Policy Research Working Paper 10927 Characterizing Green and Brown Employment in India Andrés Ham Emmanuel Vazquez Monica Yanez-Pagans Education Global Practice September 2024 Policy Research Working Paper 10927 Abstract Transitioning toward sustainable development practices is and brown employment, document patterns between and expected to result in broad changes to economic activity, within occupations, characterize workers by attributes and which will subsequently impact labor markets and change skills, and study wage differentials. The results highlight the demand for skills. India established the Skill Council the importance of monitoring green and brown jobs with for Green Jobs to identify green jobs and define the skills robust labor market monitoring systems to guide decisions required for these occupations. This paper applies the Skill on the sustainability transition and suggest key aspects to Council for Green Jobs definition of green jobs and an inter- consider when investing in green skills and the potential national definition of brown jobs to data from the 2019–20 distributive consequences of sustainability policies on the Periodic Labour Force Survey to estimate the size of green population. This paper is a product of the Education Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at myanezpagans@worldbank.org, a.ham@uniandes.edu.co, and evazquez@cedlas.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Characterizing Green and Brown Employment in India* Andrés Ham † Emmanuel Vazquez ‡ Monica Yanez-Pagans§ Keywords: climate change, sustainability, labor markets, green jobs, brown jobs, worker skills. JEL Classification: J23, J24, J49, Q01, Q56. * This document is a background paper for the World Bank’s India Country Climate and Development Report (P179782). We are grateful to Keiko Inoue, Toby Linden, Josefina Posadas, Stephane Hallegatte, Shobhana Sosale, and Xiaoyan Liang for valuable feedback and comments during peer review that have greatly improved this research. † Corresponding Author. Associate Professor, School of Government, Universidad de los Andes. Email: a.ham@uniandes.edu.co. Physical Address: Carrera 1 #19-27, Edificio Aulas, Piso 3; Bogotá, 111711, Colombia. Tel: +57 601 339-4949, Extension 5317. ‡ Senior Researcher, Center for Distributive, Labor and Social Studies (CEDLAS), Instituto de Investigaciones Económicas, Facultad de Ciencias Económicas, Universidad Nacional de La Plata. Email: evazquez@cedlas.org. § Senior Economist. Education Global Practice. The World Bank. Email: myanezpagans@worldbank.org. 1. Introduction Climate change is currently at the top of the development agenda in most countries, due to its expected consequences on multiple dimensions of well-being, including food security (Wheeler and von Braun, 2013); agricultural production, conflict, and economic growth (Carleton and Hsiang, 2016); land use (De Chazal and Rounsevell, 2009); health (McMichael and Haines, 1997); and other outcomes. The challenges that arise from climate adaptation and mitigation ambitions have led to the creation of country-specific strategies for sustainable development. However, the distributional consequences of these strategies may vary, with diverse countries requiring different strategies to combat changing climate conditions (Mendelsohn et al., 2006). Transitioning towards sustainable development requires immediate and structural changes. While the former require rapid decisions, the latter should not be overlooked. For instance, lowering emissions will likely have a direct effect on economic production and a subsequent effect on labor markets (Govindhan and Bhanot, 2014). Some jobs will be created, benefitting some workers, while others will be destroyed and thus affect other employees. To connect the transition towards sustainability and adapt to new labor markets, the “Green Jobs Initiative” was proposed by the United Nations Environment Programme, the International Trade Union Confederation, the International Organization of Employers, and the International Labour Organization, to promote opportunities, equity, and a fair transition towards sustainability (Stanef-Puică et al., 2022). This initiative intends to promote ‘green’ employment and reduce the number of ‘brown’ jobs. Green jobs are occupations that have a positive impact on the planet and contribute to environmental welfare (Bowen, 2012), while brown jobs are occupations in industries that are pollution intensive (Vona et al., 2018). Evidence on the prevalence of green and brown occupations and the skills and 2 attributes of workers in both types of jobs is essential to understand the labor market and distributional challenges of transitioning towards sustainable development (Rubio et al., 2022). This paper analyzes patterns in green and brown jobs, as well as the characteristics of their workers, using India’s national definition for green occupations and Vona et al.’s (2018) classification for brown occupations. India intends on being energy independent by 2040 and have net zero emissions by 2070. This proposed transition to a net-zero carbon economy is expected to result in significant changes to the economy and the labor market, increasing the demand for sustainable occupations that require specific skills while reducing the need for brown jobs. India has addressed green skills development through its Skill Council for Green Jobs (SCGJ), established in 2016 with a mission to identify skilling needs of service users, manufacturers, and service providers within the green business sector, and implement nation-wide, industry-led, collaborative skills, and entrepreneur development initiatives (Pavlova, 2019). The SCGJ focuses on the renewable energy, transportation, waste management, construction, and water management sectors (Bishnoi and Rai, 2022). Through collaboration between industry and its 200 training centers, 10 assessment agencies, and 400 certified trainers, the SCGJ has rolled out a range of training programs, focused on sectors such as solar and wind energy. As part of this work, the SCGJ has established a national definition of green jobs and developed qualification files for each of these occupations linked to its National Skills Qualification Framework. These qualification files clearly define the green skills and competencies required to support a sustainable transition. We employ data from the 2019-20 Periodic Labour Force Survey (PLFS) to perform our analysis. The PLFS is a nationally representative survey that provides statistics for urban and rural areas, at the state level, and by some individual attributes. We apply India’s national definition for green jobs and Vona et al.’s (2018) classification of brown jobs to occupation codes in the PFLS 3 to identify these jobs in the data. We then calculate descriptive statistics and profile workers in all jobs across sociodemographic attributes. While our results are mostly descriptive, we are unaware of previous research that studies green and brown occupations using a joint approach. We hope this evidence contributes to identify and diagnose existing challenges to green transitions and helps inform policy decisions that may facilitate the transition towards sustainable development, while highlighting the empirical challenges that arise when measuring green and brown jobs using labor force surveys frequently collected in most countries that have measurement limitations. We find that green jobs account for about 5.9 percent of all employment when using the national definition, while 4.6 percent of all jobs can be classified as brown when using Vona et al.’s (2018) definition. There is a heterogeneous distribution of green and brown occupations across sectors. Some sectors are mixed, since they include workers that are employed in both green and brown occupations. Green jobs seem to be more prevalent in states with higher per-capita GDP, but we observe no relationship between the percentage of brown jobs and local GDP. We find that green and brown workers tend to be male, younger, and slightly more educated. Green and brown jobs are slightly worse off compared to other jobs in type of contract, social security benefits, employment categories, and firm size. We find suggestive evidence of a significant wage premium for green jobs and an earnings penalty for brown jobs, which varies by skill level. The remainder of this paper is organized as follows. Section 2 describes the different approaches to measure green and brown jobs. Section 3 describes our data and how we identify green and brown jobs. Section 4 presents a descriptive analysis of green and brown jobs. Section 5 characterizes workers in green and brown occupations. Section 6 explores differences in compensation between green, brown, and all other occupations. Section 7 concludes with a discussion on our findings, their policy implications, and highlights directions for future research. 4 2. Measuring green and brown jobs Let us begin by defining what it means for a job to be either green or brown. “Green jobs” refer to occupations that are environmentally sustainable, which have a direct positive impact on the planet and contribute to overall environmental welfare. They include jobs which seek to use or develop renewable energy, conservation of resources, ensuring energy efficient means, waste management, and sustainable development. “Brown jobs” are occupations in pollution-intensive industries, which result in the contamination of water, land, and air, loss of biodiversity, exhaustion of natural resources like water, fish, land, and fossil fuel extraction, among others. While arriving at a conceptual definition of green and brown jobs is intuitively simple, there is no unique or straightforward way to identify them in practice. Identifying environmentally sustainable (green) and pollutant (brown) jobs is complex because they can be defined in multiple ways. For instance, we may consider firms’ output, their technology, workers’ skills, or occupation task content when trying to identify them in the data (Granata and Posadas, 2024). The literature identifies four potential ways to measure green jobs (Granata and Posadas, 2024). First is the output approach, which focuses on firms and the goods and services they produce. For example, all workers employed by a firm that produces electric cars would be considered to have a green job under this approach. Second, is the technology approach, that also looks at firms but focuses on their production technologies. For example, all workers in a firm that prepares food using energy efficient refrigerators would be considered green. Third, is the skills approach, focused on worker abilities required to produce goods and services. A person who has the skills required to conduct economic analysis related to the environment, independently of the sector in which she works may be classified as having a green job. Fourth is the task content approach, where the focus lies on the workers and tasks required to produce goods and services. 5 A person who conducts economic analysis (does not just have the skills to do so) related to the environment can be classified as having a green job, independently of the firm or sector in which she works. The suggested method to measure green and brown jobs in practice is the latter, the task content approach. Granata and Posadas (2024) developed a methodology that can be applied to emerging economies to identify green jobs using information collected in most labor force surveys, which tends to classify workers using the skills required for their occupations. They claim that this approach is also the most suitable one to inform skills policy because it allows understanding whether an occupation has specific tasks that promote environmental sustainability. These authors develop a dictionary of common green terms relevant to emerging economies, 5 apply the green terms dictionary to task statements of all occupations included in the ILO’s International Standard Classification of Occupations (ISCO) using text analysis, and then calculate a Green Task Intensity (GTI) index by occupation. This GTI measures the share of tasks within each occupation that directly contribute to the greening process. Formally, the Green Task Index for occupation o is defined as follows: # 0 0 = # 0 The GTI provides a list of occupations with their percentage of green tasks from 0 to 100. The higher the GTI, the larger the number of green tasks performed in that occupation. When 5 These terms are drawn from multiple sources, such as: common terms in the environmental economics literature; terms from the Green Technologies and Practices survey from the US Bureau of Labor Statistics; Occupational Information Network (O*NET) task statements from the US Department of Labor; and Skills and Competencies from the European Skills, Competencies, Qualifications, and Occupations Taxonomy (see Granata and Posadas, 2024 for details). 6 applying this procedure to 3,245 task statements in a comprehensive occupations database, 329 are classified as green using a broad version of the dictionary and 83 are classified as green under a narrow version of the dictionary. The former 329 are considered green tasks, while the latter 83 are considered strictly green tasks. To determine if an occupation is green, the GTI index should be greater than 0 under either the broad or narrow definition. Therefore, green occupations are defined as those in which workers perform at least one green task. This procedure results in a list of occupations using the ILO’s ISCO-08 at 4-digit level that are classified as green jobs (see Granata and Posadas, 2024). While there is general agreement on this task content approach when carrying out international comparisons of green jobs using labor force surveys, very few emerging countries have adopted a national classification for green jobs. India stands out in this regard, as the SCGJ has established a national definition of green jobs and clearly outlined the skills and competencies required to perform those jobs. The SCGJ’s national definition includes 44 occupations related to the production of renewable energy, environment, forest, climate change, and sustainable development. The list with these 44 green occupations, which are all identifiable when applying a crosswalk between occupational codes described below, is shown in Appendix Table A.1. 6 While the SCGJ has a national approach to identify green jobs, there is no corresponding national definition for brown jobs. We propose using the definition employed by Vona et al. (2018), whose taxonomy of brown jobs has been used by most research that aims to measure and 6 The 44 green jobs are described in detail in the SCGJ’s “Green Jobs Handbook” published in 2022. Renewable energy includes solar PV & solar thermal, wind, hydro, energy storage, biomass power/waste to energy, clean cooking stoves, biofuel, and biogas. Environmental, forest, and climate change include solid waste management, water management, e-waste management, and carbon sinks. Sustainable development includes green construction, green transportation, pollution prevention & control, green hydrogen, and energy storage. 7 study pollutant occupations and transitions towards sustainability in labor markets across the world (e.g., Mealy et al., 2018; Marin and Vona, 2019; Cerimelo et al., 2022; Rubio et al., 2022). Vona et al.’s (2018) definition of brown jobs uses the Standard Occupational Classification (SOC) from 2010, which is the standard occupation system used in the United States, at the 6-digit level. They start identifying pollution-intensive industries and then occupations within these industries. First, they select pollution-intensive industries using emissions of the six air pollutants (CO, VOC, NOx, SO2, PM10, PM2.5, and lead) and CO2 emissions. Pollution-intensive industries are defined as those with 4-digit North America Industry Classification System (NAICS) industries that are in the 95th percentile of the pollution intensity (measured in terms of emissions per worker) for at least three pollutants. Based on this, they identify a total of 62 pollution-intensive industries or ‘brown’ industries. Second, they identify brown occupations within these industries as those which are most prevalent in these 62 pollution-intensive industries. These are identified as those with a probability of working in polluting sectors being seven times higher than in any other job. This procedure results in a total of 87 brown jobs using the SOC-2010 at the 6-digit classification, which we list in Appendix Table A.2. We also require implementing a crosswalk to apply this definition since Vona et al. (2018) use different occupational codes than those found in our data. How reliable are these methodologies to accurately trace out green and brown jobs? These methods rely on the assumption that certain skills are compatible with either green or brown occupations. However, Granata and Posadas (2024) note that the same occupation may have skills compatible with both types of occupations, so while the task content approach to measurement is the most useful from a policy perspective, it does have its limitations whether it uses international standards, those for the United States or are defined by a specific country. For instance, while the national definition of green jobs in India responds directly to the needs identified by the 8 government, this taxonomy is a work in progress that will likely change over time. One way to ensure that India’s definition of green jobs is consistent is to compare it with other approaches such as the O*NET definition that uses a task content approach for the United States and the one proposed by Granata and Posadas (2024) that aims to be used widely in emerging economies. With respect to brown jobs, there are fewer available definitions to use and compare. The Indian government only identified the skills needed to conduct green jobs but did not identify those for pollutant jobs. Therefore, the available method relies on task content definitions for the United States, which may not accurately capture the industry composition and labor market structure in India. While developing a new taxonomy of brown jobs either in India or other developing countries is beyond the scope of this paper, we do believe that it is difficult to speak about the importance of a sustainable transition towards green employment without analyzing brown jobs. Therefore, measuring and characterizing both types of occupations provides a broader view of the policy challenges for environmental sustainability. 3. Data and definitions 3.1. Data We employ data from the 2019-20 India Periodic Labour Force Survey (PLFS), which is collected by the National Statistical Office of the Ministry of Statistics and Program Implementation (MoSPI). This is a nationally representative survey that covers the majority of the Indian Union, except for the villages in the Andaman and Nicobar Islands, which are difficult to access (MoSPI, 2016). The PLFS data for this round were collected between July 2019 and June 2020, and include information on household characteristics, individual-level demographics, educational attributes of respondents, as well as labor market information over the past 12 months and 7 days prior to the interview. Despite the availability of more recent rounds of this survey, we do not use them to 9 avoid potential biases due to effects from the COVID-19 pandemic, 7 which impacted how data were collected and may result in atypical estimates and wrongful conclusions. While the survey allows obtaining nationally representative statistics, it is also designed to provide reliable indicators disaggregated by urban and rural areas, at the state level, and gender. Given that the PFLS is a labor force survey, it collects detailed information on job attributes that we require to identify green and brown occupations with the methodologies described in the previous section. We rely on the primary occupation variable in the PFLS, which employs the 2004 National Classification of Occupations (NCO) at the 3-digit level to classify workers into occupations. However, several steps and assumptions are required to link observations in the PLFS data to the definitions of green and brown jobs that we described in the previous section. We describe these assumptions and empirical choices in this section, highlighting their limitations. 3.2. Identifying green occupations in the PFLS data The national definition of green jobs developed by the SCGJ employs ILO’s ISCO-08 at the 4- digit level for the classification of the occupations. However, the PLFS employs India’s National Classification of Occupations (NCO) from 2004 at the 3-digit level. Therefore, to link the SCGJ’s national definition of green jobs to the Indian PLFS data, we follow a three-stage approach. First, we use the correspondence developed by Khurana and Mahajan (2020) between NCO-2004 at 3- digits and ISCO-88 at 3-digits to translate the NCO-2004 occupation codes included in the 2019- 20 PLFS to ISCO-88 at 3-digit level. The codes are the same in 109 of the 113 occupations. Among the four that are different, two have the same description and are assigned a different number and 7 Some studies have looked at how the pandemic directly affected the green transition in India (see Kedia et al., 2020; Saxena et al., 2021, and Marimuthu et al., 2022), but this is not our objective in this paper. 10 the other two are assigned to a more general or specific group, representing a small percentage of 0.44% of all employment. 8 Figure 1. Construction of database to measure green employment Source: Own elaboration. Second, we use the ILO’s correspondence table between ISCO-88 at 4-digit level and ISCO-08 at 4-digit level to convert the ISCO-88 at 3-digit level to ISCO-08 at 4-digit level. To carry out this conversion, we assume an equal distribution of workers within each occupation, which is something done quite frequently in the labor economics literature in the absence of additional information to estimate the distribution of workers within occupations. 9 Last, we use 8 The two occupations with the same description and a different number are “Agricultural, Fishery and Related Labourers” (920 in NCO-2004, but 921 in ISCO-88) and “Subsistence Agricultural and Fishery Workers” (920 in NCO-2004, but 921 in ISCO-88). The reported occupation “Other Teaching Professionals” is coded with the number 233 in the NCO-2004 and assigned to the more general group “Teaching professionals” (code 230) in the ISCO-88, while the reported occupation “General Managers” in the NCO-2004 is assigned to the specific ISCO-88 code 131 (“Managers of small enterprises”) in the Khurana and Mahajan (2020) correspondence. For more information, see https://github.com/worldbank/gld/blob/main/Support/B%20- %20Country%20Survey%20Details/IND/EUS/Correspondence_National_International_Classifications.md. 9 This assumption is only relevant when there are correspondences between green and non-green occupations, which means that it is effectively applied in 16% of all ISCO-88 at 4-digit level occupations in the conversion ISCO-88 at 11 the ISCO-08 classification at the 4-digit level to merge the SCGJ’s national definition of green jobs to the Indian 2019-20 PLFS data. Figure 1 presents a summary of the three-stage approach we use to identify green jobs in our data based on the SCGJ's national definition. As Granata and Posadas (2024) mention, any methodology we employ to measure green jobs in household surveys requires making assumptions. In this case, since we cannot apply the SCGJ’s national definition directly to the PFLS data, we must translate occupational codes to be able to analyze green employment in India. We compare our results from this methodology with other approaches in Section 5. This comparison may lend support to the reliability of the SCGJ’s definition, although we do highlight that this area of study is active and may change quickly. 3.3. Identifying brown occupations in the PFLS data As with the SCGJ’s national definition for green jobs, the correspondence between the international definition of brown jobs and the data in the PFLS is not a one-to-one match. We need to translate brown occupations that use SOC-2010 codes at 6-digits into ISCO-88 at 3-digits. To link Vona et al.’s (2018) definition of brown jobs to the Indian PLFS data, we also follow a three-stage approach. First, we map the brown occupations identified by Vona et al. (2018) using the 2010-SOC at 6-digit to the ISCO-08 at 4-digit classification using a crosswalk from the US Bureau of Labor Statistics. When a single ISCO-08 occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is estimated, which reflects the proportion of corresponding 6-digit SOC occupations that have been classified as brown within the ISCO-08. Second, we convert this measure of brown jobs at the ISCO-08 4-digit level to ISCO-88 at 4-digit using the ILO's crosswalk. Similarly, when an ISCO-88 is associated to more than one 4-digit - ISCO-08 at 4 digits and in 28% of all ISCO-88 at 3-digit level occupations in the conversion ISCO-88 at 4- digit - ISCO-88 at 3 digits. 12 ISCO-08, we assume an equal distribution of workers in ISCO-08 occupations within an ISCO-88 occupation, and therefore use a brown proportion measure that reflects the proportion of corresponding ISCO-08 occupations that have been classified as brown within the ISCO-88. Last, brown jobs are estimated at the ISCO-88 3-digit level assuming an equal distribution of workers in 4-digit occupations within each 3-digit occupation. Figure 2 presents a summary of the approach we use to identify brown jobs based on this international definition. Figure 2. Construction of database to measure brown employment Source: Own elaboration. As with the employed definition to measure green jobs, there may be some limitations to identify brown jobs since the occupation codes proposed by Vona et al. (2018) are not the same as those found in the PFLS survey. Additionally, there is a second limitation in measuring brown jobs, since contrary to green jobs that have at least two different methodologies to identify them that we can compare, we only have one available method at the time of writing for brown jobs. The existing definition was created with the US as a reference and using it in other contexts may not accurately capture the industry composition and the labor market particularities in India. While 13 developing a new taxonomy of brown jobs is beyond the scope of this paper, we do expect future research to address this issue since studying both green and brown jobs is essential to better understand and discuss the importance of a sustainable transition for labor markets worldwide. We hope that our findings promote greater discussion not only for measuring green employment more accurately, but also on how to identify brown jobs in developing countries with greater precision. 3.4. Arriving at mutually exclusive categories Given that green jobs are defined using the SCGJ method while brown jobs are identified using Vona et al.’s definition, there is the possibility of some overlap between brown and green jobs in the PFLS data because of the assumptions that need to be made to merge the more aggregated (with 3-digit code) PLFS data to the definitions that we use (4-digits for green jobs and 6-digits for brown jobs) when using the relevant crosswalks. This leads to some occupations not being fully green nor fully brown, but a mix between the two. Some degree of overlap is reasonable given the translation required since the survey data are less disaggregated than the definitions that we employ to measure green and brown jobs, as shown by the different digits of disaggregation. About 7.9% of all occupations in the PFLS had both a percentage of green and brown tasks when applying the definitions (9 in total).10 For the 9 jobs with both green and brown tasks, we rescale the percentages so that the sum of green, brown, and other tasks equals 100. For example, if an occupation has 20% green tasks, 20% brown tasks, and 80% neither green nor brown tasks, we divided each of these numbers by 120% to rescale the weights to 16.67, 16.67, and 66.67%. This standardization ensures that we do not double count any green or brown tasks. If the 10 These occupations are: Electrical and Electronic Equipment Mechanics and Fitters, Chemical-Products Machine Operators, Rubber and Plastic-Products Machine Operators, Wood-Products Machine Operators, Printing-, Binding- And Paper-Products Machine Operators, Textile-, Fur- And Leather-Products Machine Operators, Food and Related Products Machine Operators, Assemblers, Other Machine Operators and Assemblers. 14 correspondence between surveys was a one-to-one match, this would not be necessary, but it is required in this case because of the need to translate definitions across codes. We do note that this assumption may accurately capture that the job has both green and brown tasks, but it could also reflect that the 3-digit occupation in the surveys includes narrowly defined jobs that are green and others that are brown. While we cannot disentangle which of the 9 jobs with this mix is the former or the latter, this should be an important discussion for future work in this area, since applying theoretical definitions to observational data will present similar challenges in other contexts. We return to this point when discussing the policy implications of our findings in the last section. 4. Patterns and trends in green and brown employment This section presents descriptive statistics on the size of green and brown employment in India, as well as patterns and trends to characterize these occupations. Figure 3 shows the percentage of green and brown employment after implementing the definitions discussed in the previous sections. About 5.9 percent of jobs in India are green, which amounts to approximately 27.1 million workers. In turn, 4.6 percent or 21.2 million workers are employed in brown occupations. However, most jobs in India are neither green nor brown under these definitions (89.5 percent). 11 For green jobs, we can calculate alternative statistics on the prevalence of green jobs using other definitions frequently used in the international literature. These include the O*NET definition for the US and Granata and Posadas’ (2024) green jobs taxonomy for developing countries. 12 These alternative definitions of green jobs support the finding that India’s green 11 We use the 2019 Periodic Labor Force Survey sampling weights as a base to compute all the statistics in this paper, except to estimate the total population (and number of workers), since they are underestimated in the survey. Therefore, we rescaled the sampling weights in the PFLS using the official projected population at the state-level for 2019 published in the Population Projection Report 2011-2036 of the Indian Ministry of Health and Family Welfare. 12 O*NET is an international definition of green jobs based on a taxonomy developed for labor markets in the United States (https://www.onetcenter.org/dictionary/22.0/excel/green_occupations.html). GP is an international taxonomy 15 workforce is small, ranging between 9 and 14 percent on average, suggesting that the SCGJ’s definition seems to be robust with other forms of measurement, although narrower because green employment is lower than the other definitions. The exception is the European Classification of Occupations, Skills ad Competences (ESCO) definition, which uses a less strict approach to classify a job as green than the other definitions. A simple correlation between green job definitions is positive and ranges from 0.24 to 0.29, suggesting that the definitions do have some overlap, but capture slightly different concepts of what each considers to be green skills. Figure 3. Prevalence of green and brown jobs using alternative definitions in India, 2019 100% 8% 80% 60% 86% 91% 94% 95% 92% 40% 20% 14% 9% 6% 5% 0% O*NET GP definition of ESCO definition SCGJ definition Brown jobs definition of green jobs of green jobs of green jobs green jobs Source: Own estimates using the India Periodic Labor Force Survey (2019). Notes: O*NET is an international definition of green jobs based on a taxonomy developed by the USA using occupational information. GP is an international taxonomy of green jobs developed by Granata and Posadas (2024) applying task-content text analysis to the ISCO-08 occupations. ESCO is the definition used by the European Classification of Occupations, Skills ad Competences in which green jobs are occupations for which at least one of the essential skills used is green. SCGJ refers to the government’s definition of green jobs developed by the Skill Council for Green Jobs using 4-digit ISCO-08 occupations. At-risk occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC using pollution-intensive industries. We also estimate the prevalence of green and brown jobs with the national definition using the 2007, 2009, and 2011 Employment and Unemployment Survey (EUS), as well as the 2018 and of green jobs developed by Granata and Posadas (2024) applying task-content text analysis to the ISCO-08 occupations. 16 2019 Periodic Labour Force Survey (PLFS) for India to see whether there are any trends. We find that the percentage of green jobs has increased from 5 to 6 percent in a decade. The percentage of brown occupations stays relatively constant around 4.5 percent over this period. These results suggest that green jobs are increasing at a faster rate than brown jobs, suggestively indicating that the green transition seems to be favoring growth in environmentally sustainable employment. Figure 4. Share of green and brown employment in India by sector of activity, 2019 100 3 Other Services 6 1 2 9 2 2 90 4 5 2 Public Administration 3 3 80 7 6 Financial and Business 70 30 14 Services Transport and 60 8 Communications 39 1 1 9 Commerce 50 0 Construction 40 3 30 Public utilities 53 21 48 20 Manufacturing 1 10 Mining 14 0 1 2 Agriculture Green Brown Rest Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: (1) Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. (2) Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. Next, we explore how green and brown employment is distributed by sector of activity. Figure 4 shows the share of green, brown, and rest of workers by broad economic sectors. The sectors with the largest share of green employment are construction (39 percent), manufacturing (21 percent), and agriculture (14 percent). Most brown jobs are in manufacturing (53 percent) and 17 construction (30 percent). These results show that there is a heterogeneous distribution of green and brown jobs by economic sector. In fact, looking within each sector, we find that green jobs are a larger fraction of all workers in public utilities and construction, while brown jobs are a larger share of all workers within mining and manufacturing. Moreover, the results suggest that both construction and manufacturing face considerable opportunities in driving the green transition, but also important challenges as they would need to strive to reduce their overall brown footprint. We proceed to explore the distribution of green and brown employment across geographical locations. Thirty-seven percent of green workers live in urban areas, while 42 percent of brown workers are also in these areas. Most workers in India therefore reside in rural areas, which also applies to jobs that are neither green nor brown. In unreported results, we find that there is a greater share of green and brown employment in stratums 1 and 2. We also estimate that there are slightly more green jobs in districts with a larger share of urban population (measured in quintiles). Figure 5 presents maps of green and brown employment shares by State/Union Territory (UT) to better capture geographic differences in the prevalence of green and brown employment across India. All States have some fraction of green and brown workers, but states where there is a larger prevalence of green jobs are not necessarily those where there are more brown occupations. Green workers are largely found in the UTs of Damam & Diu and Dadra & Nagar Haveli, where they represent approximately 14 and 15 percent of the workforce, respectively. Brown workers are mainly concentrated in Damam & Diu and Dadra & Nagar Haveli, making up about one-fifth of the workforce in those locations. The States with more green jobs after those are Punjab, Delhi, and Haryana; while those with more brown jobs after the top two States are Gujrat, Tamil Nadu, and Pondicheri. 18 Figure 5. Green and brown employment shares by State/Union Territory (in percent), India, 2019 Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. 19 While green and brown workers are more prevalent in the aforementioned states, their presence in the overall workforce of each state and UT is relatively modest (at most 9 percent), except for Damam & Diu and Dadra & Nagar Haveli where their representation is in the double digits (see Figure 6). This reinforces the general finding that green employment remains a small fraction of total employment in India, but that there is heterogeneity in its distribution over states. Figure 6. Green and brown employment shares within State/Union Territory (percent), India, 2019 100 90 80 70 70 67 60 84 88 85 85 90 90 90 92 95 90 93 88 93 92 89 88 87 91 92 92 87 86 86 90 92 89 92 94 91 92 89 89 50 40 30 20 16 18 10 5 6 6 5 7 7 6 6 4 2 6 4 6 5 7 14 15 4 4 4 3 4 3 4 2 3 7 3 7 8 5 3 4 3 4 10 7 2 8 7 5 8 7 9 9 0 6 6 3 4 4 4 5 5 4 5 5 6 5 7 4 4 6 6 3 7 6 Sikkim Daman & Diu Dadra & Nagar Haveli Goa Andaman & Nicober Telangana Uttaranchal Haryana Rajasthan Bihar Tripura Meghalaya Madhya Pradesh Gujrat Delhi Nagaland Himachal Pradesh Lakshadweep Kerala Uttar Pradesh Manipur Odisha Karnataka Chandigarh Assam West Bengal Andhra Pradesh Tamil Nadu Mizoram Jharkhand Pondicheri Chhattisgarh Maharastra Punjab Green occupations Brown occupations Rest of occupations Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. We also analyze whether the percentages of green and brown occupations are correlated with state per capita GDP in Figure 7. There is a positive relationship between the share of green jobs and per capita GDP, but no visible correlation between brown jobs and GDP. This suggests that areas of the country that have a higher concentration of green jobs are also the ones with the highest per capita income on average. The same does not apply when we consider brown jobs. Figure 7. Percentage of green and brown occupations and per capita gross domestic product by State/UT, 2019 12% Andhra Pradesh Tamil Nadu 10% Himachal Pradesh Punjab Haryana Delhi Green occupations (%) Tripura Uttaranchal 8% Chandigarh Jharkhand Pondicheri Goa Rajasthan 6% Uttar Pradesh Andaman & Bihar Sikkim West Bengal Nicober Chhattisgarh MaharastraKarnataka 4% Manipur Gujrat Odisha Mizoram Telangana Meghalaya Assam Nagaland 2% Madhya Pradesh Kerala 0% 4.6 4.8 5.0 5.2 5.4 5.6 5.8 Log (Per capita product) 8% Pondicheri Tamil Nadu 7% Manipur Punjab Gujrat 6% Haryana Brown occupations (%) West Bengal Delhi Jharkhand Himachal Pradesh 5% Odisha Tripura Telangana Bihar Andaman & Goa 4% Uttar Pradesh Assam Nicober Chhattisgarh Kerala Andhra Pradesh Rajasthan 3% Meghalaya Mizoram Chandigarh Nagaland Karnataka Sikkim 2% Madhya Pradesh 1% Uttaranchal Maharastra 0% 4.6 4.8 5.0 5.2 5.4 5.6 5.8 Log (Per capita product) Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India and Per Capita Net State Domestic Product in 2019-20 from Reserve Bank of India. Notes: Green occupations are those with any of the 44 green qualifications defined by the SCGJ. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. 21 5. Who works in green and brown jobs? To better understand the individuals working in green and brown employment, we now characterize them by demographics, skills, employment indicators, and wages to profile workers who are employed in green and brown jobs in India, and how they compare between each other and with other employees in occupations that are neither green nor brown. We provide a specific focus on the skill distribution within and between occupations since the transition towards sustainable development emphasizes certain required skills for workers. Table A.3 in the Appendix summarizes our main findings and compares attributes within green, brown, and the rest of occupations, conducting mean tests to identify statistically significant differences across groups. 5.1.Profile of green and brown workers in demographic attributes In Figure 8, we present the distribution of green and brown workers by gender. The disparities in labor force participation between men and women in India is well known – according to national estimates of labor force participation, in 2020, only 27 percent of females aged 15 and above participate in the labor force vis-à-vis 75 percent of men in the same age group. The strikingly low participation of women in India’s labor force reflects deep-rooted societal norms and economic barriers and is widely documented as a key barrier for economic growth (See Banerjee et al., 2013; Klasen and Pieters, 2015; Carranza, 2014; and Chatterjee et al., 2015). In terms of green jobs, this pattern is largely replicated. Green jobs are male dominated – 78 percent are held by men and 22 percent by women. This gender gap is larger when compared to employment rates in non-green occupations. Similar patterns emerge when analyzing brown jobs, where 79 percent of brown workers are men and 21 percent are women. The gender gap is also larger in brown occupations compared to other occupations that are neither green nor brown. 22 Figure 8. Share of green and brown employment in India by gender, 2019 100 90 80 70 60 78 79 73 50 40 30 20 10 22 21 27 0 Green Brown Rest Female Male Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3- digit occupation. Figure 9 presents the share of green and brown employment by age groups. Green workers are relatively younger, 43 percent are between 15 and 34 years old compared to 36 percent among non-brown nor green workers. Brown workers are also relatively younger, with 46 percent between the ages of 15 and 34. Few older workers seem to be employed in green or brown jobs, with at most 10-11 percent of workers above the age of 55 years working in these occupations. 23 Figure 9. Share of green and brown employment in India by age groups, 2019 100 11 10 90 17 80 19 18 70 21 60 27 27 50 26 40 30 28 29 25 20 10 15 17 11 0 Green Brown Rest [15-24] [25-34] [35-44] [45-54] [55+) Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6- digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO- 88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. 5.2.Profile of green and brown workers by educational attainment and skill level Figure 10 shows the distribution of workers by their educational attainment in green, brown, and other jobs. Overall, the differences in educational attainment across the three types of workers are not large. About two-thirds of workers in all categories have not completed secondary, while the remaining third have some secondary or university studies. In terms of vocational education and training received by green and brown workers, we find that about 1 in 4 green and brown workers have some formal vocational training. These vocational training programs taken by green and brown workers tend to last between 1 and 2 years and were largely funded by the workers themselves or their families (72 percent of green workers used private funds to get vocational 24 training vis-à-vis 63 percent for brown workers). However, educational attainment and worker skills are two different concepts, so we proceed to study differences by skill level. Figure 10. Distribution of green and brown employment in India by educational attainment, 2019 100 10 7 2 13 90 3 1 University incomplete or 80 23 complete 21 21 Higher than secondary but 70 not university 60 Secondary complete 22 26 20 50 Secondary incomplete 40 14 12 Primary complete 15 30 7 6 Primary incomplete 7 20 24 25 No education 10 20 0 Green Brown Rest Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. In Figure 11, we classify green and brown workers into three groups based on skills. The classification of skills that we use is based on the ISCO-08 skills classification defined using the primary job at 7-day recall. ISCO-08 classifies occupations in four categories based on two dimensions of skills – skill level and skill specialization (ILO, 2012). The four ISCO-08 skill level 25 categories are low (category 1), medium (category 2), and high (categories 3 and 4). We employ this classification of occupations by skills to study differences across workers for the remainder of this section. Low skilled workers include elementary occupations. Medium skilled workers include plant and machine operators, assemblers, skilled agricultural and trade workers, and clerical, service, and sales workers. High skilled workers include managers, professionals, and technicians. Figure 11. Share of green and brown employment in India by skill levels, 2019 A. Green employment B. Brown employment 1.68 9.15 27.16 Low skilled Low skilled 44.87 Medium skilled Medium skilled 45.98 High skilled High skilled 71.16 Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. Based on this definition, green workers are 44.87 percent low-skilled, 45.98 percent medium-skilled, and 9.15 percent high-skilled. Therefore, we identify two types of green workers based on skills – those at the lower end of the skill distribution, performing elementary occupations such as working in recycling waste collection or waste picker and those in the medium to high end such as waste managers or environmental engineers. In contrast, brown workers are 27.16 percent 26 low skilled, 71.16 percent medium skilled, and 1.68 percent high skilled, so there is much less variability in their skill levels compared to individuals employed in green occupations. Among high skilled green workers, 72 percent have tertiary education; for medium skilled green workers, 12 percent have tertiary education; and for low skilled green workers, only 3 percent have tertiary education (Figure 12). Conversely, among medium skilled brown workers, 65 percent have not completed secondary education, while 80 percent of low skilled brown employees have not completed secondary education. Figure 12. Share of green and brown employment in India by educational attainment and skill level, 2019 A. Green employment B. Brown employment 100% 3 100% 3 12 10 16 17 80% 80% 26 25 Some tertiary/post- Some tertiary/post- 72 secondary secondary 60% 42 60% 42 83 Secondary complete Secondary complete 36 42 40% 40% 8 Primary complete 8 Primary complete but secondary but secondary 6 incomple incomple 20% 19 20% 7 32 30 21 12 7 16 0% 1 0% 4 0 Low skilled Medium skilled High skilled Low skilled Medium skilled High skilled Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. We examine green and brown employment, focusing on the differences in skill levels and specific occupational fields of workers in Figure 13. 27 Figure 13. Green and brown workers by skill level and field (percent), India, 2019 A. Green employment Electrical, power and electronics 15 44 12 IT-ITeS 11 23 15 Other 7 19 17 Mechanical engineering, strategic… 13 14 Civil engineering- construction, plumbing,… 6 12 Textiles and handlooms, apparels 0 13 Allied manufacturing gems and jeweler,… 0 4 9 Artisan/creative arts and cottage-based… 1 0 11 High skilled Medium skilled Low skilled Office and business-related work 1 8 Iron and steel, mining, earthmoving and infra… 1 1 Automotive 0 2 0 Chemical engineering, hydrocarbons,… 0 1 Hospitality and tourism 2 0 Healthcare and life sciences 2 0 Telecom 0 Media-journalism, mass communication, etc 1 B. Brown employment Electrical, power and electronics 14 21 16 IT-ITeS 12 18 15 Other 8 17 15 Textiles and handlooms, apparels 0 27 Civil engineering- construction, plumbing,… 3 11 Mechanical engineering, strategic… 23 2 Allied manufacturing gems and jeweler,… 0 5 9 Artisan/creative arts and cottage-based… 1 3 9 High skilled Medium skilled Low skilled Iron and steel, mining, earthmoving and… 1 3 Chemical engineering, hydrocarbons,… 2 3 Healthcare and life sciences 2 2 1 Office and business-related work 0 2 Automotive 1 1 Security 0 3 Telecom 0 Hospitality and tourism 0 Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010- SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3- digit occupation. 28 For high-skilled positions in the green sector, the primary areas of employment are information technology (IT) and IT infrastructure services. Medium-skilled workers in the green sector are predominantly found in the electrical, power, and electronics industries. Low-skilled workers in the green sector are also primarily engaged in IT and IT infrastructure services, but additionally, they are also found in textiles and handlooms, as well as in the apparel industry. In contrast, the brown sector presents a different distribution. Medium-skilled workers in this sector are largely employed in textiles, handlooms, and apparel. On the other hand, low-skilled workers in the brown sector are typically found in electrical, power, and electronics fields. 5.3.Profile of green and brown workers in occupational and job characteristics We proceed to explore the distribution of green and brown employment across occupations, highlighting differences in skill levels, and characterize job attributes in both types of employment. Figure 14 shows that green and brown jobs are a small percentage of workers within occupations. There is a larger prevalence of these individuals as craft workers, machine operators and elementary occupations. Out of all craft operators, 10 percent are in green jobs and 18 percent are in brown employment. Eleven percent of machine operators are employed in green jobs and 22 percent in brown jobs. Twelve percent of elementary occupations are carried out by green workers and 6 percent by brown employees. In most other occupations, the participation of green and brown employment is much smaller, less than 5-6 percent across the other classifications we consider in this paper. 29 Figure 14. Share of green and brown employment in India by type of occupation, 2019 100 90 80 70 72 67 60 82 50 98 95 95 100 98 98 40 30 20 18 22 6 10 2 0 0 10 11 12 0 0 2 4 5 0 0 2 2 Managers Professionals Technicians Clerks Service and Skilled Craft workers Machine Elementary market sales agricultural operators occupations workers Green occupations Brown occupations Rest of occupations Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010- SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. Figure 15 explores whether green and brown workers in these occupations are low, medium, or high skilled. Our findings indicate a distinction in the types of roles occupied by workers in green jobs based on their skill level. Most high-skilled green workers, about 37 percent, are employed as managers and technicians. In contrast, 40 percent of the medium-skilled workers in green jobs are primarily engaged as craft workers. For low-skilled green employment, 94 percent, are in elementary occupations. When it comes to brown jobs, we find similar trends. Medium-skilled workers are predominantly machine operators, accounting for 36 percent, and craft workers, making up 61 percent. About 94 percent of low-skilled workers in the brown sector perform elementary occupations. 30 Figure 15. Share of green and brown employment in India by type of occupation and skill level, 2019 A. Green employment B. Brown employment 100 94 100 94 85 Low skilled Low skilled 80 Medium skilled 80 Medium skilled High skilled High skilled 61 60 60 37 40 33 36 40 40 28 28 23 20 20 14 5 52 01 01 00 000 0 0 0 0 0 10 000 00 00 000 0 0 1 0 0 10 0 0 Managers Technicians Skilled Machine Managers Technicians Skilled Machine agricultural operators agricultural operators Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. Next, we explore employment attributes that reflect job quality, including employment status (paid employees, non-paid employees, employers, and self-employed), having a labor contract, social security contributions, and firm size (1-5 workers, 6-9 workers, 10-19 workers, and more than 20 workers). These results are shown in Figure 16. We find that green workers are more likely to have a written labor contract (16 percent) compared to brown employees (14 percent), but not as much as workers in other occupations (23 percent). Green workers are less likely to contribute to social security than the rest of workers (22 versus 29 percent). We find that green workers tend to work as paid employees (72 percent) or self-employed workers (21 percent), while non-green workers are less likely to be hired as paid employees or salaried workers. Green workers seem to work mostly in larger firms, since the distribution of green employees is further to the right when compared to non-green jobs. 31 Figure 16. Share of green employment in India by job characteristics, 2019 A. Employment contract B. Social security insurance 100 100 90 16 14 90 23 22 20 29 80 80 70 70 60 60 50 50 40 84 86 77 40 78 80 71 30 30 20 20 10 10 0 0 Green Brown Rest Green Brown Rest Without contract With contract Without social security insurance With social security insurance C. Employment status D. Firm size 100 100 90 21 90 21 21 17 23 80 1 39 80 6 6 1 5 7 7 70 70 12 16 15 60 3 60 50 16 50 40 40 72 71 30 30 65 56 57 20 42 20 10 10 0 0 Green Brown Rest Green Brown Rest [1,5] workers [6,9] workers [10,19] workers [20+] workers Paid employee Non-paid employee Employer Self-employed Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010- SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3- digit occupation. Brown workers tend to have similar work conditions as workers in green jobs. They are less likely to have a written contract and contribute to social security but are also more likely to be paid employees and work in larger firms compared to workers that are neither green nor brown. 32 Overall, these results suggest that other workers have better job benefits but are followed by individuals employed in green jobs and those in brown jobs have the least benefits. When analyzing these indicators by skill level, we find that higher skilled workers tend to have better working conditions across the board. Higher skilled green workers are even more likely to have a written contract, contribute to social security, work as paid employees and be employed in large firms when compared to green workers with low and medium skills. The same qualitative findings also apply for brown workers with higher skills when compared to those with lower skills. Once again, such results highlight the heterogeneity within green and brown occupations. 6. Compensation for green and brown jobs Where are green, brown and other workers located on the income distribution? Figure 17 shows the distribution of employees by quintile of monthly labor income. Green and brown workers are found in the upper part of the distribution, making up 12 percent of workers in quintiles 3 and 15 percent in quintile 4. The remainder of jobs are distributed more evenly across the distribution. 33 Figure 17. Green and brown employment by quintile of monthly labor income, 2019 100 90 80 70 60 85 94 90 88 90 50 40 30 20 10 5 7 3 4 4 3 6 7 8 6 0 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Green occupations Brown occupations Rest of occupations Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. 6.1.Unconditional wage differences for green and brown workers Green workers have higher average hourly wages than brown and the rest of workers in India (Figure 18). Hourly wages are 59 rupees for green workers, 29 rupees for brown workers, and 43 rupees for workers in neither green nor brown occupations. This suggests a wage premium of almost 40 percent for green jobs and a penalty of 31 percent for brown jobs when compared to other occupations. When we observe median wages, we find that green jobs have higher wages (47 rupees) but the differences between brown jobs and the rest of jobs disappear (33 rupees per hour for both). Here, the unconditional premium and penalties are lower (9 percent premium for green workers and a 22 percent penalty for brown workers compared to workers in other jobs). 34 Figure 18. Mean and median hourly wages of workers, 2019 Mean hourly wages Median hourly wages $ 70 $ 50 $ 45 $ 60 $ 40 $ 50 $ 35 $ 40 $ 30 $ 25 $ 47 $ 30 $ 59 $ 20 $ 43 $ 15 $ 33 $ 33 $ 20 $ 29 $ 10 $ 10 $5 $- $- Green Brown Rest Green Brown Rest Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: (1) Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010- SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. (2) Mean (median) hourly wages of green, brown and rest of occupations are estimated from the constant and the coefficients of a linear (quantile) regression of hourly wages on the green job and brown job variables. However, we noted in previous sections that green and brown workers are spread out across different skill levels. In Figure 19, we plot average wages within each occupation by skill levels. We observe that in green occupations, average hourly wages are 54 rupees for low-skilled workers, 50 rupees for medium-skilled workers, and 283 rupees for high-skilled workers. In turn, mean pay for brown workers are 76, 44, and 419 rupees per hour for low, medium, and high skilled individuals, respectively. These results indicate substantial heterogeneity within occupations by skill levels. These numbers are similar when looking at median wages. Green workers earn 57, 40, and 255 rupees per hour, while brown workers earn 86, 40, and 460 rupees per hour. 35 Figure 19. Mean hourly wages of green and brown workers by skill levels, 2019 A. Green jobs B. Brown jobs Green jobs Brown jobs $ 450 $ 450 $ 400 $ 400 $ 350 $ 350 $ 300 $ 300 $ 250 $ 250 $ 200 $ 200 $ 419 $ 150 $ 283 $ 150 $ 100 $ 100 $ 50 $ 54 $ 50 $ 50 $ 76 $ 44 $- $- Low skilled Medium skilled High skilled Low skilled Medium skilled High skilled Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: (1) Green occupations are those with any of the 44 green qualifications defined by the Skill Council for Green Jobs. Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010- SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. (2) Mean (median) hourly wages of green, brown and rest of occupations are estimated from the constant and the coefficients of a linear (quantile) regression of hourly wages on the green job and brown job variables. As we noted in the last section, most green workers are in the low or medium-skill categories, with only 9.15 percent in the high category. Most brown workers are also in low or medium-skill groups, with only 1.68 percent in the high skill categories. Therefore, these wage estimates should be interpreted with caution because they are estimated on fewer observations. 6.2.Estimating conditional wage differences for green and brown workers To better approximate the earnings differential of working in green and brown jobs, we estimate Mincer equations with the logarithm of hourly wages as the dependent variable. Table 1 shows the results from two specifications: i) only controlling whether the worker is employed in a green or brown job; and ii) specification i) but adding demographic, regional, and educational attainment control variables. We also include state/UT fixed effects, to compare workers within each state. 36 Regression results confirm our descriptive findings that there seems to be a wage premium for green jobs and a penalty for brown jobs when compared to occupations not in these two categories. Without controls, the premium for green jobs is 6.54 percent while the penalty for brown jobs is 35.3 percent. Once we include controls, these estimates are 13.3 percent premium for green workers and a 21.7 percent penalty for brown workers. We also estimate quantile regressions at the median, to avoid any potential influence from outliers that may affect mean estimates in Table 2. The findings paint a similar story to the results for the conditional mean in Table 1. We confirm a pay premium for green jobs and a penalty for brown jobs. In the full specification, green workers earn 23.7 percent more at the median and brown employees earn 10.8 percent less than workers in neither green nor brown occupations, when controlling for demographics, education, region, and fixed effects. In unreported results, we also estimate Mincer regressions with different reference groups for mean and median wages, which are available upon request. The results lead to identical conclusions to the ones in Tables 1 and 2, with green jobs consistently showing a wage premium and brown jobs consistently showing a wage penalty, irrespective of the reference group and included controls. 37 Table 1. Mincer regressions of mean hourly wages, 2019 log (Hourly wages) (1) (2) (3) (4) Green job 0.0654*** 0.133*** 0.254*** 0.237*** (0.0165) (0.0149) (0.0159) (0.0156) Brown job -0.353*** -0.217*** -0.155*** -0.108*** (0.0192) (0.0165) (0.0171) (0.0165) Male 0.401*** 0.435*** 0.411*** (0.00548) (0.00566) (0.00549) Age 0.0384*** 0.0249*** 0.0366*** (0.00224) (0.00230) (0.00222) - - - Age squared 0.000333*** 0.000230*** 0.000320*** (2.83e-05) (2.92e-05) (2.80e-05) Primary incomplete 0.0208** 0.00867 (0.00894) (0.00897) Primary complete 0.0641*** 0.0425*** (0.00707) (0.00711) Secondary incomplete 0.126*** 0.0924*** (0.00627) (0.00634) Secondary complete 0.260*** 0.200*** (0.00645) (0.00659) Higher than secondary but not university 0.525*** 0.400*** (0.0193) (0.0190) University incomplete or complete 0.795*** 0.645*** (0.00797) (0.00854) Urban 0.177*** 0.221*** 0.146*** (0.00434) (0.00455) (0.00436) Medium skill 0.142*** 0.0695*** (0.00454) (0.00461) High skill 0.582*** 0.312*** (0.00683) (0.00715) Constant 4.046*** 2.417*** 2.773*** 2.409*** (0.0124) (0.0449) (0.0459) (0.0445) Observations 94,366 94,353 94,353 94,353 R-squared 0.075 0.304 0.257 0.320 Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: 1) Standard errors in parentheses. 2) *** p<0.01, ** p<0.05, * p<0.1. 3) Mincer equations estimated by quantile (median) regression, where the dependent variable is the logarithm of hourly wages in the primary occupation for workers aged 25–55. The main explanatory variable is the proportion of green jobs (panel (a)) or brown jobs (panel (b)) or both (panel (c)) in the occupation at 3 digits level of the ISCO-88 38 under the assumptions described in the methodology section. 4) Control variables in column (2) include a male dummy, age, age squared, educational dummies (no education omitted) and an urban area dummy. 5) All specifications control for State/union territory fixed effects. 39 Table 2. Mincer regressions of median hourly wages, 2019 log (Hourly wages) (1) (2) (3) (4) Green job 0.0875*** 0.146*** 0.279*** 0.263*** (0.0180) (0.0169) (0.0175) (0.0176) Brown job -0.180*** -0.140*** -0.0973*** -0.0658*** (0.0191) (0.0179) (0.0182) (0.0182) Male 0.419*** 0.440*** 0.426*** (0.00547) (0.00538) (0.00550) Age 0.0395*** 0.0253*** 0.0375*** (0.00247) (0.00247) (0.00247) - - - Age squared 0.000380*** 0.000256*** 0.000361*** (3.09e-05) (3.09e-05) (3.09e-05) Primary incomplete 0.0245** 0.0126 (0.0109) (0.0109) Primary complete 0.0490*** 0.0340*** (0.00871) (0.00874) Secondary incomplete 0.0938*** 0.0630*** (0.00758) (0.00767) Secondary complete 0.199*** 0.146*** (0.00744) (0.00763) Higher than secondary but not university 0.431*** 0.332*** (0.0178) (0.0181) University incomplete or complete 0.768*** 0.615*** (0.00827) (0.00899) Urban 0.150*** 0.165*** 0.118*** (0.00473) (0.00470) (0.00479) Medium skill 0.126*** 0.0754*** (0.00598) (0.00611) High skill 0.542*** 0.322*** (0.00722) (0.00794) Constant 3.994*** 2.477*** 2.807*** 2.475*** (0.0120) (0.0498) (0.0496) (0.0499) Observations 94,366 94,353 94,353 94,353 Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: 1) Robust standard errors in parentheses. 2) *** p<0.01, ** p<0.05, * p<0.1. 3) Mincer equations estimated by OLS, where the dependent variable is the logarithm of hourly wages in the primary occupation for workers aged 25–55. The main explanatory variable is the proportion of green jobs (panel (a)) or brown jobs (panel (b)) or both (panel (c)) in the occupation at 3 digits level of the ISCO-88 under the assumptions described in the methodology section. 4) Control variables in column (2) include a male 40 dummy, age, age squared, educational dummies (no education omitted) and an urban area dummy. 5) All specifications control for State/union territory fixed effects. While this evidence only provides correlation between the type of occupation and wages, it does shed some light on how the labor market values these occupations on average and at the median in India when we control for demographics, education, and regional variables. While further analyses are required to better understand what explains wage differentials between these occupations, there seems to be highly suggestive evidence that the market values green or environmentally friendly jobs more than brown or pollution-intensive occupations in terms of pay. 7. Conclusions and policy discussion The green transition is expected to significantly affect labor demand, creating new jobs, eliminating others, and changing the required skills in the labor market. To manage this transition effectively and prepare the current and future labor force, it will be crucial for countries to strengthen their existing Labor Market Information System (LMIS) and to establish skills development strategies that allow to align the workforce's skills with the new demands. Since its creation, almost a decade ago, the India's SCGJ has made substantial progress on the green skills agenda, setting a strong precedent for other emerging countries. In close collaboration with industry and academia, the SCGJ has identified 44 green low and medium-skilled occupations related to renewable energy, environment, forestry, climate change, and sustainable development, which are expected to reduce environmental impact and, therefore, be in high demand as the green transition evolves. The SCGJ has also developed detailed qualification files for each of these occupations, which outline their competencies standards, curricula, assessment, and certification requirements. Beyond this, the SCGJ has also rolled out a range of training programs, in collaboration between industry and its 200 training centers, 10 assessment agencies, and 400 certified trainers, with a majority focused on solar and wind energy. 41 This paper analyzes the patterns of green and brown jobs, as well as the characteristics of their workers, using India’s SCGJ definition for green occupations and Vona et al.’s (2018) international classification of brown occupations. We use data from the 2019-20 India Periodic Labour Force Survey to estimate the size of green and brown employment in India, document patterns between and within each type of occupation, characterize workers across multiple sociodemographic characteristics in both types of employment, and study wage differentials for green and brown occupations compared to other jobs in the country. To our knowledge, this is among the first research that employs mutually exclusive definitions of green and brown jobs to provide a detailed analysis of these occupations and characterizes individuals that work in either environmentally friendly (green) or pollutant (brown) jobs. Our findings indicate that green jobs account for about 5.9 percent of all employment, while 4.6 percent of all jobs can be classified as brown in India. There is substantial heterogeneity within green and brown occupations across multiple dimensions, including economic sector, geographic location, and workers’ attributes. Many economic sectors are mixed, with workers employed in both green and brown occupations. Green jobs seem to be more prevalent in Indian states with higher per-capita GDP, but we observe no relationship between the percentage of brown jobs and state-level GDP levels. We also find that green and brown workers tend to be male, younger, and slightly more educated. Green and brown jobs are slightly worse-off compared to other jobs in quality measures, including type of contract, social security benefits, employment categories, and firm size. However, green workers have slightly better working conditions than employees in brown jobs. Lastly, we also find suggestive evidence of a statistically significant wage premium for green jobs and a wage penalty for brown jobs, which varies by skill levels. 42 There are four main policy implications from the results presented in this paper. First, to support the green transition, it will be essential to strengthen existing LMIS to monitor green and brown jobs and to support the green skills development agenda. 13 While lot of progress has been made in India in recent years to create a national definition of green jobs and to develop qualifications files, moving forward it will be critical to also strengthen the LMIS to make it more adaptable and able to monitor the rapid changes in the labor market. This might involve leveraging the use of online vacancy data, conducting more regular and comprehensive surveys to monitor labor market changes, and strengthening the collaboration with the industry and the tertiary education system for a more real-time understanding of the labor market trends. Investing in more robust data and better measurement practices would provide a platform to monitor jobs and skills needed and allow policy makers to react quickly to changes in the labor market, technological changes, and other factors that might affect climate change policy. 14 Second, it will also be important that the large number of training programs that have been rolled out by the SCGJ in collaboration with industry and tertiary education centers are evaluated to ensure that individuals are demanding these training programs, that they are able to afford them, to ensure that the curricula responds to the needs of the industry, that those who enroll are able to complete the courses, and most importantly, whether the training results in employment. Third, the potential distributive consequences of the green transition must be taken into consideration when designing skills development policies so that the training programs are rolled out and targeted properly across low, medium, and high skilled groups to prevent an unequal 13 This is consistent with the findings reported by Granata and Posadas (2024). 14 A road map for the development of a robust national LMIS system is described in Testaverde et al., (2021), with an application for Indonesia. 43 transition to sustainability (Azad and Chakraborty, 2018). Promoting green skills and jobs should be carried out with the knowledge that pollution-intensive or brown jobs may be lost and that not all individuals may benefit from this transition. Experience from past economic transformations suggests that the transition away from fossil fuels may have substantial effects on the structure of employment and earnings. To better understand how the green transition may affect workers and help devise a wide range of policies to transform labor markets while protecting vulnerable workers, it is essential to consider all the potential distributive consequences of the green transition (World Bank, 2024). Lastly, one of our main results is that there is significant heterogeneity in green and brown employment across economic sectors. Future work should explore this stylized fact in greater detail to promote a discussion on whether different sectors require different approaches to transition towards sustainable practices (Varghese et al., 2018). While the green transition is a nationwide policy, there are nuanced effects that need to be considered in terms of workers, sectors, firms, and other factors such as firm size (Kumar et al., 2010). For instance, young people seem to be those acquiring green skills, but it is important to address barriers to access and inclusion and promoting a breadth of green skills that are portable for helping them succeed in future labor markets (United Nations Chidren’s Fund, 2024). Transitioning towards sustainable development requires both immediate and longer-term structural changes to avoid any potentially negative consequences for the existing and future workers. While many research and policy efforts have focused on the short-term implications of a green transition (Green, 2018), some have turned their attention towards the medium- and long- term implications of sustainable development for well-being (Marin and Vona, 2023), especially concentrating on how labor markets will transform in the next few years and decades (García- 44 García et al., 2020; Green and Gambhir, 2020; Kumar and Majid, 2020; Acemoglu et al., 2023). Ensuring a just transition towards sustainable development will require evidence on who stands to gain and lose. This evidence should be both quantitative and qualitative to ensure that we have robust estimations but also that firm and workers’ voices are heard to gather valuable insights into their experiences and challenges. Such diagnoses can help determine where and whom to invest in, to ensure that most individuals benefit from this new form of development and historical inequalities are not perpetuated. Also, different countries will require different strategies and policies to achieve their sustainability practices. India may be a particular case that need not apply to other countries in the South Asia region or in other parts of the world. Future studies could begin comparing the distinct challenges faced by countries within the same geographic region or across different ones. Only with more evidence will we be more confident that transitioning towards sustainability will result in equitable development with greater benefits than costs. 45 References Acemoglu, D., Aghion, P., Barrage, L., and Hémous, D. (2023). Climate change, directed innovation, and energy transition: The long-run consequences of the shale gas revolution (No. w31657). National Bureau of Economic Research. Azad, R. and Chakraborty, S. (2018). A policy proposal for green jobs in India: A quantitative analysis on inclusivity of green jobs. Banerjee, A. Duflo, E., Ghatak, M., and Lafortune, J. (2013). Marry for What? Caste and Mate Selection in Modern India. American Economic Journal: Microeconomics, 5(2): 33-72. Bishnoi, T. and Rai, D. K. (2022). Case Study: India. The Role of the Skill Council for Green Jobs (SCGJ) in Recognizing Green Skills and Upskilling Workers in Micro, Small and Medium Enterprises. In Recognizing Green Skills Through Non-Formal Learning: A Comparative Study in Asia (pp. 129-152). Singapore: Springer Nature Singapore. Bowen, A. (2012). 'Green' growth, 'green' jobs, and labor markets. World Bank Policy Research Working Paper, (5990). Carleton, T. A. and Hsiang, S. M. (2016). Social and economic impacts of climate. Science, 353(6304), aad9837. Carranza, E. (2014). Soil Endowments, Female Labor Force Participation, and the Demographic Deficit of Women in India. American Economic Journal: Applied Economics, 6(4), 197– 225. Cerimelo, M. M., De la Vega, P., and Porto, N. (2022). Going Green: Estimating the Potential of Green Jobs in Argentina (No. 4550). Asociación Argentina de Economía Política. Chatterjee, U. & Murgai, R., and Rama, M.G. (2015). "Job opportunities along the rural-urban gradation and female labor force participation in India," Policy Research Working Paper Series 7412, The World Bank. De Chazal, J. and Rounsevell, M. D. (2009). Land-use and climate change within assessments of biodiversity change: a review. Global Environmental Change, 19(2), 306-315. García-García, P., Carpintero, Ó., and Buendía, L. (2020). Just energy transitions to low carbon economies: A review of the concept and its effects on labour and income. Energy Research & Social Science, 70, 101664. Govindan, M. and Bhanot, J. (2014). Green jobs in India: potentials and perspectives. Granata, J. and Posadas, J. (2024). Why Look at Tasks When Designing Skills Policy for the Green Transition? World Bank Policy Research Working Paper 10753. 46 Green, F. (2018). Transition policy for climate change mitigation: Who, what, why and how. Centre for Climate Economics & Policy Working Paper. Canberra: Australian National University. Green, F. and Gambhir, A. (2020). Transitional assistance policies for just, equitable and smooth low-carbon transitions: who, what and how? Climate Policy, 20(8), 902-921. ILO (2012). International Standard Classification of Occupations: ISCO-08. Geneva: International Labour Organization. Kedia, S., Pandey, R., and Sinha, R. (2020). Shaping the post-COVID-19 development paradigm in India: some imperatives for greening the economic recovery. Millennial Asia, 11(3), 268-298. Klasen, S. and Pieters, J. (2015). What Explains the Stagnation of Female Labor Force Participation in Urban India? The World Bank Economic Review, 29(3), 449–478. http://www.jstor.org/stable/43774162 Khurana, S. and Mahajan, K. (2020). Evolution of wage inequality in India (1983-2017): The role of occupational task content (No. 2020/167). WIDER Working Paper. Kumar, V., Mandal, K., Kumar, N., and Banerjee, T. (2010). Skills for green jobs in India. ILO, Geneva, Switzerland. Kumar. J, C.R. and Majid, M.A. (2020). Renewable energy for sustainable development in India: current status, future prospects, challenges, employment, and investment opportunities. Energy, Sustainability and Society, 10(2). https://doi.org/10.1186/s13705- 019-0232-1 Marimuthu, R., Sankaranarayanan, B., Ali, S. M., and Karuppiah, K. (2022). Green recovery strategies for the mining industry of India: lessons learned from the COVID-19 pandemic. Journal of Asia Business Studies, 16(3), 428-447. Marin, G. and Vona, F. (2019). Climate policies and skill-biased employment dynamics: Evidence from EU countries. Journal of Environmental Economics and Management, 98, 102253. Marin, G. and Vona, F. (2023). Labour market implications for the sustainable transition. In Handbook on Innovation, Society and the Environment (pp. 345-357). Edward Elgar Publishing. McMichael, A. J. and Haines, A. (1997). Global climate change: the potential effects on health. Bmj, 315(7111), 805-809. Mealy, P., del Rio-Chanona, R. M., and Farmer, J. D. (2018). What you do at work matters: new lenses on labour. What You Do at Work Matters: New Lenses on Labour (March 18, 2018). Mendelsohn, R., Dinar, A., and Williams, L. (2006). The distributional impact of climate change on rich and poor countries. Environment and development economics, 11(2), 159-178. 47 Pavlova, M. (2019). Emerging Environmental Industries: Impact on Required Skills and TVET Systems. International Journal of Training Research 17 (sup1): 144–158. Saxena, A., Dutta, A., Fischer, H. W., Saxena, A. K., and Jantz, P. (2021). Forest livelihoods and a “green recovery” from the COVID-19 pandemic: Insights and emerging research priorities from India. Forest Policy and Economics, 131, 102550. SCGJ (2022). Green Jobs Handbook: A comprehensive information handbook to foster career in Green Business. Available at: https://sscgj.in/wp-content/uploads/2022/06/Green-Jobs- Hand-Book-PDF.pdf. Stanef-Puică, M. R., Badea, L., Șerban-Oprescu, G. L., Șerban-Oprescu, A. T., Frâncu, L. G., and Crețu, A. (2022). Green Jobs—A Literature Review. International Journal of Environmental Research and Public Health, 19(13), 7998. Testaverde, M., Posadas, J., & Cho, I. (2021). Toward a world-class labor market information system for Indonesia: an assessment of the system managed by the Indonesian Ministry of Manpower. World Bank Research Publication, Washington DC. United Nations Children’s Fund. (2024). Skills for the Green Transition: Solutions for Youth on the Move, UNICEF, New York. © United Nations Children’s Fund (UNICEF). Varghese, G., Gupta, S., and Sharma, K. (2018). Sectoral Approaches to Skills for Green Jobs in India. In Changing the Indian Economy (pp. 141-153). Elsevier. Vona, F., Marin, G., Consoli, D., and Popp, D. (2018). Environmental regulation and green skills: an empirical exploration. Journal of the Association of Environmental and Resource Economists, 5(4), 713-753. Wheeler, T. and Von Braun, J. (2013). Climate change impacts on global food security. Science, 341(6145), 508-513. World Bank. 2023. Toward Faster, Cleaner Growth. South Asia Development Update (October 2023). World Bank, Washington, DC. doi: 10.1596/978-1-4648-2026-7. License: Creative Commons Attribution CC BY 3.0 IGO. World Bank. 2024. Jobs for Resilience. South Asia Development Update (April 2024). World Bank, Washington, DC. doi: 10.1596/978-1-4648-2103-5. License: Creative Commons Attribution CC BY 3.0 IGO 48 Table A.1. India's national definition of green jobs developed by the SCGJ Source: List of green jobs qualifications from Skill Council for Green Jobs (2022). "Green Jobs Handbook". National Skills Qualification Framework (NSQF) Qualification files from National Qualifications Register webpage (https://www.nqr.gov.in/). Note: * https://www.openriskmanual.org/wiki/ISCO_Occupation_Group_7412.12_Wind_Turbine_Technician 49 Table A.2. Brown job qualifications in India and associated occupation based on ISCO-08 SOC 2010 SOC Title SOC 2010 SOC Title 17-2041 Chemical Engineers 51-6063 Textile Knitting and Weaving Machine Setters, Operators, and Tenders 17-2151 Mining and Geological Engineers, Including Mining Safety Engineers 51-6064 Textile Winding, Twisting, and Drawing Out Machine Setters, Operators, and Tenders 17-2171 Petroleum Engineers 51-6091 Extruding and Forming Machine Setters, Operators, and Tenders, Synthetic and Glass Fibers 19-1012 Food Scientists and Technologists 51-6093 Upholsterers 19-2031 Chemists 51-7011 Cabinetmakers and Bench Carpenters 19-4031 Chemical Technicians 51-7021 Furniture Finishers 43-5041 Meter Readers, Utilities 51-7031 Model Makers, Wood 45-4023 Log Graders and Scalers 51-7032 Patternmakers, Wood 47-4071 Septic Tank Servicers and Sewer Pipe Cleaners 51-7041 Sawing Machine Setters, Operators, and Tenders, Wood 47-5011 Derrick Operators, Oil and Gas 51-7042 Woodworking Machine Setters, Operators, and Tenders, Except Sawing 47-5012 Rotary Drill Operators, Oil and Gas 51-8012 Power Distributors and Dispatchers 47-5013 Service Unit Operators, Oil, Gas, and Mining 51-8091 Chemical Plant and System Operators 47-5021 Earth Drillers, Except Oil and Gas 51-8092 Gas Plant Operators 47-5031 Explosives Workers, Ordnance Handling Experts, and Blasters 51-8093 Petroleum Pump System Operators, Refinery Operators, and Gaugers 47-5042 Mine Cutting and Channeling Machine Operators 51-9011 Chemical Equipment Operators and Tenders 47-5051 Rock Splitters, Quarry 51-9012 Separating, Filtering, Clarifying, Precipitating, and Still Machine Setters, Operators, and Tenders 47-5061 Roof Bolters, Mining 51-9021 Crushing, Grinding, and Polishing Machine Setters, Operators, and Tenders 47-5071 Roustabouts, Oil and Gas 51-9022 Grinding and Polishing Workers, Hand 47-5081 Helpers--Extraction Workers 51-9023 Mixing and Blending Machine Setters, Operators, and Tenders 49-2095 Electrical and Electronics Repairers, Powerhouse, Substation, and Relay 51-9031 Cutters and Trimmers, Hand 49-9012 Control and Valve Installers and Repairers, Except Mechanical Door 51-9032 Cutting and Slicing Machine Setters, Operators, and Tenders 49-9041 Industrial Machinery Mechanics 51-9041 Extruding, Forming, Pressing, and Compacting Machine Setters, Operators, and Tenders 49-9043 Maintenance Workers, Machinery 51-9051 Furnace, Kiln, Oven, Drier, and Kettle Operators and Tenders 49-9045 Refractory Materials Repairers, Except Brickmasons 51-9111 Packaging and Filling Machine Operators and Tenders 49-9051 Electrical Power-Line Installers and Repairers 51-9121 Coating, Painting, and Spraying Machine Setters, Operators, and Tenders 49-9093 Fabric Menders, Except Garment 51-9191 Adhesive Bonding Machine Operators and Tenders 51-1011 First-Line Supervisors of Production and Operating Workers 51-2091 Fiberglass Laminators and Fabricators 51-9192 Cleaning, Washing, and Metal Pickling Equipment Operators and Tenders 51-3091 Food and Tobacco Roasting, Baking, and Drying Machine Operators and Tenders 51-9193 Cooling and Freezing Equipment Operators and Tenders 51-3092 Food Batchmakers 51-9195 Molders, Shapers, and Casters, Except Metal and Plastic 51-3093 Food Cooking Machine Operators and Tenders 51-9196 Paper Goods Machine Setters, Operators, and Tenders 51-4021 Extruding and Drawing Machine Setters, Operators, and Tenders, Metal and Plastic 51-9197 Tire Builders 51-4022 Forging Machine Setters, Operators, and Tenders, Metal and Plastic 53-4013 Rail Yard Engineers, Dinkey Operators, and Hostlers 51-4023 Rolling Machine Setters, Operators, and Tenders, Metal and Plastic 53-7031 Dredge Operators 51-4033 Grinding, Lapping, Polishing, and Buffing Machine Tool Setters, Operators, and Tenders, Metal and Plastic 53-7032 Excavating and Loading Machine and Dragline Operators 51-4051 Metal-Refining Furnace Operators and Tenders 53-7033 Loading Machine Operators, Underground Mining 51-4052 Pourers and Casters, Metal 53-7041 Hoist and Winch Operators 51-4062 Patternmakers, Metal and Plastic 53-7063 Machine Feeders and Offbearers 51-4071 Foundry Mold and Coremakers 53-7071 Gas Compressor and Gas Pumping Station Operators 51-4191 Heat Treating Equipment Setters, Operators, and Tenders, Metal and Plastic 53-7072 Pump Operators, Except Wellhead Pumpers 51-4192 Layout Workers, Metal and Plastic 53-7073 Wellhead Pumpers 51-4193 Plating and Coating Machine Setters, Operators, and Tenders, Metal and Plastic 53-7111 Mine Shuttle Car Operators 51-4194 Tool Grinders, Filers, and Sharpeners 17-2041 Chemical Engineers 51-6061 Textile Bleaching and Dyeing Machine Operators and Tenders 17-2151 Mining and Geological Engineers, Including Mining Safety Engineers Source: Vona et al. (2018). 50 Table A.3. Summary statistics of Green and Brown occupations in India Source: Own elaboration based on microdata from the 2019 Periodic Labour Force Survey (PLFS) for India. Notes: The column shows in parenthesis the sample standard deviations of the means and the standard error of the difference in sample means estimated as the robust standard error of the coefficient of a regression of the green/brown/rest job variable on the group membership dummy variable. (1) Green occupations are those 4-digit ISCO-08 occupations with any of the 44 green qualifications defined by the Skill Council for Green Jobs (see Table 0). To estimate the proportion of green jobs for each 3-digit ISCO- 88 (as specified in the Indian labor force survey), 4-digit ISCO-08 occupations are mapped to the 4-digit ISCO-88 classification using the ILO's crosswalk (when a single ISCO-88 occupation code is mapped to more than one ISCO-08, a green proportion measure is used to reflect the proportion of the corresponding ISCO-08 occupations that have been classified as green) and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. (2) Brown occupations are those defined by Vona et al. (2018) at the 6-digits level of the 2010-SOC, mapped to the 4-digit ISCO-08 classification using a crosswalk from the US Bureau of Labor Statistics (when a single ISCO occupation code is mapped to more than one 6-digit SOC, a brown proportion measure is used that reflect the proportion of corresponding 6-digit SOC occupations that have been classified as brown), converted to ISCO-88 at 4-digits using ILO's crosswalk and then estimated at the ISCO-88 3 digit level assuming an equal distribution of workers in 4-digits occupations within each 3-digit occupation. 51