OIT - Mobile Internet and Employment

X Mobile Internet and Employment Evidence from 52 Countries Authors / Paloma Carrillo, Marta Golin, David Bescond, Avichal Mahajan, Rémi Viné

September / 2026 ILO Working Paper 180© International Labour Organization 2026 Attribution 4.0 International (CC BY 4.0) This work is licensed under the Creative Commons Attribution 4.0 International. See: https:// creativecommons.org/licenses/by/4.0/. The user is allowed to reuse, share (copy and redistribute), adapt (remix, transform and build upon the original work) as detailed in the licence. The user must clearly credit the ILO as the source of the material and indicate if changes were made to the original content. Use of the emblem, name and logo of the ILO is not permitted in connection with translations, adaptations or other derivative works. Attribution – The user must indicate if changes were made and must cite the work as follows: Carrillo, P ., Golin, M., Bescond, D., Mahajan, A., Viné, R. Mobile Internet and Employment: Evidence from 52 Countries. ILO Working Paper 180. Geneva: International Labour Office, 2026.© ILO. Translations – In case of a translation of this work, the following disclaimer must be added along with the attribution: This is a translation of a copyrighted work of the International Labour Organization (ILO). This translation has not been prepared, reviewed or endorsed by the ILO and should not be considered an official ILO translation. The ILO disclaims all responsibility for its content and accuracy. Responsibility rests solely with the author(s) of the translation. Adaptations – In case of an adaptation of this work, the following disclaimer must be added along with the attribution: This is an adaptation of a copyrighted work of the International Labour Organization (ILO). This adaptation has not been prepared, reviewed or endorsed by the ILO and should not be considered an official ILO adaptation. The ILO disclaims all responsibility for its content and accuracy. Responsibility rests solely with the author(s) of the adaptation.

Organization (ILO). This adaptation has not been prepared, reviewed or endorsed by the ILO and should not be considered an official ILO adaptation. The ILO disclaims all responsibility for its content and accuracy. Responsibility rests solely with the author(s) of the adaptation. Third-party materials – This Creative Commons licence does not apply to non-ILO copyright materials included in this publication. If the material is attributed to a third party, the user of such material is solely responsible for clearing the rights with the rights holder and for any claims of infringement. Any dispute arising under this licence that cannot be settled amicably shall be referred to arbitration in accordance with the Arbitration Rules of the United Nations Commission on International Trade Law (UNCITRAL). The parties shall be bound by any arbitration award rendered as a result of such arbitration as the final adjudication of such a dispute. For details on rights and licensing, contact: rights@ilo.org. For details on ILO publications and digital products, visit: www.ilo.org/publns.

ISBN 9789220437445 (print), ISBN 9789220437452 (web PDF), ISBN 9789220437469 (epub), ISBN 9789220437476 (html). ISSN 2708-3438 (print), ISSN 2708-3446 (digital) https://doi.org/10.54394/00034480

The designations employed in ILO publications, which are in conformity with United Nations practice, and the presentation of material therein do not imply the expression of any opinion whatsoever on the part of the ILO concerning the legal status of any country, area or territoryor of its authorities, or concerning the delimitation of its frontiers or boundaries. See: www.ilo. org/disclaimer. The opinions and views expressed in this publication are those of the author(s) and do not necessarily reflect the opinions, views or policies of the ILO. During the preparation of this work, the authors utilized ChatGPT Enterprise for language review and refinement. Following the use of this tool, the authors carefully reviewed and edited all content as necessary and assume full

essarily reflect the opinions, views or policies of the ILO. During the preparation of this work, the authors utilized ChatGPT Enterprise for language review and refinement. Following the use of this tool, the authors carefully reviewed and edited all content as necessary and assume full responsibility for the document/publication’s content. No material subject to intellectual property rights, including third party material, was used without acknowledgment and permission. Reference to names of firms and commercial products and processes does not imply their endorsement by the ILO, and any failure to mention a particular firm, commercial product or process is not a sign of disapproval. Information on ILO publications and digital products can be found at: www.ilo.org/researchand-publications ILO Working Papers summarize the results of ILO research in progress, and seek to stimulate discussion of a range of issues related to the world of work. Comments on this ILO Working Paper are welcome and can be sent to carrillo@ilo.org.

Authorization for publication: Caroline Fredrickson, Director, Research and Statistics Department ILO Working Papers can be found at: www.ilo.org/research-and-publications/working-papers Suggested citation: Carrillo, P ., Golin, M., Bescond, D., Mahajan, A., Viné, R. 2026. Mobile Internet and Employment:

Evidence from 52 Countries, ILO Working Paper 180 (Geneva, ILO). https://doi.org/10.54394/0003448001  ILO Working Paper 180 Abstract This paper examines how the expansion of 3G mobile internet affected labour force participation and employment composition across 52 countries, predominantly upper-middle-income and high-income countries, covering about 28 per cent of the global labour force, from 2004 to

2023. Using harmonized labour force survey microdata and high-resolution regional coverage maps, we employ fixed-effects OLS regressions and an instrumental-variable strategy exploiting regional variation in lightning density. Our main finding is that 3G expansion does not have uniform labour market effects: its consequences depend critically on where countries stand in

2023. Using harmonized labour force survey microdata and high-resolution regional coverage maps, we employ fixed-effects OLS regressions and an instrumental-variable strategy exploiting regional variation in lightning density. Our main finding is that 3G expansion does not have uniform labour market effects: its consequences depend critically on where countries stand in the digital diffusion process and their level of economic development. In lower-income and lower-coverage settings, expanded connectivity raises participation and employment, with stronger gains for women, while in more developed and digitally mature settings, the same expansion reduces participation through falling employment, consistent with diminishing returns at later stages of diffusion. Employment gains in poorer contexts occur largely through own-account and family work and agriculture, pointing to more vulnerable forms of employment, whereas richer contexts see shifts toward wage employment, larger firms, and services. Gains also vary by worker characteristics, concentrating among those with basic or intermediate education while employment contracts for the highly educated and older workers in more advanced economies.

Contrary to optimistic expectations, 3G expansion does not systematically narrow gender gaps; the most robust gendered finding is a widening of already large gender gaps in youth labour market disengagement, with young women in lower-income countries becoming relatively more likely to be neither employed nor in education or training. Overall, the results caution against viewing digital infrastructure as a technology with predictable labour market benefits, and suggest that connectivity policy must be embedded in a broader strategy addressing job quality, formalisation, and equitable access. About the authors Main authors Paloma Carrillo is an Economist in the Research and Statistics Department of the International Labour Organization (ILO), where she has worked since February 2022, focusing on global and regional trends in gender differences in labour market outcomes and working conditions. She holds a PhD in economics from the Toulouse School of Economics. Marta Golin is the Insights Lead, Work, Wages and Job Creation at the World Economic Forum. Before joining the Forum, she was an Economist in the Research and Statistics Department of the International Labour Organization. She holds a PhD in Economics from the University of

holds a PhD in economics from the Toulouse School of Economics. Marta Golin is the Insights Lead, Work, Wages and Job Creation at the World Economic Forum. Before joining the Forum, she was an Economist in the Research and Statistics Department of the International Labour Organization. She holds a PhD in Economics from the University of Oxford. Her research interests and work focus on the determinants and consequences of gender differences in labour market outcomes, and which policies or interventions may help narrow existing gender gaps.02  ILO Working Paper 180 Contributing authors David Bescond is a Data Scientist in the Research and Statistics Department of the International Labour Organization. His work focuses on developing and maintaining data collection and dissemination systems, and improving the quality, accessibility, and use of international labour statistics. Avichal Mahajan holds a PhD in economics from the University of Geneva and has been an Economist in the Research and Statistics Department of the International Labour Organization since July 2021. His work focuses on the production, maintenance, and analysis of gender-disaggregated ILO modelled estimates. Rémi Viné has been a Technical Officer in the Research and Statistics Department of the International Labour Organization since February 2025 and is a lecturer of statistics at the Geneva Graduate Institute. His work focuses on international migration and capital flows. He holds a PhD in international economics from the Geneva Graduate Institute.03  ILO Working Paper 180 Abstract   01 Abouttheauthors   01 Mainauthors   01 Contributingauthors   02 Acronyms   08 X Introduction   09 X 1 LiteratureReview   13 1.1Digitalaccessandlabourmarketoutcomes   13 1.2Genderimpactofdigitalaccessinlabouroutcomes   14 X 2 DataandVariables   16 2.1ILOHarmonizedMicrodatacollection   16

1.2Genderimpactofdigitalaccessinlabouroutcomes   14 X 2 DataandVariables   16 2.1ILOHarmonizedMicrodatacollection   16 2.2Mobilecoveragedata   17 2.3Sampleselection   18 2.4Variabledefinitions   19 Outcome variables from the ILO Harmonized Microdata collection   19 Variables from geocoded data   20 Summary statistics   21 X 3 EmpiricalStrategy   25 3.1Estimationapproach   25 3.2Identifyingassumptions   26 X 4 Results   28 4.1Labourforceparticipationandlabourforcecomposition   28 4.2Mechanisms   34 4.3Heterogeneitybyindividualcharacteristics   35 4.4Employmentcomposition   38 4.5Employmentquality   41 4.6Genderratios   43 Table of contents04  ILO Working Paper 180 X 5 RobustnessandLimitations   46 5.1Robustnesschecks   46 5.2Limitations   47 X Conclusion   49 Annex   51 References   67 Acknowledgements   7105  ILO Working Paper 180 List of Figures Figure1.3GcoverageinMexicoin2009,2014and2022   17 FigureA1.Regional3Gcoverageovertimebyfrequencyoflightningstrikes   51

List of Figures Figure1.3GcoverageinMexicoin2009,2014and2022   17 FigureA1.Regional3Gcoverageovertimebyfrequencyoflightningstrikes   51 FigureA2.First-stagecoefficientswhenexcludingcountry-incomegroups   52 FigureA3.First-stagecoefficientswhenexcludingindividualcountries   5206  ILO Working Paper 180 List of Tables Table1.Summaryofcountriesinsampleandrepresentativeness   19 Table 2. Summary statistics for earliest year in sample of main dependent variable and endogenousvariablebysex   22 Table3.OLSresultsfortheimpactofregional3Gcoverageonlabourmarketoutcomes   29 Table4.IVresultsfortheimpactofregional3Gcoverageonlabourmarketoutcomes   31 Table 5. OLS results for the impact of regional 3G coverage on labour market outcomes by initial3Gcoverage   32 Table 6. OLS results for the impact of regional 3G coverage on labour market outcomes by broadcountryincomegroup   33 Table 7. IV results for the impact of regional 3G coverage on labour market outcomes by age group   36 Table 8. IV results for the impact of regional 3G coverage on labour market outcomes by educationalattainment   37 Table 9. IV results for the impact of regional 3G coverage on employment status and sectoral composition   39 Table10.OLSresultsfortheimpactofregional3Gcoverageonemploymentbyfirmsize   40

Table 9. IV results for the impact of regional 3G coverage on employment status and sectoral composition   39 Table10.OLSresultsfortheimpactofregional3Gcoverageonemploymentbyfirmsize   40 Table 11A. IV and OLS results for the impact of regional 3G coverage on weekly hours actuallyworked   42 Table 11B. IV and OLS results for the impact of regional 3G coverage on the gender ratio of monthlyearnings   43 Table 12. IV results for the impact of regional 3G coverage on gender ratios in labour market outcomes   45 Table A1. List of sample countries in the IV regression and their available regions and year observations.   53 Table A2. OLS results for the impact of regional 3G coverage on labour market outcomes by agegroup   54 Table A3. OLS results for the impact of regional 3G coverage on labour market outcomes by agegroup–Upper-middle-incomeandhigh-incomecountries   55 Table A4. OLS results for the impact of regional 3G coverage on labour market outcomes by agegroup–Low-incomeandlower-middle-incomecountries   55 Table A5. OLS results for the impact of regional 3G coverage on labour market outcomes by educationalattainment   56 Table A6. OLS results for the impact of regional 3G coverage on labour market outcomes by educationalattainment-Upper-middle-incomeandhigh-incomecountries   56 Table A7. OLS results for the impact of regional 3G coverage on labour market outcomes by educationalattainment-Lower-middle-incomeandlow-incomecountries   57 Table A8. IV results for the impact of regional 3G coverage on labour market outcomes by maritalstatus   58

educationalattainment-Lower-middle-incomeandlow-incomecountries   57 Table A8. IV results for the impact of regional 3G coverage on labour market outcomes by maritalstatus   58 Table A9. OLS results for the impact of regional 3G coverage on labour market outcomes by maritalstatus   58 Table A10. OLS results for the impact of regional 3G coverage on employment status and sectoralcomposition   59 Table A11. OLS results for the impact of regional 3G coverage on employment status and sectoralcomposition–Upper-middle-incomeandhigh-incomecountries   5907  ILO Working Paper 180 Table A12. OLS results for the impact of regional 3G coverage on employment status and sectoralcomposition–Lower-middle-incomeandlow-incomecountries   60 TableA13.OLSresultsfortheimpactofregional3Gcoverageonemploymentbyfirmsize– Upper-middle-incomeandhigh-incomecountries   61 TableA14.OLSresultsfortheimpactofregional3Gcoverageonemploymentbyfirmsize– Low-incomeandlower-middle-incomecountries   61 Table A15. OLS results for the impact of regional 3G coverage on gender ratios in labour marketoutcomes   63 Table A16. OLS and IV results for the impact of regional 3G coverage on labour force participationandstatusonthepopulationofindividualsaged15-64   64 Table A17. OLS results for the impact of regional 3G coverage on labour force participation andstatus,removinghigh-incomecountriesfromthesample   64 Table A18. OLS results for the impact of regional 3G coverage on labour force participation

Table A17. OLS results for the impact of regional 3G coverage on labour force participation andstatus,removinghigh-incomecountriesfromthesample   64 Table A18. OLS results for the impact of regional 3G coverage on labour force participation andstatus,removingcountrieswithbaselinecoverageabove80percent   65 TableA19.ComparisonofANCOVAsampleandfullanalysissample   66 Table A20. ANCOVA results for the impact of regional 3G coverage on labour force participationandstatus   6608  ILO Working Paper 180 Acronyms EMP2POP Employment-to-population ratio GSMA Global System for Mobile Communications Association ICLS International Conference of Labour Statisticians ILO International Labour Organization IV Instrumental variable ILOEST ILO Modelled Estimates LFPR Labour force participation rate NEET Not in employment, education or training OLS Ordinary least squares UNE2POP Unemployment-to-population ratio09  ILO Working Paper 180 X Introduction The rapid expansion of mobile broadband technologies over the past two decades has profoundly transformed people’s daily lives and how they communicate, work and access economic opportunities. Today, 78 per cent of the global population owns a mobile phone (ITU, 2023), and nearly 67 per cent use the internet. Mobile broadband, particularly 3G networks, have been a key driver of this transformation. Unlike 2G networks, which enable voice calls and SMS but not meaningful internet access, 3G transformed mobile phones into internet access points, supporting a wide range of online activities, such as mobile banking, e-commerce, remote work, and digital learning. These advances can influence an individual's economic well-being in multiple ways, such as reducing information frictions, lowering search and transaction costs, enabling remote work and digital entrepreneurship, and allowing consumers and businesses to engage in more

wide range of online activities, such as mobile banking, e-commerce, remote work, and digital learning. These advances can influence an individual's economic well-being in multiple ways, such as reducing information frictions, lowering search and transaction costs, enabling remote work and digital entrepreneurship, and allowing consumers and businesses to engage in more competitive marketplaces (Jensen, 2007; Aker, 2010; Bhuller et al., 2023). Mobile broadband has been particularly transformative in lowand lower-middle-income countries, where fixed-line infrastructure, such as fixed telephone lines and cables, remains limited and mobile phones constitute the primary means of internet access (Stork et al., 2013; ITU, 2023). In 2023, worldwide 95 per cent of the population was covered by at least 3G or above, although there are large discrepancies across income groups. For example, in low-income countries the population coverage reaches only 77 per cent, whereas it is nearly universal in high-income contexts (ITU, 2023). Despite its rapid diffusion, the potential economic benefits of mobile internet may not fully materialize or may do so unevenly across genders. Gender gaps in mobile phone ownership and digital literacy continue to persist. In lowand lower-middle-income countries, women are 14 per cent less likely than men to use mobile internet, as well as 14 per cent less likely to own a smartphone (GSMA, 2025). These disparities suggest that the labour market impacts of 3G coverage may differ systematically by gender. Understanding these effects is particularly relevant given the persistence of global gender disparities in labour market outcomes. In 2025, the employment-to-population (EMP2POP) ratio of men stood at 69.5 per cent globally, compared with 46.3 per cent for women, a 23.2 percentage point gap, with even wider differences in many developing countries (ILOSTAT, 2025)1. The diffusion of digital technologies such as 3G could, in principle, help narrow these disparities by facilitating access to information and new work arrangements (Henderson et al., 2026), or conage point gap, with even wider differences in many developing countries (ILOSTAT, 2025)1. The diffusion of digital technologies such as 3G could, in principle, help narrow these disparities by facilitating access to information and new work arrangements (Henderson et al., 2026), or conversely, widen them if women face greater barriers to adoption and use. A growing body of research has examined the relationship between internet access and labour market outcomes. Evidence from lower-coverage and earlier-diffusion contexts generally documents positive effects on employment, job matching, and labour force participation, with studies from both developing countries and early-stage diffusion episodes in developed economies pointing to reduced information frictions and improved market access as key mechanisms (Aker and Fafchamps, 2015; Hjort and Poulsen, 2019; Chiplunkar and Goldberg, 2022; Kuhn and Mansour, 2014). However, these positive effects appear to diminish or reverse as coverage matures: evidence from more advance or higher-coverage settings suggest smaller employment gains, consistent with diminishing returns at later stages of diffusion (Kolko, 2012). Moreover, while some studies consider gender-disaggregated outcomes, relatively little attention has been devoted to 1 Figures based on ILO Modelled Estimates (ILOEST) using the 13th International Conference of Labour Statisticians (ICLS) definitions for employment.10  ILO Working Paper 180 how digital expansion affects gender gaps themselves or to whether the positive effects documented in developing contexts extend to more advanced economies. In this paper, we analyse the expansion of 3G mobile network coverage between 2004 and 2023 across 52 countries, predominantly highand upper-middle-income countries, representing approximately 28 per cent of the global labour force and covering all levels of economic development. Using harmonized labour force survey microdata from ILOSTAT combined with high-resolution regional coverage maps, we estimate the effects of 3G expansion on labour force participation, employmentand unemployment-to-population ratios, employment composition and sectoral

Using harmonized labour force survey microdata from ILOSTAT combined with high-resolution regional coverage maps, we estimate the effects of 3G expansion on labour force participation, employmentand unemployment-to-population ratios, employment composition and sectoral allocation. For all these outcomes, we further investigate whether improved digital access has contributed to widening or narrowing existing gender gaps. The cross-country scope, long time horizon, and consistent harmonization of labour market variables allow us to provide one of the most comprehensive assessments of the labour market consequences of mobile broadband diffusion to date. Identifying the causal impact of digital connectivity on labour market outcomes presents empirical challenges, as network rollout is typically correlated with economic development, infrastructure investment, and other region-specific factors. To address this concern, we first leverage ordinary least squares (OLS) regressions that include region and time fixed effects, as well as several variables that allow us to control for other trends in telecommunication technology diffusion and broader economic development. Second, to further tighten the causal interpretation of our results, we employ an instrumental-variable strategy that exploits regional variation in lightning strike frequency to predict 3G rollout, conditional on 2G coverage and other controls. Because lightning increases the cost of constructing and maintaining telecommunications infrastructure, it generates plausibly exogenous variation in 3G expansion across regions (Chiplunkar and Goldberg, 2022; Guriev et al., 2021; Adema et al., 2022; Jain and Stemper, 2026). We note that our instrumental variable strategy loses precision when the sample is split to examine heterogeneous impacts, as the first stage coefficient becomes statistically insignificant for some smaller subsamples. Therefore, we rely on OLS regressions for all discussion of heterogeneity in the impact of 3G expansion across country income groups or by baseline level of 3G coverage. Preview of findings Global results: Our findings show that the labour market effects of 3G expansion are highly context-dependent. In pooled OLS regressions, additional 3G coverage is associated with near-zero average effects on labour force participation, reflecting a small increase in the employment-to-popPreview of findings Global results: Our findings show that the labour market effects of 3G expansion are highly context-dependent. In pooled OLS regressions, additional 3G coverage is associated with near-zero average effects on labour force participation, reflecting a small increase in the employment-to-population ratio that is broadly offset by a decline in the unemployment-to-population ratio. By contrast, the instrumental-variable estimates indicate more negative average effects: a 10-percentage-point increase in 3G coverage reduces labour force participation and the employment-to-population ratio by 1.3 and 2.7 percentage points respectively, while increasing the unemployment-to-population ratio by 1.4 percentage points. In both specifications, coverage and labour market outcomes are measured in the same survey year, so these effects are contemporaneous by construction. These differences between OLS and instrumental variable (IV) estimates are informative. Our IV strategy identifies a local average treatment effect that appears to place more weight on upper-middleand high-income contexts, which entered our sample already at advanced stages of 3G diffusion, with average national coverage of 52 and 69 per cent respectively at the time of their earliest survey year, where the marginal returns to additional coverage expansion are therefore likely to be smaller or even negative.

Country-level heterogeneity: Looking beyond pooled averages, the results consistently point to strong heterogeneity by stage of digital diffusion and level of development. In countries with11  ILO Working Paper 180 low initial 3G coverage (i.e., those countries with coverage below the sample mean of 55.5 per cent when entering our sample), additional rollout is associated with higher labour force participation and employment, with particularly strong gains for women. By contrast, in countries with higher initial coverage, additional 3G expansion is associated with lower labour force participation, driven mainly by lower employment. A similar pattern emerges when splitting the sample by country income group. In lowand lower-middle-income countries, 3G rollout is associated with higher employment and lower unemployment, whereas in upper-middleand high-income

tion, driven mainly by lower employment. A similar pattern emerges when splitting the sample by country income group. In lowand lower-middle-income countries, 3G rollout is associated with higher employment and lower unemployment, whereas in upper-middleand high-income countries the average effects are mostly null in OLS, and the negative effects from the IV regressions are concentrated in the more advanced economies that dominate the first stage. Taken together, these results suggest that the labour market effects of 3G expansion depend critically on where countries stand in the digital diffusion process: early rollout appears more likely to relax binding information and access constraints, while later rollout may coincide with diminishing returns and labour market adjustment of a different nature.

Individual-level heterogeneity: The effects of 3G expansion also vary substantially across workers and across types of employment. Looking at heterogeneity by age, the negative IV effects are concentrated among older individuals, particularly those aged 55 and above, while younger individuals show more favourable participation responses. By education, gains in labour force participation and employment are concentrated among workers with basic or intermediate education, while both labour force participation rate (LFPR) and the employment-to-population ratio contract among individuals with advanced education. Finally, the estimates suggest somewhat more favourable employment effects for married women than for unmarried women.

Employment composition: We also show that 3G expansion affects not only whether people work, but also the type of work they do. In lower-income settings, additional coverage is associated with shifts away from employee status and toward own-account and, for women, contributing family work, pointing to a possible increase in vulnerable forms of employment. In more advanced contexts, the compositional shifts are more consistent with an increase in wage employment, reallocation out of agriculture and toward services and industry, and movement away from very small firms toward medium-sized firms. We further find that 3G expansion is not associated with significant changes in job quality, as measured by average weekly hours worked and the gender gap in monthly earnings among employees, for either gender, although we find some evidence that digital access may contribute to widening gender gaps in earnings in economically advanced

small firms toward medium-sized firms. We further find that 3G expansion is not associated with significant changes in job quality, as measured by average weekly hours worked and the gender gap in monthly earnings among employees, for either gender, although we find some evidence that digital access may contribute to widening gender gaps in earnings in economically advanced contexts,