The billions that Blair is spending on his wars in Afghanistan and Iraq could have been better spent on schools and hospitals here at home.
Is this true?
For the billions we're spending in Afghanistan and Iraq, we are buying a chance of democracy in those countries. Democracy doesn't just make people happier. It also makes them richer. This matters, because the discounted present value of a rising income stream can be a very large sum indeed. Even the small chance of such a sum could be worth billions. If so, the billions we're spending in Iraq and Afghanistan is money well spent.
By contrast, spending on schools and hospitals might not be effective. Research by Ludger Woessmann (such as this) and Eric Hanushek shows that, across developed countries, there's little relationship between education spending and outcomes.
In the UK, recently, the link between spending and outcomes also seems weak. This paper (pdf) estimates that spending in the Excellence in Cities programme had a "positive but very small" effect on results. Howard Glennerster says primary school performance improved in the early years of New Labour even though education spending fell as a share of GDP - and as spending rose, the rate of improvement tailed off .
Diane Abbott should know this. One of her concerns - rightly - is the failure of black boys in school. On this count, and as a "socialist" she should therefore be aware of John Roemer's work on equality of opportunity, in which he estimates that, in the US, an equal opportunity policy requires that nine times (pdf) as much be spent on schooling blacks as whites. Why the huge difference? Because education spending has little impact on results.
Again, the same might be true in the UK. This note (pdf) says expensive City academies "may be the only way to bolster performance among the hardest-to-reach pupils.
Nor is it clear that there's a great link between health spending and outcomes. Some OECD figures are here. I reckon that, among the 31 developed countries in their sample, the correlation between health spending as a share of GDP in 2003 and life expectancy was positive, at 0.34, but barely statistically significantly different from zero (p=6.4%. t=1.94).
Now, everything I've said here is controversial. My point is merely that Ms Abbott's claim is highly dubious. It requires a lot of arguing for, and is sensitive to differences in assumptions - for example, about the probability of democracy in Iraq and Afghanistan or the productivity of health spending.
Why, then, does she seem so strident about such a suspect claim? Could it be that she thinks an increase in Labour's client base - public sector workers - is inherently good, even if it doesn't much improve health or education? Or could it be that she implicitly thinks it's better to spend money on English people than Iraqis? How odd that an article in The Voice should carry a racist undertow.