In my last blog, I looked at the stringency index of 30 countries. This is an index devised by the Blavatnik School of Government at the University of Oxford and is an aggregate of 9 measures of government enforcement on lockdowns, and closures of places where public gather in large numbers. We noted a cluster in the bottom right quarter. During the course of the next week, I took all 166 countries and made the same chart, this time using two different colours for stringency index less than, or greater than 50%. I obtained the nice chart displayed in Figure 1 using the ggplot package in RStudio.
The scatter plot is for 166 countries. The data is from 31st December 2019 to July 30th 2020, a period of 212 days. For the 40 other countries listed in the database, either the stringency index or the gdp_per_capita was not available. The three large red dots, represent Singapore, Norway and Luxembourg. The five blue dots represent Brunei, UAE, Ireland, Qatar, and Kuwait. The other 158 countries are in the bottom half of the quadrant, i.e lower gdp_per_capita. Nineteen of the 158 countries are in the left half of this group, that is in the lower stringency group. The remaining 139 countries are in the high stringency group. This represents about 84% of the countries and would lead one to conclude that most countries are enforcing lockdown security measures. The four countries with stringency index less than 25% are Belarus (8.33%, gdp_per_capita is 14.68%), Burundi (22.22%, gdp_per_capita is only 0.6%), New Zealand (19.44%, gdp_per_capita is 30.86%), and Nicaragua (11.11%, gdp_per_capita is only 4.55%). Burundi in spite of low GDP per capita index has a decent stringency index. NZ has a low stringency index, for the healthy GDP per capita value that it has, but that is perhaps because it is an island nation, and the citizens understand what is to be done in an emergency. This is, however, too small a dataset, to compare with those with higher stringency index. If we count the number of countries with stringency index less than 50% we get 22, with 144 in the higher than or equal to 50% stringency group. The mean gdp_per_capita, for the first group at 29.465, is twice that of that for the second group which 14.799. The mean stringency index of the first group, at 36.77 is half that of the stringency index of the second group, which is 74.60. This does suggest, that countries with higher gdp_per_capita, can afford to have more lax lockdown rules.
Our future data exploration would seek an association between stringency index and other variables. In the chart below, we have plotted Stringency Index, v. median value of new deaths. The USA with a stringency index of 69% and a gdp_per_capita index of 46.37% has an outlying value, of 500 deaths per day, which is most difficult to explain. India has a stringency index of 76.39% and a gdp_per_capita index of 5.5%. There are a few other outliners such as Brazil, Peru, UK, and Iran. Most countries are witnessing less than 50 deaths a day.
For my next blog, I wish to compare the emphasis on testing between select countries, over the same seven month period. Many countries did not have cases in the early months, but the comparison would be insightful for the summer months and after. Testing requires government funding, and one confirmatory scatter plot could be total_tests v. gdp_per_capita for all the countries, to see if there is a linear association between higher gdp_per_capita and a higher number of tests administered. Poorer countries cannot be blamed for not testing everyone. Even if they did, the health care facilities are not there for everyone, and it would be far better to take care of oneself at home.
Author:
Dr. Badri Toppur
Associate Professor, Rajalakshmi School of Business, Chennai
Email – badri.toppur@rsb.edu.in