Public Disclosure Authorized LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY* COVID-19 IN LAC Policy Note No.1 / April 2021 Public Disclosure Authorized Main Messages • COVID-19 closures in Latin American and the Caribbean countries helped to curb the spread of the virus, but inevitably brought negative consequences for households, principally in the form of job losses, in- come reduction, and, in some cases, food insecurity. • Future policy measures should aim to strike the right balance between saving lives and protecting livelihoods. Where closures are necessary, they should be adapted to a country’s labor market and other localized condi- tions so as to minimize profound welfare losses. • Governments should strive for robust and agile social safety net systems to be able to respond to the sudden falls in household welfare. A good start, for example, is for national security registries to be kept up to date. Public Disclosure Authorized T he COVID-19 crisis has evidenced an unavoidable trade-off for governments worldwide: the ‘choice’ between saving lives or protecting livelihoods. Government-imposed national lockdowns, travel restrictions, and so- cial-distancing measures (collectively known as ‘non-pharmaceutical interventions’) helped contain the spread of the disease. Yet, these measures also inevitably impacted the welfare of households, particularly for the most vul- nerable. Latin American and the Caribbean is one of the regions hit the hardest by the COVID-19 pandemic. The gross domestic product contracted by 7 percent in 2020, and inequality increased. In total, around 32 million additional people fell back into poverty.1 A year into the pandemic and with a renewed hike in case numbers, one lesson is critical for the region’s policymakers: the appropriate policy response should balance saving lives and protecting livelihoods. The success of these policies rests on a proper assessment of the country’s ‘initial’ economic and institutional conditions. Strict social distancing may not be realistic in countries with a high level of labor market informality or a weak health system (Barnett-Howell and Mobarak, 2020). Moreover, non-pharmaceutical interventions should be Public Disclosure Authorized complemented with social protection measures in a framework of intersectoral coordination and cross-sectoral part- nership, including with the private sector. This brief presents evidence on how closure measures and the expansion of social security programs are reflected, both positively and negatively, in the wellbeing of Latin American and the Caribbean households. 1 Measured by the $5.50 in 2011 PPP. * This note was prepared by Paola Ballon, Carolina Mejía-Mantilla, Sergio Olivieri, Gabriel Lara Ibarra and Javier Romero, under the guidance of Ximena del Carpio, with the financial support from the Latin American and Caribbean Regional Vice Presidency. 1 www.worldbank.org /BancoMundial @BancoMundialLAC LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY COVID-19 IN LAC Policy Note No.1 / April 2021 The Benefits and Costs of Closures Lockdown and safe-distancing measures implemented by governments in the region undoubtedly contributed to saving lives. Between April and September 2020, Latin America and the Caribbean registered a sustained decrease in new coronavirus cases and deaths as a result of containment and closure policies, such as school shutdowns and movement restrictions (Figure 1). As these policies became stronger (reflected in a higher stringency index), countries were able to flatten the curve and decongest their emergency health services. The result was a demonstrable decline in the spread and mortality of the disease. Figure 1: COVID-19 & Stringency in Latin America and the Caribbean April – September 20202 Daily cases Daily deaths (in Thousands) (in Thousands) 60 6 50 5 40 4 30 3 20 2 10 1 CASES DEATHS 0 0 73 75 77 79 81 83 85 87 89 91 September 2020 Stringency index April 2020 Source: Health data from WHO, Coronavirus (COVID-19) Dashboard; Stringency index from Oxford COVID-19 Government Response Tracker, Blavatnik School of Government, University of Oxford. 2 The figure plots the daily cases and deaths for the thirteen countries where the World Bank conducted High-Frequency Phone Surveys (see below) for the period April-September 2020. Each observation represents the total number of cases (or deaths) per day across the thirteen countries. The stringency index used in this brief is an overall government response index computed by the Blavatnik School of Government of the University of Oxford, United Kingdom. This index collects publicly available information on 20 indicators of government response on containment and closure policies, economic policies, and health system policies. A higher value denotes greater stringency. 2 www.worldbank.org /BancoMundial @BancoMundialLAC LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY COVID-19 IN LAC Policy Note No.1 / April 2021 The High-Frequency Phone Survey The World Bank conducted a multi-wave High-Frequency Phone Survey (HFPS) to assess the impact of the coronavirus pandemic on the welfare of Latin American and Caribbean households. Between March and June 2020, the HFPS collected nationally representative information for thirteen countries: Argentina, Bolivia, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Para- guay, and Peru. The HFPS provides information on the welfare loss that households and people experienced regarding food insecurity, changes in employment, income loss, access to health services and education, and coping mechanisms. Additionally, it documents household responses to lockdown and safe-distancing measures imposed by their governments and catalogs the behaviors they undertook to mitigate the spread of the disease. These socioeconomic impacts are available at the COVID-19 monitoring global dashboard, which provides 96 harmonized indicators across 50 countries in Latin America and the world. For further information on the HFPS, see Mejía-Mantilla et al. (2021). Despite the positive effects on public health, lockdowns and closures also resulted in job losses, family income reductions, and increases in food insecurity (Figure 2). Most countries reached the highest stringency index and job- loss levels about one month after the first lockdown was imposed. A positive and strong association between job losses and the stringency of the closures is observed throughout the region and also over time (Figure 3). However, the magnitude of employment loss was mediated by the structural characteristics of the region’s economies. Countries with high informal economies struggled, which led to substantial employment losses (c.f. Olivieri et al. (2020). For instance, in April 2020, almost 30 percent of Peruvian and Colombian workers lost their jobs, either temporarily or permanently. Countries with lower informality levels and more comprehensive access to internet ser- vices adjusted better to teleworking. Job losses in Chile, for example, were six times lower than in Colombia and Peru, despite similar closure policies. As the restrictions relaxed over time and as countries’ stringency index declined, job losses decreased across the board. By the end of July 2020, around 10 percent of workers in the region experienced unexpected unemployment (albeit rates in Colombia remained high, at 20 percent). Figure 2: Welfare losses due to COVID-19 JOB INCOME FOOD COVID-19 LOSS LOSS INSECURITY 3 www.worldbank.org /BancoMundial @BancoMundialLAC LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY COVID-19 IN LAC Policy Note No.1 / April 2021 Figure 3: Job Loss vs. Stringency in Latin America and the Caribbean Job Loss Rate (%) 30 PER COL 25 20 GTM ECU COL CRI 15 DOM BOL SLV DOM HND MEX PRY ARG 10 MEX PER HND ECU SLV CHL CRI BOL 5 GTM CHL PRY ARG 0 70 75 80 85 90 95 100 May 2020 August 2020 Stringency index Source: Job loss data from High-Frequency Phone Surveys (HFPS) Stringency index from Oxford COVID-19 Government Response Tracker, Blavatnik School of Government, University of Oxford. Loss of employment had an immediate impact on family incomes across all countries in the region. By the end of April 2020, Decrease when the stringency level was at its highest in the region, 60 to 70 percent of households experienced a in Income (% of households) decline in total family income. As countries began to reopen, households were able to find alternative employment PER 80 and income sources, leading to lower losses in household income. The most significant progress was seen in Mexico ECU and Peru. By the end of July 2020, 35 percent fewer households these two countries reported in COL a loss of income BOL 70 GTM (Figure 4). CRI PRY MEX SLV HND 60 DOM CHL 50 PER BOL HND ECU GTM SLV 40 DOM ARG CRI CHL COL PRY 4 30 MEX 20 70 75 80 85 90 95 100 www.worldbank.org /BancoMundial @BancoMundialLAC May 2020 August 2020 Stringency index CRI 15 DOM BOL SLV DOM HND MEX PRY ARG LIVES OR LIVELIHOODS? 10 MEX PER HND ECU SLV CHL CRI THE COSTS OF STAYING HEALTHY 5 BOL GTM CHL PRY ARG COVID-19 70 IN LAC 0 75 80 85 90 95 100 May 2020 August 2020 Stringency index Policy Note No.1 / April 2021 Figure 4: Income Loss vs. Stringency in Latin America and the Caribbean Decrease in Income (% of households) 80 PER ECU BOL COL 70 GTM CRI PRY MEX SLV HND 60 DOM CHL 50 PER BOL HND ECU GTM SLV 40 DOM ARG CRI CHL COL PRY 30 MEX 20 70 75 80 85 90 95 100 May 2020 August 2020 Stringency index Source: Income loss data from High-Frequency Phone Surveys (HFPS) Stringency index from Oxford COVID-19 Government Response Tracker, Blavatnik School of Government, University of Oxford. Employment losses and reduced family incomes exposed households to higher food insecurity levels. A sub- stantial share of households in the region experienced food insecurity. Honduras had the highest incidence in May Food insecurity (% of 2020, households) with more than half of households (53 percent) reporting that they ran out of food due to lack of money or resources (Figure 5). Lower rates of food insecurity followed the relaxation of lockdown measures. However, there 60 was still a considerably large percentage of households without enough resources to buy food in August 2020, with approximately one-third of households in Honduras and the Dominican Republic 50 ECU still facing food insecurity. The HND GTM in welfare, knock-on effects of food insecurity on children’s nutrition and learning outcomes result in further losses COL PER DOM which in turn affect human capital and productivity in the long term. 40 BOL SLV DOM HND 30 CRI ECU COL PRY CRI MEX SLV 20 PER GTM ARG CHL BOL 10 MEX PRY ARG 5 CHL 0 70 75 80 85 90 95 100 www.worldbank.org /BancoMundial @BancoMundialLAC May 2020 August 2020 Stringency index 60 DOM CHL 50 PER BOL LIVES OR LIVELIHOODS? HND ECU GTM SLV 40 DOM ARG CRI CHL COL THE COSTS OF STAYING HEALTHY PRY 30 MEX COVID-19 IN LAC 20 70 75 80 85 90 95 100 May 2020 August 2020 Policy Note No.1 Stringency / April 2021 index Figure 5: Food Insecurity vs. Stringency in Latin America and the Caribbean Food insecurity (% of households) 60 HND 50 ECU GTM COL PER DOM 40 BOL SLV DOM HND 30 CRI ECU COL PRY CRI MEX SLV 20 PER GTM ARG CHL BOL MEX PRY ARG 10 CHL 0 70 75 80 85 90 95 100 May 2020 August 2020 Stringency index Source: Food insecurity data from High-Frequency Phone Surveys (HFPS) Stringency index from Oxford COVID-19 Government Response Tracker, Blavatnik School of Government, University of Oxford. Fiscal Support To offset this chain of welfare losses, Latin American and Caribbean governments boosted their non-health spending in 2020 (Figure 6).969 In most cases, limited fiscal space and flexibility restricted the extent of their respective responses. Chile had the strongest response, with an additional spending (including foregone revenues) of $969 per capita. Overall, higher levels of spending are correlated with lower levels of food insecurity but not necessarily with lower levels of income losses or job losses (Table 1). This suggests income losses were substantial and, in some cases, could not be fully compensated by governments’ additional fiscal efforts. 277 301 160 132 131 128 79 78 32 38 PERU ARGENTINA CHILE COLOMBIA GUATEMALA EL SALVADOR HONDURAS COSTA RICA MEXICO BOLIVIA DOMINICAN REPUBLIC 6 www.worldbank.org /BancoMundial @BancoMundialLAC PRY CRI MEX SLV 20 PER GTM ARG CHL BOL LIVES OR LIVELIHOODS? MEX PRY ARG 10 CHL THE COSTS OF STAYING HEALTHY 0 70 75 80 85 90 95 100 COVID-19 IN LAC May 2020 August 2020 Stringency index Policy Note No.1 / April 2021 Figure 6: Additional per capita non-health spending or foregone revenues due to COVID-19 in 2020 (in US dollars) 969 277 301 160 132 131 128 79 78 32 38 PERU ARGENTINA CHILE COLOMBIA GUATEMALA EL SALVADOR HONDURAS COSTA RICA MEXICO BOLIVIA DOMINICAN REPUBLIC Source: IMF, 2020. Table 1: Correlation Matrix Additional per capita non-health spending or foregone revenues due to COVID-19 in 2020 (in US dollars) Household run out of food because of lack of money or Respondent lost job Household income decreased resources May 2020 August 2020 May 2020 August 2020 May 2020 August 2020 -0.5630* -0.5981* -0.4265 -0.1239 -0.3439 -0.1832 *indicates p<0.1, **indicates p<0.05, and ***indicates p<0.01 Source: Own estimations based on High-Frequency Phone Survey (HFPS) 7 www.worldbank.org /BancoMundial @BancoMundialLAC LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY COVID-19 IN LAC Policy Note No.1 / April 2021 Looking Forward The analysis of the High-Frequency Phone Surveys shows how the wellbeing of Latin American and Caribbean households was affected by the policy response of governments across the region, both positively and negatively. As expected, the heterogeneity in welfare losses not only mirrors the differences in countries’ policy responses but also reflects their underlying economic and institutional conditions, particularly the nature of their labor market. A vital lesson that emerges is that lockdown and safe-distancing measures help to save lives but lead to inevita- ble household welfare losses. Optimal policy responses should adapt the stringency of closures to the potential wel- fare losses for households, taking particular account of the nature of the economy. Strict security protocols, including the use of facial masks, hand hygiene, and social distancing, should be enforced to the maximum extent possible to avoid closures. Similarly, when non-pharmaceutical interventions are inevitable, their design should aim to minimize the number of jobs lost and, where possible, should be accompanied by social protection measures to support highly vulnerable populations. A second, interrelated lesson refers to the successful delivery of social safety nets. Where possible, governments should allocate extra funds to compensate for falls in household welfare. As revealed by the response to COVID-19 in Latin America and the Caribbean, the administrative and operative mechanisms that the region’s governments had in place were typically weak, insufficiently responsive, or inaccessible to the most vulnerable. For instance, several countries did not have up-to-date social security registries (Olivieri and Rivadeneira, 2021). Going forward, countries would benefit from investing in improving these registries in a framework of intersectoral coordination and by making their social protection policies and programs more inclusive. Until vaccines become available for the large majority of the population, the region still faces the challenge of fight- ing COVID-19 infections while minimizing knock-on welfare losses. Learning from the 2020 experience is essential for governments to adapt and adjust their responses so as to ensure an inclusive recovery and reduce negative long-term consequences. 8 www.worldbank.org /BancoMundial @BancoMundialLAC LIVES OR LIVELIHOODS? THE COSTS OF STAYING HEALTHY COVID-19 IN LAC Policy Note No.1 / April 2021 References Barnett-Howell, Z. and A. M. Mobarak (2020). The benefits and costs of social distancing in rich and poor coun- tries. arXiv preprint arXiv:2004.04867. Ballon, P., Cuevas R., Lara Ibarra, G., Olivieri, S. and A. Rivadeneira. The Pandemic Impacts Beyond Illness: Food Inse- curity and Joblessness in Latin American Countries – Results from a High Frequency Phone Survey. Poverty and Equity Global Practice, LAC. Mimeo Mejía-Mantilla, C, Olivieri, S. and A. Rivadeneira (2021). LAC COVID-19 High-Frequency Phone Surveys 2020, Technical note. Poverty and Equity Global Practice, LAC. Mimeo. Olivieri, S. and A. Rivadeneira (2021). COVID-19 and Forced Displacement in Ecuador: Evidence from High Frequency Phone Surveys. Mimeo. Hale, T. , Angrist, N. , Goldszmidt, R., Kira, B., Petherick, A., Phillips, T., Webster, S., Cameron-Blake , E., Hallas, L., Ma- jumdar, S. and H. Tatlow (2021). A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nature Human Behaviour World Bank. COVID-19 High Frequency Monitoring Dashboard. The World Bank Group. Washington, D.C. 2020. World Health Organisation. WHO, Coronavirus (COVID-19) Dashboard. World Health Organisation, 2021. (Accessed: 14 April 2021) 9 www.worldbank.org /BancoMundial @BancoMundialLAC