Ricardian rent theory

Lex, a loyal MR reader, asks:

My understanding of Ricardo’s theory of rents is that they are based on the least productive land used.  Shouldn’t rent be based on the most productive land not used?  Is that merely a semantic difference?

Ricardian rent theory is strange. Let’s say you have three pieces of land, yielding 3, 5, and 10. Rent will be 7, or 10-3, with apologies to Samuel Hollander. The marginal piece of land, the one yielding three, is allocated to wages and profit from capital.  (You can treat profits as an afterthought in this model.) The rest goes to landowners.

Why might that distribution result?  In the Ricardian model you can (and do) augment labor through Malthusian breeding.  So labor doesn’t have much bargaining power in the long run.  Any move in favor of wage labor creates more people and lowers the wage of labor to the return on the marginal unit of land.  If you have some extra land, even if it isn’t incredibly productive, you can always breed more people to cultivate that marginal unit.  So the value of the least valuable piece of land in essence sets the wage.

Land is inelastic in supply, and thus the landowners reap the rest of what is produced.  We can also see that the last plot of land used (and not the marginal piece of land not used) determines rent and wages, because it is the actually used piece of land which regulates population and the return to labor.

That is my quick and dirty take on Ricardo, noting that "what Ricardo really meant" became a cottage industry about one hundred and ninety years ago…