Potentially hidden unemployment in the Portuguese labor market data

October 8, 2019

The unemployment classification relies on the degree of attachment to the labor market, based on the job search criterion. An individual aged between 15 and 74 years-old is considered to be unemployed if he or she is not working, but is actively searching for a job and is available to start working within the next two weeks. Several studies have highlighted the arbitrariness of the unemployment classification criteria used in the Portuguese Labor Force Survey, considering that the non-employed population is a heterogeneous group. For example, some individuals classified as inactive may have a degree of attachment to the labor market that is close to those officially unemployed.

Among these individuals stand out the so-called “marginally-attached”, which are those that wish to work, but are not actively searching for a job or are not immediately available to start work. These include the discouraged workers which despite being available to work are not actively searching for a job. On average, about 12% of these marginally-attached individuals move to employment every quarter. This transition rate is lower than for the short-term unemployed (27.5%), but rather close to the long-term unemployed (17.1%), and clearly higher than that of the other inactive individuals (4.6%). Employment inflows originating from inactivity are relevant and higher in absolute terms than employment inflows originating from unemployment.

These subgroups in inactivity impact our view on the amount of under-utilized labor supply in the market. In our data, the marginally-attached represent roughly 6% of the non-employed population and discouraged workers almost 2%. If included in unemployment, they could potentially raise the unemployment rate by about 6 p.p. and 1 p.p., respectively.

In this paper, we conclude that the “marginally-attached” constitutes a distinct group of individuals displaying a transition behavior closer to unemployment than to the group of inactive workers that do not want work. In the Portuguese Labor Force Survey data, the classification as inactive workers of those individuals that report “waiting” as a reason for not having searched for a job, plus those who have searched for a job but are still considered to be out-of-the-labor-force, as well as those individuals that are due to start work in more than three months may not be reasonable, considering that each of them show considerable attachment to the labor market.  The paper argues in favor of rejecting the pooling of such states with the rest of the marginally-attached when thinking about the unemployment rate.

Click here to go to the paper by Fernando Martins and Domingos Seward.