Skills imbalances are largely the result of failures that characterize the education and labor markets. First, individuals and training providers frequently make decisions based on incomplete information on the returns from each investment. Second, there is a lag between the time when individuals make training decisions and when they enter the workforce. In addition, as a result of institutional constraints and information asymmetries, wages do not always reflect the skills demand of the private sector. In this context, shortages or surpluses of certain skills or occupations are common and very difficult to prevent. If these imbalances are large and persistent, there can be significant economic and social costs.
This is why many developed countries have a long tradition of initiatives designed to anticipate skills demand. Despite their limitations, these initiatives can inform public policies on skills development, investments made by education and training providers, and career and training decisions made by individuals. Given the current context of rapid technological change in which demand for skills can change rapidly and unpredictably, these initiatives are becoming even more relevant.
Different methodological approaches are described in this document with illustrative examples. Given the differing strengths and weaknesses of each of these approaches, many countries use a combination of methodologies. In general, the most successful systems efficiently combine public and private efforts. In the case of the government, two tasks are essential. The first one is the generation and processing of information which can come from a variety of sources, such as administrative data and surveys. Household labor surveys provide information on the workforce and its distribution by occupation. Employer surveys are key sources of information on economic activity by sector and employment, as well as on skills demand. Administrative information on vacancies and job seekers is needed to obtain a real measure of the degree of tightness of the labor market. In some countries, governments have also taken advantage of the increasing availability of “big data,” such as the data in digital platforms on labor supply and demand, to carry out this function.
The second task is the development of a system to classify industries, occupations and qualifications in order to make comparisons between sectors and over time. Developing a classification system of qualifications is particularly important if the ultimate goal of the system is to inform skills development policies.
The private sector also plays an essential role. First, employers play a key role in strengthening and validating projections and scenarios with specific sectorial information. The dynamism and speed of the transformation of many sectors of the economy, in a world of rapid technological changes and constant innovation, are fundamental elements to be considered in the process of anticipating skills demand. Second, private sector participation not only improves the quality of skills demand projections, but also adds value to the process itself. To the extent that the effort to identify future skills demand facilitates dialogue between the various stakeholders (government, companies, training providers), it creates a space for deliberation which undoubtedly enriches the design, monitoring and evaluation of skills development policies.
This report describes and classifies different methodologies to anticipate skills demand and characterizes the institutions leading these efforts around the world. However, it is worth mentioning that building an effective system to anticipate skills demand implies challenges that are not only technical but also institutional. Systems to anticipate skills demand should not only focus on generating quality data but also on building processes to ensure that the information is effectively used in decision making. The ultimate goal of these systems is to inform decisions made by households, firms, training providers and policymakers. To achieve this, the construction of quality data should be embedded in an institutional environment that enables stakeholders to access the relevant information and to make their decisions accordingly.