WASHINGTON, D.C. — Two days before President-elect Donald Trump’s inauguration, Republican members of the House Financial Services Committee made public on Wednesday a new set of internal Consumer Financial Protection Bureau (CFPB) documents and a new report critical of Director Richard Cordray. This is the third report the committee has published on the CFPB since November 2015.

Based on internal CFPB documents obtained by the committee, the report, “Unsafe at Any Bureaucracy, Part II: The CFPB’s Vitiated Legal Case Against Auto-Lender,” alleges that Cordray violated federal law governing agency rulemaking when the regulator published its “larger participant rule” for the auto finance market in 2015.

“Part III in this continuing series makes additional CFPB documents available to the public,” read the 30-page report’s executive summary, in part. “These documents discuss the CFPB’s disparate-impact methodology in more detail. They also reveal potential legal deficiencies in the issuance of the CFPB’s major rule authorizing it to supervise the larger participants of the auto lending market.”

The committee’s first report on the CFPB charged that Cordray was aware that the statistical method the bureau used to allege racial discrimination in auto finance, known as the Bayesian Improved Surname Geocoding, was “prone to error.” It also revealed that bureau lawyers had warned the director of the “weakness” of the disparate impact theory the regulator relied on to build its discrimination cases against auto finance sources.

The second report, issued in January 2016, showed that some checks being dispersed as part of the bureau’s $98 million settlement with Ally Financial and Ally bank had gone to white borrowers. It charged that the bureau’s theory of liability for auto financers could not meet the legal test for disparate impact.

This time, the House committee’s report charges that the CFPB failed to take steps its attorneys advised the director to undertake to comply with the Administrator Procedure Act when it issued its larger participant rule. The APA “requires agencies to publish ‘notice’ of ‘either the terms or substance of the proposed rule or a description of the subjects and issues involved’ to ‘give interested persons an opportunity to participate in the rulemaking through submission of written data, views, or arguments.’”

“Courts have held that to satisfy the [requirements] agencies must provide an opportunity for the public to comment on data and technical studies used to support a rule,” the report states, in part. “An agency must provide data supporting its conclusions, and if it fails to do so it must publish the data to re-open its comment period.”

According to the report, the bureau based its definition of “larger participants” on “quantitative information taken from Experian’s AutoCount database. During the comment period, which opened after the CFPB published its proposed rule in October 2014, the CFPB received multiple requests for a list of the institutions it believed the rule would cover. The bureau, however, did not respond with requested information because “it understood that Experian regarded the AutoCount data to be “proprietary.”

When the comment period closed in December 2014, Experian informed the CFPB it did not object to the bureau releasing the list of institutions that would be covered under its proposed threshold. In February 2015, according to the report, citing internal bureau documents, CFPB attorneys delivered a memo to Cordray recommending that he publish the list and reopen the comment period to allow for comment on the released data. The attorneys also warned “there is a cognizable risk that a court could conclude that the APA required the CFPB to do so.”

“After explaining the legal analysis supporting their conclusions that the CFPB should reopen the comment period, the attorneys outlined the possible negative outcomes of doing so, including the risk of ‘Congressional scrutiny,’ the prospect that it would ‘raise questions about whether the four prior Larger Participant rules were procedurally defective since the bureau did not solicit public comment on the identities of the entities that would qualify the supervision …,’” the report reads, in part.

Cordray signed the final rule on June 5, 2015, without reopening the comment period.

The report also offers more details on the House committee’s objection to the CFPB’s use of disparate impact, a legal theory that says a policy can be deemed discriminatory if it has an adverse impact on a protected group, even if it’s unintentional. Citing Supreme Court precedent, the report says the CFPB’s use of disparate impact “would not survive judicial scrutiny.”

A CFPB spokesperson said the bureau is reviewing the House committee’s report.

“The bureau is committed to ensuring that consumers are treated fairly in the financial marketplace, and makes a conscientious effort at all times to carry out its mission in compliance with all applicable laws,” the spokesperson said.