1004mc argument with reviewer

[Brisco58]

You do realize that the MC prior 90 days listing is the number of listings on the last day of that period, No?

In Form 1004MC, in order to provide the most accurate depiction of the “Months of Housing
Supply” as of the effective date of the appraisal, the “Total # of Comparable Active Listings”
should be based on a specific point in time. For example, when completing the “Current – 3
Months” column for “Total # of Comparable Active Listings,” the number should reflect the
listings on the most recent date in the 3-month period (which is also the effective date of the
appraisal), and not the cumulative number of listings for the entire 3-month time period. Then,
when completing the “Months of Housing Supply,” the number for the “Total # of Comparable
Active Listings” is divided by the absorption rate, which provides an accurate depiction of the
existing housing stock as of the effective date of the appraisal. (Using a cumulative number of
listings during the “Current – 3 Month” time period may result in an artificially high number for
the “Months of Housing Supply.”)

I am perfectly aware of the guidelines concerning the completion of the 1004 MC and I do not recall making any commentary to the contrary within my previous post. However, thank you for emphasizing my previous point for me. The language is simply not consistent between forms. You used the term listing without any adjective within your first flippant comment above. Please specify if you mean active, pending, AWC. The market conditions form specifically indicates the total number of active listings while the data on page two of the URAR requires reporting of the number of comparable properties currently offered for sale.

 

[Mr Rex]

This debate is absolutely hilarious. Dozens of appraiser's can't agree on how to appropriately complete a form that was designed in order to provide for the reporting of consistent market trend data. If the data within the market conditions form was intended to mirror the data contained within the top of page two of the URAR concerning the number of sales and active listings then perhaps the same language should have been utilized on both forms. The market conditions addendum clearly indicates the number of active listings within the previous 90 days while the top of page two of the URAR requires all properties currently offered for sale with no mention of 90 days or use of the adjective active.

The only thing I can say for these Fannie forms is that they are consistently inconsistent.

You might want to re-read your prior post.

 

[Eli]

Couple questions:

1. Before the 1004mc, the 1004 one-unit housing on page 1 in the neighborhood section for "high"-"low"-"pred" for items of "age" and "price", were based on what parameters?

2. What are HUD's parameters for the above items?

 

[Dale Smalley]

My page 1 market data also comes from the same data set as the mc data.

 

[Mile High Trout]

Some standard verbiage: Pre written notes: Historical days on market figures are taken from sold examples only. No withdrawn or ghost type listings were researched. / The appraiser did not use the automatically supplied MLS MC data figures. Those figures have unaffixed data points, where the active figure is not related to the sold figures. Rather; I've manually assembled the MC, to trend data list vs sales data for only successfully sold units in that time period, and not to consider what was coincidentally still left on market unsold at that time. The presented MC above was a manual workup, to express fixed data points for list vs sale, for what the homes successfully sold for. / Specific MC notes:

And you can punch in a manual median work up very easily at this site:

https://www.easycalculation.com/statistics/mean-median-mode.php
_____________________________________________________________

You've got to understand that automatic review functions have officially taken over the industry, with the advent of xml mandatory mismo or whatever it's called. At one time there was a great debate about data use and proprietary data for appraisals. Appraisers used to lock pdf reports and deal with clients who raged because they could not unlock the reports and change data, etc, etc. Now apparently it's o.k. for clients to have supreme access and data control over appraisers reports with xml. We even sign an agreement with every xml creation that 'plain text' can easily be altered. So much heat on this issue, even our software providers put a disclaimer to protect themselves, regarding all the bad things that can happen to appraisers. Appraisers; the end of the line, the last stop, the low man on the totem pole.

The above grid, MC, and 1 unit trends are easy data match ups. If they don't match, the warning is automatically generated. That's how it works, it's that simple.

So you also include a data addenda, and copy paste several one line of data into those reports. MLS#, date, price, status, year built, agla, bas GLA, beds, bath, parking.

I include my 'final data set used for 1 unit trends, above grid figures, and mc results', which is usually a highly filtered data set which expresses the high and low ranges of similarly categorized properties in that defined research area, for the past year.

Also I'll include a totally unfiltered data set in addenda, although with automatic review, I no longer inter mingle that data within form filling points.

I never remove outliers, but rather explain them away as irrelevant or potential points if subject was worse/better, etc, etc.

Usually including 4 data sets or more if possible, which is easy to copy paste into supplemental addenda multi page under AI ready forms, which acts like a WORD document within your appraisal. It's just a blank white page. Try to include one line pastes of your final filtered 1 yr, final filtered w/ 3 year, unfiltered 1 & 3 year, and possibly other 'alternative approaches' to data, if you had to use subjective searches like size vs style, or there was no clear matches, etc, etc.

Thankfully, the auto reviewers stopped trying to match a time adjustment against MC trends. What a joke that was, and it could never work. I'm always saying MC was really not very informative for this specific assignment. Even if the MC points one way or another, I'll use my professional opinion for MC box form filling. Make sure your mc form filling boxes line up with the 1 unit trends boxes, because that can be an automatic coorelation point as well. Thankfully the software is not sophisticated enough to match numbers and boxes, or the MC would absolutely not work and would bring a never ending string of correction requests as if the MC was god, and we had to check the boxes via the coincidental median indicators. And believe that, the medians are pure coincidence and a slight alteration in data research can bring forth entirely opposite MC trending results. Look at that while you're researching so you can hopefully put forth a congruent data presentation.

Nobody has dared answer the question if the percentage product of list vs sale should be a direct product of the two figures presented. It would make sense that you divide the list and sale figures to get a straight up percentage indicator. Not according to Corelogic MLS Matrix systems though. They put forth auto mc figures where the percentage indicator is not a direct product of the list vs sale figures. So essentially it's unprovable data and any appraiser using those pre supplied mc's where the percentage is not a direct product of the list and sale figures, is presenting data they cannot prove. So I print out the long form detail for MC and use a quickie pen deal where I put stars, circles, or squares with numbers, and manually calculate the dom, list, & sale, for each category, and manually fill the MC form. Takes 10 minutes with the website posted above. I'd rather go manual, than trust the computer to tell me about trending. Considering what was coincidentally left on market is a truly awful approach, but one which the big corporations like, because it routinely indicates more market strength than was was really present. Including the unsuccessful sales from that category is pure trickery! Unsuccessful and withdrawn listings have no place in trending analysis for successfully sold properties.

Total # of Comparable Active Listings
Months of Housing Supply (Total Listings/Ab.Rate)

4-6 & 7-12 = Not Researched
Not Researched
Not Researched
Not Researched

 

[Deleted member 130081]

Sometimes its helpful to read all of the instructions on a form, not just the first paragraph. At the bottom of the MC form, there is a space where it asks you to summarize your findings. If you do not know what I am talking about, that might be your problem with this whole argument right there. This portion of the form gives the appraiser the freedom to NOT use the MC data to formulate their conclusions as reported in the Neighborhood section, even though the first paragraph of the form instructions states they "must" use the data as their basis. Here is an example of how I do it and I have yet to have a revision request concerning this portion of my reports.

"The 1004MC data set is subject specific, while the predominant neighborhood value reported in the Neighborhood Section of the URAR is not, therefore a discrepancy may exist between the median 1004MC value and the reported neighborhood predominant value. The 1004MC data set (and the number or comparables active/sold reported on page 2 of the URAR) was compiled using sales from within the defined neighborhood boundaries that were built between 2005 and 2015, were between 2000-3000 sf and were not located on the water. The wide range of value can be attributed to differences in location, size, curb appeal, quality and condition. In addition to the analysis of the 1004MC data set, an additional year-over-year MLS CMA analysis was also developed that included all sales of single-family homes from within the neighborhood boundaries. That analysis found prices remained mostly stable (increasing a rate of about 1% per year), the current supply is about 4.3 months and marketing times remained unchanged, averaging 3-6 months. (It should be noted that the average marketing time is just that, an average and not typical. Marketing times in this market and throughout the region have been erratic since the crash of 2008, with some homes selling very quickly and some taking a long time to sell.) The One-Unit Housing Trends reported in the Neighborhood section of the URAR were determined after analyzing both the 1004MC statistics (subject specific) and the additional year-over-year MLS CMA (over-all market)."

I aint saying this is exactly the best or the right way to do it, (goodness no one can agree on this) but I do explain in no uncertain terms how I got there. Any reviewer coming behind me could plug in the same parameters I used and come up with the same data. Further, gosh darn it, it just makes sense to define your neighborhood boundaries consistently with the area that you used to pull your data. Cant someone in charge just come out and say that so we can all be done with this argument?

 

[Lee Lansford]

LOL - Denis. OF course Im right. LOL. hehe.

I pulled out the Fannie Mae sellers guide and looked everywhere for the word "immediate neighborhood" and found nothing.

So just to make my self better understand..The only change that Fannie Mae is now saying is that page 1 (trends), top of page 2, and the 1004mc should all be the same, right? Just want to make sure I understand exactly what this new update is so I can better argue my point.

Nothing new here. The reporting is--and has been since day 1 of the 1004MC--exactly as you state.

 

[Amy Perkins]

Here is the problem I have with the limits of the 1004MC. Here in the coastal market we have zero growth and every piece of land has been developed unless it burned down or found new technology to build on. So we have the original homes and we have custom built homes. We also have sweeping mountain ranges and a chain of islands in the sea. There is one area known as the "Mesa" and it has modest homes and some custom homes. THE MARKET FOR SOME HOMES IS NOT DEFINED BY IT'S NEIGHBORHOOD BOUNDARIES. Meaning, if I have an ocean front property, the only comparable sales are other ocean front properties, which do not have a high turn over and extend beyond several "neighborhoods" Also, with nearly every kind of property type all next to each other there are market layers. The 1004MC is called a "market data analysis" not a neighborhood data analysis. If you are appraising the biggest most stately home in a neighborhood; you might have to go to each neighborhood and find the largest most expensive home to compare to your home. If you are appraising a multiple family property, you must expand out the market and include just those. For places like Bakersfield, Sacramento, and other cookie cutter markets this works fine. I often find myself providing two 1004MCs in the addendum many times. Sometimes you can't provide meaningful data in a neighborhood because it is too small. Sometimes you need a competing market added into the market data so you can provide the most accurate data.

 

[DWiley]

Sometimes its helpful to read all of the instructions on a form, not just the first paragraph...This portion of the form gives the appraiser the freedom to NOT use the MC data to formulate their conclusions as reported in the Neighborhood section, even though the first paragraph of the form instructions states they "must" use the data as their basis.

Actually, the instructions on the 1004MC form say:

The appraiser must use the information required ON THIS FORM as the basis for his/her conclusions, and must provide support for those conclusions, regarding housing trends and overall market conditions as reported in the Neighborhood section of the appraisal report form.

Note that those instructions refer to the entire form, not just the data in the grid. Fannie acknowledges that the data in the grid alone may be insufficient for trend analysis. That is why, as you point out, there is a big narrative section on the 1004MC to include additional analysis and commentary.

 

[DTB]

Actually, the instructions on the 1004MC form say:

The appraiser must use the information required ON THIS FORM as the basis for his/her conclusions, and must provide support for those conclusions, regarding housing trends and overall market conditions as reported in the Neighborhood section of the appraisal report form.

Note that those instructions refer to the entire form, not just the data in the grid. Fannie acknowledges that the data in the grid alone may be insufficient for trend analysis. That is why, as you point out, there is a big narrative section on the 1004MC to include additional analysis and commentary.

And appraisers get stipped for ambiguous comments. Would it have been so tough to include this from your post above: "Fannie acknowledges that the data in the grid alone may be insufficient for trend analysis." ?