A Search-Theoretic Critique of Georgism

Wednesday, February 15th, 2012

Bryan Caplan presents a rather esoteric search-theoretic critique of Georgism:

Economist Henry George famously advocated a 100% (or near 100%) “Single Tax” on the unimproved value of land.  Many modern tax economists, most notably Joseph Stiglitz, conclude that George’s logic was sound: Since the unimproved value of land is perfectly inelastic, even an expropriatory tax is non-distortionary.  Economists’ main objections to Georgism are merely that (a) it is difficult to implement in practice, and (b) politically impossible.

My co-author Zachary Gochenour and I have a new working paper arguing that the Single Tax suffers from a much more fundamental flaw.  Namely: A tax on the unimproved value of land distorts the incentive to search for new land and better uses of existing land.  If we actually imposed a 100% tax on the unimproved value of land, any incentive to search would disappear.  This is no trivial problem: Imagine the long-run effect on the world’s oil supply if companies stopped looking for new sources of oil.

As I noted there, this comes down to a problem of defining unimproved:

So, a tax on the unimproved value of land doesn’t work when things that dramatically increase the value of land — finding it in the first place or finding better uses for it — aren’t treated as improvements and do end up taxed. OK.

Actually, isn’t the real problem that the State receiving these tax revenues isn’t a profit-maximizing firm? An enlightened sovereign would pay geologists to locate oil on his land, after all.

One commenter felt that there was no “new land” to be found and that there was no difference between rent, mortgage payments, and land taxes — but I disagree:

I don’t know how much “new land” is still out there for the taking, but much of the English-speaking world was “new land” not so long ago — at least as far as tax-collecting governments were concerned.

Now, if we want to compare and contrast (a) a leasehold, (b) a freehold with a mortgage, and (c) a freehold with a land value tax, we need to think about how the “rent” is determined and how it changes.

In the case of a leasehold, with literal rent, the lessee has no incentive to improve the site, because the owner will reap the benefits; the owner can increase the rent to reflect its new higher value.

In the case of a freehold with a mortgage, the nominal owner has every incentive to improve the site, because he owes the bank a fixed amount; he can sell or rent at the increased value.

In the case of a freehold with a land value tax, it depends how we compute the land value. Any improvements that aren’t exempted push us toward the leasehold case. If carving a new homestead out of the wilderness means you have to pay full rent on the newly arable land, no one will do the work of homesteading. If discovering oil means you have to “rent” the site of a potential well, then no one will go to the effort to discover oil.

These problems aren’t insurmountable; they’re just issues that come up if you propose a simple land value tax without simultaneously proposing a few complications.

Dan Klein shares Fred Foldvary’s “fairly clear presentation of a geo-rent tax proposal” in Econ Journal Watch — which is “fairly clear” to EJW readers.

I find these discussions can often devolve into near-theological debates:

One problem with using Georgist language is that it raises the philosophical question of just what unimproved means — which we can debate endlessly — when what we really want is a tax (or rent, or mortgage) scheme that aligns incentives efficiently.

It doesn’t really matter what the unimproved value of the land under Manhattan or Tokyo is — or what that even means — when what we really want is a negligible marginal tax rate on improving existing lots — or creating new lots, however that might be done.

As I’ve mentioned here before, local government could operate as a for-profit firm, like a mall-management company, with home plots as leaseholds rather than freeholds — or as something in between:

Our ideal division of property rights would align incentives so that a tenant living in a house would gain a dollar by (wisely) investing a dollar (or a dollar’s effort) in the house and would lose a dollar by neglecting maintenance by a dollar (or a dollar’s effort), and the management company running the larger neighborhood or city would similarly gain or lose depending on how well it provides “public” services.

I suggest that we could achieve this by having the city sell semi-freeholds encumbered by a flexible tax-like rent, paid to the city — but this quasi-property tax wouldn’t be based on an individual plot’s value (with improvements). Rather, it would be based on the individual plot’s acreage (area) and the average price per acre of the surrounding land.

Thus, rents would go up as property values go up, but no one property-owner would face the disincentive that comes with ordinary property taxes — doubling his own property’s value through improvements wouldn’t double his property taxes; it would only increase his property taxes infinitesimally.


  1. Sconzey says:

    Well the obvious solution is that the tax on land is based on the value at discovery, homestead, or registration and then paid in perpetuity. It’s only inflation, which means people don’t usually consider this solution.

  2. Isegoria says:

    I’m not sure what you mean by “it’s only inflation,” Sconzey.

  3. Sconzey says:

    I mean a not inconsiderable component of the long term rise in rents is due to a decline in the value of the currency.

    Of course now I’ve had a few beers it’s clear to me that the value of even unimproved land ought to rise as demand rises.

  4. Foxmarks says:

    George’s argument depends on value rising with demand. The notion of land title is predicated on a desire for exclusive use. If there was no demand for a parcel, it has no unimproved value and yields no LVT.

    The leasehold is the proper framework for applying George’s theory. What nature created, all men own in common. Government, as the agent for the commons, cannot convey title away from the commons. The freehold is an invalid construct.

    The unimproved value can be determined by markets. What another would pay for the usufruct is the “unimproved” value. That bid is the basis for assessing LVT. There would always be an incentive to find “new” land upon which nobody else would bid. There would be an incentive to bid low, as the bid determines the tax.

    By George’s reasoning, the oil in place *is* communal property. 100% rate LVT is morally justified, but probably unneccessary.

    The market price for oil includes production and delivery. If I discover a barrel of oil on my lease, you would bid something less than the market price of a bbl to obtain rights to my plot.

    At 100% rate LVT, the maximum possible benefit of discovering the bbl equals the cost of producing and delivering it to market. In other words, if you can drill better than me, you have an incentive to bid more to assume my lease than I would to keep it. If I am the more efficient driller, you cannot bid the benefit of my discovery away from me.

    Since there is profit to be made even at 100% LVT, people would still keep looking for oil.

Leave a Reply