This is the seventh blog in a series about how countries can make progress on the interlinked objectives of poverty, shared prosperity and a livable planet. For more information on the topic, read the 2024 Poverty, Prosperity, and Planet Report.
"Without data, you're just another person with an opinion.”
Data plays a pivotal role in ending poverty and boosting shared prosperity on a livable planet. Without data, policymakers cannot fully know the incidence, depth, or profiles of poverty, inequality, and vulnerability, nor the changing trends in these variables, and they will be unable to prioritize accordingly and plan effectively for the years ahead. The recently published Poverty, Prosperity, and Planet Report 2024 underscores data and solid evidence as the infrastructure for policy.
Household survey data, which are available in the World Bank’s Poverty and Inequality Platform (PIP), continue to be at the core of monitoring of the Sustainable Development Goals (SDGs), as well as the World Bank’s vision. Several SDGs rely heavily on survey data to monitor progress, assess needs, and evaluate the effectiveness of policies and interventions. These surveys are behind the construction of more than one-third of the 234 SDG indicators. Yet data deprivations in household income and expenditure surveys present challenges, as we will explore in this blog.
Despite global progress in household survey data availability, low-income settings still lag behind
Overall, the availability of household survey data has improved in many countries. There has been substantial progress in the availability of household survey data containing information on income, consumption, or both, which allows better tracking of progress toward SDG 1 (no poverty) and SDG 10 (reduced inequalities). Globally, between 1998 and 2022, the average number of available household income and expenditure survey datasets per country increased from 2.1 to 9.9, almost a fivefold increase (Figure 1). Upper-middle- and high-income countries drove this progress. More survey data have also become available for lower-income countries, with improvements in data quality, frequency, and timeliness.
However, significant data gaps remain. Although more than three-quarters of the world’s population are represented by survey data available in the Poverty and Inequality Platform (PIP) collected in 2020 or later, fewer than one-half of countries had survey data from the same period (Figure 2). This means that data availability is more limited in less populous countries. The limited availability of data reflects issues related to infrequent data collection, the lack of statistical capacity, fragile contexts, or delays in sharing the data.
The lack of survey data is particularly pronounced in settings where the need for knowledge on the levels and distribution of welfare is greatest. Less than one-half of the population in low-income, fragile countries, Small States, and countries supported by the International Development Association (IDA) have been covered by a survey since 2020. These countries have consistently had the least number of survey data since 1998 (Figure 1), and the pace of progress is slow compared with that of richer countries. Moreover, Sub-Saharan Africa and the Middle East and North Africa do not have sufficient data coverage for global poverty monitoring in recent years and therefore rely on poverty nowcasts largely based on data from before the COVID-19 pandemic.
More comprehensive welfare and livable planet metrics require more and better data
Filling the gaps in survey data is particularly relevant in light of the new, expanded World Bank vision calling for a more holistic approach to measuring well-being The World Bank’s current data on monetary well-being has been compiled primarily to measure poverty and the well-being of those at the bottom part of the distribution. The expanded vision also considers inequality and the Prosperity Gap, which capture the entire welfare-distribution.
Even when data exist, comparability over time and between countries is not always a given. For instance, countries in Latin America and the Caribbean typically use income data while those in Sub-Saharan Africa use consumption data to measure welfare. These two regions stand out as having high levels of inequality (which we’ll discuss in the next blog in this series), but the differences in their underlying welfare measures make it difficult to compare their levels of inequality and answer the question of which is more unequal. Other issues include the underreporting of top incomes and survey non-response, which could bias inequality estimates. In high-income contexts where administrative data are more readily available, tax data have been used to “correct” for this underreporting, but the best method to do this is still unclear, especially for countries without administrative data.
The World Bank’s new vision indicators are data-intensive in nature, expanding the scope to non-monetary indicators of well-being, which broaden our knowledge, for instance, on the overall living conditions of people or their ability to cope in the event of climate-related risks. These new indicators require very different and detailed types of data, such as global greenhouse gas emissions, hectares of key ecosystems, the percentage of people at high risk of climate-related hazards, and deprivations of people in access to food, nutrition, basic drinking water, sanitation services, or hygiene.
To optimize data use along all of these dimensions in the coming decade, Carletto et al. (2022) have identified priority areas for improving survey data. These include: enhancing the interoperability and integration of household surveys; designing and implementing more inclusive, respondent-focused surveys; improving sampling efficiency and coverage; expanding the use of objective measurement technologies; building capacity for computer-assisted personal interviewing, phone, web, and mixed-mode surveys; systematizing the collection, storage, and use of paradata and metadata; incorporating machine learning and AI for data quality control and analysis; and improving data access, discoverability, and dissemination.
Projections into the future come with large uncertainty, which is aggravated by lack of reliable foundational data
Clearly, data demands are increasing, thus making the data gaps larger and more pressing, especially as we look into a highly uncertain future. Figure 3, for example, shows a large variation in forecasts of average global GDP per capita under different Shared Socioeconomic Pathways (SSP). These scenarios are used extensively for depicting various pathways of how the world economy and climate could evolve in the future. GDP per capita has been shown to be a very robust predictor of poverty, and if countries’ incomes grow in accordance with the most optimistic scenarios, then global extreme poverty will be eliminated within decades. However, based on current trends, particularly in low-income countries, it appears unlikely that extreme poverty will fall drastically in the coming decades.
The uncertainty in projections increases when accounting for the interdependence of poverty, prosperity, and planetary indicators. While there is scientific consensus that global warming—caused by human-made emissions—has negative consequences for economic growth and poverty reduction, the magnitude remains uncertain, especially if so-called tipping points are reached. Therefore, it is less clear how poverty will change going forward, which largely depends on the success of mitigation and adaption efforts, and whether low-income countries can foster long-term inclusive growth. What is certain is that projections on future outcomes—be it welfare- or climate-related metrics, or the combination of both—hinges on a foundation of high-quality data today.
Figure 3: Projections of GDP per capita under different Shared Socioeconomic Pathways
Investments are needed to close data gaps
Policy makers working to alleviate poverty, build resilience, and promote sustainable well-being need accurate data to make informed decisions, particularly in an environment with increasing uncertainty, misinformation, and limited budgets. The World Bank has been playing a key role in generating open and high-quality data, for instance by providing openly-accessible development data and geospatial data platforms. The Poverty, Prosperity, and Planet Report 2024, for example, comes with a full reproducibility package, consisting of the underlying data and estimation code.
More investments are needed to produce reliable, granular, and timely household surveys, especially for low-income countries. These surveys remain the foundation for impactful, evidence-based policy making. IDA financing has been instrumental in closing data gaps for some of the poorest countries in the world, particularly in Africa. Of the 21 IDA countries that collected data available for poverty monitoring after 2021, 15 have benefitted from ongoing World Bank statistical capacity-building operations. Continued support from a strong IDA replenishment is thus essential for enhanced data availability in low-income settings, and thereby for improving people’s lives.
The authors gratefully acknowledge financial support from the UK Government through the Data and Evidence for Tackling Extreme Poverty (DEEP) Research Program.
Join the Conversation