Although this descriptive analysis might have considerable implications for policymakers and researchers, these results do not provide enough evidence about the validity or reliability of vacancy data to address skill shortages and their consequences. As is the case with data collected by other methods (e.g. surveys), information collected from job portals have limitations that affect interpretation (Chapter 4). Consequently, there is a critical need to assess the validity of the vacancy database to be sure of what it can tell us about labour demand. Thus, Chapter 8 performs extensive internal and external validity tests on the vacancy database (Henson 2001; Rasmussen 2008). First, it evaluates internal validity (consistency of the variables within the same database) via cross-tabulations and wage distribution analysis. Second, it tests external validity (representativeness) of the online vacancy information. This examination requires a comparison of the vacancy database results against other sources of information (e.g. household surveys). To do so, this book re-categorises occupations from Colombian household surveys to create updated occupational classifications that are compatible with occupational categories in the vacancy database.
After completing the described homologation, a “traditional” test is conducted by comparing the occupational structures of supply and demand. However, given the limitation of the “traditional” test, further tests are carried out to investigate the external validity of the database. Specifically, the wage distribution of labour demand and information on supply are examined to perform a relevant comparison of time series between jobs in demand, employed and unemployed individuals in the total workforce, as well as the extent of new hires (replacement demand and employment growth) by major occupational groups. These detailed tests provide information about the advantages and limitations of the vacancy database for a labour demand and skill mismatch analysis.
Once the advantages and limitations of the data are established, Chapter 9 proceeds to develop a system to identify possible skill shortages and address labour supply according to employer requirements in Colombia. First, the chapter provides a detailed description of the Colombian labour market panorama (formally or informally employed, as well as unemployed) at the occupational level. Second, it combines Colombian household survey information and the vacancy database to estimate a Beveridge curve and a set of eight (volume- and price-based) macro-indicators to identify possible skill shortages. This chapter also highlights the importance of controlling for informality when building skill mismatch indicators in a context such as Colombia. Occupations might exist with relatively low unemployment rates, but also with a relatively high informality rate, or vice versa. Accordingly, increases in the number of workers in certain occupations—for instance, those characterised by relatively low unemployment rates—might increase informality rates. Therefore, this document advises policymakers and training providers to be aware of this relevant labour market duality when providing and promoting skills. Furthermore, this chapter shows how detailed information from vacancies (job descriptions) can be used to monitor labour demand trends for skills, as well as to update occupational classifications according to current employer requirements.
Finally, Chapter 10 summarises the relevant conceptual, methodological, and empirical contributions of the book, while opening a debate on the use of novel sources of information (job portals) to fill information and analysis gaps regarding the labour market. Thus, this chapter highlights the implications of the findings for national statistics offices, policymakers, education and training providers, and career advisers. Additionally, it points out the limitations of the study and illustrates new avenues of enquiry for future research.
This comprehensive and detailed methodological and conceptual framework alongside empirical findings presents important evidence about the advantages and limitations of job portals for their use in economic analysis. It provides a basis to develop a consistent skill shortage monitoring system that can be beneficial for different countries when adopted.
2. The Labour Market and Skill Mismatches
The labour market can be defined as a “place” (not necessarily a physical place) where employers (“demand”) and workers (“supply”) interact with each other. The dynamics of this labour market are relevant for an economy as they determine different socio-economic outputs, such as productivity, unemployment, wages, and poverty, among others. Provided that the labour market influences various outcomes and different disciplines address these issues (e.g. sociology, economy, etc.), this chapter discusses labour market definitions and explains the theoretical framework adopted throughout this book to analyse labour demand based on the information found on online job portals.
The second section of this chapter explains what is understood by labour demand and labour supply in the academic literature on economics, and possible ways to statistically measure these concepts. Moreover, it defines and highlights informal economy as a key issue, especially in Latin American countries like Colombia. Subsequently, the concept of skills is introduced, and its possible implications for unemployment and informal economy are explained. With these basic definitions outlined, the third section of the chapter describes the Colombian labour market and its main outcomes, such as unemployment, wages, etc., under the assumption of perfect competition.
However, the assumptions of perfect competition are substantial and might not be appropriate for different economies such as the Colombian economy. Consequently, it is necessary to consider labour market failures—for example, imperfect information—that might appropriately explain the comparatively high rates of informal economy and unemployment levels in Colombia. Thus, the fourth section of this chapter focuses on explaining how imperfect information might increase skill mismatches and, consequently, create labour market segmentation between formal and informal workers along with a comparatively high unemployment rate, proposing thus that information failures might be one of the leading causes of high unemployment and informality rates, especially in developing countries like Colombia.
Comparable to other markets (e.g. financial markets, physical consumer markets, etc.), the labour market is composed of demand and supply (Cahuc, Carcillo, and Zylberberg 2014). The merchandise to be exchanged consists of “labour services” that represent human activities (distinguished by numbers of workers or hours of work); these human activities are one of the inputs in the production of goods and services (ILO 2018). Consequently, the dynamics between demand and supply have various implications for a range of individuals, for instance, for people with different characteristics (i.e. skills), employers who create job offers with certain requirements, and government institutions, among others. Thus, this section explains who form the labour demand and labour supply (e.g. unemployed, formal, and informal workers), as well as the relevance of skills in labour market outcomes.
In a basic economic model, people or households possess a limited quantity of “labour” that they can offer in the labour market in order to have an income to acquire goods and services (Cahuc, Carcillo, and Zylberberg 2014). Therefore, the labour supply or labour force is composed of people who offer their “labour.” As shown in Figure 2.1, the labour supply or economically active population (EAP) is composed of 1) people who do not have a job but are looking for one (unemployed)