CAF - Regulating Platform Work where Informality Matters - Evidence from Chile de 2026

First version: July, 2026

This version: September, 2026

Small sections of text that are less than two paragraphs may be quoted without explicit permission as long as this document is acknowledged. Findings, interpretations and conclusions expressed in this publication are the sole responsibility of its author(s) and cannot be, in any way, attributed to CAF , its Executive Directors or the countries they represent. CAF does not guarantee the accuracy of the data included in this publication and is not, in any way, responsible for any consequences resulting from its use.

C A F - W O R K I N G P A P E R # 2 0 2 6 / 12

Regulating Platform Work where Informality Matters: Evidence from Chile

Omar Bamieh1,5 | Juan J. Dolado2,4,5 | Álvaro Jáñez3,6 | Felix Wellschmied2,5

1Universidad Andrés Bello. 2Universidad Carlos III de Madrid. 3Stockholm School of Economics.

Omar Bamieh: omar.bamieh@unab.cl

Juan. J. Dolado: dolado@eco.uc3m.es

4CEPR. 5IZA. 6Rockwool Foundation Berlin.

Álvaro Jáñez: alvaro.Janez.Garcia@hhs.se

Felix Wellschmied: fwellsch@eco.uc3m.esC A F - W O R K I N G PA P E R # 2 0 2 6 / 1 2

F i r s t v e r s i o n : J u l y 2 0 2 6

T h i s v e r s i o n : S e p t e m b e r 1 1 , 2 0 2 6

Regulating Platform Work where Informality Matters: Evidence from Chile Omar Bamieh1,5 Juan J. Dolado2,4,5 Álvaro Jáñez3,6 Felix Wellschmied2,5 Abstract Chile’s 2022 Platform Work Law mandates firms to monitor their subcontractors’ formality and establishes maximum working hours, minimum pay, and the right to collective bargaining among these independent contractors. Unlike other countries where informality matters, Chile further stands out for providing a comprehensive dataset for platform work that tracks workers’ labor market outcomes over 2020-2024. By means of DiD and event study designs, we find that this reform induced an 8-percentage-point shift from the share of informal to formal subcontractors without altering the share of employees. Yet, we find no significant changes in collective bargaining coverage, compliance with minimum pay and maximum hours, and wage differentials between employees and subcontractors. We argue that these changes in hiring practices are consistent with a structural labor market model where firms hire employees and both formal and informal subcontractors. The calibrated model also shows that the reform reduced platforms’ profits, while strictly enforcing hours regulations would lead to very small welfare losses.

JEL codes: J21, J60

Keywords: Platforms, Subcontractors, Employees

1Universidad Andres Bello. 2Universidad Carlos III de Madrid. 3Stockholm School of Economics. 4CEPR. 5IZA. 6Rockwool Foundation Berlin. Small sections of text that are less than two paragraphs may be quoted without explicit permission as long as this document is acknowledged. Findings, interpretations and conclusions expressed in this publication are

the sole responsibility of its author(s) and cannot be, in any way, attributed to CAF, its Executive Directors or the countries they represent. CAF does not guarantee the accuracy of the data included in this publication and is not, in any way, responsible for any consequences resulting from its use. ©2022 Corporación Andina de Fomento Corresponding author: Álvaro Jáñez (alvaro.janez.garcia@hhs.se).C A F - D O C U M E N T O D E T R A B A J O # 2 0 2 6 / 1 2

E s t a v e r s i ó n : 1 1 d e s e p t i e m b r e d e 2 0 2 6

Regulación del trabajo de plataformas donde la informalidad importa: evidencia de Chile Omar Bamieh1,5 Juan J. Dolado2,4,5 Álvaro Jáñez3,6 Felix Wellschmied2,5 Abstract La Ley de Trabajo en Plataformas de Chile de 2022 obliga a las empresas a supervisar la formalidad de sus subcontratistas y establece jornadas de trabajo máximas, una remuneración mínima y el derecho a la negociación colectiva para dichos subcontratistas. A diferencia de otros países donde la informalidad es relevante, Chile destaca por disponer de una base de datos que documenta la situación laboral de los trabajadores de plataformas durante 2020-2024. Mediante especificaciones de diferencias en diferencias (DiD) y event studies, encontramos que la reforma desplazó en 8 puntos porcentuales la proporción de

subcontratistas desde la informalidad hacia la formalidad, sin alterar la proporción de asalariados. No encontramos cambios significativos en la cobertura de la negociación colectiva, el cumplimiento de la remuneración mínima y de la jornada máxima de trabajo, ni en la brecha salarial entre asalariados y subcontratistas. Estos cambios en las prácticas de contratación son consistentes con un modelo estructural del mercado laboral en el que las empresas contratan asalariados, así como subcontratistas formales e informales. El modelo calibrado muestra que la reforma redujo los beneficios de las plataformas, mientras que aplicar estrictamente la jornada máxima de trabajo generaría pérdidas de bienestar de escasa magnitud.

JEL codes: J21, J60

Keywords: Plataformas, subcontratistas, empleados

1Universidad Andres Bello. 2Universidad Carlos III de Madrid. 3Stockholm School of Economics. 4CEPR. 5IZA. 6Rockwool Foundation Berlin. Pequeñas secciones del texto, menores a dos párrafos, pueden ser citadas sin autorización explícita siempre que se cite el presente documento. Los resultados, interpretaciones y conclusiones expresados en esta publicación son de exclusiva responsabilidad de su(s) autor(es), y de ninguna manera pueden ser atribuidos a CAF, a los miembros de su Directorio Ejecutivo o a los países que ellos representan. CAF no garantiza la exactitud de los datos incluidos en esta publicación y no se hace responsable en ningún aspecto de las consecuencias que resulten de su utilización.

©2022 Corporación Andina de FomentoBAMIEH ET AL. 1 1|INTRODUCTION Around 300 million workers are currently employed through digital platforms globally, accounting for between 4.4 and 12.5 percent (%) of the workforce (Datta, Rong, Singh, Stinshoff, Iacob, Nigatu, Nxumalo and Klimaviciute, 2023). Among platform work, delivery and transportation services (transportation, hereafter) take a dominant position, partially as a result of households’ changes in behavior during the COVID-19 pandemic. A key feature of this sector is the intensive use of independent contractors. This has brought about a radical transformation in labor market relationships, as it allows platforms to sidestep traditional employment regulations and social protection (Katz and Krueger, 2019; Abraham, Haltiwanger, Hou, Sandusky and Spletzer, 2021). These common practices have raised deep concerns about the precarious conditions of many platform workers, prompting regulatory responses across continents.1 These concerns are further amplified in Latin America, where independent contractors are not only the dominant form of employment but informality is also pervasive among platform workers (see ILO, 2025), thereby leaving many of them with even less social protection. In this region—where the gig economy employs about 46 million people—Chile becomes a key case study due to its approval of one of the earliest regulations of platform work in 2022.2 Moreover, Chile stands out for the availability of uniquely detailed and comprehensive survey data on the labor market conditions of platform workers, enabling the evaluation of the Chilean Platform Work Law (Ley N° 21,431, hereafter PWL). The PWL explicitly introduces the new jurisdictional figure ofindependent platform worker in its Labor Code, establishing that platforms could no longer rely on informal independent contractors. Instead, even when hiring through subcontracting firms, they have to monitor

that their workers are registered with the tax authorities. In addition, it establishes maximum working hours, minimum pay, and the right to collective bargaining among independent platform workers. We quantify the labor market effects of this reform on transportation workers, such as Uber drivers, who represent about 84% of platform workers affected by the PWL. To that end, we combine reduced-form methods with an equilibrium structural model of the platform work market. As for data sources, we exploit the Chilean Labor Force Survey (LFS), which explicitly identifies platform work since 2020. It provides microdata on working hours, wages, employment status, formal status, and social security coverage for a nationally representative sample of the working-age population on a quarterly basis. Reflecting the widespread practice of platforms circumventing labor regulations, we show that the informality rate of transportation platform workers reached 85% before the reform, with an additional 10% of workers being formally registered as independent contractors, and the remaining 5% being employees. Accordingly, most platform workers did not have social security coverage, many earned hourly income below the legal minimum pay, one-tenth worked more hours than the legal maximum applicable to employees, and less than 0.2% participated in trade unions before the PWL. As for hours worked and wages, compared to formal platform employees, formal and informal independent contractors both work substantially longer 1For example, Australia (Fair Work Legislation Amendment Act 2024), Belgium (Wet van 28 November 2022), California (Assembly Bill 5), Colombia (Ley 2466 from 2025), France (Loi n° 2016-1088), Italy (Legge 2 November 2019, n. 128), Massachusetts (Ballot Question 3), South Korea (Employment Insurance Act No. 17859), or Spain (Real Decreto-ley 9/2021). 2Mexico (Ley Federal del Trabajo 2024) and Uruguay (Ley N° 20,396) have also enacted legislation regulating

platform work, while similar initiatives are underway in Brazil, Colombia, and Peru.BAMIEH ET AL. 2 average hours, their distributions of hours worked are more dispersed, and their mean wages are lower. We evaluate the labor market effects of the PWL using static difference-in-differences (DiD) designs that compare outcomes of treated and control workers before and after the enactment of the reform. The treated group includes low-skilled workers employed by transportation platform companies covered by the PWL. The control group consists of low-skilled, non-platform workers in that sector. The results from these DiD regressions show that, relative to the control group, the informality rate falls by 8.2 percentage points (pp.) among platform workers following the reform, which amounts to a drop of 10% in the informality rate in this sector. Hence, informality remains the dominant status among workers. The reduction stems entirely from a substitution of informal for formal contractors. The share of the latter increases by 8.0 pp., while the share of formal employees exhibits no significant changes. We also find no significant changes in wages or hours worked following the PWL. These results are robust to an event-study design, which, reassuringly, does not find significant pre-reform trends. The results are also robust to accounting for changes in the composition of workers employed in platform companies. Finally, following Deaton (1985), we apply a pseudo-panel approach that further corroborates our main findings. The finding that share of formal employees and relative wages between employees and contractors remain unchanged may sound counterintuitive at first glance, since the reporting requirements increase the average costs of contractors. Nevertheless, using an equilibrium labor-market model, we show that this outcome is what one should expect as long as platforms only partially comply with the law. In the model, platforms hire employees as well as formal and informal contractors. Hiring the latter is cheaper in terms

of labor costs; however, it comes at the cost of facing sanctions, which become tougher with the reform. The model allows us to back out this expected cost rise, which reaches 10%, reducing platforms’ expected profits. The insight for why platforms do not hire more formal employees is that, in equilibrium, they trade off the marginal costs of employees and formal contractors, neither of which is directly affected by the fines. One way the government can affect this marginal costs trade-off is by lowering the platforms’ social security contributions for employees. Consequently, we simulate a payroll tax exemption for platform employees, which corresponds with a decline of 4.4 pp. in employer payroll taxes. The reform increases the wages of employees by 3.8% and the share of employees from 5.0% to 5.2%. Hence, the scope for reducing informality through employer payroll tax cuts is rather limited. Turning to the labor law enforcement of the reform, the reduced-form evidence is that it had no effect on the share of workers exceeding maximum working hours, the share of workers earning below the minimum wage, or on trade-union participation. However, there is a rise of 4.5 pp. in social security coverage.3 We use the structural model to ask how the economy would have changed, had the government actually implemented the labor law regulation. Enforcing minimum pay would raise the relative costs of formal contractors vis-à-vis informal contractors, thus increasing 3The incomplete pass-through of formality to this coverage rate stems from the fact that, despite some workers registering with the tax authorities, and thus having a formal status, they do not contribute to the social security system.BAMIEH ET AL. 3 demand for the latter. To evaluate the effects of enforcing the maximum hours regulation of 84 hours per week, we compute the counterfactual wage subsidy that the government would have to pay to make workers indifferent to not enforcing the law. We find small

welfare losses from enforcing the PWL: the required wage subsidy is 0.07%. This small effect results from only a few contractors working longer than 84 hours. Related literature and outlineThis paper contributes to a growing literature on the characteristics of the online gig economy. One body of empirical research deals with the difficulty of measuring platform work arrangements through either conventional administrative data, specific surveys, or field experiments (Mas and Pallais, 2017; Collins, Garin, Jackson, Koustas and Payne, 2019; Katz and Krueger, 2019; Boeri, Giupponi, Krueger and Machin, 2020; Abraham, Haltiwanger, Hou, Sandusky and Spletzer, 2021). Given that Chile is one of the countries in the world with the best administrative data on platform work, our study aims at reducing measurement problems. Similar to us, there is a recent literature that highlights the high value that digital platforms provide to workers through flexible work arrangements. For example, Chen, Rossi, Chevalier and Oehlsen (2019) study the case of Uber drivers in the U.S. and show that their surplus is higher in these flexible jobs than in alternative work arrangements, which offer less flexibility. Similarly, Stanton and Thomas (2025) analyze a platform environment where buyers post one-time projects and workers compete for these projects by posting wages. They find that imposing traditional employment regulations on short-term platform work reduces both overall welfare and workers’ surplus for those valuing flexibility. Angrist, Caldwell and Hall (2021) study a yet different setup where drivers choose between permanent lease contracts (taxi driver) and revenue fees (Uber driver), finding that the worker retains work flexibility in both arrangements. Finally, Dolado, Jáñez and Wellschmied

(2025a) find that food delivery drivers place great value on the flexibility of their working hours. We add to this literature by considering an environment where platform work drives high informality rates. Another related line of research is the study of the labor market effects of on-call and zero-hours contracts by Scarfe (2019), Dolado, Lalé and Turon (2025b), and Datta (2024), which share similarities with contracts offered to independently contracted transportation workers. Under those work arrangements, firms only call-up workers when they need their services, leading to workers being operative in some periods but not in others. Just like us, these papers employ structural models to analyze the general equilibrium effects of regulating flexible contracts. However, we differ from their approach in that we add informality as an additional status to being an employee and a formal contractor and in analyzing in more detail workers’ heterogeneity in time availability. Moreover, these authors use search and matching models, whereas ours is about a frictionless labor market, which we deem more appropriate for the type of platform work under consideration. Additionally, our paper is linked to the literature on the informal economy. Ulyssea (2020) concludes that formality enforcement generates efficiency gains without necessarily increasing unemployment. Our results show that increasing the formality rate in the platform sector does not necessarily translate into output or employment gains. Moreover, our approach echoes the equilibrium models of the labor market with informal jobs (Zenou, 2008; Satchi and Temple, 2009; Albrecht, Navarro and Vroman, 2009), but differs in that we characterize a different job sorting mechanism based on contract preferences.BAMIEH ET AL. 4 Finally, our paper also speaks to other strands of the literature that deal with the effects of changing legal work-time regulations (Carry, 2022), and the modeling of hours of work

in search and matching models (Cooper, Meyer and Schott, 2017; Frazier, 2018). We depart from these works in allowing for heterogeneity regarding labor supply decisions by workers in a frictionless model and by introducing an informal labor-market segment. The rest of the paper is structured as follows. Section 2 provides some institutional background about platform work in Chile, as well as a detailed discussion of the 2022 reform under consideration. Section 3 describes the data sources and draws stylized facts about the platform workers ’ labor market. Section 4 provides reduced-form evidence about the labor market effects of the reform by means of D-i-D and event-study regressions. Section 5 lays out the quantitative model. Section 6 describes its calibration. Section 7 discusses the counterfactual results from policy experiments. Finally, Section 8 concludes. An Appendix gathers further evidence discussed in the main text. 2|INSTITUTIONAL BACKGROUND Pre-reform institutions.Over the last fifteen years—and with a marked acceleration since the pandemic episode—gig economy platforms have expanded significantly in Chile, primarily through multinational companies such as Uber, Rappi, and PedidosYa. In fact, for the dominant player, Uber, the Chilean market trails only the US and the UK, with operations in nearly three-fourths of Chile’s medium and large cities (Azuara-Herrera, Keller and González, 2019). This rapid rise in platform-based employment has led to growing concerns about how to regulate these jobs. Broadly speaking, Chile’s labor relations system is highly decentralized. This structure goes back to the so-called “Labour Plan” enacted in 1979 under the Pinochet dictatorship, which confined collective bargaining to the firm level or even lower levels (e.g., at the workplace or shop floor). Although the democratic transition expanded workers’ social rights, collective bargaining has remained largely atomized, helping employers retain a

high degree of discretion over employment matters. Regarding these employment decisions, prior to the reform, Chile maintained a dual system classifying workers either as dependent employees, subject to labor law regulation, or as independent contractors, enjoying far fewer labor law protections. To ensure tax compliance and social security coverage among independent contractors, they are required to issue so-calledBoletas de Honorarios(freelance service invoices), which were introduced in

1984. These allow independent workers to bill for their services without a formal contract with the hiring company, while ensuring tax collection. The system has been widely criticized as a form of bogus self-employment, since it enables companies to avoid formal employment relationships, given that existing jurisprudence lacks a uniform criterion about the legal definition of independent contractors (Muñoz-García, 2018).

Proving a classical subordinate employment relationship in court is yet more difficult for platform workers. Hence, platform companies classified 95% of their workforce as independent contractors rather than employees. As a result, the vast majority of platform workers were subject to minimal regulation regarding working hours and remuneration. Moreover, though independent contractors are required to be registered with the tax authorities and thus formalized, in practice, this often did not occur. Consequently, prior to the PWL, theBAMIEH ET AL. 5 informality rate in the sector reached 80%—approximately three times higher than in the Chilean economy as a whole (26%). The Platform Work Law.The high degree of informality and the lack of social protection have sparked an intense debate on how to improve the rights of platform workers while preserving the flexibility offered by these new forms of employment. As stated by the Minister of Labor in parliamentary debates preceding the law, there was a need to establish "regulation for a sector that was completely unregulated" and to "guarantee worker protections while also allowing them to harness the potential made available by globalization

and new technologies" (Biblioteca del Congreso Nacional de Chile, 2022). Law 21.431, known as the Platform Work Law, approved in March 2022 and in force since September 2022, became the first law in Latin America to establish a regulatory framework governing “work performed via digital service platforms” and, in doing so, addressing the issue of informality.4 Specifically, the law applies to companies thatmanagethe execution of services by workers for users through a digital platform, explicitly including sectors such as delivery and transport (e.g., Uber). By contrast, platforms that merely advertise others’ services without managing their execution are excluded from these regulations (e.g., Airbnb).5 Determination of employment status. To reduce the number of informal workers, a central part of the reform relied upon the amendment of the existing Labor Code through the introduction of the new jurisdictional figure of “independent platform worker.” Hence, companies may still hire eitherdependentplatform workers, whose status closely resembles that of traditional employees, or independentplatform workers, who operate as independent contractors. However, unlike traditional independent contractors, companies are required to ensure that their independent platform workers become formal. In other words, it is the platform’s obligation to verify that these contractors (i) have a valid tax identification number, (ii) are authorized to issue the aforementioned Boletas de Honorarios, and (iii) report the resulting revenue to the tax authorities. Importantly, platforms are also obliged to monitor the formality of independent platform workers employed by subcontractors, i.e., firms that hire contractors to work on the platform. Beyond these regulations aimed at formalizing independent contractors, the reform imposed additional social protections for both types of workers, which we discuss next. New regulations for dependent platform workers. Table 1 summarizes the main new regulations established in the reform distinguishing

by type of labor contract. Regarding dependent employment, the PWL introduces two main regulatory changes regarding maximum working time and minimum pay (García and Azócar, 2022). The first change relates to the definition of working time, which is expanded to include all hours during which a worker remains at the platform’s disposal, 4Note that a year later, the so-called Uber Law, approved in 2023 but still not implemented by early 2026 (therefore not affecting our reference period 2020-2024) regulates transportation apps in Chile by requiring drivers to hold a professional license, setting a maximum vehicle age (1–3 years) to improve safety, creating an official registry, and standardizing the service. The government decided to pause it due to technological problems and the risk of increasing unemployment. Moreover, it did not change the labor regulations established by the PWL. Thus, our results are immune to this law 5In this case, although the supplier of the physical service may depend on advertising platforms to access opportunities, these companies only connect supply and demand without setting trading rules, such as prices. More concretely, according to the Consejo Superior Laboral (2024), the following platform companies operating in Chile should comply with the PWL: Uber, Cabify, Beat, InDriver, Cornershop, Jumbo, Jumbo Shoppers, Rappi, Uber Eats, PedidosYa, Upgirl, Eback, Justo, Moova, Bossmap, Roadrunner, Shopper, and Delivery.BAMIEH ET AL. 6 TA B L E 1Main new regulations introduced by Platform Work Law Formal Informal Employee Contractor Panel A. Before the reform Hours worked Contractual, ¯hWorkers’ choice Workers’ choice Maximum working time 45 hours-week No No Minimum payw No No Social security taxes Employer Worker None Right to bargain collectively Yes No No Enforcement risk borne by None None Potentially none

Panel B. After the reform Hours worked Contractual, ¯hWorkers’ choice Workers’ choice Maximum working time 45 hours-week 84 hours-week No Minimum pay 1.2×w 1.2×w No Social security taxes Employer Worker None Right to bargain collectively Yes Yes No Enforcement risk borne by None None Firm Note: This table reports the main features of changes in the regulation of work arrangements following the Platform Work Law 21.431 by contract type. The source is Chapter X of Chile’s Labor Code. encompassing both active service and passive waiting periods. Notably, the sum of active and passive hours is subject to statutory maximum-hours regulations, while only active hours are remunerated. The second change concerns minimum pay, which is set at 20% above the monthly national minimum wage, being proportional to the number of active hours worked. This premium aims at compensating workers for passive working time. New regulations for independent platform workers. The PWL has introduced several regulatory changes for independent platform workers concerning maximum working time, minimum pay, and the right to collective bargaining. First, a maximum working schedule of 12 hours (within any 24-hour period) is established by enforcing a mandatory minimum disconnection time from the platform. Second, these workers are subject to the same minimum pay regulations as dependent platform workers. Third, independent platform workers have the right to unionize and bargain collectively with platforms. 3|DESCRIPTIVE FEATURES OF PLATFORM WORK Chile is one of the few countries worldwide that measures platform work in its official employment statistics. We begin by describing this dataset. Next, we document some salient features of the Chilean delivery and transport platform sector, focusing on employment, hours worked, and hourly income. 3.1|Data Our main data source is the Chilean Labor Force Survey (CLFS; orEncuesta Nacional de Empleo, ENE, in Spanish), a panel data survey conducted by the Chilean Statistical OfficeBAMIEH ET AL. 7

on a representative sample of the resident population aged 15 or older. The CLFS provides individual-level monthly information on employment status, hours worked, occupation, industry, social security coverage, and firm tax registration. Importantly, it is among the first nationally representative surveys to include a module on platform work since 2020. This module identifies the specific web platform or mobile application through which individuals work and whether platform work is their main job. We complement the CLFS data with the Income Supplement Survey (ISS; orEncuesta Suplementaria de Ingresos, ESI), an annual supplementary module that provides individuallevel information on monthly income during the last quarter of each year for a sub-sample of individuals observed in the CLFS. The income measure is the sum of total net earnings from dependent employment and total net income from self-employment, including business income and self-consumption. These income measures are net of personal income and social security taxes. We use this information to construct hourly income from the main job for each individual by dividing monthly income by the weekly ordinary hours worked multiplied by an adjustment factor of 4.35 weeks worked in a month. With this available information, we adopt the following definitions of workers’ labor market status. We defineplatform workersas individuals whose main job is with a company covered by the PWL, as listed in Section 2.6 To focus the analysis on transport and delivery workers within these platforms, we further restrict the sample to workers in low-skill ISCO08 occupations (major groups 4 through 9) in the ISIC transport and storage sector (group H). Following Statistics Chile’s standard definition, we classifyinformal workersas either dependent employees without employer-provided social security coverage or independent contractors whose activity is not registered with the tax authorities. Moreover, we consider that dependent employees have social security coverage if their employers pay health, unemployment, or pension contributions on their behalf.

We restrict the sample to working-age individuals (15–65) and focus on the period 2020q1-2024q4, the last available survey wave. Workers in agriculture and mining, as well as those involved in family businesses, are excluded. 3.2|Descriptive features Sectoral size and market structure. Prior to the PWL, the transport and delivery platf