No Adjustments In Sales Comparison?

[Mr Rex]

So toss out the high and low as outliers. Does this materially change the value center point?

 

[gregb]

What do the price per unit, price per bedroom/room and GRM'S of the comparables indicate? It appears that price paid per square foot is emphasized, not something I would usually do in my market.

 

[hastalavista]

Comparable Sale 1 $360.29 x 14.29% = $51.47
Comparable Sale 2 $273.65 x 14.29% = $39.09
Comparable Sale 3 $318.37 x 14.29% = $45.48
Comparable Sale 4 $332.92 14.29% $47.56
Comparable Sale 5 $275.21 14.29% $39.32
Comparable Sale 6 $325.64 14.29% $46.52
Comparable Sale 7 $366.89 x 14.29% = $52.41
Total: 100.00% $321.85
Rounded: $320.00
x
67,470
$21,590,400
Rounded: $21,600,000

A multifamily project based on price-per-sqaure foot?
Not necessarily what I would expect, but perhaps in that market, buyers/sellers price these properties on $/sf basis?
New York is (as far as I know) significantly impacted by rent control; hopefully and if applicable, this item is factored in somewhere.

 

[reviewbe]

The appraiser DID weight them. Just weighted all equally.

Difference between hi and lo in the 7 sales is roughly $13 (per SF?), roughly 33% of the lowest figure and about 25% of the highest. Thats a pretty big spread to have NO adjustments.

Don't know what the fee was, but regardless, for a $21M property, probably warrants at least a little bit of analysis.

 

[leasedfee]

it's possible that the comps required no adjustments. i have done a number of condos over the years that were apartment-styled and all identical matches.

Please show me that requirement. If they are all similar, he can weight them all similarly.

Agree, both possible. Mr. SandiegoBrian, you don't give sufficient evidence that the appraisal was in error. USPAP also doesn't require a grid, though I much prefer quantitative analysis. The appraisal gave a weighted average, where each comp had uniform weights. If he had weighted them, say, 5%, 11%, 17%, . . ., he'd given a weighted average with non-uniform weights. I long long ago quit weighting (e.g., 5%, 11%, 17%,. . . ) (or e.g., 14.29%, 14.29%, etc.) the comps because those weights are purely fabricated. No, zero, nada, zip market support. I reconcile to the best comp(s) for reasons stated or unstated, and dismiss the weakest comp(s) as being merely supportive. This is closer to how brokers and the buyers speak and analyze.

 

[J Grant]

What is the goal of an appraisal or a review- to present results that are credible and supported. If all the comps are equivalent to the subject, weighting them equally would be credible. However, if comp 1 and 2 were equivalent to the subject, and comp 3 and 4 have substantial differences from the subject, how could one defend weighting them equally? Comp 1 and 2 would have to be given more weight/consideration in the reconciliation.

 

[SanDiegoBrian]

Agree, both possible. Mr. SandiegoBrian, you don't give sufficient evidence that the appraisal was in error. USPAP also doesn't require a grid, though I much prefer quantitative analysis. The appraisal gave a weighted average, where each comp had uniform weights. If he had weighted them, say, 5%, 11%, 17%, . . ., he'd given a weighted average with non-uniform weights. I long long ago quit weighting (e.g., 5%, 11%, 17%,. . . ) (or e.g., 14.29%, 14.29%, etc.) the comps because those weights are purely fabricated. No, zero, nada, zip market support. I reconcile to the best comp(s) for reasons stated or unstated, and dismiss the weakest comp(s) as being merely supportive. This is closer to how brokers and the buyers speak and analyze.

Thank you.

 

[NachoPerito]

If the appraiser is not planning on making quantitative adjustments then why put in the grid at all? I am ok with qualitative adjustments only on multifamily as long as you discuss it. It looks like this appraiser didn't do qualitative or quantitative. The sales comparison approach never carries much weight anyways and adjustments are hard to quantify. But also giving equal weight to all comps... seems weak.

 

[Gobears81]

Comparable Sale 1 $360.29 x 14.29% = $51.47
Comparable Sale 2 $273.65 x 14.29% = $39.09
Comparable Sale 3 $318.37 x 14.29% = $45.48
Comparable Sale 4 $332.92 14.29% $47.56
Comparable Sale 5 $275.21 14.29% $39.32
Comparable Sale 6 $325.64 14.29% $46.52
Comparable Sale 7 $366.89 x 14.29% = $52.41
Total: 100.00% $321.85
Rounded: $320.00
x
67,470
$21,590,400
Rounded: $21,600,000

If the market is active enough to get seven comparable sales of apartments that the appraiser can argue have no significant differences, I'd do two reasonableness tests. The first would be to look at the coefficient of variance on the unadjusted per unit price, per bedroom, and per square foot. Is the price per square foot resulting in the lowest COV? Whatever results in the lowest COV offers evidence of how the market analyzes these properties (though isn't in itself the sole conclusion that said unit of comparison is the best representation of the market).

Another thing that I'd look at-assuming that the price per square foot is the best unit of comparison, what is the gross or net income per square foot for the subject and the comparable sales? If the subject property is full of 450-square foot studio units rented for $500 per month (presumably much higher in NYC) and it is being compared to a sale of solely 1,000-square foot 3-bedroom units rented for $900 per month, all else equal, that would suggest a 23% upward adjustment to the three-bedroom property, even if the buildings are otherwise entirely comparable. Personally, I wouldn't expect to see that kind of market variance in the NYC apartment market assuming that everything is implemented correctly, but I also know absolutely nothing about the NYC apartment market.

 

[Michigan CG]

I don't get the $/SF analysis. If I were to do something similar I would use the four most similar sales and do it by unit and if I am going to put a grid in I would at least put superior, inferior or similar in that grid. When there is $93/SF difference in the high and low sales prices then there has to be differences in the properties that can be measurable.

Seven sales that are that wide apart is overkill and makes the report less reliable than using the four best.

$300/SF!!!!! Holy cow, this ain't Iowa.