Versión publicada del artículo: SUPPA, Nicolai; NOGALES, Ricardo (2022) “Simulations and multidimensional poverty: reflections on a COVID-19 evaluation study”. Dimensions, 13: 25-27. Multidimensional Poverty Peer Network (MPPN). Dimensions M u l t i d i m e n s i o n a l Po v e r t y Pe e r N e t w o r k ( M P P N ) www.mppn.org February 2022 | Number 13 DISAGGREGATING BY GENDER, ETHNICITY, RACE AND CASTE IN THE GLOBAL MPI INTERVIEW WITH IVÁN OJEDA: INE DIRECTOR OF PARAGUAY THAILAND MPI SOUTH AFRICAN COVID-19 VULNERABILITY INDEX www.ophi.org.uk Contents www.mppn.org Editor: Carolina Moreno Editorial 3 Editorial Board: Sabina Alkire “The MPI is a great contribution to transparency in Gonzalo Hernández-Licona 5 public management” Interview with Iván Ojeda Mónica Pinilla-Roncancio Leaving no one behind: a case for disaggregating by gender, Proofreader: ethnicity, race and caste in the global Multidimensional 8 Maya Evans Poverty Index Kelly-Ann Fonderson by Kelly-Ann Fonderson National MPI of Thailand: The compilation Designer: process and the policy uses 11 Sandra Pérez by Rawirin Techaploog and Suphannada Lowhachai Translation: The South African COVID-19 Vulnerability Index 14 Luis Ruiz by Risenga Maluleke Cover Photo: Using Multidimensional Poverty and Vulnerability Indices unsplash.com/Pat Whelen for Equitable Policy-Making in the Context of Health Emergencies 17 by Jakob Dirksen The opinions and analysis expressed on this website bMPI supporting the private sector to achieve the and in Dimensions magazine eradication of multidimensional poverty in Costa Rica 20 are those of the authors and by Andrés Fernández do not necessarily reflect the Breakthrough: The Power to Break the Poverty Cycle views of the MPPN, its Steering 23 by John Hammock Committee, or the participant countries and organisations. Simulations and Multidimensional Poverty: Reflections on a COVID-19 evaluation study 25 by Nicolai Suppa and Ricardo Nogales OPHI Executive Education: a new journey to support leaders and policymakers in their effort to confront poverty 28 by Michelle Muschett Data of the Month 30 Follow us: News 33 facebook.com/ophi.oxford twitter.com/ophi_oxford youtube.com/OPHIOxford instagram.com/ophi_oxford 2 | DIMENSIONS FEBRUARY 2022 Simulations and multidimensional poverty: reflections on a COVID-19 evaluation study Nicolai Suppa and Ricardo Nogales T he current COVID-19 pandemic, which took hold in early 2020, put policymakers in a vexing situation. First, there is a high degree of uncertainty Data and scenarios In our study, we use micro datasets for 70 countries associated with both the effectiveness and potential of the 2020 global Multidimensional Poverty Index side-effects of most policy measures aiming to curtail (MPI) release and related changes over time, which is the spread of the virus, including the virus’ precise carefully documented in OPHI Methodological notes epidemiological characteristics. Secondly, much is at 49 and 501. The global MPI builds on 10 indicators stake, including a significant number of premature organized in three dimensions (health, education, deaths, extraordinary economic contractions, and the and living standards) and relies on the dual cutoff education of several young generations. counting approach, sometimes called the Alkire- Foster method. While being estimated since 2010, Meanwhile, policy-oriented research encounters a the global MPI was revised in 2018 to better align data scarcity, impeding grounded recommendations with the Sustainable Development Goals. The global and advice as conventional impact evaluations are MPI is estimated based on comprehensive household not yet feasible. Notwithstanding, situations like survey datasets, such as the Demographic Health these require decisions even under such adverse Survey (DHS) and the Multiple Indicator Cluster circumstances. Survey (MICS). In a recent study, we apply simulation techniques In our simulations we implement scenarios for Photo: unsplash.com/photos/Xkeczu4lvfc for a forecasted assessment of the pandemic and related two indicators, deprivation in nutrition and school policy responses on global multidimensional poverty. attendance because (i) they capture relevant short- In this article we provide a non-technical discussion term effects, (ii) there is enough reliable information of this paper, highlighting both the usefulness and to inform how they can be affected by the pandemic, challenges of simulations for multidimensional and (iii) they have a large weight in the global MPI. poverty analyses. 1 More precisely we rely on 70 countries for arriving at the results reported at the global level, whereas we make use of 97 countries during the simulation exercise itself. 25 | DIMENSIONS FEBRUARY 2022 ARTICLE For this reason, we first nowcast multidimensional poverty to 2020, and then develop and apply a model which translates the simulated shocks to 2020, allowing us to compute potential increases in different outcomes due to the shocks. Table 1 shows, for instance, that according to our upper-bound scenario, the global MPI would increase to 0.164, whereas the number of poor people would increase by 547 million, which is considerable. Aggregate Adjusted COVID-19 scenario Simulation for 2020 Selection probabilities MPI (M) Δ # poor Setback School Nutrition M*s(2020) Δ*Q s (2020) (2020−t*) attendance (%) value (million) (years) 12 – 0.114 152 3.6 Nutrition deprivation scenarios are informed 20 – 0.122 213 4.8 by World Food Programme (WFP) data on food security risks. The WFP provides pre-pandemic 50 – 0.134 310 6.4 numbers of food-insecure people in many countries. 12 50 0.146 426 8.0 We assume that food insecurity materialises in actual malnutrition among the poor and vulnerable people 20 50 0.153 469 8.8 who have managed to avoid nutrition deprivations 50 50 0.164 547 9.9 before the pandemic. Accounting for specific features of WFP data, we derive three plausible risk rates of Notes: Authors’ calculations; MPI values are popula- tion-weighted aggregates across the 70 countries, while increased nutrition deprivations: 12%, 20% and 50%. the increases in number of poor are totals across the same countries. All calculations based on UN-DESA medium-fertili- ty population projections. Risks of school nonattendance are informed by UNESCO data on daily school closures around the Source: Alkire, Nogales, Quinn, Suppa (2021) world. We derive a risk of 50% for children in school age to experience significant loss of schooling, and thus deprivation in school attendance due to school Additionally, we also compute the setback in terms Photo: www.flickr.com/photos/bcgovphotos/50724854161 closures during the first year of the pandemic. of poverty reduction. We find that, depending on the underlying scenario, the pandemic may result in up to Translation model and results an entire decade of global multidimensional poverty reduction becoming undone. In early 2020, even many of the most recent country-level micro datasets were collected several Discussion years before the pandemic. As a consequence, taking simulation results at face value would be misleading, as One implication we emphasize in the paper is that countries’ recent poverty reductions would be ignored. appropriate policy measures are needed to prevent Moreover, the simulated shocks themselves would not these new deprivations from becoming entrenched. reflect 2020 conditions.2 Previous research documents several measures to 2 Indeed, our analysis suggests a non-linear relationship between level of multidimensional poverty and the simulated shock. 26 | DIMENSIONS FEBRUARY 2022 attenuate the effect of temporary school closures on A third challenge is related to the actual school dropout rates (World Bank, 2020). implementation of the scenarios: Which individuals or households are selected to experience a new Methodologically, simulations emerge as one deprivation? How exactly is this implemented, and promising tool to better understand multidimensional why is it justified? In our schooling scenario, for poverty. First, simulations may be applied even when instance, we implement a uniform risk for all school- key information on certain aspects is unavailable. age children. This is supported by an unprecedentedly Moreover, if correctly devised, simulations may widespread school closure around the globe. Other implicitly account for various aspects of the micro exercises, such as the simulation of a specific policy data, including the joint distribution of deprivations measure, undeniably require much more elaborated (e.g. of specific deprivation profiles) and various and fine-tuned implementations to obtain truly socio-demographic aspects (e.g. age structure of the informative results. population). Appropriate policy Simulations, however, also pose important measures are needed Photo: www.flickr.com/photos/worldbank/49834249446 challenges. First, outdated micro data may not accurately reflect the current situation in a particular to prevent these new country and thus support misleading conclusions. deprivations from A translation model, as proposed in our paper, may becoming entrenched. address this issue in some instances, but simply having more recent micro data at hand may suffice. The full potential of simulation techniques for Another shortcoming of simulations is that analysing multidimensional poverty is certainly their plausibility and accuracy are not easy to assess. not yet fully developed. Addressing these and Therefore, a very clear and detailed documentation other challenges first would clearly be the next step of the underlying assumptions and the data used to forward. inform the different scenarios is vital for a transparent analysis. Otherwise, a simulation study is inevitably Suppa gratefully acknowledges funding of the “la Caixa” tainted by an unacceptable degree of arbitrariness. foundation (LCF/PR/SR20/52550004). 27 | DIMENSIONS FEBRUARY 2022 Photo: www.flickr.com/photos/bancomundiallac/50665682781 Dimensions