Any land area that is greater than that required to support the existing improvements is, by definition surplus land.
Or excess... As an ag appraiser, the problem with the neat pigeon holing of "excess" and "surplus" relegates the land component to a decidedly secondary consideration when, in fact, often the land value exceeds the value of the improvements. That according to Henry Harrison some 30 years ago is the definition of a land appraisal with improvements, not a house appraisal with land. Boy, doesn't that open up a can of worms?
Again, I would go back to the client and say, "Hey, the value lies in the land (if that is the case) and will exceed the 30% barrier (if that is a reverse mortgage issue). Because I know exactly what I saw once from an appraiser in Oklahoma. The land was about 90 acres or so and the house was old and needed substantial work. To fit the narrative and make adjusting to smaller parcels easier, the land value was said to be $400 per acre...at at time the land would have sold for $1000 or more per acre. So you had 90k land with a 30k house but flipping that on its head the land was only 36k and the house was some 80k or so. The appraiser ended up getting the FHA boot in Oklahoma and seriously, showed me the appraisal and complained about how she had been "treated" by FHA. I suggested the land value was low and perhaps they were right and we kinda left it at that. It was a few weeks before I was in that area and saw the house and then I knew for sure that the house value was overstated.
So I do not take the "excess" and "surplus" arguments very seriously. What a property COULD be (subdivided) and what it is seems somewhat strained. So I see two strategies in rural appraisal. I will describe the one that most assessors around here apply and no small number of appraisers as well. Then the method I use.
Let's assume 40 acres and a house. The assessor here breaks out land with a HL (house lot) at X dollars per acre- usually in rural areas something between $10k and $50,000. The bigger the house the more expensive the lot (?) and the cheapest is a single wide MH. The remainder 39 acres is valued as "ag land" (rates based on 10x the local pasture rentals) or for appraisers, on actual land sales. Generally that "house lot" value will include site improvements like landscaping, well, septic, and driveway.
The alternative is to treat the entire 40 acres "as if vacant and available for its highest and best use" - which is textbook language- in fact, early editions of USPAP included that until they recognized that for insurance value it wasn't really appropriate. Then the appraiser estimates the contributory value of the dwelling with an allowance for those site improvements included in the HL above.
Is one a better way than the other
in analysis? Not really. But I am circling back like Ms. Jan to the issue. Since the land is NOT subdivided to pretend that it is, smacks of a hypothetical consideration to me. It is contrary to what exists to assume a subdivision. So when you say, ....
Site is 107 acres. house needs 2-5 acres of land to support existing improvements
the real question is what is the typical lot size in the area? Is it 2-5 acres? Or are there about as many properties that are 20-100+ acres as there are <5 acre parcels and how easy is it to get a split approved?
I am arguing that the property is what the property is. And subjecting the property to an imaginary split of "surplus" land subjects it to a status as a hypothetical consideration in light of the certitude that
this imaginary split is not going to happen nor is it being proposed by the owners, is it?