Does expanding social protection schemes substantially increase informality? Whether there is such a trade-off is a key question for policymakers wanting to protect workers and their families from economic and health shocks but concerned not to generate negative side-effects.
The case for social protection is clear. There are global commitments such as the Sustainable Development Goals and the Global Partnership for Universal Social Protection (USP2030), and the experience with COVID-19 has confirmed the need to step up the pace. In 2020, 53 per cent of the global population – 4.1 billion people – were left without any form of social protection. Accessing social protection is particularly difficult for the world’s two billion workers in informal employment, who are often considered not poor enough for social assistance yet are excluded from social insurance based on formal employment. The scale of informal employment – which comprises 54 per cent of total employment in Latin America and the Caribbean – means that achieving universal social protection will require significant progress.
The potential trade-off between social protection and labor markets
The idea of a trade-off between social protection and formality is conceptually straightforward: Social protection systems that combine contributory social insurance for formal employees with non-contributory social assistance for low-income informal workers drive informality because contributions ‘tax’ formal employment, while non-contributory assistance ‘subsidizes’ informality.
From the employer perspective, this difference could provide incentives to evade labor and social security laws and lead firms to generate informal employment. From the worker’s perspective, the introduction of non-contributory social assistance benefits could make informal employment more acceptable. Self-employed workers may also choose informality to reduce taxes and contributions while still being able to access benefits.
This line of thinking was most clearly outlined in Santiago Levy’s 2008 book on Mexico’s social protection system, Good Intentions, Bad Outcomes: Social Policy, Informality, and Economic Growth in Mexico, and has been taken up in some policy guiding documents. UNDP’s 2021 Regional Human Development Report for Latin America declares that “social protection policies contribute to informality” because they “tax formality and subsidize informality” (p. 279) and the IMF’s report on The Global Informal Workforce: Priorities for Inclusive Growth claims that payroll taxes on formal sector workers “increase the cost of doing business and create double taxation of labor, thus encouraging informality” (p. 254).
There are some caveats to this hypothesized trade-off. First, the argument is based principally on a voluntarist view of informality that may over-emphasize workers’ ability to choose their employment status. It can also deflect attention from the need to reduce barriers to decent work.
Second, social protection systems specifically aim to reduce vulnerability and to improve the conditions of workers; e.g., consumption smoothing, insuring against economic and life-cycle risks, protecting against poverty, and redistributing between richer and poorer. Equitable financing requires both taxes and social contributions because workers in informal employment often have low contributory capacities.
What is the evidence for a trade-off?
The evidence on this trade-off is still limited. A 2022 review by Canelas & Niño-Zarazúa of more than a dozen empirical studies on the effects of social protection programs on formal and informal employment notes that just over half find increases in informality or decreases in formality, although almost always for specific sub-groups, such as older people.
Moreover, most empirical studies do not suggest significant employment gains as a result of cuts to contribution rates, although few studies analyze the impact on formal employment. In the case of Mexico’s Social Health Protection System (SPSS, by its acronym in Spanish) (one of the most studied programs in this discussion) almost all the evidence finds no effect on informal or formal jobs (Alonso-Ortiz & Leal, 2018; Campos-Vazquez & Knox, 2013; Azuara & Marinescu, 2013; Barros, 2009).
There are also methodological and data challenges. Almost all studies of the SPSS rely on surveys, even though Mexico does not have a representative municipal-level survey, the level where this scheme was sequentially adopted. The exception is a 2014 study by Bosch & Campos-Vazquez, which uses municipal-level administrative data from Mexico’s social security institute (Instituto Mexicano del Seguro Social, IMSS) to study Seguro Popular’s impact on formal jobs. This paper stands apart from the rest in finding that Seguro Popular decreased formal employment in small firms by four per cent, or about 17,000 formal jobs per year (they do not find any effect on the number of formal jobs on average).
New evidence on the effects on the labor market
Our study uses the roll-out of the SPSS across municipalities to quantitatively assess its impact on private sector formal employment in Mexico, using more detailed data and improved econometric methods compared to the earlier studies. Specifically, we examined whether the SPSS causally decreased the number of workers registered with IMSS; Whether workers affiliated with IMSS were more likely to leave their IMSS-registered job when the SPSS started operating in their municipality; and whether Seguro Popular increased salaries in the formal sector.
We found that the scheme did not lead to a statistically significant decrease on the number of formal jobs at the municipality level, nor on the number of workers in the private sector that quit their jobs. We found no effects in average salaries of jobs affiliated to the IMSS, further suggesting that there were no strong shifts in labor supply from the formal to the informal sector.
Overall then, the empirical evidence suggests that the expansion of the SPSS probably had small impacts on formality, possibly concentrated in specific populations and subsectors of the economy. Moreover, small incentives effects – whether they exist or not – should be contextualized within the well-documented benefits of social protection, as well as the human right to health protection.
nice article.thanks