Time Flies

It has been several very busy months since the last post. I have been busy gathering and verifying tax card information for the Geauga County Auditor in order to provide an updated database for the 2011 reappraisal in that county. In doing so, I have been logging experience hours in non-residential appraisal work, to be applied toward my application to sit for the General Appraiser certification exam.

Meanwhile, while not intending to do so, I have failed to provide a follow-up on the query that was the subject of the last post. I received the following reply (in part) to the information I sent to the troubled borrower:

Jim,

Thank you for the information and your thoughts. I will certainly take your advise [sic]. I noticed that remodeling averages are from 65%-77% depending on the project which certainly shows the appraisal was estimated much lower. I understand what the auditor's records show but they came in here and documented 4 bdrm and 3 full baths. Are we paying for an accurate appraisal? Two more questions 1) how is land value figured? And replacement cost per sq ft? Our insurance company has it much much higher. I'm just trying to wrap my mind around this as to why it came in so low. We had this same thing happen before were [sic] the first appraisal came in much lower 50-60000. We spoke with Chase and explained everything and they issued another appraisal. The appraiser who came out has intimate knowledge of our area in fact he explained the his mother is a Realtor in this area and he knows the value. It came in at $399,000.

I never did email the man back, but my notes on my thoughts have been cluttering my desk for the past 5 months, so here goes.

First off, the last few sentences of his email are very troubling to me. If, during the original loan's appraisal process, the value came in $50-60K less than what was "needed", and the bank sent out a second appraiser who "knows the value", I, as a very suspicious, untrusting, and experienced reviewer tend to immediately class the second appraiser as one who could also "hit the value", whatever it took. Sorry, I have not just one noid, but a paranoid where these matters fall.

Is the borrower "paying for an accurate appraisal"? I suppose that is a very nuanced question. By law, the client is the bank. The borrower pays for the appraisal, but the bank is the intended user. Certainly the bank should be interested in a well reasoned and factually supported opinion of value, but the words "accurate" and "opinion" do not belong in the same sentence without some other explanatory verbiage. The facts should be accurate, and the opinion should be unbiased.

Certifications 1, 4, 7, 9, 11, and 12 of the FMNA Form 1004 (3/2005) address the accuracy and adequacy of the data used in the analysis. Adjustments may be difficult in a declining market with a scarcity of sales, and the appraiser must use his best judgment. Much more attention is also being paid now by lenders to geographic competency (Certification 11) -- where does the appraiser live and normally work, and is he truly familiar with the market? n.b. -- it should be recognized that an appraiser can become extremely competent in a market at some distance from his normal base, but the onus is on the appraiser to prove such competency in both his choice of data and the resulting reasoning.

As to the value of the land, there are three normally accepted methods for determining land value (the value of the land as vacant and unimproved). The first is market sales comparison. Examination of recent sales of vacant buildable parcels should provide the appraiser with information to form an opinion of value for the subject parcel. The subject parcel had 2.65 acres; the borrower cited a 1 acre lot selling for $78,000. That would be a very poor comparable in a built-up area like the City of Solon; more comparable sales would be vacant residential lots between 2.5 and 3 acres in size, and such sales would be extremely rare. The $5750 adjustment for site size would be the appraiser's opinion of the contributory value of the site size difference; the important thing there is support for the site value which the appraiser is supposed to detail on page 3 of the Form 1004. That support is required by FNMA, regardless whether the Cost Approach is completed. (We'll discuss that a little further on.)

The other two commonly used methods for determining site value are extraction and allocation. Allocation is fairly simple and is built into most models for AVMs and CAMA. In its simplest form, when the cost to build the house, new, is calculated, a percentage figure is multiplied by the cost to give a land value. In its most reliable form, the percentage is derived from market studies of builder costs, and is adjusted internally for positive and negative land features. In practice, however, it is often a "pluck from air" (PFA) figure and, within assessment records, frequently produces hilarious results. (Sorry, but I have in mind a lot in the City of Akron which had an old house on it. The year the house was demolished, the lot was valued by the Auditor at $8,000. The next year, with a new house -- built with government grant money and not at all a market-driven project -- on the lot, the tax appraisal land value had jumped to $20,000. The mean and median price of housing in the neighborhood was unchanged. Such is the magic of carelessly applied allocation.)

Within built-up areas with few vacant land sales, extraction is the third commonly used method for determining site value. To properly do a land value extraction, sales comparisons must be run on a number of recent sales of homes where significant details of their quality and condition are known to the appraiser. The appraiser must do a Cost Approach appraisal of each of the comparable homes (comparable in this context being more applicable to the characteristics of the land than the houses), with great care taken to account for all forms of depreciation that might have affected those homes. The resultant depreciated cost of the improvements is then subtracted from the raw sales price to indicate the contributory value of the land under the home. Unfortunately, while many appraisers cite extraction as their method for determining site value, it is my experience from looking at their reports that a good many of them either have no idea what they are about, or they are incredibly lazy.

In general, FNMA does not require the Cost Approach to be completed on the Form 1004 unless it is the appraiser's opinion that it is needed to provide support for the reasoning behind the value opinion. Appraisers generally use cost tables to determine the cost to construct the home using current methods and materials, and then apply estimates of depreciation (physical, functional, and external) to arrive at a depreciated cost of the improvements. This is then added to the site value figure to indicate a market value for the home. In theory, it works if an honest site value is used and if all three forms of depreciation are properly accounted for. Lenders want to see the Cost Approach used because they feel it is a shortcut to providing a figure for which they can require the borrower to supply an insurance policy. Most appraisers are now including a disclaimer that the costs "are not to be considered indicative of insurance replacement costs", this despite the fact that the form requires them to check that the costs are either for reproduction or replacement.

An insurance appraisal is generally not a market value appraisal. Insurance policies are written to allow property owners to recover the utility of their asset if it is fully or partially destroyed. Replacement cost per square foot, for insurance purposes, must include the costs involved in removing any debris from a damaged home, and then reconstructing it to its former usefullness. This means that no depreciation can be entered into the process, since the home, when the owner again takes up residence in it, will be restored to "new" condition. The replacement cost may be, and usually is, far more than the market value of the home. If the home is purchased for $100,000, but is of such age and design and condition that to rebuild it to modern standards would cost $200,000, the borrower may only be required by the lender to insure it for the purchase price. If it burns to the ground and must be rebuilt, the insurance company will only pay the policy amount. The borrower will have to come up with the rest. If he cannot qualify for the additional loan amount, or does not have the cash, he may have to default on the original loan. Consequently, homeowners should have their houses independently appraised for insurance purposes, and under no conditions should they rely on a market value appraisal for insurance value uses.

There is one final thing cloesly related to the above which needs to be mentioned and which will make me persona non grata in some circles. The gentleman who originally wrote to me acknowledged in his rejoinder that "I noticed that remodeling averages are from 65%-77% depending on the project". A dirty little secret is that this also applies to new construction, albeit to a lesser degree. Just like a new car, which loses value the instant it is driven off the dealer's lot, in a stable market, a new house depreciates, possibly 5% - 15% depending on where it is located, the instant the buyer signs the contract with the builder. The practice of selling new homes with 100% financing is still occurring, but it is not politically expedient to suggest to FNMA that new home buyers should be required to put down 20-25% of the purchase price in cash. Especially not with the government trying to stimulate the economy by offering tax incentives in a market that is in significant oversupply.

Hi-yo, Silver! Away!