It’s not just Mentana: the real problem with non-competition clauses in employment contracts

Non-competition clauses – that is, agreements that prevent an employee from moving to a competitor or setting up a competing business for a certain period after the end of their employment – make the headlines when they involve well-known figures: there has recently been an exchange between Il Foglio and Enrico Mentana regarding whether or not a clause exists that would tie him to La7. This is not the first time. In the summer of 2023, there was a lengthy debate about the penalty clause imposed by Napoli on Luciano Spalletti, and in particular whether the national team could really be said to be ‘in competition’ with a club. However, these are stories involving superstars, who are in a position to negotiate such clauses.

The issue is much broader and affects ordinary workers. In the OECD’s Employment Outlook 2026, we published the first harmonised assessment across fifteen countries, following on from the work we carried out a few years ago with Tito Boeri and Lorenzo Luisetto, which focused solely on Italy: in our country, around one in six private-sector workers is bound by a non-competition clause, and such clauses have long since ceased to apply solely to researchers or sales managers. This is confirmed – bringing real-world validation to the rather dry statistical figures – by Roberto Benaglia, a long-serving executive at the CISL trade union, in the Diario del Lavoro, who points out that these agreements now extend to maintenance workers, IT specialists, sales staff and even warehouse workers. And this is not a phenomenon unique to Italy: across the OECD, between 14 and 35 per cent of those without a secondary school qualification report being covered by a non-competition clause. Moreover, the use of these clauses is on the rise everywhere: among companies that have changed their use of such clauses over the last five years, those that have increased their use are around five times as many as those that have reduced it.

A large proportion of these agreements would be null and void. The Civil Code requires adequate consideration and clearly defined limits in terms of subject matter, time and place; many clauses do not provide for any compensation, nor do they specify a sector or geographical area, and would not stand up in court. Yet they produce effects all the same. This is the point that should interest us most: a clause is binding not because it is valid, but because it is perceived as such. People act on the basis of what they believe to be true, and since the subject is hardly ever discussed, those who have signed assume they are bound by the agreement. They do not look for another job, they do not negotiate a pay rise, and often they do not even try.

The consequences are not limited to the individual. For most workers, the main channel for wage growth is changing employers: those who feel constrained forgo that lever. But the Employment Outlook also documents a broader effect. Where non-competition clauses are more widespread, slower productivity growth is observed through two channels: a reduced reallocation of labour from less productive firms to more productive ones, and a slower diffusion of knowledge between firms. These are statistical correlations, not proven causal links, but their consistency with mechanisms already recognised in the literature on productivity makes them difficult to ignore. The most dynamic firms grow less than they could, whilst the recovery of the others slows down. What may make sense for an individual firm – retaining staff who possess sensitive information – becomes a cost to the system when multiplied across millions of contracts. And for Italy, this is a luxury we simply cannot afford.

Leaving aside the cases of Mentana and Spalletti – who have found, or will find, a way to resolve their own issues – a reflection on the use of these clauses would be particularly useful for the rest of the workforce: on when they really make sense, and on how to prevent their misuse.