Not So Big Data Blog Ramblings of a data engineer (or two)

Wealthy countries grow more slowly than poor countries

Hey! So while we’re still experimenting and learning how to use Jekyll, I decided to write a quick post to become familiar with the post-writing workflow and reacquainting myself with Markdown. For this brief post, we’ll be looking into something mildly interesting I found while playing around with an open database.

I happened upon the Total World Population open database that is published by The World Bank (feel free to go take a look). While originally my idea was to look at the population growth of some countries over time, I discovered that the World Bank had added something interesting to the dataset - the aggregate population of countries that belong to each of their declared Income Groups, among some other notable additions which I’m sure I’ll be diving into for future posts.

For now, I calculated the population growth rate for each of these income groups, and created the visualization below (extra info! As an experiment I used PlotNine, a ggplot2 port for Python that has really impressed me):

population growth per country

Before we ask some questions, something to remember:

  • These populations are according to country economy, and do not track individuals or families. So we cannot infer anything more than what is happening within entire countries, grouped into income groups.
  • I am not accounting for countries whose economies have changed income groups over time. These do happen fairly often - on the first of July each year the analytical classification is revised by the World Bank. Each time a small handful of countries (typically 10 or so) have their classification changed. So when you see ruffles or spikes on the graph, it could’ve been an event or simply a country changing classification. I’d like to look at the general trend over a number of years, so this shouldn’t be too much of an issue, but I think it’s important to point out.

It’s rather clear that the general trend is downwards for all Income Groups besides “Low income” countries. What’s really curious is how the decline in population growth rate across “Lower middle” to “Upper middle” income has occurred at a similar rate. As is the fact the that the “Lower middle” to “Upper middle” countries diverged from a near-identical point in 1965.

So why is this the case? I’m not sure - there could be a large number of interconnected reasons. So, some speculation:

  • Having children in wealthier countries is more expensive.
  • There’s been a cultural shift to smaller families since the 1960s.
  • There has been a shift to having fewer children that are more highly qualified, than many children who are not. (i.e. specialization vs strength in numbers).
  • Richer economies are becoming poorer (unlikely)
  • Increased prevalence of family planning over time.
  • Increased demand for women to enter the workplace since the 1960s.

Although, this is of course pure speculation. I warrant you could write an entire thesis on why this graph looks the way it does if you had the motivation. Still, it is interesting.

Till next time,
Michael.