1004mc argument with reviewer

[Eli]

Gender dysphoria is everywhere these days.

You all know the problem, behind this problem, is copious numbers of appraisers define neighborhood boundaries to be any place they find sales they want to use? For example, I am planning on defining anything west of the Rockies to be "the neighborhood" from now on so that I have lots of flexibility for comp selection...

Anything to speed it up. That's the name of the game, isn't it? LOL I mean nobody cares about credibility, do they? Well, let's hope that is changing and it appears it is. We can look forward to that. Your on track more than you prolly know. And they may only use a max of 10 comps on every assignment. Just pick a few high and low sales west of the Rockies that will bracket most everything. LOL

 

[Peter LeQuire]

My all time favorite was an appraisal that defined the neighborhood as all of this county; reported the land use % as 80% residential, 20% vacant (roughly one-third is in a national park) and probably15% of the properties on the assessor's roll is commercial or multi family or agricultural.

 

[ucbruin]

Always give yourself wiggle room.

Finally!!!!

 

[Terrel L. Shields]

Often, residents/market participants in a particular area don't even agree on he boundaries of their own neighborhood or what the defining characteristics of their own neighborhood are. Its a very tough issue that we tried to address as the data guys at Fannie were salivating about the possibility of tracking defined/standardized neighborhood data. The other problem is that neighborhood boundaries can change over time

Neighborhood means a lot of different things to different people. I would never define neighborhood on the basis of census tracts, which imho are totally meaningless division in the county I live. In the two counties adjacent - very rural Oklahoma counties - there are only "census blocks" - 4 per county. Again, the demographic does not match the Census tract. One county only has 2 towns over 1,000 in population. The other has 3 but it also has a large lake and as such an entirely different demographic - non-native retirees v. indigenous native American farmers and ranchers. And between the two is a large development shared by the two counties which is an el cheapo development of mixed manufactured housing, log cabins and typical SFR's.

Trying to cubby hole such mixed properties into neat little small packages won't work. Even when I was still doing secondary market I refused to work any rural Oklahoma property because you get insane stips like asking you to find "closer" comps from towns like Chewey, Proctor, Kenwood (which is an Indian Reservation, not a town per se), and other towns of under 100 population.

"Neighborhood" is an artificial creation of the appraiser or someone who speaks in generic terms about a "neighborhood" - a group of "complimentary uses"...what does that mean? A bunch of cookie cutters in a single subdivision? Or a defined community with a common tie, such as school district or city limits complete with small retail area, commercial/industrial areas, etc. And most of us are guilty at one time or another of PFA the percentages- 40% residential; 20% vacant; 10% industrial and 30% other...After all, who wants to go into a defined area, and pull up every parcel and A-determine if it is associated with one or more other parcels; then count physically which have dwellings, which are vacant and which are non-residential?

Another thing is I see is that every community is 10% renters Every one. Nevermind the census summary claims 30% non-owner occupied. I see renters under-reported in 90% of appraisals. I've probably been guilty of it too. OTOH, what data source is going to get accurate accounting for the explicit neighborhood you defined? If your neighborhood isn't exactly the same as the census demographic neighborhood, the number could be different.

My all time favorite was an appraisal that defined the neighborhood as all of this county; reported the land use % as 80% residential, 20% vacant (roughly one-third is in a national park) and probably15% of the properties on the assessor's roll is commercial or multi family or agricultural.

Government owned land is commonly not even considered in a lot of appraisals I've seen over the past 2.5 decades. Ditto for lakes, ag land, etc. What metric is used? Numbers (gross)? Acreage? Value $ worth? It is all a big joke.

 

[DWiley]

Neighborhood means a lot of different things to different people...

Very true. In many reports that I look at the defined "neighborhood" is just the subdivision. When we refinanced last year I looked at the appraisal report done on my personal home and found that the defined "neighborhood" didn't even encompass all of my subdivision.

It seems that many are overy influenced by arbitrary distance parameters established by underwriters. As a friend of mine often says, "No one bought their house so they could live within one mile of me, so why do so many define neighborhoods that way?"

 

[Mile High Trout]

Thank you I did read those. However my issue is not if page 1, top of 2, and 1004mc should be the same. My issue is what is the "neighborhood". The review is telling me the neighborhood is basically the subdivision. My argument is that a neighborhood is defined by the boundaries that I the appraiser call out. This is all stemming from the fact that I am working with a very small town and called out the boundaries to spread over 2 miles and have included a comp that is about 2miles away and said that it is still within the neighborhood boundaries and included it in the 1004mc. So the reviewer is asking me to take it out of the 1004mc because its beyond her perception of what the "neighborhood" is defined as. So she said Fannie Mae defined the neighborhood as the "immediate neighborhood" or subdivision I guess. That is what I am trying to figure out.

Well, that's an easy one. Isn't it also a 'guideline' that 1 unit trends, above grid active & sold figures, and MC should all hail from the 'final data set'?

Therefore; Defining MC in a limited manner provides misleading results and non compliant reporting regarding the '3 necessary matches' for 1 unit, above grid, and mc.

How could you follow both rules at once, if you had to reach out of the 'micro development area' to comp out with 'the actual comparative market'?

Here is the standard text I have in MC (boilerplate), regarding how Corelogic Matrix MC analysis botched the job. I go to easycalculations.com to run manual medians for every dang report. It's easy though because it's just 10 key number data entry.

Standard MC notes: Historical days on market figures are taken from sold examples only. No withdrawn or ghost type listings were researched. The Corelogic Matrix MLS assembles MC data based on unaffixed data examples. Meaning that the sold trending is not tied directly to the sales active vs sold, but rather the Matrix MC provides anecdotal information on sold comparisons vs as of yet unsold comparisons. / My approach was to manually assemble MC figures, where the actual list vs sale for only sold units is analyzed. I feel this provides much more realistic list to sale and market price trending data compared to comparing ratios with unsold offerings. The MC is a manual work up, based on fixed analysis for list and sold pricing for the sold properties in the time periods. / Specific MC notes:

 

[residentialguy]

Well, that's an easy one. Isn't it also a 'guideline' that 1 unit trends, above grid active & sold figures, and MC should all hail from the 'final data set'?

Therefore; Defining MC in a limited manner provides misleading results and non compliant reporting regarding the '3 necessary matches' for 1 unit, above grid, and mc.

How could you follow both rules at once, if you had to reach out of the 'micro development area' to comp out with 'the actual comparative market'?

Your off a bit on the MC, Mile High. While the data for the inventory analysis of the 7-12. 4-6, & 0-3mo is specifically for comps in the neighborhood (as directed), which as you correctly stated, does not produce enough data for a reliable indicator of market trends...however the overall trends do not come from that data alone. The overall trends comes from a broader analysis that is reliable. that should be in your additional comments. Those results of the "OVERALL TRENDS" are to be reflected and supportive of page one. The results of the left side of the 1004mc form are reflected on the top of page 2. Therefore it is not limited and all three pages are in harmony with each other because the 1004MC is split.

Certainly not a perfect way to do it, but that's FNMA for you, where screwy is the one thing that they're consistent with.

 

[Rich LeMond]

Fannie does not want you to expand boundaries outside the neighborhood for the 1004MC. You could define the neighborhood very loosely, but then you may run into some other problems. I usually make a comment that the data in the neighborhood is insufficient for statistical significance, and that I analyzed the greater market and retained that in my workfile.

 

[DWiley]

I usually make a comment that the data in the neighborhood is insufficient for statistical significance, and that I analyzed the greater market and retained that in my workfile.

Doing the additional analysis is great, but a summary of that additional analysis should be on the 1004MC as well, not just in the workfile. Otherwise, the 1004MC does not support the indicated market trends, as stipulated in the 1004MC instructions.

 

[Rich LeMond]

The summary is in the form of the check boxes, which are marked based on my larger analysis, and not the data in the 1004MC. If their form requires me to squeeze a square peg into a round hole then they can't get mad at me for just jamming it in there!