Outsourcing as a ‘friction buster’ in developing labour markets

Editor’s note: For a broader synthesis of themes covered in this article, check out our VoxDevLit on Barriers to Search and Hiring in Urban Labour Markets.

In low‑ and middle‑income countries, labour markets are shaped by a pervasive problem: frictions. Many workers, particularly young men, are stuck in informal jobs, without contracts, social insurance, or clear pathways into stable work. Small firms struggle to recruit, screen, and manage staff, and matching the right worker to the right job remains a challenge. This raises a question: can specialised contractors act as markets for talent, overcoming these frictions and pulling outsiders into formal employment? 

In recent work, we use a major legal reform in Brazil to show how this can happen (Felix and Wong 2025). In 1993, a court ruling suddenly made it much safer for firms to outsource ‘non‑core’ services such as security. Focusing on security guards, which is a large, mostly formal, and tightly regulated occupation, we show that:

• Outsourcing expanded formal employment for guards by around 5%
• Much of this increase came from young men moving out of unemployment or informal work
• Average wages for similar workers barely changed
• But some older guards in well‑paying firms suffered large and lasting earnings losses when their jobs were outsourced

The central mechanism is that large, specialised security firms are better at finding, screening, and deploying workers than individual client firms. In a friction‑ridden labour market, this organisational advantage matters for economic efficiency and development. 

Brazil’s legal shock: Making outsourcing ‘safe’

Before the 1990s, Brazil’s labour courts were hostile to outsourcing. A 1986 decision (Súmula 256) essentially stated outsourcing was illegal, with a handful of exceptions. Crucially, however, judges did not apply this rule uniformly. In the South and interior of São Paulo, courts often treated outsourcing as an attempt to dodge labour protections, and made client firms pay heavy penalties. Elsewhere, courts were more permissive, and outsourcing was rarely punished. The result was that in some parts of Brazil, outsourcing security was common; in others, legal risk kept it rare.

On 17 December 1993, the Superior Labour Court issued a new ruling (Súmula 331) that changed this overnight. It:

• Legalised outsourcing of all ‘non‑core’ activities (such as security and cleaning)
• Confirmed that outsourced workers were employees of the contractor, not the client
• Made the client firm financially liable only if the contractor defaulted

Importantly, this did not weaken labour law. Contractors and client firms still had to follow the same labour code. The reform changed who could employ workers, and how legally risky outsourcing was, not the rights workers had. For security services – a classic non‑core activity for most firms – this clarity made outsourcing suddenly much more attractive, especially where courts had previously been strict.

Security guards as a case study for outsourcing

Security is a key economic activity sector in Brazil and in other high-crime countries. In 1992, one in every 20 men who were formally employed by a private sector firm were employed as security guards. That’s 5% of formal male employment, not making it a niche labour market. 

Security guards are also an especially useful occupation for studying the role of outsourcing in a high-informality context for three reasons:

• Tight regulation: Guards must be licensed by the federal police, have no criminal record, and complete specific training. This limits unrecorded ‘off‑the‑books’ outsourcing and makes the occupation comparable across regions.
• Mostly formal jobs: About 80% of guards had formal contracts during this period (versus around 40% of workers overall), allowing us to observe who moved into and out of guard jobs.

A big, visible shift: After the 1993 ruling, outsourcing of guards rose sharply in previously restrictive regions.

Specialised contractors as ‘HR technology at scale’

In most developing economies, small and medium firms face a common constraint: weak HR capacity. They hire infrequently, cannot invest much in screening or training, and have little buffer when workers quit or underperform. We argue that specialised contractors can partially solve this.

Security contractors aggregate demand from many client firms. This scale allows them to run active recruitment campaigns, especially targeting young men; use fitness tests, psychological screening, and group interviews; build large pools of guards and maintain a ‘reserve bench’ to cover absences; and move workers between sites as clients’ needs change.

In fieldwork, we found that Brazilian security contractors do exactly this. Compared with individual client firms hiring a handful of guards, contractors are simply better equipped to find and manage people.

This matters for:

• Information frictions: Employers know little about young, inexperienced workers. Contractors reduce that uncertainty through screening and repeated interactions.
• Search frictions: Young men in informal jobs may not know how to access formal security work, or lack credentials. Contractors go out and find them.
• Coordination frictions: Small firms cannot offer stable careers; contractors can offer continuity by moving guards across client sites.

The 1993 reform did not create these capabilities, but allowed contractors to scale them up nationwide.

What changed after the reform?

More outsourcing, more formal jobs

In regions where courts had previously restricted outsourcing, the share of guards employed by contractors rose sharply after 1993. Security employment there became more concentrated in large firms employing hundreds of guards. Total formal employment in security increased by about 5% and stayed higher for years. Crucially, many of these additional jobs went to workers who were previously outside the formal sector, coming from unemployment, informal work, or non‑participation.

In a setting with widespread informality, this is a central result: making it easier for specialised firms to operate expanded a regulated, contract‑based occupation without changing formal protections.

A younger workforce

Following the reform, the profile of guards shifted noticeably:

• The average guard became younger
• Tenure in the job shortened
• Employment among men aged 18–24 rose significantly
• Employment among older guards declined

In other words, the reform reallocated who held security jobs. Contractors used their screening and recruitment advantage to draw in young workers who had previously been harder to reach or riskier to hire, displacing some older incumbents.

No big fall on average wages

Despite fears that outsourcing primarily cuts wages, average pay in the occupation did not collapse. Raw average wages dipped slightly, but that was mainly because younger workers earn less. When comparing similar workers (same age and education), wages were essentially flat. This contrasts with rich‑country evidence of big wage cuts after outsourcing. Two factors likely explain the difference:

1. Security guards are a medium‑wage, licensed occupation, not an unregulated low‑wage job
2. Contractors and client firms faced the same labour law, so outsourcing did not push workers into informal or ‘gig’ status

In this context, the main effect of outsourcing was to change who got the jobs and how they were matched, rather than slashing pay across the board.

The cost: Older guards in ‘good jobs’ lost out

The friction‑reducing role of contractors came with a clear downside: some older guards in well‑paying firms bore heavy costs. We track cases where firms that used to employ their own guards suddenly cut those positions and appear to switch to contractors. Guards hit by these ‘occupational layoffs’ experienced:

• A sharp drop in the chance of having a formal job in the year of the shock
• Wage losses that persisted: Five years later they still earned roughly 10–12% less than they would have otherwise
• Frequent exits from security into other occupations or into informality/non‑employment

Over time, the total earnings loss was large – equivalent to more than a full year of pre‑layoff pay. These losses were especially pronounced for guards in the best‑paying firms, which were more likely to outsource. So, while the occupation as a whole saw more jobs and stable average wages, some incumbents lost valuable ‘good jobs’ and could not fully recover.

Policy implications for labour markets in developing countries

The Brazilian case speaks directly to labour market policy in economies with high informality and weak HR capacity:

1. Specialised firms can reduce labour‑market frictions. When allowed to operate at scale, contractors can act as intermediaries that find, screen, and place workers more effectively than most individual firms. In Brazil, this translated into more formal jobs and better access for young men who were previously on the margins.
2. Flexibility can help outsiders, but hurt insiders. The same reforms that open doors for young, informal workers can close them for older workers in good positions. Ignoring these concentrated losses risks both unfairness and political backlash.
3. Institutions shape whether outsourcing cuts wages or cuts frictions. In Brazil, contractors and client firms shared the same labour code, and the occupation was licensed and regulated. In other settings – especially low‑wage, unregulated jobs or where outsourcing is used to disguise self‑employment – outsourcing may mostly cut wages, not frictions.
4. Design reforms to harness the HR advantage, not just lower costs. For policymakers, the message is not ‘outsource everything’, but to create space for specialised firms that can genuinely improve matching and training, while maintaining protections and supporting displaced workers.

As List (2022) emphasises in thinking about scaling policies, the details of context and implementation matter. Outsourcing under a clear, common regulatory umbrella, in occupations where quality and screening are important, can be a powerful tool to reduce labour‑market frictions in developing economies. But it will create winners and losers – and good policy should plan for both.

References

Caria, S, K Orkin, A Andrew, R Garlick, R Heath, and N Singh (2024), “Barriers to search and hiring in urban labour markets,” VoxDevLit, 10(1).

Dube, A, and E Kaplan (2010), “Does outsourcing reduce wages in the low-wage service occupations? Evidence from janitors and guards,” Industrial and Labor Relations Review, 63(2): 287–306.

Felix, M, and M Wong (2025), “Outsourcing, labor market frictions, and employment,” Unpublished manuscript.

Goldschmidt, D, and J Schmieder (2017), “The rise of domestic outsourcing and the evolution of the German wage structure,” Quarterly Journal of Economics, 132(3): 1165–1217.

List, J A (2022), The voltage effect: How to make good ideas great and great ideas scale, Currency.