Chapter 2. Skills assessment and anticipation exercises

This chapter describes the fundamental aspects of the Brazilian Skills assessment and anticipation exercises. First, it discusses initiatives that have been led by the Government. Second, it describes initiatives launched by other relevant stakeholders, such as employers’ associations. The chapter analyses the specific challenges faced by these exercises along with solutions that have been found effective in other countries.


In order to design skills policies and training programmes that effectively anticipate and tackle skills imbalances, as well as respond to labour market needs, countries must thoroughly analyse their demand for and supply of skills in a systematic and regular manner. Methods to assess current and prospective skill needs are known as Skills Assessment and Anticipation (SAA) exercises.

Skills assessment and anticipation exercises have existed for more than 50 years and are carried out in virtually all OECD countries. There is no single approach to SAA exercises and the frameworks put in place by different countries vary in terms of the definitions used for skills, the time span, the frequency, the methods and data sources used and whether the exercises have a national, regional or sectoral scope (OECD, 2016).

In Brazil, as of now, there are no rigorous and systematic initiatives to carry out this kind of exercises. In this chapter, the few and scattered SAA initiatives in Brazil are described and compared to the best practices of SAA exercises internationally. Some recommendations are drawn based on these examples from abroad.

2.1. Government-led SSA initiatives

SAA exercises are tools to generate information about the current and future skills needs of the labour market and the available skill supply. The information produced by SAA exercises can serve several purposes. It can inform policy makers in charge of defining employment, education and training, and migration policy. In addition to governments, social partners can also use the information produced by SAA exercises to advise their members on skills to promote within their firms or among their workers, to inform collective bargaining processes and to disseminate the information to a wider audience. Individuals can use the information coming from SAA exercises when making educational or career choices.

2.1.1. Definition of skills for SAA exercises

Skill needs are commonly approximated by measuring which occupations are, or will be, in greater or lesser demand. Occupational forecasts are, in fact, at the origin of the skill needs exercises pioneered by the Bureau of Labour Statistics in the United States (OECD, 2016). However, it is not always clear what skills, educational qualifications or fields-of-study are the most appropriate to satisfy those occupational needs. In some countries, qualification frameworks, occupational standards or detailed descriptions about occupations have been developed to link specific occupations with specific qualifications, fields-of-study or competencies, and ultimately, determine current or future skill needs in terms of attributes that are more useful to those responsible for designing education, training and skills policies.

In Brazil, the Ministry of Labour developed the “Classificação Brasileira de Occupações” (CBO) or the Brazilian Classification of Occupations ( The CBO is a document that lists existing occupations in the Brazilian formal labour market, briefly describes the job content, as well as the education and experience requirements in terms of qualification level and field-of-study. A framework has been developed that links each occupation with related professions. Occupations in the CBO are listed with a high level of detail (up to six digits classification). The CBO has existed since 1982 and has already been updated several times so as to reflect structural changes in the labour market. The most recent update was carried in 2002.

The development of the CBO allows the Brazilian government to link trends in occupational demand with demand for specific educational degrees and levels of experience. One of the uses of the CBO, for example, has been to guide individual educational and training choices. The Ministry of Education (MEC) regularly publishes updated catalogues of vocational training courses and short free-form training courses (FIC courses): the “Catálogo Nacional de Cursos Técnicos” and the “Guia PRONATEC de cursos FIC”, respectively (further details about the later in chapters 4 to 8). These catalogues list all MEC-approved training courses, the minimum number of hours of training required, the entry requirements for each course, and a short description of the course content, among other things. For the last edition of these catalogues, in 2016 , the associated occupational CBO codes were included for some of the listed training courses. Based on that information and the regular statistics published by the Ministry of Labour about median earnings and hiring rates per occupation, individuals can make a more informed decision on which vocational training programme or FIC course to enrol into.

The disadvantage of relying on measuring skills as qualification levels, types or fields of study is that educational credentials do not necessarily map to skills required on the job and that there is a substantial variability amongst individuals with the same credentials in terms of their skills and readiness to perform a job (Quintini, 2011).

In response to such limitation, several countries link occupation-based assessment and anticipation information to specific skills through comprehensive occupational standards or descriptions of what skills are required in each occupation (OECD, 2016). In Canada, for example, the National Occupational Classification (NOC) describes the skills required by each of 500 occupational unit groups. In the United States, as another example, a database has been constructed - called O*NET - that contains detailed information about the knowledge, abilities and competencies’ requirements of more than 900 occupations (Box 2.1). Developing a more detailed description of skills associated to each occupation listed in the CBO would be one possible way forward to develop a more effective skills assessment and anticipation system in Brazil. In Italy, a survey has been implemented to regularly identify skill, knowledge, values and attitudes required by an occupation (OECD, 2016).

Box 2.1. O*NET: The Occupational Information Network

The Occupational Information Network (O*NET) is a project developed under the responsibility and sponsorship of the U.S. Department of Labor/Employment and Training Administration (USDOL/ETA). The project consists in collecting information about occupations in the U.S. labour market so as to understand how the nature of work is changing and how it impacts the workforce and U.S. economy.

Every occupation requires a different mix of knowledge, skills and abilities, and is performed using a variety of tasks. The O*NET project constructed a database containing a standardised description of almost 1 000 occupations regarding day-to-day aspects of the job and the qualifications and interests of the typical worker. Occupations are taken from the Standard Occupational Classification1 (SOC), which was last revised in 2010 to keep up with the changing occupational landscape.

The O*NET database includes variables that are comparable across occupations and describe:

  • Worker characteristics: personal and enduring characteristics required in each occupation for effective work performance. For example, abilities, preferences for work environments, work values and work styles.

  • Worker Requirements: work-related characteristics that can be acquired and developed through experience and education required in each occupation. For example, basic skills that facilitate learning, cross-functional skills, knowledge and prior education required to perform the job.

  • Experience Requirements: requirements related to previous work activities. For example, experience and training required to perform the job, entry requirements to be hired for the job, licenses, certificates or registrations required to hold the job.

The database also includes occupation-specific descriptors, such as the title and SOC identifying code, the description of required and important duties performed by workers in that occupation, alternate titles for the same job, occupation-specific tasks, tools and technologies used in each occupation.

The database is available to the public at no cost and can be downloaded at The data is continually updated with ongoing surveys of each occupation’s worker population and occupation experts. Respondents indicate the importance of a given skill, knowledge or ability for their job (on a scale from one, not important, to five, extremely important) and the level of the skill, knowledge or ability needed for their job (on a scale from zero to seven). Work styles and contexts are classified solely based on importance.

O*NET is widely considered to be the most detailed and comprehensive assessment of skills used in employment that exists (Beblavy et al., 2016; Dickerson et al., 2012). O*NET has even been regularly used for the analysis of countries other than the United States, assuming that skills needed for particular occupations in the U.S. are the same as those needed by the same occupations elsewhere. This assumption has been tested and largely holds (Cedefop, 2013; Koucky et al., 2012; Lepic and Koucky, 2013). However, in low-income countries, the skill content of certain occupations might differ substantially as the technology and regulatory context is quite distinct (Aedo and Walker, 2012; Aedo et al., 2013; Arias et al., 2014; Handel, 2016).

2.1.2. Data sources for SAA exercises

Ideally, to achieve robust and reliable results, SAA exercises should rely on a combination of both quantitative and qualitative sources of information (CEDEFOP, 2008). This reduces potential biases and helps to expand the scope of the exercises. Common quantitative sources of information include analyses of labour market information (e.g. flows in and out of employment by occupation and sector, trends in wages by occupation, trends in hours worked by occupation, etc.), vacancy surveys, employer surveys, surveys of recent graduates and administrative data on enrolments in and graduation from various levels of education (OECD, 2016). Qualitative inputs may come from interviews, focus groups, round tables, etc.

There are only a few countries that rely on a mix of quantitative and qualitative data sources in the same exercise: Australia, Flanders (Belgium), Italy and Korea. Most countries rely solely on quantitative sources of information. Some countries, such as the Netherlands, draw on qualitative sources for validation purposes rather than a direct input in itself.

But even if using only quantitative sources, it is advisable to combine different data sources. Administrative data is usually of good statistical quality and comes with a large sample or covers the entire population of interest. The information reported is reliable as it is usually coming from social security and tax records or mandatory forms. Administrative data, therefore, is ideal to track labour market trends, as well as trends in educational attainment and enrolments. However, the level of detail and information on skills contents of jobs is often limited in administrative sources.

Surveys may need to be carried out to collect more precise information on skills. This would include graduate surveys, employers’ surveys, vacancy surveys and working condition surveys. Graduate surveys, for instance, are useful to understand the employability of recent graduates and to shed light on how well the education system is aligned to labour market needs. Employers’ surveys, on the other hand, are used to ask employers about their skill shortages. Vacancy surveys are useful to understand why certain vacancies remain unfilled, and working conditions surveys can provide further information on the work content of jobs and changes experienced at the workplace. Surveys, nonetheless, may suffer from selection bias, low response rates, lack of representativeness and misreporting. Hence the need to use them together with administrative data sources.

Finally, to be able to combine administrative and survey data, and link databases at the individual level using tax identifiers for example, rules and regulations regarding the use of confidential data should be defined accordingly.

In Brazil, there are good quality administrative data sources. For instance, to track trends in wages, or flows in and out of the formal labour market, RAIS (“Relação Annual de Informações Sociais” - and CAGED (“Cadastro Geral de Empregados e Desempregados”) provide all the information needed. RAIS is an administrative form that must be submitted to the Ministry of Labour each year by all establishments with at least eleven employees. Establishments submit information about their current employees, such as wages, hours worked, occupations, etc. CAGED is an administrative form submitted to the Ministry of Labour for every individual hired or dismissed. Such trends can be complemented using survey data from PNAD (“Pesquisa Nacional por Amostra de Domicílios”), for example, to obtain an overview of what is happening in the informal labour market too. Information on employment vacancies disseminated by the Public Employment Services (PES) could also be exploited to understand trends in employment creation by region, occupation and sector of activity, although it may not be fully representative if many firms disseminate their employment opportunities using other channels than the PES.

To track enrolments and graduations in different vocational education and training programmes, data coming from the SISTEC portal (a portal developed by the Ministry of Education where technical school register courses offered, students’ enrolment, attendance, completion, etc. - can be used and provides important demographic information about the participants.

On the downside, there are no graduates surveys, employers’ surveys, vacancy surveys or working conditions surveys in Brazil. Recently, the Ministry of Industry, Foreign Trade and Services (MDIC) developed a web portal called “SuperTec( where firms can register and fill in a form to report their training needs and the reason why such training needs emerged (the firm is growing, the firm is planning new investments, the firm needs to replace some or all of its workforce, etc.). The information collected from this platform resembles the information that would be collected in an employers’ survey on skills needed. However, registration in the “SuperTec” platform is voluntary and the number of firms that have registered to submit information is relatively small. Consequently, the information collected via “SuperTec” is not representative of all firms in Brazil and its regions. Small and Medium Enterprises (SMEs) and family businesses, for example, are particularly under-represented. In fact, they have very little incentives to register (further details provided in Box 5.3). Even large firms are reluctant to register and provide information on their planned investments by fear that such information would reach their competitors.

Registrion in the platform “SuperTec” could be made mandatory for all firms, even small firms and businesses. To mitigate concerns about data confidentiality, rules and regulations concerning the use of confidential data could be strengthened and effectively enforced. Firms should obtain the guarantee that the information will only be used for policy purposes and that only aggregated results will be published, so that no individual firm can be singled out or identified. Alternatively, a random and representative sample of firms could be selected in each region and required to register and submit the training needs form via “SuperTec”. This would be equivalent to setting up an employers’ survey on skills, such as the one conducted in the United Kingdom (UK), for example.

In a country like Brazil, with a vast territory and regions that differ substantially from each other’s in terms of geographical landscape, demography, infra-structures or social capital, it is important to obtain data sources that are representative both at the national and sub-national level. In fact, skill needs are likely to differ significantly from one region to the other and this phenomena can only be captured if representative data at the sub-national level is available.

Data should be representative of the current situation and as up-to-date as possible. Recent data allows the policy response to be relevant. Nonetheless, having long time series available allows for better interpretation and understanding of trends. Long time series are also necessary if quantitative forecasting models are to be developed.

Box 2.2. The U.K. Employer Skills Survey

The Employer Skills Survey (ESS) collects information from over 87 000 establishments across the U.K. and covering all sectors of the economy, on skills needs, skills use and skills development. It is one of the largest business surveys in the world. The survey has been conducted biennially since 2011 and was last collected in 2017, under the responsibility of the Department for Education.

Interviews are conducted at the establishment level, rather than the organisational level, with employers that have at least two employees on the payroll, and with the most senior person at the site with responsibility for human resources and workplace skills. The interview is carried out in two parts: a core questionnaire and a follow-up survey looking specifically at the investment employers made in providing training to employees in the previous twelve months. Both surveys are conducted by telephone.

The questionnaires cover information on the demographics of firms, recruitment activity and difficulties filling vacancies due to skill-shortages, the skills lacking in the available labour market, the extent and nature of skills gaps within the current workforce and how these affect different occupations, how employers respond to these skills gaps, employers’ training and development activities for their employees (type of training provided and expenditure on training), as well as how employers manage, develop, engage and incentivise their staff.

When enquiring firms about specific occupations, the questionnaire uses broad occupational groups, such as “high-skill occupations”, “middle-skill occupations”, “service-intensive occupations” and “labour-intensive occupations”. The exact questionnaire used and further information can be found at:

2.1.3. Developing SAA methods

There is no single method of SAA. Several approaches and methods have been developed, some of which are descriptive of the recent past and current situation, while others are forward-looking. Most SAA systems combine several methods.

Approaches that refer to the recent past or current situation can be analysis of labour market trends (like the construction of shortages indicators for the OECD Skills for Jobs Database 2018 used in the previous chapter, analysis of flows in and out of the labour market by occupation and sector, analysis of trends in wages, etc.), or micro-level studies (for example, studies of the placement of recent graduates, or studies of the vacancies posted and hard-to-fill, employers’ surveys, etc.).

Forward-looking approaches, on the other hand, include macroeconomic modelling of future developments (potentially by economic sector, by region, by occupation or skill type), extrapolation of the occupational structure within sectors and regions using econometric methods, and foresights of skill needs within regions, sectors and occupations (using interviews, focus groups, round tables of experts, case studies and scenario development analysis, etc.).

Each approach has its own strengths and weaknesses. Analysis of trends using administrative and survey data and micro-studies using graduates’ surveys, employers’ surveys or vacancy surveys, are good methods to ensure representativeness and obtain detailed information about people’s behaviour and their perceptions. They are also good methods to obtain a more direct measure of skills. Nonetheless, such methods require technical expertise in survey design and conduction (weighting, questionnaire design, interview training, etc.), as well as in the analysis of survey outcomes. It may also be problematic to get responses and large samples are needed to get robust data and results. Therefore, these methods may end up expensive.

Forecasting quantitative models are reliable, comprehensive, consistent and transparent. However, these methods require consistent time series on labour markets (sector, occupation, qualification, etc.) and population (age, gender and labour market participation), expertise in building and running quantitative forecasting models, as well as statistical and programming experience. Such methods can also be costly and give a false impression of precision. Furthermore, not every concept is easily quantifiable.

Qualitative methods (such as interviews, panels of experts, etc.) have the advantage of not requiring specific data. These methods may be able to address problems and concerns more subtly and in greater depth. Additionally, these methods are useful to exchange views and can also provide a holistic picture of skill needs. However, synthesizing qualitative outcomes often proves challenging and results may not be easily extrapolated to other contexts. These methods can also be unrepresentative, provide only a partial view or suffer from potential bias and subjectivity. It is also difficult to conduct such methods in a systematic and consistent way.

In Brazil, there are some elements in place to conduct skills assessment analyses that refer to the current situation and recent past, although no system has been put in place so that such analyses are conducted regularly and consistently, and so that results feed into policy design. Furthermore, forward-looking approaches are missing and there is no department in the government that has been specifically attributed the responsibility of conducting SAA exercises to inform employment, education, training and migration policies altogether. This is one of the biggest challenges to the Brazilian skill development system.

One country that has been particularly successful in developing an effective SAA system, and that has some similarities with Brazil in terms of vastness of geographic area, disparities across regions and population diversity, is Australia. In Australia, the SAA system is well-developed and a wide variety of exercises, combining both quantitative and qualitative data, are carried out at the national, sub-national and sectoral level (Box 2.3).

The Australian SAA system is, nevertheless, resource intensive. It is not always easy to find the right people to acquire and analyse data, just as to communicate the results appropriately. The SAA system in Australia also suffers from the drawback of being occupation-based. A more skilled-based approach would be recommended. Developing a detailed skill-based occupational classification, such as O*NET, to link the results from an occupation-based SAA system, such as the Australian one, to skills, would be a promising avenue.

Box 2.3. The Australian SAA system

The production of skill needs information in Australia is under the responsibility of the Department of Jobs and Small Business which has its own team and budget for the purpose. Other stakeholders also lead exercises, including the Australian Bureau of Statistics (ABS), the National Centre for Vocational Education Research (NCVER), the Department of Education and Training (DET), as well as state-level authorities and industry committees, which provide complementary information (OECD, 2018).

The backbone of the SAA exercise conducted by the Department of Jobs and Small Business is an employers’ skills survey called “Survey of Employers who have Recently Advertised” (SERA). The SERA is a telephone-based interview of employers who have recently advertised vacancies in selected skilled occupations, as they are in a good position to comment on current recruitment. It is conducted by the Department of Employment through its State and Territory Labour Economics Offices (LEOs) and its National Office.

Selected occupations are taken from the Australian and New Zealand Standard Classification of Occupations (ANZSCO) at the six-digit level. Only relatively large occupations are selected (those with national employment of at least 1 500 individuals) and which are considered skilled occupations (that require at least three years of post-compulsory-school education and training). While occupational coverage varies from year to year, there is a core of around 80 occupations which have been assessed annually. These are mostly professionals, technicians and trades workers.

Researchers contact employers who have recently advertised a vacancy for the occupation being assessed. Vacancies are selected from a wide range of sources including metropolitan and regional newspapers and internet sites. When there are many advertisements in the occupation being assessed, a sample is randomly selected. When there are not enough advertisements for a particular occupation, employers are cold canvassed to discuss their recruitment experiences and expectations. Contact with employers is discussion based, rather than a formatted survey, as this allows for the identification of issues which are relevant for each particular occupation (although interviewers are provided with a list of recommended questions as guidance). Key issues discussed includes the proportion of vacancies filled, the number of applicants, the number of suitable applicants, skill and qualification requirements of the position, etc. Key statistical and qualitative information from these discussions is recorded and analysed.

To reduce the influence of seasonal factors, contacts are made at approximately the same time each year for each group of occupations. The exact number of employers contacted depends on the number of people employed in the occupation, the number of employers and the number of vacancies advertised. Attempts are made to survey an appropriate number of employers from both metropolitan (state and territorial capital cities) and regional areas.

Cold canvassing for occupations with low vacancy numbers consists of employers being asked whether they have advertised vacancies in the target occupation in the last six months. If they have, the standard discussion over key issues takes place. Otherwise, discussions focus on the likelihood of them recruiting in the next six months, their expectations of difficulty filling vacancies, whether they have potential employees in the target occupation approaching them for work, their experience with staff turnover in the target occupation, etc.

This procedure provides a consistent methodology for research across states and territories which can be analysed over time to determine trends in skilled labour markets. Nevertheless, to reduce potential bias, the Department of Jobs and Small Business complements the information collected through SERA with data from the Australian Bureau of Statistics, namely the Labour Force Survey, to look at changes in employment levels, trends in wages, hours worked, among other things, at the national level and separately by targeted occupation. Researchers from the Department of Jobs and Small Business also take account of other data sources, such as industry activity statistics and projections and graduate employment outcomes. Finally, results from the analysis are discussed with key associations to ensure that they are consistent with their perceptions and expectations.

All the information collected and analysed is also used to feed into a quantitative forecasting model developed by the Department of Jobs and Small Business. This model forecasts employment trends at the state and regional level, by skill level and by detailed industry and occupation. It consists in the average of an ARIMA (Autoregressive Integrated Moving Average) model and a ESWDT (Exponential Smoothing with Damped Trend) model that places further weight on recent trends rather than historical time series. Forecasts are conducted every year and for a time span of five years. At the end, results are manually adjusted for known future industry and regional developments.

Based on all the information gathered and processed, researchers from the Department of Jobs and Small Business rate each targeted occupation in terms of shortage. This rating exercise is used to define, for example, the Skilled Occupations List (SOL) which is then considered for a range of employment, education and training, and migration policies. Further details can be found at:

2.1.4. Time, geographical and sectoral coverage

To develop an effective SAA system, Brazilian authorities will have to consider the frequency which the SAA exercises should be repeated, the time span of the forecast and foresight exercises2, as well as the extent of the coverage: whether it is conducted at the national level, state level, and municipality level, and finally, separately by sector of activity.

Skill needs anticipatory exercises can be considered short-term (six months to two years), medium-term (two to five years) or long-term (five or more years). Short-term SAA exercises are less common, perhaps because they differ little with the assessment of current skill needs. Medium-term SAA exercises are the most common in OECD countries (OECD, 2016).

SAA are generally carried out on an annual basis. Long-term forecasts covering a time span of ten years or more are run less frequently, but they should still be updated regularly to take into account new developments.

National-level coverage may be useful for broad education and training policy, as well as overall labour market monitoring. However, national-level assessments overlook specific skills needs that may be present in a particular region or sector. Furthermore, labour market mobility usually occurs within sector and within region. Therefore, skill imbalances observed in one region or sector may not be present in another (OECD, 2016). This could be particularly relevant in Brazil. In this context, regional specific and sector-specific exercises are strongly recommended as they can facilitate more targeted policy making.

To achieve national, regional and sectoral coverage, most countries usually carry national assessments whose results can be disaggregated at the regional and/or sectoral levels. In other countries, national exercises are complemented by independent regional and/or sectoral analyses (OECD, 2016). In Norway and Switzerland, for example, those are carried out by professional associations in an ad-hoc basis. When carrying out independent regional- and sectoral-specific exercises, it is important to avoid potential incompatibilities with national results. Such incompatibilities could prevent reaching a consensus on what skills are most needed. For that purpose, some adjustments may need to be made case-by-case.

In Brazil, lack of human and financial resources may constitute a barrier to carry out independent and complementary regional and sectoral SAA analyses. To minimise the costs, complementary independent sector-specific SAA analysis can be carried out for two or three sectors each year, following a rotating structure. This rotating structure has been adopted in Finland, for instance.

In terms of regional coverage, one possibility would be for several municipalities to team-up and aggregate resources so as to develop a local-based SAA framework (statistical infra-structure, network of local stakeholders, local platforms to exchange ideas, information, perceptions and reach a consensus on local skill needs, publication of a common report, etc.). These regional-specific SAA frameworks would not necessarily need to stick to administrative state circumscriptions. For example, border municipalities from different states, but who are well-connected in terms of infra-structures, important flows of workers commuting across them, local firms trading with each other, etc. would be good candidates to team-up and develop a local SAA framework. State governments should support the creation of such local SAA networks, for example, through actions of awareness for their importance and potential impact on regional and national policy, or by organising training workshops where SAA methods are explained. Staff involved in the development of local SAA frameworks could meet regularly across the country to share and learn from everyone’s experience. There are several example of collaboration agreements to develop regional-specific SAA exercises in Sweden (Box 2.4).

Box 2.4. Aggregating interests at the local level: examples from Sweden

In Sweden, the education system is decentralised. While the decentralised system grants ample freedom to municipalities to allocate funds at the local level, it led to the emergence of competition between schools to attract students, both within municipalities and across municipalities within states, so as to secure funding.

The emergence of competition across municipalities has led them to pursue objectives that are not aligned with the national ones, namely attracting students at all costs rather than planning educational offer to tackle skills imbalances in the medium to long-run using SAA information (OECD, 2016).

As a result, platforms to coordinate local interests, produce synergies and reduce competition were developed. Teknikcollege and the regional framework in the Gothenburg region are two successful examples.


The idea behind the concept of Teknikcollege is that education providers from a minimum of three municipalities are required to co-operate through explicit agreements and offer joint technology-oriented courses at different levels. For example, one school can provide the physical infra-structure, while another school provides teaching staff and the third one uses its local network to place students on internships. The different education providers then share the costs and benefits obtained.

The education providers from different municipalities are encouraged to collaborate and form a local network with firms so that they can align their course offer to the needs of employers in their geographical area. For that purpose, regional steering groups were formed by members coming from the industry, education providers, municipalities and regional government. During the meetings of these steering groups, members discuss what are the skills needed locally, what education providers can offer, what investments would be needed to meet local skills needs, and whether synergies between schools across municipalities would help to offer courses that are aligned with these skills needs.

The regions and municipalities are also encouraged to form part of a larger national network with the goal of promoting the exchange of experiences and provide quality assurance of the different local education providers in a coordinated manner. Further information can be found at

The Gothenburg regional network

The Gothenburg regional network consists in an explicit platform to establish a joint planning of the educational offer in the region, so as to meet local labour market needs one to five years ahead, while taking into account the specificities of the regional school system. The network is composed by various local stakeholders, including firms and municipalities. The network is built so as to preserve a good balance between smaller and larger municipalities.

When the network meets, members are tasked to propose ideas on what issues should be prioritised at the regional level. Together, they must draft an agreement on the content of the regional objectives to be passed on to the political steering group of the region. Once the regional objectives are approved, they are sent back to each municipality which is, finally, in charge of enforcing it through the use of its own local resources.

One of the strengths of this approach is that while many Swedish municipalities lack the resources to use SAA information, through the Gothenburg Regional network they are able to benefit from shared statistical information, analytical capabilities, experience interpreting results, etc.

2.1.5. Disseminating SAA results

In most countries, results from SAA exercises are shared and disseminated through the publication of reports and the development of websites. In some countries, public media (TV, radio, newspapers or magazines) are also used to disseminate the results to a wider audience. In the Netherlands, the government also uses social medial channels to share the information collected through SAA exercises (OECD, 2016).

Effective web dissemination usually centralises all the information coming from SAA exercises in one single platform. This is the case, for example, in Canada, with the web platform “Job Bank” (Box 2.5). Web platforms that are particularly successful tend to have more than one interface depending on the targeted audience or to allow for an interactive experience, tailoring the information to users’ interests (OECD, 2017). Other countries have adopted this approach, namely in Austria, with the “Qualification Barometer” web platform, and in Bulgaria, resulting from the collaboration between the Bulgarian Industrial Association, the Confederation of Independent Trade Unions (CITUB) and the Confederation of Labour “Podkrepa” (ILO, 2017).

Other dissemination methods sometimes used are the organisation of seminars, conferences, workshops and panel discussions (for example in Canada and Slovenia), or making experts available to explain SAA results to the press (as is the case in Norway).

In Brazil, the Ministry of Labour (MTb) has recently launched an on-line platform called National Observatory of the Labour Market (“Portal do Observatório Nacional do Trabalho”) that provides information on earning per occupation and earning per level of education. The platform also disseminates information on movements in the Brazilian formal labour market, namely hirings and job separations. Such information is computed from administrative data sets such as RAIS, CAGED and data from “Cadastro Único”, the social security records where all individuals that receive some sort of welfare benefit are listed. The information is published through tables of statistics that can be downloaded and press articles that summarise and interpret the information. The purpose of this platform is to make labour market information available to state governments and municipalities’ managers. However, information cannot be easily interpreted for a non-informed audience and the platform has not been designed for the general public. For instance, it does not allow one to easily compare himself or herself with workers in other regions or other occupations.

Box 2.5. The web platform “Job Bank” in Canada

The platform “Job Bank” ( has three interfaces: “Job Bank for employers”, “Job Bank for job seekers” and “Labour Market Information”.

For employers, the platform allows them to post vacancies on-line so as to disseminate the information to potential candidates. It also provides a job matching service to help recruiters find potential candidates with specific characteristics. The platform also provides information on business financing programmes, employment standards (labour law, health and safety regulations, among other things), training opportunities for the firms’ workers, etc.

For job seekers, the platforms lists all available job opportunities with the possibility of searching according to certain criteria. It is also possible for job seekers to create a resume using a professional template that they can download and print or directly use to apply for jobs on the platform. The platform publishes tips on how to effectively look for a job or how to prepare for an interview. This information is adapted for individuals belonging to particular groups, such as veterans, students and youth, newcomers to Canada, senior workers, persons with disabilities and indigenous people.

Finally, the platform has a whole interface designed to disseminate labour market information:

  • It is possible to explore jobs or career options that match individuals’ skills and knowledge. Individuals can tick out of a list of different skills and knowledge that they possess and visualise the results of available vacancies that demand such skills and knowledge.

  • Individuals can enter an occupation’s name and learn about available jobs in that occupation, associated average wages and career prospects, as well as skills and job requirements for that occupation. The platform allows the possibility of searching only in a restricted geographical area.

  • It is also possible to enter one’s current wage and compare it to individuals in similar occupations in other parts of the country or between occupations in the same geographical area.

  • Individuals can look for high-paying education programmes (based on median earnings and by educational attainment level) or plan their career with the “Career navigator quizzes”.

  • Individuals can compare fields of study according to median earnings, earnings range, and percentage employed, percentage unemployed and percentage not looking for a job after course completion.

All the information used to develop this web platform comes from Government-led SAA exercises at the national and regional governments exercises conducted locally. The information can be easily used by any individual, but also by social partners, regional government staff and national government staff.

Independently, the Ministry of Labour also manages a web portal of employment vacancies called “Emprega Brasil( This web portal is led by the Public Employment Services (SINE). Through this portal, workers can search for employment vacancies, information about unemployment benefits and the minimum salary programme. The portal also redirects individuals to the webpage of “Escola do Trabalhador” (worker’s school), a platform of e-learning courses made available for free and for anyone who wants to enrol ( E-courses offered through this platform correspond to free-form professional qualification courses and they are certified by Brasília’s University (“Universidade de Brasília”). To enrol in these e-learning courses, there are no entry requirements and individuals can progress at their own pace. In the same portal, “Emprega Brasil”, employers can post employment vacancies, verify the resume of a particular worker and whether it matches administrative records, as well as submit compulsory information regarding job hirings and separations.

The two web portals, “Portal do Observatório Nacional do Trabalho” and “Emprega Brasil” are not connected. Consequently, it is not possible to link the occupations associated with highest median earnings and hiring rate according to “Portal do Observatório Nacional do Trabalho” with on-line vacancies or related e-learning courses listed in “Emprega Brasil”, for example.

In parallel, the Ministry of Education (MEC) also runs a web platform called SISTEC ( which displays information about vocational educational and training programmes across the country. Through this platform, it is possible to view all vocational courses leading to a secondary education degree that are available in each state, the schools that offer these courses and the number of students enrolled in each. In fact, it is compulsory for technical schools to register and enter all the details of their vocational course offer. In some cases, when the schools also offer free-form professional qualification courses (the short courses called “Cursos FIC”), these courses will also be listed. However, there is no exhaustive listing for FIC courses as technical schools that do not offer secondary-level vocational education programmes are not obliged to register in the portal. Vocational training courses listed in the SISTEC portal are not linked to particular occupations, so it is not possible to infer the associated expected earnings or career prospects, nor identify high-paying vocational education programmes.

There exists yet another web portal called “Educa Mais Brasil( that provides extensive information regarding scholarships to enrol in basic, secondary, tertiary and vocational education, as well as in professional qualification courses. Through this webpage, individuals can register, obtain information about which scholarships they would be eligible to and be redirected to the relevant webpages to apply for these scholarships.

Finally, MDIC is planning to further develop its work-in-progress web platform “SuperTec” so as to use it as a dissemination tool. Based on the information collected through firms on training needs, the platform would spread information on the occupations which are most in demand by firms in all regions of the country. The idea is that the platform would then establish a link with the related training courses available, listing the technical schools in which such courses are available and the entry requirements. The platform has also been prepared so that individuals can look out for information about each occupation, such as median earnings and number of recent hirings. Lack of financial and human resources is preventing the platform from developing further and being implemented on a large scale.

The different web portals developed by MTb, MEC and MDIC could be improved following the Canadian example. Coordination between the three ministries so as to interact the different portals they manage could lead to the development of a unified web platform with all the relevant information for employers, job-seekers, students, families and workers. Centralising the information in one web portal would make it much easier to obtain the relevant information for all audiences. Interacting the information from different portals in one website would also improve the quality of the information provided and increase its potential uses and applications. As it stands at the moment, initiatives from different ministries overlap in some aspects (for e.g. with the dissemination of information regarding earnings and hiring rate per occupation), while none of them really manages to connect skills needed, training courses available, career counselling and employment opportunities.

2.2. Stakeholders involvement in SAA initiatives

In most countries, several actors are involved in the development of skills assessment and anticipation exercises. They commonly include different ministries, public employment services, regional or sector-specific institutions, and social partners. These stakeholders are also generally involved in the discussion of the results and the development of policy responses.

There are several mechanisms to spur the discussions and to facilitate consensus over which skills are needed across different stakeholders. Formal mechanisms include legal norms that oblige the government to consult with other stakeholders around skill issues, or yet the inclusion of stakeholders in advisory boards to different ministries. Informal mechanisms include the setting up of work groups, round tables or skills councils.

Skill councils, for example, are employer-led or tri-partite organisations involving representatives from employers, workers and government or educational institutions. They are generally publicly-funded, but can receive some additional funding from its private sector members. Skill councils provide a platform for the discussion of skills-related challenges of specific sectors or regional areas, as well as the development of a joint policy response (OECD, 2016). Such skill councils have been established at the sector-level in the United Kingdom and in Finland, for example.

In some countries, employer organisations or individual employers lead their own independent SAA exercises. This approach is less common, however, as in most OECD countries SAA exercises already exist and employers do not see the value added of building their own models.

In Brazil, given the lack of regular and consistent SAA exercises, some regional offices of SENAI (see Box 1.1) have developed their own SAA methods. Such methods have proved very effective in identifying skills needs in the manufacturing sector and aligning SENAI’s training offer to the needs of local labour markets. In fact, the SAA method developed by SENAI São Paulo, for example, has been mentioned in several publications as best practice example to develop a SAA framework (see ILO, 2017).

Box 2.6. SENAI’s SAA exercises

SENAI developed a framework called “Mapa do Trabalho Industrial” (Map of manufacturing work), that tries to anticipate what skills will be most important in the manufacturing sector between 2017 and 2020. This framework employs mixed methods and combines information from interviews with senior managers and local policymakers, as well as data on employment creation per sector. The information gathered is then analysed and translated into a list of training needs across different geographic areas. Table 2.1 presents the main areas of training expected to be in high demand between 2017 and 2020, as identified by SENAI in 2016.

Table 2.1. Areas of training expected to be in high demand between 2017 and 2020

Area of training


Environment and production

Metalworking and mechanics


Clothing and footwear

Information Technology and Communication



Petrochemicals and Chemistry

Wood and Furniture

Paper and Printing


Research, Development and Design

Source: “Portal da Indústria

SENAI also has a national platform to track their former students, called “Sistema de Acompanhamento de Egressos do SENAI” (SAPES) or SENAI’s Graduates Tracking System ( Through this platform, SENAI units in each state collect data on employment rates, occupational patterns and earnings for graduates from all of SENAI’s training courses. The platform also allows SENAI to identify former students’ current employers, so that they can be contacted and asked to participate in regular satisfaction surveys. The information collected is used by SENAI to evaluate the quality of their training courses, as well as to understand which graduates are more easily absorbed in the labour market and enjoy better employment conditions. The results from such analyses can feed back into training courses’ planning and content development.

Besides the nation-wide “Mapa do Trabalho Industrial” and the management of the SAPES platform, some state-level SENAI offices have also developed their own SSA exercises. In fact, SENAI branches in each state have some autonomy to decide which training courses to open, allowing them to adapt to local demand for skills, although they usually try to keep their activities aligned to national guidelines.

SENAI from Santa Catarina

In Santa Catarina, a wealthy state in the south of Brazil, SENAI representatives make decisions on which courses to offer based on direct contacts with local employers. Often, senior management staff from local firms contact SENAI directly to ask for the opening of specific courses. Otherwise, SENAI Santa Catarina sends out surveys to employers in order to identify the key skills needed.

Additionally, in 2013, SENAI Santa Catarina conducted an ad-hoc study called “Programa de Desenvolvimento Industrial Catarinense” or Programme for the Industrial Development in Santa Catarina (PDIC 2022). The purpose of this large study was to identify trends in the state industrial production and strategic industrial sectors ( The study combined the perceptions from employers and experts gathered through workshops and round tables, with the collection of quantitative data. After analysing the information collected, several road maps were published providing general guidelines on how to boost each key sector’s productivity and competitiveness. These guidelines sometimes included references to skills needs.

SENAI from São Paulo

In São Paulo, SENAI developed a rigorous and systematic framework to identify skill needs. This framework combines different methods and data from different sources.

The first step of their framework uses administrative data from the Ministry of Labour (RAIS) to look at trends in the flows in and out of employment by occupation and within the manufacturing sector, separately by region within the state of São Paulo.

The second step consists in connecting the information from the occupational analysis with national and international databases on skills use. At this point, they use different data sources such as O*NET and CBO to understand how trends in occupations translate into trends in skills.

Additionally and independently, SENAI São Paulo also sends online questionnaires to firms in the sector and state to enquire them about industrial developments, future skills needs and hiring difficulties related to lack of skills.

The final stage of the SAA framework consists in combining the quantitative and qualitative data collected, which will be used as background material for the appreciation of a technical committee. Experts with different occupational backgrounds make up these technical committees. They are expected to discuss the results and agree on what the ideal profiles of workers in the manufacturing sector will be in the coming years. Those ideal professional profiles are then used to define which technical courses to offer and how to develop their content.

SAA exercises conducted by SENAI’s regional offices are quite focused on the particular objective of identifying skills needs in the manufacturing sector to define their own courses’ content. This means that such exercises may lack the flexibility or broadness in scope to be directly applicable for other policy purposes or useful to other actors. Nevertheless, SENAI’s SAA model could be adapted by the Brazilian government to be used at the national scale and for other sectors of activity. Alternatively, the Brazilian government could develop a SAA governance structure so that a national SAA framework is complemented by sector-specific exercises led by institutions from the S-System, for example. This would require closer collaboration between the government and S-System institutions, as well as the development of coordination mechanisms to avoid overlapping initiatives and the inefficient use of resources.


Aedo, C. and I. Walker (2012), Skills for the 21st century in Latin American and the Caribbean. Directions in Development; human development. World Bank. License: CC BY 3.0 IGO.

Aedo, C. et al. (2013), From occupations to embedded skills: a cross-country comparison. Policy Research Working Papers, World Bank, Washington DC.

Arias, O. S. et al. (2014), Back to work: growing with jobs in Europe and Central Asia, World Bank Publications, Washington DC.

Beblavy, M. et al. (2016), What are the new occupations and the new skills? And how are they measured?, State of the art report.

CEDEFOP (2008), Systems for Anticipation of Skill Needs in the EU Member States, CEDEFOP Working Paper, No. 1.

CEDEFOP (2013), Quantifying skill needs in Europe, occupational skill profiles: methodology and application.

Dickerson, A. et al. (2012), Developing occupational skills profiles for the UK: a feasibility study, UK Commission for Employment and Skills, Evidence Report 44.

Handel, M. J. (2016), “The O*NET content model: strengths and limitations”, Journal of Labour Market Research, Vol. 49, pp. 157-176.

ILO (2017), Skill needs anticipation: systems and approaches, Geneva.

Koucky, J. et al. (2012), Occupational skills profiles: methodology and application, Contribution to the concept, structure and quantification of skill needs in Europe.

Lepic, M. and J. Koucky (2013), Occupational skills profiles: country specific OSP for the Czech Republic and Related Documentation for 2013 country workbooks, Charles University in Prague, Faculty of Education.

OECD (2018a), Getting Skills Right: Australia, Getting Skills Right, OECD Publishing, Paris,

OECD (2018b), Getting Skills Right: Future-ready adult learning systems, OECD Publishing, Paris. Forthcoming.

OECD (2017a), Getting Skills Right: Good Practice in Adapting to Changing Skill Needs: A Perspective on France, Italy, Spain, South Africa and the United Kingdom, Getting Skills Right, OECD Publishing, Paris,

OECD (2017b), Getting Skills Right: Skills for Jobs Indicators, Getting Skills Right, OECD Publishing, Paris,

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

OECD (2016b), Getting Skills Right: Sweden, Getting Skills Right, OECD Publishing, Paris,

Quintini, G. (2011), Right for the job: over-qualified or under-skilled? OECD Social, Employment and Migration Working Papers, No. 120, OECD Publishing, Paris,


← 1. The Standard Occupational Classification (SOC) system is a federal statistical standard used by federal agencies to classify workers into occupational categories for the purpose of collecting, calculating, or disseminating data (see All workers are classified into one of 867 detailed occupations according to their occupational definition. To facilitate classification, detailed occupations are combined to form 459 broad occupations, 98 minor groups, and 23 major groups. Detailed occupations in the SOC with similar job duties, and in some cases skills, education, and/or training, are grouped together. 

← 2. Forecast and foresight exercises both look into the future. Forecast exercises aim at providing general indications about future trends. Foresight exercises provide a framework for discussion between stakeholders or experts to jointly think about future scenarios, and eventually, to develop priorities that can be translated into policy actions.

End of the section – Back to iLibrary publication page