By Paulino Díaz, MPP’20, former Communications Consultant at Golin

As the world grapples with the spread of COVID-19, social distancing (SD) has been championed as an effective preventative measure to slow down the pandemic. While economists recognize that SD carries a significant economic toll, there are other tradeoffs that have not been sufficiently acknowledged. Numerous studies have shown that in developing countries, short-term fluctuations in GDP per capita have large negative impacts on infant mortality (O’Hare, et al., 2013). The following analysis is focused on Mexico, a country in which 40% of the population lives on less than $1.90 USD a day and is extremely vulnerable to short-term fluctuations in income. For Mexicans living in poverty, changes in GDP per capita due to enforced SD policies could carry lethal consequences. As Mexican leaders determine the appropriate response to the pandemic, they must understand that preventing deaths from COVID-19 through SD measures could come at the expense of children’s lives lost as a result of economic hardship. 

Most economists agree that flattening the infection curve via social distancing will inevitably push the economy towards a recession, and if a recession is already underway it will exacerbate its effects (CEPR, 2020). Tragically, for Mexico, the decision is not as straightforward as choosing between high unemployment rates and saving lives. The United Nations (UN) Standing Committee on Nutrition has demonstrated that recessions in developing nations are significantly associated with malnutrition among pregnant women and children (UN, 2009). In Mexico, approximately 40% of the population (51 million) lives in poverty, and 7.6% lives in extreme poverty (9.4 million) (CONEVAL, 2016). This vulnerable segment of the population lacks proper access to water, health, social security or quality housing, and survives on less than $1.90 USD a day. For children living in these impoverished households, the economic hardship caused by SD poses an immediate risk that can lead to malnutrition, starvation and even death.

A systemic review and meta-analysis of multiple estimates of the relationship between income and the mortality rate of children under five (U5M) in developing countries found that a 1% change in GDP per capita is equivalent to a -0.28% (95% CI, −0.37% to −0.19%) change in U5M (O’Hare, et al., 2013). Starting in March, the projections for 2020 GDP growth in Mexico have ranged from +0.9% to -8% (El CEO, 2020). Using the average of these projections (-2.44%) and the estimates from O’Hare et al. and CEPR, we derived a simple model that highlights the potential impact of SD on U5M rates in Mexico. These are the estimates:

Figure 1

The estimates in Table 1 and Figure 1 can be best understood as potential outcomes based on several assumptions and limitations. Additional details can be found below in the appendix.

In 2018, Mexico’s U5M was 12.7% for every 1000 live births (UNICEF, 2018), and there were 2,162,535 live births (INEGI, 2018). This means that a 0.000425 change in the mortality rate is equivalent to the life of one Mexican child under 5 years old. The Mexican government’s Phase 3 response to the pandemic is expected to last 4 weeks, and it could easily cause a 50% decline in economic activity, as it will enforce a generalized quarantine that keeps a majority of the Mexican labor force at home and not working (El Financiero, 2020). Based on the Table 1 estimates, a 50% shutdown of the economy for 1 month will kill 188 children under the age of 5 (a 0.08 increase in child mortality). The Mexican government needs to seriously consider this human cost when evaluating the implementation of additional SD measures, and provide an unprecedented level of support to the families of the children that are most at risk.

APPENDIX: Model assumptions and limitations

In the United States, CEPR economists estimate a 6.5% decrease in annual GDP growth due to shutting down 50% of the economy for one month, and an additional 3.5% decrease from a 25% shutdown for a second month (CEPR, 2020). Starting in March, the projections for 2020 GDP growth in Mexico have ranged from +0.9% to -8% (El CEO, 2020). Using the average of these projections (-2.44%) and the estimates from O’Hare et al. and CEPR, we can derive a simple model that highlights the potential impact of SD on U5M rates in Mexico:

Where ∆GDP_SD is the economic impact of the social distancing policy and ∆U5M₂₀₁₈ is the under-five mortality rate in 2018.

To determine the change in GDP (∆GDP) due to social distancing we can generalize CEPR’s findings and assume that after the first month at 50% capacity an additional monthlong shutdown of 25% of the economy increases the economic slowdown by 53% (3.5 / 6.5 = .53). For the purpose of this analysis, we will assume that the impact of SD on the economy increases at a marginally increasing rate (more on this later). To exemplify this, we can deconstruct the impact of each additional unit (1 unit = a 25% shutdown for 1 month) of SD into two parts:

• A constant impact of -1.22 (50% of -2.44) of GDP for each additional unit of SD
• A growing impact equivalent to a 3% increase over the previous effect (-2.44 – 2.44*.03 -1.22= ~ -3.74)

From this setup the following formula for changes in GDP caused by SD can be derived:

Where SD is equal to the strictness of the social distancing policy, and ∆GDP_SD-1 is the change in GDP caused by the previous SD policy. For a strictness of SD equal to 1, we assume the economic impact is just the average of the various projections to date.

For this to analysis to hold true, other assumptions are necessary:

1. The economic recession is unavoidable, and Mexico will be severely impacted by the global downturn. Tourism alone accounts for 8.5% of Mexico’s GDP (OECD, 2017) and this activity has evaporated overnight. An additional 2.7% of Mexico’s GDP comes from remittances, which will also be severely impacted by factors outside of Mexico’s control, such as the economic shutdown in the U.S.
2. The Mexican government lacks both the means and the competence to swiftly provide the economic relief needed to offset the economic effects of SD. Approximately 60% of the country’s population is employed by the informal economy (INEGI, 2018). This sector of the economy heavily relies on in-person activities that lack social security or unemployment benefits. This means that the government cannot simply use the country’s current social safety net to provide timely government relief that 6 out of every 10 Mexican workers would require. Existing mechanisms to provide a social safety net and economic relief would need to be adjusted in order to help the most vulnerable populations. However, the Mexican government has indicated that it won’t issue additional debt, implement a fiscal reform or pursue additional spending due to COVID-19 (France 24, 2020).
3. The economic impact of social distancing is increasing at a marginally increasing rate[JAJ1] [PD2] [PD3] . In Mexico, there are approximately 5 million SMEs (El Universal, 2020). These businesses have been severely impacted by the pandemic, and their earnings have disappeared overnight. As SD measures get extended, many business owners will have to fire workers to stay afloat, or in many cases file for bankruptcy. In turn, this creates persistent negative spillover effects on the rest of the economy that compound over time (JEEA, 2020).
4. Some level of social distancing is unavoidable. Even if the government decided to fully re-open the economy, many consumers and business owners would be reluctant to resume their activities out of fear of getting infected – and would self-enforce social distance. The calculations in Table 1 account for this by not allowing for a level of SD equal to 0. 
5. The limited data available shows it is highly unlikely that social distancing could directly save the lives of children under 5 years old. To date, COVID-19 has not claimed a single life under 10 years old (Worldometers, 2020), and very few lives under 20. Children have a higher risk of dying due to the economic effects of SD than from the disease itself. 
6. There are significant limitations to this analysis given social distancing could have a positive effect on children’s lives. For example, it could be the case that by preventing an overload of the health system, SD can ensure that children suffering from other illnesses can get the care they need. Similarly, functioning hospitals could also mean that newborn babies receive better care. Additionally, in Mexico, where multi-generational households are common, elderly people tend to have very important roles and often help take care of the young children that they live with. Social distancing can help save elderly people’s lives, and thus the policy could indirectly help vulnerable children cope with the crisis by keeping their caregivers safe and at home. These effects should not be dismissed and an effort to understand their magnitude would help determine the best response to the pandemic. Unfortunately, there’s too much uncertainty to properly estimate these effects at the moment, which means the estimates in Table 1 could suffer from upward bias.

Furthermore, in the context of Mexico, social distancing might not be as effective as in other countries, and could likely exhibit decreasing marginal effects. This is sensible for two reasons:

1. Housing conditions in Mexico limit the amount of SD that is possible. Approximately 2 million homes in Mexico consist of a single room, and almost half a million of those homes are occupied by 5 or more inhabitants (INEGI, 2010). Mexico City alone, which has one of the highest population densities in the world, hosts thousands of these one-room households. Under such crowded conditions, a degree of socializing is unavoidable – and thus the marginal effect of SD decreases as the policy gets stricter.
2. In Mexico, compliance with social distancing measures decreases, as the duration of the crisis increases. Economist Carlos Alberto Bandala from Universidad La Salle estimates that the average Mexican has roughly enough savings to survive 15 days without income (INFOBAE, 2020). When faced with a choice between starving or complying with SD policies, it is unrealistic to expect low-income people in Mexico to stay at home for more than two weeks.

So far, this analysis has shown that stricter SD policies accelerate the economic recession (and its impact on child mortality) at a marginally increasing rate. Figure 1 should be a worrisome projection for Mexican leaders. As social distancing policies get stricter, the human cost of such policies increases exponentially, while its marginal public health benefits are likely to decrease. In other words, as the measures that prevent deaths from COVID-19 become less effective, Mexican society must pay a higher price to implement these policies. If Mexico is not careful, there could be lethal unintended consequences for the country’s children – who have a small direct risk of dying from the disease, but who could end up paying for these preventative measures with their lives.

---