Second, it will discuss the current meaning of employability, showing how, faced with the demands of operationality, a series of works and developments in practices are beginning to give it the meaning of a collective construction.
1.1. One hundred years of trial and error between the individual and the collective: seven operational definitions of employability
Faced with a vague concept, poorly defined or not defined at all, but clearly covered by various meanings such as employability, one possible strategy is to stick to operational versions, by which we mean the presence for each of them of three directly interrelated components: a definition, a statistical translation and consequences for economic and social policy. This leads us to identify at least seven distinct versions.
1.1.1. Seven versions in three waves during the 20th century
A systematic study presenting the details of the corresponding bibliographical references (Gazier 1999, pp. 62–67) and summarized in (Gazier 2017) has endeavored to distinguish these different versions, and to give them a name that makes it possible to differentiate them. The names chosen are thus intended to clarify this range and were not produced or used by the various people who developed and used this or that version, who mostly confined themselves to talking about “employability” without seeming to note that it was a specific version.
The first version (E1) dates back to the 1900s and persisted, especially in the United Kingdom and the United States, until the early 1950s. It was a simple dichotomous perception of employability. A person was – or was not – employable, that is, able-bodied and immediately available for work. The statistical translation of this “dichotomous employability” gradually emerged around three criteria that became common in a number of studies conducted in the United States during the Great Depression of the 1930s: age criteria (being between 15 and 64 years old), the absence of physical or mental disability, and the absence of strong family constraints such as, for mothers, having young children to raise. People experiencing poverty can then be channeled differently: the “unemployable” could receive emergency social assistance, while the “employable” were first assigned to public works and then, when these had been interrupted, sent back to the labor market. This concept has been widely criticized, partly because it was formulated without reference to the labor market context and partly because it did not provide for any gradation between employability and unemployability.
Modern versions of the concept began in a second wave of usage and elaboration around the 1950s and 1960s, which immediately extended beyond the English-speaking world, and included contributions from many countries, notably France. Three very different types of employability were identified and used by social workers, employment policy makers, statisticians and physicians.
First, the second version, E2, which can be described as “medicosocial employability”. Mainly developed by doctors and rehabilitation practitioners, and aimed at the disabled, this version introduces a quantitative scale from the outset: one can be more or less employable, and this assessment forms the basis of the intervention which aims to improve employability. In concrete terms, it involves scores on a series of items constituting an individual employability test: the abilities of a more or less disabled person are scored in different areas (visual, auditory, cardiac, motor skills, etc.), and also abstraction skills, and the ability to read and write. This is an individual employability test: the abilities of a more or less disabled person are noted in various areas which cover physical as well as mental impairments. Depending on the deficiencies identified, those that can be treated or compensated for are selected and a program of action is drawn up.
This version was almost immediately doubled by a third, more general version, mainly aimed at unemployed people in difficulty. In fact, it is possible to introduce into the scale, with various weightings, items relating to handicaps that are no longer medical but social: we then move on to deficiencies in terms of qualifications, and also in terms of mobility, presentation and so on. For example, a person who does not have a driving license, or who has a criminal record or a history of drug use may be considered to have low employability. This E3 employability can be described as “labor policy employability”. It is intended to measure the distance between the characteristics of an individual and the imperatives of production and acceptability in the labor market. Here again, we can intervene by selecting the items on which action is possible (e.g. training programs or simply driving lessons, or even clothing advice).
However, the E2 and E3 versions, which have been developed mainly in the United States, have the limitation that they are strictly concerned with the individuals whom we wish to bring closer to employment, and thus here again implicitly take as given the state of the labor market and the possible prejudices of employers.
A fourth variant was developed more specifically in France during the 1960s and takes a very different approach to the problem, starting from the collective dimensions. It is called E4, “employability-flow”. In this version, the focus is on the speed with which a group of unemployed people finds a job. This can be measured by the proportion of a given group of unemployed persons – for example, the unemployed over 50 years of age – who have been unemployed for more than one year. This statistic of unemployability (rather than employability) has the advantage of immediately linking the situation of the unemployed to that of the labor market (more or less good economic conditions, more or less selectivity). It can then be broken down into the relative disadvantages of this or that sub-group of unemployed or even of individual unemployed (differential employability).
Versions E2 and E3, on the one hand, and E4, on the other hand, were developed separately, although E4 emerged as a reaction against the practices and representations associated with E3.
The 1970s saw the entry into crisis, in the United States, of E2 and E3, mainly because the scores obtained in the various individual employability tests proved to be rather poorly predictive of an individual’s success on the labor market. Symmetrically and a little later, during the 1980s, it was employability-flux (E4) that went into crisis, particularly in France, when massive and lasting unemployment took hold in Europe. Indeed, it seemed increasingly disheartening to record a long-term deterioration in the employability of the unemployed and to appreciate only a collective dimension that depended mainly on the slowdown in growth: how then could we act to help these people when the paths to a rapid relaunch of activity were closed? This is probably why the statisticians who used this definition in France eventually abandoned it. In fact, they renamed their indicator “reclassification difficulties” in the mid-1980s.
More recently, a third wave, during the 1980s and 1990s, more inter-nationalized, with Canadian contributions in particular, brought three new versions of employability.
First, a series of American studies have proposed since the end of the 1970s a more neutral and complete statistical definition of employability: E5, which can be described as “employability – performance on the labor market”. It consists of establishing three specified probabilities for a group or an individual for a time interval, taking into account available statistical information on employment trajectories: the probability of obtaining one or more jobs, the probable duration of these jobs in terms of working hours, and the probable hourly wage. Multiplying these three probabilities together gives a synthetic indicator of the ability of a person or group to extract income from the labor market. The advantage of this measure is that it does not focus on the probability of finding work alone, and also introduces some minimal indications of the “quality” of the job (duration and wage). It does not postulate a priori any link between individual skills, collective situations or economic or social policy actions and the outcome on the labor market. In this sense, it is neutral and cannot guide action: it can only be used for the retrospective evaluation of a particular intervention program.
This is not the case for two more recent versions developed mainly in the early
