Lender says no time of sale adjustments!

[Steve Owen]

What lenders are really saying if they don't want market conditions (time) adjustments is that they want comps that are new enough not to require them. Even if you know that the market is going up by 4% per year, it is not appropriate in most circumstances to take the adjustment on a house that only sold three months ago (just my opinion) because the market is not perfect. It has been standard (right or wrong) in many parts of the country to ignore time adjusment unless the sale is more than one year old. This may be something that is different in your area, or it may be changing. But, I still don't think it's a good idea to take the adjustment for a recent sale. In my business I often use commercial or industrial comps as much as five years old. When a comparable sale is that old, you can look back over time and see what the increase (or decrease) in prices has been. I really have a hard time with thinking you could do that with a sale less than nine months old. No one is good enough to have their finger on the pulse of the market that tightly. (Just my opinion, again.)

 

[Austin]

George: Here is the problem I have with your situation of hyperinflated prices: If you get into a short-term excess demand situation in a local area as you describe, two possible things can happen: 1. If the appraisers make time adjustments to accommodate the high demand, when the market returns to equilibrium, then the lenders are stuck with an inventory of homes with more owed than they are worth. 2. On the other hand, if you follow my theory and stick to the cost approach theory, if prices go up, then the buyers can pay the difference and assume the risk. The difference is either the buyers assume their own risk and consequences of their decisions or the taxpayers and public assume the burden. Plus the fact, if you accommodate the hyperinflation you are just adding fuel to the fire.
I grew up with a friend and went from the 1st grade through college with him. In about 1971, he moved to California. He came home and showed me a picture of his new $70,000, 1,500 sf rancher. Ten years later he come home and told me he just acquired a big equity line credit loan based on a $225,000 appraisal on his house. Then about a year later the phone rings at 3:00 AM and it is my buddy from California. He wants to borrow $2,500 before 9:00AM Monday morning. I ask what for? He says to make his house payment. I ask what happened? He said he lost his job. I said, “If you can’t make the payments, why don’t you sell the house?” He said, “because I owe $200,000 on it and it is only worth $120,000.”
The moral of the story is you can’t get out of balance with the national or world economy, because if you do, this is the result. How many instances of this situation does it take to get the attention of people in California? Must be something in the water out there.

 

[George Hatch]

Steve,

It goes without saying that appraisers are typically going to present the most current data they can get their hands on. I highly doubt the Rijman was applying a time adjustment in a situation with (3) 30-day old sales. I don't think anyone would argue that a time adjustment for a 3-month old sale in a 4% annual increase market is unnecessary. Especially when the market segment involves a $175,000 value range. What's that come out to, less than $2,000? But a 6-month old sale in a 20% market with $800,000 homes is a different story (and yes, we have tracts of $800,000 homes here). That adjustment is more like $80,000. Sometimes ya gotta do what ya gotta do.

This is not to say that I am a big time adjustment user. Fact is, I rarely use that method myself; except in those few cases where I am forced to use a really dated sale, the adjusted value of which totally throws off my analysis in comparison to the more recent data. Instead, I present all my sales comparables chronologically and try to throw in a pending sale. If there is a trend, it should readily apparent. This allows me to justify going to the top of the adjusted range after accounting for any other differences. IMO, if we can't demonstrate a trend in the subject's market segment, we shouldn't be trying to account for it. I prefer to not make a lot of adjustments and instead rank the subject within the range whenever possible. But that doesn't mean that making individual adjustments is by any means the wrong way to analyze the value. It's just another tool. The difference between a novice and a journeyman is having the right tools for the job at hand and knowing how and when to use them to their maximum utility. Even regression analysis has its place. I just don't happen to think that place is everywhere.


$.02

George Hatch

 

[George Hatch]

Austin,

Whoops, you got your latest post up while I was writing my last pot.

I would completely agree with you that our job description does not include trying to guess where the market is going. It is not our job to facilitate a speculative venture. It is our job to report what is actually happening based on the data at hand. If the data points one way, we can hardly go the other way. If the market is increasing, then its increasing. It is not our job to underwrite the deal and protect the lender's portfolio. It's our job to give them the information they need to make an informed decision. Of course the market will fluctuate, that's what the market does. Some more than others. Fortunately, our work is supposed to be judged by what is actually happening as of the date of the appraisal. Perfect 20-20 hindsight 2 years later doesn't count.

Besides, short term demand is within the eye of the beholder. We are currently in our 5th year of price increases. Because of the costs of development coupled with a relative lack of developable land, its very expensive to build a house here. Supply and demand, and we do have an imbalance. Although some of the upper price ranges have slowed or stabilized altogether, other segments are still going up, as in closed sales in the same trend pattern. What are we supposed to do with that already closed data, ignore it because we think it can't go on? The buyers and sellers in these markets surely aren't ignoring it.

As for your friend's California experience, you are referring to the '80s meltdown where creative financing was the norm, 16% financing was common and appraisers commonly didn't account for those conditions in their analyses. There were probably some other appraisal errors on the appraiser's part as well. Truth is, your friend also didn't account for that, and neither did that lender. It looks like he had a 14.6% interest rate. The lender should have figured that interest rate would require a strong long-term debt service. I wonder if that lender made it through the 90s? There were plenty of errors to go around. Incidentally, if your friend had been able to hold on for a few more years, he would have truly tripled his money. Maybe better. My grandmother bought a house in Huntington Beach, CA., in 1972 for $50,000; it sold in 1987 for $250,000. I'll bet that same house today is worth $500,000, real money, with lots of comps. It's just a tract home on a small lot, no view nor is it even in a great location. Run of the mill.

I think the 90's recession is a slightly better indicator because of the lack of creative financing and onerous interest rates. Here in California, the late 80's was a boom mentality with a lot of fear buying going on. The sae thing is happening today. That doesn't mean those values weren't there then and aren't there now. Previous performance is no guarantee of future performance. That's one reason why they charge interest, because there's some risk. How much interest and how much risk are decisions they are suppoed to make because it's their money. I spend a lot of time discussing this same issue with my clients, as I'm sure you do. If they can't see that there is some risk to a 90% loan right now, then they are ignoring our warnings. That makes it their problem. It becomes our problem when we start telling them only what they want to hear. It makes no difference whether they force us to do it or we volunteer to do it because we are afraid for our jobs.

Anyhow, I agree that hyperinflation in any market is not sustainable and that appraisers shouldn't fuel the fire by going beyond an already identified trend. But if a lender wants to protect themself during perilous times, and they should, they should simply cut the loan or eliminate the program. They shouldn't ask the appraiser to assume that responsibility by reporting a value that doesn't exist. We have to do our job competently; they should do the same.


George Hatch

 

[Alan Simmons]

For the past month my OM class (basically applying various QM to OM) focused on forecasting. Great class; the professor is a VP for a NASA contractor and has his Ph.D. in Engineering. This class is better than anything similar I ever had getting my Finance degree.

We studied several methods used to recognize trends (+/-), then used various techniques to check them for seasonal or random variations. Finally we used MAD and MSE to check the error to determine the best forecasting method.

I have only had one opportunity to apply this new knowledge so far. It really is very simple math and statistics, but it ended the disagreement very quickly.

 

[rijman]

Another update, broker calls my office to tell us that Flagstar backed down off the "time of sale adjustments" and the loan is going to fund with the appraisal as is. The broker also indicated Flagstar is now re evaluating their policy against time of sale adjustments. (No one at the corporate office I talked to could even confirm this policy existed!). I am not sure what part I played in this, I did leave three detailed messages to the top QC people in the corporate office, yet none of my calls were returned, and I talked directly to 2 senior UW's at that office. Mysteriously my broker was contacted with the good news. My point was that a lender who requires appraisals conform to USPAP and FNMA/FHLMC guidelines yet dictates no time of sale adjustments is creating a catch 22. Meeting the lenders guideline can directly conflict with USPAP and FNMA/FHLMC guidelines.

George, I agree 100%. Who are we to decide what adjustments are made, the market dictates that. To arbitrarily decide an adjustment for time of sale won't be made is dictating the market rather than being an objective unbiased reporter of data. If a time of sale adjustment is warranted and clearly supported by market data we are required to make that adjustment, even our Limiting Conditions and Appraiser's Certification says we must make adjustments where warranted that affect the subject's value.

It is not our job to look out for the lender or investor. Our job is to provide an unbiased appraisal report in conformance with USPAP and the Limiting Conditions/Appraiser's Certification contained within every appraisal report. I am appraising as of a specific date. If the market turns down tomorrow that is not my concern for appraisals done in the past, that is the lender and investors problem.

This is a good day. I will take my little victory and celebrate until the next tough assignment or lender condition hits my desk, which should be all of 15 minutes.

Regards,

 

[Austin]

Alan: There is a better way to check the accuracy of regression or statistical methods than test statistics. Once you find the regression equation for the trend, apply it to the input data and see how well it predicts the input data and observe the deviation of the difference between the actual and predicted data. The perfect trend estimate is the equation that can predict the input data with little or no variance. This is significant in gauging appraisal regression results because we are not dealing with random variables, we are dealing with highly selected and verified variables that we know affect price and how much they affect price. People have always criticized regression because they didn’t like the results of the test statistics. In my opinion, that is not a valid criticism because the tests statistics don’t mean the same thing with know non random variables. For example, a 2%+- error to a base of 1,000 is plus or minus 20. With selected subjective data a 2%+- variance to a base of 100 is plus or minus 2. Same test statistics will be reported for both examples, but as you can see, the results mean two totally different things. Then too, there are no test statistics for the present appraisal method. It is a good thing too, because the present sales comparison sequence of adjustments is totally irrational and illogical. In my opinion!

 

[Alan Simmons]

Austin:

We are only up to Chapter 17. Please do not give away the end of the book.

In my finance classes we were constantly being taught that price (value) was a function of future expectations, discounted. Basically that the price of gas at the pump today is based upon the perceived price six months from now. Also that past performance was cute but not really a good indicator of the future. Instead we were taught to study leading economic indicators. Not that this is any more reliable, but Time Adjustments appear to be attempts by Appraisers to reconcile this.

I do find it funny sometimes when Appraisers make time adjustments. It seems (and I know this is a GENERALZATION) that if the market is so dynamic as to require a time adjustment, then there should be abundant sales and listings to justify such. Thus minimizing the need for an adjustment.

There was an article two years ago warning Appraiser about using those government "increasing average price statistic." Those reports only indicate that the mean price for houses in an area has increased X%, not that all houses in an area have increased X%.

A Reviewer gigged me last year for NOT making a Time Adjustment on an eight-month-old sale. I have been tempted to post this Reviewer's comments (on this and other items) several times.

 

[rijman]

Alan, time adjustments are no more than recognizing a market change from one period to another. The future does not factor into this whatsoever. We are gauging changes between the off market dates of comparables up to the appraisal date. Have values changed during that period? I don't believe appraisers or lenders have the right to arbitrarily decide whether or not time adjustments are made. That is dictated by the market. Appraisers are unbiased reporters of market data. If values are increasing at 1% per month does the appraiser have the right to decide not to recognize this and make the appropriate adjustments simply because they choose not to make time adjustments? That could be construed as a violation of USPAP and clearly not in conformance with the standard LImiting Conditions and Appraiser's Certification within every FNMA/FHMLC form report.

As far as having sufficient recent sales, our increasing values in southern California are a result of limited supply, therefore limited sales. It is not uncommon to lack a recent similar sale that has closed within 2 months. A two month old closing could actually have gone off market 3-4 months ago during which time values have increased at 1%/month in many of my areas. As for active listings, they help just as pending sales but we all know that the value has to be supported with closed sales data, that's the FNMA/FHLMC way.

I really believe many appraisers do not make time of sale adjustments because they are afraid lenders won't accept them and because they don't know how to properly support them. Neither are acceptable reasons to ignore market changes because they can keep the appraiser from reaching "market value," which is the main goal of our lender based appraisal reports.

Regards,

 

[Austin]

Alan: Let me tell you my time adjustment story. In the early 1970’s, I was in a family brokerage-auction-appraisal business. I didn’t want to do appraisals but our office was next door to a second mortgage lender and my dad didn’t want to do it, so he made me do it. It was good money, about $150-$200, back then using two comps. Just a FNMA form with no details, very simple, no reviewer, no USPAP.
Another real estate agent in town saw how much money I was making so he decides to go into the appraisal business. He moves in on my second mortgage client and takes about half the business. I was talking to the load officer one day and he said he used the other appraiser because his appraisals were much higher than mine and he could lend more money. He knew the appraisals were wrong, but that was not his concern.
I ask if I could see one of the other appraiser’s reports. He showed me one.
What this guy was doing was using Marshall and Swift cost manual and he didn’t know how to use it. He was using a good quality classification to appraise fair quality houses and coming up $20,000 higher per house than I was. When he got to the sales comparison approach, he made a $20,000 time adjustment to balance things out.
The LO won an award as being the leading LO for the company for the entire southeastern region of the US that year. He out loaned every other LO in the region. As he told me; if the worst happens, it is that appraisers problem, not mine. Now, why didn't the worst happen? Because other appraisers were making time adjustments too.
I think that is what most appraisers use the time adjustment for as well as: Design & appeal, site view; quality of construction; and functional utility. If the data won’t make the deal work, these adjustments will. In 28 years of appraising, I have never seen any justification for a time adjustment on a residential appraisal. I still contend it is not possible to measure as of date of appraisal. If you go back and study data for the preceding years, you way support a general rising price trend, but as to a specific property as of a certain date using sales less than 6 months old-no way.
My daughter graduated from college in 1998, and was house shopping in a large metro area of Virginia. She had a very good Realtor helping her. Daughter found the house she liked and called me in to check it out. The Realtor knew I was an appraiser and had her story ready. She said the house sold three years ago at $125,000, and if you figure a 3% increase per year that means the house is worth about $137,500 now. I ask her where the 3% figure came from. She replied: everybody knows that, it is standard practice. That is the only logical justification for a time adjustment I have ever heard, and it does not make sense. You make a time adjustment because everybody else makes time adjustments. Well if she was using the subject as a comp I could see a time adjustment in this environment, but if the other comps are less than six months old means a time adjustment of 1.5% which is totally perfunctory and meaningless. In my opinion.
PS: This is also the best argument I have ever heard for using three year old comp sales. If you know a time adjustment of 3% is built in, then what is the big deal about current sale data? In my opinion.