This is already implemented in many areas of modern life, especially by governments (Obamacare, progressive taxation, social security, etc), and it has a major negative side effect of incentivizing poverty: either real (by encouraging the young to work as little as possible) or faked (by encouraging those with accumulated wealth to conceal it as much as possible).

My own personal view is that using starvation to incent work is acceptable only if society is unable to produce enough food for everyone; in that case, those few points to the GDP have real impact. I personally don't feel that it's okay to use starvation to incent work when that work largely produces luxuries, but that's just me, and everyone has an opinion on this one.

[1]Interestingly, historically this aphorism has been a leftist saying... that is, the rich should have to work, too. Leaving aside my controversial views of the bible, in more recent history, John Smith used it to this meaning in Jamestown, and then in the early 20th century, Lenin argued that it was a fundamental principle necessary for socialism to work. Both were talking about situations were people were starving (the early days of Jamestown and the early days of the Russian revolution, respectively) More recently, it's used as a rightest saying, arguing that the poor ought to work or starve, something that was the assumed truth in earlier ages.

I think most people would agree with you on that one. And the fact that people in "rich" countries are still food-insecure is horrible.

One problem is that there isn't a clear bright line between "starvation" (in the sense of malnutrition) and "nutritious but not very tasty food" and "nutritious and tasty but not very pretty food". And you can keep going in that vein to things like "not having a flat-screen tv", say, which I think most people would classify as quite different from "starvation". But there isn't an obvious cutoff point on the way.

Oh, and what we consider sufficiently nutritious today may not be so considered tomorrow. Both literally (at some point nutrition science discovered vitamins and then trace minerals, etc), but people nowadays also get into arguments about whether internet access is a fundamental right like food or a luxury. And whether personal transportation (e.g. cars) is a fundamental right. What will people think about access to a self-driving car 50 years from now: luxury or fundamental right?

In practice, what we consider below-acceptable standards of living (which we as a society need to subsidize until they reach acceptable levels) are a good bit higher than what was considered totally acceptable 50 years ago. That's a natural consequence of society growing richer, of course and I would say it's a _good_ thing.

But here's the question that bothers me: Had we frozen per-capita production at 50 years ago levels, while subsidizing living standards to a level above the 50-years-ago acceptable minimum but below the today acceptable minimum, such that vast numbers of people today would be forced to live below what we consider an acceptable living standard today, would that be a net gain? In the short term, clearly yes (we subsidized people at a higher level). In the medium term (to today), it seems clear to me the answer is no. In the longer term (200 years from now, say), I have no idea.

It's hard to tell even post facto, much less a priori, whether we're over-subsidizing or under-subsidizing to achieve maximal happiness. And that's even if we can all agree on a timeframe. Combine that with the fact that different people already have different definitions of what constitutes minimum acceptable living standards, and it becomes very hard to apply ethical considerations to this problem in a principled way. :(