2.1 Negative net flows

[jm-rivera]

Some countries already pay more in outflows than they receive in inflows, and the trend is showing many more countries could go that way.

We need a way to visualise / present the data which shows that many countries are in negative net flows. We may want to show countries that are close to negative net flows as well... though only if it makes sense.

This chart is an example designed for exploration... but it's not very clear - not least because it uses GDP in order to plot the data in the same scale.

The point of a chart here would be to show the number of countries and the extent to which they are in the negative space. Could we consider a choropleth map? Just an idea, open to whatever makes sense.

[Mattie-P]

Reluctant to produce a map. Worry that, alongside all the usual issues with using maps as a data vis, it fails to show (1) how the situation changes over time, and (2) the scale to the issue.

Currently playing with the ideas of both a beeswarm comparing 2010, 2022 and 2025 (Y axis) compared to net transfers on the X axis. Could also do some form of binned distribution along the x axis (i.e. 0%<=x<0.5% of GDP, 0.5%<x<=1% etc). I think because 2010 will be more or less all positive net flows, and then around 25 odd in 2022 will show net negative, and 35 or so in 2025, both these graphs can clearly depict a movement of from positive to negative net flows. These thoughts are inspired by Luca's gender equality graph here.

I am unsure whether to show all the positive values just yet. On the one hand, if you only show negative values (but in absolute terms), its a bit more intuitive for the reader to read from left to right and see the issue getting worse as you go further right. But then with annotation we could easily get around this. On the other hand, if you show the positive net flow countries, it gives more context to the situation, and will (hopefully) show more of a movement towards the negative net transfers side. Will have a play around once I have the data.

There then is the optionality to add in colour on either region or income level (I think the latter is better). We could use this to emphasise the LICs and LMICs that recieve negative net flows in each time frame.

My main concern is the use of GDP as a scaler. Without a scaler, the axis are too broad. Need to have a play around with this as well once I have the data.

FYI @jm-rivera @lpicci96

[Mattie-P]

The core issue with this data visual is we are trying to show too many things at once. Based on the initial issue, we wanted to show: (1) the number of countries that have negative net transfers (2) how the number of countries has changed over time (either a time series or a two/three year comparison) (3) how the scale of this issue has worsened i.e. have flows got more negative (and potentially will worsen in the future using projections).

The text surrounding this chart has changed as the drafts have gone on. Now, it states: "In 2022, 26 countries paid a total of $48 billion more to service their external debts than they received in new external finance – meaning they were already in negative net transfers (NNT). This could increase to 34 countries in 2025, paying a total of $75 billion in NNT if financing flows do not increase." As it stands, based on this, I think the beeswarm is the best chart.

To try answer this, I tried multiple chart types (NB: because of time constraints, I have done very limited formatting. Luca, do not come at me for not yet applying your guidelines...):

(i) Beeswarm -

Graph shows a comparison of 2010 to 2022 when looking at net flows as a share of GDP. Not the most intuitive scale, but offers a good way to scale each countries unique situation, and place all charts on the same scale. Using logs does not work well given the number of countries at, or near, 0.

There were too many data points to accurately show all of this trend. I capped the x axis at 2% (as it ranges up to around 40%) which worked to an extent, and we are only interested in the countries near the 0 line. I was not fully convinced on this chart type but now I have tried other styles, feel it is actually one of the better chart types

After deciding this chart was too busy, I wanted to find a way of removing some of the 'busy-ness' and decided to try a histogram...

(ii) Histogram -

Again, this chart shows net flows as a share of GDP to create some scale). Here, i've capped the x axis at 5%.

My issue with this graph is the obscure chart type. It is not instantly clear what this graph is showing and risks being too confusing. I do feel it accurately shows the trend, but quite expensively. It also has limited interactability (is that a word?) i.e. user cannot see which countries are at the lower end, unlike the beeswarm.

Based on this, I figured it would be better to try a simpler chart type and remove one of the elements listed above. I decided to focus on (1) here.

(iii) Scatter - I tried both log scales and scaling by gdp and neither charts really worked. Happy to revisit this chart if needed, but unsure this is the way to go. It fails to show the increase in number of countries beyond the line.

Then decided perhaps it was better to show trend (3), by showing how the total amount of net negative flows is increasing. Tried this as both a steamgraph and an area chart, but decided the area chart is more intuitive.

(iv) area chart (ignore the breakdown by continent) / line chart- Chart clearly shows how negative net transfers has increased over time, but it fails to show how many/which countries are holding up this trend. I am unsure if (iv) is the way to go. I do not think this chart is relevant to the text anymore.

Please any feedback would be very very welcome @jm-rivera @lpicci96

[Mattie-P]

Final chart