Poverty Headcount Research Articles

Foreign Aid and Poverty Reduction in West African Nations: Insights and the Need for Policy Re-Orientation

Poverty, a deteriorating healthcare system and insufficient educational standards are among the major macroeconomic problems facing African countries. Foreign aid from Western countries to African countries has been hailed as one of the solutions to these problems. Thus, this study investigated the impact of foreign aid on poverty using a panel data set of 14 West African nations. Foreign aid was disaggregated according to the reason for which it was given while poverty was measured with three variables namely; poverty headcount, infant mortality, and literacy rate. Annual data was sourced from the WDI (2022), WGI (2022) and OECD (2022), scoped 2008-2020. Using a two-step system GMM, the study found that total aid insignificantly reduces poverty headcount in the presence of reduced corrupt practices. Furthermore, foreign aid allocated to health and education was found to lower infant mortality and improve literacy rates, respectively; however, these effects were only significant in the condition of reduced corrupt practices. The study thus concludes that foreign aid works better when corrupt practices are reduced, and it recommends that policymakers in west- African countries should consider corruption control policies so that aid allocation is properly managed to improve its impact on poverty headcount, infant mortality, and literacy rate.

Understanding Vulnerability to Poverty, COVID-19’s Effects, and Implications for Social Protection: Insights from Ghana

The Coronavirus pandemic has created new vulnerabilities and deepened existing ones. In Ghana, the overall headcount poverty decline has disguised a large group of vulnerable households now threatened with falling back into poverty given the effects of the COVID-19 pandemic. This article draws on new household panel data to analyse transient poverty, the drivers of vulnerability before and since the COVID-19 pandemic and provides implications for social protection. Using poverty transition matrices, we first measure poverty episodes which help to not only track the movements in and out of poverty but also to identify and categorize households into distinctive groups. We then apply logistic regression to examine the determinants of vulnerability to poverty before evaluating COVID-19 vulnerabilities in the context of social protection in Ghana. We find that a large group of struggling households exist between the poor and the stable middle classes that have been overlooked by cash transfer programmes. We characterize this group and show that current vulnerability measurements and social protection design need adjustments. We also find that poverty and vulnerability to it is no longer a rural phenomenon as the transient poor since the COVID-19 pandemic are mostly located in urban areas and largely not covered by current social protection systems. Both vertical and horizontal expansion of social policy is required to reach these new groups and new locations. For reducing poverty and limiting vulnerability in developing country contexts, our findings imply that (a) new measurements and targeting may be required, (b) vulnerable informal groups in urban areas need to be included, and (c) new types of social protection for struggling households are required to prevent their falling back into poverty in the event of aggregate shocks.

Collateral damage? Welfare effects of the Ukraine war on Pakistan

The 2022 Ukraine conflict has contributed to a major spike in international commodity prices. In this study, we conduct a top-down, macro-micro simulation analysis by combining Computable General Equilibrium simulations with data from the 2018–19 Pakistan Household Integrated Economic Survey to analyze the impact of global price shock to four major commodities, namely wheat, vegetable oil, petroleum, and fertilizers. We come up with evidence for a significant, non-negligible negative impact of the price shock, both at the aggregate and the household level: Real GDP shrinks and real household consumption and income decrease. The drop in consumption and income is more visible among farm households, with the two falling by 5% and 3.48% respectively. The consumption and income of the above-median income households fall more steeply, and income inequality decreases. The USD3.2 headcount poverty rate at the 2018 purchasing power parity increases by 1.15%. Urban households show the worst decline in food security, both in incidence and intensity. The role of petrol prices in driving poverty among rural farm households and that of food stuff (wheat and vegetable oil prices) in aggravating urban food security is particularly evident. These findings highlight the poverty and food security-worsening dimension of the Russo-Ukrainian conflict.

Analisis Pengaruh Indeks Pembangunan Manusia, Laju Pertumbuhan Penduduk, dan Tingkat Pengangguran Terbuka terhadap Kemiskinan di DKI Jakarta

The poverty rate has increased rapidly again due to the Covid-19 Pandemic. DKI Jakarta is one of the top 5 provinces with the lowest poverty rate in Indonesia, even though it has high complexity in administering its government. Because of these achievements, this study aims to find factors that influence poverty so that it can become a reference for other local governments in alleviating poverty. This research was conducted using panel data regression with a fixed effect model. The results showed that simultaneously poverty headcount index in the t-1 period, human development index (HDI), open unemployment rate, and population growth rate significantly affect the poverty headcount index. While partially, open unemployment and population growth rate positively and significantly affect the poverty headcount index. On the other hand, the poverty headcount index in the t-1 period and HDI positively and insignificantly affect the poverty headcount index. This research shows that local governments must focus more on policies to reduce unemployment and population growth rate in the context of alleviating poverty.

The Conditional Influence of Poverty, Inequality, and Severity of Poverty on Economic Growth in Sub-Saharan Africa

Poverty and inequality represent major policy syndromes that are relevant in the achievement of most United Nations’ Sustainable Development Goals (SDGs) in sub-Saharan Africa (SSA), while economic growth is also essential for the achievement of attendant SDGs. This study extends existing literature by assessing the conditional influence of poverty, income inequality, and severity of poverty on economic growth. The focus is on 42 countries in SSA with data from 1980 to 2019. The Gini index is used to measure income inequality. Poverty is measured in terms of the poverty headcount ratio, while the severity of poverty is computed as the squared of the poverty gap index. The empirical evidence is based on quantile regressions to assess how income inequality and poverty dynamics affect economic growth throughout the conditional distribution of economic growth. Our main finding shows that the negative response of economic growth to poverty is a decreasing function of economic growth. In other words, the incidence of poverty in reducing economic growth decreases with increasing levels of economic growth. In two specifications, the effect of inequality is negative in bottom quantiles and positive in top quantiles of the conditional distribution of economic growth. Policy implications are discussed, especially as it pertains to (1) the relevance of poverty in mitigating economic growth in SSA contingent on initial levels of economic growth and (2) comparative incidences of poverty and inequality in affecting economic growth.

The financial burden of out of pocket payments on medicines among households in Ethiopia: analysis of trends and contributing factors

BackgroundIn Ethiopia, out-of-pocket (OOP) payment is the key means of healthcare financing, and expenses on medicines are a crucial component of such payment. This study aims to investigate the financial implications of OOP payments on medicines for Ethiopian households.MethodsThe study involved a secondary data analysis of the national household consumption and expenditure surveys of 2010/11 and 2015/16. The "capacity-to-pay" method was used to calculate catastrophic OOP medicine expenditures. The extent of economic status related to catastrophic medicine payment inequity was calculated using concentration index estimation. The impoverishment consequences of OOP payment on medicine were estimated using poverty headcount and poverty gap analysis methods. Logistic regression models were used to identify the variables that predict catastrophic medicine payments.ResultsMedicines accounted for the majority of healthcare spending (> 65%) across the surveys. From 2010 to 2016, the total percentage of households facing catastrophic medicine payments decreased from 1% to 0.73%. However, the actual number of people expected to have experienced catastrophic medicine payments increased from 399,174 to 401,519 people. Payment for medicines pushed 11,132 households into poverty in 2015/16. The majority of disparities were explained by economic status, place of residence, and type of health services.ConclusionOOP payment on medicine accounted for the majority of total health expenses in Ethiopia. A high medicine OOP payment continued to push households into catastrophic payments and impoverishment. Household seeking inpatient care, those with lower economic status and urban residents were among the most affected. Hence, innovative approaches to improve the supply of medicines in public facilities especially those in urban settings and risk protection mechanisms for medicine expenditures particularly for inpatient care are recommended.

Poverty decompositions with counterfactual income and inequality dynamics

AbstractTraditional poverty accounting decomposes changes in a country's poverty headcount ratio into changes in income and inequality. We argue that this approach is unsatisfactory from the perspective of policy analysis because it compares a country in two points of time without taking the country's initial situation, and hence its potential for poverty reduction, into account. We thus suggest comparing traditional poverty decompositions with a counterfactual situation. This counterfactual indicates what a country starting from its initial situation could be expected to achieve in terms of income, inequality, and, hence, poverty developments. We construct those counterfactuals by modeling income and inequality trends characterized by convergence and a “Kuznets” relationship between inequality and development. Parameters in those relationships are estimated using PovcalNet survey data from 144 countries and we construct our counterfactual poverty predictions for 71 developing countries. While there is overall a tight relationship between actual developments and counterfactuals, we identify several cases, where both deviate from each other and discuss the policy implications. We also check for commonalities in differently performing countries and find that those who fell particularly short of expectations often underwent political transition and state fragility.

Causes and Measures of Poverty, Inequality, and Social Exclusion: A Review

Prevailing measures on the topics of monetary and non-monetary poverty—as well as economic and carbon inequality—are being critically assessed under sustainable development goals (SDGs) with a worldwide perspective. On the one hand, the poverty headcount ratio and the indices poverty gap, poverty severity, and Watts are assessed as core poverty indices. On the other hand, important inequality measures such as the Gini index and the Palma ratio are evaluated in order to find their potentials for policymaking. Furthermore, social exclusion (with the extreme forms of aporophobia and homelessness) has detrimental ramifications on human wellbeing and disturbs public provision policies. Thus, poverty, inequality, and social exclusion are integral parts of SDGs (1, 6 and 10), emphasizing the multidisciplinary nature of the above issues. Additionally, intrinsic elements of Agenda 2030 and the measuring of poverty, inequality, and polarization would significantly improve integrated policy pathways in the national and international fora.

Effect of Foreign Direct Investment and Trade Openness on the Poverty Alleviation in Burundi – Sub African Country: ARDL (Co-integration) Approach

International Trade and Foreign Direct Investment openness are crucial for economic growth and development. These are equally significant for developed and under-developed countries. The study’s major goal was to look at how foreign direct investment, inflation, GDP and trade openness affected the poverty alleviation in Sub-Saharan African Countries. A country where half of the population lives below the poverty. Annual data for the period 1990 to 2021 gathered from World Development Indicators database to investigate the long run & short run effects of the independent variables for the Sub-Saharan African countries by using Auto-regressive and Lag distributive (Co-integration) model. The findings revealed negative long run influence of foreign direct investment and GDP to poverty head count ratio. The shreds of Evidence also showed that FDI has not shown any distinctive effect on poverty in a shorter period. Moreover, GDP, trade openness and inflation are also effective in reducing poverty in short run. While inflation urges the investors to do more investment in production sector that reduces unemployment and poverty in the end. The study suggested the government officials, policymakers, and investors, to invest more in the poor countries. As FDI is crucial in creating jobs for the unemployed population that lead to the rise in income and living standard of the people. Moreover, the study suggested questioning some other factors causing poverty.

Multidimensional poverty is associated with dementia among adults in Afghanistan

Multidimensional poverty is associated with dementia, but no evidence is available for countries in conflict. A cross-sectional study was conducted in two provinces of Afghanistan between February 15th 2022 and April 20th 2022 among adults age 50 and older. Multidimensional poverty included six dimensions of well-being and 16 indicators of deprivation. The Rowland Universal Dementia Assessment Scale measured dementia. Poverty between adults with and without dementia was examined, adjusting for sex. Associations between dementia and poverty were investigated using multivariate regression model. Of the 478 adults included, 89 (52.7%) had mild, and 25 (14.8%) had moderate to severe dementia. More women than men had mild (52.7% vs 33.3%) and moderate-to-severe dementia (14.8% vs 5.8%). Approximately 33.9% adults with mild and 51.2% adults with moderate-to-severe dementia were found to be deprived in four or more dimensions compared to 21.8% without dementia. The difference in four dimensions of multidimensional poverty between adults with mild and moderate-to-severe dementia and adults without dementia was respectively 59.5% and 152.88%. Education, employment, health, and living conditions were the main contributors to the adjusted poverty head count ratio. Multidimensional poverty in four or five dimensions was strongly associated with dementia among older adults particularly over 70 years old (odds ratio [OR], 17.38; 95% CI, 2.22-135.63), with greater odds for older women overall (OR, 2.69; 95% CI, 1.76-4.11). Our findings suggest that early improvement in social determinants of health through targeted structural policies may lower dementia risk later in life. Specifically, better access to free, quality education, healthcare, and basic living standard together with employment opportunities could reduce risk of dementia. The present study was funded by a grant from the Alzheimer Association (AARG-NTF-21-851241).

The impact of health reform on poverty estimates in Iran: Implications for monitoring the first goal of Sustainable Development Goals.

Iran launched a series of health reforms called Health Transformation Plan (HTP) in order to improve financial protection and access to health care in 2014. This study aimed to investigate the extent of impoverishment due to out of pocket (OOP) payments during 2011-2016 and to assess the implications of health expenditures to overall national poverty rate before and after the HTP implementation, with a focus on monitoring the first Sustainable Development Goals (SDGs). The study relied on data from a nationally representative household income and expenditure survey. Two measures of poverty were estimated in this study: the prevalence (poverty headcount) and the intensity of poverty before and after OOP health payments (poverty gap). Proportion of the population fell into poverty due to OOP spending for health care from total poor population calculated for 2years before and after the HTP implementation using three poverty lines suggested by the World Bank for global poverty comparison ($1.90, $3.2 and $5.5 per day in 2011 purchasing power parity (PPP) $). Our results indicate that the incidence of impoverishing health expenditures has remainedrelatively low during 2011-2016. At the 2011 PPP $5.5 daily poverty line, the average incidence rate at the national level was 1.36% during the period. The percentage of population impoverished due to OOP health expenditures increased after HTP implementation, irrespective of the poverty line used. However, the proportion of individuals that pushed further into poverty decreased after HTP implementation. It was estimated that around 12.5% of total poor population fell below poverty line due to paying OOP payments in 2016. Although health care costs are not major causes of impoverishment in Iran, the relative impact of OOP spending for health is not negligible. To attain SDG 1, pro-poor interventions that aim to reduce the burden of OOP payments should be advocated and implemented with an inter-sectoral approach.

Towards accessibility or affordability? Multidimensional energy poverty across the South African urban–rural divide

This study combines energy affordability and accessibility into a multidimensional energy poverty measure, which we stratify by rural and urban locale. Accessibility considers a number of binary indicators related to the type of energy used for a series of household activities, while affordability is determined by the ratio of household required energy expenditure to total expenditure. We employ an equivalence scale approach to estimate household energy requirements using publicly available household expenditure survey data. Our results suggest extensive urban–rural disparities across our multidimensional indicators—37% of rural households are affordability deprived, which is nearly double urban affordability deprivation; the rural–urban differences are at least double, when it comes to clean cooking, lighting, space heating, water heating and multidimensional headcount poverty. After splitting the households by degrees of energy poverty, it is found that the extreme energy-poor are more likely to be income-poor. However, urban extreme energy-poor is driven by affordability deprivation, while more than half of the rural extreme energy-poor are deprived in both affordability and accessibility.

Food Security and Poverty: Searching for a Linkage in the Poverty Prone Areas in Bangladesh

The public and private level cross-sectional shreds of evidence on poverty are available at the national and sub-national levels in Bangladesh. Bangladesh Bureau of Statistics provides us a cross-sectional measure of poverty almost after 5 years. While food poverty has been analyzed at various levels but the discussion on food security got scanty importance. Since poverty and food security is not static and mostly dynamic, an in-depth analysis of those two issues in the Rangpur regions (poverty-prone areas) gives us insights into the understanding of long-root causes of poverty and food security. The samples of the study have been drawn from the lower segment of the population. The FGT measures show that the consumption based poverty headcount is about 73 with a poverty gap of 22.5 and a squared poverty gap of 9.0. The income based measures (USD 1.90) also show that over fifty per cent of the households have less than USD 1.90 earning a day, and nearly two-thirds of the households have less than USD 2.15 earning a day. The poverty gap and squared poverty gap are relatively large compared to the national level statistics. The near around half of the households have occasional moderate food insecurity. The study shows that regional heterogeneity, and household attributes determine the state of food security and poverty. The poverty and food insecurity are found interlinked – the low food security is associated with extreme poverty.

Financial inclusion and economic development: Turkey and Greece

Purpose- Financial inclusion means individuals and businesses have access to useful and affordable financial products and services to deliver their needs in a responsible and sustainable way. A financial sector is measured and compared on four main features; debt is the size of financial institutions, access is the access and use of financial services by the users, efficiency is the efficiency in the provision of financial services, and stability is the stability in the provision of financial services. The purpose of this paper is to measure the level of financial inclusion of Turkey and Greece from 2000 to 2020 and compare its relationship with the economic growth and income inequality of both countries. Methodology- The World Bank data covering the 2000-2020 period is extracted from Turkey and Greece from the world bank report. The whole financial system for both countries is defined as a combination of banks, nonbanks financial institutions, and stock exchange markets. The related indicators for each of the subsectors of the financial system are determined for banks, nonbanks financial institutions, and stock exchange markets. Thus, 32 indicators for banks, 6 indicators for nonbanks, and 16 indicators for stock exchange markets are determined for the financial inclusion index. All indicators are in percentages. All individual indicators are summed for the computation of subsectoral indexes and then the growth rate in each subsectoral indexes are computed. The growth rates of each subsectoral index are summed and weighted by the subsectoral asset sizes or trading volüme. Finally, the causal relationship between the financial inclusion index, Gini coefficient, Poverty Headcount ratio, and GDP per capita was examined. Findings- The average growth rate for the financial inclusion index for the 21 years is 2,83% for Turkey and 0,97% for Greece. According to the analysis, we found that the financial inclusion index Granger-cause GDP per capita, Gini index Granger-cause financial inclusion index and there is a bidirectional relationship between the financial inclusion index and Poverty Headcount ratio for Turkey. On the other hand, there is a bidirectional relationship between GDP per capita and the financial inclusion index and a bidirectional relationship between the financial inclusion index and the Poverty Headcount ratio for Greece. Conclusion- Financial inclusion simply means a larger size of financial institutions and a variety of financial products and services available for the use of adult individuals, businesses, and governmental agencies. Economic growth is supported and accelerated by an increase in financial inclusion. The empirical analysis supports the literature that the growth in the financial inclusion index enhances a higher growth in GDP and a much higher growth in GDP per capita for both Turkey and Greece. The project titled “Istanbul as an International Financial Center” may easily improve the level of financial inclusion in Turkey. Keywords: Financial inclusion, economic growth, income inequality, financial indicators, Turkish and Greek financial markets JEL Codes: G40, G41

Do unilateral trade preferences help reduce poverty in beneficiary countries?

This paper has investigated the utilization of non-reciprocal (or unilateral) trade preferences (NRTPs) provided by QUAD countries on poverty in recipient countries. It uses a panel dataset of 77 beneficiaries of NRTPs over the period of 2002–2019, and considers two main blocks of NRTPs, namely ‘Generalized System of Preferences’ (GSP) programs and ‘other trade preferences programs’. The analysis relies on two main indicators of poverty, i.e., the poverty headcount ratio at $1.90 and the poverty gap at $1.90, but also provides a robustness check using indicators of poverty at $3.20 and $5.50. Empirical findings obtained from the use of the two-step generalized method of moments indicate that over the full sample, an increase in the utilization rates of both GSP programs and other trade preferences programs is associated with poverty reduction in beneficiary countries, with the magnitude of this effect being higher for least developed countries (LDCs) than for other countries in the full sample. In addition, the simultaneous utilization of GSP programs and other trade preferences programs exerts a strong poverty reduction effect in beneficiary countries. Finally, the effect of the utilization of each type of NRTPs on poverty works depends on the beneficiary country’s level of economic complexity, and the greater the level of economic complexity, the larger is the poverty reduction effect of the utilization of these NRTPs.

Influence of Poverty Gap and Severity Index on Production among Cassava based Farmers in Southern Nigeria

Background: Poverty has been a prevailing challenge among individuals and farmers most especially in the Nigerian economy. The study analyzed poverty status of cassava farmers in Agrarian cassava region in Cross River State, Nigeria. Methods: A multistage sampling technique was adopted in selecting 120 respondents using structured questionnaire between 2019 and 2022. Data obtained were analyzed using both descriptive and inferential statistics. Result: Results showed that the poverty line estimated across all gender was N 17,770.80 and 64.15% of the male farmers were non-poor while 35.85% of them were poor. Male recorded the highest poverty head count ratio (P0) of 0.358. The poverty gap (P1) and severity of poverty (P2) among male farmers were 9.8 and 4.8 per cent, respectively. Similarly, poverty head count (P0), poverty gap (P1) and severity of poverty (P2) for female farmers was 32.8%, 9.7% and 4% respectively. Drudgery/ inadequate access to modern equipment (2.74), high cost of labour (2.64), high perishability of cassava (2.60), high cost of transportation (2.55), fluctuation in prices (2.43) and lack of improved variety (2.38) were the serious constraint faced by cassava farmers during their production in the study area. Policies should be directed towards enhancing farmer income and reducing income inequality through access to credit facility for cassava based farmers.

Empirical investigation of dynamic relations among foreign direct investment, economic growth and poverty prevalence using African data

This study examined dynamic relations among foreign direct investment (FDI), economic growth and poverty headcount ratio using a sample of 24 African Countries. To achieve this goal, historical data of the variables regarding cross section of countries were collected over 14-year period. The stratified random sampling technique was employed in selecting the sample. Following pre-regression diagnostics, we specified Vector-Auto-Regression model for computation of coefficients of the variables in dynamic relations. These were complimented with computation of relative impulse response function and forecast-error-variance decomposition of regression estimates. We found that FDI did not granger-cause economic growth among African economies, just as growth was found not sustainable and inclusive enough as to achieve substantial poverty reduction. More-so, evidence appeared to support earlier isolated findings that FDI has been largely exploitative and attracted to economies with high growth rates and low poverty ratios. The study reinforces earlier isolated findings that it is not necessarily ‘growth’ that results in decline in poverty prevalence but ‘sustained economic growth’. Hence most developing nations that depend on annual fiscal plans for poverty reduction may consistently miss development targets. Again, contrary to widely held view that foreign direct development leads to economic growth, the study established exceptional case for Africa.

Assessing the Effects of Financial Inclusion on Reducing Poverty and Income Inequality in South Asia: Evidence from a CS-ARDL Approach

This study aims to examine the effects of financial inclusion on poverty reduction and income inequality with a focus on South Asian economies from 2004 to 2020 by employing dynamic panel models augmented with a comprehensive composite financial inclusion index and a set of well-known macroeconomic predictors. The primary results confirm the long-run relationships between financial inclusion, poverty headcount ratio and the Gini index, while the results of the cross-sectionally augmented autoregressive distributed lags model indicate that financial inclusion is significantly effective in reducing poverty and income disparities. Moreover, the findings reveal that real per capita income, institutional quality, age dependency ratio, credit to the private sector, inflation rate, school enrolment rate, trade openness, ratio of mobile users, and information and communication technologies are highly significant in intermediating the aggregate effects of financial inclusion on reducing poverty and income inequality. The findings clearly show that the greater impact of financial inclusion on reducing poverty and income inequality is primarily based on the development, enhancement and promotion of financial inclusion in South Asian economies, where the financially excluded segment constitutes a significant proportion of society.

Estimation of national and subnational all-cause mortality indicators in Nepal, 2017

BackgroundDespite the civil registration and vital statistics (CRVS) system in Nepal operating for several decades, it has not been used to produce routine mortality statistics. Instead, mortality statistics rely on irregular surveys and censuses that primarily focus on child mortality. To fill this knowledge gap, this study estimates levels and subnational differentials in mortality across all ages in Nepal, primarily using CRVS data adjusted for incompleteness.MethodsWe analyzed death registration data (offline or paper-based) and CRVS survey reported death data, estimating the true crude death rate (CDR) and number of deaths by sex and year for each province and ecological belt. The estimated true number of deaths for 2017 was used with an extension of the empirical completeness method to estimate the adult mortality (45q15) and life expectancy at birth by sex and subnational level. Plausibility of subnational mortality estimates was assessed against poverty head count rates.ResultsAdult mortality in Nepal for 2017 is estimated to be 159 per 1000 for males and 116 for females, while life expectancy was estimated as 69.7 years for males and 73.9 years for females. Subnationally, male adult mortality ranges from 129 per 1000 in Madhesh to 224 in Karnali and female adult mortality from 89 per 1000 in Province 1 to 159 in Sudurpashchim. Similarly, male life expectancy is between 64.9 years in Karnali and 71.8 years in Madhesh and female male life expectancy between 69.6 years in Sudurpashchim and 77.0 years in Province 1. Mountain ecological belt and Sudurpashchim and Karnali provinces have high mortality and high poverty levels, whereas Terai and Hill ecological belts and Province 1, Madhesh, and Bagmati and Gandaki provinces have low mortality and poverty levels.ConclusionsThis is the first use of CRVS system data in Nepal to estimate national and subnational mortality levels and differentials. The national results are plausible when compared with Global Burden of Disease and United Nations World Population Prospects estimates. Understanding of the reasons for inequalities in mortality in Nepal should focus on improving cause of death data and further strengthening CRVS data.

Measuring consumption over the phone: Evidence from a survey experiment in urban Ethiopia

The paucity of reliable, timely household consumption data in many low- and middle-income countries have made it difficult to assess how global poverty has evolved during the COVID-19 pandemic. Standard poverty measurement requires collecting household consumption data, which is rarely collected by phone. To test the feasibility of collecting consumption data over the phone, we conducted a survey experiment in urban Ethiopia, randomly assigning households to either phone or in-person interviews. In the phone survey, average per capita consumption is 23 percent lower and the estimated poverty headcount is twice as high than in the in-person survey. We observe evidence of survey fatigue occurring early in phone interviews but not in in-person interviews; the bias is correlated with household characteristics. While the phone survey mode provides comparable estimates when measuring diet-based food security, it is not amenable to measuring consumption using the ‘best practice’ approach originally devised for in-person surveys.