New market analysis changes.

[alebrewer]

My bolding.

Don't speak it into existence! All jokes aside, that is an advanced concept that a former Forumite used to beat that drum incessantly for years and taught classes promoting the practice. Memories.....

I only use them infrequently - and generally when I'm appraising a property that is on the edge of a market's GLA range - high or low. Beats having to make GLA AND functional adjustments. For GSE/HUD work, though, I always just apply consistent adjustments and then throw in a functional adjustment if necessary.

 

[Non Sequitur]

I've done it. Just like you wouldn't adjust the same for baths in a 3 bedroom 10 bath home, you might adjust differently for a home with larger than typical GLA or lower than typical GLA to the point that it is in the expected range for the market.

We are talking about in the same report, basically adjusting a Comp1 at a different GLA rate than Comp 2 and so on.

Edit: See alebrewer's post above.

 

[shrubberyvaluation]

We are talking about in the same report, basically adjusting a Comp1 at a different GLA rate than Comp 2 and so on.

Edit: See alebrewer's post above.

Sure, if there are diminishing returns. If Comp 2 is larger than typical for the area and is an overimprovement, the GLA adjustment might be different, at least for part of the GLA difference. Of course its better not to use that comp, but sometimes its all you have. Just like if the subject has 2 full baths, comp 1 has 1 full bath, and comp 2 has 3 full baths. The market might value the 2nd bath more than having a third bath. You could have different adjustments. Marginal utility, same with GLA.

Now if comp 1 and 2 have the same GLA and are adjusted different then that I dont get unless the quality is much different perhaps?

 

[CGinMN]

A serious question for all of the high horse appraisers:

How long have you been supporting and including nonlinear market adjustments that are sensitized for seasonal influences? What has been the response from underwriters and reviewers when you included 3 different time adjustments in a report? And how many listing are you gridding as support for changing markets?

I've been doing this my whole career. I never got any pushback, but I wasn't doing a lot of lending work up until 5-7 years ago.

If you overlay the monthly median sale price trend against a linear trendline, what do you see?


Half the year sales are falling below trend, and half the year they are above trend. This is mostly because of the compositional effects of the housing stock. Nicer/bigger homes selling in March-August compared to September-February. However there are differences in the median seller motivation as well. Neither of these lines perfectly demonstrates price change. In reality, the measured change in the market is something more like the red line.

Most of the increase is happening March, April, May, then stabilizes, sometimes with a pullback between Thanksgiving thru January (not shown). Let's assume the linear market appreciation is +6%/year, and you are appraising a home in the very early spring market. Appraisers can make linear adjustments, but a January sale is likely going to adjust below trend, and the August sale might be above trend until late March when the pricing surpasses the prior year. So, if you are doing a linear adjustment, you would ideally account for this within your reconciliation. Or you can make non-linear adjustments based what's actually happening (red line), which requires getting on the phone and talking to agents and gridding out sales that are under contract. Once adjusted, these contracts will likely support upward adjustments to the prior year's sales that is above the linear change. That adjustment can be supported by sensitivity, much like many other adjustments are developed. If you're just using the prior year's linear appreciation rate without acknowledging this, you are missing the early spring market, which is why the frequency of low appraisals follows a seasonal trend.

I review appraisals and most appraisers are not doing this level of analysis, and frankly that's ok because you can figure out ways to work through it using reconciliation. But it's important to at least have some knowledge of these trends if you are an appraiser in a heavily seasonal market. Fannie's illustration sucks because it is confusing for a number of reasons. However, I think the point is more to prepare lenders for the possibility of non-linear market adjustments, rather than to show appraisers how to do things.

 

[shrubberyvaluation]

I've been doing this my whole career. I never got any pushback, but I wasn't doing a lot of lending work up until 5-7 years ago.

If you overlay the monthly median sale price trend against a linear trendline, what do you see?
View attachment 96363

Half the year sales are falling below trend, and half the year they are above trend. This is mostly because of the compositional effects of the housing stock. Nicer/bigger homes selling in March-August compared to September-February. However there are differences in the median seller motivation as well. Neither of these lines perfectly demonstrates price change. In reality, the measured change in the market is something more like the red line.
View attachment 96367
Most of the increase is happening March, April, May, then stabilizes, sometimes with a pullback between Thanksgiving thru January (not shown). Let's assume the linear market appreciation is +6%/year, and you are appraising a home in the very early spring market. Appraisers can make linear adjustments, but a January sale is likely going to adjust below trend, and the August sale might be above trend until late March when the pricing surpasses the prior year. So, if you are doing a linear adjustment, you would ideally account for this within your reconciliation. Or you can make non-linear adjustments based what's actually happening (red line), which requires getting on the phone and talking to agents and gridding out sales that are under contract. Once adjusted, these contracts will likely support upward adjustments to the prior year's sales that is above the linear change. That adjustment can be supported by sensitivity, much like many other adjustments are developed. If you're just using the prior year's linear appreciation rate without acknowledging this, you are missing the early spring market, which is why the frequency of low appraisals follows a seasonal trend.

I review appraisals and most appraisers are not doing this level of analysis, and frankly that's ok because you can figure out ways to work through it using reconciliation. But it's important to at least have some knowledge of these trends if you are an appraiser in a heavily seasonal market. Fannie's illustration sucks because it is confusing for a number of reasons. However, I think the point is more to prepare lenders for the possibility of non-linear market adjustments, rather than to show appraisers how to do things.

Sure but how useful it is to go back 5 years for most appraisals? How does that assist with market condition adjustments. It can give some insight on seasonal fluctuations, but whether its seasonal or not doesn't matter as an opinion of market value is for a specific date.

 

[CGinMN]

Good post alebrewer.

And I don't know anyone or have reviewed any appraisal supporting or applying nonlinear market adjustments. And I have reviewed 1,000s of reports in multiple states, not a single one with nonlinear market adjustments.

You've never reviewed an appraisal where the appraiser didn't adjust one sale but adjusted another? This is non-linear.

 

[Tom D]

Sure, if there are diminishing returns. If Comp 2 is larger than typical for the area and is an overimprovement, the GLA adjustment might be different, at least for part of the GLA difference. Of course its better not to use that comp, but sometimes its all you have. Just like if the subject has 2 full baths, comp 1 has 1 full bath, and comp 2 has 3 full baths. The market might value the 2nd bath more than having a third bath. You could have different adjustments. Marginal utility, same with GLA.

What are you suburban and rural people working with. I should never complain about, and i don't, my princess appraiser life doing big urban little row homes. All my comps look alike with most GLA differences less than 200 GLA, lots ignored. I gotta start giving back some money to the AMC abused appraiser funds. Let me think, nay.

 

[shrubberyvaluation]

You've never reviewed an appraisal where the appraiser didn't adjust one sale but adjusted another? This is non-linear.

Well I have where the recent comps are not adjusted, never thought of that as non linear.

 

[Terrel L. Shields]

 

[Joe Flacco]

You've never reviewed an appraisal where the appraiser didn't adjust one sale but adjusted another? This is non-linear.

As far as I know, you and I are the only people that look at the data in a seasonal context.