Securing the Livelihoods of Informal Economy Workers in Times of Global Changes

Report

5 October 2026

5. Informal workers and emerging risks in the era of environmental degradation and climate change

Copy link to 5. Informal workers and emerging risks in the era of environmental degradation and climate change

Abstract

Environmental degradation and climate hazards – such as droughts, floods and extreme heat – pose urgent and growing threats to economic activity and work worldwide. These hazards are particularly pronounced in developing countries, with acute consequences for workers in the informal economy. Drawing on data from the Key Indicators of Informality based on Individuals and their Households (KIIbIH), this chapter starts by documenting the strong dependence of informal economy workers on the natural environment for their livelihoods. It then assesses the level of risk faced by informal workers in relation to droughts, floods and heat stress based on three dimensions: exposure, sensitivity and adaptive capacity. By highlighting the structural factors that underpin these risks and vulnerabilities, the chapter provides a range of short‑ and long‑term evidence-based policy options to strengthen the resilience of informal economy workers to environmental degradation and climate shocks.

In brief

Copy link to In brief

Environmental degradation threatens the livelihoods of informal economy workers

Informal workers’ livelihoods strongly depend on the natural environment

Across KIIbIH countries, 36% of all workers are employed in sectors that rely directly or indirectly on natural and environmental resources. Informal workers (41%) are far more likely than formal ones (12%) to operate in these sectors, reflecting their involvement in agriculture.
Environmentally linked jobs are associated with lower earnings, higher poverty rates and higher household informality. On average, the poverty rate of workers operating in environmentally linked sectors (36%) is double that of those in sectors with no environmental link (18%).
Reliance on the environment is compounded in rural areas, among low‑educated workers, and among own‑account and contributing family workers.

Droughts and floods disproportionately affect informal workers

More than 40% of informal workers face medium to high levels of risk from droughts and floods, compared with less than 2% of formal workers. This asymmetry stems from the limited adaptive capacity of informal workers and their large sectoral sensitivity to environmental hazards.
Should exposure to weather events intensify, the level of risk faced by many informal workers could rapidly increase, putting their livelihoods under greater strain.
People living in fully informal households are far more vulnerable to droughts and floods. In cases of shocks during which all members may be affected simultaneously, these households have a much lower ability to offset income losses through alternative sources of income, with potentially large negative consequences on the well-being of all household members.

Heat stress is increasingly responsible for large productivity losses among informal workers

Informal workers are overrepresented in physically demanding jobs, the activities in which productivity losses due to heat stress are largest.
In 2025, excessive heat reduced working hours by 2.2% across KIIbIH countries, corresponding to a GDP loss of USD 510 billion in purchasing power parity (PPP) terms. This is equivalent to about 60 billion hours of work lost or about 28.8 million full‑time jobs. Heat-related productivity losses in 2025 were substantially larger for informal workers (2.4%) than for formal ones (1.4%).
Under current labour‑market structures, productivity losses are projected to increase to 2.8% by 2055, equivalent to an additional 8 million full‑time jobs lost.

Securing the livelihoods of informal workers calls for policy responses that combine preparedness, mitigation and adaptation to environmental shocks

To secure informal workers’ livelihoods in the face of environmental degradation and climate risks, governments should focus on five policy objectives: i) consolidate access to traditional and new social protection tools; ii) invest in skills, training and human capital; iii) strengthen occupational safety and health policies and early warning systems; iv) establish comprehensive agricultural and rural development policies to build rural resilience; and v) invest in resilient and quality urban infrastructure.

Introduction

Copy link to Introduction

Technological innovation has helped reduce some of the environmental damage associated with industrialisation. Still, the combined effects of multiple factors throughout the 20th and into the 21st centuries – in particular accelerated industrialisation, natural resource depletion, urban sprawl and mass consumption – have placed growing pressure on the natural environment. Emerging environment-related risks have turned into multi-dimensional threats to societies worldwide, and to workers in particular.

Environmental degradation is a complex and multifaceted concept. Broadly, it refers to the deterioration of the environment caused by human or natural causes, including resource depletion, ecosystem and habitat destruction, species extinction and rising pollution. According to Yeganeh (2020[1]), degradation can be categorised into two types – pollution and deterioration – across five dimensions: atmosphere, hydrosphere, lithosphere, biosphere and food. For example, global warming results from atmospheric deterioration, intensifying climate-related events such as heat waves, droughts and floods. Similarly, land degradation – primarily driven by human activity – falls under lithosphere deterioration and encompasses phenomena such as deforestation and desertification.

Individuals and households face varying levels of risk from adverse events triggered by environmental degradation and climate change, including droughts, floods and extreme temperatures. From a livelihood perspective, these risks arise across three dimensions: exposure, sensitivity and adaptive capacity. Exposure to shocks refers to the presence, frequency and severity of environmental threats. Sensitivity reflects the extent to which individuals and households are likely to be harmed by such threats. Adaptation relates to the capacity of individuals, households and communities to cope with and recover from environmental impacts. Together, the notions of sensitivity and adaptive capacity form the intermediary concept of vulnerability towards environmental degradation and climate change, whose formal definition has evolved over time. Vulnerability can take many forms and span multiple dimensions of individuals’ lives – including non-economic ones (Canpolat et al., 2025[2]). This chapter looks specifically at environment-related vulnerability through the lens of the potential impacts that environmental degradation and climate hazards can have on workers’ incomes. (Annex 5.A provides more details on the concepts of risk and vulnerability and the approach adopted in the analysis).

Within this perspective, the situation of workers in the informal economy requires specific attention. The argument developed in this chapter can be summarised as follows. By and large, environmental degradation and resource depletion disproportionately affect some of the world’s poorest countries (exposure), in which work is often mostly informal. Workers in the informal economy contribute in many ways to environment sustainability through the nature of their tasks and often have a smaller carbon footprint than formal workers. However, they tend to be concentrated in activities that are directly linked to the natural environment, which makes them particularly sensitive to environmental degradation and climate change (sensitivity). Finally, informal workers and their households typically lack access to social protection systems, which severely limits their ability to cope with and recover from the socio-economic consequences of environmental degradation (adaptation). This limited adaptive capacity is often even more constrained for women, who may face additional barriers linked to unequal access to resources, unpaid care responsibilities and discriminatory social norms.

Importantly, this chapter analyses informal workers’ links to the environment insofar as these links can heighten their sensitivity to environmental degradation. Yet, these same links also position informal economy workers as key agents of the green growth agenda and of efforts to mitigate environmental degradation (World Bank, 2024[3]; Tucker and Anantharaman, 2020[4]; Longondjo Etambakonga and Roloff, 2019[5]). In urban areas, for example, the informal economy generates substantial social and environmental value. Activities such as waste picking, recycling and street vending are crucial components of local economies. They can help reduce cities’ carbon footprints while providing valuable services at a cost that remains accessible to most residents.

This chapter aims to document the links between the informal economy and the natural environment and assess the level of risk faced by workers in the informal economy as a result of environmental degradation. It then reflects on possible solutions to support their livelihoods and those of their households at a time when more and more countries are pursuing green transition strategies. Drawing on KIIbIH data, the first section examines the relative importance of the natural environment for the livelihoods of informal workers. The second and third sections investigate the levels of risk faced by workers in the informal economy in relation to two specific environmental hazards: extreme weather events and heat stress. The chapter concludes by identifying short- and long-term policy options to help reduce and mitigate the vulnerabilities of informal economy workers to environmental degradation.

Workers of the informal economy rely heavily on the natural environment for their livelihoods

Copy link to Workers of the informal economy rely heavily on the natural environment for their livelihoods

To measure the links of workers to the environment, jobs are classified into three categories based on their codes in the fourth revision of the International Standard Industrial Classification of All Economic Activities (ISIC‑04). Sectors with a direct environmental link either rely directly on natural resources or play a key role in environmental management. Sectors with an indirect environmental link are those in which most activities depend on the environment without directly exploiting natural resources (Figure 5.1). All remaining sectors are classified as having no environmental link. Together, sectors with direct and indirect links are referred to as environmentally linked sectors.

Figure 5.1. Sectoral linkages to environment
Copy link to Figure 5.1. Sectoral linkages to environment

Many workers depend on the natural environment for their livelihoods, especially informal workers

Across the KIIbIH countries for which sectoral data are available, about one-third of workers rely on the environment for their jobs and livelihoods. On average, 32% of workers’ primary jobs have a strong, direct link to the environment, while an additional 4% are indirectly linked (Figure 5.2). Reliance on the environment varies widely across countries but is particularly elevated in Africa and, to a lesser extent, in Asia. In Mali, Lao PDR, Madagascar and Niger, more than two-thirds of the labour force work in sectors that are directly or indirectly linked to the environment, reflecting the central role these activities play in national labour markets.

Figure 5.2. About one-third of workers rely on the environment for their livelihoods
Copy link to Figure 5.2. About one-third of workers rely on the environment for their livelihoods

Share of workers aged above 15 years by sectoral link to the environment, latest year available

Most environmentally linked jobs are concentrated in agriculture, but not exclusively so. In low‑ and middle‑income countries characterised by a high overall dependence of the labour market on the environment, this reliance primarily stems from the size of the agricultural sector (which is naturally classified as directly linked to the environment). However, dependence on the environment extends beyond agriculture. In Burkina Faso, Liberia, Mali, Mongolia, Togo and South Africa, more than 4% of jobs directly linked to the environment are outside agriculture, often reflecting the importance of the extractive sector in these countries. Many jobs are also concentrated in sectors that indirectly rely on the environment or are located downstream in agricultural value chains. Despite the important role of these jobs, notably for food systems, they often face bias from national and local policymakers and lack adequate public-sponsored support during adverse events such as the COVID-19 pandemic or climate-induced disruptions (OECD/UN-Habitat, 2022[7]). Overall, across KIIbIH countries, up to 4% of jobs are in sectors that depend indirectly on natural resources, ranging from waste management to food processing, wholesale of food and beverages, ecotourism, and so forth. In Cameroon, Ghana, the Maldives and Togo, KIIbIH data show that these indirect sectors account for more than 8% of total employment.

Workers who operate in an environmentally linked sector are more likely to be in informal employment, and this likelihood increases as the sectoral link to the environment tightens. Across KIIbIH countries, the informal employment rate stands at 84% among sectors with a strong and direct environmental link. The rate drops to 70% among sectors with an indirect link and to 63% for sectors with no link to the environment (Figure 5.3). This cross-sectoral pattern holds true across nearly all KIIbIH countries, notably when the profile of the labour market is more diverse. By contrast, in Africa, where informality is widespread across the economy, differences across sectors are more limited.

Figure 5.3. Informal employment is more common among environmentally linked sectors
Copy link to Figure 5.3. Informal employment is more common among environmentally linked sectors

Share of workers aged above 15 years in informal employment by sectoral link to the environment, latest year available

Overall, these informal employment rates translate into informal workers being more concentrated than formal ones in jobs that depend on the natural environment. Across KIIbIH countries, on average, 41% of informal workers (44% for men and 37% for women) hold a job that has a direct or indirect link to the environment, including 37% in jobs with a direct link. In comparison, only 12% of formal workers (15% for men and 9% for women) work in a sector that has a direct or indirect link to the environment, including 9% with a direct link. This asymmetry primarily stems from the larger shares of informal workers who operate in the agricultural sector (36%) compared with formal ones (7%).

Environmentally linked jobs tend to pay less, which disproportionately affects informal workers

In general, KIIbIH data show that workers who operate in sectors linked to the environment earn less than the rest of the labour force, which puts informal workers at a greater risk of low-pay jobs. On average, across the 14 countries for which sectoral and income data are available (most of which are located in Latin America and the Caribbean), workers who operate in sectors with a direct environmental link earn 82% of the median labour income compared with 91% of the median labour income for workers in sectors with an indirect environmental link, and 109% of the median labour income for workers in other economic sectors (Figure 5.4).

Figure 5.4. Workers operating in environmentally linked jobs earn less than the rest of the labour force
Copy link to Figure 5.4. Workers operating in environmentally linked jobs earn less than the rest of the labour force

Labour income of workers aged above 15 years by sectoral link to the environment as a share of national median labour income, latest year available

These differences in labour income across sectors conceal large discrepancies between informal and formal workers. The earning disadvantage in environmentally linked sectors seems to be mostly observed among informal workers. Across these same 14 countries, formal workers, including those who operate in environmentally linked sectors, earn far more than the national median labour income. Conversely, informal workers, regardless of whether or not they belong to an environmentally linked sector, tend to earn systematically less than the national median labour income. Moreover, the informal economy tends to be characterised by a much clearer labour income penalty associated with operating in an environmentally linked sector. In Nicaragua, informal workers who have a job with a direct link to the environment earn, on average, 50% of the national median labour income, whereas informal workers in a sector indirectly linked earn 93%, and informal workers in sectors with no link earn 100%. By contrast, formal workers, regardless of which sector in which they operate, earn on average more than 150% of the national median labour income. A similar gradient is found in Bolivia, where informal workers respectively earn 51% (direct link), 82% (indirect link) and 96% (no link) of the national median labour income, compared with 137% (direct link), 150% (indirect link) and 167% (no link) for formal workers.

Working poverty is also more common in environmentally linked sectors. Across KIIbIH countries, the average poverty rate among workers whose main job is directly tied to the environment stands at 36%, compared with 24% for workers whose main job is only indirectly linked, and 18% for workers in sectors with no clear link (Figure 5.5). This pattern is consistent across all KIIbIH countries for which data are available. In many African countries (such as Cameroon, Madagascar, Mali, Senegal, Sierra Leone, Togo and Zambia), the gap between workers in direct link and no link sectors is larger than 30 percentage points. In 221 of 40 countries with available data, the poverty rate of workers in environmentally linked sectors is at least twice that of no link workers.

Figure 5.5. Workers operating in environmentally linked jobs are more likely to be poor
Copy link to Figure 5.5. Workers operating in environmentally linked jobs are more likely to be poor

Poverty rates at national poverty line among workers aged above 15 years by sectoral link to the environment, latest year available

For both formal and informal workers, environmentally linked jobs are concentrated among rural men with low educational attainment

Across KIIbIH countries, rural workers are overrepresented in jobs that are directly linked to the environment. In fact, 72% of workers in these jobs live in rural areas, including 75% of informal workers and 50% of formal workers (Figure 5.6). This pattern largely reflects the central role of agriculture (and more generally of the entire primary sector), which is classified as directly linked to the environment and remains concentrated in rural areas. Evidence from KIIbIH also shows that rural workers are more likely to be informally employed and to face greater gaps in access to education and training (see Chapter 2).

Figure 5.6. Rural men with low levels of education and employed as own-account workers are more likely to have jobs with a direct link to the natural environment
Copy link to Figure 5.6. Rural men with low levels of education and employed as own-account workers are more likely to have jobs with a direct link to the natural environment

Distribution of informal (Panels A, C, E and G) and formal (Panels B, D, F and H) workers aged above 15 years according to the environmental link of their job and their location, gender, educational attainment and employment status

Men are more likely than women to work in sectors that directly rely on the environment. On average, men account for about two-thirds (66%) of workers whose job has a direct environmental link. The share rises to 77% among formal workers but remains similar among informal workers (66%) (Figure 5.6). This overrepresentation mainly reflects men’s greater involvement in agriculture and, above all, their disproportionate presence in mining and quarrying activities. The gender pattern differs slightly regarding jobs with an indirect link, notably among informal workers, where women account for 55% of employment. This largely reflects women’s stronger participation in activities related to the wholesale of agricultural products, food, beverages and tobacco.

These KIIbIH findings are mostly in line with other sources. Taking a different lens and focusing on the agrifood system, the Food and Agriculture Organization (FAO) estimates that men account for 58% of employment in agrifood systems (FAO, 2026[8]). However, because of more limited employment options outside of agrifood systems, women’s employment tends to be disproportionately concentrated in such activities, notably in countries where agricultural production remains dominant. Women’s informal employment is also particularly pronounced in non-agricultural agri-food systems.

Workers’ educational profiles vary markedly in relation to how strongly their jobs are linked to the environment. For both informal and formal workers, educational attainment rises as the environmental link of jobs weakens (Figure 5.6). In effect, jobs directly linked to the environment are disproportionately held by workers with low levels of education. Across KIIbIH countries, about two-thirds of workers (66%) in directly linked sectors have no education or only primary education, while only a small minority (6%) have completed tertiary education. This pattern is particularly pronounced for informal workers in directly linked sectors, among whom around 41% have no formal education and fewer than 3% have a tertiary degree (Figure 5.6). Formal workers in direct sectors are better educated than their informal counterparts, but still less so than workers operating in other parts of the economy, where secondary and tertiary education account for a much larger share of employment. Overall, this results in workers with low levels of education being far more likely to be employed in sectors that are directly linked to the environment. Across KIIbIH countries, nearly half of workers with no education and more than one-third of those with only primary education work in directly linked sectors, while this share declines sharply with education and reaches very low levels among workers with tertiary education.

Likewise, workers’ employment status varies strongly with how closely their jobs are linked to the environment. Jobs directly linked to the environment are disproportionately held by own‑account workers and contributing family workers (i.e. helping a family or household member run a business, farm or job, without regular payment), while employees are much more prevalent in sectors with no environmental link. Across the KIIbIH, more than two-thirds of workers in directly linked sectors are either own‑account workers (43%) or contributing family workers (23%), compared with around one-third in sectors with no environmental link (Figure 5.6). In the informal economy, own‑account and contributing family workers account, respectively, for nearly half of employment in directly linked sectors and one-quarter of it. By contrast, among formal workers, employment in directly linked sectors is dominated by employees, reflecting the limited presence of non‑standard employment statuses outside informality. Here again, it translates into own‑account workers and contributing family workers being far more likely than employees to be concentrated in environmentally linked sectors. Across KIIbIH countries, 35% of own‑account workers and 59% of contributing family workers operate in jobs directly linked to the environment, whereas most employees (84%) work in sectors with no environmental link. This pattern is especially strong in the informal economy, where non‑wage workers are heavily concentrated in environmentally dependent activities.

Informal workers’ reliance on the natural environment is compounded by the household dimension

Access to resources, livelihoods, labour diversification and coping capacities are largely influenced by the composition of household and living arrangements. This is why workers’ reliance on the environment depends not only on the sector in which they operate, but also on whether they live with other household members who also rely on the environment for their livelihoods and may be affected by shocks simultaneously.

Evidence from KIIbIH data points to a strong overlap between environmental dependence at the job level and informality at the household level, with informal households being more likely to depend on the environment for their livelihoods. On average, 39% of workers who live in informal households – in which all working members of the household are in informal employment – operate in jobs with a direct environmental link, compared with 17% in mixed households – in which some working members are in informal employment and others are in formal employment. The share drops to only 8% in formal households (Figure 5.7). Conversely, workers living in formal households are overwhelmingly concentrated in sectors with no environmental link, which account for 90% of their employment.

Figure 5.7. Informal households depend more on the environment than other types of households
Copy link to Figure 5.7. Informal households depend more on the environment than other types of households

Share of workers aged above 15 years by sectoral link to the environment and the informality status of the household in which they live, latest year available

This overlap between environmental dependence and household informality holds across regions, although its intensity varies. In Africa and Asia, around half of workers who live in informal households are employed in sectors directly linked to the environment (56% and 49%, respectively). In the Americas, the share remains substantial at 24%. Across all regions, workers in mixed households occupy an intermediate position, with a higher likelihood of working in environmentally linked sectors (21%) than those in formal households (11%), but substantially less than those in fully informal households (43%). As potential environmental shocks are more likely to simultaneously affect both labour income and household resources, this concentration of workers with environmentally linked jobs in informal households reinforces their vulnerability. Conversely, formal households tend to benefit from greater labour diversification and insulation from such risks.

Informal workers face high risks from droughts and floods

Copy link to Informal workers face high risks from droughts and floods

Climate change is already affecting every region across the globe. Evidence from historical observed climate data shows an increase in frequency of intensity of droughts and heavy precipitation events, which can lead to inland floods (IPCC, 2023[9]). Human influence has also raised the likelihood of compound extreme events, including the frequency of concurrent heatwaves and droughts and compound flooding in some locations (IPCC, 2023[9]).

The effects of extreme weather events are not uniform across households. Beyond the direct destruction of assets and infrastructure, such events trigger impacts on labour demand and supply, on-the-job productivity, income and vulnerability among the self-employed, and labour reallocation (Feriga, Lozano Gracia and Serneels, 2024[10]). The incidence of these labour-market and income shocks largely depends on where people live, the sectors and jobs they rely on, and the resources and institutions they can draw on to cope. Recognising this heterogeneity is central for policymakers and to design adequate adaptation policies: when the most exposed groups are also those with the least capacity to smooth consumption and recover, climate shocks can translate into persistent welfare losses and widen existing socio-economic divides.

A growing body of evidence shows that extreme weather events can have sizeable effects on incomes and inequality, often hitting those with fewer resources the hardest. While labour earnings may be partly stabilised through transfers or substitution across activities – and capital income may, in some contexts, be more exposed to disaster-related losses – the distributional impacts vary markedly depending on the type of disaster and the local context. Studying flash floods in Brazil, Wagner (2025[11]) shows that income losses are concentrated at the bottom of the income distribution, largely because widespread labour informality leaves lower-income workers – especially those in the informal sector – less protected from disruptions. Heightened income losses are associated with inadequate housing, lack of access to compensation schemes from public and private institutions, and exclusion from formal disaster responses. In the United States, Pleninger (2022[12]) shows that natural disasters primarily affect middle income households, which may help explain why aggregate inequality measures sometimes show little change following a disaster. Importantly, effects differ widely across sectors, and some workers may even experience short-term cyclical gains. The construction sector, for instance, where informal employment is often prevalent, frequently sees a surge in labour demand in the aftermath of disasters (Wagner, 2025[11]; Groen, Kutzbach and Polivka, 2020[13]). Following Türkiye’s 2023 earthquakes, Akarsu et al. (2026[14]) find that women, younger and less-experienced workers, as well as those earning lower wages, were hit hardest, facing both higher job losses and steeper wage declines. By contrast, construction workers experienced relative gains. Beyond income effects, Higa (2025[15]) highlights that in Peru, high temperatures reduce hours worked but the effect is primarily concentrated among informal jobs rather than among weather-exposed industries. The findings suggest that informal workers are more affected (i.e. working fewer hours) irrespective of the sector in which they work. This points towards multiple impact channels: demand shocks, infrastructure gaps (notably at home) and care burdens intensified by heat.

Exposure to extreme weather events and their frequency shape the overall risk faced by workers as well as their resilience (or lack thereof). Multiple and severe disasters tend to generate larger income losses (Pleninger, 2022[12]). If countries are unable to restore a fair distribution of income and resources, notably where inequality is already high, as well as broader access to social protection and health services, increasingly recurrent shocks can crowd out adaptation efforts and further undermine resilience (Cappelli, Costantini and Consoli, 2021[16]). Yet, prior experience can also help attenuate impacts, as areas that have experienced natural disaster in the past are often better prepared than locations where a “surprise” effect can have dramatic human and economic consequences (Pleninger, 2022[12]).

Against this backdrop, this section examines the risks from droughts and floods faced by workers, notably those in informal employment, using a three-dimension framework of exposure, sensitivity and adaptive capacity tailored to weather events (Box 5.1).

Box 5.1. Measuring drought and flood risks: Concepts and methodological approach

Copy link to Box 5.1. Measuring drought and flood risks: Concepts and methodological approach

The methodology to measure the levels of risk from weather events builds on a theoretical framework combining three core dimensions: exposure, sensitivity and adaptation. Together, the dimensions of sensitivity and adaptive capacity form the sub-concept of vulnerability. This framework is then adapted and operationalised in the context of droughts and floods. Annex 5.A provides more details on the risk framework adopted and the approach implemented to measure risk levels from droughts and floods.

Exposure is measured using data from the Emergency Events Database (EM‑DAT)¹ for the period 2000-23. For each KIIbIH country, data on droughts and floods are aggregated at the sub-national administrative level, using the Global Administrative Unit Layers (GAUL) level 1 classification developed by the FAO. These data are then linked to population data using the same GAUL framework. For both drought and floods, exposure is measured by normalising the total annual duration of events over the reference period and classifying the resulting indicator into three exposure levels: low (0-1 month per year); medium (1-2 months per year); and high (more than 2 months per year).
Sensitivity is measured through the sectoral link of jobs to droughts and floods – i.e. the degree to which a sector could be negatively affected by these hazards. Jobs are classified into three categories – direct, indirect and no/weak sensitivity – based on their ISIC‑04 codes. This sensitivity classification is tailored specifically and separately for droughts and floods, and slightly differs from the environmental link classification outlined in the first section of the chapter. Annex 5.A provides more details on the sensitivity classification of sectors for both droughts and floods.
Adaptive capacity is measured using the formality or informality status of workers aged 15 years and above as a proxy. As informal workers are usually excluded from employment-based social insurance, their adaptive capacity to job loss and income shocks is considered lower than that of formal workers.

All three dimensions are combined to build three levels of risks from weather events – low, medium and high (Figure 5.8).

Figure 5.8. Definition of levels of risks from weather events
Copy link to Figure 5.8. Definition of levels of risks from weather events

Exposure to droughts and floods varies widely across countries in both frequency and intensity

Countries differ widely in their exposure to these shocks, both in frequency (how often they occur) and intensity (how severe they are). EM‑DAT data for 2000-23 show clear contrasts between the two hazards. Droughts are relatively rare but long‑lasting.2 Over this period, KIIbIH countries experienced (on average) about one drought, with a mean duration of under six months. Floods, by contrast, were more frequent, with an average of seven events per country, but of much shorter duration, with a mean duration of only one month. On average, KIIbIH countries experienced 0.5 months per year of droughts and 0.4 months of floods, with extremely large variations across countries (Figure 5.9).

Data also suggest that regions face different types of challenges. Among KIIbIH countries, African countries are the most affected by droughts. Between 2000 and 2023, 73 of the 22 African countries included in the analysis experienced average yearly drought durations exceeding one month. On average, the African region faced a yearly drought duration of 0.9 months, compared with 0.5 months in the Americas, 0.4 months in Asia and less than 0.1 months in Europe. In several African countries, including Ethiopia, Kenya, Malawi and Niger, droughts are both recurrent and long‑lasting, reflecting their slow‑onset character and the sustained pressure they place on livelihoods that depend on rain‑fed agriculture and natural resources. By contrast, flood exposure is more unevenly distributed across regions and driven by a smaller number of highly affected countries such as Bolivia, the People’s Republic of China (hereafter “China”), Colombia, Kenya, Malawi and Niger, in which the average yearly flood duration exceeds one month. This diversity of countries reflects the frequency of localised flood events occurring across different parts of large and climatically diverse territories. While individual flood events are typically short‑lived, their recurrence throughout the year can result in high cumulative exposure, increasing the likelihood of repeated disruptions to work and income. Overall, the coexistence of slow‑onset and rapid‑onset hazards implies that workers may face different types of climate‑related risks over time, often through distinct transmission channels.

Figure 5.9. Exposure to droughts and floods varies substantially across countries
Copy link to Figure 5.9. Exposure to droughts and floods varies substantially across countries

Normalised yearly duration of droughts and floods over the period 2000-23

Together, these patterns highlight that risks from weather events are shaped not only by whether countries are exposed to droughts or floods, but also by the nature of that exposure. Droughts tend to generate prolonged and widespread stress, particularly in low‑income and agriculture‑dependent settings, while floods create repeated, localised shocks that can disrupt livelihoods multiple times within a year. Examination of population exposure shows no major differences between formal and informal workers. On average, across the KIIbIH, the shares of formal and informal workers who live in areas where the average yearly duration of droughts and floods was larger than one month per year are nearly identical (Figure 5.10). In other words, in most countries the informality status cuts across geographical locations, exposing all workers to the same risks. Yet, for informal workers, who often lack savings, insurance and adequate housing conditions, as well as access to social protection and infrastructure, both types of exposure can translate into heightened income volatility and limited capacity to recover between shocks.

Figure 5.10. Exposure of informal and formal workers to droughts and floods is similar
Copy link to Figure 5.10. Exposure of informal and formal workers to droughts and floods is similar

Share of informal and formal workers aged above 15 years who, on average over the period 2000-23, have been exposed to more than one month per year of droughts (Panel A) and floods (Panel B)

Informal workers are particularly sensitive to weather events

Although exposure captures how often and intensely countries are affected by droughts and floods, it does not necessarily indicate which workers are most likely to be affected. The effects of weather events also depend on workers’ sensitivity, that is, the extent to which their jobs are concentrated in sectors that can be directly or indirectly affected by these hazards (see Annex 5.A for more details on the classification of drought- and flood-sensitive sectors). The dimension of sensitivity is independent of that of exposure. For example, a worker may be employed in a drought-sensitive sector without living in a drought-prone area. Conversely, even in areas facing similar levels of exposure, workers may experience very different risks, depending on their sector of employment and their employment status.

Figure 5.11. Informal workers are disproportionately employed in sectors exposed to droughts and floods
Copy link to Figure 5.11. Informal workers are disproportionately employed in sectors exposed to droughts and floods

Share of informal and formal workers aged above 15 years employed in sectors that have a direct or indirect link to droughts (Panel A) and floods (Panel B), latest year available

Data show that sensitivity to weather events varies markedly between informal and formal workers. For both types of events and across all regions, informal workers are consistently more likely than formal workers to be employed in sectors that can be directly or indirectly affected by these hazards. This sensitivity largely reflects the disproportionate employment of informal workers in agricultural activities. For droughts, on average, 45% of informal workers operate in drought‑sensitive sectors compared to 13% of formal workers (Figure 5.11, Panel A). The gaps between informal and formal workers attain 51 percentage points in Africa and 36 in Asia. Overall, in 174 of 40 countries for which sectoral data are available, more than half of informal workers are employed in drought‑sensitive sectors, while the share of formal workers operating in these sectors rarely exceeds 20%. For floods, a similar pattern exists although the sensitivity of both informal and formal workers is larger, reflecting the wider set of sectors that could be potentially affected by floods. On average across KIIbIH countries, 62% of informal workers operate in flood‑sensitive sectors compared with 25% of formal workers (Figure 5.11, Panel B). In nearly all countries, more than half of the informal labour force operates in such sectors.

A higher sensitivity to droughts and floods, combined with reduced adaptive capacity, places informal workers at greater risk

Measuring risk as the combination of exposure, sensitivity and adaptive capacity shows that around one in three workers across KIIbIH countries face medium to high levels of risk from droughts and floods.

These risks are more pronounced for informal workers, who typically lack access to employment protection, insurance mechanisms or adaptive technologies, and are also more exposed to disruptions caused by droughts or floods. Large variations exist across regions, with informal workers from African and Asian countries facing the highest levels of risk (Figure 5.12). Across 36 countries with available data, 10% of informal workers face high levels of risk from droughts and 32% face medium levels, compared with about 1% of formal workers who face either medium or high levels (Figure 5.12, Panel A). Similarly, 5% of informal workers face high levels of risk from floods and 38% face medium levels, compared with less than 2% of formal workers (Figure 5.12, Panel B). Of 36 countries analysed, more than half of informal workers face medium to high levels of risk from droughts in 115 countries and from floods in 146 countries.

Figure 5.12. Informal workers face greater risks from droughts and floods than formal ones
Copy link to Figure 5.12. Informal workers face greater risks from droughts and floods than formal ones

Share of informal and formal workers aged above 15 years by the level of risk they face from droughts (Panel A) and floods (Panel B)

The decomposition by risk dimensions highlights that medium-to-high risks are primarily driven by the combination of workers’ informal status (low adaptive capacity) and their sectoral link to the hazard (high sensitivity), rather than by exposure alone. For both droughts and floods, most workers facing medium risk levels are informal workers who operate in sectors that can be directly affected by these hazards, but who live in areas with low exposure. Workers who face high risk levels tend to have a similar profile (informal with direct sectoral link) but live in areas that are more prone to droughts and floods. Conversely, risks rarely exceed low levels among workers with only a weak sectoral link to either hazard, regardless of where they live.

These insights suggest that should weather events intensify, the level of risk faced by many workers could rapidly increase. This is notably the case for informal workers who work in a drought- or flood-sensitive sector and already face medium risk levels at current exposure levels. For both hazards, these workers account for about one-quarter of the labour force across KIIbIH countries. A worsening in hazard frequency, duration or intensity would shift their exposure component upward, tipping a sizeable share of the workforce into high risk levels – even without changes in the sectoral or informal composition of the labour market. Importantly, under such a scenario, formalisation alone would not necessarily eliminate risks. If exposure becomes medium to high in hazard-prone areas, workers whose livelihoods depend on hazard-sensitive sectors would still face elevated risks of income and employment disruption, even if they formalise.

If the aggravation of events were very large, a second, more structural dynamic could emerge. Informal workers currently with low exposure and no (or weak) sectoral sensitivity, but low adaptive capacity, could progressively “move” along the risk continuum and face, over time, medium levels of risk as exposure expands geographically and temporally. In other words, workers currently shielded by their location could become more regularly affected, increasing their level of risk. In this regard, adaptation strategies should target not only workers at high risk today but also anticipate future transitions by prioritising: i) informal workers in hazard-sensitive sectors who already sit close to the medium/high threshold; and ii) informal workers with a weak sectoral link but who live in areas in which exposure is likely to rise.

Overall, these dynamics are likely to be non-linear and path dependent. Repeated or overlapping shocks can erode coping capacity (e.g. savings, assets, health and debt position), which can progressively reduce adaptive capacity – including among workers who are currently formal. As a result, future increases in exposure may translate into larger-than-proportional increases in medium to high risk, especially for groups already concentrated near the risk thresholds.

Informal workers’ risks from droughts and floods are amplified at the household level

A disproportionate concentration in households in which all members are informally employed exacerbates the risks of informal workers from droughts and floods. Across KIIbIH countries, 44% of workers who live in fully informal households face medium to high risks from droughts, compared with 16% of those in mixed households and less than 1% of those in fully formal households (Figure 5.13, Panel A). The pattern is identical regarding floods, with workers who belong to fully informal households facing far greater risks than the working members of other households (Figure 5.13, Panel B).

Figure 5.13. Fully informal households face greater risks from droughts and floods
Copy link to Figure 5.13. Fully informal households face greater risks from droughts and floods

Share of informal and formal workers aged above 15 years who face medium and high levels of risk from droughts (Panel A) and floods (Panel B) by the informality status of the household in which they live, latest year available

Poverty is also a key dimension of workers’ vulnerability to weather events, especially for those in informal employment – with large overlaps across these two realities. On average across KIIbIH countries, more than half of workers living in households belonging to the poorest welfare quintile (Q1) face medium to high levels of risks from droughts and floods (Figure 5.14, Panels A and B), and nearly all of them are informally employed. As household welfare rises, the share of workers facing medium to high risks declines. In the richest quintile (Q5), fewer than 20% of workers face medium to high risks from either hazard. The gradient is particularly steep in Africa, where the share of workers facing medium to high risks from droughts falls by around 50 percentage points – from over 75% in the poorest quintile to about 25% in the richest. This highlights the combined effects of widespread informality, the central role of agriculture within the poorest segments of the population, and the continent’s relatively high exposure to droughts and floods.

Poverty rates are much higher among workers who face greater environmental risks, especially among at‑risk informal workers. On average across KIIbIH countries, poverty affects 39% of informal workers who face medium to high risks from droughts (37% for floods), compared with 22% of informal workers who face low risks (23% for floods). By contrast, 28% of formal workers who face medium to high risks from droughts are poor (7% for floods).

Figure 5.14. Poorer households face greater risks from droughts and floods
Copy link to Figure 5.14. Poorer households face greater risks from droughts and floods

Share of workers aged above 15 years by the level of risk they face from droughts (Panel A) and floods (Panel B), across household welfare quintiles

Heat stress disproportionately affects the productivity and income of informal economy workers

Copy link to Heat stress disproportionately affects the productivity and income of informal economy workers

One major consequence of climate change is global warming, which increases both the frequency and intensity of extreme heat events, including heatwaves (IPCC, 2023[9]). As these events multiply, the risk of heat stress rises because workers are exposed to high temperatures for longer periods, and potentially at levels that exceed what the human body can safely tolerate. Heat stress is a health hazard that occurs when the heat received by the body exceeds its capacity to dissipate excess heat through physiological mechanisms such as sweating and cardiovascular regulation, without causing impairment (Kjellstrom et al., 2016[18]). As extreme heat events become more frequent, more intense and last longer, periods during which ambient temperatures, humidity and solar radiation exceed these physiological thresholds become more common.

Such excessive heat can have severe short- and long-term health consequences. It notably impairs the body’s ability to maintain a stable internal temperature, leading to physical strain. In severe cases, it can cause adverse health outcomes ranging from mild conditions such as heat rash, heat cramps and heat exhaustion to potentially fatal heatstroke (NIOSH, 2026[19]). Beyond acute health outcomes, growing evidence also underlines a range of longer-term and chronic health consequences of repeated heat exposure, such as developing serious and debilitating chronic diseases, impacting the cardiovascular and respiratory systems, and negatively affecting kidney function (Flouris et al., 2024[20]). For pregnant women, heat stress can lead to a wide range of adverse health effects, including preterm birth, low birth weight, congenital anomalies and still birth (Lakhoo et al., 2025[21]).

Heat stress can also severely degrade labour productivity. Excessive body temperature and dehydration not only increase the risk of heat exhaustion and heat stroke, but may also slow work, raise the likelihood of errors while working, and increase the risk of occupational injuries (Kjellstrom et al., 2017[22]). The productivity loss due to workers slowing down already results in billions of potential hours of work lost. In turn, this can be expressed in millions of full-time equivalent job losses or in terms of monetary economic loss. Based on this chain of causality, the International Labour Organization (ILO) estimated that global economic losses linked to heat stress could rise from USD 280 billion in 1995 to USD 2.4 trillion by 2030 (ILO, 2019[23]). Similarly, the Lancet Countdown estimated that reduced labour capacity due to heat exposure resulted in global potential income losses of around USD 1 trillion in 2024, equivalent to 0.97% of global gross domestic product (GDP) – with the largest relative losses occurring in less advanced economies (Romanello et al., 2025[24]). In aggregate, these economic impacts of heat may be even larger and extend well beyond individual effects on labour productivity (Callahan and Mankin, 2022[25]; Miller et al., 2021[26]; Diffenbaugh and Burke, 2019[27]). For instance, excess heat that leads to school closures or reduced working hours in certain sectors may aggravate economic losses.

The impacts of heat on labour productivity depend strongly on the nature of the job and the conditions under which it is performed. Productivity losses are particularly severe in activities that involve heavy physical effort (such as agriculture, construction and mining), which tend to employ more men, and other labour‑intensive occupations, notably indoor manufacturing activities without cooling (such as textile factories, notably in Southeast Asia), where women are often overrepresented. In these jobs, tasks often require sustained physical exertion in environments in which exposure to high temperatures cannot easily be avoided or mitigated. In these contexts, heat stress can rapidly translate into sharp productivity losses, shorter working hours or work stoppages, with direct consequences for earnings, especially where income is closely tied to daily output or hours worked. Other factors may amplify these effects. Working outdoors without shade or indoors with poor ventilation, and the need to wear heavy or protective equipment, can all intensify heat stress by limiting the body’s ability to dissipate heat. Importantly, high temperatures can also impair productivity in less physically demanding jobs. In situations of inadequate indoor cooling or ventilation, excessive heat can make it difficult to perform even basic office and desk‑based tasks due to mental fatigue, reduced concentration and discomfort.

Building on the approach developed by Kjellstrom et al. (2017[22]) and ILO (2019[23]), which examined heat-related productivity losses across all workers regardless of informality status or location, this section applies a similar framework to estimate productivity losses associated with historical and projected heat exposure, with a specific focus on informal workers (Box 5.2).

Box 5.2. Measuring heat-related productivity losses: Concepts and methodological approach

Copy link to Box 5.2. Measuring heat-related productivity losses: Concepts and methodological approach

The methodology to estimate productivity losses associated with heat exposure builds on the approach developed by Kjellstrom et al. (2017[22]) and ILO (2019[23]). In these studies, temperatures and associated productivity losses are computed at a fine spatial resolution using climate grid cells. Population data are then overlaid at the grid‑cell level using Columbia University’s Gridded Population of the World dataset. To estimate productivity losses across sectors, national employment‑to‑population ratios are applied uniformly across all grid cells for agriculture, construction, industry and services, constituting a strong and potentially biased assumption.

While the methodology developed in this chapter follows a similar overall approach to compute temperature levels and associated productivity losses, it produces these estimates at the level of countries’ administrative areas. In parallel and at the same spatial scale, the approach relies on KIIbIH geolocated household microdata to generate population and sector-specific labour estimates. Arguably, this major update makes it possible to more precisely link climate data with population information, and to analyse the productivity impacts of heat exposure across a broader range of social, demographic and economic dimensions, including workers’ household environments.

The method developed in this chapter follows three main steps:

Computation of heat exposure: Hourly wet bulb globe temperatures (WBGT) are computed using historical climate data and bias‑adjusted future climate projections for the period 2010 to 2069, providing a consistent measure of heat stress over time and across locations. WBGTs are calculated for each small‑resolution climate grid cell and for a daily theoretical 12‑hour working period. This step provides a consistent measure of heat stress over time and across locations.
Estimation of productivity losses: Heat exposure is translated into daily productivity losses using workload‑specific response curves derived from epidemiological studies that distinguish three types of workers depending on the level of physical intensity their job involves: light = 200 watts (W); moderate = 300 W; and heavy = 400 W. For each grid cell, heat-related productivity losses are computed separately for each type of worker. To capture long-term climate trend, productivity losses for all three levels are then aggregated into two 30-year period averages. Within this chapter, the period 2010‑39 is referred to as 2025, and the period 2040‑69 as 2055.
Matching climate and labour data to a common spatial classification: Finally, climate grid cells are matched to sub‑national administrative areas, using the GAUL level 1 classification developed by the FAO as a spatial reference. In parallel, KIIbIH data are geolocated and matched to the same list of sub‑national administrative areas. For each sub‑national administrative area, this spatial correspondence makes it possible to obtain two estimates: i) heat-induced productivity losses by level of sectoral physical intensity (200 W, 300 W and 400 W); and ii) the sectoral composition of the labour force. By combining these two measures, it is possible to derive estimates of total productivity loss for each sub-national administrative area, as well as at the country level.

Annex 5.B presents the methodology in more detail, including data sources and assumptions.

Extreme temperature predominantly affects countries and areas with a large informal economy

Under the intermediate‑emissions scenario (SSP2‑4.5),7 global projections indicate that temperatures will exceed 30°C during an average of 11.9 days in 2030 (3.3% of the year) and 35°C during 5.4 days (1.5% of the year), with considerable variations across regions and income-groups (OECD, 2026[28]). Low- and lower-middle income countries are notably projected to experience more additional hot days compared with upper-middle and high-income countries (Maes et al., 2025[29]). Likewise, the SSP2‑4.5 scenario projects that Africa will experience 19.5 days above 30°C and 13.2 days above 35°C in 2030, whereas Europe will experience, respectively, only 3.1 and 0.6 days on average. In general, many economies with large informal sectors are located in tropical and subtropical regions, where baseline temperatures are high and heat extremes occur more frequently. As a result, the projected share of days in 2030 with temperatures above 30°C and 35°C is considerably higher in countries with high rates of informal employment (Figure 5.15).

Figure 5.15. Countries in which the informal economy dominates are also more exposed to hot days
Copy link to Figure 5.15. Countries in which the informal economy dominates are also more exposed to hot days

Correlation between the informal employment rate and the share of days in 2030 with median temperatures above 30°C (Panel A) and 35°C (Panel B), and the projected change in the number of days above these thresholds between 2030 and 2050 (bubble size)

Climate projections suggest that countries with current high informality will experience the steepest increases in extremely hot days in the coming decades. Between 2030 and 2050, the yearly share of days above 30°C and 35°C is expected to rise more sharply in these economies than in those with lower informality levels. For both temperature thresholds, Figure 5.15 shows that the largest bubbles – representing the biggest percentage‑point increases in the annual number of days above these thresholds – correspond to countries that either have a very large informal economy or already experience many hot days, or both.

Informal workers are concentrated in jobs that are physically demanding

Not only are countries with a large informality footprint more exposed to extreme heat, but informal workers are concentrated in economic sectors in which excessive heat can severely hinder work. Following the approach of Kjellstrom et al. (2017[22]) and ILO (2019[23]), this analysis divides sectors of the economy into three categories according to the level of physical effort necessary to perform a job: light, moderate and heavy (see Box 5.2 and Annex 5.B for more details).

Across KIIbIH countries, on average, 36% of workers aged above 15 are employed in sectors involving heavy physical effort, such as agriculture and construction. A further 12% work in sectors characterised by moderate physical effort, including mining and quarrying; manufacturing; electricity, gas, steam and air‑conditioning supply; and water, sewage, waste and remediation activities. These shares vary widely across regions and countries, reflecting differences in economic structure and the size of the rural sector. In many Africa and Asian countries, more than half of all workers are employed in jobs involving heavy physical effort. By contrast, in most Latin American and Caribbean countries, this share falls below 30% and is less than 20% in half of them.

Informal workers predominantly hold jobs involving moderate and heavy physical effort (Figure 5.16). On average, 55% of informal workers are employed in sectors that require either heavy or moderate physical intensity, including 44% in heavy‑intensity sectors and 11% in moderate‑intensity sectors. Among formal workers, by contrast, only 26% are employed in sectors with moderate or heavy physical intensity (11% in heavy‑intensity and 15% in moderate‑intensity sectors). This uneven sectoral distribution of informal and formal workers translates into higher informality rates in more physically demanding activities. Across the KIIbIH, 86% of workers in the high-intensity sectors of agriculture and construction are informally employed, compared with 64% in moderate‑intensity sectors and 62% in light-intensity sectors. In reality, these averages conceal substantial cross‑country variation. In most African countries, the gap in informality rates between heavy- and light-intensity sectors remains relatively small, underlining the widespread nature of informality across these economies. Conversely, gaps exceed 30 percentage points in countries such as Argentina, the Bahamas, Brazil, Jamaica, the Maldives, Mongolia and Suriname.

Figure 5.16. Informal workers are more likely to hold jobs that involve heavy physical effort
Copy link to Figure 5.16. Informal workers are more likely to hold jobs that involve heavy physical effort

Share of informal (Panel A) and formal (Panel B) workers aged above 15 by the level of physical intensity of their jobs, latest year available

Productivity losses associated with heat are disproportionately larger for informal workers

For any given job, the impact of heat on labour productivity depends heavily on the level of physical effort required. According to the International Organization for Standardization, heat stress hinges on three key parameters: i) the characteristics of the environment; ii) the heat generated inside the body through physical activity (often referred as the metabolic rate); and iii) the clothing worn, which affects the body’s ability to exchange heat with the environment (ISO, 2017[31]). In this section, the environmental heat level is measured using the wet bulb globe temperature (WBGT), one of the most widely used heat‑stress indices in occupational health (see Box 5.2 and Annex 5.B for more details). The metabolic rate is assessed following the approach of Kjellstrom et al. (2017[22]) and ILO (2019[23]), which applies risk functions that translate heat exposure into productivity losses – i.e. the reduction in work capacity. These risk functions derive from epidemiological studies and are estimated for the three levels of physical intensity described in the previous sub-section: light (200 W), moderate (300 W) and heavy (400 W) (see Box 5.2 and Annex 5.B for more details). The third parameter, clothing insulation, is assumed to be constant and standard across workers.

On average across KIIbIH countries, excessive heat reduced total working hours by 2.2% in 2025 (Figure 5.17), amounting to about 60 billion hours of work lost and equivalent to about 28.8 million full‑time jobs. Assuming that reductions in labour input translate proportionally into output losses, the associated combined GDP losses are estimated at about USD 500 billion in PPP terms, out of a combined total GDP of USD 24 000 billion (PPP) for KIIbIH countries.

Figure 5.17. Productivity losses due to excessive heat are substantially larger for informal workers
Copy link to Figure 5.17. Productivity losses due to excessive heat are substantially larger for informal workers

Heat-related productivity losses of informal, formal and all workers aged above 15, 2025

These losses vary widely across countries, reflecting both differences in heat exposure and the structure of national labour markets. On average in 2025, productivity losses reached 1.1% across American countries covered by the KIIbIH, 2.9% across Asian ones, and 3.1% across African ones. In Brazil, Ghana, Indonesia, Mexico Myanmar, Thailand and Viet Nam, the size of the labour market and the level of losses translate into more than one million jobs lost. Because of differences in economic size, the largest monetary losses in 2025 occur in Asian KIIbIH countries, with productivity losses corresponding to an estimated GDP shortfall of about USD 340 billion (PPP) under the assumption that output is proportional to labour input (out of a combined total GDP of around USD 9 200 billion in PPP). In Indonesia, Thailand and Viet Nam, these losses each exceed USD 80 billion in PPP. By comparison, GDP losses are estimated at around USD 130 billion (PPP) in the Americas and USD 40 billion (PPP) in Africa.

Heat-related productivity losses are substantially larger for informal workers than for formal ones. On average across the KIIbIH countries, informal workers lost 2.4% of their working hours in 2025, compared with 1.4% among formal workers (Figure 5.17). This pattern holds true in every KIIbIH country, with the gap between informal and formal workers exceeding 3 percentage points in countries such as Mali and Niger.

The impacts of heat on labour productivity are particularly strong for jobs that involve heavy physical effort. On average across KIIbIH countries, current climatic conditions are estimated to lead to productivity losses of more than 4% for these jobs (Figure 5.18), exceeding 7% in countries such as Benin, Gambia, Ghana, Thailand and Togo. The impacts are particularly pronounced in countries facing particularly severe heat conditions.

Figure 5.18. Informal workers in physically demanding jobs suffer from the largest heat-related productivity losses
Copy link to Figure 5.18. Informal workers in physically demanding jobs suffer from the largest heat-related productivity losses

Heat-related productivity losses of informal, formal and all workers aged above 15 by the level of physical intensity of their jobs, 2025

Heat-related productivity losses are not homogeneous across formal and informal workers

Because of the strong linkages between rural employment and agriculture, productivity losses due to heat are substantially larger for rural workers than for urban ones. On average across KIIbIH countries, rural workers lost 2.9% of their working hours in 2025, compared with 1.6% among urban workers (Figure 5.19, Panel A). Losses among informal rural workers are particularly pronounced in countries that combine high exposure to extreme temperatures with a rural labour market that is poorly diversified and dominated by agricultural activities. In Ghana, Thailand and Togo, productivity losses among informal rural workers exceeded 6%.

Heat‑related productivity losses also vary sharply by educational attainment. For both informal and formal workers, losses rise as education levels fall (Figure 5.19, Panel C). This pattern is especially marked among informal workers: across the KIIbIH, informal workers with less than primary education experienced productivity losses of 2.3% in 2025, compared with 2.1% for those with primary education, 1.6% for those with secondary education, and 1% for those with tertiary education. These differences largely reflect the concentration of low‑educated informal workers in manual and physically demanding activities that require a low level of qualifications but are highly sensitive to elevated temperatures.

The data further show that own‑account and contributing family workers face substantially higher productivity losses than other workers. Across the KIIbIH, their respective heat‑related productivity losses are estimated at 2.4% and 3%, compared with 1.6% among employees (Figure 5.19, Panel D). This mainly reflects that agricultural jobs are disproportionately held by own‑account and contributing family workers, most of whom are informal, while employees are more prevalent in sectors with no environmental link (see Figure 5.6) and in clerical tasks that require only light physical effort. Incidentally, own-account and contributing family workers, notably when informally employed, tend to be the least protected by occupational safety and health policies. The 2006 Thai Occupational Standard, for example, orders work interruptions and workplace adjustments according to specific WBGT thresholds, and mandates employers to deploy personal protective equipment, post warning notices and arrange health check-ups for employees. However, it does not mention self-employed workers. Likewise, India’s approach to occupational health and safety is outlined in both the 2020 Occupational Safety, Health Working Conditions Code and the 2009 National Policy on Occupational Safety, Health and Environment at Workplace. In principle, these policies recognise unorganised workers; in practice, they are designed for establishments with 10 or more workers, typically excluding home-based workers (Sinha et al., 2026[34]).

Figure 5.19. Informal and rural workers with low levels of education and in vulnerable forms of employment experience larger heat-related productivity losses
Copy link to Figure 5.19. Informal and rural workers with low levels of education and in vulnerable forms of employment experience larger heat-related productivity losses

Heat-related productivity losses of informal, formal and all workers aged above 15 by socio-demographic characteristics, 2025

The limitations of the model applied may lead to an underestimation of the losses faced by many own‑account and contributing family workers, particularly in urban settings. The current level of spatial and sectoral detail – in both climate and labour data – does not allow to capture the specific heat exposure of certain urban occupations (e.g. home‑based workers, street vendors and market traders). Yet evidence from Bangkok (Thailand) and Delhi (India) shows that these workers are among the most exposed to heat (Box 5.3). Moreover, their vulnerability is often compounded by overlapping disadvantages, including limited access to shade, ventilation and basic services, poor housing conditions, and restricted access to health care.

Box 5.3. The vulnerability of urban informal workers to heat stress: A granular perspective from WIEGO

Copy link to Box 5.3. The vulnerability of urban informal workers to heat stress: A granular perspective from WIEGO

Cities are hotspots of climate risk, and a growing body of evidence shows that informality shapes such risk across housing, industry, livelihoods, infrastructure and ecological systems (Satterthwaite et al., 2018). While increasing attention has been paid to informal settlements as sites of both vulnerability and innovation (IPCC, 2022[35]), far less is known about how informal livelihoods shape risk, experience climate impacts and contribute to adaptation. Recent research by Women in Informal Employment: Globalizing and Organizing (WIEGO) with street vendors and home-based workers in Bangkok and Delhi identifies three interconnected channels through which heat shapes productivity for urban workers in the informal economy: infrastructure deficits, demand disruptions and intensified care burdens (WIEGO, 2026[36]).1 Standard productivity measures, which narrowly focus on labour inputs, tend to overlook these constraining factors (Chen et al., 2025[37]).

Across both cities, infrastructure deficits were a key factor affecting heat-related shifts in productivity. For home-based workers, housing conditions directly shape heat exposure. In Delhi – where most research subjects lived in high-density settlements with limited or irregular access to basic services – only one-third reported access to clean water during the heat period. Six in ten respondents spent more than two hours per day collecting water, nearly double the share in milder conditions. Home-based workers who reported increased time collecting water were more likely to report reductions in work hours. Street vendors depend on public space and access to shade, water and toilets. In Delhi, over 70% of surveyed vendors lacked access to these basic services, with women facing systematically lower access. The largest gender gap was in access to shade, at nearly 20 percentage points. Vendors without shade were also more likely to report reduced working hours during the heat period.

A second channel operates through demand. In Bangkok and Delhi, 80% and 96% of vendors, respectively, reported a decline in customers, with similar patterns observed across genders. Among home-based workers in Delhi, 74% reported reduced work orders during the heat period, almost double the proportion observed during milder periods. While fluctuating demand and irregular access to raw materials are common features of this work, these findings suggest that heat amplifies these challenges. Evidence from previous crises, such as the COVID-19 pandemic, suggests that heat may affect supply-side factors by disrupting the mobility of workers and intermediaries, delaying the delivery of inputs and exacerbating unequal value-chain dynamics, such that unfulfilled orders can lead to future cancellations.

Finally, heat-related illness increases care responsibilities and can reduce labour availability, particularly for women. In Bangkok, 64% of female street vendors and 77% of female home-based workers reported increased direct care responsibilities (e.g. taking care of elders, children or the sick) during extreme heat. Workers reporting increased care responsibilities were also more likely to reduce their working hours. In Delhi, home-based workers in larger households (more than six members) were more likely to miss work due to illness. While this pattern may reflect the direct health impacts of heat exposure, it also suggests that household composition mediates impacts and the ability to cope.

Together, this evidence shows that heat-related productivity losses are shaped by the interaction of gender and informality, operating through unequal access to infrastructure, demand disruptions and unpaid care responsibilities. These findings are consistent with emerging evidence from other contexts, including econometric analyses from Peru (Higa, 2025[15]) and survey-based research on home-based workers (HomeNet South Asia, 2022[38]) and street vendors (Ghosh, 2024[39]) in South Asia. More research is needed to understand how productivity impacts vary across different groups of workers in the informal economy and through the multiple channels identified here.

Note:

1. The evidence is based on two survey findings. In Bangkok, 496 street vendors (70% women) and 530 home-based workers (84% women) were surveyed in May 2025 and asked to recall their experiences during April, the city’s hottest month. In Delhi, approximately 500 street vendors (36-41% women) and 300 home-based workers (all women) were surveyed before and after the summer heat in the same markets and neighbourhood clusters. This allowed a cohort analysis of aggregate changes between milder and extreme heat conditions.

In aggregate, male workers tend to experience larger productivity losses than women, largely because they are overrepresented in sectors such as agriculture and construction. On average across KIIBIH countries, male workers lost 2.4% of their working hours in 2025, compared with 2% for female workers (Figure 5.19, Panel B). In the Americas, although overall productivity losses are lower than in Africa and Asia, the shortfall for informal male workers is twice as large as that for informal female workers. This gender gap is closely linked to gender‑based occupational segregation, where men and women tend to be concentrated in different sectors and types of work (OECD, 2023[40]; OECD, 2024[41]). Often, men are overrepresented in jobs traditionally associated with physically demanding activities. Worldwide, in 2023, men accounted for 92% of total employment in the construction sector and 86% in mining (OECD, 2023[40]). Such concentration exposes these workers to higher potential productivity losses under extreme heat.

However, the aggregate gender gap conceals very different realities across employment statuses. When disaggregating data by type of employment, the gender gap in productivity losses persists for employees and own-account workers but tends to disappear – and in some cases reverse – among contributing family workers (Figure 5.20). This is notably the case across several African and Asian countries included in the analysis, such as Cameroon, Gambia, Indonesia and Thailand, where female contributing family workers (i.e. helping a family or household member run a business, farm or job, without regular payment) experienced larger heat‑related productivity losses than their male counterparts in 2025. This is not trivial, as women are far more likely to be engaged in this form of work. In 2023, for example, only 7% of employed Southeast Asian men worked as contributing family workers, compared with 20% of women (OECD, 2024[41]).

Figure 5.20. Gender-based differences in heat-related productivity losses largely depend on the employment status
Copy link to Figure 5.20. Gender-based differences in heat-related productivity losses largely depend on the employment status

Heat-related productivity losses of informal workers aged above 15 by gender and employment status, 2025

Importantly, the gender gap in productivity losses fails to consider the significant heat‑related vulnerabilities women face through their unpaid domestic work. Globally, women spend 2.6 times more hours on unpaid care and domestic tasks than men (OECD, 2023[40]). This work has long been excluded from GDP and remains largely invisible in monetary terms (van de Ven, Zwijnenburg and De Queljoe, 2018[42]). Across the OECD, unpaid housework has been valued at between 15% and 27% of GDP in 2021 (OECD, 2021[43]) but, overall, data remain scarce and limited to few countries with reliable information. As a result, current measures likely underestimate the full productivity losses women experience, including in unpaid labour. A 2023 study covering India, Nigeria and the United States found that 75% of women’s heat‑related productivity losses stem from their unpaid work, revealing how much of the burden remains unmeasured (Adrienne Arsht-Rockefeller Foundation Resilience Center, 2023[44]). The study suggested that accounting for unpaid work would raise women’s estimated heat‑related productivity losses by 260%, compared with a 76% increase for men.

Heat stress not only has direct and largely unaccounted effects on women’s productivity, it may also indirectly amplify their caregiving burden. Evidence on these indirect effects remains scarce, primarily because of the difficulty of measuring and monitoring time devoted to unpaid care and domestic activities.8 However, growing evidence suggests that, as children, elderly and working-age adults face greater risks of heat exhaustion, heat-related illnesses and co-morbidities, the resulting additional care needs are likely to fall disproportionately on women’s shoulders, increasing the already unequal share of unpaid care and domestic work they undertake (OECD, 2023[40]; HomeNet South Asia, 2022[38]). Heat-related disruptions, such as school closures, may further exacerbate the demand for care and force women to reduce or forgo income-generating work.

These unaccounted impacts are often most severe in low-income and informal households, particularly those living in urban slums or informal settlements built with heat-retaining materials and lacking proper ventilation or cooling. For women who remain at home to perform unpaid care and domestic tasks, this translates into heightened vulnerability to heat. For other workers who spend long hours in physically demanding jobs under extreme heat and then return to dwellings that stay hot well into the night, the result is a cycle of continuous exposure and insufficient recovery (Meade et al., 2025[45]). For workers in the informal economy – especially those living in informal settlements – climate exposure accumulates across both workplace and home, reducing productivity, increasing health risks, and eroding long-term earning potential.

Effective policy responses must therefore integrate all dimensions of vulnerability, including the household environment, living conditions, access to basic services, the distribution of unpaid care and domestic work, and workplaces outside formal regulations. Importantly, these policies should systematically apply a gender lens since women and men may experience these vulnerabilities through different channels, some of which may remain hidden or overlooked due to limited data. Without this broader perspective, interventions risk overlooking the conditions in which informal workers live and the cumulative exposure that shapes their overall capacity to cope with extreme heat.

Climate projections suggest that the impacts of heat stress on informal workers may aggravate by 2055

Across KIIbIH countries included in the analysis, rising temperatures already increased the share of total working hours lost from 1.7% in 1995 to 2.2% in 2025. Under the intermediate‑emissions scenario (SSP2‑4.5), the “middle‑of‑the‑road” development pathway – in which social, economic and technological trends continue broadly in line with historical patterns – rising temperatures are expected to raise the share of total working hours lost to 2.8% in 2055 (Figure 5.21, Panel A). Under current labour market structures, this amounts to about 17 billion additional hours of work lost because of extreme heat, rising from 60 billion hours in 2025 to 77 billion in 2055. This is equivalent to more than 8 million additional full‑time jobs lost.

Losses will be far more severe in several countries, particularly in Africa and Asia. In Benin, Ghana, Mali, Myanmar, Thailand, Togo and Viet Nam, productivity losses are projected to exceed 5%, while estimates for Niger reach more than 7% (Figure 5.21, Panel A). Because economies differ in size and population, the distribution of lost hours and jobs is uneven. Most jobs losses will occur in large economies experiencing rapid temperature increases. By 2055, Brazil, Indonesia, Mexico, Thailand and Viet Nam are each expected to lose more than 500 000 additional full‑time jobs.

Although all workers will be affected, informal workers will face the steepest increases under current labour‑market structures. On average across KIIbIH countries, their productivity losses are projected to rise by 0.6 percentage points, compared with an increase of 0.4 percentage points for formal workers (Figure 5.21, Panel B). In Niger, the increase for informal workers exceeds two percentage points.

Several structural factors may amplify these losses by 2055. First, these projections are based on an intermediate‑emissions scenario (SSP2‑4.5) consistent with a global temperature increase of around 2.7°C by 2100 relative to 1850-1900 (Chen et al., 2023[46]). Under a different and more intense pathway, global temperatures could rise by more than 4°C, substantially increasing productivity losses. Second, the impacts of rising temperatures are uneven across regions and countries. Less‑developed economies are more exposed to extreme heat and often face deep‑seated socio‑economic vulnerabilities, intensifying the consequences of heat stress. Projections in this chapter assume constant population and labour-market structures, but demographic dynamics may substantially compound and modify these patterns. For example, many African countries face the largest projected increases in heat stress while experiencing rapid population growth, which could result in a disproportionate increase of heat-related productivity losses. By contrast, in rapidly ageing regions projected to experience more frequent and intense heat events – e.g. Southeast Asia – a shrinking labour force may reduce heat-related productivity losses but amplify and aggravate the health effects of heat exposure as older people have more difficulties in adapting to high heat levels (Cheng et al., 2018[47]). Third, and closely related to the limitations of assuming constant population and labour-market structures, future economic growth – or lack thereof – may profoundly alter the structure of the economy and employment. As a result, estimated productivity losses could differ markedly under alternative growth and structural change scenarios. Finally, poor housing and lack of urban planning may aggravate the effects of heat in the future. Informal settlements often rely on materials that trap heat, have poor ventilation, and offer limited access to basic services such as electricity or cooling. Dense urban layouts, combined with limited green spaces and extensive paved surfaces, amplify local temperatures through the urban heat island effect. Together, if left unaddressed, these factors may increase continuous exposure to heat at home and at work, reducing workers’ ability to rest, recover and maintain productivity (Meade et al., 2025[45]).

Figure 5.21. Productivity losses due to heat are projected to increase by 25% by 2055
Copy link to Figure 5.21. Productivity losses due to heat are projected to increase by 25% by 2055

Heat-related productivity losses of all (Panel A) and informal (Panel B) workers aged above 15, 1995, 2025 and 2055

Conclusion and policy discussion

Copy link to Conclusion and policy discussion

Environmental degradation and climate events such as droughts, floods and extreme heat pose urgent and growing threats to human lives, economic activity and work worldwide. These threats are particularly pronounced in developing and emerging economies, with particularly acute consequences for workers in the informal economy. New evidence from the KIIbIH shows that informal workers depend heavily on the natural environment for their livelihoods and face far greater exposure to environmental risks than formal workers. Their vulnerability is characterised by structural and often overlapping disadvantages, including limited social protection coverage, lower levels of education, discriminatory social norms, concentration in fully informal households, lack of access to quality infrastructure and poor housing conditions.

The scale and distribution of risks across informal workers vary considerably depending on the type of event considered and the characteristics of workers. Informal agricultural workers are among the most vulnerable to all types of events, with direct and potentially dramatic consequences of hazards on their income and livelihoods. The concentration of informal workers in agriculture makes agricultural and rural development policies central pillars of any resilience strategy. These risks also differ by gender, with men more often concentrated in outdoor agricultural activities and women more exposed through downstream agri-food activities, such as processing and sales, as well as informal service and vending activities, especially in urban areas. Many informal urban workers (e.g. home‑based workers, street vendors and market traders) are among the most exposed to heat stress. Their vulnerability is often compounded by limited access to shade, ventilation and basic services (e.g. drinkable water), poor housing conditions, and restricted access to health care. Low-paid and poor informal workers are also disproportionately concentrated in environmentally linked sectors as well as drought- and flood-sensitive sectors, exacerbating their vulnerability to climate hazards.

In this context, securing the livelihoods of informal economy workers against environmental risks requires both a holistic approach that addresses the core dimensions of resilience – adaptation and mitigation – and tailored solutions that tackle the specific risks faced by diverse groups within the informal economy. Both approaches should be locally led and systematically informed by a gender lens, to ensure that policies reflect local realities and account for the different ways in which women and men experience, anticipate and cope with environmental risks. Against this backdrop, a strategy could be articulated around five major policy areas: i) consolidate access to traditional and new social protection tools; ii) invest in skills, training and human capital; iii) strengthen occupational safety and health policies and early warning systems; iv) establish comprehensive agricultural and rural development policies to build resilience; and v) invest in resilient urban infrastructure.

Consolidate access to traditional and new social protection tools

Traditional social protection instruments, such as unemployment benefits, cash transfers, food assistance and public works programmes, must remain the cornerstone of resilience towards environmental degradation. Ultimately, there is no better preparation than ensuring that existing social protection systems cover as many people as possible.

Recent policy development has framed adaptive social protection – i.e. systems that scale up automatically in response to climate shocks – as a core climate strategy rather than a residual safety net (ILO, 2024[48]). Adaptive social protection rests on two central principles. First, it should entail a wide range of programmes, from productive inclusion to active labour market programmes and insurance schemes. Second, investment in those delivery systems should occur in advance (OECD, 2024[49]). In this regard, the role of social protection should be understood as a continuum, with the right mix of interventions before and after a risk materialises. Notable examples include Ethiopia's Productive Safety Net Programme and India's Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA), which provide income support while building productive assets that can reduce future vulnerability (IPCC, 2022[35]). In Kenya, the Hunger Safety Net Programme (HSNP) functions as a traditional cash-transfer programme but integrates a drought-related scale-up mechanism that extends coverage to additional beneficiaries depending on the drought’s severity – assessed through remote sensing data monitoring of vegetation conditions (OECD, 2024[49]).

Beyond traditional social protection instruments, new approaches such as parametric (or index-based) insurance9 may be well suited to protect informal workers from extreme weather events, particularly heat. These products may be offered by private insurers, public schemes or public-private partnerships. Unlike conventional indemnity insurance, parametric products automatically pay out a fixed amount when a pre-defined physical threshold or index value is reached – such as daytime and/or nighttime temperatures, humidity level, rainfall deficit or flood gauge reading – without requiring individual loss assessment (McLeod and Goering, 2025[50]; OECD, 2025[51]). This makes them potentially far better suited for workers of the informal economy, although little evidence on their impact exists at present. Several experiments have incorporated women’s unique exposure patterns and needs into the trigger mechanisms and payout structures, suggesting that parametric insurance may be particularly effective in addressing the gender dimension of extreme weather events (Suzumu and Chanel, 2024[52]).

Recently, several such programmes have been launched in India. In 2023, the Self-Employed Women’s Association (SEWA), in partnership with the Climate Resilience Center and the insurer Blue Marble, piloted the Extreme Heat Income Insurance in Ahmedabad and the state of Gujarat (Climate Resilience Center, 2026[53]; Dabrowski, 2023[54]). The scheme covers 21 000 female workers, including waste recyclers, street vendors, salt pan miners, construction workers and home-based workers. It provides income compensation when temperatures exceed a pre-defined threshold. Rather than preventing work outright, the payout is designed to offset lost earnings and cover additional costs – e.g. medical care, higher electricity bills, extra water – that extreme heat imposes on low-income households. The approach also incorporates heat early warning systems and practical equipment such as heat-resistant tools. One of the notable innovations of the Extreme Heat Income Insurance programme is that Blue Marble developed a forecast-based heat stress product that enables pay-outs to be made ex-ante rather than ex-post (Raithatha, 2023[55]). In Ahmedabad, the Mahila Housing Trust developed a comparable product, the Climate Risk Insurance solution (Raithatha, 2024[56]).

To be effective, parametric heat insurance products should incorporate several key design features, including premium subsidies, alignment with heat action plans and product bundling. Premium subsidies provided by a government, donor or sponsor are often critical, particularly during the product’s inception phase and for individuals with low levels of financial literacy (Garcia Ocampo and Lopez Moreira, 2024[57]). To improve understanding of payout triggers and encourage appropriate protective behaviour, product designers should also align insurance parameters with early warning systems and heat action plans. Finally, parametric products are often most effective when bundled with other social protection and/or financial instruments, as well as livelihood-support measures such as credit or maternity benefits. Bundling allows innovative products to leverage existing, trusted delivery channels, such as health and life insurance schemes or cash transfer programmes. This “portfolio effect” can work in both directions: traditional social protection gains a channel for reaching informal workers who might not otherwise access it while the parametric product benefits from the credibility and delivery infrastructure of established programmes.

Scaling parametric insurance for heat and other natural hazards may require updates to regulatory frameworks. In many countries, bringing these products to market would require amending insurance legislation to allow payouts based on pre-defined index triggers without the need to assess actual losses or damages (Garcia Ocampo and Lopez Moreira, 2024[57]). Official supervisory capacities may be limited, notably regarding index selection as well as the accuracy and reliability of payout triggers. At present, regulatory approaches vary substantially across jurisdictions: some countries have enacted specific laws and regulations, while others regulate these new products under broader insurance laws or through legal opinions and pilot projects (Garcia Ocampo and Lopez Moreira, 2024[57]).

The long-term financial sustainability of these new products remains largely uncertain. Most are recent and rely heavily on subsidies, either through lower premiums for insured individuals or through top-up mechanisms that increase payouts beyond what collected premiums would normally cover. While subsidies and support from governments, donors or sponsors can improve affordability and encourage uptake, reliance on such support raises questions about whether these programmes can remain financially viable over time.

Invest in skills, training and human capital

Over the long term, skills development can reduce informal workers’ structural vulnerability to environmental degradation and climate risks. Many informal workers with low levels of education are concentrated in sectors that depend heavily on natural resources or are highly exposed to droughts, floods and extreme heat. Training can help lower this sensitivity in two ways: by improving workers’ ability to adapt within their current livelihoods and by expanding their access to less climate-sensitive jobs over time.

For workers who are likely to remain in agriculture, skills policies should focus on practical, climate-adaptive practices. These include soil conservation, improved water management, efficient irrigation, agroecological techniques, crop diversification, and the use of drought-resistant or heat-tolerant varieties (see below section on “Establish comprehensive agricultural and rural development policies to build resilience”). Such training can help protect productivity and incomes without requiring a full sectoral move, which may be unrealistic for many rural informal workers in the short term. Extension services, farmer field schools and co-operatives can play key roles in delivering this support, especially when they combine technical advice with access to inputs and market information.

At the same time, skills development should support gradual livelihood diversification. Informal workers need access to training that improves their prospects in sectors with lower climate sensitivity and stronger productivity potential. This can include technical and vocational education and training in food processing, repair and maintenance services, care services, construction adapted to climate-resilient infrastructure, renewable energy, waste management and other green or local service activities (AUC/OECD, 2024[58]). Priority should be given to transferable skills such as literacy, numeracy, digital skills, financial literacy and business management. These skills can help informal workers adapt to changing labour demand, shift activities when needed and strengthen their ability to manage shocks.

Skills policies should also account for gender-specific barriers to training and labour-market transitions. Gender divides remain a major constraint to skills development, including in foundational skills, participation in technical and vocational education and training (TVET), and access to more productive activities (AUC/OECD, 2024[58]). To address the typical barriers faced by informal female workers – lower literacy levels, limited mobility, care responsibilities, and weaker access to information, finance and employer networks – training programmes should include targeted outreach to women, flexible schedules, childcare support and links to women’s organisations, co-operatives and local employers. They should also promote women’s participation in climate-resilient and higher-productivity activities, such as agri-food processing, repair services, renewable energy, waste management and local care services. However, training alone may not be sufficient where informal laws, customary practices or social norms restrict women’s mobility, control over income, or ability to work outside the household. Skills policies should therefore be paired with community engagement, support from local women’s organisations, and measures that address discriminatory norms and practices.

Overall, skills policies need to recognise the realities of informal work, which means that training programmes should be short, modular, locally delivered and compatible with workers’ schedules. They should also reduce participation costs through stipends, transport support, or links with cash-transfer and public employment programmes. For workers with low literacy, training should rely on practical demonstrations, peer learning and local languages rather than classroom-based instruction alone.

Finally, skills policies should be closely linked to formalisation and certification. Recognition of prior learning can help informal workers document skills acquired on the job and improve their access to better-paid work, contracts or public support programmes. Stronger linkages with the private sector as well as co-operatives and worker organisations can strengthen the credibility of training and help align skill supply with the labour-market demand (AUC/OECD, 2024[58]).

Strengthen occupational safety and health policies and early warning systems

Occupational safety and health

Legislation on occupational safety and health (OSH) is a cornerstone of effective policy responses to environmental degradation and climate‑related hazards. OSH policies define acceptable working conditions and set benchmarks for employers and workers, with the aim of preventing accident, injury and illness in the workplace by reducing exposure to potential hazards (ILO, 1981[59]). Most existing OSH frameworks, however, are not designed for climate-driven risks. Heat stress, for instance, is rarely covered by dedicated standards in low- and middle-income countries, despite its already significant productivity costs. In addition, even in countries in which policymakers have integrated specific climate risks into OSH and environmental legislation, risks and hazards may evolve and intensify, making it necessary to re‑evaluate guidelines or create new regulations (ILO, 2024[60]; ILO, 2026[61]). OSH frameworks also rarely account for the gender-specific nature of climate-related occupational risks, even though women and men often work in different occupations, face different workplace exposures, and experience different constraints related to working time, unpaid care responsibilities, workplace design and access to protective equipment (ILO, 2014[62]).

In practice, implementing robust OSH frameworks often involves one or a combination of the following: establishing specific temperature thresholds with mandatory rest, shade and hydration provisions for both indoor and outdoor work; setting protocols for suspending work during extreme weather events, including flood warnings; and designing requirements for employers to assess and communicate climate-related exposure risks. In this regard, ILO Conventions No. 155, No. 184 and No. 187 form the basis of strong and resilient OSH policies and management systems (ILO, 2006[63]; ILO, 2001[64]; ILO, 1981[59]). Although employer-based protections (including OSH guidelines) are weak or absent for most informal workers, promoting these legal requirements can have a powerful “lighthouse effect.” In effect, establishing a standard for the formal economy signals a socially acceptable standard to the informal economy.

To effectively reach informal workers, OSH policies and programmes must be adapted to their unique characteristics. Informal jobs are heterogeneous, often mobile and conducted in public spaces. In turn, informal workers usually face information barriers, limited bargaining power and low trust in public institutions. It is therefore crucial to simplify requirements, prioritise high‑risk hazards, and embed OSH messages within existing contact points and community structures, including primary health care services and social protection programmes. Evidence from Bangkok and Delhi, for instance, shows that information shared through community networks and market associations reaches informal workers more reliably than top-down government communication (Valdivia et al., 2025[65]; Sinha et al., 2026[66]).

Workers' organisations and collective action can play crucial roles in shaping effective OSH policies and in negotiating agreements and devising strategies that go beyond OSH policies while effectively protecting informal workers from the consequences of climate‑related hazards. Collective organisations, such as trade unions, cooperatives and informal worker associations, are essential to: negotiate better working conditions (including adapted work schedules and regular breaks); access shared services (such as cooling infrastructure); benefit from hydration and sanitation strategies; and organise mutual support systems (Valdivia et al., 2025[65]; Sinha et al., 2026[66]).

Early warning systems

In the context of extreme weather events, multi-hazard early warning systems are an essential complement to OSH standards. Location- and occupation-specific alerts – when tied to temperature thresholds with clear, actionable guidance on rest, hydration and task adjustment – can reduce health impacts and productivity losses. For heat, most occupational injuries occur before the official authority declares a heatwave (Flouris et al., 2024[20]). It is therefore crucial to trigger alerts and undertake preventive measures during hot periods and not only during heatwaves (when crisis response plans are usually activated). For droughts and floods, warnings that cover seasonal outlooks and their expected effects on agricultural calendars can help farmers adjust planting decisions, activate insurance cover, or access support in advance of shocks, shifting from reactive response to anticipatory risk management (FAO and WMO, 2025[67]).

The design of warning systems should account for the fact that many informal workers have limited digital skills or low literacy, which can prevent critical information from reaching them. Effective systems should utilise multiple channels, including community radio, SMS alerts in local languages, market associations and extension services. Critically, they should deliver messages that are simple, actionable and tailored to specific occupations (Valdivia et al., 2025[65]). Integrating OSH guidance (e.g. recommendations on rest, hydration and/or flood evacuation) into early warning messages could further increase the practical value of each alert.

Establish comprehensive agricultural and rural development policies to build resilience

Agricultural practices and climate risks are closely intertwined. Globally, the growth of the agri-food industry and the substantial productivity gains achieved during the last 50 years have largely come at the cost of the degradation of ecosystems (Davis et al., 2025[68]). Current intensive and unsustainable practices – e.g. poor soil management, absence of crop rotation, and misuse or overuse of chemical fertilisers and pesticides – have notably worsened the quality of soil, water and air. In combination, these factors have increased the vulnerability of the agricultural sector to extreme weather events and climate risks (OECD, 2026[69]). In return, the intensification of climate-related and extreme weather events now threatens agriculture productivity and farmers’ livelihoods.

Agricultural practices

New and different environment-friendly approaches are needed, typically entailing practices such as organic farming, agroecology and regenerative agriculture (OECD, 2026[69]). In general, these techniques require more labour inputs than in conventional practices, which would provide more on-farm jobs and wage work. They also tend to require different types of inputs (e.g. fertilisers and pesticides) and techniques (e.g. crop diversification and cross-crop cultivation) that are less harmful for the environment than those used in conventional farming. Socially, these approaches tend to produce a large range of positive outcomes, from better gender inclusion to the promotion of local and indigenous agricultural knowledge and the development of local agri-food value chains.

The potential economic risks associated with shifting towards more sustainable approaches should be clearly understood and carefully managed. Changing farming practices represents a substantial risk for farmers in terms of agricultural knowledge, yield and income. For example, many business models of organic farming depend on a price premium; if that premium declines over time or fails to materialise, the returns to organic farming may fall below those of conventional farming. For informal workers with limited resources and little financial cushion, this risk may simply be too high to absorb. Overall, two complementary elements are vital to success. First, farmers must clearly understand the trade-offs of different sustainable agriculture practices. In parallel, public policy should help reduce transition risks by supporting farmers during the adjustment period and providing incentives to adopt new practices.

Agri-food value chains

Transitioning successfully to sustainable agricultural practices would require boosting the market demand for sustainable food products. Reliable certification and labelling systems can help inform customers on the use of sustainable agricultural practices. Strengthening local supply chains through better linkages among actors (e.g. farmers, processors, distributors and retailers) could improve the quality of products, minimise losses and improve farmers’ direct access to markets. Internationally, trade agreements that prioritise sustainable agriculture could support such practices by shifting the priority from quantity to quality.

The development of local value chains rests on investments in agricultural infrastructure. At the macro level, sustainable agricultural livelihoods largely depend on access to a broad and diverse set of infrastructures, including roads, water supplies, irrigation canals and market linkages. Often, these investments directly enhance agricultural productivity and resilience while also strengthening the institutional and social fabric of rural communities. Conversely, underdeveloped infrastructure, inadequate distribution systems, low production technology and insufficient institutional capacity can constitute major threats to food security (JICA, 2026[70]).

Irrigation and water management

Access to reliable water management infrastructure is particularly critical. It constitutes a powerful structural determinant of agricultural resilience, notably in locations prone to heat stress and droughts. The Intergovernmental Panel on Climate Change (IPCC) identifies the shift from rainfed to irrigated systems as one of the most widely adopted adaptation responses in agriculture (IPCC, 2022[35]). Yet in much of sub-Saharan Africa and parts of South and Southeast Asia, the majority of smallholder farmers remain entirely dependent on rainfall, making their livelihoods inherently fragile in the face of climate variability (FAO, 2023[71]).

Although conventional irrigation is often unaffordable for most informal smallholders, innovative and sustainable solutions exist. For example, the community-based smallholder irrigation (COBSI) approach, developed by the Japan International Cooperation Agency (JICA), represents a fundamental departure from the assumption that irrigation requires large-scale, externally financed, and technically complex infrastructure (JICA, 2017[72]; JICA, 2024[73]). First piloted in Malawi in 2002 and subsequently in Zambia in 2009, COBSI enables farmers to construct functional diversion weirs using locally available materials such as wood, stone, clay and grass – at near-zero cost. These simple weirs serve as an entry point to irrigated agriculture, allowing communities to gain experience with water management and irrigated cropping. The model then operates gradually: where conditions and resources permit, these simple structures are upgraded to more permanent weirs with more complex designs. In Zambia, by the end of the initial COBSI project period (2009-11), 568 irrigation sites had been developed or rehabilitated using this method, with most households shifting from rainfed, slash-and-burn cultivation to irrigated agriculture within a few years. As a comprehensive approach COBSI also strengthens community group dynamics by bringing people together to construct, manage and maintain shared water infrastructure. This builds social capital, itself a critical component of climate resilience. The approach has also been linked to the Smallholder Horticulture Empowerment and Promotion (SHEP) methodology, transitioning farmers from subsistence to market-oriented horticulture, with nutritional awareness components additionally integrated.

Drought-resistant and heat-tolerant crops

The increased use of drought-resistant and heat-tolerant seeds is a key practice that could help reduce agricultural sensitivity to climate shocks. Drought-resistant and heat-tolerant crop varieties allow farmers to maintain meaningful production under conditions that would otherwise cause near-total crop failure. Across the world, several initiatives are underway to develop and disseminate climate-resilient crop varieties. In Africa, for instance, the Accelerating Impacts of CGIAR Climate Research for Africa (AICCRA) project has focused on validating and disseminating climate-smart technologies across Ethiopia, Ghana, Kenya, Mali, Senegal and Zambia (World Bank, 2024[74]). Likewise, the FAO's Global Framework on Water Scarcity in Agriculture (WASAG) programme has launched trials of indigenous drought-resilient crops – e.g. sorghum and cowpeas – targeting women and youth farmers, recognising that traditional crop knowledge is itself a resilience asset (FAO, 2026[75]).

However, the gap between the development of drought-resistant varieties in research institutions and their practical availability to smallholder farmers remains substantial. In most low- and middle-income countries, improved varieties rarely reach informal smallholder farmers due to weak national seed regulatory systems, inadequate seed multiplication, limited distribution infrastructure, high commercial seed prices and limited agricultural extension capacity (Sutton, Lotsch and Prasann, 2024[76]). The IPCC also underlines that cultural and economic factors – including farmers’ risk aversion, lack of extension support and weak input supply chains – could constitute crucial barriers to the adoption of such varieties (Sutton, Lotsch and Prasann, 2024[76]). Collective organisations, such as farmers’ cooperatives, can play a key role by aggregating demand for seeds and irrigation at a scale individual smallholders cannot reach alone.

Rural development and rural-urban linkages

Beyond agriculture, building the resilience of informal workers requires broader territorial approaches to rural development. While climate-smart agriculture, irrigation and resilient crops can strengthen farm livelihoods, they are unlikely to be sufficient on their own. Many rural households already depend on a combination of farming and non-farm activities, including trade, transport, agro-processing, construction, tourism and personal services. Supporting the development of these activities can reduce dependence on climate-sensitive agriculture, diversify income sources and create new opportunities for productive employment. In this regard, rural development policies should not focus solely on increasing agricultural productivity, but also on fostering local entrepreneurship, skills development, access to finance and the growth of rural non-farm economies (OECD, 2019[77]).

Such territorial approaches acknowledge that rural and urban areas are deeply interconnected. Intermediary cities, in particular, are at the heart of food systems, acting as interfaces at which rural areas source their agricultural inputs and farmers access local markets (OECD/UN-Habitat, 2022[7]). These areas also play a particularly important role as hubs for food processing, logistics, education, health care and business services (OECD, 2019[77]). Strengthening rural-urban linkages through investments in transport, digital connectivity, market infrastructure and local value chains can help connect rural producers and workers to larger markets, services and employment opportunities, thereby transforming agricultural growth into broader local economic development.

These linkages highlight that resilience cannot be built in rural areas alone. The territorial perspective underscores the need to integrate climate adaptation, economic development and spatial planning across rural and urban areas rather than treating them as separate policy domains. As workers, goods and services move between territories, the quality of urban infrastructure increasingly shapes the ability of households, including rural ones, to cope with environmental risks. Investments in resilient transport systems, housing, public services and urban infrastructure are therefore a critical complement to rural development policies, particularly in intermediary and rapidly growing cities that connect rural and urban economies (OECD/UN-Habitat, 2025[78]; OECD/UN-Habitat, 2022[7]).

Invest in resilient and quality urban infrastructure

The quality of infrastructure and buildings in which informal workers live and work is a crucial dimension of their vulnerability to extreme weather events. Urban informal workers often face a double exposure, both at work – whether they operate in a fixed place of work or in the public space – and at home. Workplace‑focused measures are therefore essential but are not sufficient on their own because heat exposure and its consequences extend well beyond the job site. For example, dwellings built with heat-retaining materials, minimal insulation and inadequate ventilation can reach intolerable temperatures during heatwaves, often exceeding outdoor ambient temperatures (Valdivia et al., 2025[65]; Sinha et al., 2026[66]). In dense, low-income urban neighbourhoods, the urban heat island effect can substantially raise temperatures. Meanwhile, for workers in flood-prone areas, poorly sited and structurally inadequate housing means that flooding simultaneously destroys assets and disrupts livelihoods. Informal workers often face compounding exposure as they tend to be concentrated in the most hazardous parts of cities – e.g. floodplains, steep slopes and poorly drained lowlands – without access to basic infrastructure such as water drainage or electricity (IPCC, 2022[35]).

A prerequisite is to ensure dwellings and places of work have access to the basic infrastructure on which urban livelihoods depend: piped water, drainage, paved roads, electricity and emergency access (Satterthwaite et al., 2018[79]). In cases of floods, for example, lack of access to these services can contribute to the spread of disease in the aftermath and delay recovery. Likewise, lack of access to electricity during heatwaves can prevent households and businesses from installing or operating cooling systems.

Informal settlement upgrading – i.e. systematic improvement of housing stock, tenure security, services and public spaces – is one of the most effective vehicles for climate adaptation available to city governments (OECD/UN-Habitat, 2022[7]). In most places, the urban housing deficit is not in quantity but in quality. In 2018, the Inter-American Development Bank estimated that 94% of the total urban housing deficit in Latin America and the Caribbean was associated with inadequate living conditions in existing informal settlements while 90% of housing solutions involved the construction and delivery of new units instead of retrofitting (Adler et al., 2018[80]).

Upgrading such informal settlements and providing basic services requires the engagement of multiple layers of governance. Since the adverse impacts of extreme weather events are observed at different scales, urban adaptation that is both pro-poor and pro-informal workers is needed across all stakeholders, from individual and household level to neighbourhood, community and city level. To remain feasible and realistic, these adaptation measures should integrate coping measures, incremental adjustments and transformational changes. This co-ordination among diverse actors and across spatial layers is crucial to ensure infrastructure investment incorporates climate risk from the outset, including through flood risk mapping, heat island analysis and community participation at the planning stage. Failing to do so can inadvertently lock in new vulnerabilities. For example, road investments can accelerate stormwater runoff while drainage improvements in one location can worsen flooding downstream (Satterthwaite et al., 2018[79]).

In practice, improving housing and workplace buildings can draw upon diverse techniques and methods. In the case of heat, demand for cooling will grow sharply as temperatures rise, but access to effective cooling remains deeply unequal, mainly because of the cost of air conditioners or coolers. Electric fans, the most common fallback in low-income households, are ineffective above 35°C – when cooling is most needed – since they circulate hot air without reducing indoor temperature. Policies should therefore focus not only on providing affordable electricity and access to electricity-powered cooling mechanisms, but should also prioritise passive cooling approaches that do not depend on access to electricity or households’ welfare (ADB, 2022[81]). These passive strategies include thermal insulation, natural ventilation design, shading, cool roofs and reflective surfaces. Importantly, most of these potential gains are locked in at the design stage. The role of building codes and minimum energy performance standards for new construction is crucial to ensure passive solutions are progressively implemented as new dwellings and buildings are built.

In informal settlements, cool roof solutions are particularly interesting as they are relatively inexpensive, fast to deploy and do not require major structural modifications (ADB, 2022[81]). Evidence from a range of major cities in the Global South shows that cool roof retrofits could reduce up to 91% of annual heat exposure (Nutkiewicz et al., 2022[82]). For existing housing stock, publicly subsidised retrofit programmes that prioritise ventilation improvements and reflective surfaces offer a practical route to reduce chronic heat exposure for the most vulnerable residents.

At the community or city level, adaptation to heat can be achieved through various levers, including decongestion of public space, use of cool or permeable pavement, provision of shaded and resting spaces (e.g. benches) in public urban spaces, and the installation of vegetation and water areas where possible (ADB, 2022[81]). Cooling centres – e.g. community spaces with fans, drinking water and basic facilities – can also provide critical relief during heatwaves for outdoor workers and local residents in overheated dwellings who have no private alternative (Engel, Anzilotti and Palmieri, 2025[83]). Green infrastructure can also deliver a wide range of co-benefits that are particularly relevant to informal workers: it absorbs stormwater during flood events, reduces air pollution, improves mental health outcomes, and provides physical spaces for outdoor work and rest during extreme heat (Hunter et al., 2023[84]).

Strengthening urban resilience to climate risks also requires rethinking transportation systems and the way public space is organised. In 2025, in the context of intermediary cities of Kenya and Mozambique, the OECD and UN Habitat identified four priority areas of action: i) rethinking how road space and public space are allocated; ii) structuring urban development and infrastructure investment around accessibility rather than distance; iii) expanding and improving sustainable mobility options; and iv) reinforcing governance and institutional alignment (OECD/UN-Habitat, 2025[78]). These high-leverage goals are strategic system shifts that can unlock broader transformation. While each city and urban area may have its own constraints and specific objectives, these areas of action tend to converge towards similar priorities: reclaiming road space for people, linking housing and transport planning, and improving public and shared mobility.

Annex 5.A. Environmental and climate risks: Concepts and application

Copy link to Annex 5.A. Environmental and climate risks: Concepts and application

Conceptual approach

Copy link to Conceptual approach

The concept of environmental and climate risk has evolved over time. The field of environment uses a wide array of terms – risk, vulnerability, sensitivity, resilience, adaptation and adaptive capacity – for which relationships are not always clear and meanings may vary when used in different contexts or by different authors. In its sixth and latest assessment (2022), the Intergovernmental Panel on Climate Change (IPCC) revisited the definition of risk as a function of hazard, exposure and vulnerability (IPCC, 2022[35]). This approach establishes vulnerability as both a component of risk and a stand-alone crucial concept. Defined as the propensity or predisposition to be adversely affected, vulnerability itself encompasses a variety of concepts, including sensitivity or susceptibility to harm and lack of capacity to cope and adapt. This definition remains very similar to the description of vulnerability in the IPCC’s third assessment (2001), which describes vulnerability as a function of the character, magnitude, and rate of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity (Brooks, 2003[85]).

The chapter examines specifically at environment-related risks through the lens of potential impacts that environmental degradation and climate shocks can have on workers’ incomes. Although risks and vulnerabilities can take many forms and span multiple dimensions of individuals’ lives – including non-economic ones (Canpolat et al., 2025[2]) – the analysis focuses primarily on the risks faced by the informal economy from environmental degradation and hazards. For this reason, the chapter uses income vulnerability as its main entry point and examines, by extension, individuals’ ability to economically sustain their livelihoods and those of their households in the event of adverse shocks.

The chapter adopts a definition of risk broadly in line with the IPCC’s definition, relying on three key dimensions: exposure, sensitivity and adaptation (Annex Figure 5.A.1). This is consistent with the frameworks used in the research on environmental risks. For instance, the University of Notre Dame’s ND-GAIN index assesses the vulnerability of a country for a large range of indicators and areas across three cross-cutting components: the exposure of the sector to climate-related or climate-exacerbated hazards; the sensitivity of that sector to the impacts of the hazard; and the adaptive capacity of the sector to cope or adapt to these impacts (Chen et al., 2024[86]). Likewise, the OECD Development Centre (2024[87]) adopted a definition of risk that relies on three key factors: exposure (the presence, occurrence and magnitude of environmental degradation); sensitivity (the harm caused by a given level of environmental degradation); and adaptation capacity (the ability of individuals or systems to cope with the impacts of environmental degradation).

Annex Figure 5.A.1. Risk as the outcome of three key elements – exposure, sensitivity and adaptive capacity
Copy link to Annex Figure 5.A.1. Risk as the outcome of three key elements – exposure, sensitivity and adaptive capacity

Exposure is defined as the presence, occurrence and magnitude of environmental degradation or hazards. How exposure is measured depends on the type of event considered. To assess exposure to heat stress, the chapter uses the wet bulb globe temperature (WBGT) – a composite indicator capturing the combined effects of air temperature, humidity, wind speed and solar radiation. For droughts and floods, exposure is measured using the normalised total yearly duration of events, which reflects both their frequency and persistence. In addition, the level of exposure to a given hazard can vary substantially across locations, depending on the type of event being analysed. Typically, regions or countries that are severely exposed to droughts are not necessarily exposed to the same extent to floods, extreme temperatures, storms, sea‑level rise or other hazards.

Sensitivity is defined as the potential harm caused to individuals’ livelihoods by a given level of environmental degradation. From a labour perspective, sensitivity refers to whether jobs and labour income could be potentially impacted by environmental degradation. This dimension is strongly linked to the concept of dependency towards and reliance upon environmental resources. Importantly, the level of sensitivity can vary depending on the type of environmental degradation or climatic shock to which a given individual, household or community is exposed. Typically, based on the economic sector of occupation, a worker may be highly sensitive to a certain type of environmental degradation, such as heat stress, but less sensitive to another type such as floods.

Adaptation is the ability of individuals or systems to cope with the impacts of environmental degradation. This approach is broadly in line with the IPCC, which defines adaptation in human systems as the process of adjustment to actual or expected climate events and associated effects in order to moderate harm or take advantage of beneficial opportunities (IPCC, 2022[35]). In this chapter, adaptation capacity is assessed from a labour-market perspective and is primarily understood as operating in the informal economy or not. The use of the informality status is a proxy for access – or lack thereof – to social protection, which is a crucial factor for individuals to avoid harm and/or recover from it. At the same time, workers of the informal economy are more likely to experience a wide range of deprivations in other key socio-economic dimensions such as poor living conditions, lower educational attainment, reduced access to public services and infrastructure, etc. These limitations further hamper informal workers’ ability to cope with environmental degradation.

Methodology to measure risk from weather events

Copy link to Methodology to measure risk from weather events

To measure risk from weather events, the chapter adopts a definition based on the three core dimensions previously outlined and tailors them to the available data.

Exposure is measured using data from the Emergency Events Database (EM-DAT) for the period 2000‑23. Produced by the Centre for Research on the Epidemiology of Disasters (UCLouvain / CRED, 2025[17]), EM-DAT documents the occurrence and impacts of more than 27 000 mass disasters worldwide since 1990, including droughts and floods. The database draws on multiple sources, such as UN agencies, non‑governmental organisations, reinsurance companies, research institutes and press agencies.

Based on harmonisation work undertaken by Teber et al. (2025[88]), EM-DAT data for droughts and floods are geocoded using the spatial standard of Global Administrative Unit Layers (GAUL) developed by the Food and Agriculture Organization (FAO). These hazards are linked to GAUL level 1 areas, which correspond to the first national administrative level (e.g. provinces, regions). For each KIIbIH country, data on floods and droughts are aggregated at the sub-national level. Exposure is measured by normalising the total yearly duration of events over the reference period and classifying the resulting indicator into three exposure levels: low (0-1 month/year), medium (1-2 months/year) and high (more than 2 months/year).

In parallel, population data from the KIIbIH are geocoded and harmonised to match GAUL level 1 classification (Annex Table 5.A.1). Weather event data are then linked to the KIIbIH population data using the same GAUL framework.

ISO-3	Country	Year	Region	Number of GAUL area covered	Share of population geo-located (%)	Share of labour force geo-located (%)
ALB	Albania	2012	Europe	12	100	100
ARG	Argentina	2023	Americas	24	100	100
ARM	Armenia	2023	Asia	11	100	100
BEN	Benin	2021	Africa	12	100	100
BFA	Burkina Faso	2021	Africa	13	100	100
BGR	Bulgaria	2023	Europe	No data geo-located
BHS	Bahamas	2013	Americas	No data geo-located
BOL	Bolivia	2022	Americas	9	100	100
BRA	Brazil	2023	Americas	27	100	100
BRB	Barbados	2016	Americas	11	100	100
CHL	Chile	2022	Americas	16	100	100
CHN	China	2020	Asia	21	100	100
CMR	Cameroon	2007	Africa	10	100	100
COL	Colombia	2023	Americas	32	98	100
CRI	Costa Rica	2023	Americas	No data geo-located
CYP	Cyprus	2023	Asia	No data geo-located
DOM	Dominican Republic	2018	Americas	10	100	100
ETH	Ethiopia	2018	Africa	11	100	100
GHA	Ghana	2013	Africa	10	100	100
GMB	Gambia	2015	Africa	8	100	100
GTM	Guatemala	2022	Americas	22	100	100
HND	Honduras	2019	Americas	16	24	44
HRV	Croatia	2023	Europe	No data geo-located
IDN	Indonesia	2014	Asia	19	94	95
IND	India	2012	Asia	32	100	100
JAM	Jamaica	2019	Americas	14	100	100
KEN	Kenya	2015	Africa	47	100	100
KHM	Cambodia	2019	Asia	24	100	100
LAO	Lao PDR	2012	Asia	17	100	100
LBR	Liberia	2016	Africa	15	100	100
LVA	Latvia	2023	Europe	No data geo-located
MDG	Madagascar	2012	Africa	6	100	100
MDV	Maldives	2019	Asia	18	100	100
MEX	Mexico	2022	Americas	32	93	100
MLI	Mali	2021	Africa	9	100	100
MMR	Myanmar	2015	Asia	3	100	100
MNG	Mongolia	2021	Asia	21	100	100
MWI	Malawi	2019	Africa	3	100	100
NAM	Namibia	2015	Africa	14	100	100
NER	Niger	2018	Africa	8	100	100
NGA	Nigeria	2015	Africa	35	100	100
NIC	Nicaragua	2014	Americas	No data geo-located
PER	Peru	2023	Americas	22	100	100
PRY	Paraguay	2024	Americas	16	100	100
ROU	Romania	2023	Europe	No data geo-located
RWA	Rwanda	2016	Africa	5	100	100
SEN	Senegal	2021	Africa	14	100	100
SLE	Sierra Leone	2018	Africa	4	100	100
SLV	El Salvador	2023	Americas	14	100	100
SUR	Suriname	2022	Americas	10	100	100
TGO	Togo	2021	Africa	5	100	100
THA	Thailand	2017	Asia	77	100	100
TZA	Tanzania	2019	Africa	31	100	100
UGA	Uganda	2019	Africa	4	100	100
URY	Uruguay	2023	Americas	19	100	100
VNM	Viet Nam	2016	Asia	63	100	100
ZAF	South Africa	2016	Africa	7	100	100
ZMB	Zambia	2015	Africa	10	100	100

Adaptive capacity is measured using the formality or informality status of workers aged 15 years and above.

Sensitivity is measured through the sectoral link of jobs to droughts and floods, defined as the extent to which economic activities are likely to be negatively affected by these hazards. Jobs are classified into three categories – direct, indirect, and no or weak sensitivity – based on codes from the fourth revision of the International Standard Industrial Classification of All Economic Activities (ISIC‑04). ISIC is a hierarchical classification system that organises economic activities into four levels, ranging from broad categories at level 1 to detailed activities at level 4. The sensitivity classification is developed separately for droughts and floods and is tailored to each hazard.

For droughts, all activities within “agriculture, forestry and fishing” (ISIC‑04 Sector A) are classified as having direct sensitivity (Annex Figure 5.A.2). Activities with indirect sensitivity include selected manufacturing and service sectors. In manufacturing (Sector C), five divisions and all of their level‑3 and level‑4 subcategories are classified as indirectly sensitive: manufacture of food products (C10); beverages (C11); tobacco products (C12); wood and wood products excluding furniture (C16); and paper and paper products (C17). In services, all food‑related activities within accommodation and food services (I56, Sector I), as well as several level‑3 and level‑4 activities from sectors G, N and R, are also classified as indirectly sensitive to droughts (Annex Figure 5.A.2). The rest of the economy is classified as only weakly sensitive to droughts.

Annex Figure 5.A.2. Sectoral linkages to droughts
Copy link to Annex Figure 5.A.2. Sectoral linkages to droughts

For floods, all activities within “agriculture, forestry and fishing” (Sector A) and “mining and quarrying” (Sector B) are classified as having direct sensitivity (Annex Figure 5.A.3). In addition, two level‑3 activity groups related to transportation systems are identified as directly sensitive to floods. Activities with indirect sensitivity include selected manufacturing and service sectors. All activities within “construction” (Sector F), “accommodation and food service activities” (Sector I) and “real estate activities” (Sector L) are classified as indirectly sensitive to floods. The same five manufacturing divisions (Sector C) identified as indirectly sensitive to droughts are classified as indirectly sensitive to floods, along with all water‑ and sewage‑related activities (E36, E37 and E38, Sector E). Several level‑2, level‑3 and level‑4 activities from Sectors G, N, R and T are likewise classified as indirectly sensitive to floods (Annex Figure 5.A.3). All remaining economic activities are classified as having only weak sensitivity to floods.

Annex Figure 5.A.3. Sectoral linkages to floods
Copy link to Annex Figure 5.A.3. Sectoral linkages to floods

Annex 5.B. Methodology to estimate productivity losses associated with heat

Copy link to Annex 5.B. Methodology to estimate productivity losses associated with heat

This annex presents the methodology used to estimate labour productivity losses associated with heat stress. Productivity losses are defined as reductions in work capacity resulting from exposure to high levels of heat during working hours. The approach builds on the framework developed by Kjellstrom et al. (2017[22]) and the ILO (2019[23]), while introducing important methodological adaptations designed to better capture heterogeneity in labour‑market exposure to heat.

Conceptual approach and comparison with existing methodologies

Copy link to Conceptual approach and comparison with existing methodologies

Relative to the ILO methodology, the approach adopted in this chapter involves a trade‑off between spatial precision in climate exposure and precision in population and labour‑market characteristics. On the climate side, heat exposure is aggregated from fine‑resolution climate grid cells to sub‑national administrative areas using a “contain” approach, whereby only grid cells fully contained within a given Global Administrative Unit Layers (GAUL) area are retained and median values across these cells are used to characterise local heat exposure. This aggregation reduces within‑area climate variability compared with a purely grid‑based analysis as implemented in Kjellstrom et al. (2017[22]) and ILO (2019[23]).

On the population side, the methodology substantially improves the measurement of exposure and impacts across workers. Instead of relying on national employment‑to‑population ratios applied uniformly across space, as in the ILO approach, the methodology applied here links productivity losses to geolocated individual‑level data from the KIIbIH household survey. This allows productivity impacts of heat stress to be weighted by the actual spatial and sectoral distribution of workers and to be analysed across specific sub‑groups of the labour force, including formal and informal workers, as well as across a broader range of social, demographic and economic characteristics.

The methodology therefore prioritises a more accurate representation of who is exposed to heat and how exposure translates into productivity losses across different segments of the labour force, while maintaining consistency with the core climate-productivity relationships established in the literature.

Methodology to compute productivity losses

Copy link to Methodology to compute productivity losses

The methodology consists of four main steps: i) measurement of heat exposure using WBGT; ii) translation of heat exposure into productivity losses at the climate grid‑cell level; iii) spatial aggregation of productivity losses to administrative areas; and iv) linkage of climate‑based losses with labour‑market microdata (Annex Figure 5.B.1).

Annex Figure 5.B.1. Process to compute productivity losses based on climate and KIIbIH data
Copy link to Annex Figure 5.B.1. Process to compute productivity losses based on climate and KIIbIH data

Measurement of heat exposure

Raw climate data are extracted from the Copernicus Climate Data Store (CDS) for the period 2010-69, using historical data (2010-25) from the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) dataset, and projected data (2026-2069) from the Coupled Model Intercomparison Project Phase 6 (CMIP6) dataset (Copernicus Climate Change Service CDS, 2026[32]; Copernicus Climate Change Service CDS, 2026[33]). CMIP6 data are extracted from the MPI-ESM1-2-LR model of the Max Planck Institute for Meteorology for the intermediate‑emissions scenario (SSP2‑4.5). Annex Table 5.B.1 provides information on the historical and projected raw data extracted from Copernicus CDS.

Dataset	Code	Name	Unit	Description
ERA5
	t2m	2m temperature	Kelvin (K)	The temperature of air at 2 meters above the surface of land, sea or inland waters is calculated by interpolating between the lowest model level and the Earth's surface, taking account of atmospheric conditions.
	d2m	2m dewpoint temperature	Kelvin (K)	The temperature to which the air, at 2 meters above the surface of the Earth, would have to be cooled for saturation to occur is a measure of the humidity of the air. Combined with temperature, it can be used to calculate the relative humidity. 2m dewpoint temperature is calculated by interpolating between the lowest model level and the Earth's surface, taking account of atmospheric conditions.
	u10	10m u-component of wind	Meter per second (m/s)	The eastward component of the 10-meter wind corresponds to the horizontal speed of air moving towards the east, at a height of 10 meters above the surface of the Earth.
	v10	10m v-component of wind	Meter per second (m/s)	The northward component of the 10-meter wind corresponds to the horizontal speed of air moving towards the north, at a height of 10 meters above the surface of the Earth.
CMIP6
	tas	Near-surface air temperature	Kelvin (K)	The temperature of air at 2 meters above the surface of land, sea or inland waters is calculated by interpolating between the lowest model level and the Earth's surface, taking account of atmospheric conditions.
	tasmax	Daily maximum near-surface air temperature	Kelvin (K)	Daily maximum temperature of air at 2 meters above the surface of land, sea or inland waters.
	huss	Near-surface specific humidity	dimensionless	Amount of moisture in the air near the surface, divided by the amount of air plus moisture at that location.
	psl	Sea level pressure	Pascal (Pa)	Pressure of the atmosphere at the surface of the Earth, adjusted to the height of sea level.
	sfcWind	Near-surface wind speed	Meter per second (m/s)	Magnitude of the two-dimensional horizontal air velocity (u- and v- component) near the surface.

Source: (Copernicus Climate Change Service CDS, 2026[32]) and (Copernicus Climate Change Service CDS, 2026[33]).

To ensure that the output of the climate model aligns with historical data, projected data are bias-corrected using quantile delta mapping method. First, ERA5 and CMIP6 historical simulations are re-gridded to a common spatial resolution. Monthly histograms are constructed for each grid cell and variable, using fixed bin widths. These histograms are aggregated over the calibration period and converted into empirical quantiles. Second, quantile delta mapping method is applied to projected CMIP6 data. For each grid cell, projected values are mapped from the CMIP6 historical distribution to the ERA5 reference distribution, while preserving projected changes in quantiles. Relative humidity is bias‑corrected indirectly by correcting relative humidity distributions and converting them back to specific humidity. This approach ensures physical consistency between temperature and humidity.

Heat exposure is measured using the wet bulb globe temperature (WBGT), a composite indicator that captures the combined effects of air temperature, humidity, wind speed and solar radiation on human heat stress. WBGT is widely used in occupational health research and forms the basis of the productivity-heat response functions adopted in this analysis.

The standard formula to compute the outdoor WBGT is:

$W B G T = 0.1 \times T_{d b} + 0.7 \times T_{w b} + 0.2 \times T_{g}$

Where:

$T_{d b}$ = dry bulb temperature
$T_{w b}$ = wet bulb temperature
$T_{g}$ = globe temperature (radiant heat)

The dry bulb temperature ( $T_{d b}$ ) corresponds to the ambient air temperature.

The wet bulb temperature ( $T_{w b}$ ) reflects the lowest temperature that can be reached by evaporative cooling and is influenced by both air temperature and humidity. It can be estimated using the Stull (2011[89]) approximation, especially in large-scale modelling:

$T_{w b} \approx T_{d b} \times \tan^{- 1} (0.151977 \times \sqrt{R H + 8.313659}) + \tan^{- 1} (T_{d b} + R H) - \tan^{- 1} (R H - 1.676331) + 0.00391838 \times {R H}^{1.5} \times \tan^{- 1} (0.023101 \times R H) - 4.686035$

Where:

$T_{d b}$ = dry bulb temperature
$R H$ = relative humidity

Based on Parish and Putnam (1977[90]), relative humidity ( $R H$ ) is itself calculated as:

$R H = \frac{{e (T}_{d p})}{{e (T}_{d b})} \times 100$

Where:

${e (T}_{d p})$ = actual vapour pressure, i.e. vapor pressure at dewpoint temperature ( $T_{d p}$ )
${e (T}_{d b})$ = saturation vapour pressure, i.e. vapor pressure at dry bulb temperature ( $T_{d b}$ )
Both actual and saturation vapour pressure can be estimated for any temperature level $T$ using the Clausius-Clapeyron formula (Bolton, 1980[91]):

$e (T) = 6.112 \times e x p (\frac{17.67 \times T}{T + 243.5})$

For CMIP6 data (for which dewpoint temperature is not available), the computation of ${e (T}_{d p})$ can be further transformed to instead rely only on near-surface specific humidity ( $q$ ) and sea level pressure ( $p$ ):

$e (T_{d p}) = \frac{q \times \frac{p}{100}}{0.622 + 0378 \times q}$

The globe temperature ( $T_{g}$ ) can be estimated using wind speed:

$T_{g} = T_{d b} + 1.1 + 0.02 \times W S$

Where:

$T_{d b}$ = dry bulb temperature
$W S$ = wind speed

For CMIP6 data, near-surface wind speed is already available. For ERA5 data, it can be computed from the u- and v-component of wind with:

$W S = \sqrt{u^{2} + v^{2}}$

Based on these formulas, WBGT is calculated for a daily theoretical 12‑hour working period, following the standard assumptions made in Kjellstrom et al. (2017[22]) and ILO (2019[23]). For historical climate data (ERA5), hourly WBGT values are computed for hours spanning 7:00 to 19:00 in local time. For projected climate data (CMIP6), hourly values of the raw data are not available. Instead, the method estimates the typical hourly distribution of heat levels in each grid cell from 7:00 to 19:00 by applying the “4+4+4” method. This method divides the 12-hour daily working period into three 4-hour sub-periods: during 4 hours, WBGTs are assumed to be close to the daily maximum WBGT; during 4 hours, WBGTs are assumed to be close to the daily mean WBGT (early morning and early evening); during the remaining 4 hours WBGTs are assumed to lie halfway between the daily mean WBGT and the daily maximum WBGT. The resulting three temperature datapoints computed are assumed to represent the 12-hour distribution of temperatures.

Translation of heat exposure into productivity losses

Hourly WBGT values are translated into labour productivity losses using workload‑specific response curves derived from epidemiological and occupational health studies (Kjellstrom et al., 2017[22]). These response functions describe the reduction in work capacity – or productivity loss – associated with increasing heat stress and differ according to the physical intensity of work.

Three categories of workers are distinguished based on metabolic workload: light work (200 watts [W]), moderate work (30 W) and heavy work (400 W) (Annex Figure 5.B.2).

Annex Figure 5.B.2. Estimated productivity losses at different level of physical intensity: 200 W, 300 W and 400 W
Copy link to Annex Figure 5.B.2. Estimated productivity losses at different level of physical intensity: 200 W, 300 W and 400 W

For each climate grid cell and each day, hourly productivity losses are computed separately for the three workload categories by applying the corresponding response curve to the observed WBGT values. These hourly losses are aggregated to daily and annual productivity losses at the grid‑cell level.

To capture long-term climate trends, rather than short‑term variability, annual productivity losses are averaged over three 30‑year periods. A 30-year period aligns with what the climate science community regards as the minimum time period over which a long-term climate trend, as opposed to weather or extreme events, can be demonstrated (ILO, 2019[23]). Throughout the chapter, the period 1980-2009 is referred to as 1995, the period 2010-39 as 2025, and the period 2040-69 as 2055.

Spatial aggregation of climate data

Climate grid cells are spatially matched to sub‑national administrative areas using the GAUL spatial developed by the FAO. Productivity losses are linked to GAUL level 1 areas, corresponding to the first national administrative level (e.g. provinces, regions).

To reconcile the spatial resolution of climate grid cells with administrative boundaries, a “contain” approach is applied. Only grid cells fully contained within a given GAUL area are retained. For each GAUL area and each period (1995, 2025 and 2055), GAUL-level productivity losses are obtained by computing the average, median, minimum and maximum values across all contained grid cells. The median value is used as the baseline measure in the analysis, as it provides a robust representation of typical exposure while limiting the influence of local extremes.

Linkage with labour‑market microdata

The final step links climate‑based productivity losses with individual‑level labour‑market data from KIIbIH household surveys. In total, 50 surveys in the KIIbIH database are spatially harmonised by assigning individuals and households to GAUL level 1 areas based on their place of residence. Annex Table 5.A.1 in Annex 5.A reports the results of this spatial harmonisation process and the share of the population successfully geolocated.

Productivity losses computed at the GAUL level for 1995, 2025 and 2055 are matched to individuals based on their location of residence and sector of activity. Based on codes from the fourth revision of the International Standard Industrial Classification of All Economic Activities (ISIC‑04), jobs are mapped to the three workload categories defined above. Activities in agriculture, forestry and fishing (Sector A) and construction (Sector F) are assumed to involve heavy physical work. Activities in mining and quarrying (Sector B), manufacturing (Sector C), electricity, gas and air conditioning supply (Sector D), and water supply and waste management (Sector E) are assumed to involve moderate physical work. All remaining activities are assumed to involve light physical work (Annex Figure 5.B.3).

This linkage allows productivity losses to be weighted by the actual spatial and sectoral distribution of workers and enables analysis of heat‑related productivity impacts across specific labour‑market sub‑groups, including formal and informal workers.

Annex Figure 5.B.3. Level of physical intensity of ISIC-04 sectors
Copy link to Annex Figure 5.B.3. Level of physical intensity of ISIC-04 sectors

References

[81] ADB (2022), Beating the heat: Investing in pro-poor solutions for urban resilience, Asian Development Bank, Manila, Philippines, https://www.adb.org/sites/default/files/publication/815696/beating-heat-pro-poor-solutions-urban-resilience.pdf.

[80] Adler, V. et al. (2018), Housing, what’s next? From thinking the unit to building the city, Inter-American Development Bank, Washington, DC, https://publications.iadb.org/es/publications/english/viewer/Housing_Whats_Next_From_Thinking_The_Unit_to_Building_The_City.pdf.

[44] Adrienne Arsht-Rockefeller Foundation Resilience Center (2023), The scorching divide: How extreme heat inflames gender inequalities in health and income, Climate Resilience Center, Atlantic Council, Washington, DC, https://onebillionresilient.org/extreme-heat-inflames-gender-inequalities.

[14] Akarsu, O. et al. (2026), “The Labor Market in Times of Natural Disaster: Unequal Impacts of Türkiye’s 2023 Earthquakes”, Working Paper, No. 26/07, Central Bank of the Republic of Türkiye, Istanbul, Türkiye, https://www.tcmb.gov.tr/wps/wcm/connect/EN/TCMB+EN/Main+Menu/Publications/Research/Working+Paperss/2026/26-07.

[58] AUC/OECD (2024), Africa’s Development Dynamics 2024: Skills, Jobs and Productivity, AUC, Addis Ababa/OECD Publishing, Paris, https://doi.org/10.1787/df06c7a4-en.

[91] Bolton, D. (1980), “The Computation of Equivalent Potential Temperature”, Monthly Weather Review, Vol. 108/7, pp. 1046-1053, https://doi.org/10.1175/1520-0493(1980)108%3C1046:TCOEPT%3E2.0.CO;2.

[93] Bowen, T. et al. (2020), Adaptive Social Protection: Building Resilience to Shocks, World Bank, Washington, DC, http://hdl.handle.net/10986/33785.

[85] Brooks, N. (2003), “Vulnerability, risk and adaptation: A conceptual framework”, Working Paper, No. 38, Tyndall Centre for Climate Change Research, Norwich, UK, https://www.researchgate.net/profile/Nick-Brooks-3/publication/200032746_Vulnerability_Risk_and_Adaptation_A_Conceptual_Framework/links/0fcfd50ac169e15865000000/Vulnerability-Risk-and-Adaptation-A-Conceptual-Framework.pdf.

[25] Callahan, C. and J. Mankin (2022), “Globally unequal effect of extreme heat on economic growth”, Science Advances, Vol. 8/43, https://doi.org/10.1126/sciadv.add3726.

[2] Canpolat, E. et al. (2025), Understanding Social Vulnerability for more Effective Climate Strategies: Lessons from CCDRs in Southern and Eastern Africa, World Bank, Washington, DC, https://doi.org/10.1596/43184.

[16] Cappelli, F., V. Costantini and D. Consoli (2021), “The trap of climate change-induced ’natural’ disasters and inequality”, Global Environmental Change, Vol. 70, p. 102329, https://doi.org/10.1016/j.gloenvcha.2021.102329.

[86] Chen, C. et al. (2024), University of Notre Dame Global Adaptation Initiative: Country Index Technical Report, Notre Dame Global Adaptation Initiative (ND-GAIN), University of Notre Dame, Notre Dame, IN, https://gain.nd.edu/assets/581554/nd_gain_countryindex_technicalreport_2024.pdf.

[46] Chen, D. et al. (2023), “Framing, Context, and Methods”, in Masson-Delmotte, V. et al. (eds.), Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK and New York, NY, https://doi.org/10.1017/9781009157896.003.

[47] Cheng, J. et al. (2018), “Heatwave and elderly mortality: An evaluation of death burden and health costs considering short-term mortality displacement”, Environment International, Vol. 115, pp. 334-342, https://doi.org/10.1016/j.envint.2018.03.041.

[37] Chen, M. et al. (2025), “Informal Enterprises Generate Jobs Yet Face Overarching Bias”, WIEGO Blog, Women in Informal Employment: Globalizing and Organizing (WIEGO), https://www.wiego.org/blog/informal-enterprises-generate-jobs/ (accessed on 1 July 2026).

[53] Climate Resilience Center (2026), Extreme Heat Protection Initiative, Climate Resilience Center, Atlantic Council, https://onebillionresilient.org/project/extreme-heat-protection-initiative/ (accessed on 1 April 2026).

[33] Copernicus Climate Change Service CDS (2026), “CMIP6 climate projections”, Copernicus Climate Change Service (C3S) Climate Data Store (CDS) (database), https://doi.org/10.24381/cds.c866074c (accessed on 1 April 2026).

[32] Copernicus Climate Change Service CDS (2026), “ERA5 hourly data on single levels from 1940 to present”, Copernicus Climate Change Service (C3S) Climate Data Store (CDS) (database), https://doi.org/10.24381/cds.adbb2d47 (accessed on 1 April 2026).

[54] Dabrowski, J. (2023), Fighting extreme heat with parametric insurance: protecting the livelihoods of women in India’s informal economy sector, Climate Resilience Center, Atlantic Council, https://onebillionresilient.org/2023/03/07/fighting-extreme-heat-with-parametric-insurance/ (accessed on 1 April 2026).

[68] Davis, B. et al. (2025), “Resilient and inclusive rural transformation: Exploring pathways for sustainable development”, Global Food Security, Vol. 46, p. 100868, https://doi.org/10.1016/j.gfs.2025.100868.

[27] Diffenbaugh, N. and M. Burke (2019), “Global warming has increased global economic inequality”, Proceedings of the National Academy of Sciences, Vol. 116/20, pp. 9808-9813, https://doi.org/10.1073/pnas.1816020116.

[83] Engel, R., E. Anzilotti and M. Palmieri (2025), “From Work to Transportation, Extreme Heat Is Reshaping Urban Life”, World Resources Institute, https://www.wri.org/insights/extreme-heat-reshaping-urban-life (accessed on 1 April 2026).

[8] FAO (2026), “Gender indicators 2000–2024 – June 2026 update”, FAOSTAT Analytical Briefs, No. 128, Food and Agriculture Organization of the United Nations (FAO), Rome, https://openknowledge.fao.org/items/23877723-7b44-47b0-b1bf-5e731a7f5863.

[75] FAO (2026), WASAG - The Global Framework on Water Scarcity in Agriculture, Food and Agriculture Organization of the United Nations (FAO), https://www.fao.org/wasag/awsame-initiative/indigenous-drought-resilient-and-nutritious-crops/en (accessed on 1 April 2026).

[71] FAO (2023), The State of Food and Agriculture 2023, Food and Agriculture Organization of the United Nations (FAO), Rome, https://doi.org/10.4060/cc7724en.

[67] FAO and WMO (2025), Extreme heat and agriculture report: Highlights, Food and Agriculture Organization of the United Nations (FAO) and World Meteorological Organization (WMO), Rome and Geneva, https://openknowledge.fao.org/items/436c55ae-6662-4f7e-9fac-0588db782f5e.

[10] Feriga, M., N. Lozano Gracia and P. Serneels (2024), “The Impact of Climate Change on Work: Lessons for Developing Countries”, Policy Research Working Paper, No. 10682, World Bank, Washington, DC, http://hdl.handle.net/10986/40978.

[20] Flouris, A. et al. (eds.) (2024), Heat at work: Implications for safety and health. A global review of the science, policy and practice, International Labour Organization (ILO), Geneva, https://www.ilo.org/sites/default/files/2024-07/ILO_OSH_Heatstress-R16.pdf.

[57] Garcia Ocampo, D. and C. Lopez Moreira (2024), “Uncertain waters: can parametric insurance help bridge NatCat protection gaps?”, FSI Insights, No. 62, Financial Stability Institute (FSI) of the Bank for International Settlements (BIS) and Secretariat of the International Association of Insurance Supervisors (IAIS), https://www.iais.org/uploads/2024/12/FSI-IAIS-Insights-on-parametric-insurance.pdf.

[39] Ghosh, M. (2024), Heat Havoc: Investigating The Impact on Street Vendors, Greenpeace India, Bangalore and National Hawkers Federation, Delhi, https://www.greenpeace.org/static/planet4-india-stateless/2024/06/09966a5b-heat-havoc_website_use.pdf.

[13] Groen, J., M. Kutzbach and A. Polivka (2020), “Storms and Jobs: The Effect of Hurricanes on Individuals’ Employment and Earnings over the Long Term”, Journal of Labor Economics, Vol. 38/3, pp. 653-685, https://doi.org/10.1086/706055.

[15] Higa, M. (2025), “Is it too hot to work? Evidence from Peru”, Environment and Development Economics, Vol. 31/2-3, pp. 117-136, https://doi.org/10.1017/s1355770x25100272.

[38] HomeNet South Asia (2022), Impact of climate change on urban home-based workers in South Asia: Study of women home-based workers living in slums and informal settlements in Bangladesh, India and Nepal, HomeNet South Asia, Port Louis, Mauritius, https://wrd.unwomen.org/sites/default/files/2022-06/Impact-of-Climate-Change-on-Urban-Home-based-Workers-in-South-Asia.pdf.

[84] Hunter, R. et al. (2023), “Advancing urban green and blue space contributions to public health”, The Lancet Public Health, Vol. 8/9, pp. e735-e742, https://doi.org/10.1016/s2468-2667(23)00156-1.

[61] ILO (2026), “Conclusions of Meeting of Experts on Occupational Safety and Health in Extreme Weather Events and Changing Weather Patterns (Geneva, 20-24 April 2026)”, No. MEEWE/2026, International Labour Organization (ILO), Geneva, https://www.ilo.org/sites/default/files/2026-06/MEEWE_AdoptedConclusion-EN.pdf.

[30] ILO (2026), “SDG indicator 8.3.1 - Proportion of informal employment in total employment by sex and economic activity (%) - Annual”, ILOSTAT (database), https://ilostat.ilo.org/topics/informality (accessed on 1 April 2026).

[60] ILO (2024), Ensuring safety and health at work in a changing climate, International Labour Organization (ILO), Geneva, https://www.ilo.org/sites/default/files/2024-07/ILO_SafeDay24_Report_r11.pdf.

[48] ILO (2024), World Social Protection Report 2024-2026: Universal Social Protection for Climate Action and a Just Transition, International Labour Organization (ILO), Geneva, https://www.ilo.org/sites/default/files/2024-09/WSPR_2024_EN_WEB_1.pdf.

[23] ILO (2019), Working on a warmer planet: The impact of heat stress on labour productivity and decent work, International Labour Organization (ILO), Geneva, https://www.ilo.org/sites/default/files/wcmsp5/groups/public/@dgreports/@dcomm/@publ/documents/publication/wcms_711919.pdf.

[62] ILO (2014), The gender dimension: Integrating the gender perspective in OSH policies, International Labour Organization (ILO), Geneva, https://www.ilo.org/resource/article/gender-dimension-integrating-gender-perspective-osh-policies (accessed on 1 July 2026).

[63] ILO (2006), “Promotional Framework for Occupational Safety and Health Convention, 2006 (No. 187)”, 95th Session of the International Labour Conference (15 June 2006), International Labour Organization (ILO), Geneva, https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C187.

[64] ILO (2001), “Safety and Health in Agriculture Convention, 2001 (No. 184)”, 89th Session of the International Labour Conference (5-21 June 2001), International Labour Organization (ILO), Geneva, https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO:12100:P12100_INSTRUMENT_ID:312329:NO.

[59] ILO (1981), “Occupational Safety and Health Convention, 1981 (No. 155)”, 67th Session of the International Labour Conference (3-24 June 1981), International Labour Organization (ILO), Geneva, https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO::P12100_INSTRUMENT_ID:312300.

[9] IPCC (2023), “Section 2: Current Status and Trends”, in Core Writing Team, H. Lee and J. Romero (eds.), Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Intergovernmental Panel on Climate Change (IPCC), Geneva, pp. 35-115, https://doi.org/10.59327/ipcc/ar6-9789291691647.

[35] IPCC (2022), Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK and New York, NY, https://doi.org/10.1017/9781009325844.

[31] ISO (2017), “Ergonomics of the thermal environment — Assessment of heat stress using the WBGT (wet bulb globe temperature) index”, ISO, No. 7243:2017, International Organization for Standardization, Geneva, https://www.iso.org/standard/67188.html.

[70] JICA (2026), Overview: Analysis of agriculture sector before JICA global agenda, Japan International Cooperation Agency (JICA), https://www.jica.go.jp/english/activities/issues/agricul/overview.html (accessed on 1 April 2026).

[73] JICA (2024), Expansion of Community-Based Smallholder Irrigation Development Project (E-COBSI), Japan International Cooperation Agency (JICA), https://openjicareport.jica.go.jp/pdf/12375887_01.pdf.

[72] JICA (2017), Technical Cooperation Project on the Community-Based Smallholder Irrigation (T-COBSI), Japan International Cooperation Agency (JICA), Zambia Office, https://www.jica.go.jp/Resource/project/english/zambia/017/materials/c8h0vm0000b4gz82-att/leaflet.pdf.

[18] Kjellstrom, T. et al. (2016), “Heat, Human Performance, and Occupational Health: A Key Issue for the Assessment of Global Climate Change Impacts”, Annual Review of Public Health, Vol. 37/1, pp. 97-112, https://doi.org/10.1146/annurev-publhealth-032315-021740.

[22] Kjellstrom, T. et al. (2017), “Estimating population heat exposure and impacts on working people in conjunction with climate change”, International Journal of Biometeorology, Vol. 62/3, pp. 291-306, https://doi.org/10.1007/s00484-017-1407-0.

[21] Lakhoo, D. et al. (2025), “A systematic review and meta-analysis of heat exposure impacts on maternal, fetal and neonatal health”, Nature Medicine, Vol. 31/2, pp. 684-694, https://doi.org/10.1038/s41591-024-03395-8.

[92] Lee, J. et al. (2023), “Future Global Climate: Scenario-based Projections and Near-term Information”, in Masson-Delmotte, V. et al. (eds.), Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK and New York, NY, https://doi.org/10.1017/9781009157896.006.

[5] Longondjo Etambakonga, C. and J. Roloff (2019), “Protecting Environment or People? Pitfalls and Merits of Informal Labour in the Congolese Recycling Industry”, Journal of Business Ethics, Vol. 161/4, pp. 815-834, https://doi.org/10.1007/s10551-019-04339-w.

[29] Maes, M. et al. (2025), “Monitoring exposure to future climate-related hazards: Forward-looking indicator results and methods using climate scenarios”, OECD Environment Working Papers, No. 264, OECD Publishing, Paris, https://doi.org/10.1787/b9ba6ee0-en.

[50] McLeod, K. and L. Goering (2025), “As Extreme Heat Threats Grow, Can Heat Insurance Offer Protection?”, Journal of City Climate Policy and Economy, Vol. 3/2/3, pp. 215-222, https://doi.org/10.3138/jccpe-2024-0044.

[45] Meade, R. et al. (2025), “Responding to rising heat in workplaces and homes of low income workers”, BMJ, Vol. 391, p. e086218, https://doi.org/10.1136/bmj-2025-086218.

[26] Miller, S. et al. (2021), “Heat Waves, Climate Change, and Economic Output”, Journal of the European Economic Association, Vol. 19/5, pp. 2658-2694, https://doi.org/10.1093/jeea/jvab009.

[19] NIOSH (2026), Heat-related Illnesses, National Institute for Occupational Safety and Health (NIOSH), Centers for Disease Control and Prevention (CDC), Department of Health and Human Services (DHHS), https://www.cdc.gov/niosh/heat-stress/about/illnesses.html (accessed on 1 April 2026).

[82] Nutkiewicz, A. et al. (2022), “Cool roofs can mitigate cooling energy demand for informal settlement dwellers”, Renewable and Sustainable Energy Reviews, Vol. 159, p. 112183, https://doi.org/10.1016/j.rser.2022.112183.

[69] OECD (2026), Improving Environmental and Climate Outcomes in Agri-food Projects: A Guide for Development Practitioners, OECD Publishing, Paris, https://doi.org/10.1787/a4e26a65-en.

[6] OECD (2026), “Key Indicators of Informality based on Individuals and their Households (KIIbIH)”, OECD International Development Statistics (database), https://data-explorer.oecd.org (accessed on 1 July 2026).

[28] OECD (2026), “Number of hot days”, Projected exposure to extreme temperature (database), https://data-viewer.oecd.org?chartId=98d2d4d0-357d-4e38-b7ef-60819a695936 (accessed on 1 April 2026).

[51] OECD (2025), Economic Outlook for Southeast Asia, China and India 2025: Enhancing Disaster Risk Financing, OECD Publishing, Paris, https://doi.org/10.1787/6fc95782-en.

[41] OECD (2024), SIGI 2024 Regional Report for Southeast Asia: Time to Care, Social Institutions and Gender Index, OECD Publishing, Paris, https://doi.org/10.1787/7fc15e1c-en.

[49] OECD (2024), “The role of social protection in the fight against hunger and poverty: Summary and key messages”, 5th Policy Dialogue on Social Protection and Development (PD-SPD), OECD Development Centre and USP2030, https://www.oecd.org/content/dam/oecd/en/events/2024/12/5th-oecd-policy-dialogue-social-protection-and-dev/Summary%20and%20key%20messages.pdf.

[87] OECD (2024), Towards Greener and More Inclusive Societies in Southeast Asia, Development Centre Studies, OECD Publishing, Paris, https://doi.org/10.1787/294ce081-en.

[40] OECD (2023), SIGI 2023 Global Report: Gender Equality in Times of Crisis, Social Institutions and Gender Index, OECD Publishing, Paris, https://doi.org/10.1787/4607b7c7-en.

[43] OECD (2021), Bringing Household Services Out of the Shadows: Formalising Non-Care Work in and Around the House, Gender Equality at Work, OECD Publishing, Paris, https://doi.org/10.1787/fbea8f6e-en.

[77] OECD (2019), OECD Principles on Rural Policy, OECD Publishing, Paris, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/rural-service-delivery/principles-rural-policy-brochure.pdf.

[78] OECD/UN-Habitat (2025), Climate Mitigation in Intermediary Cities: Policies for Systemic Change in Kenya and Mozambique, OECD Publishing, Paris, https://doi.org/10.1787/3bdd564f-en.

[7] OECD/UN-Habitat (2022), Intermediary Cities and Climate Change: An Opportunity for Sustainable Development, OECD Publishing, Paris, https://doi.org/10.1787/23508323-en.

[90] Parish, O. and T. Putnam (1977), “Equations for the determination of humidity from dewpoint and psychrometric data”, NASA Technical Note (TN), No. D-8401, National Aeronautics and Space Administration (NASA), Washington, DC, https://ntrs.nasa.gov/citations/19770009916.

[12] Pleninger, R. (2022), “Impact of natural disasters on the income distribution”, World Development, Vol. 157, p. 105936, https://doi.org/10.1016/j.worlddev.2022.105936.

[56] Raithatha, R. (2024), “Index insurance innovation: Protecting families against extreme heat in India”, Micro Insurance Network, https://microinsurancenetwork.org/posts/index-insurance-innovation-protecting-families-against-extreme-heat-in-india (accessed on 1 April 2026).

[55] Raithatha, R. (2023), “Innovation in microinsurance: Reflections from ICII 2023”, Micro Insurance Network, https://microinsurancenetwork.org/posts/innovation-in-microinsurance-reflections-from-icii-2023 (accessed on 1 April 2026).

[24] Romanello, M. et al. (2025), “The 2025 report of the Lancet Countdown on health and climate change: climate change action offers a lifeline”, The Lancet, Vol. 406/10521, pp. 2804-2857, https://doi.org/10.1016/s0140-6736(25)01919-1.

[79] Satterthwaite, D. et al. (2018), “Responding to climate change in cities and in their informal settlements and economies”, Paper prepared for the IPCC for the International Scientific Conference on Cities and Climate Change in Edmonton (March 2018), International Institute for Environment and Development (IIED), https://www.iied.org/sites/default/files/pdfs/migrate/G04328.pdf.

[34] Sinha, S. et al. (2026), “Working in the Heat: Home-Based Workers and Climate Justice in Delhi”, WIEGO Policy Brief, No. 35, Women in Informal Employment: Globalizing and Organizing (WIEGO), Manchester, UK, https://www.wiego.org/research-library-publications/home-based-workers-heat-climate-delhi/.

[66] Sinha, S. et al. (2026), “Working in the Heat: Street Vendors and Climate Justice in Delhi, India”, Policy Brief, No. 34, Women in Informal Employment: Globalizing and Organizing (WIEGO), Manchester, UK, https://www.wiego.org/research-library-publications/working-heat-street-vendors-delhi-india/.

[89] Stull, R. (2011), “Wet-Bulb Temperature from Relative Humidity and Air Temperature”, Journal of Applied Meteorology and Climatology, Vol. 50/11, pp. 2267-2269, https://doi.org/10.1175/jamc-d-11-0143.1.

[76] Sutton, W., A. Lotsch and A. Prasann (2024), Recipe for a Livable Planet: Achieving Net Zero Emissions in the Agrifood System, World Bank, Washington, DC, http://hdl.handle.net/10986/41468.

[52] Suzumu, M. and S. Chanel (2024), “From Vulnerability to Empowerment: The Role of Parametric Insurance for Women in Fiji”, United Nations Capital Development Fund (UNCDF), https://www.uncdf.org/article/8682/from-vulnerability-to-empowerment-the-role-of-parametric-insurance-for-women-in-fiji (accessed on 1 April 2026).

[88] Teber, K. et al. (2025), “Geo-Disasters: Geocoding climate-related events in the international disaster database EM-DAT”, Big Earth Data, pp. 1-16, https://doi.org/10.1080/20964471.2025.2576274.

[4] Tucker, J. and M. Anantharaman (2020), “Informal Work and Sustainable Cities: From Formalization to Reparation”, One Earth, Vol. 3/3, pp. 290-299, https://doi.org/10.1016/j.oneear.2020.08.012.

[17] UCLouvain / CRED (2025), Emergency Events Database (EM-DAT), Brussels, Belgium, https://www.emdat.be (accessed on 1 December 2025).

[94] UN Women (2021), Measuring Time Use: An assessment of issues and challenges in conducting time-use surveys with special emphasis on developing countries, United Nations Entity for Gender Equality and the Empowerment of Women (UN Women), Global Centre of Excellence on Gender Statistics (CEGS), Mexico City, https://data.unwomen.org/sites/default/files/documents/Publications/Measuring%20time%20use.pdf.

[65] Valdivia, M. et al. (2025), “Climate Change and Heat Stress: Impacts on Home-Based Workers and Street Vendors in Bangkok, Thailand”, WIEGO Policy Brief, No. 33, Women in Informal Employment: Globalizing and Organizing (WIEGO), Manchester, UK, https://www.wiego.org/wp-content/uploads/2025/10/WIEGO-Policy-Brief-33-Climate-Change-and-Heat-Stress.pdf.

[42] van de Ven, P., J. Zwijnenburg and M. De Queljoe (2018), “Including unpaid household activities: An estimate of its impact on macro-economic indicators in the G7 economies and the way forward”, OECD Statistics Working Papers, No. 2018/4, OECD Publishing, Paris, https://doi.org/10.1787/bc9d30dc-en.

[11] Wagner, P. (2025), “The hole dug deeper: Flash floods, income disparities, and labor informality in Brazil”, Ecological Economics, Vol. 237, p. 108689, https://doi.org/10.1016/j.ecolecon.2025.108689.

[36] WIEGO (2026), Climate Change and the Urban Informal Economy: Understanding Impacts and Strengthening Resilience, https://www.wiego.org/project/climate-change-urban-informal-economy/ (accessed on 1 July 2026).

[74] World Bank (2024), Advancing Climate-Smart Agriculture Technologies in Africa, https://www.worldbank.org/en/news/press-release/2024/03/08/advancing-climate-smart-agriculture-technologies-in-africa (accessed on 1 April 2026).

[3] World Bank (2024), Unseen Green Jobs: A Study on Informal Waste Workers in Vientiane Capital, Lao PDR, World Bank, Washington, DC, http://hdl.handle.net/10986/41754.

[1] Yeganeh, K. (2020), “A typology of sources, manifestations, and implications of environmental degradation”, Management of Environmental Quality: An International Journal, Vol. 31/3, pp. 765-783, https://doi.org/10.1108/meq-02-2019-0036.

Notes

Copy link to Notes

← 1. Benin, Bolivia, Brazil, Cameroon, El Salvador, Ghana, Jamaica, Lao PDR, Madagascar, Maldives, Mali, Namibia, Nicaragua, Niger, Paraguay, Senegal, Sierra Leone, Thailand, Togo, Uganda, Viet Nam and Zambia.

← 2. While the distinctive pattern of droughts – relatively rare but long-lasting – likely reflects their slow-onset nature, it may also point to important limitations of EM-DAT. To be included in EM-DAT, an event must meet at least one of the following inclusion criteria: at least ten deaths (including dead and missing); at least 100 people affected (people affected, injured or homeless); and/or a call for international assistance or an emergency declaration. Unlike floods, which are visible events that often immediately lead to impacts and damages, droughts are silent, slow onset events, the consequences of which are much more difficult to track and reveal.

← 3. Ethiopia, Kenya, Madagascar, Malawi, Namibia, Niger and South Africa.

← 4. Benin, Cameroon, Gambia, Ghana, Lao PDR, Madagascar, Mali, Mongolia, Nicaragua, Niger, Peru, Sierra Leone, Thailand, Togo, Uganda, Viet Nam and Zambia.

← 5. Cameroon, Lao PDR, Madagascar, Mali, Mongolia, Niger, Sierra Leone, Thailand, Uganda, Viet Nam and Zambia.

← 6. Bolivia, Cameroon, Colombia, Gambia, Ghana, Lao PDR, Madagascar, Mali, Mongolia, Niger, Sierra Leone, Uganda, Viet Nam and Zambia.

← 7. The SSP2‑4.5 scenario is one of five core scenarios based on Shared Socio-economic Pathways (SSPs) developed by the Intergovernmental Panel on Climate Change (IPCC), using simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The SSP2‑4.5 scenario represents a “middle‑of‑the‑road” development path in which social, economic and technological trends continue broadly in line with historical patterns. It assumes moderate challenges to both mitigation and adaptation, with radiative forcing reaching 4.5 W/m² by 2100. In this scenario, global emissions peak around mid‑century and then decline, but not fast enough to avoid significant warming beyond 2°C (Chen et al., 2023[46]; Lee et al., 2023[92]).

← 8. As of 2021, the United Nations estimated that 82 countries had conducted time-use surveys at national level, corresponding to only 38% of countries worldwide. About half of them (44 countries) had repeated their time-use survey at least once (UN Women, 2021[94]).

← 9. Parametric insurance as discussed in this chapter refers to insurance schemes that provide payouts directly to individuals. This differs from how parametric insurance is usually understood within adaptive social protection (Bowen et al., 2020[93]), under which it often serves as a market-based financing mechanism for governments. In the latter framework, payouts triggered by catastrophic or adverse events can be used to finance the vertical or horizontal expansion of social protection programmes, in line with the objectives of adaptive social protection.

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & events

News & events

Featured events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment