Mary Cummins, Real Estate Appraiser, Animal Advocates, Los Angeles, California

Mary Cummins, Real Estate Appraiser, Animal Advocates, Los Angeles, California
Mary Cummins, Real Estate Appraiser, Animal Advocates, Los Angeles, California

Friday, June 18, 2021

False statement by President Joe Biden on home valuation and race - by Mary Cummins

Home Valuation is not related to race, President Joe Biden

UPDATE: 11/08/2021 Farther below in this article it's proven that Andre Perry's "paper" about the value of white and black owned homes is based on major statistical errors. Perry compared the values of a very small sample of two intentionally extreme groups which Perry himself chose. Perry also did not control for the other variables especially location which we all know is the main contributor of value. There is no way to control for location. Perry also used Zillow data. Zillow stated they had a median error rate of 6.9% today and previously it was over 14%. Within the last week it was shown that Zillow's data is totally unreliable today. Zillow used their own data to buy and sell homes for profit or at least that was the plan. They ended up selling them for less than what they bought them. This proves Zillow's data which Andre Perry relied upon is deeply flawed. That makes his paper even more meaningless and worthless.

ORIGINAL: I'm glad that on June 17, 2021 President Joe Biden made Juneteenth a Federal holiday. It's been a holiday in most states including California for years. Unfortunately during that presentation President Joe Biden made a totally false statement about home valuation and race when he stated the following (sic),

" belongs to an aggressive effort to combat racial discrimination in housing. Finally address the cruel fact home owned to this day by a black American family is usually appraised at a low rate for a similar home owned by a white family in a similar area."

Someone repeated a misleading statement related to the misleading Brookings report to Joe Biden. A paper was written by Andre Perry who is promoting and selling a book on the subject. It's not peer reviewed research or a legitimate paper released by a University. The "report" misstates an alleged relationship between home valuation and race. The report only showed that all people of color including Latinos, Native Americans, Alaskans ... tend to buy and own less expensive homes in less expensive areas. The home values were determined from Zillow listing price robots who didn't see the homeowners or the homes. The values were list prices which are agreed upon by the homeowners and their own real estate agent and the homeowners estimate of value of their own homes. Real estate appraisers had nothing to do with the values. Race and color were not a factor in the calculation of home values. Perry doesn't say if these are active or off market listing prices. There is error in both prices but off market list prices has a much, much larger error making them meaningless. And again Zillow even states their Zestimates should never be considered an "appraisal" and are "no substitute for an appraisal." Perry also doesn't bother to mention that anyone can edit Zillow's data. I could go in there and say a home actually has ten bedrooms and not two. 

UPDATE: Yesterday July 15, 2021 Andre Perry stated that "media articles" appear to indicate that real estate appraisers low ball the value of homes owned by black people. He didn't say his research proved this. He inferred that "white staging" homes for real estate inspections seemed to show this. 

The mathematical formula used to determine value is the most common asset valuation formula used worldwide for thousands of years. That formula compares one item to another similar item. It's also called matched pairs analysis. Homes of the same size, bed/bath count, condition, amenities were compared to other similar homes which recently sold in the SAME EXACT AREA. Homes in certain geographical areas are worth less than others in more desirable geographical areas with different characteristics. The mathematical calculation has absolutely nothing to do with race or color. In this instance they used Zillow listing prices per square foot. All appraisers know the "a" in "Zillow" stands for "accuracy." Zillow "zestimates" are not accurate at all. They also used ACS data which is the value the actual homeowner thinks their own home is worth. 

A home in South Los Angeles is worth less than a home of the same size, condition and features in Beverly Hills. The main difference is the value of the land and not the house. Here in California the land makes up over 70% of the total value. A vacant lot in South Los Angeles is worth less than a vacant lot of the same size in Beverly Hills. Everyone knows home value is based on three main factors, i.e. location, location, location.

There is a correlation between all people of color, new immigrants, people who don't speak English, people with less education... and level of income and net financial worth. ALL OF THESE PEOPLE tend to have less income and a resulting lower net worth. Because they have less income and less financial worth they tend to buy less expensive homes which they can afford in less expensive areas. Andre Perry committed the main fatal statistical sin confusing correlation with causation. That makes his "report" totally meaningless and useless. There are even more grave errors in his analysis which will be examined below. 

In Andre Perry's paper he used the separate inaccurate Zillow home valuation listing prices per square foot and self reported home values from the American Community Survey. The ACS is sent to a sample of 3.5 million people in the nation in the mail. A response to their written questionnaire is mandatory but most people throw it away assuming it's junk or scam mail. All of the information is self reported including when the home was built, number of bedrooms, rooms... Here's a sample form. Based on my experience most homeowners don't know all of this information off the top of their head. They generally don't know what their home is worth today. I would not trust the results of the ACS to be accurate about real estate values or home characteristics. Perry refers to the ACS results as the Census results. The ACS is not part of the Census. 

Perry compared those values to the racial makeup of the areas at large. Again, Zillow valuations are not accurate and listing prices don't equal home valuations. Homes don't always sell for listing price. Perry used price only per square foot which is never used in appraising because it's not accurate for home valuation. We weight comparable home sales based on similarities. Self reported home values are also very inaccurate. 

UPDATE: 07/30/2021 July 29, 2021 Andre Perry was on a panel talking about this issue. He stated, 

"Appraisers have the freedom to discriminate. In the price comparison approach they compare homes in area to another in the same area. If you only compare homes in a discriminated area to the same area, discrimination continues. There are multiple ways to arrive at value. We look across metropolitan areas instead of just the same neighborhood to get our values."

This proves that Perry knowingly compared different areas to each other. Perry only looked at 113/384 metropolitan areas which he chose based on his % black limit. Of course different areas would have different values. Everyone knows the main factor in value is location, location, location. Perry didn't control for location because it's impossible. If all the areas were truly similar, they would have had the same values. Perry basically admitted his research was flawed. UPDATE

Perry allegedly compared the values of alleged "similar" homes in "primarily black" areas to the same in "primarily white" areas. He claims the homes and neighborhoods had the exact same characteristics except for color of homeowner. Perry alone made that determination based on no real science. He ended up comparing 10% of all homes which he stated are in "primarily black" areas to an even smaller percentage of homes which he stated are in "primarily white" areas. The "primarily white" areas have 0-1% black people. He doesn't even bother to state the racial makeup of the 99-100%. Are they white Latinos, Asians, light Native Americans...? We don't know. Perry then took that data and extrapolated, stated it represented the entire USA. Let's look closer at the groups he compared. 

Perry stated an area was "primarily black" if 50% or more of the population identify as "black." He compared that to an area with less than 1% black residents. Perry actually stated zero black residents in the report. How many places have zero black people? I find it hard to believe they are any unless they are very tiny towns in the middle of nowhere. Perry clearly intentionally chose two very extreme comparison groups in order to create the most extreme statistics possible to fit his preconceived "results." That would be like comparing only very tall people to very short people to see if there's a difference in their lives. Of course there will be a difference. This is clearly an example of "garbage in, garbage out" i.e. GIGO statistics. Perry actually stated the areas that are "primarily black" are "more segregated." If you are looking at an area that is 50% plus black of course it's more segregated. 

Below is Table 2 from the report for comment. It shows the 10 metropolitan areas with the largest and smallest differences in the value of homes. Notice how widely the home "values" range in only the 99%+ white areas. What is causing the wide range of values in the white only area? It's not black people. The values range from $61,000 to $783,000. The same is evidence in the mainly black areas with a range $59,000 to $403,000. Perry stated he controlled for all factors except black or white skin color yet the "same homes with same amenities" are worth very different values in different areas. Could it be that there are other factors involved besides skin color? Maybe it's the location.  

Here is Table 6, the 10 metropolitan areas with the most and least devaluation of homes. The ones with the most alleged devaluation due to black skin color are depressed areas. Some neighborhoods in Detroit, Michigan? The city is auctioning off vacant homes for $1,000 just so they don't have to demolish them. I would bet that people who could afford to leave those specific neighborhoods left. Those that couldn't are still there. 

Per the chart, "Devaluation measure estimates median list price per sq foot after adjusting for home and neighborhood quality." We still don't know exactly how the home and neighborhood qualities were derived. Perry doesn't show the math. 
Perry has now added an "anti black sentiment index" from Stephens-Davidowitz. The "paper" is a New York Times op-ed opinion piece (pay wall). It's not peer reviewed research. Anyone can write an op-ed piece. The paper is based on "the percent of Google search queries that include racially charged language." The author Seth Stephens-Davidowitz then compared the search results to Obama's votes. His conclusion was anti black sentiment cost Obama 4% in votes in 2008 and 2012. Obama won. Did Obama win in spite of this or maybe he won because people wanted a black President and he was extremely well qualified? Perry is not only cherry picking his groups for comparisons but also his reference material. 

How much of the US fell into only those two very specific small categories of 50%+ black verses 0-1% white? Perry states that 10% of the US falls into the category of "primarily black." An analysis of only 10% of the country does not represent the entire country, see Perry's map below. Per the map this 10% appears to only be in the East and a little in the South. This is still only the percentage of the population living in the specific area and not the percentage of all black homeowners in that area or the entire country. Most homes in less wealthy areas which tend to have more people of color are actually owned by white investors or large public and private corporations. Perry then compared those values to areas with only 1% black people. Again, that would be a very small group for comparison. Of course it will show differences when you intentionally choose two very small and  extremely different groups for comparison. This is very misleading statistical analysis. It's as bad as any Fox News "statistics" or poll.


What about the white, red, yellow person who lives next door to the black person in an identical home in the "primarily black" area? What about the other <50% in the area who are not black? From Perry's report "Though most residents are Black (because he chose 51%+ black areas)(14.4 million non-Hispanic Blacks) by definition, approximately 5 million non-Black Americans live in majority Black neighborhoods." The 5 million non-Black person's home is worth exactly the same as the black person's if it's the same size and condition. The white person's home value is not "caused" by being black because they're not black. They're white or other. The home was not "devalued" or valued by black skin color of the homeowner.

If you actually read the 28 page report, Perry out of one side of his mouth admits "Majority
black neighborhoods do exhibit features associated with lower property values, including higher crime rates, longer commute times, and less access to high-scoring schools and well-rated restaurants." He then states that only explains half of the alleged devaluation of homes. Those are just the characteristics of poorer areas in general. I live in a high crime area in Los Angeles, California with poor schools. It's 80% Latino. The homes are valued less here but it's not because of black homeowners. 

Perry goes on to say there is a "significant correlation between the devaluation of homes in black neighborhoods and upward mobility of black children in metropolitan areas with majority black neighborhoods." The real correlation is any and all children living in poorer areas will find upward mobility very difficult. It has to do with more than just the value of the homes in the area. It has to do with the people being poorer. Homes in poor areas are just worth less. One should also note that most people living in poorer areas don't even own a home. They're renters in poor areas. That's the main reason it's difficult to move up the ladder. 

Here is another example of poor data used. Perry used the EPA walkability index to determine if an area had amenities such as "access to shopping, jobs and schools." This is how he compared areas with the "same amenities." The EPA index only shows the "suitability for walking as a means of travel." It doesn't show if there are lots of great local shops, cafes, great high paying jobs or great schools. It just shows if you can walk to the bus stop to take the bus to your job, school or store. Two homes with the same walk score can be worth very different amounts. Walk score does not equal value. In poor areas there are few grocery stores, retail chain stores or shopping malls. The reason is high crime, theft, higher insurance rates, higher expenses and less projected net profits which keep major chain stores away. Most people in poorer areas don't have cars. They can't just drive to the next city to go shopping or escape from hurricane Katrina. More info on the walkability analysis here

I remember the Los Angeles Rodney King riots. There were a few chain stores there like Payless Shoes. They were looted along with many other stores then burned to the ground. They didn't reopen. The area is still a grocery store desert. There are only liquor stores and a few non-chain fast food restaurants in the better pockets of the area near the freeway on ramps. I appraise in these areas. This is a huge issue in Los Angeles. 

Here is yet another example of Perry selectively using statistics. He states there is a correlation between a high number of violent crimes, black areas and reduced valuations. He also stated there is a correlation between high number of property crimes in high valuation non black areas. For that reason he only included violent crimes in his analysis of black areas. Then he didn't really include "crime" in his analysis. His "conclusion" is crime doesn't explain the devaluation of homes in primarily black areas. That's actually the opposite of what he said in another section of his report. A high violent crime area does explain a lower home valuation area. That's why all the home sites include the crime statistics. I would not live in a high crime area if I could afford to live elsewhere. 

On top of the useless statistics Perry then used the results of his limited analysis to extrapolate devaluation to 113 other metropolitan areas if they had at least one majority black area. He took very bad math from a few small extreme areas to state that many other areas he did not analyze have his same preconceived results. Andre Perry's entire paper should be dismissed because of the intentionally misleading statistics used to fit Perry's personal agenda to sell his book.

One more ridiculous comparison. Someone with less money will buy a less expensive car which they can afford. People of color tend to have less money as do some others who are not of color. Kelly Blue Book uses a computer program to calculate the value of your car. You type in make, model, year, mileage, condition and it tells you what it's worth. They will tell you the value of a Rolls Royce or a Ford Escort. They don't know the color of the person typing in the information. Is Kelly Blue Book "intentionally devaluing" cars owned by black people when they tell the owner of a 1990 Ford Escort that their car is worth less than a 2020 Rolls Royce? No. Real estate robot appraisals are the same. Actual appraisers follow the same mathematical formula as the robot appraisals. Land in Beverly Hills is the Rolls Royce of land and worth more than land in South Los Angeles. It's perceivable that all homes in less desirable areas could be worth $48,000 less than homes in more desirable areas per Perry's report.

A last thing to consider is that people who own homes in less expensive areas paid less for the homes originally compared to homes in more expensive areas. They bought a home they could afford. When they go to sell that home they will sell it for the same price as a similar home in the area which is market value. If they bought a home in South LA, it will sell for less than a home in Beverly Hills. Did someone "devalue" or "steal" the difference in price between their home and a home in Beverly Hills? No homes were "devalued" by $48,000 by robot appraisals, appraisers or anyone. 

Andre Perry has done very poor biased research. The report is therefore meaningless. Perry is guilty of using misleading statistics and data to prove his preconceived bias that black homes are devalued compared to white homes only because they are owned by black people. Perry used small extremely biased sample sizes. He organized the data to omit any findings that contradict the result he tried to prove. He manipulated the results to influence perception through misleading graphs and visuals. He used faulty correlations and causations to create false statistics. He committed every single statistical sin in the book. The Brookings Institute should retract his report. 

At the June 20, 2019 House Finance Committee meeting Andre Perry spoke about the alleged devaluation of black owned homes. There were real estate appraisal industry experts at the meeting. I watched the meeting and took notes, see above link. When the speakers were asked by a Committee member if they thought appraisers discriminate against black people in arriving at their home's value Perry was the only person who raised his hand in a room of real estate valuation experts. Perry was the only non expert in the room. 

Perry clearly does not understand real estate valuation. He is trying to use his report full of bad statistics to state that racist appraisers intentionally devalue the homes of black people by $48,000. He stated that in the Committee meeting. He stated the purpose of the report was to give "Black homeowners and policymakers a target price for redress." Oddly enough in the meeting Congresswoman Rashida Tlaib stated that the government should just add $48,000 to the value of all black people's homes. One doesn't have to be a real estate appraiser to realize how ridiculous that would be. What about the white person who has an identical home next door that's worth the same? Does he also get $48,000?

If Joe Biden or anyone wants to tackle the real issue of the correlation between people of color and income, net worth, they need to deal with those issues.The issue is most people with less income and money are people of color including many other colors and races other than black. Those issues have nothing to do with robot home appraisal analysis or home appraisers. They would need to make sure everyone makes the same amount of money regardless of any other factor. Until that happens all people with less money will buy and own less expensive homes in less expensive areas. People with no money won't buy or own any home at all. Most people in the US don't even own a home. Some don't even rent a house but maybe rent a tiny apartment or else they're homeless. Don't forget about those people.

The point of this article was to analyze the Brookings report and prove Joe Biden's statement incorrect. No one should ever rely on this misleading report. The media and others today have been stating that white real estate appraisers are intentionally appraising homes owned by black people lower than market value because all appraisers are white racists. No appraiser appraised any home in the report. The alleged valuations came from misleading computer analyzed Zillow listing prices per square foot and the ACS which uses self reported home values from the homeowner. Any low balling would therefore be caused by Zillow or the homeowner. 

These attacks on appraisers must stop. While racism certainly exists and there are racist appraisers not all appraisers are racists who devalue homes owned by black people. Most of us are hard working appraisers who always provide the best appraisal possible which complies with all the laws for everyone and anyone. 

* I'm a Democrat and voted for Joe Biden. I just disagree with Biden on this specific issue. I believe he was given an incorrect talking point. I'm also a Latino real estate appraiser. I appraise in primarily lower income areas which are mainly POC. I live in a lower income, high crime, very mixed area in South Los Angeles. 

** I went to the University of Southern California. I took statistics and a graduate class in research as an undergraduate. My research was selected by two Psychology faculty members to present at the Western Psychology Conference for Undergraduate Research at the University of Santa Clara in 1984.  The subject of my research was Classical Conditioning as a Factor in Response Bias. I used chi-square tests on my data. My Faculty Sponsor was Herm Turk. I do know a thing or two about statistics and research. 



Perry, Andre M. (2020). Know Your Price: Valuing Black Lives and Property in America’s Black
Cities. Washington, D.C.: Brookings Institution Press

Sweeney, Maureen (2019). Is Race Baked into Big Data?  My Letter to Housing SubCommittee.

Mary Cummins of Animal Advocates is a wildlife rehabilitator licensed by the California Department of Fish and Game and the USDA. Mary Cummins is also a licensed real estate appraiser in Los Angeles, California.

Google+ Mary Cummins, Mary K. Cummins, Mary Katherine Cummins, Mary Cummins-Cobb, Mary, Cummins, Cobb, wildlife, wild, animal, rescue, wildlife rehabilitation, wildlife rehabilitator, fish, game, los angeles, california, united states, squirrel, raccoon, fox, skunk, opossum, coyote, bobcat, manual, instructor, speaker, humane, nuisance, control, pest, trap, exclude, deter, green, non-profit, nonprofit, non, profit, ill, injured, orphaned, exhibit, exhibitor, usda, united states department of agriculture, hsus, humane society, peta, ndart, humane academy, humane officer, animal legal defense fund, animal cruelty, investigation, peace officer, animal, cruelty, abuse, neglect #marycummins #animaladvocates #losangeles #california #wildlife #wildliferehabilitation #wildliferehabilitator #realestate #realestateappraiser #realestateappraisal #lawsuit


  1. Thank you Mary for sharing this and speaking up. I see that admin turned off comments in our female appraiser group on FB. Sad, as this kind of vitreol affects us all.

    1. People have been using the Brookings report as if it were a peer reviewed research paper. The public needs to know it's just the work of one person who used very questionable statistics to come to incorrect conclusions.