Evidence from 37 low- and middle-income countries shows that cash transfer programmes were associated with a 20% reduction in mortality for adult women and an 8% reduction in mortality for children aged <5 years
Eradicating poverty by 2030 features prominently as the first of the UN’s Sustainable Development Goals (SDGs). Achieving this goal is pivotal for many other SDGs since poverty makes it harder for people to meet basic needs, avoid hunger, and access health and education. Despite decades of progress, however, 659 million people (8.5% of the world’s population) lived in extreme poverty in 2019 – a number that rose even further (to 724 million, or 9.3% of the world’s population) during the COVID-19 pandemic (United Nations 2023). With slow and uneven progress in poverty reduction in the past decade, developing strategies to alleviate poverty remains as essential as ever.
In the past two decades, social protection programmes – particularly cash transfers – have become a centerpiece of many countries’ efforts to reduce poverty and protect against the many consequences of poverty. Examples of cash transfer programmes include one-time cash grants to pregnant women, recurring payments to households with children or to poor and vulnerable families, as well as payments that are conditional on children attending school regularly. A recent World Bank report documented 962 cash transfer programmes in 203 countries in 2020-21, with 1.36 billion people—17% of the world’s population—receiving cash transfers (Gentilini 2022). The COVID-19 pandemic was widely seen as a “game changer” for cash transfers, as 672 of the 962 cash transfer programmes were newly introduced during the pandemic and many others were expanded.
Research on cash transfers has kept pace with the expansion in cash transfer programmes – there was a 26-fold increase in publications on this topic since 1980, with 95,000 publications in 2021 alone (Gentilini 2022). We know that cash transfers are highly effective not only in reducing poverty but also in improving school attendance, child nutrition, healthcare utilisation, women’s empowerment, and mental health outcomes among beneficiaries (Bastagli et al. 2016, McGuire et al. 2022). But most studies have focused only on the effects of cash transfers on direct beneficiaries, even though cash transfers also have general equilibrium effects by stimulating greater economic activity in communities where beneficiaries reside (Egger et al. 2022). In the case of infectious diseases like HIV, they also reduce new infections (Richterman and Thirumurthy 2022). Ultimately, we know surprisingly little about the overall effects of cash transfer programmes on population-level mortality rates, an important health indicator that is essential for understanding how poverty alleviation may contribute to improving population health.
A new analysis of data from 37 low- and middle-income countries
Country-specific evaluations of cash transfers programmes typically lack the large sample sizes or long-term follow-up needed to assess the effects of such programmes on mortality rates. In our research (Richterman et al. 2023), we overcome this challenge by combining population-representative survey data from many low- and middle-income countries (LMICs) and examining what happens to mortality rates following the introduction of large-scale cash transfer programmes run by governments.
To estimate mortality rates from all causes, we relied upon data from Demographic and Health Surveys (DHS) conducted in many LMICs about every 5 years. Adult mortality was inferred from information reported by respondents about their siblings’ vital status. Child mortality was determined similarly with information about respondents’ birth histories. We used these data to construct longitudinal datasets for adults and children with observations at the level of the person-year, with an indicator variable for survival in each year. For the LMICs with DHS data between 2000-2019 and no pre-existing cash transfer programmes, we also identified and characterised all major government-led cash transfer programmes introduced during this period by manually reviewing primary (e.g. government reports) and secondary (e.g., United Nations databases) sources. For each cash transfer programme, we estimated the most recent impoverished population coverage by dividing the total number of beneficiaries by the number of people living below the international poverty line.
Altogether, 37 LMICs were included in our study. Among these, 16 countries introduced large-scale cash transfer programmes during the study period (Figure 1), about half of which were unconditional. Cash transfers covered a median of 27% of the impoverished population (interquartile range 16–100%), and the maximum transfer amount per beneficiary had a median that equated to 10% of per capita GDP. The individual-level datasets we assembled had over 30 million person-years of information for 4.3 million adults, with 126,714 deaths in total (42 per 10,000 person-years). For children, there were 16 million person-years of information for nearly 3 million individuals, with 162,488 deaths (99 per 10,000 person-years).
Figure 1: Countries with available DHS data that did not have pre-existing cash transfer programmes at the start of the study period (2000). Countries in blue implemented cash transfer programme(s) with impoverished population coverage of 5% or greater between 2000-2019, and had available mortality data during the cash transfer period. Countries in red did not implement such programmes between 2000-2019, or implemented programmes during the study period but only during years without available mortality data.
We used the difference-in-differences methodology to then examine the effects of cash transfer programmes on mortality rates. This is a quasi-experimental methodology that can estimate causal effects from observational data by comparing differences in outcomes between intervention and comparison groups during pre-intervention and post-intervention periods. We confirmed several assumptions that are important for ensuring the validity of this approach. Our primary explanatory variable was a binary variable indicating whether a cash transfer programme that reached at least 5% of a country’s impoverished population was active in the survey respondent’s country during a given year. Our primary goal was to determine the association between the presence of a cash transfer programme in a country and mortality rates in that country. In addition, we also examined the effects on mortality rates over time and the effects on various sub-groups. Our analyses also included several country- and individual-level covariates that may confound relationships between cash transfer programmes and mortality (e.g. country-level GDP per capita, funding for HIV/AIDS, and governance indicators). We also confirmed that inclusion of health expenditures per capita – which may decrease or increase along with cash transfer programme coverage – did not affect our estimates.
Government-led cash transfer programmes led to large declines in all-cause mortality rates
Our primary finding was that cash transfer programmes were associated with a 20% reduction in the overall risk of death among adult women and an 8% reduction in the risk of death among children aged <5 years (Figure 2). These striking estimates translate into large numbers of lives saved given the underlying mortality rates in the countries we studied and compare favourably to the effect sizes of other effective health policy interventions (Figure 2) (Bendavid, Holmes, Bhattacharya, & Miller, 2012; Jakubowski, Stearns, Kruk, Angeles, & Thirumurthy, 2017; McGovern & Canning, 2015). In contrast, for adult men, children aged 5-9 years, and children aged 10-17 years, cash transfer programmes were not associated with significant changes in mortality.
Figure 2: Relative reductions in mortality associated with large-scale, government-led cash transfer programmes in green (Richterman et al. 2023), compared with other major health policy interventions in orange (Bendavid et al. 2012, Jakubowski et al. 2017, McGovern & Canning 2015).
When examining the year-by-year effects of cash transfer programmes, we observed significant declines in mortality within 2 years after programmes were introduced. These effects grew even larger over time. Declines in mortality rates over time were also detected for men. The temporal patterns in the mortality effects of cash transfers are notable for several reasons. First, short-term evaluations of cash transfer programmes may not detect effects on mortality rates. Second, the results are consistent with the possibility that cash transfer programmes may induce changes in behavior (such as increased healthcare utilisation or investment in education) that lead to longer-term reductions in mortality.
Our study had several other findings with notable policy implications. Cash transfer programmes with and without conditionality had similar effects on mortality rates. Since over 80% of cash transfer programmes globally do not include conditions (Gentilini 2022), this is encouraging evidence that conditionality is not essential for observing health benefits. In addition, the effects on mortality were much larger when cash transfer programmes served a larger proportion of the population or when cash transfer amounts were larger. Programmes that are more ambitious in scope therefore have much greater potential to impact population health. Finally, cash transfer programmes had larger effects in countries with low health spending and low life expectancies. Thus, people living in nations with little health-care infrastructure or substantial public-health challenges could benefit especially from such programmes.
Combatting poverty can save lives
In an effort to alleviate poverty and combat growing inequality, policymakers and stakeholders in many settings – ranging from high-income countries to low-income countries – are now considering the merits of introducing or further expanding cash transfer programmes. Policy measures like universal basic income and guaranteed income are also being piloted in various places. The overall costs and sustainability of such programmes is often a central concern, and our research shows that an important and largely unnoticed benefit of cash transfer programmes may come in the form of lower mortality rates – not merely among beneficiaries but rather at the population level. Whilst formal cost-effectiveness analyses based on our findings remain necessary, a rough calculation using the total cost of cash transfers and the underlying mortality rates in the study countries implies a cost of $11,000 per life saved. This is a number that compares favorably to the most cost-effective health charities (GiveWell 2023) and health aid programmes (United States Department of Health and Human Services 2023), even though it is a substantial underestimate of the cost-effectiveness of cash transfers since it does not include the effects of cash transfers on other outcomes (e.g. better education outcomes) or the general equilibrium effects of cash transfers.
If current trends continue, 575 million people will still be living in extreme poverty and only one-third of countries will have halved their national poverty levels by 2030. Cash transfer programmes may be essential for making progress in reducing poverty. Our research shows that they may also save many lives, thus providing an important new perspective on how cash transfer programmes could address multiple SDGs by reducing poverty and improving health.
References
Bastagli, F, J Hagen-Zanker, L Harman, and V Barca (2016), "Cash transfers: what does the evidence say?".
Bendavid, E, C B Holmes, J Bhattacharya, and G Miller (2012), "HIV Development Assistance and Adult Mortality in Africa", Jama, 307(19): 2060-2067.
Egger, D, J Haushofer, E Miguel, P Niehaus, and M Walker (2022), "General Equilibrium Effects of Cash Transfers: Experimental Evidence From Kenya", Econometrica, 90(6): 2603-2643.
Gentilini, U (2022), "Cash Transfers in Pandemic Times: Evidence, Practices, and Implications from the Largest Scale Up in History", Available here.
GiveWell (2023), "How We Produce Impact Esimates", [Online], Available: https://www.givewell.org/impact-estimates.
Jakubowski, A, S C Stearns, M E Kruk, G Angeles, and H Thirumurthy (2017), "The US President’s Malaria Initiative and under-5 child mortality in sub-Saharan Africa: A difference-in-differences analysis", PLoS Medicine, 14(6): e1002319.
McGovern, M E and D Canning (2015), "Vaccination and all-cause child mortality from 1985 to 2011: global evidence from the Demographic and Health Surveys", Am J Epidemiol, 182(9): 791-798.
McGuire, J, C Kaiser, and A M Bach-Mortensen (2022), "A systematic review and meta-analysis of the impact of cash transfers on subjective well-being and mental health in low- and middle-income countries", Nature Human Behaviour, 6(3): 359-370.
Richterman, A, C Millien, E F Bair, G Jerome, J C D Suffrin, J R Behrman, and H Thirumurthy (2023), "The effects of cash transfers on adult and child mortality in low- and middle-income countries", Nature, 618(7965): 575-582.
Richterman, A and H Thirumurthy (2022), "The effects of cash transfer programmes on HIV-related outcomes in 42 countries from 1996 to 2019", Nature Human Behaviour, doi:10.1038/s41562-022-01414-7.
United Nations (2023), "The Sustainable Development Goals Report: Special Edition", [Online], Available: https://unstats.un.org/sdgs/report/2023/.
United States Department of Health and Human Services (2023), "What is PEPFAR", [Online], Available: https://www.hiv.gov/federal-response/pepfar-global-aids/pepfar/.