2014 is latest possible base year due to WIOD-SEA data

[t-youngman]

The model uses WIOD-SEA data from 2014 for a few investment and employment variables. The project this data comes from has stopped so the data is no longer being updated. Replacing this with an alternative source is required to run the model from later years.

Other data sources mostly go up to 2020, 2021 or 2022. For these it is a case of downloading the latest releases.

Eurostat is used quite heavily - will need to check that Brexit doesn’t cause issues with UK data coming from Eurostat in recent years.

[t-youngman]

Could the OECD IO tables replace some WIOD-SEA variables?
ONS data can replace most Eurostat files

[jose-moran]

I'm not sure. The problem is that SEA data is not the same thing at all as the data in IO tables. The SEA data tells you things like number of employees or the value-added split between labour and capital. I am sure this data can be found elsewhere, but I think it would mean individual datasets for each country (or, in the best case, some Eurostat database for the EU). This then leads to the problem of checking whether those datasets are harmonised.