The paper – which can be found here – seeks to quantify the level of labour market concentration across a wide range of occupations and for almost every commuting zone in the US. In a nutshell, it finds that labour market concentration is high, and that higher concentration is associated with significantly lower posted wages. Given current high concentration levels in labour markets, mergers have the potential to significantly increase labour market power. The authors argue that this type of analysis could be used by antitrust agencies to assess whether mergers can create anti-competitive effects in labour markets. With this, they seek to challenge how little attention antitrust regulators have devoted to labour markets, despite the labour economics literature finding that firms can have substantial market power in these markets.
The paper is structured as follows:
- Section 2 describes the data and how the paper will go about measuring labour market concentration. They use proprietary data from CareerBuilder.com, the largest online job board in the US, containing about 35% of the total number of employment vacancies. The authors use commuting zones to define geographic labour markets, and 6-digit SOC codes to define markets by occupational category. The data includes the number of applicants for each vacancy, which allows the authors to calculate labour market tightness. They estimate posted wages for a given market on the basis of advertised salaries. They rely on the Herfindahl-Hirschman Index as the baseline measure of concentration in a labour market. HHI is calculated on the basis of the share of vacancies of all the firms that post vacancies in that market.
- Section 3 analyses the relationship between labour market concentration and posted wages. The paper identifies a negative correlation between salaries and market concentration, consistent with standard oligopsony theory. In other words, higher labour market concentration is associated with significantly lower real wages. The authors find that the effect of an increase in market concentration on wages is expressed both directly through lower wages for each job title, and by an increase in the posting of lower-wage job titles. They also note that commuting zones around large cities tend to have lower levels of labour market concentration than smaller cities or rural areas.
- Section 4 performs robustness tests and addresses the limitations of the analysis, which I am sure will be of interest to those of us who are statistically inclined. Such controls include, among others, alternative market definitions, removing markets where the number of hiring firms is too low, and correcting for local averages.
- Section 5 concludes that there are significant disagreements about how competition agencies should deal with labour markets. The authors suggest that their results on the anti-competitive effects of concentration on the labour market could be important, and that the type of analysis they pursue could be used to incorporate labour market concentration concerns into merger control analysis.
Comment: In short, this paper sets scene: it raises an issue (lack of competition in labour markets), develops a method to identify the circumstances under which that issue will be problematic (the statistical analysis they pursue), and advances a high level suggestion on how to address it (incorporating their method into antitrust analysis, and taking into account of the effects of mergers on labour markets).
What the paper does not do is develop an argument on how competition law should take into account the effects of mergers on labour markets in greater detail. That is the topic of the papers reviewed in the two posts below.