Dodgy TUC Graphs, GDP by Expenditure Edition
which they use to make various claims about GDP by expenditure, including this zinger:
There is little evidence that government consumption rose as a share of GDP in the ten years before the crash, and a consistent share of GDP of under 25% is well below that of many other advanced economies.
This is a highly misleading claim, and it’s odd that left-wingers would want to make this claim. “After thirteen years of New Labour, what did we achieve? No increase in government spending!”
Here is the trick the TUC are pulling. Imagine an economy with just consumption spending where we start with nominal GDP of £1000/year. The first year is like this:
Government consumption spending is £100, which pays for the education of 10 students.
Market sector consumption spending is £900, which pays for 900 apples.
Now fast forward a few years, but presume nominal GDP is held constant:
Government consumption is now £200, which is still educating just 10 students.
Market sector consumption is now £800, but there’s a massive productivity improvement on the farm, so this pays for 2000 apples.
Total output has gone up, and you get to pick your statistic to fit your bias. As a proportion of total output, government consumption has fallen. As a proportion of total spending, government consumption has risen.
That is a crude reflection of what happened under New Labour, which raised public sector pay (in education and the NHS particularly) without achieving a commensurate increase in measured public sector productivity. That’s a perfectly defensible thing to do, if you think, say, public sector pay was too low before. Plus, measuring public sector output is a particularly meaningless exercise; there have been endless arguments about whether the kids are better educated, how you measure health outcomes, and so on.
This graph shows the change in nominal government consumption as a proportion of nominal GDP, and yes, ignoring everything after 2008 is reasonable because of the erratic nominal GDP growth since then:
The TUC paper also uses the volume series to draw demand-side conclusions about the trend in household consumption and investment, which just does not make much sense as far as I can see.