Can we measure the evolution of job quality during the Great Recession and what are main trends?

Quinne — Tue, 11 Oct 2016 21:24:35 +0000

Can we measure the evolution of job quality during the Great Recession and what are main trends?
by Christine Erhel and Mathilde Guergoat-Larivière

Aggregate analysis of job quality using synthetic indexes provides interesting comparative results and identifies some drivers of job quality. It shows for instance that the correlation between job quality and the economic cycle is positive in general: countries where employment is higher and unemployment lower usually have relatively high levels of job quality. Over the years 2005-2010, these indexes generally reveal a slight decrease in job quality that may be partly related to the crisis.

However several reasons call for disaggregating the analysis. Composition effects cannot be corrected at the macro level, which raises concerns when one wants to study the relationship between job quality trends during a recession. Stability or even improvements in aggregate job quality may hide some job destructions in the low-wage low-quality jobs. Job quality improves but only because bad jobs are destroyed…Two other (contradictory) effects are also relevant to explain trends in job quality in the crisis. Rising unemployment can deteriorate the bargaining power of workers leading to decreases in job quality (in terms of working conditions, hours of work, wages etc.). But at the same time, if subjective indicators of job quality are used (based on workers’ declarations), a “perception” effect can also be at play: when more and more workers are unemployed and the situation on the labour market is worsening, people who still have a job may evaluate their job quality in a more favourable way. When aggregate measures are used, it is impossible to disentangle between these two effects…

Another reason why it is important to disaggregate the analysis of job quality trends is the heterogeneity of job quality across social groups: women, youth, low educated are usually more concerned by bad jobs and such inequalities might be even higher in bad times.

What are the results of analyses looking at job quality trends at the individual level? Previous research from Quinne researchers, analysing the evolution of job quality at the beginning of the crisis (between 2007 and 2009), using EU-SILC data, has shown that some socio-economic groups are more affected by decreasing trends in job quality (other things being equal), especially youth, older workers and low-educated workers. Women seem less affected by these negative trends than men but are more likely than men to become unemployed or inactive over the period. Cross-country heterogeneity in job quality trends can be related to economic trends (the size of unemployment variation) and, to a minor extent, to the employment distribution by sectors. Some labour market institutions also seem to play a role in explaining the evolution of job quality in times of crisis: employment protection legislation (as defined by the OECD) prevents individual transitions to non-employment (and has no direct effect on job quality) while public expenditure per unemployed slightly reduces the risk of job quality deterioration.

See
ERHEL C., GUERGOAT-LARIVIERE M., LESCHKE J.,WATT A. (2012), “Trends in job quality during the Great Recession: a comparative approach for the EU”, Document de travail du CEE, november 2012, n°161

The coming of the Robots: what implications for job quality?

By Rafa²

With the advancement of artificial intelligence, cloud learning and robot engineering there is a growing public concern, shared by social scientist and policy makers alike [1], about the employment implications of the new developments in computerization and robotization of activities so far performed by workers.

In 2013, Carl B. Frey and Michael A. Osborne, of Oxford University[2], published a paper aiming at estimating how susceptible different jobs may be to computerization. According to their estimates around 47% of total US employment was at risk. The question we would like to rise in this blog is what kind of jobs, in terms of job quality, are more susceptible of automatization. In order to do so, we have related the probability of computerization as estimated by Frey and Osborne with the Job Quality Index of such occupations as measured by Muñoz de Bustillo et al. (2011) at the European level[3]. Due to the different occupation classification used in the US and EU, our analysis has been limited to 39 occupations[4].

Figure 1 reproduces the relation between the risk of computerization (or risk of automation) and the JQI of the occupation in 2010. The figure speaks for itself. There is a clear inverse association between the probability of computerization and job quality, i.e. those jobs with higher risk of being in the future performed by machines are characterized by lower levels of job quality.

To give a flavor of the results, the occupation with lowest probability of computerization, health professionals (0.022) has a relatively high JQI (0.59); the same can be said of chief executives, senior officials and legislators, with a JQI of 0.69 and a relatively low risk of computerization of 0.087. On the other end we find occupations such as Food preparation assistants (high risk: 0.860, low job quality: 0.27) or cleaners and helpers (high risk: 0.605, low job quality: 0.325).

Therefore, this quick inspection suggests that statistically speaking, robotization will have more impact in low quality jobs. Thus, it can be argued that it will contribute to the process of freeing human beings from the drudgery of hard and unpleasant work (as it has done in the past, i.e. of the legions of workers that cleaned the streets of horse manure before the arrival of cars[5]). The figure also shows, however, that mid and high quality jobs can also be affected, albeit to a much lower extent.

Figure 1. Probability of computerization and IJQ in 39 occupations.

Source: Author´s analysis from Frey and Osborne (2013) and EWCS microdata.

A final caveat: the results presented above refer only to the types of job potentially destroyed by computerization, without taking into consideration the jobs that will most likely be created in the process of building the machines that will substitute worker (generally high quality jobs) or other jobs created indirectly by the expanding consumption capacity of society. That’s a complex issue that we leave for another blog piece.

Rafa²is formed by Rafael Muñoz de Bustillo and Rafael Grande, University of Salamanca. Spain

[1] See, for example, Jeremy Rifkin´s the End of Work, Andrew McAfee and Erik Brynjolfsson, The Second Machine Age, or the Rise of the Robots by Martin Ford.

[2] Carl B. Frey and Michael A. Osborne (2013): [3] Muñoz de Bustillo, R., Fernández-Macías, E., Antón, J. I., and Esteve, F. (2011): Measuring more than money. The social economics of job quality, Chelthenham: Edward Elgar.

[4] The authors want to thank Martina Bisello and Enrique Fernández-Macías (of [5] On this issue see Joel Tarr and Clay McShane (1997): “

Quinne's blog

Can we measure the evolution of job quality during the Great Recession and what are main trends?