Outsourcing Ban Improved Wages and Benefits

This article by Manuel Gonzalez appeared in the June 29, 2025 edition of Sin Embargo.

The reform prohibiting outsourcing, or labor subcontracting in Mexico, one of the work arrangements that facilitated abuses by companies and employers, celebrated its 4th anniversary in April since it was approved by Congress, and its first fruits can already be harvested: workers’ salaries have increased and they have access to better benefits, social security and even higher wages for those who earn the least.

In April 2021, Mexico banned outsourcing, that is, the transfer of one’s own workers to another company and the prohibition of representing oneself as an employer without having any productive activity. Since then, companies have transferred and recognized— with exceptions —as their own workers all those workers who perform core activities related to their corporate purpose and main economic activity.

Subcontracting was maintained, but only for specialized services and works “that are not part of the corporate purpose or the predominant economic activity.” However, companies wishing to provide these specialized services must register with the Ministry of Labor and Social Welfare (STPS), with prior proof of compliance with labor, social security, and tax obligations.

In May of that year, the guidelines were published, and starting in November, companies had to have moved beyond the official ban. Every three years, companies providing specialized services must update their registration with the STPS.

In a study published by the National Bureau of Economic Research, a private, non-profit research organization in the United States, a group of experts concluded that, thanks to this ban on labor outsourcing in Mexico, “outsourcing rates plummeted, companies added previously outsourced workers to their payrolls, wages increased significantly (especially for lower-wage workers), and companies began making mandatory social security, benefit, and profit-sharing payments.”

“These gains did not come at the expense of employment or production, which remained stable, nor did they cause a change in the composition of inputs,” the study adds.

Companies Abused Their Power

The wage increases they observed “were larger among workers at the bottom of the distribution.” This, coupled with the fact that neither productivity, output, nor other factors changed, suggests that the reform “not only reduced wage inequality, but also returned bargaining power to workers.”

Moreover, pay increases were especially concentrated in firms that previously held the greatest power, those with the largest cuts, or those that paid workers far less than their marginal productivity.

But despite the increase in labor costs, the data show no reduction in the use of other factors or in productivity, experts say. “This supports the view that, before the reform, firms were able to contain wages through market power, rather than reflecting underlying economic constraints,” that is, an abuse by employers endorsed during the PRI-PAN administrations and the so-called neoliberal era.

Costs, But Many More Benefits

The reform, the study says, did have its costs. “We observed a small but significant decrease in capital investment and a slight increase in the likelihood of firms leaving the market,” it concludes.

These negative effects were most pronounced among smaller or lower-revenue firms, suggesting that some had relied on the cost savings from outsourcing to stay afloat. However, the magnitude of these disadvantages appears “modest,” experts say, compared to the significant improvements in wages and employment formalization.

This case of a nationwide ban demonstrates, they conclude, that labor market regulation, “when carefully targeted and effectively applied,” can significantly improve workers’ living conditions without harming employment, something that critics of the reform, such as the Employers’ Confederation of the Mexican Republic (Coparmex), predicted would put “millions of jobs” “in uncertainty” with this reform.

And one of the keys to this legislative change being effective, experts say, is clear legal definitions, rigorous oversight, and the administrative capacity to implement it.

But the study warns that there are also risks of going too far, such as “excessively rigid labor protections,” driving companies toward informal hiring or limiting job creation. “Finding the right balance is essential,” they stated, something that, based on the results shown, was achieved in the case of Mexico.

What it best demonstrates, they conclude, is that modest, well-designed interventions can succeed where previous, weaker regulations failed: “The formal inclusion of workers not only increases their pay, but also transforms the distribution of power and profits in the labor market.”

Banxico’s Diagnosis

In December 2024, the Bank of Mexico (Banxico) also analyzed the impact of the outsourcing ban in Mexico. The results show that workers who transitioned from outsourcing to direct-contract employment “were more likely to maintain their employment in the formal private sector compared to a counterfactual scenario without reform.”

That is, thanks to this hiring, cuts were avoided and formal jobs were also maintained, something the private sector had said would happen, but the opposite would happen if the reform were approved, as it was.

Furthermore, Banxico concluded, the wages of workers who transitioned to direct employment due to the reform increased between 3 and 4 percent, although with variations by gender, age, industry, company size, and region. “These findings highlight how labor regulations that restrict schemes such as outsourcing can influence job stability and wage growth,” it noted.

The ban on subcontracting also led to an average increase of approximately 3 percent in the wages of regularized workers. The study details that these wage increases were evident from the first months after implementation. The wage effects of the reform in companies that regularized workers in newly created employer registries were initially greater than in those with previous registries, although they converged after six months.