Without reading through "all" of the posts in this thread, I have gathered that the original author's contention was that EO can be only attributable to the land.
With that in mind, let me pose the question of an industrial property - let's say a pulp mill. Five years ago, Company ABC pays $80 million for this existing pulp mill. Three years later, for various reasons, the market price of pulp deteriorates substantially. Causes could be related to numerous issues like increased global capacity, changes in governmental regulations, more modern plants developed in the southern hemisphere, whatever. But the fact remains that prices of the product produced plunge. Would that effect the price buyers are willing to pay for a pulp mill in the US? Of course it would.
Now, does it effect land value? It might or might not - if the land is valued as vacant, and if there are alternative uses for the site, the state of the pulp industry might have very little - if any- effect on the land value. However, this situation certainly effects the value of the improvements. Thus, it is clear that External Obsolescence can be attributable to Improvements - which was the main point here.
There were many other points illustrated (locational factors already being considered in the land, as vacant, etc). But I think it is quite obvious that External Obsolescence issues can/do effect Improvements. Much of the conversation focused on EO in terms of "location" (like heavy traffic or a railroad track). However, when you use a different form of EO (like changes in economic conditions), it becomes very clear that EO does effect Improvements.
Patrick Price