Economists tracking the impact of Brexit created a really interesting technique, that I'd never have thought of, and that I really like.
They took the other countries in the G7 and said "what combination of these economies was the UK economy most like, up to 2016?"
They took the other countries in the G7 and said "what combination of these economies was the UK economy most like, up to 2016?"
It's like a recipe, for example,
The UK = 1/5 of Canada, 2/5 of France, 1/10 of Germany, 1/10 of Japan, 1/5 of USA, 0 of Italy.
The UK economy is really like France, a bit like Canada and the USA, hardly like Japan and Germany, nothing like Italy.
The UK = 1/5 of Canada, 2/5 of France, 1/10 of Germany, 1/10 of Japan, 1/5 of USA, 0 of Italy.
The UK economy is really like France, a bit like Canada and the USA, hardly like Japan and Germany, nothing like Italy.
An original paper on this is "The Economic Consequences of the Brexit Vote". They create a "doppelganger" UK, then compare how it performed compared to the real UK since 2016. It's such a good paper. centreformacroeconomics.ac.uk/Discussion-Pap… 
So, I've been thinking.... we can use this "doppelganger" approach to estimate the impact of all kinds of things.
And the "Solver" add-in for Excel makes it really easy.
And so... … because I am somewhat unusual... … I present SimWales!
And the "Solver" add-in for Excel makes it really easy.
And so... … because I am somewhat unusual... … I present SimWales!
I've taken the GVA/head data for all the English regions, and created a SimWales -- which is a weighted combination of those regions' economy up to 2011 (when the Welsh voted to *really* get devolution going).
Wales is about 4/9 West Midlands + 3/9 North East + 1/9 London.
Wales is about 4/9 West Midlands + 3/9 North East + 1/9 London.
The Solver add-in for Excel is so good. I used to write optimisation problems like this in Python, or Matlab, or C# and it was a pain to set up. In Excel it's so simple by comparison.
Anyway, here's the result... both Wales and SimWales have grown by 15% since 2011. No significant difference.
I done Scotland too. 2011 when the SNP became the majority government. Pleasingly Scotland's economy is a linear combination of only two English regions. It's half London, half North East England. It has also not significantly overperformed its doppelganger. Just 1% more growth.
If we had UK regional economic data back to, say, 1987, we could do a good doppelganger evaluation of Scottish devolution. Twenty years of data to create a doppelganger from, then twenty years since then to see if it has made any difference.
