LSE - Small Logo
LSE - Small Logo

Rodrigo Aguilera

August 19th, 2020

COVID-19 in Latin America: were we doomed from the start?

4 comments | 38 shares

Estimated reading time: 7 minutes

Rodrigo Aguilera

August 19th, 2020

COVID-19 in Latin America: were we doomed from the start?

4 comments | 38 shares

Estimated reading time: 7 minutes

Profile photo of Rodrigo AguileraEven though different countries responded very differently to the emergence of coronavirus, its impact has been devastating virtually everywhere in Latin America. Underlying factors like labour informality, compounding health issues, low healthcare spending, multi-generational households, and economic openness have made the region’s experience of the crisis especially grave, writes Rodrigo Aguilera.

Along with the US, Latin America has emerged as the main battleground in the global fight against COVID-19. Despite the delayed arrival of the virus in the region, Brazil and Mexico now have respectively the second and third highest numbers of total deaths, whereas Peru and Chile are amongst the worst performers in terms deaths per capita. Alarmingly, the infection curves in most of these countries show no sign of being under control. Economic reopening has also proven far more problematic than in Europe, with even an early success story like Costa Rica reporting a spike in cases and deaths since loosening its lockdown in June.

Daily COVID-19 deaths in Latin America for selected countries (log scale)
Source: Our World in Data

Variations in regional responses

These dismal results come despite a very uneven pattern of responses: indifference and denial in countries like Nicaragua and Brazil versus swift and comprehensive lockdowns in places like Argentina, Chile, and Peru. Many of these responses were undertaken well before European counterparts had acted, and at a point when there were far fewer reported cases and deaths.

Peru, for example, declared a national emergency and began its nationwide lockdown on 16 March, with just 85 cases and no deaths. In stark contrast, the UK had over 6,000 cases and 359 deaths when its lockdown began a week later on 23 March.

Nor have governmental attitudes to the pandemic been the determining factor. Much has been said about the weak responses of Jair Bolsonaro in Brazil and Andrés Manuel López Obrador (AMLO) in México, both of whom have appeared to be concerned primarily with the economic impact of lockdowns and their own image and popularity. But in terms of deaths per capita, Brazil and Mexico trail Peru and Chile, both of which mounted far quicker responses, offered more comprehensive economic support to workers, and now boast higher testing rates.

It is worth noting that by virtue of being federal republics, state governments in Mexico and Brazil have had considerable influence on their local pandemic responses. Bolsonaro and López Obrador may have stolen the spotlight due to their irresponsible antics, but the actions taken on the ground by local governments were far more important than many casual observers may have believed.

Coronavirus deaths vs testing to August 2020 in Latin America
Source: Worldometer

Was Latin America doomed from the start?

There is a case to be made, however, that the region was always destined to come unstuck when faced with a silently transmitted virus whose co-morbidities are so closely matched to the region’s health profile.

The first weakness, transmission, is directly related to the capacity of regional governments to effectively enforce lockdown, with all the disadvantages that underdevelopment entails. The factor most frequently cited in this regard is the region’s high level of informality, which affects over 40 per cent of the labour force even in wealthier countries like Chile. In poorer countries like Honduras, Nicaragua, and Bolivia, informality exceeds 70 per cent.

Informality rates of various Latin American countries in 2017
Source: ECLAC 2017 or latest available figure (Mexico data from INEGI Q4 2017)

Informality complicates lockdown because the general lack of universal social security means that informal workers receive no income support. And since most informal jobs are in hands-on sectors like retail and construction, working from home becomes virtually impossible. Taken together, these two factors make adherence to lockdown rules a literal matter of life and death for millions of people. Governments have been somewhat mindful of this conundrum, which is why many have enforced lockdowns only loosely. But even those workers opting to venture out have found little demand for their wares in near-empty streets.

Even if the region had better resisted the post-1980s ideological disdain for universal social security, governments would have faced other hurdles. Low levels of financialisation mean that millions of Latin Americans lack bank accounts, making it impossible for them to access unemployment support remotely. Fewer still have the means to shop online, so they are obliged to hit the streets to stock up on provisions. Then again, many households even lack refrigerators, which makes stocking up impossible; instead they simply have to make more trips to the shops and buy less on each visit. All of these factors increase the risk of contagion in ways that Westerners have been able to avoid.

Source: Global Obesity Observatory

As for co-morbidities, Latin America finds itself in the unenviable position of having third-world incomes with first-world disease profiles. The region has some of the world’s highest rates of obesity (with Mexico second in the world to the US) as well as widespread diabetes and hypertension, all of which have been found to worsen the outcomes of COVID-19. This has also blunted what might otherwise have been the region’s most potent weapon against the disease: youth. Instead, even relatively young people exhibit many of these co-morbidities much earlier than they should, particularly those in their 40s (an otherwise low-risk age group). This potential benefit is further offset by Latin America’s high proportion of multi-generational households, particularly where younger members are forced to work and may inadvertently spread the disease to elderly relatives.

Graph showing rates of hypertensive heart disease in various Latin American countries
Source: Global Burden of Disease

Finally, healthcare spending varies considerably across the region, but even this does not appear to fully explain different outcomes: the region’s top healthcare spenders have not necessarily performed much better than those spending less, whether as a share of GDP or in per capita terms. What is true is that no country in Latin America spends as much as it should, since even in GDP terms not a single country spends as much as the OECD average.

Graph showing coronavirus deaths in relation to healthcare spending in Latin America
Source: World Bank, Worldometer

One might point out – notwithstanding the horror stories from Ecuador’s early outbreak and remote Amazonian cities like Manaus (Brazil) or Iquitos (Peru) – that the region has generally managed to avoid the “collapse” of its healthcare systems, but even here there is a caveat. The ability of hospitals to cope is related to the region’s far higher share of deaths occurring outside of hospitals.

Whether because of transportation issues or concerns over the cost or quality of care, many Latin Americans have chosen to avoid going to the hospital despite falling ill. This has undoubtedly led to a higher number of avoidable fatalities. Patients can die from a lack of oxygen even in milder cases that don’t require mechanical ventilation, whereas side-effects like strokes and renal failure can also be fatal outside of a hospital setting. It is estimated that eight out of ten COVID-related deaths in Mexico have occurred outside hospitals, and there is no reason to suspect this would be particularly different elsewhere in Latin America.

Were Latin American economies doomed as well?

The economic fallout from the pandemic is yet another unfortunate consequence of the disease and one that impacts the region disproportionately. According to the IMF’s most recent estimates, Latin America faces the most severe recession of any world region (a contraction of 9.4%), dwarfing its experience even of the 2009 global recession or the 1982 debt crisis.

As with the health crisis, much of this was unavoidable. With their largely open economies and high dependency on tourism, the region’s economies would have been dealt a hammer blow even had there not been a single local case. And though widespread informality does mean that some economic activity slipped through the cracks, this was nowhere near enough to compensate for the effect of lockdowns on the formal sector, which contributes the lion’s share of GDP and fiscal revenues.

Three buskers in protective face masks play to passers-by in Quito's Centro Historico
“Even where restrictions have been loosened, there is no guarantee that consumers will venture out if the risk of infection is still high.” (Quito, Terán Bryan Esteban/shutterstock.com)

Failure to bring down the infection curve has also meant that that Latin America’s lockdowns have been more prolonged. Even where restrictions have been loosened, there is no guarantee that consumers will venture out if the risk of infection is still high. If a second wave hits the Europe over the coming months, it will at least come after a brief economic revival in the summer. Latin America may not see any semblance of normality until a working vaccine is released.

Lessons unlearned

Accepting the premise that the region’s health crisis and economic crisis were largely unavoidable should not be seen as a justification for incompetence. There is no excuse for Bolsonaro’s Trump-like attacks on state governors who were taking the crisis seriously. Mexico’s AMLO was wrong to refuse to follow hygiene rules (except when in the US) and to claim that religious amulets were enough to protect him. Nicaragua’s Daniel Ortega should not have buried his head in the sand and disappeared from public view for 34 days at the start of the crisis.

But it does seem that even if these countries had done everything right, the pandemic might not have played out much differently than it has. With so many factors at play, and with so many aspects of the disease still poorly understood by science, there is still no conclusive answer to why some countries have done so poorly despite significant efforts, whereas others have seen similar outcomes while seeming to do much less. That said, it is undeniable that those countries that did well responded early and put the right policies in place. Not all the region’s COVID failings were based on bad policies, but all of its successes were based on good ones.

That a less developed country like Paraguay has done so well is also encouraging because it shows that success action against COVID-19 does not depend on income levels or institutional development alone. It is also true that some of the countries that were hit hardest and earliest, like Chile, are now seeing a reduction in their infection curve, whereas early successes like Argentina and Colombia have seen consistent rises.

But the clearest lessons are for those countries that have struggled from the start. It should now be obvious to Latin America, as for the UK and the US, that taking half-measures or delaying action for fear of the economic consequences ends up being far costlier in the long-run.

 

Notes:
• The views expressed here are of the authors rather than the Centre or LSE
• Please read our Comments Policy before commenting

About the author

Rodrigo Aguilera

Rodrigo Aguilera is a Mexican-born, London-based economist who has worked as an international economist for Chatham House and the Economist Intelligence Unit, where he was the lead analyst for Mexico from 2012 to 2017 (as well as covering other countries such as Chile, Peru, and Venezuela). Rodrigo holds a BSc in Economics from Universidad de las Américas-Puebla and an MSc in Social Policy and Development from LSE.

Posted In: COVID19 | Society

4 Comments