= Subscriber content; or subscribe now to access all American Banker content.

CFPB Staff Evaluations Show Sharp Racial Disparities

WASHINGTON Since the Consumer Financial Protection Bureau burst onto the financial stage a few years ago, it has made a steady stream of controversial moves.

None are more despised by bankers than the agency's use of statistical differences in the loan terms offered to different ethnic groups to sue creditors for unintentional racial bias.

In an ironic twist, it turns out that the CFPB's own managers have shown distinctly different patterns in how they rate employees of different races, according to confidential agency data obtained by American Banker.

Specifically, CFPB managers show a pattern of ranking white employees distinctly better than minorities in performance reviews used to grant raises and issue bonuses. Overall, whites were twice as likely in 2013 to receive the agency's top grade than were African-American or Hispanic employees, the data shows.

What's more, those disparities are only one of many serious personnel problems plaguing the CFPB. Inside the agency, morale is poor and management has been accused in several cases of favoring Caucasian men and of creating a hostile work environment. That's according to interviews with a dozen current and former staffers across six departments, all of whom requested anonymity over concerns about retaliation.

Employees have filed 115 official grievances with the National Treasury Employees Union (NTEU) since last August, the CFPB says. If unofficial complaints that haven't yet worked their way through the system are included, the number exceeds 200, according to information obtained by American Banker.

Most of the complaints pertain to allegations of unequal pay and raise questions about the recent performance reviews.

The NTEU "has identified disparities in performance ratings that appear to negatively impact non-whites and females (to some degree)," the local union said in a January email to members, which was obtained by American Banker.

CFPB spokesman Sam Gilford says the agency is still analyzing the performance evaluation data and indicated that it's preliminary and could change "depending on the outcome of pending reviews and appeals."

"The CFPB is committed to fairness and equity in the workplace as well as the marketplace," Gilford said. "Just as we often remind lenders that strong compliance management systems are critical to ensure compliance with consumer protection laws, the bureau has taken a compliance management approach in monitoring and evaluating its own performance rating process."

Stark Differences

The most concrete data available on the CFPB's employee evaluations relates to 2013. The agency rated its more than 1,100 staffers on a scale of 1 to 5 and grants greater benefits, including raises and bonuses, to those who receive higher scores.

White employees scored markedly higher than minorities. Overall, 74.6% of whites received ratings of 4 or 5, versus 65.5% of Asians, 65.2% of Hispanics and 57.6% of African-Americans, according to an internal CFPB report obtained by American Banker.

The discrepancies were even greater at the ratings range's extremes. At the top, one-fifth of white employees, or 20.7%, received a 5 and were dubbed "role models" compared with 10.5% of African-Americans and 9.1% of Hispanics.

In contrast, a relatively high proportion of minority employees received 3 ratings the lowest grade given out in large numbers. In total, a rating of 3 was given to 42.4% of African-Americans, 34.5% of Asians, 34.8% of Hispanics and 24.4% of Caucasians.

Breakdowns of employees who received 1 or 2 ratings were unavailable because they represented small slices of the agency's population likely a few dozen staff. Overall, the evaluations covered 778 white employees, 191 African-Americans, 110 Asians and 68 Hispanics. The data did not cover small groups identified as "two or more races" and "other race."

The statistics themselves do not prove that CFPB managers are discriminating intentionally against minority employees. Yet they do indicate that racial disparities can be just as easily identified within the CFPB's ranks as among the lenders the bureau regulates. The agency has pressed such claims under a controversial legal theory known as disparate impact the assertion that different results for different racial groups are themselves a type of wrongful bias, even if they are unintentional.

"The level of hypocrisy at this agency is shocking," said a current agency employee who spoke on condition of anonymity. "If it was a lender and had similar statistics, it would be written up, immediately referred to the Justice Department, sued and publicly shamed."

Added another agency employee, "If we're telling banks to own it [statistical evidence of racial bias], then why don't we own things too?"

The CFPB is likely to face pressure to answer that question as the figures circulate.

"In Washington, the hypocrisy side of this is the easiest storyline to write and the foes of the CFPB will use this as latest round of ammunition against the agency," says Edward Mills, a financial policy analyst at FBR Capital Markets. "This will certainly be something even the Democrats will have concerns about, because you don't want to lose support over personnel issues."

For its part, the bureau will likely move quickly to try to fix its personnel problems and move on, he added.

The CFPB's Gilford said the agency voluntarily collected performance evaluation data and hired an outside firm to assess it further.


(18) Comments



Comments (18)
I retired from the federal government twenty years ago for this very reason and it has gotten worst. IF we do not open up the discussion on Cultural Marxism i.e. political correctness it will destroy us!!!
I arrived at a military installation in 1992, as a fire chief and evaluated my staff of assistant chiefs. My assistant chief for fire prevention was a good man however, he was very, very weak and did not have the necessary depth of knowledge, skills and abilities to serve as an effective leader. He was a minority thus he was "protected" by the system and that was twenty years ago! I worked with some EXCELLENT minorities in the Air Force and Army but I never experienced how hard it would be to downgrade this man. Keep in mind, he was a great guy but he simply did not know his job. Go to YouTube and watch CULTURAL MARXISM: The Corruption of America.
Posted by TopAssistant | Monday, May 26 2014 at 9:23AM ET
haven't heard back from the editor yet
Posted by jim_wells | Tuesday, April 08 2014 at 10:17PM ET
@jim... whats the link? or is it not published yet?
Posted by Mike_D | Tuesday, April 08 2014 at 6:48PM ET
@ Mike_D Funny you should pose this question, as I just submitted an article tentatively titled, "CFPB: Cop on the Beat or Bully on the Block."
Posted by jim_wells | Tuesday, April 08 2014 at 5:24PM ET
@Jim... i know im a month late on the discussion, the larger issue is even if there was intent to discriminate what can congress do about it, besides reporting it in a article, what are you going to do about it? they, the CFPB, can do whatever the hell they want without any repercussions. If they want to discriminate, they are legally allow to. Remember this Gestapo machine has ZERO Oversight. The house oversight committee invited reps from the CFPB (i like that word invited, because they cant do anything else but to invite them) to go over the discrimination "issue". So what happen??? they essentially gave the house oversight committee the middle finger to them and didnt even show up for the hearing, because they can.

This "agency" drummed up by doddy and uncle frank and spearheaded by elizabeth warren are certainly not helping the average joe? they have created more confusion, more paper work, higher cost and thus higher rates in the mortgage world. I can be here all day going on on their uselessness, but i digress. And yes they are total hypocrites, "do as i say, not as i do."
Posted by Mike_D | Tuesday, April 08 2014 at 5:12PM ET
Rather than being a criticism of this article, I believe Mr. Lindees's comments captured the irony that the reporter endeavored to identify. Namely that the CFPB now appeared to be guilty of the same type of discriminatory behavior for which it was hounding financial institutions.
Posted by jim_wells | Monday, March 10 2014 at 1:52PM ET
While the article and many of the comments are interesting, all seem to miss the key point. Just because you can create a statistical disparity, does not mean there is a problem. This is the flaw with disparate impact, the newest tool used by regulators like the CFPB to find discrimination when there is no intent, or even an active effort to avoid discrimination. In the case of Ally's recent settlement, the CFPB used this approach with confidence. "Whether or not Ally consciously intended to discriminate makes no practical difference to consumers," said CFPB Director Richard Cordray. "In fact, we do not allege that Ally did so. Yet the outcome, and the harm to consumers, is the very same here."

The importance of this revelation and its impact on banks is that under the current standard of the CFPB, if harm is caused, even by well meaning individuals, then the institution is at fault. While I believe there are many legitimate reasons the data might appear as it does, it is clear that some individuals have been harmed or at least could claim they were. By the standard used by the CFPB itself, discrimination has therefore occurred. At best, this demonstrates the fallacy of disparate impact as a standard for regulation and enforcement. At least, it brings the moral authority of the CFPB into question. We can hope that the CFPB will respond to this discovery with self reflection and a more rational approach to protecting our consumer rights.
Posted by Eric Lindeen | Monday, March 10 2014 at 11:51AM ET
@SB2014 - you may be taking "control" out of context. By control, I (as well as others I assume) am referencing experiential control, not any type of manual control in the process.

In order to draw valid inferences from data, you have to apply statistically valid methods.

I don't argue that there may indeed be a disparity among races in terms of performance rankings. However, in order to make that conclusion, many more factors need to be understood.
Posted by sam_sach | Monday, March 10 2014 at 9:22AM ET
Wow. The radio silence on this issue from Democrats and civil rights organizations is quite deafening. Indeed, the message reverberating is that we should care about racial disparities only when caused by familiar opponents, which is quite astonishing considering that the disparities alleged here are larger than most disparities that have served and will serve as the basis of government enforcement actions :)Ironically, what's being argued by some of you now are the same arguments typically made by those
"familiar opponents" to defend the alleged discrimination. Rather than hold your government agency responsible, you support arguments that do nothing but disservice minorities and the constituencies that you have been elected to serve. You argue now about the small sample size of minorities, which is interestingly a "variable" that the agency controls. You argue now that the analysis should control for rank, title, education pay grade, etc. I'm sure the agency can find as many controls needed to justify the disparities, as can those "familiar opponents." But as a regulator would say, disparities caused by factors that correlate to race is the same thing as saying that race is causing the disparities. And any "controls" applied should be "controls" that are relevant. Are we saying that job performance is explained by job title and pay grade? And if so, perhaps we should perform a "deeper analysis" to examine where minorities fall on the pay grade and job title range since these factors are allegedly skewing the analysis. And then the most hypocritical argument is that all of these minorities deserved their lower rankings, which is the same minority pathology argument advanced by our familiar foes just in a "different suit." Instead of looking at the structural conditions of the agency that may be causing the differentials, you blame the differentials on presumed gaps in performance and values. Here is some advice, rather than looking for ways to justify the disparities, one should try holding their regulators to the same standards that would be applied in any other context. To do so, would be the responsible thing to do.
Posted by SB2014 | Sunday, March 09 2014 at 3:11PM ET
American Banker should consider basing their claims on a statistically valid analysis. This is a pretty big claim to make without doing the following:

1) Controlling for rank
2) Controlling for division/unit
3) Running proper statistical tests (e.g. chi-sq/g-test or fisher's exact)

Looking forward to the response some statistician at the CFPB is undoubtedly drafting :)
Posted by sam_sach | Friday, March 07 2014 at 4:18PM ET
The hypocrisy angle started when the agency turned what should have been unbiased regulatory examinations into inquisitions by including enforcement attorneys on all examination teams. Hiring from the social activist and enforcement ranks rather than specific financial market professionals only exacerbated the hypocrisy and bias. What a shame that an agency supposedly dedicated to protecting consumers, long-ignored by other federal financial regulators, could be so set on self-destruction.
Posted by jim_wells | Thursday, March 06 2014 at 3:51PM ET
uesider is almost right on the money. I would add this: I would want to see the population of the managers/supervisors that actually did the evaluations. This would not prove beyond a shadow of a doubt that a problem existed or did not exist, but it would shed some light on the issue.

Also, American Banker, if playing to it's clients, is doing them no favors. All this will do is add fuel to the fire. Do we want to go to war when we hold almost no trump cards and they control the game?
Posted by robrose | Thursday, March 06 2014 at 2:17PM ET
What bothers me most is that the American Banker obtained CONFIDENTIAL information from the Bureau.....So much for privacy of the individuals.
Posted by rsquared | Thursday, March 06 2014 at 12:55PM ET
Frankly, with a sampel this small (191 people) you can't draw much of a quantitative conclusion. This is different for larger entities like banks with can thousands of employees, but with only 1100 or so I would caution strongly against drawing any sort of conclusion when a change of ten grades can push your stats by as much as 5 points in one direction or another.
Posted by editengine | Thursday, March 06 2014 at 12:49PM ET
Disparate impact in HMDA data is evidence of structural racism when you account for other variables via regression. This piece is pretty weak though, an agency that is 16% African American can hardly be compared with an industry that overages less than half that. Additionally, the CFPB workforce is diverse throughout the entire structure, unlike the financial industry which is only diverse at the lower end. I have heard there have been problems with burnout, especially in the rules writing teams but otherwise I wouldn't give too much credence to anonymous sources when we have pretty good data on the diversity of the workforce and the diversity of the banks which they regulate. There simply if no comparison, you can review the latest GAO report on the topic here. It would see that if you are writing an article on this subject you would cite relevant literature. http://www.gao.gov/assets/660/653814.pdf
Posted by editengine | Thursday, March 06 2014 at 12:26PM ET
I get that morale is very low and turnover quite high at the CFPB, and that is certainly problematic. But the rest of this article is just dumb and a cheap shot at an easy target. Frankly, it smells like the American Banker is just shilling for and pandering to an industry upset at the CFPB. The fact that members of one race received better grades on their employee evaluations than members of other races does not indicate discrimination and has absolutely nothing to do with the CFPB's fair lending enforcement or disparate impact. It is, after all, possible that the employees who received higher marks deserved them, a possibility not even mentioned in the article, likely out-of-fear of being politically incorrect. The article's conclusion is reminicent of community activist who argue that HMDA data showing higher denial rates for one race is conclusive evidence of discrimination. That position, like the article's, ignores the underlying facts that might give rise to the final outcome.
Posted by uesider | Thursday, March 06 2014 at 10:06AM ET
Power corrupts, absolute power corrupts absolutely.

It would be interesting to see what this same statistical snap-shot would reveal about the financial regulators who seem to be "pearly white" as well - in mind, body and attitude.
Posted by mdillon | Thursday, March 06 2014 at 9:55AM ET
They're not borrowers so the Fair Lending and non-discriminatory laws do not apply here. But, truly, the fox is indeed in the henhouse.
Posted by lenard_poon | Thursday, March 06 2014 at 8:13AM ET
Add Your Comments:
Not Registered?
You must be registered to post a comment. Click here to register.
Already registered? Log in here
Please note you must now log in with your email address and password.