large share of international students, enlarging the domestic pool. International students are attractive for OECD host countries beyond the economic benefits of their skills and ideas. In 2011, half of the OECD countries for which the data are available charged higher fees for international students (OECD, 2016d).

Some developing and emerging economies see a large share of their most talented students leave to study abroad, where they can benefit from high-quality programmes. If these students do not return, their countries lose part of their initial education investment. Nonetheless, the possibility of studying abroad may increase people’s incentive to invest in education at home, as it opens a prospect of higher expected return on education. More people may decide to enrol in tertiary education, even though not all of them will pursue education abroad, increasing the supply of tertiary educated people in the origin country.

This “return on migration” for the origin country is possible if the probability of emigration for students with tertiary education is sufficiently low. However, the incentive to study abroad could encourage students to choose a field of study that responds to migration prospects. When foreign and domestic needs for skills differ, the mismatch between the supply and demand for skills in the origin country can become large and induce what is called “brain waste” (Docquier and Rapoport, 2012).

The development of GVCs benefits from and contributes to the internationalisation of tertiary education. Offshoring activities shift the skills demand across industries and across countries (Chapter 2), altering the efficiency of educational investment. In the country where multinationals are located, three main changes are likely to happen. First, the offshoring of activities means that some investment in education, typically in specific vocational skills, is lost if the corresponding jobs are offshored. Second, education and training policies may be needed to help the workers who lose their jobs to develop skills that will help them to move to another industry. Third, the increase in demand for skills induced by the new distribution of production – usually skills associated with more complex tasks – may raise demand for other or new education programmes. In contrast, in countries to which activities are offshored, the need for some education programmes increases, but these countries may not have the capacity to develop such programmes.


Countries aim at gaining comparative advantages through the skills of their population but GVCs make the production process less dependent on local skills. In this context, the social returns of education to countries are more uncertain while the private returns of education may increase. This uncertainty can affect the governments’ education finance strategies as well as people’s educational decisions. Countries can co-operate to design arrangements that benefit both sides.

An agreement to jointly design and finance education programmes needs to build on mutual recognition of the distribution of benefits and costs arising from the internationalisation of tertiary education and of the production process. First, the financial resources collected by universities in OECD member countries from international students are large whereas the cost of education in non-OECD countries is usually lower. Second, the internationalisation of tertiary education and of the production process generates opportunities in developing and emerging economies but can also lead to costs, especially if skills in these countries are downgraded. Third, the productivity gains of offshoring can be higher for offshoring countries if they can rely on a skilled labour force in countries to which activities are offshored.

Co-operation arrangements can take various forms. They can involve discussions between governments and firms from various countries on the specific skills needs implied by offshoring, and how they can be met. They can also entail more formal agreements in which the costs of education programmes are shared and offshoring countries are involved in designing education programmes in countries to which activities are offshored. Some studies have proposed to create a Global Skill Partnership under which two countries can agree on sharing the costs and benefits of creating skills for both countries’ needs, while preserving workers’ freedom of movement (Clemens, 2015).

A formal agreement on co-operation for education has to build on a partnership between different stakeholders that have complementary roles in GVCs. Partnership between public and private entities is vital because each can improve the quality of training. A dialogue between private-sector partners and public institutions is also important to guarantee that the educational programmes are portable. Moreover, private-sector partners play an important role in identifying the skills mix they need whereas public institutions (in particular in countries to which activities will be offshored) should ensure that students are not tied to the employers. Such an agreement should guarantee successful placement for students in the global labour market. There are several examples of international co-operation in VET, for instance the German-Thai Dual Excellence Education programme (European Commission, 2015).

Education institutions can learn from each other, enhancing the quality of education in both institutions. Being involved in education programmes in countries to which activities are offshored can be a way to maintain knowhow in this type of programme, especially for VET programmes. In addition, some activities and skills may be offshored today but onshored – brought back to the domestic market – tomorrow. And if know-how in VET programmes is retained, it is easier to fulfil needs for related education programmes that may emerge in the future.

For emerging and developing economies, the loss of talented students can be partly offset by remittances, especially if they flow towards education. The country of origin could inform students about opportunities, once they are working, to send money home. Filipino migrants have been found to be more willing to remit to beneficiaries in their origin country when the money is “labelled” to be used for educational expenses (De Arcangelis et al., 2014). The country of origin could also foster intangible remittances, such as ideas and knowledge, by maintaining strong ties with students abroad.

Aligning migration policies with international competitiveness objectives


Migrants are increasing the supply of skills in many destination countries. The number of migrants with tertiary education living in OECD countries almost doubled between 2000 and 2010, an increase much larger than that in the native-born population (Arslan et al., 2016).

People born in the same country who each move to different countries develop networks that help to spread ideas, knowledge and technology. Networks between migrants from the same country of birth, in particular skilled migrants, stimulate trade and foreign direct investment by removing informational and cultural barriers (Javorcik et al., 2011; Foley and Kerr, 2013). Migrants also increase cultural diversity in firms, which can boost productivity (Alesina, Harnoss and Rapoport, 2016; Ottaviano and Peri, 2005).

Migrants are more entrepreneurial than native-born citizens so they can boost innovation, spurring economic growth. Immigrant groups are more engaged in self-employed enterprises than natives in many OECD countries, including Australia, Canada, the United Kingdom and the United States (Kerr and Kerr, 2016). This trend has strengthened over time. In the United States, immigrants’ share of entrepreneurship rose from 17% in 1995 to 27% in 2008 (Fairlie and Lofstrom, 2013).5 Migrant entrepreneurship is often driven by a lack of other employment opportunities, particularly for low-skilled migrants, but also by migrants’ social networks, which provide information, financial support and a customer base (Kerr and Mandorff, 2016). Thanks to their knowledge of global markets, migrant entrepreneurs orient their sales to international customers. A study of 7 600 firms in London showed that companies with foreign-born owners (including those with some UK-born owners) are more likely to introduce new products and services and to sell to the international market than firms with only UK-born owners (Nathan and Lee, 2013).


It is important to avoid misalignment between migration policies and international competitiveness objectives. In recent years, some OECD countries have revised their policy frameworks for skilled workers and international students (OECD, 2016e). In particular, there has been a clear trend to make it easier for international students to remain in the country after graduation. Labour migration schemes have been adjusted, generally to favour skilled workers. Changes have gone in various directions for less skilled labour.

Facing the risks and implications of offshoring

Participation in GVCs includes the offshoring of activities, which lowers the demand for some skills. GVCs also increase the interconnections between countries and thereby the uncertainty surrounding the demand for skills. Skills policy changes in a country affect its skills endowment and, in turn, its comparative advantages in GVCs but also those of its trading partners. Overall, the development of GVCs leads to structural changes that can be costly for some groups of workers, especially in the short term. Well-designed policies have a vital role to play in reducing these costs.

Finding a balance between preventive and curative policies to support workers at risk of displacement


Some workers face a greater risk than others that their jobs will be shifted offshore, either because of the type of tasks they perform on the job (Chapter 2) or because of workers’ own characteristics.

Older workers and those with low education levels face a higher risk of displacement, take longer to get back into work and suffer greater and more persistent earnings losses (OECD, 2013c). While young people also face a higher risk of displacement than prime-aged workers, they generally find work quickly after displacement, often in jobs with greater skills requirements than their previous jobs.

Women are generally no more likely to be displaced than men, once other factors are taken into account, such as the type of contract they hold before displacement. However, women are more likely than men to become disconnected from the labour market and experience longer spells of inactivity after displacement.

Very often workers’ characteristics and the type of task they perform on the job are mutually reinforcing factors: low-skilled older workers perform routine intensive tasks that can be offshored and face difficulties finding a new job because of their low skills.

Displaced workers tend to have longer unemployment spells and higher earning losses than the other unemployed workers. The duration of the subsequent unemployment spell increases with job tenure, because workers with high tenure have a higher level of specific human capital investment in their firms and in their industry or occupational sectors (OECD, 2013c; Cavaco, Fougère and Pouget, 2009).


Well-designed policies can help displaced workers find a new job but it is also crucial to provide organised learning opportunities, beyond initial formal education, to help vulnerable workers to adapt and change job if needed, as the social cost of job losses for these workers can be high. Both types of policies are discussed in this section.

Most OECD countries have policies to support displaced workers and some have been beneficial. In France, several programmes that target displaced workers have been gradually introduced. The initial programme increased the likelihood of finding a job with a permanent contract (Cavaco, Fougère and Pouget, 2009).6 The programme provided immediate, individual support to displaced workers for six to eight months, beginning just after dismissal, including retraining, often vocational training, and help in looking for a new job. In the United States, the focus has been on limiting the risk that displaced workers experience a loss in earnings, as some of their skills are not transferable to emerging industries. The government has allocated grants to community colleges to train displaced workers for in-demand jobs. These programmes have increased re-employment earnings of participants (Jacobson, Lalonde and Sullivan, 2005).

Labour market policies can help displaced workers find a new job if they are well designed. Some important elements include an obligation to look for a job, access to high-quality job search counselling, and strong and modern public employment services. When specific programmes are proposed, they should include some work experience and labour market training (OECD, 2015d).

Workers who are displaced because their industry or occupation is exposed to offshoring need to develop new skills to raise their employability. In some cases, short training programmes can be sufficient to improve workers’ skills. In other cases, workers may need to go back to education, such as upper secondary or post-secondary VET. It is crucial to provide organised learning opportunities for adults beyond initial formal education. Workers in high-technology sectors need to keep pace with rapidly changing techniques. Workers in low-technology industries and those performing low-skilled tasks must learn to be adaptable.

VET programmes can better develop a broad range of skills – including foundation literacy and numeracy skills, and vocational skills – than short-term labour market programmes, so they can be an important policy for some groups of displaced workers. In most EU countries, upper secondary VET is available for adult learners, either through the general VET system or through programmes dedicated to them (European Commission/EACEA/Eurydice, 2015b). Several countries have also a specific upper secondary programme for youth who have left school early, such as the second chance schools in France (OECD, 2015a).

Overcoming obstacles to adult education and training


In most countries, workers whose skills levels are already high participate the most in adult education (OECD, 2013b). Participation in adult learning increases with the use of reading activities in everyday life (Figure 4.15). Those who use their reading skills a lot in everyday life are 2.5 times more likely to participate in adult learning than those with low use of reading skills. In addition, participation in adult learning is predominantly job-related – driven by motivation to improve career prospects and get or change a job – and as such is closely linked to employment status, among other factors. Participation rates for the employed are higher than for the unemployed and for adults who are not in the labour force (Figure 4.16).

Figure 4.15. Adult participation in education and training by frequency of use of reading skills in everyday life
Percentage of adults, 25-64 year-olds, 2012 or 2015

Note: Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders. Highest frequency refers to reading daily or weekly and the lowest frequency refers to no reading, or reading rarely or less than once a month.

Source: OECD (2016d), Education at a Glance 2016: OECD Indicators, Table C6.1,

Figure 4.16. Adult participation in education and training by employment status
Percentage of adults, 25-64 year-olds, 2012 or 2015

Note: Chile, Greece, Israel, New Zealand, Slovenia and Turkey: Year of reference 2015. All other countries: Year of reference 2012. Data for Belgium refer only to Flanders and data for the United Kingdom refer to England and Northern Ireland jointly.

Source: OECD calculations based on the Survey of Adult Skills (2012, 2015),; OECD (2014e), Education at a Glance 2014: OECD Indicators,

There is a virtuous cycle for high-skilled workers: they have the skills to learn more and by learning more, they raise their skills. In contrast, low-skilled workers or the non-employed face a vicious cycle. They are trapped in situations in which they do not benefit from training and therefore their skills remain weak or may even deteriorate.


Policies to support workers at risk of displacement have to break the vicious cycle that keeps low-skilled workers out of adult learning. Data for EU countries show that work and family responsibilities are important barriers to adult participation in lifelong learning (European Commission/EACEA/Eurydice, 2015b). Distance learning programmes can offer solutions to those with work and family responsibilities. Another important step is to develop agreement between employers and unions, as is done in the Nordic countries, to offer more flexibility for undertaking lifelong learning on the side of both employers and adult education and training providers.

To facilitate employment and career transitions, it is also crucial to offer opportunities to start or continue higher education later in life, which requires adapting the requirements to enter higher education. The validation of non-formal and formal learning is vital, not only to ease the access of adult learners to higher education but also to ensure that they have the skills needed to succeed in their learning programme.

Financing is often an impediment to adult education and training. Investing in skills involves a range of costs, including not only lost earnings and tuition fees but also potentially higher future taxes. These costs need to be balanced by higher future wages, tax credits to undertake education, and education grants.

An OECD study has formulated several indicators that measure the impact of tax and spending policy on people’s incentives to invest in skills (OECD, 2017). In particular, the Breakeven Earnings Increment indicator measures how much earnings need to increase so that people earn back the costs of an investment in skills over their remaining years in the workforce. A 50-year-old worker reskilling in later life through a one-year course of study to change career needs to achieve much higher earnings gains than a 27-year-old taking a similar one-year course of education or a 32-year-old undertaking a short course of job-related training (Figure 4.17). This is because training to change career is less frequently eligible for tax deductions than job-related training and because a 50-year-old worker has fewer years in which to recoup the cost of education.

Figure 4.17. Financial incentives to invest in job-related education and lifelong learning

Note: Data on job-related education are for a 32-year-old single taxpayer with no children, who undertakes a short course of job-related education, earning 95% of the average wage over the year while they study. Data on lifelong learning are for a 50-year-old single taxpayer with no children, who undertakes a one-year course of non-job-related education, earning 25% of the average wage during schooling. This figure shows results that incorporate tax deductions and tax credits for direct costs, tax exemptions for scholarship income, and reduced taxes on student wage income. Tax incentives in the personal income tax system are incorporated, but not the social security contribution system. They do not incorporate skills tax expenditures that subsidise parental spending on education or that subsidise firm spending on education. It is assumed that the skills investment is financed wholly with savings: students do not incur any debt to make a skill investment.

Source: OECD (2017), Taxation and Skills,

As governments provide many tax measures to support investment in skills – such as tax deductions of skills expenses, or tax exemptions for scholarship income – good design of these provisions is vital in ensuring their effectiveness (OECD, 2017). Existing skills tax expenditures are often only available for training connected to a workers’ current employment, and may be ineffective in assisting workers who need or want to change careers. These tax provisions may reduce labour market flexibility and exacerbate skills mismatches. Skills tax expenditures often provide larger benefits to those with larger taxable incomes, and may provide more benefits to those in secure employment than to those in casual employment. Ensuring access to skills investment for those who have difficulty borrowing is crucial. Income-contingent loans may be an efficient and equitable policy instrument to achieve this aim.


Countries can shape their performance in GVCs through effective and well-co-ordinated skills policies. The risks of misalignment between skills policies and international competitiveness objectives are large, however: GVCs and trade issues pertain to ministries with their own set of policies outside the skills area, while education, research and labour ministries in charge of most skills policies generally focus on employment and innovation outcomes. To make the most of GVCs, a “whole of government” approach is needed in which stakeholders collaborate to take into account the country’s current positioning in GVCs, the strengths and weaknesses of skills policies, other types of policies that affect the country’s performance in GVCs, and the opportunities for further specialisation.

While countries compete against one another within GVCs, the fragmentation of the production process enables them to specialise and find niches in which they are competitive. Co-operation on the design and financing of education and training programmes can lead to solutions that benefit both sides: countries that play a leading role in GVCs benefit from the quality of skills in other countries and from maintaining their expertise in the teaching of technical skills that might have been offshored; countries to which activities are offshored benefit from education programmes of higher quality, which can help them improve their position in GVCs. In addition, co-operation between multinationals, governments and emerging countries to develop social standards and obligations can lead to better job quality in countries to which activities are offshored. Co-operation on education and training programmes, as well as on social standards, can ensure a more equal distribution of the gains generated by GVCs.

The development of GVCs exposes some workers to the risk of lower wages and displacement, which can have political repercussions. While the effect of GVCs on inequalities within countries is not clear-cut (Chapter 2), rising trade integration can lead to polarisation of politics (Autor et al., 2016). Adverse economic shocks related to international trade may cause voters to support positions that lean towards political extremes on the left or right. This can put huge challenges on education systems and skills policies. This chapter shows that more needs to be done to strengthen the quality and co-ordination of skills policies. At the same time, several innovative and well-designed policies are already in place that target specific groups or on a small-scale basis. For many countries, the challenge is to raise the quality of these policies on a broader scale.


Adalet McGowan, M. and D. Andrews (2015), “Labour market mismatch and labour productivity: Evidence from PIAAC data”, OECD Economics Department Working Papers, No. 1209, OECD Publishing, Paris,

Adascalitei, D. and C. Pignatti Morano (2016) “Drivers and effects of labour market reforms: Evidence from a novel policy compendium”, IZA Journal of Labor Policy, Vol. 5/15.

Alesina A., J. Harnoss and H. Rapoport (2016), “Birthplace diversity and economic prosperity”, Journal of Economic Growth, Vol. 21/2, pp. 101-138.

Arslan, C. et al. (2016), “International Migration to the OECD in the Twenty-First Century”, KNOMAD Working Paper 16, KNOMAD, Washington, DC.Autor, D. et al. (2016), “Importing political polarization? The electoral consequences of rising trade exposure”, NBER Working Paper, No. 22637, The National Bureau of Economic Research, Cambridge, MA.

Bartelsman, E., G. Pieter and J. de Wind (2010), “Employment protection, technology choice, and worker allocation,” IZA Discussion Papers, No. 4895, Institute for the Study of Labor (IZA), Bonn.

Blattman C. and S. Dercon (2016), “Occupational choice in early industrializing societies: Experimental evidence on the income and health effects of industrial and entrepreneurial work”, NBER Working Paper, No. 22683, NBER Working Paper, No. 22637, The National Bureau of Economic Research, Cambridge, MA.

Bloom, N. and J. Van Reenen (2007), “Measuring and explaining management practices across firms and countries”, Quarterly Journal of Economics, Vol. 122/4.

Bloom N. et al. (2013), “Does management matter? Evidence from India”, Quarterly Journal of Economics, Vol. 128/1.

Bloom, N. et al. (2012), “Management practices across firms and countries”, NBER Working Papers, No. 17850, The National Bureau of Economic Research, Cambridge, MA.

Borgonovi, F. et al. (2017), “Youth in transition: How do some of the cohorts participating in PISA fare in PIAAC?”, OECD Education Working Paper, No. 155, OECD Publishing, Paris,

Cavaco, S., D. Fougère and J. Pouget (2009), “Estimating the effect of a retraining program on the re‐employment rate of displaced workers”, IZA Discussion Papers, No. 4227, Institute for the Study of Labor (IZA), Bonn.

CEDEFOP (2008), “Systems for anticipation of skill needs in the EU Member States”, CEDEFOP Working Papers, No. 1, European Centre for the Development of Vocational Training, Thessaloniki.

Clemens, M. (2015), “Global skill partnerships: A proposal for technical training in a mobile world”, IZA Journal of Labor Policy, Vol. 4/2.

Danish Agency for Science Technology and Innovation (2013), The Hidden Champions – Danish Industrial Export Successes, Copenhagen.

De Arcangelis, G. et al. (2014), “Directing remittances to education with soft and hard commitments”, Journal of Economic Behavior and Organization, Vol. 111, pp. 197-208.

De Backer and Destefano (forthcoming), “The links between global value chains and global innovation network: An exploration”, OECD Science, Technology and Industry Working Papers.

Docquier, F. and H. Rapoport (2012), “Globalisation, brain drain, and development”, Journal of Economic Literature, Vol. 50/3, pp. 681-730.

Dougherty and Reddy (2011), “The impacts of state performance funding systems on higher education institutions: Research literature review and policy recommendations”, CCRC Working Paper No. 37, Community College Research Center, Teachers College, Columbia University, New York, December.

Eurofound (2010), “Very atypical work: Exploratory analysis of fourth European Working Conditions Survey”, Background Paper.

European Commission (2014), The ERASMUS Impact Study: Effects of Mobility on the Skills and Employability of Students and the Internationalisation of Higher Education Institutions, Publications Office of the European Union, Luxembourg.

European Commission (2015), Building Knowledge on International Cooperation in VET, Publications Office of the European Union, Luxembourg.

European Commission/EACEA/Eurydice (2015a), The European Higher Education Area in 2015: Bologna Process Implementation Report, Publications Office of the European Union, Luxembourg.

European Commission/EACEA/Eurydice (2015b), Adult Education and Training in Europe: Widening Access to Learning Opportunities, Eurydice Report, Publications Office of the European Union, Luxembourg.

European Union (2009), “The little scientist’s house”, Compilation of good practice on fostering creativity and innovation in the fields of learning and cultural awareness,

Fairlie, R. and M. Lofstrom (2013), “Immigration and entrepreneurship “, in B. Chiswick and P. Miller (eds), The Handbook on the Economics of International Migration, North Holland, Amsterdam.

Foley, C.F, and W.R. Kerr (2013), “Ethnic innovation and U.S. multinational firm activity”, Management Science, Vol. 59/7, pp. 1529-1544.

Garber, G.R. (2013), “Non-compete clauses: Employee mobility, innovation ecosystems, and multinational R&D offshoring”, Berkeley Technology Law Journal, Vol. 28/4.

Heckman, J.J. and T. Kautz (2013), “Fostering and measuring skills: Interventions that improve character and cognition”, NBER Working Paper, No. 19656, The National Bureau of Economic Research, Cambridge, MA.

Hoogendoorn, S. and M. van Praag (2012), “Ethnic diversity and team performance: A field experiment”, IZA Discussion Paper, No. 6731, Institute for the Study of Labor (IZA), Bonn.

Horwitz K. and B. Horwitz (2007), “The effects of team diversity on team outcomes: A meta-analytic review of team demography”, Journal of Management, Vol. 33/6, pp. 987-1015,

ILO (2015), World Employment and Social Outlook 2015: The Changing Nature of Jobs, International Labour Organization (ILO), Geneva.

Jacobson, L., R.J. Lalonde and D.J. Sullivan (2005), “The returns to community college schooling for displaced workers”, Journal of Econometrics, No. 125, pp. 271-304.

Javorcik, B.S. et al. (2011), “Migrant networks and Foreign Direct Investment”, Journal of Development Economics, Vol. 94/2, pp. 231-241.

Johnstone, D.B. (2004), “The economics and politics of cost sharing in higher education: Comparative perspectives”, Economics of Education Review, Vol. 23, pp. 403-410.

Johnstone, D.B. and P. Marcucci (2010), Financing Higher Education Worldwide: Who Pays? Who Should Pay?, Johns Hopkins University Press, Baltimore, MD.

Kerr, W. and S.P. Kerr (2016), “Immigrant entrepreneurship”, in J. Haltiwanger et al. (eds), Measuring Entrepreneurial Businesses: Current Knowledge and Challenges, Studies in Income and Wealth (NBER), University of Chicago press, Chicago, IL.

Kerr, W. and M. Mandorff (2015), “Social networks, ethnicity, and entrepreneurship”, NBER Working Papers, No. 21597, The National Bureau of Economic Research, Cambridge, MA.

Kuczera, M, S. Field and H.C. Windisch (2016), Building Skills for All: A Review of England, OECD Skills Studies, OECD, Paris,

Le Donné, N., P. Fraser and G. Bousquet (2016), “Teaching strategies for instructional quality: Insights from the TALISPISA link data”, OECD Education Working Papers, No. 148, OECD Publishing, Paris,

Marx, M., D. Strumsky and L. Fleming, (2009), “Mobility, skills, and the Michigan non-compete experiment”, Management Science, Vol. 55/6, pp. 875-89.

Nathan, M. and N. Lee (2013), “Cultural diversity, innovation, and entrepreneurship: Firm-level evidence from London”, Economic Geography, No. 89, pp. 367-394.

OECD (forthcoming a), OECD Reviews of School Resources: The Funding of School Education, OECD Publishing, Paris.

OECD (forthcoming b), Engaging Employers in Putting Talents to Better Use in the Workplace, OECD Publishing, Paris.

OECD (2017), Taxation and Skills, OECD Tax Policy Studies, No. 24, OECD Publishing, Paris,

OECD (2016a), PISA 2015 Results (Volume I): Excellence and Equity in Education, PISA, OECD Publishing, Paris,

OECD (2016b), Getting Skills Right: Assessing and Anticipating Changing Skill Needs, OECD Publishing, Paris,

OECD (2016c), OECD Science, Technology and Innovation Outlook 2016, OECD Publishing, Paris,

OECD (2016d), Education at a Glance 2016: OECD Indicators, OECD Publishing, Paris,

OECD (2016e), International Migration Outlook 2016, OECD Publishing, Paris,

OECD (2015a), OECD Skills Outlook 2015: Youth, Skills and Employability, OECD Publishing, Paris,

OECD (2015b), The ABC of Gender Equality in Education: Aptitude, Behaviour, Confidence, PISA, OECD Publishing,

OECD (2015c), OECD Science, Technology and Industry Scoreboard 2015: Innovation for growth and society, OECD Publishing, Paris,

OECD (2015d), “Activation policies for more inclusive labour markets”, in OECD Employment Outlook 2015, OECD Publishing, Paris,

OECD (2014a), Skills beyond School: Synthesis Report, OECD Reviews of Vocational Education and Training, OECD Publishing, Paris,

OECD (2014b), Employment and Skills Strategies in the United States, OECD Publishing, Paris,

OECD (2014c), OECD Employment Outlook 2014, OECD Publishing,

OECD (2014d), OECD Science, Technology and Industry Outlook 2014, OECD Publishing,

OECD (2014e), Education at a Glance 2014: OECD Indicators, OECD Publishing, Paris,

OECD (2013a), PISA 2012 Results: Excellence Through Equity (Volume II): Giving Every Student the Chance to Succeed, PISA, OECD Publishing, Paris,

OECD (2013b), OECD Skills Outlook 2013: First Results from the Survey of Adult Skills, OECD Publishing,

OECD (2013c), OECD Employment Outlook 2013, OECD Publishing, Paris,

OECD/World Bank Group (2015), “Inclusive global value chains: Policy options in trade and complementary areas for GVC integration by small and medium enterprises and low-income developing countries”, OECD and World Bank Group report prepared for submission to G20 Trade Ministers Meeting, Istanbul, Turkey, 6 October 2015,

Ottaviano, G.I. and G. Peri (2005), “Cities and cultures”, Journal of Urban Economics, Vol. 58/2, pp. 304-337.

Samila, S. and O. Sorenson (2009), “Non-compete covenants: Incentives to Innovate or Impediments to Growth”, Danish Research Unit for Industrial Dynamics Working Paper, No. 10/02.

Sieger, P., U. Fueglistaller and T. Zellweger (2016), Student Entrepreneurship 2016: Insights From 50 Countries, International Report of the GUESSS Project 2016, KMU-HSG/IMU, St.Gallen/Bern.

Siemens Stiftung (n.d.), “Little scientists’ house: Discovering the world with scientific passion”, Siemens Stiftung website,

Starr, E., N. Bishara and J.J. Prescott (2016), “Noncompetes in the U.S. labor force”, Working Paper,

Taglioni, D. and D. Winkler. (2016), Making Global Value Chains Work for Development, Trade and Development series, World Bank, Washington, DC.

Tang, H. (2012), “Labor market institutions, firm-specific skills, and trade patterns”, Journal of International Economics, Vol. 87/2, pp. 337-351.

UKCES (2015), “Global value chains and the future of high skills: Evidence from Singapore and implications for the UK”, Briefing Paper, UK Commission for Employment and Skills (UKCES), London.

UKCES (2010), Skills for Jobs: Today and Tomorrow, National Strategic Skills Audit for England 2010, UK Commission for Employment and Skills (UKCES), London.

UK Office for National Statistics (2016), “Contracts that do not guarantee a minimum number of hours: September 2016”, UK Office for National Statistics.

Vincent-Lancrin, S. (2016), “Open higher education: What are we talking about?”, Background Paper for OECD conference on “Opening Higher Education: What the Future Might Bring”, Berlin, 8-9 December 2016.

Wächter, B. and F. Maiworm (2014), “English-Taught Programmes in European Higher Education: The State of Play in 2014”, ACA Papers on International Cooperation in Education, Brussels.


← 1. The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

← 2. Global University Entrepreneurial Spirit Students’ Survey is a global survey of students’ entrepreneurial intentions and activities immediately after graduation and five years later. It started in 2003 and is undertaken by the Swiss Research Institute of Small Business and Entrepreneurship at the University of St. Gallen. In the 2016 edition, 122 509 students from 1 082 HEIs from more than 50 countries participated (Sieger, Fueglistaller, and Zellweger, 2016).

← 3. According to this study, in Austria for example, 50% of so-called marginal workers are estimated to be unemployed people, retirees or students. Moreover, studies in Austria have found that nearly half of all marginally employed people have more than one employer and that 4.3% have as many as three employers at the same time.

← 4. The Delphi method entails a group of experts who anonymously reply to questionnaires and subsequently receive feedback in the form of a statistical representation of the “group response”, after which the process repeats itself. The goal is to reduce the range of responses and arrive at something closer to expert consensus. The Delphi Method has been widely adopted and is still in use today.

← 5. This increase in share has been partly driven by the decrease in business creation by natives from the Great Recession starting in 2007 in the United States (Fairlie and Lofstrom, 2013).

← 6. The initial programme was the “convention de conversion”. It has been changed into a “convention de reclassement personnalisé”, which includes a social benefit in addition to retraining and job search actions.