Reforming America’s lenders of second-to-last resort

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Last week the US Federal Housing Finance Agency presented its report on the future of the Federal Home Loan Banks. An FHFA report on FHLBs might sound a little FML, but genuinely interesting things are afoot. Really!

FHLBs are private but government-sponsored banking co-operatives set up to support the US housing market after the Depression, using their implicit sovereign guarantee to borrow cheaply and lend the money on to their members to support residential mortgages. In other words, they’re obscure but important — and increasingly controversial in some circles.

As Alphaville wrote earlier this year, over the years they have morphed into unofficial de facto lenders-of-second-to-last-resort to the US banking industry.

That’s arguably destabilising in a crisis: it lets troubled lenders mask problems, undermines the Federal Reserve formal backstop role, and increases the cost of the clean-up if the borrowing bank collapses (because the FHLB advances get repaid before everyone, even the Fed). Viz, Silicon Valley Bank.

Not great, in other words. The FHFA — which regulates the FHLBs — now wants to change things, and sketched out what wants to do in a comprehensive report on Nov 7. It has been in the pipeline for more than a year but it’s obviously lot more timely after the spring’s banking shenanigans.

The mission creep into emergency lending is a major theme. As the report notes (with Alphaville’s emphasis below):

The FHLBanks’ advance activity is funded primarily by the issuance of bonds and shortterm notes, and there are practical limits to the amounts that the FHLBanks can issue in a single day. When member demand for advances is high, these limitations can become acute, particularly late in the day when debt markets have slowed or closed.

The FHLBanks can better serve their members if they coordinate with their large depository institution members and the members’ prudential regulators to make certain that these members have established protocols to meet their emergency liquidity needs from the Federal Reserve discount window when necessary. Ensuring that the FHLBanks are not acting as lenders of last resort for institutions in weakened financial condition will allow the FHLBanks to use their available liquidity to provide financing to all members so they can continue to serve their communities.

During the 2023 market stress caused by multiple regional bank failures, it became apparent that several large depository members were effectively using the FHLBanks as their lender of last resort. These members did not have agreements in place or collateral positioned to borrow from the Federal Reserve discount window. Accordingly, FHFA will provide guidance to the FHLBanks to work with their members and the members’ primary federal regulators to ensure all large depository members have established protocols to borrow from the Federal Reserve discount window so that these institutions’ borrowing needs continue to be met. Additionally, FHFA expects the FHLBanks to negotiate appropriate agreements with the regional Federal Reserve Banks to ensure expedited movement of collateral if a member’s lending activity must be moved to the Federal Reserve discount window.

Elsewhere, the FHFA sounded sceptical proposals to let mortgage REITs and non-bank mortgage lenders become members of the system, and said that it wanted to make sure that existing members maintain 10 per cent of their assets in residential mortgage loans

This may sound like a small point, but at present you just need to have 10 per cent to be admitted into the FHLB fold in the first place, but what happens after that doesn’t matter. Here’s JPMorgan on the implications of this change.

. . . Thinking through the possibilities, around a third of commercial banks currently wouldn’t meet the 10% threshold; forcing banks to maintain a 10% resi loan share would likely make them reconsider whether they want to hold more loans (to get over the hurdle) or fewer (since they might not receive as favorable funding on what they do hold). For those looking to scale up their holdings, the better regulatory prospects for CLNs point to an appealing way to keep more loans on balance sheet at a lower risk weight (and higher ROE). On the insurance side, none of the public insurers are close to the 10% limit, which means it would probably be a heavy lift for them to scale up to maintain eligiblity. However, if maintaining FHLB membership and liquidity is viewed as particularly important, that would be beneficial for loan demand.

All these reforms sounds eminently sensible, but analysts at Barclays have raised some fascinating but potentially major unintended consequences of the proposals.

For example, if tottering banks cannot tap their local FHLB for funding — and the stigma of going to the Fed’s discount window remains undimmed — this could exacerbate ructions in funding markets as they scramble for alternatives.

Here’s Barclays’ short-term rates savant Joe Abate, with FTAV’s emphasis:

It is said that you can bring a horse to water but you can’t make it drink. Our sense is that this is true about discount window borrowing. While banks’ ability to use the FHLB as a lender of last resort will be more limited in the future, it is not clear that banks will be any more willing to borrow at the discount window. In recent funding events, healthy banks have used advances or other Fed liquidity programs and avoided the Fed’s primary credit program. All things equal, we suspect that limiting FHLB lending to term funding will mean that future funding shocks could be larger and potentially more widespread as banks chase shrinking market sources of liquidity to avoid going to the discount window. Banks may respond by holding thicker precautionary liquidity buffers at the Fed, increasing their minimally comfortable level of bank reserves.

Separately, Barclays’ interest rate derivatives team highlight another even wonkier but possibly problematic side-effect.

The FHLBs raise the money for their members by issuing their own bonds and notes. And an increasingly large proportion of this debt is callable — almost $380bn at the moment, according to FHLB Office of Finance

In April 2023, nearly 85 per cent of all callable fixed income issuance in the US was by FHLBs funding their member advances, and even today they account for about 50 per cent.

Barclays analysts say this matters because all this callable supply in practice helps dampen interest rate volatility — especially at times of broader financial stress. The implications are “potentially significant,” writes Amrut Nashikkar and Eveline Dong

For fear of mangling this we’re going to quote directly from the report:

An investor who buys a callable bond is indirectly selling embedded volatility into the vol market, because a significant proportion of callable issuance tends to get swapped. A surge in callable issuance that occurs precisely when the demand for interest rate vol rises helps stabilize the vol market.

. . . The proposed changes will likely take time to filter through to the market because they require concrete policy action and implementation. If changes to the operation of the FHLB system result in reduced debt issuance during times of crisis, this would also mean callable debt issuance, and hence vol supply, during stress events. In a future banking system stress event, when rate uncertainty is high and market participants are looking for tail risk hedges in short-term rates, dealers may find themselves with limited access to an important source of interest rate volatility. The result should be a higher ex-ante vol risk premium in policy-sensitive parts of the vol surface, which tends to coincide with the tenors and expiries over which issuance takes place. Further, since stress events are associated with rallies, this could especially affect skew in the vol surface.

Suddenly we have an overpowering urge to just write a simple “stocks go up/down” story.