COVID-19 and the Economy: insights from a cross-country view

I thank my friend Luca Sartorio for giving me access to his data on the stringency of COVID- related lockdowns, as part of his reseatch at the Torcuato di Tella University in Buenos Aires.

COVID-19 has upended the entire world in 2020. The economic news from Q2 of 2020 are bleak: most countries in the world witnessed a sharp contraction of their GDP growth, resulting from the crisis. In this blog post, I correlate this decline in GDP growth with characteristics of the pandemic (e.g., incidence and death rates in countries), and the macroeconomic conditions in select countries (e.g., foreign dependence, stringency of the lockdowns) in select countries. Due to the unavailability of comprehensive data on Q2 2020 growth, I was only able to gather data on 70 countries. Some regions are thus over-represented in my dataset (e.g., OECD countries) and others are under-represented (e.g., Africa). For a complete list of data sources, please see the last section.

GDP declined sharply in most countries in Q2 of 2020

As the figure below shows, the growth in GDP for Q2 of 2020 was negative for most countries in this dataset. Note that this dataset is not comprehensive and only contains 70 countries. While the data is incomplete, we can still conclude that in most regions of the world, GDP contracted very sharply in Q2 of 2020, as the repercussions of the COVID-19 crisis was felt everywhere. From the figure below, it looks like South America suffered the largest contraction in GDP growth, while Africa suffered the least.

Want to read this story later? Save it in Journal.

Correlating the pandemic and economic growth: The hardest-hit countries suffered the most

The first figure below plots GDP growth in Q2 of 2020 against the total cases per 1 M population for the countries in my dataset. We observe a negative relationship between growth and total cases per 1 M population. That is, the more cases a country registered, the worse its GDP growth was in Q2 of 2020. This supports the hypothesis that the hardest-hit countries are the ones that suffered the most in terms of GDP growth.

The next figure plots the GDP growth in Q2 of 2020 against the total number of deaths per 1 M population, and shows a similar, if not sharper, negative relationship than the one I describe above. That is, the countries that reported more deaths did suffer a greater decline in GDP growth in Q2 2020.

The two figures above make intuitive sense: the harder-hit countries not only saw their resources diverted to combat the healthcare crisis, but probably also suffered from the economic consequences of the crisis. In the next sections, I attempt to understand which of these economic consequences are more correlated with the decline in GDP growth in Q2 2020.

Foreign economic dependence plays a role: more so for trade than tourism

My first idea was to look at tourism as percentage of GDP, and correlate it with the GDP decline in Q2 2020. The hypothesis was that the more a country was reliant on tourism, the more likely it was to see its GDP sharply decline. The figure below tells me otherwise: there is no downward relationship between the decline in GDP and a country’s reliance on tourism. This is probably because the hardest-hit countries are large industrialized countries that do not depend on tourism as much as the more developing economies.

Next, I turned my attention to dependence on trade, as measured by the percentage of GDP that is accounted for by trade (sum of imports and exports). Here I do find a negative relationship between GDP growth and dependence on trade: the countries that depend on foreign trade were worse off economically than those that weren’t. Again, due to the fact that large manufacturing countries are hardest-hit, this makes sense, as supply chains and global trade was brutally disrupted by the pandemic.

Lockdowns and other government measures are negatively correlated with economic growth

To understand the relative stringency of the lockdowns and other government measures to curb the pandemic, I borrow Oxford University’s stringency index, a composite score that looks at various measures undertaken by governments around the world. These include border closures, imposed lockdowns, school closings, etc. I got access to clean and organized data from this research project at the Torcuato di Tella University in Argentina.

The first figure below looks at the average stringency index across continents as of September 1st, 2020. As you can see, Europe had the lowest stringency measures (probably because most of the European countries relaxed restrictions during the 2020 summer), while the highest stringency measures were recorded in Latin America (in some countries like Colombia and Peru, the lockdown imposed back in March is still in effect). Here, again, it is worth noting that Africa is under-represented (with only the top 10 hardest-hit countries on the continent). It is entirely possible that taken as a whole, Africa has the lowest stringency index in the world given its relatively low exposure to the pandemic (fingers crossed it stays that way!).

Next, I look at the correlation between GDP growth and stringency index, and a mild negative relationship for countries with indices below 60, and a sharp negative relationship for those with indices above 60. This supports the hypothesis that stringency measures, while necessary to contain and curb the pandemic, inadvertently had a negative impact on the world economy.

The old tale applies here too: Correlation does not imply Causation

What I did in this blog post is look at the correlation of only certain factors with economic growth in Q2 of 2020. The above descriptions by no means conclude that the disease burden or stringency measures caused the economy to plunge. There could be a myriad of other factors at play here, such as the composition of the economy, testing and contact tracing, the truthfulness of government when reporting both disease and economic data, etc.

A more comprehensive view would include a larger set of variables, and would conduct a regression analysis in order to isolate the effects of each factor. Still, I do think the above analysis is useful for the curious like me :)

If you’ve read all the way to here, thank you! I would love for you react and share if you can! You can also get in touch with me at kenza.bouhaj@gmail.com if you’d like access to the data or the code.

Data Sources

Coronavirus Data (Cases, Deaths): Worldometers Coronavirus tracker

Q2 2020 GDP Growth: OECD Data Library and other country-specific sources (news releases, Central Bank data, etc)

Tourism and Trade as a percentage of GDP: World Bank Data

Stringency index: Data from the Torcuato di Tella University Research Project.

More from Journal

There are many Black creators doing incredible work in Tech. This collection of resources shines a light on some of us: