Bad advice from Fannie--"Multiple Parcels" from Dec. 2019 'Appraiser Update'

[J Grant]

After Andrei answers the "best comps" question that answer will trigger several more questions to follow; each being triggered by the response to the one before. I'm about 4 moves ahead of him at the moment. I'm going to march him straight into a corner.

George I respect your acumen, but why do these threads need turn into a duel with a winner and a loser and the winner claims a "victory" by proving someone else is "wrong"? Why not just discuss the issues and consider what someone else says, even if we don't agree with what they say?

 

[George Hatch]

What message are they conveying? PS they may not be conveying the same message- and it would be discredited if based on bluster about how others do not understand MV and difference of MV and value in use, HBU, how either affects the MV opinion etc. To answer the CA scenario :

The CA has an improvement replacement value and a land value. Obosuly the larger worth of the 2 lots would be reflected in the total land value. estimate in the CA

The duplex/triplex question : if the extra site CAN have another duplex or triplex (but on eff date is vacant ), the income approach only applies to the existing duplex/ triplex. Just as income approach on a single site duplex reflects only what exists, not some imaginary not yet existing addition. Same as a vacant site CAN have a house built on it, we don't include the value of the not built house for vacant site value.

There is no answer for bluster such as post 736, "You're going over his head. Seriously. ". They to resort to bluster because the fact is USPAP allows a single MV opinion for two or more properties sold together. And one of those properties might be below it 's HBU if sold alone. That is part of the appraisal analysis. They can not seem to grasp that even though the parcel might have a different or higher value if sold separately, the MV opinion is for the lot sold with the improvement, and the lot's contribution may be the same as, or less than, or more than, the $ market value opinion lot would be if sold alone.

That's not how it works with the appraisals of assemblages, outside of FannieWorld, anyway. I mean, blocks of parcels under the same usage, like clocks of condos or blocks of vacant parcels can often be valued using sales of other such assemblages. But once it comes to most other property types the normal process is to first value each component, then develop the discount rate for the assemblage. This is especially true when it comes to disparate uses.

So with the duplex+lot scenario we account for the contributory value of the additional lot in the Income Approach by importing the land value from a land sale analysis, and then if there's a discount for the assemblage of disparate uses we develop that and import that, too.

Really, it's no different than what you would do if you had an SFR and a cell tower lease. You'd find comps for the SFR and you'd find other sales of SFRs with cell towers. It will be unlikely that those SFRs are directly comparable to your's but you can compare those transactions to their respective comps to identify the contributory value of the cell tower (or extra lot) to the whole.

SFR+ orchard or grove, same process. SFR+Gas Station, same process

These are established methods and techniques for valuing these types of appraisal problems.

 

[Andrei Fin]

Finally. Thank you

I never ignore sales history or the example it sets. But, we're looking for most probable as demonstrate by the group, not the one. And don't forget, the broker in the 2016 sale was smart enough to list both properties separately so they were at least *considering* the alternative that you are so casually dismissing.

Anyways, in addition to your subject's sale history, if you're doing an appraisal, what will all your best comps in the SCA look like? A or B?

Trouble with math? Your "group" consists of one dissimilar sale. One is not a group. You can cheat and use your fingers if needed.

What's a better comp, a 100% identical sale two years back or a very dissimilar sale now? The answer is A, the time adjustment is one of the most easiest and reliable to calculate. The best comps would look like B, however we don't have any in this case, only the subject's own prior sale. The comp on Walnut looks very little like B. And since you are only interested in rehashing this: the reason I didn't bother to look is that I was never asked to appraise the property. Meanwhile you botched your comp search so badly I didn't know whether to laugh or cry. Partly due to lazy methodology and partly due to lack of geographic competency (which you would be quick to crucify anyone else for).

Carnivore's marketing time scenario is totally flipped against reality, and it is the reason the Garfield lot listing was withdrawn, and why you can't show me any examples where the lots went to separate buyers but I can show you 100 examples where they went to the same buyer. Because although the typical seller may be interested in maximizing profit among other things, the typical buyer is neither interested in developing a vacant lot, nor living next door to a construction site.

And yes, as J Grant pointed out the "over your head" comment says it all. Coming from someone who can't tell the simple difference between two questions, "what are the two parcels worth if sold together" and "what are the two parcels worth if sold separately".

Regarding a duplex/triplex tell me this. You have an illegal triplex zoned duplex, does this mean you comp it out with only duplexes because that's the HBU? No, the best comps are going to be other illegal triplexes (given that the jurisdiction allows them to transfer, which most do in CA). And don't try to tell me that no lender will take it, because you are clearly much more interested in theoretical practice than what the lender wants.

 

[George Hatch]

George I respect your acumen, but why do these threads need turn into a duel with a winner and a loser and the winner claims a "victory" by proving someone else is "wrong"? Why not just discuss the issues and consider what someone else says, even if we don't agree with what they say?

Because there are some idiots following this thread who are finding validation for doing the same from Andrei's foolishment. Letting it stand unchallenged as if that's a legitimate solution is a bad plan. The conclusion that the extra lot is nothing more than a lawn will be true in many cases (that's fine), but it won't be because the appraiser actually went through the process to actually eliminate the other possibilities in order to get to "most probable" (that's not fine).

To reiterate, if "most probable" really is based on the extra lot being a lawn then that's great. But skipping the analysis it takes to come to that conclusion is not fine. Basing your decision to do that on some uncertainty and nervousness about overvaluation, or in advocacy of being "conservative" is unethical.

 

[J Grant]

That's not how it works with the appraisals of assemblages, outside of FannieWorld, anyway. I mean, blocks of parcels under the same usage, like clocks of condos or blocks of vacant parcels can often be valued using sales of other such assemblages. But once it comes to most other property types the normal process is to first value each component, then develop the discount rate for the assemblage. This is especially true when it comes to disparate uses.

So with the duplex+lot scenario we account for the contributory value of the additional lot in the Income Approach by importing the land value from a land sale analysis, and then if there's a discount for the assemblage of disparate uses we develop that and import that, too.

Really, it's no different than what you would do if you had an SFR and a cell tower lease. You'd find comps for the SFR and you'd find other sales of SFRs with cell towers. It will be unlikely that those SFRs are directly comparable to your's but you can compare those transactions to their respective comps to identify the contributory value of the cell tower (or extra lot) to the whole.

SFR+ orchard or grove, same process. SFR+Gas Station, same process

These are established methods and techniques for valuing these types of appraisal problems.

I agree with the above. Which is reflected where I stated CA would show the value of the two lots as the land value estimate portion.

 

[J Grant]

Because there are some idiots following this thread who are finding validation for doing the same from Andrei's foolishment. Letting it stand unchallenged as if that's a legitimate solution is a bad plan. The conclusion that the extra lot is nothing more than a lawn will be true in many cases (that's fine), but it won't be because the appraiser actually went through the process to actually eliminate the other possibilities in order to get to "most probable" (that's not fine).

To reiterate, if "most probable" really is based on the extra lot being a lawn then that's great. But skipping the analysis it takes to come to that conclusion is not fine. Basing your decision to do that on some uncertainty and nervousness about overvaluation, or in advocacy of being "conservative" is unethical.

I agree, but I don't think Adrei is saying he would skip the analysis. But if he (or anyone does ), than yes that would be unethical/not a credible solution to assignment.

 

[George Hatch]

Trouble with math? Your "group" consists of one dissimilar sale. One is not a group. You can cheat and use your fingers if needed.

What's a better comp, a 100% identical sale two years back or a very dissimilar sale now? The answer is A, the time adjustment is one of the most easiest and reliable to calculate. The best comps would look like B, however we don't have any in this case, only the subject's own prior sale. The comp on Walnut looks very little like B. And since you are only interested in rehashing this: the reason I didn't bother to look is that I was never asked to appraise the property. Meanwhile you botched your comp search so badly I didn't know whether to laugh or cry. Partly due to lazy methodology and partly due to lack of geographic competency (which you would be quick to crucify anyone else for).

Carnivore's marketing time scenario is totally flipped against reality, and it is the reason the Garfield lot listing was withdrawn, and why you can't show me any examples where the lots went to separate buyers but I can show you 100 examples where they went to the same buyer. Because although the typical seller may be interested in maximizing profit among other things, the typical buyer is neither interested in developing a vacant lot, nor living next door to a construction site.

And yes, as J Grant pointed out the "over your head" comment says it all. Coming from someone who can't tell the simple difference between two questions, "what are the two parcels worth if sold together" and "what are the two parcels worth if sold separately".

Regarding a duplex/triplex tell me this. You have an illegal triplex zoned duplex, does this mean you comp it out with only duplexes because that's the HBU? No, the best comps are going to be other illegal triplexes (given that the jurisdiction allows them to transfer, which most do in CA). And don't try to tell me that no lender will take it, because you are clearly much more interested in theoretical practice than what the lender wants.

Oh, we'll get to your subject's sale history soon enough, and I predict you're not going to like how that turns out, either. That's a couple steps ahead, and if you can't already see how I use a sales history to isolate the contributory value of some variable then you're going to see a new-to-you technique today.

In the meantime, your subject's sales history isn't going to help you duck the question of what to do in your SCA.

For now the choice is A or B, and both are identical to each other except that the lot area for one is on a single parcel and the lot area on the other is on 2 parcels. You already gave it up that B was the more similar comparable. So the answer of which types of sales (when available) you would want to put into your SCA should be right on the tip of your tongue.

 

[Andrei Fin]

I agree, but I don't think Adrei is saying he would skip the analysis. But if he (or anyone does ), than yes that would be unethical/not a credible solution to assignment.

George's argument for the last 200 posts is that Fannie Mae's instructions are evil based on the fact that I skipped the analysis on a house that I was never asked to appraise. What can you do?

We should also keep in mind that these instructions have been in place for a very long time, this newsletter was just a clarification.

 

[George Hatch]

George's argument for the last 200 posts is that Fannie Mae's instructions are evil based on the fact that I skipped the analysis on a house that I was never asked to appraise. What can you do?

You posted your example and cited it in support of your opinion. I am challenging the manner in which you developed and have been defending that opinion. I asked for the example because I intended to analyze it.

Basically, I'm using your rationale to demonstrate how dumb Fannie's widget approach is. That rationale is functioning as my crash test subject. This is an instructional strategy I used to considerable effect for many years. I always solicited the opposing view in order to compare it to the course material in order to show which worked better or made more sense. Bring it up, air it out with each side of the argument giving it their best shot and disassembling it based on its weaknesses.

 

[Andrei Fin]

You posted your example and cited it in support of your opinion. I am challenging the manner in which you developed that opinion. I asked for the example because I intended to analyze it.

Basically, I'm using you to demonstrate how dumb Fannie's widget approach is. You're my crash test subject.

I never developed an opinion because I was never asked to appraise the property. What you're missing is that Garfield was not offered as a subject property but rather it was offered as a comparable.