Join In On The Action "Register Here" To View The Forums

Already a Member Login Here

Board index Forum Index
User avatar
Adjutant
 
Posts: 3646
Joined: 17 May 2013, 3:32 pm

Post 01 Mar 2018, 10:31 am

A few years ago in Glendale a car dealership told a client that loans did not get approved for clients with names ending in "ian" and "yan". You guys and your fantasy race-neutral car buying world...
User avatar
Ambassador
 
Posts: 21062
Joined: 15 Jun 2002, 6:53 am

Post 01 Mar 2018, 10:37 am

freeman3 wrote:A few years ago in Glendale a car dealership told a client that loans did not get approved for clients with names ending in "ian" and "yan". You guys and your fantasy race-neutral car buying world...


Glendale, an area with which I'm familiar, has an Armenian community to be sure.

I'd be interested in documentation.

The one thing I've always found about every business: they're very eager to make money. If you're in an area with a decent number of Armenians and won't do business with them, that seems, well, stupid.
User avatar
Ambassador
 
Posts: 15994
Joined: 15 Apr 2004, 6:29 am

Post 01 Mar 2018, 10:56 am

bbauska wrote:Please show where the race is a component of the loan process resulting in higher loan costs based only on the race.

It would look like this:
Person A has a FICO of 725 and is black. Their costs are x.
Person B has a FICO of 725 and is white. Their costs are y.

Compare x to y on a statistically significant sample size. (alliteration intended)

This paper from Columbia Law cites studies:

http://web.law.columbia.edu/sites/defau ... ership.pdf

the most comprehensive study of markups was released last year by Vanderbilt business professor Mark A. Cohen. Professor Cohen studied more than 1.5 million General Motors Acceptance Corporations (“GMAC”) loans made between 1999 and April 2003. [9] Cohen’s study revealed that African-Americans are three times as likely as similarly situated white customers to be charged an interest rate mark-up on their loans financed by the General Motors Acceptance Corporation. According to the report, discrimination in the GMAC loans was across the board, regardless of the profession of a buyer or model of car purchased. Cohen’s report concludes that after conducting “numerous statistical tests” the higher interest rate charged to African-Americans cannot be explained by “creditworthiness or other legitimate business factors.”[10]


Richard Voith, senior vice president of Econsult, a financial-research firm in Philadelphia, has found results similar to those in the Cohen report. Voith’s study was limited to Hispanic car buyers in Chicago. The study, based on millions of Ford Motor Credit loans made between 1997 and 2001, found that buyers with Hispanic surnames paid an average of about $ 266 more per loan than did non-Hispanics with similar credit histories. [18] The study also found that the average loan mark-up for borrowers with Hispanic surnames was 3.46 percentage points. [19] The average non-Hispanic customer got a 2.78 percentage point mark-up.[20]
User avatar
Adjutant
 
Posts: 3646
Joined: 17 May 2013, 3:32 pm

Post 01 Mar 2018, 11:04 am

The case settled. Client's credit was denied even though it was pretty good, statement was made, case settled early before we got to taking a systematic look at how people with similar credit scores of different races were treated. Why would a business treat a person differently with the same credit score based on race? Could be an anecdotal bad experience. Could be that that they have a gut feeling that somehow that a person of a certain race lives in a certain area that makes them more of a credit risk or that in some way a person's race makes them a poorer risk. This is their business, it's their money, and their gut feelings are what matter--not some PC belief that everyone with the same credit score should be treated exactly the same. I can also tell you there are differences in how insurance companies settle cases based on race. If you're Japanese...they love you. Very low claims rate. Hispanic? Not so much.
User avatar
Ambassador
 
Posts: 21062
Joined: 15 Jun 2002, 6:53 am

Post 01 Mar 2018, 11:12 am

freeman3 wrote: If you're Japanese...they love you. Very low claims rate. Hispanic? Not so much.


Now, that's interesting.

I wonder if there is a "repossession" factor in all of this. Do repo-rates vary significantly by race? I really don't know.
User avatar
Ambassador
 
Posts: 21062
Joined: 15 Jun 2002, 6:53 am

Post 01 Mar 2018, 11:14 am

danivon wrote:
bbauska wrote:Please show where the race is a component of the loan process resulting in higher loan costs based only on the race.

It would look like this:
Person A has a FICO of 725 and is black. Their costs are x.
Person B has a FICO of 725 and is white. Their costs are y.

Compare x to y on a statistically significant sample size. (alliteration intended)

This paper from Columbia Law cites studies:

http://web.law.columbia.edu/sites/defau ... ership.pdf

the most comprehensive study of markups was released last year by Vanderbilt business professor Mark A. Cohen. Professor Cohen studied more than 1.5 million General Motors Acceptance Corporations (“GMAC”) loans made between 1999 and April 2003. [9] Cohen’s study revealed that African-Americans are three times as likely as similarly situated white customers to be charged an interest rate mark-up on their loans financed by the General Motors Acceptance Corporation. According to the report, discrimination in the GMAC loans was across the board, regardless of the profession of a buyer or model of car purchased. Cohen’s report concludes that after conducting “numerous statistical tests” the higher interest rate charged to African-Americans cannot be explained by “creditworthiness or other legitimate business factors.”[10]


Richard Voith, senior vice president of Econsult, a financial-research firm in Philadelphia, has found results similar to those in the Cohen report. Voith’s study was limited to Hispanic car buyers in Chicago. The study, based on millions of Ford Motor Credit loans made between 1997 and 2001, found that buyers with Hispanic surnames paid an average of about $ 266 more per loan than did non-Hispanics with similar credit histories. [18] The study also found that the average loan mark-up for borrowers with Hispanic surnames was 3.46 percentage points. [19] The average non-Hispanic customer got a 2.78 percentage point mark-up.[20]


This is compelling.

However, as it appears to be 15 years old, it does raise the question as to whether this is still the practice.
User avatar
Adjutant
 
Posts: 3646
Joined: 17 May 2013, 3:32 pm

Post 01 Mar 2018, 11:27 am

Personal injury claims rates do vary by race. It seems impossible that they would do it directly that way...maybe some other non-racial factor that gets tied into race, I don't know. Maybe claims rates by zip code? I have no idea. But there are noticeable differences.

And these are observations over 25 years and (per your observation above) I might have to keep assessing whether that's still the case. I am getting to be a old fogey with fixed ideas..
Last edited by freeman3 on 01 Mar 2018, 11:35 am, edited 1 time in total.
User avatar
Ambassador
 
Posts: 21062
Joined: 15 Jun 2002, 6:53 am

Post 01 Mar 2018, 11:30 am

freeman3 wrote:Personal injury claims rates do vary by race. It seems impossible that they would do it directly that way...maybe some other non-racial factor that gets tied into race, I don't know. Maybe claims rates by zip code? I have no idea. But there are noticeable differences.


I have some Armenian friends I'll be seeing in a week. As you know, the community in SoCal is pretty tight. I'll see if I can pick up some anecdotal info.
User avatar
Ambassador
 
Posts: 15994
Joined: 15 Apr 2004, 6:29 am

Post 01 Mar 2018, 11:31 am

Doctor Fate wrote:This is compelling.

However, as it appears to be 15 years old, it does raise the question as to whether this is still the practice.
To which I would answer, what evidence do we have that this has changed in the period since?

In another area of lending, mortgages, this University of Chicago paper - http://home.uchicago.edu/~buchak/papers ... nation.pdf - shows that while the differential reduces when there is greater competition, it still persists, falling by only about 25%.

So, even if there was greater competition in auto-finance than 15 years ago, there is reason to conclude that (absent other major changes to the way that auto-finance operates) racial discrimination would still be a feature.

What other changes can you suggest?
User avatar
Ambassador
 
Posts: 21062
Joined: 15 Jun 2002, 6:53 am

Post 01 Mar 2018, 12:12 pm

danivon wrote:
Doctor Fate wrote:This is compelling.

However, as it appears to be 15 years old, it does raise the question as to whether this is still the practice.
To which I would answer, what evidence do we have that this has changed in the period since?

In another area of lending, mortgages, this University of Chicago paper - http://home.uchicago.edu/~buchak/papers ... nation.pdf - shows that while the differential reduces when there is greater competition, it still persists, falling by only about 25%.

So, even if there was greater competition in auto-finance than 15 years ago, there is reason to conclude that (absent other major changes to the way that auto-finance operates) racial discrimination would still be a feature.

What other changes can you suggest?


Many things change over time.

Here's the thing with academic papers: I'm not going to deconstruct them. If I find ONE that rebuts yours, do I win?

The answer, of course, is "No."

If you believe your paper is objective, then fine. Good for you. If you believe a greedy, money-grubbing capitalist cares more about discriminating than making money, nothing I can say or show is going to change your mind.

That's what we call "wasting time."
User avatar
Statesman
 
Posts: 11324
Joined: 15 Aug 2000, 8:59 am

Post 01 Mar 2018, 3:37 pm

bbauska
Please show where the race is a component of the loan process resulting in higher loan costs based only on the race.


http://www.responsiblelending.org/other ... arkups.pdf

[quote]Interest rate markups also create the potential for discriminatory outcomes. In the past decade,eleven major lenders that participate in indirect financing have settled class action lawsuits alleging racial discrimination in how markups were assigned to their loans. Loan-level data showed that African-Americans and Latinos disproportionately received interest rate markups more frequently and to a greater degree than their similarly-situated white counterparts.6
For each lender, the lawsuits resulted in some degree of
self-imposed interest rate markup cap to remedy the discrimination.24
It was determined that markup caps between 2 and
3 percentage points were enough to dampen the appearance
of racial bias. However, while rate markup caps may have
addressed disparate impact for extreme cases, it still left the
opportunity for wide discretion in assigning rate markups for all consumers. Note that the average
rate markup in 2009 stood at 2.47 percentage points, slightly below the typical markup cap. Even
with caps at that level, markups were able to garner $25.8 billion from consumers.
Still, even with a typical 2.5% markup cap on 60-month loans, the average new car deal will
create over $1,700 in additional interest payments—a figure which increases for consumers with
poor credit. For a family on a tight budget, extending the debt burden an extra $1,700 could be a
determining factor for the sustainability of that loan.
User avatar
Ambassador
 
Posts: 15994
Joined: 15 Apr 2004, 6:29 am

Post 01 Mar 2018, 10:00 pm

Doctor Fate wrote:
danivon wrote:
Doctor Fate wrote:This is compelling.

However, as it appears to be 15 years old, it does raise the question as to whether this is still the practice.
To which I would answer, what evidence do we have that this has changed in the period since?

In another area of lending, mortgages, this University of Chicago paper - http://home.uchicago.edu/~buchak/papers ... nation.pdf - shows that while the differential reduces when there is greater competition, it still persists, falling by only about 25%.

So, even if there was greater competition in auto-finance than 15 years ago, there is reason to conclude that (absent other major changes to the way that auto-finance operates) racial discrimination would still be a feature.

What other changes can you suggest?


Many things change over time.

Here's the thing with academic papers: I'm not going to deconstruct them. If I find ONE that rebuts yours, do I win?

The answer, of course, is "No."

If you believe your paper is objective, then fine. Good for you. If you believe a greedy, money-grubbing capitalist cares more about discriminating than making money, nothing I can say or show is going to change your mind.

That's what we call "wasting time."
something tells me you didn't waste much time to read the latest one, which is saying that capitalism - or rather more competition - has reduced discrimination, just not by a lot.

It is not about capitalism / socialism, but human prejudice. You can deny it exists or has an effect if you like.
User avatar
Ambassador
 
Posts: 4961
Joined: 08 Jun 2000, 10:26 am

Post 02 Mar 2018, 7:59 am

Slightly raising the level of the conversation, here's an interesting op ed I saw earlier this week.

https://www.wsj.com/articles/trump-can- ... ge=1&pos=1

I don't fully agree with this -- we do have this legacy of slavery and segregation over here -- , but I appreciate the level of thought. Here's an excerpt for those who cannot access
Identity politics—the artificial segmentation of Americans into antagonistic groups organized along often imagined ethnic, racial and sexual categories—is tearing America apart. President Trump can do something about it.
Government played a key role in creating these identities. The establishment of a new official taxonomy of Americans started roughly in 1966, when the Equal Employment Opportunity Commission began asking companies with more than 100 employees to collect information through the EEO-1 form on “Negro, American Indian, Oriental and Spanish-surnamed” employees. What began as an effort to track how policies affected people thought to be disadvantaged easily but tragically slid into government-sanctioned promotion of victimhood and racial preferences. The goal of the Civil Rights Act of 1964, to prohibit racial discrimination, was turned on its head....

Starting in 1980 the census began tabulating all residents into groups that correspond to a vague and unscientific color code: white, black, brown, yellow and red.
If you don’t think of yourself that way, the government will do it for you. There’s a box on the census for “some other race,” but the bureau explains: “When Census 2010 data were edited to produce the estimates base, respondents who selected the Some Other Race category alone were assigned to one of the OMB mandated categories.”

For people who tick multiple boxes—permissible since 2000—OMB has instructed the Census Bureau to “allocate” responses that “combine one minority race and white” to “the minority race.” As Mr. Hollinger puts it, “thus the federal government quietly reinserted into the tabulation of the census the principle of hypodescent”—the technical term for the old segregationist one-drop rule—“that the opportunity to make ‘more than one’ was publicly said to repudiate.”
Until the Trump administration stopped it last month, the census was preparing to add in 2020 yet another vast pan-ethnic grouping—“Middle East or North Africa”—for residents with ancestry anywhere between Morocco and Iran. That would have made a minority of everyone from Rep. Darrell Issa to the late Steve Jobs.
Every level of American government now follows this scheme, as do most other major institutions. Public schools promote the invidious idea that all subjects, even math, should be taught differently to children depending on where administrators place them on the pentagon. Universities have become cultural battle zones where students search for victim status rather than truth. And if you work for a large organization, there’s someone in your human-resources department whose job is to put you into one of the government-created silos.
User avatar
Ambassador
 
Posts: 15994
Joined: 15 Apr 2004, 6:29 am

Post 02 Mar 2018, 9:58 am

Ray Jay wrote:Slightly raising the level of the conversation, here's an interesting op ed I saw earlier this week.

https://www.wsj.com/articles/trump-can- ... ge=1&pos=1

I don't fully agree with this -- we do have this legacy of slavery and segregation over here -- , but I appreciate the level of thought. Here's an excerpt for those who cannot access
Identity politics—the artificial segmentation of Americans into antagonistic groups organized along often imagined ethnic, racial and sexual categories—is tearing America apart. President Trump can do something about it.
Government played a key role in creating these identities. The establishment of a new official taxonomy of Americans started roughly in 1966, when the Equal Employment Opportunity Commission began asking companies with more than 100 employees to collect information through the EEO-1 form on “Negro, American Indian, Oriental and Spanish-surnamed” employees. What began as an effort to track how policies affected people thought to be disadvantaged easily but tragically slid into government-sanctioned promotion of victimhood and racial preferences. The goal of the Civil Rights Act of 1964, to prohibit racial discrimination, was turned on its head....

Starting in 1980 the census began tabulating all residents into groups that correspond to a vague and unscientific color code: white, black, brown, yellow and red.
If you don’t think of yourself that way, the government will do it for you. There’s a box on the census for “some other race,” but the bureau explains: “When Census 2010 data were edited to produce the estimates base, respondents who selected the Some Other Race category alone were assigned to one of the OMB mandated categories.”

For people who tick multiple boxes—permissible since 2000—OMB has instructed the Census Bureau to “allocate” responses that “combine one minority race and white” to “the minority race.” As Mr. Hollinger puts it, “thus the federal government quietly reinserted into the tabulation of the census the principle of hypodescent”—the technical term for the old segregationist one-drop rule—“that the opportunity to make ‘more than one’ was publicly said to repudiate.”
Until the Trump administration stopped it last month, the census was preparing to add in 2020 yet another vast pan-ethnic grouping—“Middle East or North Africa”—for residents with ancestry anywhere between Morocco and Iran. That would have made a minority of everyone from Rep. Darrell Issa to the late Steve Jobs.
Every level of American government now follows this scheme, as do most other major institutions. Public schools promote the invidious idea that all subjects, even math, should be taught differently to children depending on where administrators place them on the pentagon. Universities have become cultural battle zones where students search for victim status rather than truth. And if you work for a large organization, there’s someone in your human-resources department whose job is to put you into one of the government-created silos.


To me, I think they have it a bit back to front. The collection of statistics based on these characteristics means we have data to show how things are. Before 1966 there clearly was discrimination in the US based on race, and it isn't going to just disappear through wishful thinking or assuming markets will be a great equaliser. Human tribalism is ingrained and pervasive.

Sometimes the efforts to identify and mitigate such prejudices go to far or are counter-productive. But corporations in the free market follow these practices for another reason - if there is unfair discrimination, it means that people who are less able or qualified will push out others who are better than they are. That is not meritocracy.

I'm currently in India. If you think the US has a class / discrimination issue, or the UK, then India has it much worse. It is really hard to counter the prejudices that history and culture imprint on to society. Pretending not to "see" the differences is just a way of ignoring the reality. Recognising them has its risks, but does mean we can measure the effects.
User avatar
Adjutant
 
Posts: 3646
Joined: 17 May 2013, 3:32 pm

Post 02 Mar 2018, 12:02 pm

That article is kind of like the Brahmin telling Untouchable to stop trying to divide with complaining about how things are...

We just had an election won a president catering to white prejudices, so now the issue is minirities making all the fuss? Another right-wing fantasy that the government has created racism/prejudice with its policies. There seems to be a tendency in the US to try and say we live in a post-racial society, no way anyone's views could be influenced by prejudice, and the real problem is minorities complaining about racism holding them back , with the exception of a very few.

Better to accept the reality that humans are inherently tribal (as Danivon notes) and deal with reality instead of a fantasy. It's a lot better than it used to be by prejudice did not end just because it is not overtly considered acceptable (which is a good thing by the way). You don't get to MLK' s Dream World by pretending you have already reached it...