Are Appraisals Really Relevant?

[Vermonter]

The 'error rates' of off-market property for AVMs is 7% (Zestimate & Redfin, Realtor.com and Movoto don't give their error rate and are probably even higher than 7%). That is way too high for prudent lending decisions. Appraisers know the immediate neighborhood market better and recognize upgrades better than AVMs. Reports by appraisers are more 'relevant' than using AVMs.

What's the error rate on human appraisals? We know it's not ZERO

We have to realistically look at the entire picture. Collateral value is a just a piece of the pie. 20 years ago it was a much bigger piece, but as automation improves other aspects of the lending pipeline that piece will become smaller. As a borrower becomes less of a default risk, the accuracy of the collateral value becomes less important.

 

[J Grant]

What's the error rate on human appraisals? We know it's not ZERO

We have to realistically look at the entire picture. Collateral value is a just a piece of the pie. 20 years ago it was a much bigger piece, but as automation improves other aspects of the lending pipeline that piece will become smaller. As a borrower becomes less of a default risk, the accuracy of the collateral value becomes less important.

That is a baseless assumption on your part.

Without MV benchmark, however acheived, for property valuation it will lead to higher losses, because the prices have vastly increased even on starter homes compared to 20 years ago. The Collateral is a huge piece of the pie. Without it, we are talking about the high interest rates and short term repayment terms of non collateral loans.

 

[Zoe]

That is a baseless assumption on your part.

Without MV benchmark, however acheived, for property valuation it will lead to higher losses, because the prices have vastly increased even on starter homes compared to 20 years ago. The Collateral is a huge piece of the pie. Without it, we are talking about the high interest rates and short term repayment terms of non collateral loans.

Yeah, lenders base the interest rate and terms of loan on appraisal and other factors. If a prudent buyer and seller are doing cash transaction or owner financing or whatever, the appraisal is very important. If they are doing refinance, appraisal is very important.

 

[J Grant]

With regard to how to address the fact that market value is a concept and even the best supported MV can have a variance, the loan officers should be allowed to add or subtract up to a 3% cap from the MV opinion, depending on borrower credit, property type risk, etc.

Actually, lenders are allowed to do this now, but they can not freely do it as they would be responsible for the value if they change it up or down.

Either give the lender wiggle room or the RE agents and lendes should inform buyers ( and then sellers ) that when a borrrower finances , there is no guarantee the $ mv opinion will be the target amount needed for a loan, and thus parties can re negotiate a price or come up with more cash.

The competition to appraisals are WAIVERS, which do not have to meet the regulations appraisals have, and thus the target value that makes a deal work - a loan officer value or the SC price is the property value - ( as long as it fits within the FF AVM)

 

[Vermonter]

hat is a baseless assumption on your part.

If you say so....

Prevalence of GSE Appraisal Waivers

The use of appraisal waivers at the GSEs has exploded in recent months and now accounts for over 40% of all valuations.

www.aei.org

 

[Zoe]

If you say so....

Prevalence of GSE Appraisal Waivers

The use of appraisal waivers at the GSEs has exploded in recent months and now accounts for over 40% of all valuations.

www.aei.org

I won't touch that. I'll refer back to homogeneous and heterogenous markets.

I believe a waiver is plausible in a very homogeneous market. When you get in very heterogenous markets, a waiver is useless and very risky to the overall real estate market. Who determines these waivers?

Who does it? Does God do it?

 

[DTB]

Most here know that I have been property since I was in High School, my Dad always touted real estate investments and had quite a few apartments aside from his main business.

When, at age 41, I started appraising I was fully impressioned that I was THE D3P. Did my very best on appraisals, never considered myself a number hitter but also realized that the market participants and the data isn't wrong very often.

By age 42-43 I was disillusioned that my report was merely a compliance document. Oh sure, I was important when they needed me, for a rush job or last second "revision".

Since I was buying and selling a fair amount of foreclosures at the time, I came to realize that the appraisal written 2-3 years ago had very little bearing on the lenders' risk of loss.

Worst case scenario there is a SFR with an opinion of value reported that is 10-15% pushed and the borrow put down 3%. Unless it's a first payment default there is little correlation back to that appraisal written 3 years ago. The house is probably stripped, the lender paid the taxes and insurance since the default. Does anyone think that they are expecting that value form 3 years ago to cover them? Impossible, I say.

The only issue is that becomes the gotcha moment for the appraiser, now you are suddenly important again. The will sift through that old appraisal to see if they can hook you something. Aha! if you the appraiser didn't do A, B or C we would have never been in this mess!

Never had it happen to me but in my market and price point I would likely to have been happy to buy t from them and make them whole.

It's a system that I was good at, made a lot of money doing but by age 50 just couldn't stomach it anymore. Now I make my money or I don't but it's solely up to me.

 

[Vermonter]

I won't touch that. I'll refer back to homogeneous and heterogenous markets.

I believe a waiver is plausible in a very homogeneous market. When you get in very heterogenous markets, a waiver is useless and very risky to the overall real estate market. Who determines these waivers?

Who does it? Does God do it?

Doesn't matter. The point is the increase. Granted covid accelerated waiver use and has since dropped from the height, but today is still higher compared to 2019. No reason to think that won't continue.

 

[ZZGAMAZZ]

I asked Perplexity what Redfin and Zestimates error rate was.


"How Zestimates ‘Error Rate’ Is Determined

Zestimate error rate refers to how closely Zillow’s property valuation algorithm matches actual sale prices of homes. This is a statistical measurement, reflecting Zillow’s ongoing effort to track the accuracy of its automated estimates.
Calculation Method
• Zillow calculates the median error rate by comparing its Zestimates (the prices it predicts) with the actual sale prices of homes that have recently sold.
• For instance, if the Zestimate was $500,000 and the home sold for $510,000, the error would be (|500,000 - 510,000| / 510,000), or 1.96%. Zillow runs this calculation for a large sample of recent transactions, then finds the median value from all those error percentages.
• Median error rate means that half of the Zestimates are within this percentage of the sale price, and half are not.
Different Rates for On- and Off-Market Homes
• For on-market homes (actively listed for sale), the median error rate is 1.83–1.94%.
• For off-market homes (not currently listed), the error rate is higher: around 7.01–7.06%.
What “Error Rate” Indicates
• If the error rate is 1.9%, this means half of Zestimates for on-market homes are within 1.9% of the actual sale price, and half are more than 1.9% off. The same logic applies to the 7% error rate for off-market homes.
• Zillow also reports the percentage of homes where the Zestimate was within 5%, 10%, or 20% of the sales price. For most major markets, more than 95% of Zestimates fall within 10% of the final sale price.
Factors Affecting Error Rate
• **Accuracy depends on available ** Areas with more MLS data, recent comparable sales, and user-fed real estate information have lower error rates.
• Uniqueness of properties: Unique or rarely sold properties usually have higher error rates, especially if recent comparable sales are unavailable.
• Market activity and data quality: Frequent sales and good, up-to-date data improve algorithm accuracy.
In essence, the Zestimate error rate is a statistical comparison of predicted property values versus real sales, reported as a median percentage across thousands of transactions. It’s regularly updated and varies based on market conditions, data coverage, and property type."

Very comprehensive analysis. Thank you. [But I wonder whether the "confounding" factor is addressed, i.e., if the Zillow estimate is used by seller to establish the list price.]

 

[ZZGAMAZZ]

What's the error rate on human appraisals? We know it's not ZERO

We have to realistically look at the entire picture. Collateral value is a just a piece of the pie. 20 years ago it was a much bigger piece, but as automation improves other aspects of the lending pipeline that piece will become smaller. As a borrower becomes less of a default risk, the accuracy of the collateral value becomes less important.

The Forum previously described the potential error in the formula based upon contract prices that are revised, usually downward, to reflect appraisal values that fail to support the original offer. [Q... Are sellers n buyers notified of the result of an AVM at the same point in the transfer process as if it was an appraiser's OV?]