FHFA Updates FHLBank Membership Rules – NMP Skip to main content

FHFA Updates FHLBank Membership Rules

Jan 12, 2016

The Federal Housing Finance Agency (FHFA) has released a final rule amending its regulation on Federal Home Loan Bank (FHLBank) membership by financial institutions.

The updated rule follows more than 1,300 comment letters received by the regulator after it issued a proposed rule in 2014. The final rule does not include provisions from the 2014 proposed rule requiring financial institutions to retain FHLBank membership by maintaining ongoing minimum levels of investment in specified residential mortgage assets.

"The statutory requirements for members to continue their commitment to housing finance can be addressed by monitoring the levels of residential mortgage assets they hold and we, therefore, decided not to include the ongoing investment requirements in the final rule," said FHFA Director Melvin L. Watt.

But the new rule features a provision that redefines "insurance company" that excludes captive insurance companies. The FHFA argued that captive insurers primarily focus on underwriting insurance for its parent company or for other affiliates, instead of the general public.

Rather than immediately ban captive insurers from membership, the new rule allows FHLBank captive insurer members that joined before the new rule takes effect to terminate their membership within the next five years. In making this announcement, Watt added that Congress has the authority to amend the Federal Home Loan Bank Act to allow captive insurers the right to become FHLBank members again.

“The MBA is disappointed, to say the least, in the final rule on FHLBank membership,” said David H. Stevens, CMB, president and CEO of the Mortgage Bankers Association (MBA). “When Congress established the FHLBank membership framework, it didn’t limit membership to only certain insurance companies. Today, without direction from Congress, FHFA decided to narrow membership and exclude important members of the system.”

Rob Nichols, president and CEO of the American Bankers Association (ABA), countered the sentiments of Stevens and the MBA.

“We’re pleased that FHFA listened to the more than 1,300 comment letters from ABA, state associations and banks opposing this provision. There was no demonstrable need for the proposed changes, which contradicted congressional intent and would have harmed the FHLBs, their member banks and the communities they serve,” said Nichols in a statement.

“In today’s marketplace, we need a FHLB system that serves the wide variety of lending institutions active in today’s housing finance market, including captive insurance companies, REITs, independent mortgage bankers, and other entities, all of which provide major sources of liquidity and are core components in the 21st century FHLBank system,” said Stevens. “We will continue to work with Congress on this issue to address the shortcomings of today’s rule.”

About the author
Published
Jan 12, 2016
MISMO Updates Business Glossary To Support AI, eMortgages

New definitions covering eHELOCs, remote online notarization, valuation modernization, and compliance initiatives aim to improve consistency

MISMO Launches AI Governance Framework For Mortgage Lenders

New FRAME toolkit gives lenders, servicers, and technology providers a roadmap for managing AI risk while supporting innovation

CFPB Tells Lenders Immigration Status Can Factor Into ATR Analysis

CFPB frames immigration status as a potential ability-to-repay factor when future U.S.-based income is at risk

UAD 3.6 Deadline Nears; First American Earns Verification

First American's ACI Sky Workbench gains verification ahead of the Nov. 2 implementation date for the GSEs' updated appraisal reporting requirements

MISMO Introduces New Loan Boarding Standard

Wrapper Files support standardized data transfers between origination and servicing systems, with potential savings of $60 to $160 per loan

The GLBA Compliance Gap Your AI Deployment Just Opened

Old statutes, new models, and the vendor contract you signed before machine learning became operational