Another Appraiser Bias Academic Paper

[Fernando]

sorry - couldn't hear you. My ears got in the way.

Have to admit that I find you not as MAGA as the worst of them.

 

[alebrewer]

Have to admit that I find you not as MAGA as the worst of them.

I'm disappointed.

 

[Fernando]

I'm disappointed.

Your one of few MAGAs with sense of humor that can make me laugh.

 

[Mark K]

"Paper" was obviously funded by the Soros cartel or by the NAACP, or both.

 

[djd09]

and they would like to thank the appraisal foundation, fannie, freddie, revaa, crn, and the lying media

 

[Fernando]

and they would like to thank the appraisal foundation, fannie, freddie, revaa, crn, and the lying media

Yeah, Fox the lying media.

 

[Dublin ohio]

Yeah, Fox the lying media.

WTF does FOX have to do with it?

 

[Fernando]

Someone mentioned lying media. Reflex response.

 

[Elliott]

Brent, et al are probably working on their Phd's. Here are the problems I noticed right out of the gate:

1. So how do they know appraisers are white, black, or green?

3.3. Identifying Appraiser Race
Although the property owner’s race is disclosed in the New Century data, we do not directly observe the race
or ethnicity of the appraiser. However, we can infer the appraiser’s race and ethnicity using the Bayesian
Improved First Name Surname (BIFS) classifier approach, which is similar in spirit to the methodology used
by regulators to determine consumer race and ethnicity (Consumer Financial Protection Bureau, 2014). As
noted by Ambrose, Conklin, and Lopez (2021), Bayesian-based classification methods have also been used
to infer an individual’s race or ethnicity in various court cases (e.g., Guardians ***’n of N.Y.C. Police Dep’t
v. Civil Serv. Comm’n, 1977).

2. And this is why they should be laughed out of the esteemed "Journal." In the 'work' data they show their graphs, and probably they don't know their R^2's are shown. Their data correlation has R^2s of 0.02 to 0.18. [The GSE's studies on appraiser race discrimination have these similar R^2 0.02 and 0.04 correlations that 'prove' nothing].






3. Well, how significant is their correlation? This is what AI's Perplexity says about correlation:



....................
My best friend has a Phd and my daughter was a candidate, so I know a little about the process. I'm sure these 'finance academics' think they found the Holy Grail (just like the gal from NM), that will move them up the Phd ladder, but I'd be surprised this survives peer review, but who knows what passes as academic intelligence now a days in universities. When I was in high school, when a buddy tried to push something that was obviously false, one of our favorite retorts was, "Keep f-ing that chicken."

 

[shrubberyvaluation]

Well, I guess it isn't over. Published in Review of Finance yesterday.

Abstract:
Do Appraiser and Borrower Race Affect Mortgage Collateral Valuation?

We examine racial bias in property appraisals using a national sample of refinanced mortgages from 2000 to 2007. Our data allow us to observe the race of the appraiser and homeowner in a setting where the appraiser’s valuation conveys critical information to the lender. After conditioning on a rich set of controls and individual appraiser fixed effects, we observe lower appraised values (relative to benchmark valuation estimates) for Black owned homes. Hispanic and Asian owned homes are also under appraised, but to a lesser extent. We find no evidence that minority valuation discounts lessen when the homeowner and appraiser share the same race, suggesting implicit bias is not solely attributable to White appraisers. We also show that minority valuation discounts are not driven by a small subset of appraisers; rather, they are widespread.

https://academic.oup.com/rof/article/doi/10.1093/rof/rfaf046/8316104

A biiiiiiiig flaw I see from the get-go; Basing a current paper on any data pre-2008 financial crises screams "I know nuthin' about real estate."

So they are saying diversity will change nothing.

I have a feeling their controls and benchmark valuation estimates are not good enough.

So all races supposedly value minority homes lower even if in the same race group, no thought that maybe their value estimates are the issue.