How to define relevant labour markets?

The application to employers of competition rules and their conduct, including concentrations (see here) has been called the “new frontier for competition policies”. In the last 15 years, this so-called “labour antitrust” has grown rapidly, especially on the otherside of the Atlantic (see here). ‘All eyes are on labor’ in the United States even in 2024 (see here).

Competition between employers is receiving an increasing attention also on the old continent. We have seen enforcement in several European countries (see this). This summer, the European Commission reported that it had begun its first formal investigation into no poach agreements. The Commission also published an antitrust policy brief earlier this year.

Yet the revised Market Definition Note, which the Commission published in this year, does not include markets where firms compete for workers. It is a shame that the Commission did not include labour markets in its revised Market Definition Notice (Notice) published this year. This would have given credibility to their mission to intervene on these markets, and would have also provided guidance to other enforcers under EU competition law. In my latest working paper

, I reflect on this omission and discuss how to define relevant labor markets. The labour markets must be defined in cases where the conduct or concentration is being investigated concerns the purchasing of labor, i.e. employer conduct or employer concentrations.

There are good reasons to intervene against anti-competitive conduct and concentrations of employers. First, the exercise of market power by employers may have negative effects on downstream consumers (see this

). This will be the case when the exercise leads to a lower output of the undertakings in question, which is then not offset by their downstream competitors.

EU competition law should nevertheless defend workers against competitive harm even if consumers are not impacted negatively (see here). This can occur, for example, when undertakings are able reduce wages without reducing the amount of procured labor and, therefore, the produced output. The law ought to protect workers even in such situations not only because most people receive income primarily from work (see here

) but also because we tend to spend a good portion of our lives at work and hence care deeply about our working conditions (see here).To be sure, the relevant market does not need to be defined in all enforcement cases against employers. The Notice states that the Commission will not define the relevant market in cases where agreements restrict competition by their object. And the Commission considers this category to include also agreements to fix wages or not to poach each other’s workers (see here and here

).However, employers may also violate EU competition law by agreements that restrict competition by effect, by abusing their dominant position and by concentrations that significantly impede effective competition. The relevant labour market must then be determined. The Commission has, in fact, focused its attention primarily on agreements that restrict competition by object. This is demonstrated by the fact that the policy brief, which was mentioned above, was dedicated exclusively to these agreements. Other enforcers, however, have been more aggressive. For instance, the Dutch competition authority is looking

into how journalists will be impacted by a media concentration. Market Definition Notice and Labour Markets

The Notice does not say anything about labour markets. The only provision that is somewhat related to this is paragraph 7, where it states that the guidance in the Notice applies also to the definition of “purchasing markets”, i.e. The Notice acknowledges the existence of relevant labour markets, since employers purchase their employees on labour markets. Since labour markets are markets on which employers purchase labour, the Notice arguably recognises the existence of relevant labour markets.

Paragraph 7 further advises to define purchasing markets ‘taking into account their specificities’. However, such advice is of little value if these specifics are not stated. The only specificity of purchasing markets mentioned by the Notice is the obvious point that actual competitors of a buyer are given by whom the sellers consider as alternative buyers.

Labour markets are missing from the Notice despite express requests to include them. The European Trade Union Confederation criticized

the absence of labour markets as part of the stakeholder consultation 2020 on the previous notice. When the draft new Notice published for comments in 2022 did not change anything in this respect, the criticism was voiced

also by the Catalan Competition Authority.This is different from the United States, where labour markets are addressed by antitrust soft law. They were mentioned in the Merger Guidelines, published by the US Department of Justice jointly with the Federal Trade Commission towards the end of 2023. The guidelines discuss labour markets in a very brief manner, but they do have an impact on the definition. The relevant labour market is a group of jobs that workers can switch between with relative ease (see here

). The question then is with which other undertakings are these employers competing for the respective labour. The first dimension to be discussed is the occupation dimension. This corresponds to the product dimensions of product markets. As observed by the Notice, ‘in purchasing markets the substitution assessment focuses on alternatives available to suppliers, rather than alternatives available to customers.’ The occupation dimension is thus given by the alternatives available to workers.Which alternatives are available to suppliers is always constrained by the customers: suppliers can switch only to those customers who want to buy from them. The same is true for labour markets. The alternatives available to workers are determined by the employers’ hiring preferences. In this respect, labour markets are similar to markets that sell differentiated products. In these markets, buyers don’t see the goods as identical and instead distinguish between them. Consider the market for cars. “The buyer cares about identity, nature, features, and the product in question, the car” (see here

). The alternatives available to car suppliers are then very much contingent on the preferences of buyers.

Employers have preferences over various worker characteristics. Employers are most likely to need certain skills. In this regard, a position at another company will be in the same market as jobs in the undertaking in question if the workers who hold these positions have the necessary skills. Some employers have additional requirements that go beyond skills, such as matching personality traits with the culture of their workers (see here

). All these additional requirements imposed by employers limit the options available to workers, and thus narrow the scope of the labour market. workers. Rethink the car market. Car sellers don’t usually care who buys their car. Which alternatives are available to suppliers thus depends just on the preferences of buyers: whether a buyer is in competition with another one depends on whether the latter prefers the same car suppliers as the former (see here).

Workers, in contrast, do have preferences over various aspects of jobs (see here

). The financial compensation is a major consideration. The actual content of the work is also important. Workers may also have preferences regarding practical issues, such as shift flexibility and vacation time (see here

). However, they may also have a preference for a particular workplace culture (see this

) or even the identities of other workers (see this). Other relevant features of a job may include job security, job autonomy, job prestige, work intensity, participation in training, skill development, relationship with the employer, and potential adverse effects of work on private life (see here

).The preferences of workers further narrow down the scope of labour markets. The preferences of workers further narrow down the scope of labour markets. This dual limitation can lead to a limited number of employment options for a worker. This is why labour markets can be rather narrow in their substantive scope (see here).

Geographic dimension

The question which alternative jobs are available to workers has also a geographic dimension. A job is only an alternative if the worker is willing to commute to it or even relocate there (see here). This will vary between different groups of workers. “For example, younger workers might be more willing to travel farther or move from one place to another while older workers with children or married may be less mobile” (see here). Workers with higher levels of education tend to be more mobile (see here). As people are not very mobile, labour markets tend to be quite narrow. Some even argue that the geographical scope of labour markets are generally smaller than those of product markets. To define a market, you can ask what is the smallest group of products that a hypothetical monopsonist would find profitable to implement a non-transitory small but significant increase in price. Analogously, to define a labour market, it is possible to ask what the smallest set of jobs is over which a hypothetical single employer would find it profitable to implement a small but significant and non-transitory decrease in wages (SSNDW) (see here).

Labour markets and downstream product markets

One should not infer too much about the relevant labour market from information about the downstream product markets on which the respective undertaking is active. They can have a wide range of geographical scope. For instance, mines hire miners locally but distribute their products even globally (see here

)

In addition, two undertakings may compete as sellers but not as employers and vice versa. For the first possibility, it can be achieved when two undertakings produce the exact same output, but use different technology to do so, and therefore need workers with different skill sets. Consider, for example, different types of power stations. Or, the companies may need the same skills, but they are so far apart that their employees will not be willing to switch. This occurs when undertakings use the same workers to produce products that cannot be substituted. This can be illustrated by the US case concerning the e-commerce platform eBay and the financial software company Intuit, who were in competition for computer engineers (see here).The relevant labour market may hence correlate poorly with the industry to which the given undertaking belongs. Some workers, especially those with low skills, may change occupations. For example, some retail workers might start working in the food service industry (see here). The same occupation can be offered in different industries. For example, “an accountant may leave his accounting position at a manufacturing firm to join an accountancy firm” (see here

). That is to say that labour markets should be defined in their own right.

Conclusion

In principle, EU competition law is enforceable also against employers. The relevant legal provisions are sufficiently flexible, and we have seen some enforcement actions already. To make sure that interventions against employer market-power reach their full potential, it is important to recognize the possibility of such interventions across the entire EU competition law. This includes the definition of relevant labour markets. The Commission must therefore quickly find a means to declare that it is willing to define relevant labour markets. We are still a long way from this being true in Europe.

Story originally seen here