(Part 1 is here)
In my school days my friends and I had a kind of running joke amongst us where the punchline was ‘Answer the question set!’ referring to the first challenge of any exam question, that of actually understanding what you’re being asked! I feel our authors could have paid more attention to this concept…
To be clear, when referring to “ground rent” in this paper we mean the net present value of the land devoid of physical improvements such as buildings.
Note we have not complicated the models by differentiating between improvements and the land’s “unimproved” value…
The first of these quotes is alarming. Why the authors redefine ground rent to be something other than….well…..rent for ground, I have no idea (it certainly doesn’t help understand the topic at hand, as we’ll discover), and in particular, removing the time continuous nature of ground rent is just asking for trouble. The second, however, is an insult to academics everywhere. Heaven forbid a paper challenging the idea that land has unimproved value should stoop itself to actually modelling the thing it is challenging! That just wouldn’t be cricket, what?!
If they are going to not bother actually doing what they said they would, they at least have a good reason, right?
We consider the problem of a risk-neutral prospective purchaser of land who compares his opportunity cost to the value of some parcel:
E(V) = E(I) − P (1)
Where V is the lands value, E(I) is the expected income of the land, and P is the land’s price.
Note there are some assumptions that become evident here. Pay attention to E(I), the expected income. An important question to ask is ‘Where does this income come from?’. Whatever it’s magnitude, and however accurate the expectation, it must have a source. Caplan and Gochenour’s answer to this question limits their analysis.
Landowners will engage in costly search to discover more information about the land (Stigler 1960): for instance, if the land contains natural resources, or if the land can support a new type of building. Rather than think of natural resources as an inherent quality of the land, from an economic perspective it is more accurate to think of them as being produced through the process of discovery and efficient extraction. [emphasis added]
So we are all aware, this is the crux of the entire argument put forth by our authors. This is what they’re trying to ‘prove’ so to speak; the basis for saying that LVT is distortionary. This statement is rather easily dismissed, however, by considering that the value of performing a search depends greatly on what is there to be found and by definition you don’t know what’s there to be found until you look. Thus, searching, like pretty much every other economic activity becomes more or less effective depending on where you search.
You can spend as long as you like searching a used ice-cream tub for some delicious vanilla goodness, but you’ll wind up disappointed. Search the freezer aisles of Tesco, however, and it becomes a very simple, cheap and rewarding endeavour! Which leads us back to the basic principle: The extent that the effectiveness of an activity depends on where it is performed is the extent to which the more effective locations will charge for the privilege of their use.
As this is really the entirety of their argument, we could stop here…..but where’s the fun in that?! 😉
He searches for higher value uses for the land until the search costs exceed the expected beneﬁts,
V = E[I(S)] − E[C(S)] − P (2)
Where E(I(S)) is the expected income for a given level of searching S > 0 and C(S) is the search cost.
Now we’ve added the costs and incomes arising from searching, the question of where the income comes from becomes ever more relevant. Why is it that performing a search for greater capabilities of a given plot results in an expected increase in income from ownership? A couple more quotes:
Any “rent” that would accrue to the landowner after he discovers and brings to market any properties of the land are really a quasi-rent (Marshall 1920): while superﬁcially similar to a rent, the reward of charging P > MC for, say, natural resources discovered on the land are a necessary incentive to search for these resources.
Since any resources found would be taxed at 100%, there is no beneﬁt to
searching, and in equilibrium no search occurs and the price of land becomes
Hopefully by now you’ve seen the problem here. Caplan and Gochenour see the additional resources capabilities found as simply a thing to be sold on. To them, the income comes from selling up. There is no consideration of the actual process of production that must be employed to realise the discovered opportunities.
The authors are only considering one model of organisation, that of landlord-directed production and trade. If all production and trade decisions are made by landlords, then taxing the value of the land does indeed leave the landlord with no incentive to get involved with production and trade, including seeking out new opportunities. Searching for new capabilities of a plot is not unique in this context.
The landlord-directed model, however, is fallacious as I’ve discussed before. It is land-users who not only ultimately decide how to use land and whether to search, but are the ones actually performing the production and the searching. The question is whether the taxation of ground rents remove the incentive from the land-user to utilise a location and seek further opportunities within it.