Guido Fawkes claims flat taxes lead to growth. They do not as data from the IMF shows. Flat tax countries are growing at a slower rate than the rest of the world.
Earlier today, Guido Fawkes drifted from his muck raking comfort zone to offer an economic theory based on three data points. In response to blogs from myself and Labour List about Britain’s slow growth performance, the normally sharp blogger implied that the Baltic countries’ flat taxes were the reason behind their high growth rates.
Any statistician will tell you that a correlation does not make a causation. The problem for Guido is that he doesnt even have a correlation. There are 21 countries with flat taxes. Projections from the IMF show that their average expected growth rate for 2011 is 4.1 per cent.
By contrast, the global economy will grow by 4.3 per cent this year so flat tax economies are underperforming the rest of the world.
There’s a decent debate to be had about why the Baltic countries are growing faster than other European economies but to claim that it’s all down to their flat tax makes about as much as sense as claiming that heavy taxation is bad for the economy – another Guido staple.
13 Responses to “Flat taxes do not deliver growth”
Will Straw
Thanks for the comments:
James – the onus is on Guido to support his claim with proper academic evidence rather than just asserting that heavy taxes dampen growth or that flat taxes enhance it. The Lane Kenworthy blog puts a significant fly in the ointment of that theory. If you have an academic paper showing a causal link between low taxes and high growth which shows that Sweden & Denmark are just outliers, I’d love to see it.
Oxford Kevin – Exactly. As I pointed out to the TaxPayers Alliance on twitter, the GDP collapse in 2009 was -18% in Latvia, -15% in Lithuania and -14% in Estonia.
13eastie – A regression with a dummy variable of flat taxes (as the independent variable) and growth levels (as the dependent variable) would end up looking exactly the same as the averages above. The p value would fail to reject the null hypothesis precisely because there is no statistically significant link between the two variables.
13eastie
Ludicrous study design e.g. sample size of 1 (2011) gives any null hypothesis a decent chance of surviving, Will.
But not for the right reasons.
That Guido’s message is garbled doesn’t necessarily make it wrong, but your argument against it shares and exceed its failings.
You’re playing the man, not the ball.
Leon Wolfson
@3 – Right, and he hasn’t. It’s a knife in the kidneys.