Why this site exists

None of the data here is mine. It belongs to the World Bank, Eurostat, the ECB, the BIS, Statistics Canada, the USGS and the others named on every page. What I have tried to add is smaller than a dataset and, I think, still worth doing: keeping the source and the licence attached to the number.

I teach international business with R and Python, and I write research on trade, value chains, corridors and geoeconomics. The same thing happens every term and in every project. The data exists, it is public, and it is genuinely well made by the institutions that publish it. Then a week disappears into finding it, reconciling four different country codes, working out whether a figure is in millions or units, and discovering afterwards that the licence did not permit what was already done with it.

That week is not research. It is not teaching either. It is the part nobody writes up, repeated in parallel by people who will never meet, and it is where quiet errors get in.

What this is not

It is not a new dataset. It is not a new index, and it does not rank countries by anything I invented. Where a number is derived rather than collected, the method is written down and the derivation is labelled as one. Where a source could not be verified, its data is held back rather than published on an assumption.

What it tries to do

Three things, none of them clever. Put the sources in one place with their units reconciled. Keep the licence with the data all the way to the download, so a person can tell what they may do with it before they build on it. And be explicit about what is missing, because a gap you can see is more useful than a gap you cannot.

The licence gate is the part I would defend. It is not a paywall — everything here is free — and it exists because "open" and "you may do anything with this" are not the same sentence. Some of the best data on this site is non-commercial only. Saying so plainly seemed better than letting someone find out later.

Who this is for

I teach in Montréal, and my students can reach almost any dataset through a library subscription they never see the price of. That is not the ordinary case. For most students and researchers in the world, the barrier to working with international economic data is not skill or interest — it is a licence they cannot buy, an API key tied to an institution they are not part of, or a paywall between them and a number that a government published for free.

I have in mind particularly the students I work with through Data Science for Women in Africa. They write good code and ask good questions, and then hit a wall that has nothing to do with either. This site is a small attempt at removing one brick from that wall: the data here is free, its terms are stated plainly, and nothing about it assumes you belong to a wealthy institution.

That is also why the limits are written down as carefully as the findings. Somebody working without a subscription cannot cross-check a number against three commercial sources. The least this site can do is be honest about what it does not know, so that what it does know can be relied on.

Corrections welcome

This will contain mistakes. A concordance will map a country wrongly, a unit will be off, a licence summary will be too generous. If you find one, I would rather know. The methodology note on each dataset says how its figures were assembled, which is the fastest way to check whether I have got it wrong.

Thierry Warin, PhD — Full Professor, HEC Montréal. Scientific Director of Mondo, in a voluntary capacity.
Mondo is hosted at CIRANO, on servers in Montreal, and powered by QSnxt.ca.
warin.ca · ORCID · GitHub