Working Without Borders: The Promise and Peril of Online Gig Work CHAPTER 4 How Inclusive Is the Online Gig Economy? 4.1 INTRODUCTION By providing flexibility in location and time, reduced friction in matching customers and clients, and low entry barriers, online gig work provides opportunities for individuals who face constraints in accessing the local offline labor market. Women prefer flexible work arrangements to balance household responsibilities. Youth work on online gig platforms to try dif- ferent occupations and learn skills for future career development. People with disabilities and those in rural areas who face mobility barriers and have limited locally available job opportunities could get access to a broader job market through online platforms. Additionally, people use gig work to earn supplemental income. This chapter discusses how online gig workers compare with other workers in six aspects of inclusion (age, gender, skills, location, language, and employment and income patterns), using available data from the latest labor force and household surveys from the I2D2 database.92 This chapter examines the following: 1. Differences between online gig workers on local and global platforms, 2. Differences between online gig workers and • Workers in the services sector, • Informal workers, and • Workers with similar occupations, who were identified by matching the typical task categories found on online gig work platforms (including business and professional services, information technology (IT) and software development, and microtasks) to similar occupational codes (the mapping of occupational codes is provided in appendix G and has some limitations). 4.2 METHODOLOGY The analysis is based on data from several survey instruments: • Global RDIT survey in 17 countries. The primary data source for this analysis is a global RDIT web survey conducted by the team in 17 countries in six regions, using random domain intercept technology (RDIT; see appendix D). 92  he International Income Distribution Database (I2D2) developed by the World Bank is a collection of harmonized T household and labor force surveys (LFSs). 77 Chapter 4 How Inclusive Is the Online Gig Economy? • Ten platform-based surveys, including nine online freelancing and one microwork platform survey conducted between April and December 2022 (Table 4.1). All nine online freelance platforms were regional/local in nature. The surveys were conducted in collaboration with the nine freelancing platforms, relying on a variety of distribution channels, including emails sent by the platforms to gig workers and promotion of the survey on the platforms. The survey conducted in the microwork platform was posted as a task, and online gig workers were invited to complete the survey just as they would complete any other task. The number of responses across the surveyed platforms varied from fewer than 50 (in four platform surveys) to more than 700 (in four platform surveys, with the highest number for one survey being 3,600). The analysis used the platform surveys with high response rates (see appendix E for a detailed description of the platform surveys). • Five country-level deep dive surveys conducted in collaboration with World Bank country teams from Social Protection and Jobs (SPJ), Social Sustainability and Inclusion (SSI), and Digital Development (DD). The country deep dives were done in Bangladesh, Indonesia, Kosovo, Malaysia, and Pakistan. The team received platform data from Malaysia-based platform eRezeki (2016–20) and the GLOW PENJANA program93 (2020–21), provided by the Malaysia Digital Economy Corporation (MDEC) and analyzed with support of World Bank colleagues in Malaysia. In Indonesia, the study team collaborated with the SPJ team, who conducted a large survey of over 4,000 informal work- ers; the SPJ team also supported the effort with data analysis. In Pakistan, we worked with the SSI country team, which had implemented an operation in Khyber Pakhtunkhwa (KP) and was keen to roll out an end-of-operation survey. We worked with the team to conduct the survey. In Kosovo, we worked with the DD team to trace beneficiaries of a DD pilot on gig work. In Bangladesh, we worked with client counterparts in the Ministry of Information and Communications Technology (ICT) to roll out a small-scale survey of gig workers. See appendix E for further details. • Aggregate data from platforms provided by four online gig work platforms and projects. • Interviews with 28 platforms, including 24 regional/local platforms and 4 global platforms. Semistructured interviews were conducted with the founder, CEO, or senior management of each platform between summer 2021 and autumn 2022 (see also chapter 3 and appendix F for a detailed overview). • Focus group discussions with select gig workers. Focus group discussions were conducted with Kenyan online freelancers using the Onesha platform in December 2022 and with Pakistan- based online freelancers using a variety of gig work platforms in August 2022. Limitations. The analysis in this chapter has some data limitations. First, the comparison of online gig workers to workers with similar occupations is restricted to eight countries for which the labor force surveys (LFSs) and household surveys contained enough information on occupational codes for an accurate analysis. The eight countries are Argentina, Bangladesh, India, Mexico, Pakistan, the Philippines, South Africa, and Tunisia (see appendix D). Second, the comparison between online gig workers and informal sector workers is restricted to four regions on the basis of data availability: Africa, Latin America and Caribbean, Middle East and North Africa, and South Asia (see also appendix D, which provides further details on the methodology for analyzing the global RDIT survey data and limitations). 93 T  he GLOW PENJANA program was developed by MDEC as a spin-off to the eRezeki platform to support individuals affected by the COVID-19 pandemic. The program provides training to aspiring online gig workers. 78 Working Without Borders: The Promise and Peril of Online Gig Work TABLE 4.1: Platforms featured in the study (includes survey data and data provided by the platform) Platform Region Type of data Elharefa MENA Survey (n = 41) and platform data eRezeki platform EAP. These are initiatives of the Platform data and GLOW PENJANA Malaysian government agency program MDEC to support online gig work. Flexiport SA Survey (n = 11) and platform data Jolancer AFR Survey (n = 19) and platform data Microworkers Global microwork platform Survey data (n = 1,073) Onesha AFR Survey (n = 82) and platform data SheWorks! LAC Survey (n = 36) and platform data SoyFreelancer LAC Survey data (n = 325) Truelancer SA Survey (n = 746) and platform data Workana LAC (with a regional office in EAP Survey (n = 3,697) and platform data collected in as well) collaboration with the Inter-American Development Bank Wowzi AFR Survey (n = 960) and platform data YouDo ECA Platform data Source: Study team compilation. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SA = Southeast Asia; MDEC = Malaysian Digital Economy Corporation. 4.3 AGE Online gig work platforms tend to attract youth. Most online gig workers tend to be youth under the age of 30, mostly students or young professionals at the beginning of their careers. More than half of online gig workers are under 30, and the results hold true across most regions except for East Asia and Pacific, where the share of youth is slightly smaller (48 percent; figure 4.1). In this respect, there is no significant difference between global platforms and regional/local platforms. FIGURE 4.1: Age composition of online gig workers in the global survey 100 4 1 5 1 90 17 14 80 35 27 30 Share of workers (%) 70 39 23 60 17 50 40 79 30 63 66 53 53 48 20 10 0 EAP LAC ECA SSA SAR MENA Age groups 15–29 30–44 45–54 55–64 65+ Source: Global RDIT survey conducted by the study team. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. 79 Chapter 4 How Inclusive Is the Online Gig Economy? Online gig workers are younger than workers in the services sector, workers with similar occupations in the labor market, and workers in the informal sector. Across regions, the por- tion of youth among online gig workers is significantly greater than that in the services sector and in the informal sector (Figure 4.2). Results from eight countries show a similar pattern of a significant share of online gig workers younger than workers with similar occupations in the labor market, and in some cases the difference is sizeable (Figure 4.3). For instance, over 63 percent of online gig workers in Mexico, Pakistan, and Tunisia are under 30, a much larger proportion than in the labor force (between 15 and 33 percent). FIGURE 4.2: Age composition of online gig workers, by region a. Compared to workers in the services sector Online gig workers 63 30 14 1 SSA Service sector workers 29 39 17 10 6 Online gig workers 53 39 4 13 ECA Service sector workers 20 38 23 16 3 Online gig workers 48 17 35 0 EAP Service sector workers 25 43 23 7 1 Online gig workers 53 23 17 3 3 LAC Service sector workers 30 32 18 11 9 Online gig workers 79 14 5 1 MENA Service sector workers 23 37 23 12 4 Online gig workers 66 27 1 6 0 SAR Service sector workers 36 31 16 10 7 0 20 40 60 80 100 Share of workers (%) Age groups 15–29 30–44 45–54 55–64 65+ b. Compared to informal workers Online gig workers 45 54 1 SSA Informal sector workers 16 76 8 Online gig workers 40 57 3 LAC Informal sector workers 18 75 7 Online gig workers 49 51 0 MENA Informal sector workers 21 76 2 Online gig workers 51 47 2 SAR Informal sector workers 14 82 4 0 20 40 60 80 100 Share of workers (%) Age groups 15–24 25–64 65+ Source: Study team analysis of Global RDIT survey and labor force and household surveys. Note: The values for online gig workers by region differ between the two figures because the comparator countries vary in data availability. The online gig worker estimates include the same countries in each region as those for which the team had labor force surveys. For a list of countries and labor force surveys used, please refer to appendix D, specifically tables D.4 and D.5. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. 80 Working Without Borders: The Promise and Peril of Online Gig Work FIGURE 4.3: Age composition of online gig workers compared to workers in similar occupations, by country Online gig workers 44 29 20 4 3 ARG Similar occupation workers 19 40 23 14 4 Online gig workers 39 48 10 20 BNG Similar occupation workers 52 29 11 5 2 Online gig workers 57 32 0 12 0 IND Similar occupation workers 26 44 20 8 2 Online gig workers 64 30 12 2 MEX Similar occupation workers 33 39 17 8 3 Online gig workers 67 10 14 6 3 PAK Similar occupation workers 32 44 17 7 1 Online gig workers 61 22 12 05 PHL Similar occupation workers 41 39 13 6 1 Online gig workers 32 31 3 26 7 SAF Similar occupation workers 20 43 25 10 1 Online gig workers 63 28 3 4 2 TUN Similar occupation workers 15 48 25 10 1 0 20 40 60 80 100 Share of workers (%) Age groups 15–29 30–44 45–54 55–64 65+ Source: Study team analysis of global RDIT survey and labor force and household surveys. Please see table D.6 in appendix D for the list of countries and labor force surveys used. Note: ARG = Argentina; BNG = Bangladesh; IND = India; MEX = Mexico; PAK = Pakistan; PHL = the Philippines; SAF = South Africa; TUN = Tunisia. Data from platform-based surveys also confirm the greater proportion of youth. For instance, over half of the respondents on Truelancer, an online freelancing platform based in India, were youth, with an even higher proportion (61 percent) for the global microtask platform Microworkers (see Figure 4.4, panel a). Microwork is seen as a good source of supplementary income for young people (Cedefop 2021). Wowzi, a Kenya-based platform specializing exclusively in “influencer” marketing tasks, had almost 90 percent youth freelancers (or influencers)94 because of its focus on new social media skills. The Latin American platforms Workana and SoyFreelancer also showed significant shares of young workers: 50 and 40 percent, respectively.95 The study team’s country deep dives confirm the dominance of youth on gig platforms (Figure 4.4, panel b). More than half of the survey respondents in Bangladesh were 20- to 30-year-olds, while in Pakistan, both the average and the median ages of respondents to the team’s survey were 26 years. In Indonesia, over 50 percent of the online gig workers are below 30 years old, compared to 24 percent of the informal-sector workers. Existing studies on global trends in gig work suggest a similar age pattern, with online platform workers tending to be below the age of 35.96 94  he share is based on the number of freelancers using Wowzi who provided information about their age to the platform. T The proportion is confirmed by data collected through a survey conducted by the World Bank on the Wowzi platform. 95  The data presented are based on an internal survey conducted by Workana Latin America among its user base and confirmed through the survey conducted by the World Bank and Inter-American Development Bank for this study. 96 Several studies confirm this profile, for instance ILO (2021a, 2021b), Goldfarb (2019), and in the European Union, Pesole  et al. (2018), Urzì Brancati, Pesole, and Férnandéz-Macías (2020), and Cedefop (2021). 81 Chapter 4 How Inclusive Is the Online Gig Economy? FIGURE 4.4: Age distribution of online gig workers a. In selected platform surveys Wowzi 89 10 10 Microworkers 61 34 4 10 Truelancer 54 36 7 21 0 20 40 60 80 100 Share of workers (%) b. In country deep dives Malaysia 60 32 6 20 Bangladesh 58 40 10 Pakistan 74 25 10 Indonesia 55 37 8 0 0 20 40 60 80 100 Share of workers (%) Age groups 15–29 30–44 45–54 55–64 65+ Source: Analysis of platform surveys and country deep dives conducted by the study team. Note: Data for Malaysia indicate registered users on the eRezeki platform in 2020. Digital gig work attracts young people for several reasons. The study survey found three key reasons that online gig work platforms appeal to youth: the chance to learn new digital skills, espe- cially for someone at the beginning of their career; the flexibility of online work; and the ability to earn additional income. Most youth gig workers have another job or are students, findings that are similar to those of other studies (ILO 2021b). In countries with high youth unemployment rates, gig work could provide a path to integrate youth into the labor market.97 Opportunities in the online gig economy can play an important role in countries struggling with high levels of youth unemployment or underemployment. For countries with growing cohorts of youth, online gig work can provide young people with work opportunities beyond what is available in the traditional labor market (UNDESA 2022). Countries struggling with high youth unemployment rates or high rates of youth not in employment, education, or training (NEET), ­ 97  ee ILO news release, “Global Youth Unemployment is on the Rise Again,” August 24, 2016, https://www.ilo.org/global/ S about-the-ilo/newsroom/news/WCMS_513728/lang--en/index.htm. 82 Working Without Borders: The Promise and Peril of Online Gig Work like Nigeria (36 percent) and Pakistan (34 percent),98 could provide targeted support to youth to access online gig jobs (figure 4.5; see also chapter 7). FIGURE 4.5: Proportion of youth in the working-age population and NEET rate among youth in the 17 countries in the global survey 40 NGA PAK 35 ZAF 30 PHL BGD KEN LBN Neet rate among the youth (%) MEX EGY 25 IND VEN ARG MAR 20 TUN RUS CHN 15 UKR 10 5 0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 The share of youth in the working-age population (%) Sources: ILOSTAT and UNDESA. ILOSTAT data are from 2021 and 2022; UNDESA data are from 2022. Note: ARG = Argentina; BGD = Bangladesh; CHN = China; EGY = Arab Republic of Egypt; IND = India; KEN = Kenya; LBN = Lebanon; MAR = Morocco; MEX = Mexico; NGA = Nigeria; PAK = Pakistan; PHL = the Philippines; RUS = Russian Federation; TUN = Tunisia; UKR = Ukraine; VEN = República Bolivariana de Venezuela; ZAF = South Africa. NEET = not in employment, education, or training. 4.4 GENDER Globally, women participate in online gig work to a greater extent than in the general labor market. The survey found that 42 percent of online gig workers are women, a larger pro- portion than in the global labor force (39.7 percent as of 2021).99 By region, the share of women in online gig work varies between 19 percent in the South Asia region and 56 percent in the Middle East and North Africa (figure 4.6, panel a). In some cases, the portion of women in online gig work is significantly greater than that for the services sector (in East Asia and Pacific and the Middle East 98 I LO, “ILO Modelled Estimates (ILOEST database),” 2022, https://ilostat.ilo.org/resources/concepts-and-definitions/ ilo-modelled-estimates/. 99 World Bank, WDI database. Estimates are based on data obtained from the ILO and the United Nations Population  Division, https://data.worldbank.org/indicator/SL.TLF.TOTL.FE.ZS. 83 Chapter 4 How Inclusive Is the Online Gig Economy? and North Africa; figure 4.6, panel a) and the informal sector (in the Middle East and North Africa; figure 4.6, panel b). The share of women among gig workers is greater on global platforms than on regional platforms (45 versus 27 percent). FIGURE 4.6: Share of female online gig workers, by region a. Compared to female workers in the services sector Online gig workers 60 40 ECA Service sector workers 52 48 Online gig workers 61 39 LAC Service sector workers 60 40 Online gig workers 81 19 SAR Service sector workers 74 26 Online gig workers 46 54 EAP Service sector workers 61 39 Online gig workers 44 56 MENA Service sector workers 82 18 Online gig workers 72 27 SAR Service sector workers 55 45 0 20 40 60 80 100 Share of workers (%) b. Compared to female workers in the informal sector Online gig workers 51 49 SAR Informal sector workers 50 50 Online gig workers 61 39 MENA Informal sector workers 61 39 Online gig workers 87 13 LAC Informal sector workers 40 60 Online gig workers 73 27 SSA Informal sector workers 81 19 0 20 40 60 80 100 Share of workers (%) Male Female Source: Study team analysis of global RDIT survey and labor force and household surveys. See tables D.4. and D.5 in appendix D. Note: The values for online gig workers by region differ between the two figures because the comparator countries vary in data availability. The online gig worker estimates refer to the same countries in each region as those in the labor force surveys (LFSs). For a list of countries and LFSs used, please refer to appendix C, specifically tables C.4 and C.5. ECA = Europe and Central Asia; EAP = East Asia and Pacific; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. 84 Working Without Borders: The Promise and Peril of Online Gig Work The potential of online gig work to support female labor force participation has not been fully tapped. Results from eight countries in the team’s global survey100 show that while women are starting to participate to a greater extent in the online gig economy than in the general workforce in similar occupations, they remain underrepresented in several countries (figure 4.7). In Argentina, Bangladesh, and Pakistan, women account for greater shares of online gig workers than in the broad labor force. In Argentina, in fact, women account for almost two in three online gig workers (57 percent). At the same time, in countries including India, the Philippines, South Africa, and Tunisia, the share of women in the online gig economy is much more limited than the share of women in similar occupations in the workforce at large. Other studies have found overall similar results101 and have pointed to a smaller proportion of women (2 in 10) in online gig work in developing countries (ILO 2021b). In India, fewer than 2 in 10 platform workers were women (ILO 2021a). Among the G20 countries, Italy has the largest portion of women online gig workers (58 percent) (ILO 2021a). FIGURE 4.7: Proportions of female online gig workers compared to female workers in similar occupations in selected countries Online gig workers 35 65 ARG Similar occupation workers 43 57 Online gig workers 59 41 BGD Similar occupation workers 69 31 Online gig workers 81 19 IND Similar occupation workers 72 28 Online gig workers 69 31 MEX Similar occupation workers 62 38 Online gig workers 80 20 PAK Similar occupation workers 89 11 Online gig workers 55 45 PHL Similar occupation workers 39 61 Online gig workers 48 52 ZAF Similar occupation workers 39 61 Online gig workers 72 28 TUN Similar occupation workers 53 47 0 20 40 60 80 100 Share of workers (%) Male Female Source: Study team analysis of global RDIT survey and labor force and household surveys. See table D.6 in appendix D for the list of countries and labor force surveys used. Note: ARG = Argentina; BGD = Bangladesh; IND = India; MEX = Mexico; PAK = Pakistan; PHL = the Philippines; ZAF, South Africa; TUN = Tunisia. Some countries and gig platforms are doing better in including women. The country deep dive in Indonesia shows a greater share of women in online gig work than in the informal sector (50 versus 31 percent). In Malaysia and Latin America, online gig work enables more women to engage in paid work than the general labor market does (Figure 4.8). In Malaysia, the eRezeki and 100  he comparison was developed for those countries for which the labor force and household surveys contained enough T information on occupational codes for an accurate analysis. 101 ILO (2021b) found that 4 in 10 online gig workers are women.  85 Chapter 4 How Inclusive Is the Online Gig Economy? GLOW PENJANA programs (online gig work programs supported by the Malaysian government) show a percentage of women users (over 50 percent) higher than the general labor force participation of women (38 percent). A higher percentage of women is also reported for SoyFreelancer (52 percent) and Workana (49 percent). On YouDo, a Russian online gig work platform, however, the vast majority of registered users (71 percent) are male. Compared to the share of women in the offline Russian labor force (48.6 percent), women engage to a lesser extent on YouDo. FIGURE 4.8: Women’s participation in the labor force and in online gig work platforms 49 Latin America 52 41 58 Malaysia 51 38 29 Russian Federation 49 0 10 20 30 40 50 60 70 Share of workers (%) Workana SoyFreelancer GLOW eRezeki YouDo Country/Region average Sources: Country/regional averages were retrieved from WDI. The percentages of women gig workers by platform are based on platform and survey data collected for this study. Note: The country/region average shows the share of women in the total workforce in 2021. The key drivers of women’s participation in this market are the ability to earn additional income and the flexibility online work offers. The team’s global survey shows that women most value those two attributes of online gig work. Women are more likely than men to do online gig work because they want to earn additional income and because they don’t have other job opportunities, while men appreciate more the ability to learn new digital skills and the chance to be one’s own boss (figure 4.9, panel a). Data at the platform level provide further evidence. For women working on Workana, flexibility in location and time was a more important motivating factor figure 4.9, panel b). Flexible working hours can help women balance their caregiving responsibilities (­ with the need to earn a living (Anwar and Graham 2020). In Africa, household survey data from nine countries102 from 2017 and 2018 show that women are driven mainly by the need to control their schedule (over 60 percent), whereas this reason carries less weight for men. Conversely, the most important reason for men to join gig work platforms is to gain work experience for future job opportunities (over 65 percent of men compared to approximately 30 percent of women). However, flexibility comes with a caveat. When flexibility leads to fragmented work schedules, it may have a negative impact on the speed with which tasks are completed and on earnings; women tend to be particularly affected (Adams-Prassl 2020). 102  he nine countries are Ghana, Kenya, Mozambique, Nigeria, Rwanda, Senegal, South Africa, Tanzania, and Uganda. T The survey was conducted by Research ICT Africa, an ICT policy think tank. The data cover not only online web-based platform workers, but also location-based platform workers (Chen, forthcoming). 86 Working Without Borders: The Promise and Peril of Online Gig Work FIGURE 4.9: Main reason for doing online gig work by gender a. Global survey 34.8 To get additional income or higher pay 35 26.1 Flexibility on time and location 23.6 13.5 To learn new digital skills 15.4 12.2 Allow me to be my own boss 15 13.5 No job opportunity 11 0 5 10 15 20 25 30 35 40 Share of workers (%) b. Workana survey 49.9 Flexibility on time and location 47.4 27.1 To get additional income or higher pay 28.7 9 To learn new digital skills 10.1 9.5 Allow me to be my own boss 8.8 4.6 No job opportunity 5 0 10 20 30 40 50 60 Share of workers (%) Women Male Source: Study team analysis of global RDIT survey and the Workana survey conducted by the study team. Note: The gender difference in the Workana survey is statistically significant at 5 percent for flexibility on time and getting additional income. In the global gig worker survey, males were more likely than their female counterparts to report that their motivation for engaging in online gig work is driven by the desire to be their own boss, have location flexibility, and learn new digital skills. These differences are statistically significant at 5 percent, with weights applied. On the other hand, females are more likely than males to report that the lack of job opportunities is a driving factor for their engaging in online gig work. A more proactive and intentional approach to enroll women can make digital work more gender inclusive. One example of active support for the participation of women in online gig work is the Latin American platform SheWorks!. While the platform is not exclusively for women, most of the online gig workers using it are women because of the platform’s emphasis on flexible working hours and the marketing strategy reflected in the platform’s name. Networks and successful women freelancers sharing their experience with other women can be a catalyst for promoting the opportu- nities of online gig work among women (see Box 4.1 for an example from Pakistan). 87 Chapter 4 How Inclusive Is the Online Gig Economy? BOX 4.1: JOURNEYS OF SUCCESSFUL WOMEN ONLINE FREELANCERS IN PAKISTAN Two successful women online freelancers in Pakistan—Laraib Afzal and Anum Bakhtiar— started their online careers after studying software engineering and being faced with limited work opportunities in the field of information technology (IT). They joined the most popular online gig work platform in Pakistan, Fiverr, with very limited experience in online freelancing but with the desire to learn and to access more jobs in their preferred fields. Becoming an online freelancer involved a significant amount of self-learning and learning by doing. Laraib developed her graphic design skills by watching YouTube videos, and both women learned to improve their freelancer profiles by analyzing other profiles and deriving best practices. While the start of their journeys was difficult and at times disheartening, with no or very few low-value orders received, by persevering in the process and continually learning, both Laraib and Anum managed to build successful profiles. In addition to their technical skills, soft skills have played a major role in securing their success, particularly skills in communication, managing clients, and having confidence in interactions with clients. In growing their business, management skills also became quite important, especially for overcoming challenges related to fluctuating income and the need to build a diverse portfolio of clients. Anum is now running her own business in the world of online freelancing, specializing in graphic design and developer jobs. She currently works with several other women, training them in graphic design and in how to succeed in receiving jobs through Fiverr. Online freelancing is no longer the main career for Laraib, but she sees it as a valuable activity next to her full-time job, as it allows her to keep improving her skills and developing new ones. She is also seeking to further develop her experience as an online freelancer and establish an agency account in order to work with other online freelancers and share her acquired knowledge of the field. 4.5 SKILLS AND EDUCATION Workers with a variety of skill levels are participating in the online gig economy, especially those with high-school-level education. Over 70 percent of online gig workers do not have a tertiary education degree (Figure 4.10). The participation of workers with basic and intermediate education shows that there are opportunities and there is growing awareness of online gig work across varied educational backgrounds. The fact that the team’s global survey was conducted in multiple languages, not just in English, could explain the difference between our survey findings and the literature.103 Knowledge of English in countries where English is not an official language may be correlated with a higher level of education. 103 The ILO estimated in 2021 that over 60 percent of gig workers attained at least one university degree (ILO 2021b).  88 Working Without Borders: The Promise and Peril of Online Gig Work It is also important to assess whether skill levels affect the intensity with which people do gig work. Given the task-based nature of gig work, those doing gig work as a primary job may be different from those who do gig work sporadically. To understand work intensity, the gig workers were classified as main, secondary, or marginal workers depending on the extent to which gig work contributed to their overall income and the number of hours they worked on gig tasks (see Table 4.2). Workers with tertiary education are more likely to do online gig work as a main occupation than those with less education (Figure 4.10). TABLE 4.2: Intensity of online gig work based on income earned as a share of personal income and hours worked Less than 10 hours Between 10 and More than 20 hours a week 20 hours a week a week Less than 25 percent of personal income Marginal Secondary Secondary 25 to 50 percent of personal income Secondary Secondary Main More than 50 percent of personal income Secondary Main Main Source: Adapted from Urzì Brancati, Pesole, and Férnandéz-Macías 2020. FIGURE 4.10: Educational backgrounds of online gig workers and intensity of online gig work 70 64 60 57 52 50 40 30 20 19 20 13 15 10 11 9 9 9 10 5 3 4 0 Postgraduate Bachelor's Vocational High School Below high Degree Degree Training school Main Secondary Marginal Source: Global RDIT survey conducted by the study team. Note: Results are shown as percentages. Local platforms tend to attract a slightly larger share of workers with intermediate education (high school and vocational) than global platforms. Almost half of the gig workers on local platforms have vocational or high-school-level training, while global gig platforms tend to attract slightly more diverse workers, at both the high-skills end (workers with a bachelor’s degree) and the low-skills end (workers with below-high-school education) (Figure 4.11). However, the differences remain minor and may be due to the tasks available on regional/local platforms versus global plat- forms and the level of education required to complete such tasks (see chapter 3 for a discussion of tasks on global and regional/local platforms). 89 Chapter 4 How Inclusive Is the Online Gig Economy? FIGURE 4.11: Educational backgrounds of online gig workers using global and regional/local platforms 30 25 26 25 24 24 24 22 21 20 19 50 10 8 8 5 0 Graduate Bachelor's Vocational High School Below high degree degree Training school Global Regional Source: Study team analysis of global RDIT survey conducted by the team. Note: Results are shown as percentages. On average, online gig workers are more educated than workers in the services sector and the informal sector. In most regions, the share of online gig workers with advanced education is greater than that of workers in the services sector; Europe and Central Asia and East Asia and Pacific are the exceptions (figure 4.12, panel a). Online gig workers are significantly more educated than workers in the informal sector (only 3 to 12 percent of informal workers have advanced education) (figure 4.12, panel b). FIGURE 4.12: Educational backgrounds of online gig workers, by region a. Compared to workers in the services sector Online gig workers 3 34 63 SSA Service sector workers 16 30 34 20 Online gig workers 13 37 50 ECA Service sector workers 3 43 54 Online gig workers 6 67 27 EAP Service sector workers 1 25 39 35 Online gig workers 16 50 34 LAC Service sector workers 4 35 32 29 Online gig workers 1 62 36 MENA Service sector workers 14 17 35 35 Online gig workers 20 44 36 SAR Service sector workers 17 39 19 24 0 20 40 60 80 100 Share of workers (%) Less than basic Basic Intermediate Advanced (Continued) b. Compared to workers in the informal sector 90 Online gig workers 3 34 63 SSA Online gig workers 1 62 36 MEN Service sector workers 14 17 35 35 Online gig workers 20 44 36 SAR Service sector workers 17 39 19 24 Working Without Borders: The Promise and Peril of Online Gig Work 0 20 40 60 80 100 Share of workers (%) Less than basic Basic Intermediate Advanced FIGURE 4.12: (Continued) b. Compared to workers in the informal sector Online gig workers 3 34 63 SSA Informal sector workers 20 55 22 3 Online gig workers 16 50 34 LAC Informal sector workers 11 51 26 12 Online gig workers 1 62 36 MENA Informal sector workers 32 23 33 12 Online gig workers 20 44 36 SAR Informal sector workers 36 45 12 7 0 20 40 60 80 100 Share of workers (%) Less than basic Basic Intermediate Advanced Source: Study team analysis of global RDIT survey and labor force and household surveys. See tables D.4. and D.5 in appendix D. Note: The values for online gig workers by region differ between the two figures because the comparator countries vary in data availability. The online gig worker estimates refer to the same countries in each region as those in the labor force surveys (LFSs). For a list of countries and LFSs used, please refer to appendix C, specifically tables C.4 and C.5. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. Microtasks provide more opportunities than more-complex online freelancing tasks for low- skilled workers. Microwork generally includes repetitive, routine tasks, such as data classification, that can be performed relatively easily by following a set of instructions. Workers doing online micro- tasks tend to have a lower level of education (77 percent have only high school or less education, and only 15 percent have university-level education) than online gig workers who conduct complex tasks such as IT and software development (almost 40 percent have university-level education) and business and professional management (36 percent of gig workers have university-level education; see figure 4.13). The ILO also shows that online gig workers who do microtasks tend to be less edu- cated than online gig workers who do more-complex freelancing tasks (64 percent of microworkers are highly educated, compared to 83 percent of freelancers) (ILO 2021b). 91 Chapter 4 How Inclusive Is the Online Gig Economy? FIGURE 4.13: Highest level of education attained by online gig workers and their main online gig tasks IT, software development and tech 39 61 Design, multimedia and creative tasks 24 76 Business and professional management 36 64 Sales and marketing support 35 64 Writing and translation 32 67 Business and professional support 23 76 Data entry, administrative and clerical tasks 30 70 Online microtasks 15 86 0 10 20 30 40 50 60 70 80 90 100 Share of workers (%) University degree Below university Source: Analysis of global RDIT survey conducted by the study team. Note: IT = information technology. Microtasks can help drive the inclusion of low-skilled workers. Data from the eRezeki platform and GLOW PENJANA program in Malaysia suggest that over 50 percent of online gig workers carry out data entry and clerical tasks rather than more-complex digital tasks or digitally enabled tasks such as delivery and domestic services. In comparison, only 8.3 percent of the overall labor force in Malaysia carries out similar tasks104 (clerical support105), suggesting that online gig work opens up new oppor- tunities for gig workers that are otherwise not that common in the general labor market. From this perspective, online gig work can also provide more opportunities for workers who are not highly skilled. This is particularly relevant since the majority of workers by occupation in Malaysia are concentrated in services and sales (24.3 percent), an occupation group that generally relies on workers with secondary education or postsecondary, nontertiary education. While administrative and clerical occupations are not among the most common in Malaysia, they are accessible since they do not require a high level of skills and thus may provide opportunities for a broad range of workers in the labor market. Online digital work replicates the occupational segregation observed in the offline labor market, with men dominating tasks that require higher-technology skills (such as IT and software development) and that pay more. On Workana, for example, the share of men doing IT-related tasks is very high compared to that of women (44 versus 5 percent). In contrast, the propor- tion of women working in sales and marketing, data entry, and online microtasks is higher than that of men. Similarly, on SoyFreelancer, another Latin American platform, IT-related tasks offer higher pay than data entry and online microtasks. In Malaysia, women also do data entry and administrative and clerical tasks to a greater extent than men on the GLOW program106 (figure 4.19). Globally, women gig workers generally perform work in legal services, translation, writing and editing, business ser- vices, and sales and marketing more than men do, while men dominate work related to technology and data analytics (ILO 2021b). 104 B  ased on data from 2020 (Department of Statistics Malaysia 2020). 105 Data entry, administrative, and clerical tasks are equivalent to the job of clerical support workers, as defined by the  International Standard Classification of Occupations ISCO-08, which include general office clerks, data entry clerks, secretaries and such (ILO 2012). The International Standard Classification of Occupations-ISCO-08 is available at https://www.ilo.org/ilostat-files/ISCO/newdocs-08-2021/ISCO-08/ISCO-08%20EN%20Vol%201.pdf. 106 The GLOW PENJANA program was developed by MDEC as a spin-off to the eRezeki platform to support individuals  affected by the COVID-19 pandemic. The program provides training to aspiring online gig workers. 92 Working Without Borders: The Promise and Peril of Online Gig Work FIGURE 4.14: Share of users by gender and workstream, GLOW PENJANA program, Malaysia, 2021 100 5 4 7 7 7 6 90 13 12 80 14 70 17 16 19 60 50 40 30 58 62 54 20 10 0 All Female Male Design, Media and Architecture Writing and Content Creation Websites, IT and Software Sales and Marketing, Social Media, SEO Data entry, Admin and Virtual Assistant Source: Study team analysis based on Malaysian Digital Economy Corporation (MDEC) data. Note: IT = information technology; SEO = search engine optimization. Gig work requires more than just digital skills. In the study surveys, socioeconomic skills in particular are consistently mentioned as necessary for success on digital platforms. For Workana workers, communication skills and time management were listed as most important, alongside a set of other skills such as self-confidence; this observation holds true across education levels and genders (see Figure 4.15). FIGURE 4.15: Top skills for succeeding in online gig work, by education level and gender of online gig workers on Workana a. Importance of skills across education levels 90 86858786 8685 80 8383 81 79 7575 7473 7169716971 727269 80 67 67 6666 68 6766 63 65 62 62 62 60 57 5657 58 Percentage 70 50485050 42 60 50 40 en e sk ion sk ion ag n tio e ki l en t en lf g l cy l s ca em lien em im tia ric gu y i id Se En e n ta ni t ills ce ills lls t e n ish t t an nc go P ag T ica tia ag C gi h in fici di Tec l l ie go un o c a fi c Pr Ne m nf lo ro m co an an ne P Co m m Graduate degree Bachelors's degree Vocational High school Below high school (Continued) b. Importance of skills across gender 93 Time management 88 82 80 Communication skills 87 50 40 en e sk ion sk ion ag n tio e ki l en t en lf g l cy l s ca em lien em im tia ric gu y i id Se En e n ta ni t ills ce ills lls t e n ish t t an nc go P ag T ica tia ag C gi h in fici di Tec l l ie go un o c a fi c Pr Ne m nf Chapter 4 How Inclusive Is the Online Gig Economy? lo Pro m co an an ne Co m m Graduate degree Bachelors's degree Vocational High school Below high school FIGURE 4.15: (Continued) b. Importance of skills across gender Time management 88 82 80 Communication skills 87 Self confidence 71 80 Negotiation skills 70 70 Technical digital skills 68 56 63 Client management 73 60 Proficiency in local language 71 Price negotiation 56 60 Proficiency in english 49 46 0 10 20 30 40 50 60 70 80 90 100 Percentage of workers Male Female Source: Study team analysis based on Workana survey data. Note: Values are percentages of respondents; respondents could choose multiple options. The survey results indicate that there are statistically significant gender differences in all of the skills that were identified as very important, except for negotiation skills and the ability to speak and read English, for which there were no significant differences (5 percent level) observed. 4.6 SPATIAL INCLUSION Online gig work creates work opportunities beyond major cities. The global survey was able to track a respondent’s location; the survey automatically recorded geolocation data for each respondent. The team used the location data to classify gig workers into three types of cities: (a) capital cities, (b) secondary cities (cities that are not the capital city but among the top 10 largest cities in a given country), and (c) tertiary cities (smaller cities and towns beyond the capital city and the top 10 largest cities in a given country). The data show that more than 6 in 10 gig workers live in tertiary cities and over a quarter live in a secondary city (Figure 4.16, panel a). Patterns may differ at the platform level, but generally a good share of online gig workers come from cities other than the capital. On the India-based Truelancer platform, for instance, more than 60 percent of the online gig workers surveyed live in secondary or tertiary cities and villages; 40 percent live in capital cities. Nevertheless, there are strong differences between regions. The vast majority of online gig workers in Europe and Central Asia, East Asia and Pacific, and Latin America and the Caribbean are based in tertiary cities (fFigure 4.16, panel b). However, in Sub-Saharan Africa and in the Middle East and North Africa, a much greater share of online gig workers is in capital cities than in the other regions (42 and 45 percent, respectively). There is no major difference between the location of gig workers on global platforms and regional platforms. The spread of gig workers across both major and minor cities within countries shows that online gig work can bring tangible benefits for workers beyond the main economic centers or capital cities. 94 Working Without Borders: The Promise and Peril of Online Gig Work FIGURE 4.16: Distribution of online gig workers by city size a. Distribution of online gig workers by city size (%) Capital city 9.8 27.5 Secondary cities 62.6 Tertiary cities b. Distribution of online gig workers by city size and region 100 90 17 29 80 54 70 64 41 60 81 25 50 98 40 30 42 21 42 45 20 10 16 15 1 3 5 0 ECA EAP SAR LAC SSA MENA Capital city Secondary cities Tertiary cities Source: Global RDIT survey conducted by the study team. Note: Secondary cities in this context refer to the top 10 largest cities in a given country except for the capital. Tertiary refers to the rest of the smaller cities and towns. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. While remote online work can provide more job options for rural workers, the availability of digital infrastructure and devices is one of the main constraints. The spatial distribution of online gig work is dependent on the level of internet penetration, rural electrification, and the overall level of economic development of the country. With greater availability of internet access, greater levels of rural electrification, and higher income per capita, gig workers tend to be more spread out in secondary and tertiary cities in the country (figures 4.17 and 4.18). A study conducted with US platform workers also found that the least urbanized areas with poor infrastructure and lower levels of education are least likely to participate in online platform work (Braesemann et al. 2022). A digital divide between urban and rural areas still exists in developing countries. The difference in access to the internet between urban and rural areas is marginal in developed countries (89 and 85 percent, respectively), but in developing countries the disparity is much wider (72 and 34 percent, respectively) (ITU 2021). The difference in the enabling environment and access to the Internet may limit opportunities in developing countries that lack the infrastructure to support online gig work. 95 Chapter 4 How Inclusive Is the Online Gig Economy? A study based on data from a major global platform suggests that online gig job projects flow to the capital cities in the Global South to a greater extent than in other regions of the countries, with capitals attracting 15 times as many projects.107 FIGURE 4.17: Spatial distribution of gig workers within countries 100 5 90 6 18 22 22 26 Share of workers (%) 80 36 41 44 70 59 60 54 60 34 22 70 71 66 60 75 78 41 50 46 40 89 28 51 46 30 15 12 57 33 20 47 15 18 40 38 11 30 10 29 28 26 28 20 13 15 13 11 14 13 0 4 1 a ria p. Ru Ban cco Fe esh n ut ine Pa a an a RB Ar ico a ilip a s a n ne ny ric isi in di in tio no Re ge st ex a nt In Ch Ve Tun a, o ad pi Ke Af ra ba r ki or Ni Uk M ab el ge gl de Le h M zu Ar Ph ne So t, n yp ia ss Eg Capital city Secondary city Tertiary city Source: Global RDIT survey conducted by the study team. FIGURE 4.18: Relationship between spatial distribution of gig workers within countries and key infrastructure and economic development factors a. Share of gig workers in the top five cities and internet penetration rate in each country 50 KEN NGA Share of workers (%) 40 EGY ZAF MAR BGD 30 PAK LBN IND RUS UKR 20 TUN PHL ARG MEX CHN 10 20 40 60 80 100 Internet penetration rate b. Share of gig workers in the top five cities and (Continued) rural electrification rate in each country 50 KEN NGA EGY Share of workers (%) 40 ZAF MAR 107 B  raesemann, Lehdonvirta, and Kässi (2022) used data from onePAK gig jobs tend major global platform and found that BGD LBN to30be clustered in capital cities. Their study used different indicators of concentration and used data from only one IND platform, while our survey, conducted in 12 languages, reached a larger proportion of people in smaller cities. In RUS UKR addition, 20 the Global South classification used in this paper does not account for several countries, including China, TUN India, and South Africa, which are included in the team’s estimates based on the global survey and which PHL carry ARG significant weights in the team’s analysis. MEX 10 CHN 96 0 20 40 60 80 100 Rural electrification rate ZAF Share of worker BGD 30 PAK LBN IND RUS UKR 20 Working Without Borders: The Promise and Peril of Online Gig Work TUN PHL ARG MEX CHN 10 20 40 60 80 100 FIGURE 4.18: (Continued) Internet penetration rate b. Share of gig workers in the top five cities and rural electrification rate in each country 50 KEN NGA EGY Share of workers (%) 40 ZAF MAR PAK BGD LBN 30 IND RUS UKR 20 TUN PHL ARG MEX 10 CHN 0 20 40 60 80 100 Rural electrification rate c. Share of gig workers in the top five cities and per capita income (purchasing power parity) in each country 50 KEN NGA Share of workers (%) 40 EGY MAR ZAF PAK 30 BGD LBN IND RUS UKR 20 TUN PHL ARG MEX CHN 10 0 10000 20000 30000 40000 Per capita income PPP Countries Trendline Source: Analysis based on the global RDIT survey conducted by the study team and WDI data. Note: The analysis is restricted to the percentage of gig workers in the top five cities in each of the countries in the global survey. ARG = Argentina; BGD = Bangladesh; CHN = China; EGY = Arab Republic of Egypt; IND = India; KEN = Kenya; LBN = Lebanon; MAR = Morocco; MEX = Mexico; NGA = Nigeria; PAK = Pakistan; PHL = the Philippines; RUS = Russian Federation; TUN = Tunisia; UKR = Ukraine; ZAF = South Africa. Gig work could provide some temporary opportunities for a particularly vulnerable group— namely, refugees, who often face difficulties in integrating in the local labor market and for whom location is thus a barrier to traditional work. An International Finance Corporation (IFC) report (IFC 2021) analyzing the experience of women refugees in Jordan and Lebanon empha- sizes that while the digital economy may hold promise for refugees, at least as a temporary source of income, there are still barriers to be overcome to integrate refugees into the economy (such as easing legal restrictions on the type of work that refugees can carry out and improving knowledge about the refugee demographic). Box 4.2 presents key initiatives promoting online gig work as an opportunity for refugees and other displaced people. 97 Chapter 4 How Inclusive Is the Online Gig Economy? BOX 4.2: ONLINE GIG WORK AS AN OPPORTUNITY FOR REFUGEES Online gig work can be a solution to the entry barriers of local traditional labor markets for refugees and displaced people. Several initiatives around the world are tapping this potential, through a combination of training programs directly geared to or open to refugees, among other participants, and access to online gig job opportunities. Humans in the Loop is a social enterprise founded in 2017 and based in Bulgaria (Humans in the Loop 2020). It is active in Iraq, the Syrian Arab Republic, and Türkiye and trains and employs displaced people to work on data annotation projects for artificial intelligence start-ups. Humans in the Loop takes a two-pronged approach to fostering access to online gig work opportunities for refugees by providing low-entry-barrier jobs, such as easy-to-complete data annotation online tasks, and by offering training opportunitiesvthat focus on digital skills, English language skills, and career guidance. The organization currently employs over 250 refugees, migrants, internally displaced people, and vulnerable locals; its workforce has grown from 167 in 2019. In addition to providing employment opportunities, Humans in the Loop had trained 137 people as of June 2022. The organization pays particular attention to the challenges faced by women and ensures that at least 50 percent of participants in the training and employment programs are women. In 2020, women made up 54.6 percent of its workforce (Humans in the Loop 2020). Gaza Sky Geeksa is an initiative of Mercy Corps founded in 2011 in Gaza and currently operating in Gaza, the West Bank, and East Jerusalem. Gaza Sky Geeks supports freelancers, founders, and coders working online and in the tech field. For online freelancers, Gaza Sky Geeks offers two types of programs: the Freelance Academy,b a three-month mentorship program, and the Code Academy, courses to improve programming skills. The Freelance Academy helps aspiring online freelancers understand the essentials of online freelancing platforms, how to build a competitive profile, and how to apply for jobs, communicate with clients, and negotiate. The Freelance Academy partners with Upwork and supports freelancers in setting up their accounts. The Freelance Academy has trained 2,225 online freelancers, 61 percent of whom were women. Through the Coding Academy, Gaza Sky Geeks provides two courses on web development: a foundational course for those without experience and an advanced course for students with some experience. More than 130 students have graduated from the Coding Academy. Gaza Sky Geeks has also supported refugees in using online gig opportunities. For instance, in 2021, 35 refugees and internally displaced people in Iraq completed the Freelance Academy program, delivered remotely with support from the Mercy Corps Iraq team.c Success stories of Gaza Sky Geeks also show their impact in the Palestinian refugee camp of Al Faraa, where Gaza Sky Geeks organized a four-day boot camp to boost online freelancing skills.d (Continued) 98 Working Without Borders: The Promise and Peril of Online Gig Work BOX 4.2: ( Continued ) The Dadaab Collective provides an interesting example of leveraging training and the agency approach to online gig work to support refugees and displaced people. The Norwegian Refugee Council and the International Trade Centre, with funding from the Dutch Ministry of Foreign Affairs, have been training refugees in the Dadaab refugee camp in Kenya for online freelancing as part of the Refugee Employment and Skills Initiative (RESI).e The initiative provides courses for young refugees to develop skills that are sought-after on online gig work platforms, including graphic design, digital marketing, data entry, translation, and digital journalism and photography. The technical courses are complemented by trainings in soft skills and business skills to empower refugees to pursue online freelancing. The key to integrating the students into the market for online gig jobs, however, is not solely the training, but a cooperative of freelancers to support and motivate them to work. The cooperative, the Dadaab Collective, brings together the graduates of the training program and is independent and run solely by youth. The organization facilitates the sourcing of jobs among its members and is registered as an agency for Upwork.f By simplifying the process of searching for jobs, the agency model may be particularly useful for ensuring that less experienced graduates can learn and be motivated by graduates of the program who have gained experience in online freelancing, increasing their chances of success in the early stages of freelancing after having finished their training. a. See https://gazaskygeeks.com. b. See https://gazaskygeeks.com/freelance/. c. See “Letter from the Director,” January 5, 2022, https://www.linkedin.com/pulse/. letter-from-director-gaza-sky-geeks/?trk=organization-update-content_share-article. d. “Rapid Success in Just Two Years of Freelancing!, May 12, 2022, https://www.linkedin.com/pulse/ rapid-success-just-two-years-freelancing-gaza-sky-geeks. e. Paul Ireland, “Meet the Refugees Joining the Digital Economy,” NRC, March 31, 2021, https://www.nrc.no/ perspectives/2021/meet-the-refugees-joining-the-digital-economy/. f. Dadaa Collective Freelancing Agency, Upwork, https://www.upwork.com/ag/dadaabcollectiveagency/. 4.7 LANGUAGE Language can be a significant barrier to accessing online gig work opportunities. Some 33 percent of online gig workers confirm that one of the main challenges they face on global ­ platforms is English language skills. The global supply of online gig work is dominated by workers in English-speaking countries. Three countries in particular—India, Bangladesh, and Pakistan—account for over 50 percent of the supply of online gig work on the basis of data collected by the Online Labour Index (OLI),108 signaling that workers from non-English-speaking countries are likely to face language barriers to enter the online gig work market. 108 T  he OLI collects data from the five largest English-language online gig work platforms and six non-English-language platforms (three in Russian and three in Spanish), http://onlinelabourobservatory.org/oli-supply/. 99 Chapter 4 How Inclusive Is the Online Gig Economy? Surveys conducted in English tend to not only exclude non-English-speaking populations but also might underestimate the overall size of the online gig workforce. The study team’s global survey was translated into 12 languages to ensure a wider reach. In addition, the team was keen to reach gig workers who may be working on regional/local platforms. A substantial number of responses (57 percent) were in languages other than English (figure 4.19). For countries where English is not the official language or a widely used language, English-only surveys could neglect a significant proportion of the online gig work population (China, Ukraine, República Bolivariana de Venezuela; Figure 4.20). FIGURE 4.19: Languages of responses received to the global survey 45 42.8 40 35 30 25 20 15 13.8 13.3 10 8.6 7.1 4.1 5 2.0 1.9 1.5 0.8 0.4 0.3 0.1 0 ish sh c n rin ch la an g i du a ili nd bi us lo ia ah ng ni en ni da Ur gl a ss ga Hi Ha a Ar Sw Ba i En Ru ra Fr an Sp Ta Uk M Source: Global RDIT survey conducted by the study team. Note: Values are percentages. FIGURE 4.20: Distribution of languages of responses by online gig workers by country 100 0 0 4 6 90 21 26 31 80 47 70 60 78 78 87 89 89 91 92 92 50 100 99 96 94 99 40 80 74 69 30 53 20 10 22 22 13 12 11 9 8 8 1 0 ca a Pa a an ilip a ng es sh n sia t Fe xico n co a RB e a yp in in in di ny ri no tio n ri de oc st ge ni nt In Ch ra Eg a, pi Af Ke ba e ra ki Tu or la Uk Ni ge M el de Le h M zu Ar ut Ph Ba ne So n Ve ia ss Ru English Non-English Source: Global RDIT survey conducted by the study team. Note: Values are percentages. 100 Working Without Borders: The Promise and Peril of Online Gig Work Local platforms could help bring non-English-speaking people to digital platforms. Data from the global survey on differences between workers on global versus regional/local platforms provide supporting evidence. Two-thirds of online gig workers in the global survey who work on regional/local platforms completed the survey in a language other than English, while 50 percent of workers on global platforms responded in English. Platforms in Latin America and the Caribbean have especially catered to local-language speakers. On Workana, English is among the lowest-ranked skills needed to succeed in online gig work; in comparison, Spanish is considered more important by online gig workers on Workana (see figure 4.20). Similarly, on SoyFreelancer, survey respon- dents see English language skills as less important than other skills (such as communication skills, time management, and Spanish language skills). The lesser importance of English language skills in the region may be a sign of the growing maturity of the regional online gig work market and the diversity of work opportunities in the local language. The availability of work opportunities in the local language on Workana could contribute to a greater inclusion of workers in the (online) labor market. 4.8 EARNINGS AND INCOME Online gig work is an important means of earning supplemental income. Gig work is a sec- ondary activity for 4 in 10 workers (figure 4.21, panel a), which means they spend 10 to 19 hours and earn 25 to 50 percent of their income through gig work; workers with uneven work patterns are also considered in this group (people spending little time but earning a large share of their income from gig work, or spending substantial time but earning a small share of their income from online gig work; see table 4.2). Around one in three online gig workers is engaged in online work as their main activity, earning a majority of their income from or spending the majority of their working time (more than 20 hours a week) on online gig work, and more than one-quarter do online work only sporadically (that is, as a marginal activity, earning less than 25 percent of their income from and spending less than 10 hours a week on online gig work). A greater share of workers on regional/ local platforms carry out online gig work only as a marginal activity compared to workers on global platforms (46 versus 24 percent), while greater shares of workers on global platforms conduct online gig work as a main or secondary activity. Intensity of gig work differs regionally. In East Asia and the Pacific, a greater share of online gig workers engage in online work as their main occupation (39 percent), while in the South Asia region most online gig workers do such work only marginally (53 percent; figure 4.21, panel b). A comparable study from Europe estimated the share of main gig workers at 11 percent based on data from 2018 and found that most gig workers were secondary gig workers.109 109 C  aveat: this figure also includes workers who perform location-based gig work, based on data collected through a survey conducted in 16 European countries (Urzì Brancati, Pesole, and Férnandéz-Macías 2020, 16). 101 Chapter 4 How Inclusive Is the Online Gig Economy? FIGURE 4.21: Share (%) of online gig workers by intensity of work based on the global RDIT survey a. By intensity of work Marginal Main 26.9 32.38 40.6 Secondary b. By intensity of work and region 100 11 18 33 27 80 35 53 60 58 43 42 41 37 40 25 20 39 26 28 31 31 22 0 SAR LAC SSA ECA MENA EAP Share of workers (%) Main Secondary Marginal Source: Global RDIT survey conducted by the study team. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MENA = Middle East and North Africa; SAR = South Asia region; SSA = Sub-Saharan Africa. Evidence at the platform level also confirms that online gig work is used primarily to earn supplemental income. On SoyFreelancer, a Latin American gig platform, two out of three online gig workers report having another job. Half of them report working for an employer, and over one-quarter report running their own business. On Workana, for most of the respondents, earnings from gig work account for less than 25 percent of their household and individual income, with no significant variations across gender. Still, for almost a quarter of workers, online gig work is the main source of their income. This is consistent with other estimates of 10 percent110 to 30 percent (ILO 2021b). In Bangladesh, respondents to the study survey earned a significant share of their income from freelancing platforms. 110 This figure does not distinguish between location and web-based online gig work (Goldfarb 2019).  102 Working Without Borders: The Promise and Peril of Online Gig Work On average, online gig workers report earning Tk 82,943 per month (equivalent to US$967) from freelancing platforms,111 while the average monthly household income is estimated at Tk 16,000 (equivalent to approximately US$150).112 In Pakistan, the study survey finds that monthly earnings of online gig workers are substantially higher than those of informal workers. Over 90 percent of the informal workers earn less than US$200 per month, while the same parameter is only 41 percent for online gig workers, as shown in figure 4.22. FIGURE 4.22: Monthly incomes of online gig workers compared to informal workers in the Khyber Pakhtunkhwa province, Pakistan 50 40 30 20 10 0 0–$100 $100–$200 $200–$500 $500–$1000 $1000+ Informal workers (LFS, 2020) Online gig workers (KP) Sources: Survey conducted by study team in Khyber Pakhtunkhwa province, Pakistan, and Pakistan labor force survey (LFS), 2020. Note: The figure compares the wages of informal workers in the Khyber Pakhtunkhwa province of Pakistan to the information on income provided by online gig workers in the Khyber Pakhtunkhwa province who completed the gig worker survey conducted by the study team. We conducted a pooled regression analysis by combining data from the Khyber Pakhtunwa survey and the LFS for the KP region. We controlled for age, education, and marital status and found that online gig workers are more likely to be in higher income brackets than similar workers in the informal sector. USD = US dollars. In Africa, household survey data from nine African countries from 2017 and 2018 show that the income earned through gig economy activities is essential for the majority of gig workers (figure 4.23). FIGURE 4.23: Importance of income earned through gig economy activities (% of gig economy participants) It is an important component of my budget, but not essential 16% It is essential for meeting my 53% basic needs It is nice to have, but I 31% could live comfortably without it Source: Chen, forthcoming. 111  he average earnings of online gig workers are based on self-reported information collected through the survey, and T biases can exist. 112 The figure for the average household income is based on the latest information available from the Bangladesh Bureau of  Statistics, through the Household Income and Expenditure Survey from 2016, http://data.bbs.gov.bd/index.php/catalog/182. 103 Chapter 4 How Inclusive Is the Online Gig Economy? With targeted initiatives, online gig work can help bring unemployed people back into the labor market. The eRezeki program of Malaysia is an excellent example of a country that has intentionally used online gig work to increase access to jobs. The program was set up by MDEC to foster the inclusion of underserved citizens, especially low-income citizens, in the labor market. Between 2016 and 2020, on average one in three workers on eRezeki was unemployed upon registering on the platform. In 2019 and 2020, eRezeki took a more targeted approach to engaging users, which resulted in a much larger share of unemployed people joining the platform (in 2019, three in four workers who registered on the platform were unemployed). (More details are given in chapter 7.) In terms of earnings, the gender pay gap among online gig workers is lower than in the general labor market. Data for online gig workers from Argentina using Workana show that, on average, a female online gig worker’s wages are equivalent to 68 percent of her male counterpart’s. In contrast, that figure is only 62 percent for the general labor force (figure 4.24). The same is true for online gig workers from Brazil and Mexico using Workana, though the magnitudes differ. Nevertheless, there is still a considerable wage gap between men and women, even in the online gig economy. FIGURE 4.24: Women’s wages as a percentage of men’s wages for online gig workers using Workana compared to national LFSs 90 82 80 73 72 73 71 68 70 62 59 60 50 40 30 20 10 0 Argentina Brazil Colombia Mexico Online gig workers Total labor force Source: Study team analysis of Workana survey and the latest available national labor surveys in the selected countries, conducted with an Inter-American Development Bank team. Note: The earnings of online workers in the Workana survey are indirectly inferred by asking them, “What is the minimum monthly salary that a full-time salaried job would have to offer for you to stop doing freelance work on Workana (in USD)?” LFS = labor force survey; USD = US dollar. Gig work is becoming increasingly competitive as the supply of gig workers increases. The COVID-19 pandemic exacerbated some of the existing trends in online platform work and increased competition. The notion of remote online work has become more widespread because of the pan- demic and policies to reduce social contacts (Fairwork 2021), but issues of oversupply of labor are increasing, as evidenced by the platform country surveys conducted by the team and other studies (Stephany et al. 2020). In Bangladesh, respondents to the survey confirm that they were affected by COVID-19, primarily by the increase in competition. At the platform level on Workana in Latin America, there is a similar perspective (see figure 4.25). More than one-third of the respondents find that COVID-19 increased competition among freelancers. 104 Working Without Borders: The Promise and Peril of Online Gig Work Workers in developing countries would like to do more gig work but find it hard to access enough well-paying tasks. Skills and reputation are the key assets of online gig workers, but rep- utation is not always easy to build. The anonymous and sporadic nature of gigs means that a prior reputation is critical for access to better-paid or longer-term work opportunities (Wood et al. 2019). This pressure of building a reputation or rating leads to significant stress for gig workers, who often work on short notice and at odd hours or on unfair terms simply to avoid low ratings (Wood and Lehdonvirta 2021). This risk is amplified by the limited transparency in platform policies and processes behind the rating systems (Sutherland et al. 2020; Wood and Lehdonvirta 2021). FIGURE 4.25: Impact of the COVID-19 pandemic on online gig work, according to workers on Workana More competition after COVID as more 36.1 freelancers joined platform No effect of Covid on my freelancing job 26.6 Reduction in Number of projects received 12.4 during COVID Increase in number of projects received 11.0 since COVID Earned less money as freelancer 7.2 Earned more money as a freelancer 6.7 0 5 10 15 20 20 30 35 40 Source: Study team analysis based on Workana survey data. Note: Values are percentages. In terms of career prospects, freelancing is a career path for some online gig workers, though not most. More than one in three online gig workers in Pakistan strive to be entrepreneurs, wanting to start their own agency or grow their existing online freelancing agency. Another 35 percent would like to earn more from their online gig work. Interviews with women online freelancers in Pakistan also show how online freelancing can become not just an activity to earn additional income, but also a career in its own right, allowing women to become entrepreneurs (see box 4.1). Data at the platform level shows that preferences may vary, however. Over 50 percent of respondents in the surveys conducted on Workana and Wowzi confirm that they want to increase their earnings from online gig work, but only about 1 in 10 online gig workers on either platform wants to start or grow a freelancing agency (figure 4.26). On SoyFreelancer, the vast majority wish to grow and earn more as a freelancer (64 percent). Another 20 percent of respondents would like to go beyond the platform work and start their own business. 105 Chapter 4 How Inclusive Is the Online Gig Economy? FIGURE 4.26: Career aspirations among survey respondents on Workana and Wowzi 60 51.3 50.3 50 40 30 19.4 20 17.8 14.3 14 12.8 9.8 10 6.7 3.6 0 Workana Wowzi Workana Wowzi Workana Wowzi Workana Wowzi Workana Wowzi I want to find a I want to I want to learn I will continue I want to earn better full-time job start/grow more digital skills working more money as (not freelancing). myown freelancing so I can work for a intermittently a full-time agency in the company in future as a freelancer. freelancer. future. (not freelancing). Source: Study team analysis based on Workana and Wowzi survey data. 4.9 CONCLUSION Our study finds that online gig work is dominated by youth, giving them the chance to earn money and learn new skills and the flexibility to earn while studying or doing another job. While men make up most of the online gig workers, women are participating in the online gig economy to a greater extent than in the general labor market in similar sectors and occupations. Although still dominated by people with higher education levels, the online gig economy can provide opportunities to a variety of skill levels, particularly of those with high school education. More than 6 in 10 online gig workers are based in tertiary cities—in other words, smaller cities and towns other than the capital or the top 10 largest cities in their country, which points to the role that online gig work could play in addressing regional inequalities in access to jobs. Regional/local platforms offer more opportunities for non-English-speaking workers, thus enabling inclusion in countries where English is not the main language. Overall, gig work remains a secondary means of earning income for most, but not all, gig workers. 106 Working Without Borders: The Promise and Peril of Online Gig Work References Adams-Prassl, Abigail. 2020. “The Gender Wage Gap on an Online Labour Market: The Cost of Interruptions.” CEPR Discussion Papers 14294. Centre for Economic Policy Research, Brussels. https://cepr.org/publications/dp14294. Anwar, Mohammad Amir, and Mark Graham. 2020. “Hidden Transcripts of the Gig Economy: Labour Agency and the New Art of Resistance among African Gig Workers.” Environment and Planning A 52 (7): 1269–91. https://doi.org/10.1177/0308518X19894584. Braesemann, Fabian, Vili Lehdonvirta, and Otto Kässi. 2022. “ICTs and the Urban-Rural Divide: Can Online Labour Platforms Bridge the Gap?” Information Communication and Society 25 (1): 34–54. https://doi.org/10.1080/1369118X.2020.1761857. Braesemann, Fabian, Fabian Stephany, Ole Teutloff, Otto Kässi, Mark Graham, and Vili Lehdonvirta. 2022. “The Global Polarisation of Remote Work.” PLOS ONE 17 (10). https://doi.org/10.1371/ journal.pone.0274630. Cedefop. 2021. “Skill Development in the Platform Economy: Comparing Microwork and Online Freelancing.” Cedefop Research Paper 81, Luxembourg. https://doi.org/10.2801/592284. Chen, Rong. Forthcoming. “Who Are Participating in and Benefiting from the Gig Economy?”. Department of Statistics Malaysia. 2020. “Labour Force Survey Report, Malaysia, 2020.” https:// www.dosm.gov.my/v1/index.php?r=column/cthemeByCat&cat=126&bul_id=dTF2dkJpcUFYU- WRrczhqUHVpcDRGQT09&menu_id=Tm8zcnRjdVRNWWlpWjRlbmtlaDk1UT09. Fairwork. 2021. “Work in the Planetary Labour Market. Fairwork Cloudwork Ratings 2021.” https://fair. work/en/fw/publications/work-in-the-planetary-labour-market-fairwork-cloudwork-ratings-2021/. Goldfarb, Danielle. 2019. “A New Way to Measure Online ‘Gig’ Work.” Presentation to ILO-UNECE Steering Group on Measuring Quality of Employment, November 6, 2019. https://unece.org/ fileadmin/DAM/stats/documents/ece/ces/ge.12/2019/S3_RIWI.pdf. Humans in the Loop. 2020. “Impact Report 2020.” https://humansintheloop.org/resources/ whitepapers-and-reports/impact-report-2020/. IFC. 2021. “Barriers and Opportunities to Refugee Women Engaging in the Digital Economy in Jordan and Lebanon.” IFC, Washington, DC. https://www.ifc.org/wps/wcm/connect/topics_ext_content/ ifc_external_corporate_site/gender+at+ifc/resources/barriers+and+opportunities+to+refugee+- women+engaging+in+the+digital+economy+in+jordan+and+lebanon. ILO. (International Labour Organization). 2012. “The International Standard Classification of Occupations: ISCO-08.” https://isco-ilo.netlify.app/en/isco-08/. ILO (International Labour Organization). 2021a. “Digital Platforms and the World of Work in G20 Countries: Status and Policy Action.” ILO, Geneva. https://www.ilo.org/global/about-the-ilo/ how-the-ilo-works/multilateral-system/g20/reports/WCMS_829963/lang--en/index.htm. ILO (International Labour Organization). 2021b. World Employment and Social Outlook 2021: The Role of Digital Labour Platforms in Transforming the World of Work. Online Labour Observatory. Geneva: ILO. https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/docu- ments/publication/wcms_771749.pdf. 107 References Pesole, A., M. C. Urzí Brancati, E. Fernández-Macías, F. Biagi, and I. González Vázquez. 2018. Platform Workers in Europe: Evidence from the COLLEEM Survey. Publications Office of the European Union. https://doi.org/10.2760/742789. Stephany, Fabian, Michael Dunn, Steven Sawyer, and Vili Lehdonvirta. 2020. “Distancing Bonus or Downscaling Loss? The Changing Livelihood of Us Online Workers in Times of COVID-19.” Journal of Economic and Human Geography 111 (3): 561–73. https://doi.org/10.1111/tesg.12455. Sutherland, Will, Mohammad Hossein Jarrahi, Michael Dunn, and Sarah Beth Nelson. 2020. “Work Precarity and Gig Literacies in Online Freelancing.” Work, Employment and Society 34 (3): 457–75. https://doi.org/10.1177/0950017019886511. UNDESA (United Nations Department of Economic and Social Affairs), Population Division. 2022. “World Population Prospects 2022 Summary of Results.” United Nations, New York. https:// www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/wpp2022_sum- mary_of_results.pdf. Urzì Brancati, M. C., A. Pesole, and E. Férnandéz-Macías. 2020. “New Evidence on Platform Workers in Europe. Results from the Second COLLEEM Survey.” Joint Research Centre Paper JRC118570, Publications Office of the European Union, Luxembourg. https://doi.org/10.2760/459278. Wood, Alex J., Mark Graham, Vili Lehdonvirta, and Isis Hjorth. 2019. “Good Gig, Bad Gig: Autonomy and Algorithmic Control in the Global Gig Economy.” Work, Employment and Society 33 (1): 56–75. https://doi.org/10.1177/0950017018785616. Wood, Alex J, and Vili Lehdonvirta. 2021. “Platform Precarity: Surviving Algorithmic Insecurity in the Gig Economy.” Online Workshop at the University of Sheffield, jointly hosted by the Sheffield Institute for Commercial and Corporate Law (SICCL) and the Sheffield Political Economy Research Institute (SPERI). 108