Policy Research Working Paper 11173 The Future of Poverty Projecting the Impact of Climate Change on Global Poverty through 2050 Johanna Fajardo-Gonzalez Minh C. Nguyen Paul A. Corral Rodas Poverty and Equity Global Department July 2025 Policy Research Working Paper 11173 Abstract Climate change is increasingly acknowledged as a critical bear the brunt, contributing the largest number of poor issue with far-reaching socioeconomic implications that people, with estimates ranging between 40.5 million and extend well beyond environmental concerns. Among the 73.5 million by 2050. Another significant finding is the most pressing challenges is its impact on global poverty. disproportionate impact of inequality on poverty. Even This paper projects the potential impacts of unmitigated small increases in inequality can lead to substantial rises in climate change on global poverty rates between 2023 and poverty levels. For instance, if every country’s Gini coeffi- 2050. Building on a study that provided a detailed analysis cient increases by just 1 percent between 2022 and 2050, of how temperature changes affect economic productivity, an additional 8.8 million people could be pushed below the this paper integrates those findings with binned data from international poverty line by 2050. In a more extreme sce- 217 countries, sourced from the World Bank’s Poverty and nario, where every country’s Gini coefficient increases by 10 Inequality Platform. By simulating poverty rates and the percent between 2022 and 2050, the number of people fall- number of poor under two climate change scenarios, the ing into poverty could rise by an additional 148.8 million paper uncovers some alarming trends. One of the primary relative to the baseline scenario. These findings underscore findings is that the number of people living in extreme the urgent need for comprehensive climate policies that poverty worldwide could be nearly doubled due to climate not only mitigate environmental impacts but also address change. In all scenarios, Sub-Saharan Africa is projected to socioeconomic vulnerabilities. This paper is a product of the Poverty and Equity Global Department. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at pcorralrodas@worldbank.org, jfajardog@worldbank.org, and mnguyen3@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Future of Poverty: Projecting the Impact of Climate Change on Global Poverty through 2050* Johanna Fajardo-Gonzalez, Minh C. Nguyen, and Paul A. Corral Rodas JEL Classification: Q54; I32; O1 Keywords: Climate change; temperature; poverty; inequality * This paper is a product of the Poverty and Equity Global Department. The names of the authors have been ordered using the American Economic Association Author Randomization Tool, per confirmation code: d9YfWniXZzwV. Corresponding author: Paul Andres Corral Rodas – pcorralrodas@worldbank.org 1. Introduction The average global temperature has increased by more than one degree Celsius since pre-industrial times (Masson-Delmotte et al. 2021), with far-reaching implications for human welfare. Substantial evidence indicates that rising temperatures and changing precipitation patterns negatively impact economic growth, labor productivity, and agricultural yields, and are expected to disproportionately affect poorer countries (Kotz, Levermann, and Wenz, 2022; Waldhoff et al., 2020; Lee, Villaruel and Gaspar, 2016; Burke, Hsiang, and Miguel, 2015; Dell et al., 2014). These climate-related impacts have the potential to derail development progress, especially in settings with limited adaptive capacity. The relationship between climate change and poverty has been extensively studied, revealing a strong bidirectional link between the two (Onyeaka et al., 2024; Barbier and Hochard, 2018; Seaman et al., 2014; Hope, 2009; Ahmed, Diffenbaugh, and Hertel, 2009). For instance, Hallegatte and Rozenberg (2017) highlight how climate change exacerbates poverty by increasing climate variability and extreme weather events, which erode people's adaptive capacity and create persistent poverty traps. Their findings build upon Hallegatte et al. (2016), who emphasize that climate shocks—such as food price spikes, reduced agricultural yields, and increased health risks—pose long-term threats to poverty reduction efforts. A key finding from Hallegatte and Rozenberg (2017) is that, by 2030, climate change could push between 35 million and 122 million people into extreme poverty, primarily due to agricultural disruptions and rising food prices. Jafino et al. (2020) provide a revised estimate, suggesting that climate change could push between 32 million and 132 million people into extreme poverty by 2030, with food price increases, natural disasters, and health impacts— including those intensified by the COVID-19 pandemic—playing significant roles (Jafino et al., 2020). Further emphasizing the link between temperature increases and poverty, Dang et al. (2025) analyze data from over 1,594 subnational units and find that a 1°C increase in temperature is associated with an increase of the number of global poor due to global warming in the range of 127.5 million to 202.1 million people by 2030, using the $2.15 (2017 USD PPP) per day poverty threshold. Their findings highlight heightened vulnerability in Sub-Saharan Africa, where climate change is projected to have the most severe poverty-inducing effects, although the impact on inequality varies across regions depending on policy interventions. While studies have explored climate impacts on either poverty or inequality, fewer have modeled how these two forces may interact over time. Climate change has been shown to exacerbate income inequality—particularly in agriculture-dependent economies—where rainfall anomalies and temperature increases affect lower-income households more acutely (Diffenbaugh and Burke, 2019; Wildowicz-Szumarska and Owsiak, 2024; Palagi et al., 2022). Yet, most global modeling efforts treat poverty and inequality separately or rely on stylized assumptions to estimate inequality trajectories. This study advances the literature by jointly examining the impact of climate change on poverty and inequality through 2050. Building on the climate–income projection framework of Burke, Hsiang, and Miguel (2015), it simulates how long-run temperature increases affect GDP per capita, and how 2 those shocks translate into changes in poverty under varying assumptions about income distribution. Importantly, this study also explores the role of inequality in amplifying poverty outcomes, offering projections under different Gini scenarios to reflect uncertainty in future distributional trajectories. Research suggests that under climate change scenarios, rising food prices may significantly increase global poverty, particularly in the developing world, where food consumption makes up a large share of household expenditures (Blanz and Kalkuhl, 2022). In addition, projections for Sub- Saharan Africa indicate that increasing temperatures could intensify chronic poverty rates, further deepening economic disparities (Dang et al., 2024b). Thus, this study quantifies how different levels of income inequality could affect global poverty projections, providing policy makers with concrete insights into the potential consequences of inaction. Addressing climate change without tackling income disparities may lead to worsening poverty traps in the most vulnerable regions (Hallegatte, Fay, and Barbier, 2018). Therefore, an integrated approach—linking climate adaptation policies with inclusive economic growth strategies—is essential to mitigate both climate and inequality-driven poverty risks. To implement these projections, this study combines two key datasets. First, it draws on the global climate–income projections of Burke, Hsiang, and Miguel (2015), which provide country-level estimates of the impact of rising temperatures on future GDP per capita. Second, it integrates harmonized microdata from the World Bank’s Poverty and Inequality Platform (PIP), which includes income distributions for 217 countries. Unlike studies that rely on modeled inequality inputs—such as those from the International Futures model—the approach in this paper uses observed within- country income distributions to simulate poverty impacts. This enables more precise country- and region-specific estimates grounded in actual welfare levels, enhancing both the granularity and the policy relevance of the analysis. Burke, Hsiang, and Miguel (2015) provide insights into how economies might respond to rising temperatures through their analysis of data from 165 countries over five decades. 1 They identify a strong, non-linear relationship between annual average temperature and GDP per capita growth, with economic productivity peaking around 13°C and declining sharply at higher temperatures. Under a business-as-usual emissions scenario, their projections suggest that global economic output could be reduced by 23 percent by 2100 compared to a world without climate change. In most scenarios, projected global income losses exceed 20 percent, underscoring the potentially severe economic consequences of inaction. While the main specification assumes that rich and poor countries respond similarly to temperature changes, the economic impacts are substantially larger for poorer countries, as these are predominantly located in already warm regions. Interestingly, when the model allows for heterogeneous responses by income level or geography, warming appears even more damaging for colder, wealthier countries—though these projections carry greater uncertainty due to data limitations. 1 For countries lacking growth projections, the region’s average outcome is used. 3 Looking toward 2100, Burke, Hsiang, and Miguel’s (2015) projections suggest a concerning trend. Unmitigated climate change could render a staggering 77 percent of countries poorer in per capita terms than they would be without climate change. Even countries with high baseline growth and rapid income convergence could potentially be worse off. This sobering reality particularly affects countries with higher average temperatures. One of the most striking findings is that if climate change continues unchecked, the world’s poorest 40 percent of countries could lose nearly 75 percent of their relative income by 2100—compared to a scenario without climate change. Meanwhile, the wealthiest 20 percent might experience modest gains, thanks to their currently cooler climates. This study uses 2022 as a baseline, when an estimated 715 million people lived on less than $2.15 per day (2017 USD PPP). From that point forward, poverty projections are based on country-level income forecasts from Burke, Hsiang, and Miguel (2015). Initial simulations assume fixed within- country income distributions, but subsequent scenarios relax this assumption to explore the role of changing inequality, yielding significantly larger poverty impacts. It is important to note that the analysis in this paper focuses on prospective impacts of climate change from 2022 onward and does not account for past climate-related effects. While existing literature projects that climate change could push between 32 million and 132 million people into poverty by 2030, depending on the severity of impacts and policy responses (Jafino et al., 2020), there is limited empirical evidence quantifying how many people have already fallen into poverty due to climate-related shocks in previous years. Consistent with the projected trajectory, the projections in this study suggest that by 2025, global poverty could reach 584.8 million under a high-emissions scenario, compared to 565.2 million under the baseline—implying that roughly 20 million additional people could be driven into poverty in just a few years due to continued inaction on climate mitigation. Looking ahead to 2030, the global population in extreme poverty is expected to be 389.2 million under the high-emission scenario versus 351.0 million under the baseline scenario. By 2050, the numbers are projected to be 95.0 million people under the high-emission scenario and 52.3 under the baseline emission scenario. Regional Projections: • Sub-Saharan Africa: The region is anticipated to bear the highest poverty burden in all scenarios, with 294.1 million people (20.0 percent of the region’s population) in poverty by 2030 and 73.5 million (3.5 percent of the region’s population) by 2050 under the high- emission scenario. Numbers are lower in the baseline, 269.7 million (18.4 percent) by 2030 and 40.5 million (1.9 percent) by 2050. • South Asia: The number of people in poverty is projected to be 48.8 million (2.3 percent) by 2030 under the high-emission scenario versus 38.7 million (1.8 percent) under the baseline, and 3.4 million (0.1 percent) by 2050 under the high-emission scenario versus the zero- poverty baseline by 2050. 4 • Latin America and the Caribbean: Projections for 2030 under the high-emission scenario are 12.7 million (1.9 percent) and 11.1 million (1.6 percent) under the baseline. By 2050 projections suggest nearly 4 million (0.5 percent) under the high-emission scenario and 2.4 million (0.3 percent) under the baseline. • East Asia and the Pacific: Extreme poverty induced by climate change could affect 6.3 million people (0.29 percent) by 2030 under the high emission scenario (5.6 million or 0.26 percent under the baseline), and 0.9 million (0.04 percent) by 2050 (0.5 million or 0.02 percent under the baseline). It is important to mention that the Burke, Hsiang, and Miguel’s growth projections under the SSP5 baseline scenario assume rapid economic growth and technological advancement, potentially producing optimistic poverty scenarios that may not reflect realistic future trajectories. For instance, these projections suggest only 52.3 million people would remain in extreme poverty by 2050 even without policy action on climate change, a figure that may underestimate persistent poverty challenges in many regions. In the absence of better alternatives with comparable global coverage and methodological consistency, this study utilizes these projections while acknowledging their limitations. Readers should therefore interpret the absolute poverty numbers with caution and focus more on the relative differences between scenarios, which highlight the additional burden imposed by climate change. In projecting poverty rates through 2050, this study also focuses on how changes in income inequality could amplify climate-driven poverty impacts. Recent evidence suggests that temperature and precipitation anomalies are drivers of between- and within-country inequality (Gilli et al., 2024; Paglialunga et al., 2022; Diffenbaugh and Burke, 2019). This creates a potential feedback loop: climate change exacerbates inequality, while rising inequality leaves the poorest populations increasingly vulnerable to climate impacts. The relationship between inequality and poverty extends beyond pure economic factors to encompass differential access to coping mechanisms. Wealthier households can often buffer climate shocks through their access to assets, social networks, and government support. In contrast, poorer households typically lack the financial and social capital needed for climate adaptation, leaving them more vulnerable to extreme and frequent weather events such as floods, droughts, and heatwaves that can disrupt livelihoods and deepen poverty. Another major finding is that even small increases in inequality can lead to disproportionately large increases in poverty. For instance, if every country’s Gini coefficient (a common measure of income inequality) increases by just 1 percent between now and 2050, an additional 8.8 million people could be pushed below the international poverty line, compared to the case of no poverty-worsening inequality. In a more extreme scenario, where every country’s Gini coefficient increases by 10 percent between now and 2050, the number of people falling into poverty could rise by an additional 148.8 million, compared to the case of no poverty-worsening inequality. These projections underscore the fact that worsening inequality magnifies the impact of climate change on poverty. The remainder of this paper is organized as follows: Section 2 presents the data sources, and Section 3 outlines the climate scenarios. Section 4 examines poverty simulations under different climate 5 scenarios, while Section 5 expands the analysis to incorporate changes in income inequality. Finally, Sections 6 and 7 discuss the uncertainty of the results and provide concluding remarks. 2. Data Sources The analysis in this paper uses two key data sources. First, the GDP impact data is derived from Burke, Hsiang, and Miguel (2015), whose research provides one of the most comprehensive assessments of the nonlinear effects of temperature on economic productivity. Drawing on over 50 years of data from 165 countries, their study identifies a critical temperature threshold between 20°C and 30°C. Within this range, economic productivity is optimized; however, as temperatures exceed 30°C, there are significant declines in labor productivity and agricultural yields. Conversely, in cooler regions with temperatures below 20°C, modest warming can initially boost productivity and crop yields, reflecting the nonlinear relationship between temperature and economic outcomes. The second key data source is the World Bank’s Poverty and Inequality Platform (PIP), which compiles harmonized household survey data from a broad range of countries. Maintained by the World Bank, the PIP provides nationally representative information on income, consumption, and other socioeconomic indicators. This paper draws on a dataset of 217 countries that captures the global distribution of welfare from 1990 to 2024, with each year including 1,000 welfare bins (based on income or consumption) per country. 2 In the PIP, in cases where household survey data for a given year are unavailable, welfare aggregates from the most recent available year are projected forward using growth rates from national accounts—typically real GDP per capita or real Household Final Consumption Expenditure (HFCE) per capita. However, not all economies have usable household surveys or the necessary price data (such as credible PPPs or CPIs) to estimate international poverty. To ensure full coverage in regional and global poverty estimates, the PIP imputes a poverty headcount for these countries based on their region’s population-weighted average. 3 This paper uses the 2022 distribution file from PIP for the main analysis. The PIP provides binned data for all 165 countries studied in Burke, Hsiang, and Miguel (2015). The 52 countries not directly matched in the PIP are still included in the final analysis by assigning them the mean projected poverty rates of their respective regions. Together, these data sources allow for a detailed analysis of how climate change may affect both national economies and individual households. By integrating macroeconomic projections with microeconomic data, the projections presented in this study should be interpreted as illustrative scenarios that highlight potential poverty impacts under specific climate and inequality assumptions. They are not intended as precise forecasts, but rather as exploratory estimates that help quantify the scale and direction of risks in the absence of strong mitigation or adaptation. 2 Further details on the construction of this binned dataset can be found in Mahler, Nishant, and Lakner (2022). 3 A summary of the PIP methodology is available in Appendix 1. 6 3. Climate Change Scenarios This study bases its projections on two “business-as-usual” emissions scenarios. The SSP5-RCP8.5 High Emission Scenario is based on the RCP8.5 (Representative Concentration Pathway) and SSP5 (Shared Socioeconomic Pathway) climate scenarios, while the SSP5 Baseline Scenario relies solely on SSP5. To examine the potential impacts of unmitigated climate change on GDP, this study draws on the work of Burke, Hsiang, and Miguel (2015). 4 The SSP5 Baseline Scenario represents a world characterized by rapid economic growth and technological advancement, but with a continued dependence on fossil fuels as the primary energy source. This scenario assumes that while global economic development accelerates, it is unevenly distributed, leading to significant regional inequalities. Wealthier nations and regions experience substantial economic gains and technological advancements, enabling them to better adapt to the impacts of climate change. In contrast, poorer countries and regions face heightened vulnerability, as their adaptive capacity fails to keep pace with the increasing frequency and intensity of climate- related shocks. SSP5 is also useful for exploring the interaction between socioeconomic trajectories and the physical impacts of climate change. For example, under SSP5, global economic growth could potentially reduce poverty rates in some regions, but the reliance on fossil fuels would exacerbate climate change, leading to severe environmental and social consequences in already vulnerable areas. This pathway illustrates how economic growth alone is insufficient to shield against climate risks without coordinated global efforts to reduce emissions and address structural inequalities. The RCP8.5 (Representative Concentration Pathway) scenario represents a high-emissions future in which global temperatures rise by approximately 4°C by the end of the century. This scenario assumes limited or no meaningful efforts to mitigate greenhouse gas emissions, with continued reliance on fossil fuels and minimal progress in reducing global carbon intensity. Under RCP8.5, the physical impacts of climate change, such as extreme heat, shifting precipitation patterns, rising sea levels, and intensified weather events, are projected to reach their most severe levels. The RCP8.5 scenario is particularly relevant for understanding the risks faced by low- and middle- income countries, where adaptive capacities are often limited due to financial, infrastructure, and institutional constraints. In such regions, the impacts of rising temperatures and extreme climate events are likely to disproportionately affect vulnerable populations, exacerbating poverty, food insecurity, and economic instability. RCP8.5 provides insight into the most severe outcomes if global mitigation efforts fail, highlighting the urgent need for adaptation strategies, especially in regions already operating near critical temperature thresholds. It is worth emphasizing that the SSP5 scenario embodies particularly optimistic assumptions about economic convergence and poverty reduction that may not materialize in practice. Under these assumptions, extreme poverty would nearly disappear in most regions by 2050 even without climate 4 Burke, Hsiang, and Miguel (2015) referred to the scenario based on SSP5 as the 'Projected per capita GDP without climate change' scenario. However, it is important to note that this scenario does include climate change and is characterized by high emissions due to the prioritization of economic development over environmental sustainability. 7 mitigation—a trajectory that appears increasingly unlikely given current trends. Nevertheless, by comparing the SSP5-RCP8.5 scenario against this optimistic baseline, the analysis provides valuable insights into the potential magnitude of climate-driven poverty impacts, even if the absolute poverty levels in both scenarios should be interpreted with caution. In summary, while SSP5 and RCP8.5 both describe high-emission futures, SSP5 focuses on socioeconomic pathways leading to high emissions, and RCP8.5 focuses on the resulting radiative forcing and climate impacts. 5 SSP5 is often paired with RCP8.5 to create a consistent high-emission scenario that combines socioeconomic assumptions with climate projections. Importantly, both scenarios assume limited global action on adaptation and mitigation, framing the results as illustrative of outcomes in a “no-policy change” world. 6 4. Simulating the Impacts of Climate Change on Poverty This analysis employs a top-down macro-to-micro simulation approach, linking projected changes in GDP per capita—driven by temperature shocks—to household-level welfare data. This methodology facilitates cross-country comparability and leverages the rich detail of the Poverty and Inequality Platform (PIP). However, it is important to situate this approach within the broader literature on climate-poverty modeling. Alternative “bottom-up” methods, such as those used in the Shock Waves report (Hallegatte et al., 2016) and Jafino et al. (2020), directly simulate the impacts of climate change on households by modeling channels such as consumption losses, health shocks, asset destruction, and sector-specific vulnerabilities. These approaches can capture more granular effects that may not be reflected in GDP aggregates but that nonetheless affect poverty dynamics. By contrast, the top-down approach used in this paper prioritizes consistency with macroeconomic climate scenarios and is well suited for generating systematic, large-scale projections. It does, however, risk underestimating the impact of localized, non-GDP-driven climate shocks. The two approaches are therefore complementary, offering distinct insights into the complex relationship between climate change and poverty. The methodology for projecting the poverty impacts of climate change also involves two key approaches. In the first approach, explained in this section, economic growth (measured by GDP per capita) is assumed to be evenly distributed across the population. This "neutral distribution approach” provides a baseline projection, assuming no changes in income inequality within countries. 7 The second approach, described in Section 5, relaxes this assumption by incorporating potential changes in income inequality. Using a method that simulates changes in Gini via a Growth 5 The measure of additional energy on the Earth’s atmosphere reaches 8.5 watts per square meter by 2100. 6 Appendix 3 summarizes the regional GDP per capita growth projections under both the SSP5 baseline and SSP5-RCP8.5 scenarios, based on Burke et al. (2015). These projections serve as the foundation for the simulated poverty outcomes presented in this paper. Highlighting the divergence in growth trajectories between the two scenarios helps illustrate the mechanism by which climate-induced GDP losses affect poverty. Readers are encouraged to interpret the poverty results in light of these underlying macroeconomic assumptions. 7 The World Bank uses the neutral distribution approach to estimate global poverty counts for countries without survey data for the desired year (Prydz et al., 2019). An evaluation of poverty projection methods in Latin America from 2003 to 2014 found this approach to perform well (Carusso et al., 2017). While the approach relies on a relatively strong assumption that could introduce bias over the long term (Karver, Kenny, and Sumner, 2012), its simplicity is a strength in the face of the significant uncertainties involved in long-term poverty projections. 8 Adding more complex assumptions could introduce biases that might either compound or offset each other, with limited insight into which outcome is more likely. Incidence Curve, this approach allows growth to vary across the income distribution, enabling the analysis to simulate how worsening inequality—potentially exacerbated by climate change—may amplify poverty outcomes. Together, these two approaches provide complementary insights into the distributional consequences of climate-induced economic shocks. To simulate the impacts of climate change on poverty, the economic growth projections from Burke, Hsiang, and Miguel (2015) are first applied to the household survey data from the PIP. Under the neutral distribution approach, it is assumed that economic growth is evenly distributed across the population. This means that all households experience the same proportional change in income, based on the projected growth rate for their country. 8 Under the neutral distribution approach scenario, the analysis models how changes in national income, driven by climate-induced productivity losses, translate into changes in household-level income and consumption. Under the neutral distribution approach, expenditure or income per capita for household in country in time under climate scenario is equal to: = 0 �(1 + ) (1) =1 Where is the income per capita growth rate provided by Burke, Hsiang, and Miguel (2015) under climate scenario for year in country , and is the pass-through between economic growth rate and growth rate of household income. With income for time it is then possible to derive poverty headcount. To derive the total number of poor for a given year, the model relies on population projections from the United Nations World Population Prospects (UNDESA, 2024). Using the method outlined above, it is estimated that by 2050, climate change could push an additional 42.8 million people below the international poverty line of $2.15 per day (2017 PPP) (Figure 1), which is the difference in the number of poor people between the SSP5 Baseline scenario (52.3 million poor people) and the SSP5-RCP8.5 High Emission scenario (95.0 million poor people). For the $3.65 and $6.85 international poverty lines, the estimated number of additional poor under climate change are 194.2 million and 692.9 million people, respectively. Figure 2 illustrates the estimated number of individuals living in extreme poverty across various regions. By 2050, the Sub-Saharan Africa region is projected to have the highest number of people in extreme poverty, with 73.5 million individuals in a neutral distribution scenario with high emissions. This is followed by the Middle East and North Africa, with 7.0 million; Latin America and the Caribbean, with 3.9 million; South Asia, with 3.4 million; East Asia and the Pacific, with 0.9 million; and Eastern Europe and Central Asia, with 0.2 million. High-income countries (Figure 2, panel g) 8 The analysis applies standard passthrough rates to translate projected changes in GDP per capita into changes in household welfare. A passthrough rate of 1.0 is used for countries with income-based welfare aggregates, and a rate of 0.7 for those using consumption-based aggregates. These parameters are derived from historical relationships observed in non-climate-related growth episodes. However, it is important to acknowledge that climate-induced economic shocks may differ in nature from typical growth fluctuations. In particular, climate shocks may disproportionately affect certain sectors or regions and may be more difficult to smooth through household coping mechanisms, especially in contexts with limited adaptation capacity. As such, the passthrough assumptions introduce uncertainty into the projections and may not fully capture the heterogeneous ways in which climate impacts translate into household-level outcomes. In the absence of climate-specific passthrough estimates, these values provide a practical benchmark but should be interpreted with 9 caution. exhibit a steady pattern throughout the study period, contributing an average of 6.1 million people by 2050. Previous estimates suggested that the number of people in extreme poverty in Sub-Saharan Africa could reach 224.4 million by 2030 (Jafino et al., 2020). The estimates in this paper, which suggest that the number of people in extreme poverty in this region could range between 269.7 million and 294.1 million by 2030, are relatively close to those provided by Jafino et al. Figure 1: Global Number of Poor Under Climate Change (International poverty line: $2.15 a day, 2017 PPP) Source: Authors’ calculations based on income per capita projections from Burke, Hsiang, and Miguel (2015) and country level household survey data from the Poverty and Inequality Platform (PIP). Note: Simulations assume economic growth (GDP per capita) would be evenly distributed among the entire population. Population numbers correspond to UN World Population Prospects. The regional differences in projected poverty impacts reflect underlying structural and economic disparities. In Sub-Saharan Africa, the projected increase in poverty is driven by a combination of high baseline poverty, demographic pressures, limited economic diversification, and strong dependence on climate-sensitive sectors such as rain-fed agriculture. These structural vulnerabilities are compounded by constrained adaptive capacity, as documented in previous research (Hallegatte and Rozenberg, 2017; Dang et al., 2024) and are already reflected in the regional GDP projections used in this analysis. In contrast, regions such as Europe and Central Asia (ECA) and high-income countries exhibit minimal changes in projected poverty levels. This is largely due to 10 their lower baseline poverty rates, more diversified economies, stronger institutional capacity, and broader access to social protection systems. Moreover, the use of the international poverty line of $2.15 per day (2017 PPP) means that relatively few individuals in these regions are close to the threshold, resulting in limited sensitivity to income shocks at this level of welfare. 9 These patterns emphasize the importance of structural context in shaping regional exposure to climate-related poverty risks. Figure 2: Number of Poor Under Climate Change, by Region (International poverty line: $2.15, 2017 PPP) a. East Asia and Pacific b. Eastern Europe and Central Asia c. Latin America and the Caribbean d. Middle East and North Africa 9 At the $3.65-per-day poverty line, projections under a neutral distribution scenario with high emissions by 2050 suggest the number of poor could rise to 267.6 million in Sub-Saharan Africa, 30.0 million in the Middle East and North Africa, 11.7 million in Latin America and the Caribbean, 67.6 million in South Asia, 6.6 million in East Asia and the Pacific, and 1.0 million in Eastern Europe and Central Asia. At the higher $6.85-per-day poverty line, the projected figures increase to 833.1 million, 98.9 million, 40.4 million, 586.4 million, 61.9 million, and 4.9 million, respectively. 11 e. South Asia f. Sub-Saharan Africa g. High-Income Countries 12 Source: Authors’ calculations based on income per capita projections from Burke, Hsiang, and Miguel (2015) and country level household survey data from the Poverty and Inequality Platform (PIP). Note: Simulations assume economic growth (GDP per capita) would be evenly distributed among the entire population. It is important to emphasize that the poverty projections presented here are driven exclusively by temperature-induced impacts on GDP per capita but do not incorporate assumptions about structural adaptation or mitigation. As such, the analysis abstracts from a wide range of potential responses to climate change—including infrastructure investments, behavioral adjustments, or policy interventions—that could alter the economic consequences of warming. The projections also do not account for other climate-related risks such as extreme weather events, disease outbreaks, or food system disruptions, which may further influence poverty but are not captured in aggregate GDP projections. While this modeling approach ensures internal consistency and tractability, it should be interpreted as providing a conservative estimate of climate-related poverty impacts. This conservatism holds despite the optimistic growth trajectory assumed under both climate change scenarios, which may understate the poverty risks in the absence of robust adaptation and mitigation efforts. 5. The Inequality Dimension of the Climate-Poverty Nexus Climate change is also expected to worsen global inequality. It is not just between-country inequality (Figure 3), which is expected to worsen as colder and richer countries are forecasted to become wealthier and the hotter poorer countries are projected to become poorer, but within-country inequality is also expected to worsen. For example, Palagi et al (2022) find that climate change, through extreme precipitation events, is poised to intensify within-country income inequality, particularly in economies with high agricultural dependence, such as those in Africa, where a 1.5- SD increase in precipitation could have a 35-fold stronger impact on the bottom income shares compared to less agriculturally reliant countries. Nevertheless, there are others who find channels, such as migration, which may reduce income inequality. Shayegh (2017) suggests that climate change-induced migration in developing countries could reduce income inequality, as skilled individuals, with greater migration opportunities, experience increased returns on skill acquisition, leading to a potential reduction in local income inequality through enhanced education investment and fewer children among non-migrating low-income individuals – the so-called quantity-quality trade-off. Overall, the poor are expected to fare worse as they have fewer coping mechanisms and lack the human capital necessary to adequately respond to shocks. 13 Figure 3: Between country inequality (Gini) 90 77.0 80 69.4 70 69.1 (country-level population weighted) 60 50 Gini Index 40 40.5 30 SSP5-RCP8.5 High Emission 20 SSP5 Baseline 10 0 2020 2030 2040 2050 2060 2070 2080 2090 2100 Year Source: Authors’ calculations based on income per capita projections from Burke, Hsiang, and Miguel (2015) and country level household survey data from the Poverty and Inequality Platform (PIP). Note: The numbers are calculated at the national level, meaning they do not account for differences within countries. The projections, based on Burke, Hsiang, and Miguel (2015), show that hotter countries are likely to fare much worse than cooler ones. Since there is a strong correlation between temperature and income levels, this disparity is expected to deepen inequality between economies. To simulate changes in within-country inequality, this analysis applies a convex Growth Incidence Curve (GIC) approach directly to the welfare distributions in the Poverty and Inequality Platform (PIP) household survey data. This method enables distributional adjustments that reflect hypothetical changes in the Gini coefficient while avoiding restrictive parametric assumptions such as log- normality. 10 To simulate the joint effects of climate-induced income changes and evolving within-country inequality, this study applies a redistribution-adjusted growth rule as specified in Lakner et al. (2022). The starting points for the GIC-adjusted simulations for a given year are the resulting means from Eq 1: � . The welfare vector is modified to arrive at the Gini adjusted welfare vector : = (1 + Δ ) − Δ ( � ) (2) 10 Appendix 2 describes a parametric approach to account for changes in the income distribution more explicitly. Instead of using a growth incidence curve, a log-normal welfare distribution is assumed. The estimates displayed in Figure A1 suggest that an annualized 1 percent increase in the Gini coefficient by 2050 could result in nearly 3.3 million more people living in extreme poverty by 2030 and 3.3 million more by 2050, compared to a scenario where within-country income inequality remains unchanged – i.e. the neutral distribution scenario. The situation deteriorates further with a 10 percent increase in the Gini coefficient by 2050, potentially leading to approximately 38.1 million more people in extreme poverty by 2050. The wide range of the projected number of poor individuals underscores the uncertainties 14 inherent in these projections, even without accounting for the uncertainty in projected GDP figures. where Δ is the annualized percentage change in the Gini necessary to arrive at a change of Gini of ∈ [0,1] by year 2050. 11 In essence, individuals below the mean experience a decline in welfare, while those above the mean see an increase—yet average welfare remains unchanged. Assuming that everyone’s income is increased at a rate , the national GDP per capita growth rates in each the SSP5 Baseline scenario and the SSP5-RCP8.5 High Emission scenario are used to estimate between 2023 and 2050. Compared to traditional parametric approaches, this GIC approach operates directly on observed microdata and is agnostic to the underlying distributional form. This transformation maintains the rank order of the distribution while altering the relative income shares to reflect greater inequality. It allows inequality to be adjusted in a transparent, policy-relevant manner without compromising empirical realism. Importantly, the shape of the convex GIC reflects emerging evidence that climate shocks tend to disproportionately impact lower-income groups (Palagi et al., 2022), making it particularly well-suited to simulate distributional changes under climate stress. Poverty simulations under climate change scenarios could vary significantly as national income inequality worsens. The estimates suggest that increasing within-country inequality may lead to considerably worse outcomes (Figure 4). Specifically, an annualized one percent increase in the Gini coefficient by 2050 could result in nearly 5.4 million more people living in poverty by 2030 and 8.8 million more by 2050, compared to a scenario where within-country income inequality remains unchanged – i.e. the neutral distribution scenario. The situation deteriorates further with a 10 percent increase in the Gini coefficient by 2050, potentially leading to approximately 148.8 million more people in extreme poverty by 2050. The wide range of the projected number of poor individuals 11 The maximum change modeled here is an increase in Gini of 20 percent. 15 underscores the uncertainties inherent in these projections, even without accounting for the uncertainty in projected GDP figures. Figure 4: Number of poor ($2.15, 2017 PPP) under SSP5-RCP8.5 High Emission Scenario (changes in Gini using a convex GIC) Source: Authors’ calculations based on income per capita projections from Burke, Hsiang, and Miguel (2015) and country level household survey data from the Poverty and Inequality Platform (PIP). Note: Each line represents a one percent increase in the national Gini index between 2022 and 2050, starting from 1 and ending in 20 percent (maximum worsening assumed in darkest line). The total increase in inequality is annualized over the study period. 6. Limitations of the Study One important limitation of the analysis is the omission of demographic reweighting in the simulations. By holding survey weights constant, the approach focuses solely on the income channel through which climate change affects poverty. While this simplifies attribution and ensures consistency in the comparison across scenarios, it does not account for potential demographic 16 transitions—such as migration, changes in age structure, or occupational shifts—that could influence poverty trajectories. 12 While some studies incorporate projections of such structural changes to assess how they affect vulnerability to climate impacts (Hallegatte and Rozenberg, 2017; Rozenberg and Hallegatte, 2015; World Bank, 2016), doing so requires detailed, forward-looking projections of demographic and labor market indicators. This data is not consistently available across countries and is therefore beyond the scope of the present analysis. As a result, these dynamics are not captured in the simulations, and their omission represents a limitation. Prior work—such as Shock Waves (Hallegatte et al., 2016) and Hallegatte and Rozenberg (2017)—has shown that demographic and socioeconomic transitions can critically shape the poverty impacts of climate change. Future extensions of this framework could incorporate demographic reweighting to explore how evolving population and labor market structures interact with climate-related risks. This analysis focuses on a high-emissions trajectory (SSP5-RCP8.5) and its associated baseline; however, it does not include an intermediate emissions pathway such as SSP2-RCP4.5. The inclusion of a mid-range scenario would provide a more realistic counterfactual and allow for a broader range of poverty projections under varying levels of mitigation effort. However, consistent regional GDP projections under alternative climate pathways are not readily available using the same modeling framework. Future work could address this gap by incorporating intermediate pathways to better reflect plausible emissions trajectories and policy scenarios. The projections in this study are based on the temperature–growth relationships estimated by Burke et al. (2015). While this framework provides a tractable and widely used approach to assessing climate impacts on GDP, it is subject to several important limitations. It focuses solely on long-run temperature effects, excluding other climate stressors such as precipitation variability, extreme events, or sea level rise. The model also operates at the national level and does not account for spatial heterogeneity or adaptation responses, which may attenuate or exacerbate the projected impacts. More recent studies—including Kotz et al. (2024), Waidelich et al. (2024), and Nath et al. (2024)—offer alternative modeling approaches that address some of these concerns by incorporating subnational variation, multiple climate stressors, and endogenous adaptation mechanisms. As such, the poverty impacts presented in this paper should be interpreted as conservative estimates, conditional on the assumptions embedded in the Burke et al. (2015) framework. It is also important to underscore that the results presented in this paper are not precise forecasts, but indicative scenarios based on a set of stylized assumptions. Multiple layers of uncertainty shape the projections, including the trajectory of global emissions, the magnitude of climate-induced GDP impacts, the evolution of income inequality, and countries’ adaptive capacity. The projections rely 12 Demographic reweighting alone does not capture deeper structural transitions such as shifts from rural to urban areas, changes in employment sectors, or education attainment. Rather, it adjusts household weights to ensure that the survey sample aligns with projected demographic aggregates (e.g., age, gender, location) based on UN population forecasts. Additionally, the act of reweighting can mechanically affect poverty and inequality estimates—independently of any macroeconomic or climate-related shocks—by changing the composition of the surveyed population. 17 on macroeconomic models that simplify complex dynamics and do not account for all channels of climate vulnerability or policy responses. As such, the findings should be interpreted with caution and viewed as illustrative of the potential range of poverty outcomes under different climate and inequality scenarios. The projections presented here also differ substantially from projections undertaken in the Country Climate and Development Reports (CCDR). 13 Under the CCDR income projections are often accompanied by other information that allows for a, perhaps, more nuanced projection of poverty within each country. 14 The poverty projections in this study differ from the CCDRs primarily because it relies on the Burke et al. (2015) temperature-based macroeconomic forecasts, which assume limited adaptation and focus solely on long-run GDP impacts from temperature changes. By contrast, CCDRs typically incorporate country-specific macroeconomic models, consider a broader range of physical climate risks (including extreme events), and include detailed policy-based adaptation and mitigation pathways. The analysis in this study is complementary: it provides a standardized global benchmark under business-as-usual assumptions, whereas CCDRs provide context-specific scenarios grounded in local development plans, sectoral strategies, and more recent growth forecasts. As with any forward-looking analysis, there are several sources of uncertainty in the projections presented in this report. These uncertainties arise from the unpredictable nature of climate change itself as well as the distributional assumptions made. One key source of uncertainty is the rate of future economic growth. The projections in this report rely on GDP forecasts derived from Burke, Hsiang, and Miguel (2015), which assume certain patterns of economic development. However, these forecasts are inherently uncertain, as future growth will be influenced by a range of factors, including technological innovation, political stability, and global trade patterns. If growth rates deviate from these assumptions, the poverty outcomes could differ significantly from the projections. In addition, there is significant uncertainty around the future trajectory of climate change itself. The projections in this report are based on the RCP8.5 and SSP5 scenarios, which represent high- emissions futures. However, the actual path of global emissions will depend on the success of international efforts to mitigate climate change. If global emissions are reduced more rapidly than expected the impacts of climate change on poverty could be less severe than projected. Conversely, if emissions continue to rise unchecked, the poverty impacts could be even worse than anticipated. 13 The World Bank’s Country Climate and Development Reports (CCDRs) provide in-depth analyses of how climate change impacts countries' development prospects and outline actionable strategies to promote sustainable, resilient growth. 14 The Country Climate and Development Reports (CCDRs) typically use a micro-simulation framework combined with the CGE model (MANAGE) or a Macro-Fiscal Model (MFMOD) to project poverty. This top-down macro-micro simulation approach, based on Bussolo, de Hoyos, and Medvedev (2010), distributes macroeconomic shocks across populations using household survey data. It incorporates sectoral employment and wage dynamics from the macro model, alongside demographic changes, to simulate the impact of macroeconomic factors on poverty and inequality. The method involves six steps: calibrating sample weights, projecting consumption, aligning labor shares with macroeconomic projections, estimating income for dependent and self-employed workers, calculating 18 household income and consumption, and ultimately projecting poverty rates and inequality indices. Finally, there is uncertainty around the role of adaptation and mitigation measures. While this study assumes limited adaptation, there is the potential for governments, communities, and households to take proactive steps to reduce their vulnerability to climate change. For example, investments in climate-resilient infrastructure, agricultural technology, and social protection systems could help mitigate some of the negative impacts of climate change on poverty. However, the effectiveness and scale of these interventions remain uncertain. 7. Concluding Remarks This paper presented global and regional projections of the potential poverty impacts of climate change through 2050, using a macro-to-micro simulation approach that links temperature-driven GDP shocks to household-level income and consumption data. The findings suggest that under a high-emissions, limited-adaptation scenario (SSP5-RCP8.5), climate change could significantly slow progress in poverty reduction—particularly in regions with high baseline poverty, low adaptive capacity, and strong dependence on climate-sensitive sectors. The projections show that climate-induced income losses alone could push an additional 41 million people into extreme poverty by 2050. When accounting for potential increases in inequality, the number of additional poor could rise by up to 148.8 million, under the assumption that every country’s Gini coefficient increases by 10 percent between 2022 and 2050. These impacts are concentrated in Sub-Saharan Africa, South Asia and Latin America and the Caribbean, where the effects of climate change are likely to interact with structural vulnerabilities and weaker social protection systems. The results presented in this study are intentionally conservative in scope. By relying on the Burke et al. (2015) estimates, the analysis focuses solely on the long-run effects of rising temperatures on economic growth and associated poverty impacts, without accounting for broader and potentially more acute channels through which climate change may affect human well-being. Moreover, the SSP5 baseline assumes sustained economic growth and limited global mitigation, potentially understating the scale of future poverty in scenarios where growth is slower or climate impacts more severe. These limitations underscore the importance of interpreting the results as indicative of the incremental poverty that could arise from unmitigated temperature increases alone highlighting the need for both mitigation and adaptation to safeguard progress in poverty reduction. In addition to the global and regional estimates presented in this study, country-level poverty projections have been generated for all countries included in the analysis. These estimates are intended to serve as reference points for long-term climate risk discussions and may be particularly useful for World Bank country teams and policy dialogue. While these projections are based on stylized assumptions and should be interpreted with caution, they offer a consistent “off-the-shelf” benchmark for understanding the potential magnitude of climate-induced poverty in the absence of strong mitigation or adaptation. These country-level results will be made available as supplementary material or upon request. Importantly, they are not intended to replace detailed country-specific 19 modeling exercises—such as those conducted under the Climate Change and Development Reports (CCDRs)—but rather to complement them by providing a standardized view of global risks under a high-emissions, limited-adaptation scenario. This paper also aims to support both policy and research communities by offering projections that can be used in complementary ways. For operational teams and World Bank staff, the results provide a globally consistent benchmark to inform climate vulnerability assessments, CCDRs, and long-term strategic planning. For researchers and technical audiences, the analysis highlights key methodological challenges—such as climate-specific passthrough modeling, inequality dynamics, and demographic transitions—and identifies areas for future research to enhance predictive accuracy and relevance. The findings of this paper underscore the urgent need for coordinated global action to mitigate the poverty impact of climate change. Without significant policy interventions, climate change is likely to push millions of people into poverty, particularly in low- and middle-income countries. The most vulnerable populations, including those in Sub-Saharan Africa and Southeast Asia, will bear the brunt of these impacts, further entrenching global inequality. Addressing the poverty impacts of climate change requires a multi-faceted approach. First, it is essential to strengthen social safety nets and provide targeted support to the poorest and most vulnerable populations. This includes expanding access to healthcare, education, and financial services, as well as improving the resilience of livelihoods through investments in agriculture, infrastructure, and climate adaptation measures. Second, efforts to reduce income inequality must be central to any poverty reduction strategy. As the projections in this report show, even small increases in inequality can lead to large increases in poverty. Policies that promote inclusive economic growth, reduce barriers to education and employment, and enhance social protection systems will be critical for ensuring that the benefits of economic development are shared more broadly. Finally, international cooperation will be essential for addressing the global nature of the climate challenge. High-income countries, which have contributed the most to global emissions, have a responsibility to support low- and middle-income countries in their efforts to adapt to climate change. This includes providing financial resources, technology transfer, and capacity-building support to help countries build resilience to climate shocks and transition to low-carbon development pathways. 20 References Ahmed, S., Diffenbaugh, N., and Hertel, T. (2009). Climate Volatility Deepens Poverty Vulnerability in Developing Countries. Environmental Research Letters, 4, 034004. https://doi.org/10.1088/1748- 9326/4/3/034004. Barbier, E., and Hochard, J. (2018). The Impacts of Climate Change on the Poor in Disadvantaged Regions. Review of Environmental Economics and Policy, 12, 26 - 47. https://doi.org/10.1093/reep/rex023. Blanz, A., and Kalkuhl, M. (2022). Climate Change and Long-Run Inequality: The Case of Food Prices in The Developing World. ETH Working Papers. Burke, M., Hsiang, S., and Miguel, E. (2015). Global Non-Linear Effect of Temperature on Economic Production. Nature, 527(7577), 235–239. Bussolo, M., de Hoyos, R. E., and Medvedev, D. (2010). Economic Growth and Income Distribution: Linking Macro-Economic Models with Household Survey Data at The Global Level. International Journal of Microsimulation, 3(1), 92–103. Caruso, G. D., Lucchetti, L. R., Malasquez, E., Scot, T., and Castaneda, R. A. (2017). But? what is the poverty rate today? testing poverty nowcasting methods in Latin America and the Caribbean (No. 8104). The World Bank. Dang, H.H., Hallegatte, S., and Trinh, T.-A. (2024a). Does Global Warming Worsen Poverty and Inequality? An Updated Review. Journal of Economic Surveys, 1–33. https://doi.org/10.1111/joes.12636 Dang, H. H., Hallegate, S., Nguyen, M. C., and Trinh, T. A. (2025). Does Hotter Temperature Increase Poverty and Inequality? Global Evidence from Subnational Data Analysis, Papers 2025-08, Centre for Health Economics, Monash University. Dell, M., Jones, B. F. and Olken, B. A. What Do We Learn from The Weather? The New Climate-Economy Literature. J. Econ. Lit. 52, 740–798 (2014). Diffenbaugh, N. S., and Burke, M. (2019). Global Warming Has Increased Global Economic Inequality. Proceedings of the National Academy of Sciences, 116(20), 9808-9813. Gilli, M., Calcaterra, M., Emmerling, J., and Granella, F. (2024). Climate Change Impacts on The Within-Country Income Distributions. Journal of Environmental Economics and Management, 127, 103012. Hallegatte, S., Fay, M., and Barbier, E. (2018). Poverty and Climate Change: Introduction. Environment and Development Economics, 23, 217 - 233. Hallegatte, S., and Rozenberg, J. (2017). Climate Change Through a Poverty Lens. Nature Climate Change, 7(4), 250–256 Hallegatte, S., Bangalore, M., Bonzanigo, L., Fay, M., Kane, T., Narloch, U., Rozenberg, J., Treguer, D., and Vogt- Schilb, A. (2016). Shock Waves: Managing The Impacts of Climate Change on Poverty. World Bank Hope, K. R. (2009). Climate Change and Poverty in Africa. International Journal of Sustainable Development and World Ecology, 16(6), 451–461. https://doi.org/10.1080/13504500903354424 21 Jafino, B. A., Walsh, B., Rozenberg, J., and Hallegatte, S. (2020). Revised Estimates of The Impact of Climate Change on Extreme Poverty By 2030. The World Bank. Karver, J., C. Kenny, and A. Sumner. 2012. MDGS 2.0: What Goals, Targets, and Timeframe? IDS Working Papers 2012 (398): 1–57. doi:10.1111/j.2040- 0209.2012.00398.x. Kotz, M., Levermann, A. and Wenz, L. (2024) The Economic Commitment of Climate Change. Nature 628, 551– 557. Kotz, M., Levermann, A., and Wenz, L. (2022). The Effect of Rainfall Changes on Economic Production. Nature, 601(7892), 223-227. Lakner, C., Mahler, D. G., Negre, M., and Prydz, E. B. (2022). How Much Does Reducing Inequality Matter for Global Poverty?. The Journal of Economic Inequality, 20(3), 559-585. Lee, M., Villaruel, M., and Gaspar, R.E. (2016). Effects of Temperature Shocks on Economic Growth and Welfare in Asia. Socially Responsible Investment eJournal. Mahler, D., Yonzan, N., and Lakner, C. (2022) The Impact of COVID-19 on Global Inequality and Poverty. Policy Research Working Papers; 10198.  World Bank, Washington, DC.  Last accessed at: https://openknowledge.worldbank.org/entities/publication/54fae299-8800-585f-9f18-a42514f8d83b License: CC BY 3.0 IGO.” updated with Poverty and Inequality Platform (September 2024). Nath, I. B., Ramey, V. A., and Klenow, P. J. (2024). How Much Will Global Warming Cool Global Growth? (No. w32761). National Bureau of Economic Research. Onyeaka, H., Nwauzoma, U., Akinsemolu, A., Tamasiga, P., Duan, K., Al-Sharify, Z., and Siyanbola, K. (2024). The ripple effects of climate change on agricultural sustainability and food security in Africa. Food and Energy Security. Paglialunga, E., Coveri, A., and Zanfei, A. (2022). Climate Change and Within-Country Inequality: New Evidence from A Global Perspective. World Development, 159, 106030. Palagi, E., Coronese, M., Lamperti, F., and Roventini, A. (2022). Climate Change and The Nonlinear Impact of Precipitation Anomalies on Income Inequality. Proceedings of the National Academy of Sciences, 119(43), e2203595119. Prydz, E. B., D. M. Jolliffe, C. Lakner, D. G. Mahler, P. Sangraula, National Accounts Data Used in Global Poverty Measuremen, Global Poverty Monitoring Technical Note; no. 8. Washington, D.C. : World Bank Group. 2019. Seaman, J.E., G.E. Sawdon, J. Acidri, and C. Petty. (2014) The Household Economy Approach. Managing the impact of climate change on poverty and food security in developing countries. Climate Risk Management, Volumes 4–5, Pages 59-68, ISSN 2212-0963, https://doi.org/10.1016/j.crm.2014.10.001. Shayegh, S. (2017). Outward Migration May Alter Population Dynamics and Income Inequality. Nature Climate Change, 7(11), 828-832. UNDESA - United Nations, Department of Economic and Social Affairs, Population Division (2024). World Population Prospects 2024, Online Edition. Waidelich, P., Batibeniz, F., Rising, J., Kikstra, J. S., and Seneviratne, S. I. (2024). Climate damage projections beyond annual temperature. Nature Climate Change, 14(6), 592-599. 22 Waldhoff, S., Wing, I., Edmonds, J., Leng, G., and Zhang, X. (2020). Future Climate Impacts on Global Agricultural Yields Over The 21st Century. Environmental Research Letters, 15. https://doi.org/10.1088/1748- 9326/abadcb. Wildowicz-Szumarska, A., and Owsiak, K. (2024). Impact Of Climate Change on Income Inequality. Implications for rich and poor countries. Economics and Environment. Wollburg, P., Hallegatte, S., and Mahler, D.G. Ending extreme poverty has a negligible impact on global greenhouse gas emissions. Nature 623, 982–986 (2023). 23 Appendix 1. The Binned PIP Data As of June 2024, the World Bank’s Poverty and Inequality Platform (PIP), formerly known as PovcalNet, provides the most recent and standardized estimates of global poverty indicators. The PIP is based on harmonized multitopic household income and expenditure surveys, which serve as the primary data source for measuring monetary poverty. These surveys are compiled in the Global Monitoring Database (GMD), with data typically collected by national statistical offices. The GMD surveys are processed and harmonized by the Data for Goals (D4G) team, in coordination with the six regional statistics teams of the Poverty and Equity Global Practice. The Global Poverty and Inequality Data (GPID) team within the Development Economics Data Group (DECDG) contributes additional historical data, including pre-1990 estimates and recent surveys from the Luxembourg Income Study (LIS). To support cross-country and temporal comparability, key sociodemographic and welfare variables are harmonized to the extent feasible. Despite efforts to standardize survey instruments and definitions, residual methodological differences may lead to discrepancies between PIP estimates and those reported in other sources. Although country coverage varies across survey years, the alignment of survey-year estimates to a standardized reference year allows for consistent cross-sectional and longitudinal analysis of monetary poverty across a stable set of countries. 1 Measures of economic growth are also recovered. The previous extrapolation method for global poverty estimation adjusted the welfare measures obtained from household surveys by a common scale factor, which was equal to a measure of growth derived from national accounts data: growth in real Household Final Consumption Expenditure per capita (referred to as 'growth in HFCE') for countries outside Sub-Saharan Africa, and growth in real GDP per capita (referred to as 'growth in GDP') for countries in Sub-Saharan Africa (Prydz et al. 2019). The new method recommends using HFCE data for upper-middle-income and high-income countries and GDP data for low-income and lower-middle-income countries. Additionally, it introduces two different "passthrough" rates. A passthrough rate is a factor that determines the share of growth from national accounts that "passes through" to the survey welfare measure. For countries using consumption aggregates to calculate growth in welfare, the passthrough rate is 0.7, while the rate is 1 for countries using income aggregates. 1 For detailed information on the line-up method, see the PIP methodological handbook. The line-up method uses growth in national accounts to extrapolate and interpolate household income/consumption from the survey years, as described in Mahler and Newhouse (2024) and the PIP methodological handbook which is available at https://datanalytics.worldbank.org/PIP-Methodology/. Appendix 2. An Alternative Approach to Incorporating Changes in Inequality An additional approach undertaken incorporates changes in inequality explicitly. This is done under the assumption that welfare is log-normally distributed. Under this assumption, the Gini is given by (Crow and Shimizu,1987): = 2Φ � �−1 √2 The Gini is then assumed to change by x percent by 2050. This change is annualized and is given by: = 0 (1 + ) 1 where = (1 + )2050−2022 − 1. With the corresponding Gini value for each year and for each country, it is then necessary to back out the corresponding value of for each year. Then all that is needed is the average of the natural log of welfare ( ) from Eq. 1 for each year to obtain the corresponding poverty rate: ln () − = Φ� � (2) To ensure comparability with the neutral distribution scenario (where the Gini remains unchanged), the adjusted Gini-based poverty estimates are aligned with the neutral distribution scenario estimates. This is achieved by computing the ratio of the neutral distribution projections to the projections from Equation A2 under a no-change-in-Gini assumption. Each adjusted Gini poverty rate is then multiplied by this ratio, aligning the estimates with those of the neutral distribution scenario. Figure A1: Number of poor ($2.15, 2017 PPP) under SSP5-RCP8.5 High Emission Scenario (annualized changes in Gini) Source: Authors’ calculations based on income per capita projections from Burke, Hsiang, and Miguel (2015) and country level household survey data from the Poverty and Inequality Platform (PIP). Note: Each line represents a one percent increase in the national Gini index between 2022 and 2050, starting from 1 and ending in 20 percent (maximum worsening assumed in darkest line). The total increase in inequality is annualized over the study period. Appendix 3. Regional GDP Growth Projections Average Annual GDP per Capita Growth (percent), 2023-2050 SSP5-RCP8.5 Region SSP5 Baseline High Emission East Asia and Pacific EAP 5.44 4.54 Europe and Central Asia ECA 3.68 4.09 Latin America and Caribbean LAC 4.5 3.45 Middle East and North Africa MNA 4.47 3.45 High-Income Countries OHI 2.02 1.83 South Asia SAS 6.12 5.08 Sub-Saharan Africa SSA 6.79 5.46 Note: Projections based on Burke et al. (2015) temperature–growth estimates.