Is The Cost Approach A Real Thing?

[Terrel L. Shields]

if you're doing a CA only appraisal,

Well...not necessarily. Again, if I know what the building cost when new...then I can get a book value from a cost book plus use the historical index to double check that. I know what land is worth "as if vacant and available for its HBU"... and so...the issue is depreciation. But isn't that an issue with the sales approach too? I mean how many times in using comps that are somewhat different are we at a loss to account for one thing or another?

Log homes. I was asked today about a log home that a guy had tried to finance, and the lender refused to lend the money because it was log. And how many comps will we have for a geodesic dome? And underground home? A mansion? I know a place on a lake with 200 acres at the time. Cost north of $30 million. 20 years later the builder died. The estate tried to sell for $8 million. They ended up auctioning and it brought $3 million which was rejected. Several years later it was butchered up into lots and sold with an educational provider and Walmart executive buying the house and a few surrounding acres. There is and was no way to accurately value the property CA, SA, IA...nothing worked to be "accurate."

You could guess at depreciation using the CA with as much accuracy as you could guess the functionality and externalities at play. Perhaps more. Unique estates are a dart board exercise and the best you can do is to warn the client that there is a huge element of guessimation in your estimate, or for that matter, anyone else's estimate.

Another one I remember. An old retired superintendent turned Realtor, up in his 80s, predicted that a nearby historical ranch would bring $5000 an acre. The ranch, known as the Colcord Ranch and home of an early Oklahoma developer (the Colcord Building in OKC is his brainchild) and rancher, sold for a fraction that at auction. Not a distressed sale. Several other Realtors has also estimated that there would be a premium for being such a historic place. As it turned out, didn't make any difference. What would have been a good pre-auction estimate? Well probably just the typical large tract value LESS about 10% (since auction prices are often less...) would have been a good guess. About half the high prediction.

 

[alebrewer]

Well...not necessarily. Again, if I know what the building cost when new...then I can get a book value from a cost book plus use the historical index to double check that. I know what land is worth "as if vacant and available for its HBU"... and so...the issue is depreciation. But isn't that an issue with the sales approach too? I mean how many times in using comps that are somewhat different are we at a loss to account for one thing or another?

Log homes. I was asked today about a log home that a guy had tried to finance, and the lender refused to lend the money because it was log. And how many comps will we have for a geodesic dome? And underground home? A mansion? I know a place on a lake with 200 acres at the time. Cost north of $30 million. 20 years later the builder died. The estate tried to sell for $8 million. They ended up auctioning and it brought $3 million which was rejected. Several years later it was butchered up into lots and sold with an educational provider and Walmart executive buying the house and a few surrounding acres. There is and was no way to accurately value the property CA, SA, IA...nothing worked to be "accurate."

You could guess at depreciation using the CA with as much accuracy as you could guess the functionality and externalities at play. Perhaps more. Unique estates are a dart board exercise and the best you can do is to warn the client that there is a huge element of guessimation in your estimate, or for that matter, anyone else's estimate.

Another one I remember. An old retired superintendent turned Realtor, up in his 80s, predicted that a nearby historical ranch would bring $5000 an acre. The ranch, known as the Colcord Ranch and home of an early Oklahoma developer (the Colcord Building in OKC is his brainchild) and rancher, sold for a fraction that at auction. Not a distressed sale. Several other Realtors has also estimated that there would be a premium for being such a historic place. As it turned out, didn't make any difference. What would have been a good pre-auction estimate? Well probably just the typical large tract value LESS about 10% (since auction prices are often less...) would have been a good guess. About half the high prediction.

And I guess cost equals value?

 

[Terrel L. Shields]

And I guess cost equals value?

when cost is the best you got, prove it wrong... I mean the problem is the same for the CA and the SA. If you don't have sales info to back up the report, then the accuracy goes to pot. But with plenty of sales data, I can extract the depreciation from sales, I can extract and differentiate the external from the functional from the physical as readily with the CA as with the SA.

 

[alebrewer]

when cost is the best you got, prove it wrong... I mean the problem is the same for the CA and the SA. If you don't have sales info to back up the report, then the accuracy goes to pot. But with plenty of sales data, I can extract the depreciation from sales, I can extract and differentiate the external from the functional from the physical as readily with the CA as with the SA.

Exactly. Cost does not equal value in almost any conceivable situation, but the CA assumes it does. You are a smart guy, T. I'm not sure why you keep defending an indefensible position except that you are trying to sell your wares...

 

[Terrel L. Shields]

I'm not sure why you keep defending an indefensible position except that you are trying to sell your wares...

With identical data, the CA should be very close to the SA...after all, you need to extract accrued depreciation from sales. duh.

 

[Raven59]

Is CA a legitimate concept? If it is not, the GSE's are going to make it important. By utilizing a data collector and a few land sales and cost data, the GSEs now have an alternative product to counter the biased SC approach...no need for an appraiser. It's why you heard several questions about alternatives to SC in the bias hearing.

 

[alebrewer]

Is CA a legitimate concept? If it is not, the GSE's are going to make it important. By utilizing a data collector and a few land sales and cost data, the GSEs now have an alternative product to counter the biased SC approach...no need for an appraiser. It's why you heard several questions about alternatives to SC in the bias hearing.

I think you're probably right. And the reason the CA is a viable option for eliminating perceived bias is because it IS so easily manipulated. With the SCA, there is no way around using truly 'comparable' sales. With the CA, you can just make stuff up.

 

[alebrewer]

With identical data, the CA should be very close to the SA...after all, you need to extract accrued depreciation from sales. duh.

How do you extract accrued depreciation from sales when you're appraising a prison?

 

[Alsie35]

Nevermind

 

[Terrel L. Shields]

How do you extract accrued depreciation from sales when you're appraising a prison?

Doesn't that apply to the sales approach as well as the cost approach? Which one is going to identify functional or external obsolescence?