Market Conditions: Am I on the right track?

[Bill_FL]

You have shown what the median and average sales prices were, but what did that buy? What was the median home? If last year, the median home price was $200,000 and was a 1700sf, 3/2, but this year, the median price dropped to $195,000, but it only bought a 1450sf 3/2 , did values fall? The data you have presented does not yet have any real meaning.

 

[Riick]

You have shown what the median and average sales prices were, but what did that buy? What was the median home? If last year, the median home price was $200,000 and was a 1700sf, 3/2, but this year, the median price dropped to $195,000, but it only bought a 1450sf 3/2 , did values fall? The data you have presented does not yet have any real meaning.

You do the best you can, with the data you have, and you make it understandable to your reader.
Start talking mode, median, standard deviations, t-tests, and such and their minds just shut-down.

More likely in most places, what's happening is:

Last year the median home price was $200,000 and was a 1,700sf, 3/2, but this year, the median price dropped to $195,000, & it bought 1,850sf 3/2.5
​

 

[Bill_FL]

Which is why I like to also show $psf in the trend. I also like to show, last year what that median home was as well as what it is today. I bet most people, if they would do that, would be suprised. I bet they would also be suprised to then figure out what the median home last year is selling for today.

remember, the median is simply the middle sale. In my market, the high end of the market has slowed most dramatically. Less than 25% of the volume from previous year or two. You take those sales out and the median shift is pretty big.

 

[Rufus]

Overkill and relatively useless. They don't care about the market five years ago. They want to know about the changes in the past year. You would be better served to break it down to the previous year and then break the year prior to the effective date down into quarters to show current market trends. You are analyzing the current market as compared to the trend for your comparable sales in the past year, six months and three months.


My opinion is that the whole thing is a collosal waste of time for you and the reader.

It will not impress anybody, unless it deals with the current market and recent past market as it pertains to the data in the report.

Breaking the data down into quarters of the year is a good idea for large areas, however, expanding the subject neighborhood beyond reasonable boundaries to do so is not appropriate in my opinion. If I did that with the data presented then each quarter would have so few sales that the margin of error would be so large as to make it completely meaningless.

The thing I see in so many reported statistics is that they only compare raw numbers to determine an increase or decrease. If we start with the null hypothesis that values are stable and then use proper statistical analysis, which takes into account the margin of error in the sample size, it will only indicate a Type B Correct Decision when there is sufficient evidence in the noncritical region. At that point the data can be considered statistically significant to predict increasing or decreasing values.

You have shown what the median and average sales prices were, but what did that buy? What was the median home? If last year, the median home price was $200,000 and was a 1700sf, 3/2, but this year, the median price dropped to $195,000, but it only bought a 1450sf 3/2 , did values fall? The data you have presented does not yet have any real meaning.

Of course I am biased here, but I would not go so far as to say my analysis does not have any real meaning. You do have a good point that I should look into the average size of the dwellings for the corresponding time frames. I'm not yet convinced that will be meaningful, but until I look at the data it I shouldn't form an opinion.

 

[Mztk1]

I have to agree with the others that say it is overkill and not in simple enough terms.

I was just brought onto a bank's rotation by a lender because my reports are thorough, I was told. The last appraiser "killed" five deals in a row, which is normal these days so it didn't bother them, but the originators and loan officers who have to explain the appraisal to the owner and tell them why it appraised low, didn't have enough sound or clearly written reasoning in the report. A new bank policy prohibits production staff from calling the appraiser to discuss the results, so when they called the owner to explain the appraisal, they couldn't. In short the guy was canned because he just stated (or summarized) his market statements and didn't show the data and facts to support it. I bet when he stops getting the work he will think it is because he killed a bunch of deals in a row. Totally wrong.

So to answer you question, Are you on the right track? Absolutely. But I think more and more with the loan officers out of the loop to speak with us, we are going to have to be writing for the lay reader to understand our reasoning, not the guru reviewer or the seasoned and highly analytical underwriter, if we want to stay in demand.

Like Bill Potts said, I too include the average size of the house and the median size of the house that sold, the average price, median price, average price per square foot, median price per square foot. It is amazing how wildly different these figures can be. In one of my counties (though I break things down to markets and only include county data for a trend analysis), the median value has dropped from $165,000 to $150,000 in the past 4 months with about 300 sales a month. But the price per square foot has remained fixed at about $95. Someone can easily misread the market to a 36% decline a year when in fact it is relatively stagnant.

But that might not work for you. It looks to like you have minimal data. I like to break things down the way Riick did with a year-to-date over year-to-date, or even quarter over quarter, because it compares apples to apples. Plus, if my trend is showing a stable market for a fourth straight month in the county, I'll analyze that four+ month period in the report specific to my market to see if the local area has stabized too.

I like how you broke out the REOs. That is difficult to do with my MLS because of a glitch in the system. Supposedly they were working on fixing it.

There are so many ways to skin this cat, and frankly once you settle on a way, the market changes a bit and you have to adapt your method just a little if you want to really understand it correctly.

 

[Blue1]

Hi Rufus,

You are on the right track but your explanation is way to technical IMO. The information you provided should be in your file...Then you need to summarize it so that the intended user can understand it.

Good luck...

 

[Metamorphic]

I dont think that what you're trying to present is too technical. But it sounds a little awkward because of they way you're saying it. For instance. I re-wrote your first paragraph.

"Sales data from the subject neighborhood suggests two distinct markets are present. One market comprised of REO and pre-REO properties (REO), and another market comprised of non-REO, non-distressed sale properties (Normal). Market data for both these markets are presented separately below. The ratio of REO to Normal sales is approximately 7/3 (71.7% REO). There are more REO sales in the current year than in previous years. In the previous 4 years REO’s comprised between 52.1% and 60.3% of the market. Currently, there are 44 active listings; 13 of the 44 are flagged as REO's; there were also 6 REO's listed as Pending sales (Source:MLS). "



Its pretty much all the same but I tried to simplify the sentence structure and I made sure there was only one idea in each sentence.


But you should probably put in a couple sentences up front to summarize what you're saying below. Its the old tell them what you're going to tell them, then tell them, then tell them again" thing. But when you're doing that you usually want to make the first "tell them", a simplified version for the people with short attention spans that will never read the whole thing.




For instance I might start with: "The market data presented below describes a market with the following conditions:
- Home prices have been declining since 2005.

- The rate of decline has been ____ in the last year and ____ in the last 6 months.

- The market has 2 distinct segments REO/short sales, and normal sales.
- REOs sell for approximately 1/3 the price of Normal sales.

- There are are approximately 5 sales occurring per month; there are 44 active listings.
- The market is in oversupply with a ____ month absorption. "

Then you go on with all the detail.

That kind of approach makes everybody happy.

 

[Rufus]

Jim Klos and Metamorphic,

Thank you! Very helpful information.

 

[Tim Hicks (Texas)]

Breaking the data down into quarters of the year is a good idea for large areas, however, expanding the subject neighborhood beyond reasonable boundaries to do so is not appropriate in my opinion. If I did that with the data presented then each quarter would have so few sales that the margin of error would be so large as to make it completely meaningless.

The thing I see in so many reported statistics is that they only compare raw numbers to determine an increase or decrease. If we start with the null hypothesis that values are stable and then use proper statistical analysis, which takes into account the margin of error in the sample size, it will only indicate a Type B Correct Decision when there is sufficient evidence in the noncritical region. At that point the data can be considered statistically significant to predict increasing or decreasing values.


Of course I am biased here, but I would not go so far as to say my analysis does not have any real meaning. You do have a good point that I should look into the average size of the dwellings for the corresponding time frames. I'm not yet convinced that will be meaningful, but until I look at the data it I shouldn't form an opinion.

Breaking things down for an entire year will skew your data also. Some markets have dramatic corrections. Some markets take time to change. I know of several markets in my area with declines that did an immediate correction about eight months ago and then leveled off. So sales in the last six months require no adjustments, but over a eight months ago need a large adjustment. Underwriters expect us to adjust in a line, which was not appropriate. I know of a few that continually dropped and continue to drop. This can only be shown by breaking data down to smaller time lines.

Data from five years ago is relatively useless in a current volatile market.

Also, breaking data down to smaller increments of time helps to prove a stable market which is what most of my area actually has and no underwriter wants to believe.

Typing excess data that is over the head of the reader only earns the trash can and will never impress anybody. Ask Otis Key and his 25 pages of addenda.

 

[CANative]

I don't have much education and my writing reflects that. I think that's why most underwriters and lenders understand what I write.