Richard Murphy claims the multiplier matters.

 
He makes that claim in an article entitled “The multiplier shows that not all government spending is equal.”

I normally agree with Murphy, particularly since he started backing MMT. But his ideas on the multiplier leave room for improvement.

One of his main points seems to be that if government raises spending by £Xpa, and if the multiplier is large enough, then the increased economic activity will raise tax paid to government by roughly £Xpa, in which case that extra government spending may pay for itself, which is allegedly a plus. Incidentally that's a point to which Ann Pettifor has attached importance in some of her articles.

But suppose the multiplier is relatively small, and extra public spending DOES NOT pay for itself in the above sense. Why would that matter? All government and central bank need do is create and spend more base money so as to raise demand by the required amount. And creating base money (i.e. central bank issued money) costs nothing in real terms as Milton Friedman pointed out.  

As I explained in a Ralphonomics article entitled “The multiplier is totally irrelevant” over ten years ago, the multiplier, contrary to Richard Murphy’s claims, is irrelevant, at least as far as the above "pay for itself" point is concerned.