The episode of BBC Radio 4's Analysis programme which aired tonight at 1930 GMT had a scary false dichotomy between the gold standard and our current monetary supply system.
It is true that it doesn't make sense to have a money supply without a physical referent, but that doesn't make the gold standard any more logical. If money supply was related to the amount of gold that has been discovered, the absurd situation arises that discovering a quantity of a comparatively useless metal would increase the size of the economy, and once all gold has been discovered, the economy cannot grow.
Relating money supply to the price of gold is nonsensical, because the price of gold must be in units related to the price of gold (as money is on the gold standard). Or, if the price of gold could float freely, money supply would be related to a free floating variable.
In the gold standard, gold had to be shipped. Ridiculous. All this relatively useless stuff did was sit in vaults wile it wasn't being transferred betwwen them. No wonder governments realised it was utterly stupid, and sold the gold without admitting it.
But the gold standard proponents were right that money supply should not be unrestrained. It is a starting point to say that it should be related to the amount of goods and services being produced. The problem here is the problem of growth in GDP - the planet is finite so far as physical stuff goes, so insofar as it relates to stuff, money supply should not increase. Services are a different matter. Technology is a case in point. Microproccessor based technology is getting smaller as it gets more powerful. It sis clear that, over time, we can get more service per unit of stuff. If we could quantify service, we might have a guide to the quantity of money that there should be, but how do you compare a basic human need (nutrition, say) against a higher order need or service (finding out where the nearest restaurant is using your smart phone).
For me, human well-being is the key and for the moment, I'm thinking money supply should be a function of the number of people alive and their age. I'm sure this is less than perfect, but it is a starting point.