Time Adjustment , huh?

[Fernando]

Yes, here in my city, that's what is happening, where there is an undersupply of housing, there are multiple bids all the time. The bad thing is out of town REITs come through and bid waaay over market and they are all cash. They rent the homes at twice the marker rent to multiple people.

San Francisco list price has been listed way below market value for long time, at least 10 years. It has become more common everywhere here in recent years.
Because cap rates are so low here, REITs and outsiders are not a major buyer. Also, San Francisco has strict rent control and tenant protections and city regulations that investors BEWARE.

 

[Doug Wegener]

If you don't adjust for SF when less than 50 SF or 100 SF difference, then adjusting for short term "time" is a guess. The variation of prices in the inefficient market we call housing means we cannot really call such variations on a daily, weekly or even monthly basis without an element of uncertainty - aka - the margin of error is high. That's why the oft quoted praxis was that the appraiser was 'good' if they got within 5% of the "true" value. So you are not going to be particularly accurate attempting to make time adjustments monthly. Even Case-Shiller and the NAR project their numbers by adjusting for the season... are you?

"Even Case-Shiller and the NAR project their numbers by adjusting for the season... are you?""............................No, I don't seasonally adjust comps. Over the years, I have seen a lot of appraisals by a lot of different appraisers, many with time adjustments. I have never seen time adjustments made seasonally, by anyone. The appraisal classes I have taken do not teach that either. Now, I am not saying you are wrong and that it shouldn't be done that way, but I will hazard a guess and say in 95%+ of appraisals it is not done that way. If it should be done that way, nearly EVERYBODY is doing it wrong except Case Shiller, NAR, and Zillow.

 

[Tom D]

all the trend statistics you see from realtor.com or your MLS are based on a median, or average, of the total. you cannot just go seasonal, or skip the last 3 months, because you haven't done that statistic for that time period. some of you don't understand that you can't be that precise with any time period within a time period. i would love to see your statistic work file. most likely some of you are making a guess at what happened, or is happening. if the rate was higher at the beginning of the year, then slowed down, then your average or mean adjusts for that. so instead of 1% a month it becomes .37 a month like it has in my areas. you can easily show that number, our MLS provides those numbers, even the individual zip codes. but, i also do a micro of the neighborhood.
some of you in trying to be so precise can't prove how you came with a gross number to overall adjust. keep it simple and provable. some of you are doing the opposite.
i should call yous the statitistics 1/10%ers, just like how the ansi people think they are determining an exact GLA.

 

[CGinMN]

Seasonally adjusting a trend line is different than seasonally adjusting comps. So I’m not really sure where to stand, Doug. Many appraisers seasonally adjust a trendline by grouping 12 months of data into one period. In doing so, their comps are not adjusted for seasonality. It’s a fast and cheap way to derive a time adjustment, but it’s prone to error. Still better than nothing. It’s not a market conditions analysis though.

So many appraiser don’t understand why it’s important to look at more than median sale price data. You have to analyze DOM, price ratios, price drops, seller concessions - basically all the data points the MC form asks for. You gonna do that plus gather other information, like talking with realtors of competitive listings pendings (the most direct comps) and asking questions, such as: how many showings, how many offers, what’s the contract price (or ballpark), what was your 2nd best offer, how does this compare to 6 months ago or 12 months ago, what is you observation about market changes this season, etc…

I never really cared what case shiller says, since they are analyzing macro trends that are different from what I’m being asked, which is derive a time adjustment for a specific market segment.

most likely some of you are making a guess at what happened, or is happening. if the rate was higher at the beginning of the year, then slowed down, then your average or mean adjusts for that.

Your problem, like many appraisers, is that you’re too focused on median sale price as the only piece of data in a market conditions analysis. The fact you think others can’t know better, but you’re not analyzing this other information, tell me that you’re the one who is probably guessing.

 

[Terrel L. Shields]

but I will hazard a guess and say in 95%+ of appraisals it is not done that way. If it should be done that way, nearly EVERYBODY is doing it wrong except Case Shiller, NAR, and Zillow.

If you don't look at the numbers in a statistical light, you are capturing as much "noise" as you are real increase. And if you have enough sales to truly run a statistically significant data set, then you probably have enough sales within the previous 2 months or so to value the property without resorting to older sales. And if you are using only sales within the DOM you almost certainly will not be able to separate the market conditions adjustment from that statistical noise that is a given in an inefficient market. Further, once the market 'turns' (which you should be looking for anytime the number of listings increase, DOM increases, and sales decrease) can you fish out the "correct" market conditions adjustment that very month?

Again, if you have a $300,000 sale which is 2 months old in a 1% a month market, that's what? $3,000??? And if you have a $300k home on a typical lot here, the land price is 33% possibly? So you have a SF adjustment that could be from $50-100/SF. So if you don't adjust for 50SF difference ($2,500-5,000) why would you adjust for some tiny amount of time?

The more adjustments we make, the less reliable is the estimate mathematically- simple as that. We are pouring adjustments upon top of adjustments in a whack-a-mole fashion to arrive at a value that probably is biased by our process.

 

[Tim Hicks (Texas)]

If the 1004MC shows lower indicated median pricing for 0-90 days than 91-180 days, why should you make a market increase adjustment? If lenders want us to use the 1004 MC as a tool, use it as a tool. The easiest thing to do is just use sales within 30 days.

 

[Sadie]

I have seen this more than once. I am looking at a report which says "Sales older than 90 days were adjusted 1% per month to account for changes in market conditions".

Huh? Where on earth did the idea come from that you don't adjust for sales less than 90 days. It makes no sense in an increasing market. Does anybody know? As I mention, I have seen this on more than one occasion.

I have heard of it but never done it. It may not be reflective of the market. I bet that is the way this particular appraiser was taught. You need to make market condition adjustments as the market indicates.

 

[Member258]

This does seem odd, but with no more information, my first question would be is, how did they arrive at 1%. Did the appraiser provide any market support.?That is what I would ask. What the basis is for 90 days? Was your market really flat up until three months ago? If this is for residential appraisal I’ve got to believe you can get historic housing trend data or find a few paired sales from the local MLS. If this is a commercial assignment, there are plenty of resources and published market studies to support market conditions. Sounds boilerplate…perhaps this was a copy/paste From another template report and overlooked in the review process? Good lk!

 

[Terrel L. Shields]

Was your market really flat up until three months ago?

Unless the market is going up rapidly then small incremental changes are buried in the statistical noise of the mere difference in prices. Look at your last appraisal. What was the spread between your high and low comp? zero? I bet not. I bet you had differences of several thousands of dollars. So what is that variation within the mean of the sales used? 2%? 4% 10%??? Well, you are not going to make any adjustment smaller than you margin of error between your comps that is legitimately and statistically defensible.

As it stands, in NE Oklahoma I am seeing a very flat market since January. NW Arkansas OTOH, still going up measurably but sales numbers are lower... I think the market is turning but it will take until August or so to see.

 

[Tom4value]

6 months ago the market was different, during slow winter season and mortage rates were lower so time adjustments can be higher.
Past few months, rates went up and start of home buying season so no need to make time adjustments.

Ding, ding, ding! We have a winner! Appraisers when figuring out market adjustments always assume that it changes at a constant rate. This is how they get into trouble.