Fed under pressure to rethink TLAC rules

US becomes the first G20 nation to propose total loss absorbing capacity rules, but deems existing debt ineligible for inclusion

Paul Tucker: "TLAC isn’t about faith; it’s about contracts"

  • The Federal Reserve has proposed its total loss absorbing capacity rules for the eight US global systemically important banks (G-Sibs).
  • The US regulator could require banks to meet over half of the TLAC requirement with long-term debt rather than equity.
  • Eligible debt is subject to tight restrictions on acceleration clauses and securities issued under foreign law are not permitted to be used as TLAC.
  • The Fed has not confirmed whether existing debt can be grandfathered under the new rules, and it estimates that up to $120 billion in new eligible debt securities will need to be floated.
  • Bank lobbyists refute that number and claim that the true TLAC shortfall is actually $363 billion, with 90% of outstanding G-Sib debt ineligible to be included as TLAC.

Even by Wall Street standards, $363 billion is a big number. That is the estimated shortfall in outstanding long-term debt that needs to be made up by the eight US global systemically important banks (G-Sibs) by 2019, in order to satisfy their total loss-absorbing capacity (TLAC) requirements.

The deficit, calculated by a group of five US bank trade associations, is three times the size of the Federal Reserve’s own estimate. In its October 2015 rule, proposing the TLAC requirements, the US central bank estimated the collective TLAC shortfall among the eight US G-Sibs would be only $120 billion – a gap the regulator “assumed that covered bank holding companies (BHCs) would fill by replacing existing ineligible debt with eligible external long-term debt”.

The wide gulf between these two numbers neatly encapsulates the battle that is brewing between Federal Reserve economists and New York bankers over TLAC.

On one side, government regulators insist G-Sibs are not holding enough debt securities that could be converted into equity or written off during resolution proceedings to avert the need for a taxpayer-funded bailout.

On the other, banks insist they are better capitalised today than they were pre-crisis, and liabilities such as equity and existing outstanding debt securities should be eligible to be counted as TLAC instruments, rather than forcing G-Sibs to issue hundreds of billions in – what they see as unnecessary – new eligible long-term debt.

“In the TLAC formulation, as envisioned by the Financial Stability Board (FSB), one-third of TLAC will be covered by senior bonds. The US goes beyond the 33% requirement of the FSB, and I think that is because they start from the assumption that capital is very often not there when you activate resolution procedures, so long-term debt acts as a second line of defence to replenish capital,” says Santiago Fernández de Lis, chief economist of financial systems and regulation at BBVA Research in Madrid.


The Fed’s insistence that G-Sibs retain higher levels of long-term debt is at the heart of industry protests over the TLAC proposal. Their main concern is that US authorities are applying a much more stringent long-term debt requirement than the baseline standard set out in the FSB TLAC term sheet, published in November 2015.

The FSB states the minimum external TLAC requirement – loss-absorbing assets held by third parties outside the bank, which can be written off or converted into equity during a resolution – must be equivalent to at least 16% of the group’s risk-weighted assets (RWAs), rising to at least 18% by January 2022.

The FSB was far less prescriptive on the actual liabilities comprising this TLAC, however. It states that common equity Tier 1 capital can be used to meet TLAC requirements under certain circumstances, but “there is an expectation” that a G-Sib would meet at least 33% of its minimum TLAC requirement with long-term debt.

The Federal Reserve goes far beyond this one-third baseline. Its requirements apply differently, depending on the composition of a G-Sib’s balance sheet, but a PwC research note from November 2015 sums up the effective debt requirement levied in the proposed rule: “The amount of required long-term debt must be greater than 7% to 10.5% of consolidated RWAs and 4.5% of the G-Sib’s consolidated leverage exposure. A firm subject to the 10.5% long-term debt requirement must therefore hold well over half of its 18% required TLAC in the form of long-term debt, which is significantly higher than the FSB proposal’s 33%.”

The line that equity will have evaporated by the time of resolution, so should not be used as TLAC, is a false argument. Regulators have the ability to invoke Title II when there is a risk of failure
Legal counsel for an industry association

The logic behind the Fed’s insistence on G-Sibs holding large amounts of long-term debt as TLAC is clear. Allowing banks to fulfill TLAC requirements with large amounts of equity increases the risk that little loss absorbency capacity will remain to recapitalise the firm and conduct an orderly resolution. Requiring banks to hold a large proportion of long-term debt is one way to combat this risk, but bank lobbyists insist the Fed could structure its rules to allow large volumes of equity to be used as a TLAC instrument in a safe and prudent manner.

“The line that equity will have evaporated by the time of resolution, so should not be used as TLAC, is a false argument. Regulators have the ability to invoke Title II when there is a risk of failure – it does not have to be at the point of failure. Likewise, the regulators are supervising the G-Sibs, so they can also prohibit banks from taking write-downs against equity if they need to. We are kind of baffled by this long-term debt requirement. There is nothing comparable to this in the FSB term sheet,” says the legal counsel for one industry association.

Banks also have grave concerns over the tough eligibility requirements that the Fed rule imposes on debt securities in order to meet its TLAC standard. Structured notes, bonds with credit-sensitive features or contractual provisions for conversion into equity are all prohibited from being included in the TLAC stack.

A more problematic restriction is an outright ban on the use of any debt securities that give the bondholder a contractual right to accelerate the repayment of the debt. The Fed’s thinking here is again understandable. Acceleration clauses allow bonds to be repaid early if certain circumstances are met, depleting the available debt stack that regulators would be able to use as TLAC in a resolution.

Ironically, the one acceleration clause that is permissible under the proposed Fed rule is early repayment triggered by an insolvency event. Debt with this clause can be included in the TLAC stack because the insolvency of the BHC would coincide with the firm’s entry into a resolution proceeding, “in which case the payment obligations would generally be stayed and the debt would remain available to absorb losses”, the Fed rule reads.

Another troubling exclusion is the requirement that only debt instruments “governed by the laws of the US or any state thereof” are counted as eligible securities under TLAC.

The rule’s only explanation for taking this stance is that eligible debt securities should “consist only of liabilities that can be effectively used to absorb losses during the resolution of a covered BHC under the US bankruptcy code or Title II, without giving rise to material risk of successful legal challenge”.

The inference in this sentence seems clear: the Fed has taken the position that allowing long-term debt issued outside of the US to be bailed-in as TLAC assets introduces legal uncertainty, and could leave a US resolution vulnerable to litigation in foreign jurisdictions.

The exclusion is all the more notable given the extensive work undertaken by supervisors in the G20 countries to develop home-nation/host-nation regulatory co-ordination through the single-point-of-entry (Spoe) resolution regime, under which one home regulator takes charge of the resolution of the entire G-Sib by assuming control over the BHC.

Broad prohibition

The aggregate impact of these exclusions in disqualifying outstanding long-term debt from inclusion as eligible TLAC assets is staggering. According to a joint analysis, conducted by five bank trade associations, the eight US G-Sibs had $964 billion of long-term debt outstanding in the market as of September 2015. Under the current Fed proposal, however, $866 billion of this debt (90%) would not qualify as eligible debt securities for TLAC purposes due to acceleration clauses and other prohibitions.

The same analysis found that of the $793 billion in ineligible plain-vanilla long-term debt securities (excluding structured notes) issued by US G-Sibs in September 2015, $95 billion (12%) were bonds governed by foreign law.

The trade associations concluded that if the Fed’s final TLAC rule was relaxed to allow debt with embedded acceleration clauses and foreign law debt securities to be eligible for inclusion, the long-term shortfall would tumble from $363 billion to just $56 billion – halving the Fed’s own estimate of $120 billion.

Even if the Fed was open to relaxing its eligibility standards, one great unknown still remains: would the regulator allow US G-Sibs to grandfather across already-issued long-term debt? The Fed did not contemplate grandfathering existing debt in its proposed rule, but it did ask consultation respondents to provide feedback on whether it should be permitted. The response from industry has been emphatically in favour of such a measure.

“Grandfathering existing long-term debt into eligible debt securities under the Fed rule would make a big difference. That would bring the industry’s projections into harmony with the Fed’s estimates and lessen the expected shortfall on eligible debt securities. There would still have to be some new long-term debt issuances, but it would be in the range of $150 billion to $200 billion, as opposed to the $634 billion in new issuances [between now and 2019] we are estimating would be required. That would ameliorate a lot of our concerns,” says Robert Hatch, director and counsel for regulatory affairs at the Financial Services Roundtable in Washington DC.

This $634 billion figure is comprised of $271 billion in ineligible securities maturing between now and 2019, which could be refinanced with TLAC-eligible debt, plus the $363 billion net shortfall.

Permitting grandfathering to occur would only be meaningful, however, if the restrictions on acceleration clauses and foreign law securities are also dropped. Without this, such a provision would provide only temporary relief.

Intended consequences

The potential for up to $363 billion in new US G-Sib debt that must be eligible for bail-in to flood the system makes it almost a certainty that banks will have to pay more to issue debt. This is especially true given that credit markets will be able to roughly approximate the amount of paper that each G-Sib will need to issue for external TLAC purposes and price their bids accordingly.

“The TLAC requirements are going to make debt more expensive for G-Sibs, but they are meant to, because debt will no longer carry a de facto government guarantee. During the crisis it wasn’t the banks that were bailed out – it was the bondholders. Equity holders were either wiped out or hugely diluted, but bondholders got rescued. Perhaps they expected to be rescued, which meant they didn’t charge enough for the risk and that made it too cheap for the banks to lever up. We are supportive of the Fed’s approach in setting out higher TLAC standards,” says Paul Tucker, chairman of the Systemic Risk Council and former deputy governor of the Bank of England.

How much the yields on new long-term debt will increase remains to be seen. But accepted regulatory wisdom is that G-Sibs have benefited from underpriced bond issuances for years on account of the implicit subsidies they have enjoyed as too-big-to-fail banks, which could be expected to be rescued by the Fed in the event they got into trouble.

The increase in the cost of issuing debt is consequently seen by US regulators as pricing in the true risk presented by G-Sib bonds, absent a discount for expected government support now that a more credible resolution regime has been established for these banks.

Internal TLAC

Once the external TLAC has been issued by the G-Sib, the BHC has to determine what to do with the cash proceeds. For US banks, the issue is a relatively simple determination around how much TLAC cash to preposition with its operating subsidiaries and how much cash to retain at the holding company as contributable assets.

The matter becomes more complex, however, in a cross-border context, where a foreign BHC has to implement a resolution regime for an intermediate holding company (IHC) housing operating subsidiaries in another jurisdiction, such as the US. This is where internal TLAC, the second element of the FSB regime, comes into play.

Internal TLAC is the mechanism by which the operating subsidiaries of a G-Sib, such as its national bank or broker-dealer, are kept in business without interruption while the holding company is put into resolution.

In its simplest form, an operating subsidiary sells debt securities to its own holding company. In exchange for issuing these internal TLAC securities, the overseas parent provides the US subsidiary with cash. The subsidiary deploys that cash in its regular course of business and records the internal TLAC as a liability owed back to the holding company on its balance sheet.

Should the BHC be put into resolution under a Spoe strategy, these liabilities would effectively be wiped from the balance sheet of the operating subsidiary, thereby recapitalising it and enabling it to continue operating.

The concept is simple, but resolution experts caution the Fed has yet to define the exact mechanics whereby any losses experienced at a US operating subsidiary would be transmitted up to the foreign holding company to ensure losses accrue to the parent, rather than the affiliates.

“It is not clear how an IHC upstreams losses to the parent BHC. This is not just a weakness of the Federal Reserve proposal, but it is also unclear in the global FSB framework, and this is something that needs to be defined. Once you have issued internal TLAC, it becomes an asset of the BHC, but a liability of the IHC that could be activated under certain circumstances to recapitalise the IHC if necessary. But how will this trigger be defined? Who will have the capacity to activate the trigger? All these things remain to be defined,” says one regulatory specialist at a European bank.

It seems crazy that the Fed, the Bank of England and the European Central Bank can’t get their act together about how they would work together in the next crisis

Internal TLAC also plays an important role in engendering trust between home and host regulators. The international consensus to implement a regime where foreign subsidiaries would essentially be recapitalised by inflicting losses on a domestic parent bank presents a novel test of the confidence that regulators have in one another’s competency to resolve huge banks in an orderly fashion.

“TLAC isn’t about faith; it’s about contracts. When I was involved with this as a policymaker, the goal was to take faith out of the equation. In some respects, policymakers are trying to set up a structure where home and host regulators have to come clean with each other in advance, rather than discovering whether or not they can trust each other in the middle of a crisis. This process will reveal to both parties whether they can or cannot work with each other,” says the Systemic Risk Council’s Tucker, who was involved in the development of resolution planning during his time at the Bank of England.

“I personally think this is the most important regulation to come out of the Federal Reserve in some time, because this is the nub of the entire reform structure. We are discovering whether countries are capable of working together to come up with a regime where groups are resolved by putting losses on bondholders, so that vital services for households and businesses can be sustained without them paying for it as taxpayers. The Fed’s rule is pretty good in achieving that goal,” he adds.

Other commentators take the opposite perspective, viewing internal TLAC as symptomatic of a complete lack of trust between regulators. The FSB proposed setting internal TLAC in a range of 75–90% of what the IHC’s standalone TLAC requirement would be, to be decided by home and host supervisors in a crisis-management group. Instead, the Fed has opted for 90% without consulting home supervisors.

Pre-funding overseas IHCs with loss-absorbing debt before a crisis largely obviates the need for co-operation between regulators during a resolution since the bail-in funds are already in place. An even less charitable interpretation is that the hundreds of billions in new TLAC debt issuance is a burden that banks are facing simply in order to make the work of regulators in a resolution easier.

“What the Fed is signaling to other central banks here is a message that in the next crisis it will be the lender-of-last-resort to banks in the US, but it will not provide liquidity outside of the US. The Fed will not be concerned [with] what is happening in London and I think that is kind of selfish,” concludes one bank lobbyist.

“It just seems crazy that the Fed, the Bank of England and the European Central Bank can’t get their act together about how they would work together in the next crisis, and instead are forcing the banks to restructure themselves in a way that makes it easier for the regulators to do their jobs. They’re doing this to make it easier for regulators to react in the next crisis, and I would argue that they are doing that at the expense of damaging the day-to-day functioning of the G-Sib banks,” the lobbyist adds.

Fed TLAC rules present headaches for Mpoe banks

Amid bank complaints regarding the amount of new long-term debt that banks will have to issue to meet the Federal Reserve’s proposed total loss absorbing capacity (TLAC) rule, one aspect has received less attention: the potentially adverse impact on banks following a multiple-point-of-entry (Mpoe) resolution strategy.

All eight US global systemically important banks (G-Sibs) are implementing a single-point-of-entry (Spoe) resolution plan, under which regulators would assume control of the holding company of a collapsing bank and establish a bridge institution to oversee the winding down of the firm and the sale of its assets.

Some foreign G-Sibs – notably Santander and HSBC – have opted to follow an Mpoe strategy, under which multiple different legal entities of the bank are resolved individually. Foreign subsidiaries and affiliates of G-Sibs undergoing an Mpoe strategy would be resolved by the host regulator, so the US intermediate holding company (IHC) would be resolved by the either the Fed, the Federal Deposit Insurance Corporation or through bankruptcy.

This creates a problem, however. The Fed’s TLAC rule only envisions the IHCs of foreign banks issuing internal TLAC back to the parent holding company. The rule does not propose IHCs can issue long-term debt to third parties in the same way that bank holding companies (BHCs) will be obliged to, in order to raise external TLAC.

Since the BHC of a G-Sib pursuing an Mpoe resolution plan would not intend to preposition internal TLAC in its US subsidiary, concerns have been raised that such an IHC may find itself unable to raise sufficient TLAC to effect an orderly resolution if it is unable to issue external debt to build its own TLAC stack.

“Our position is that if a foreign bank following Mpoe is going to have an IHC resolved here in the US, they should not be forced to issue internal TLAC to the parent. They should be allowed to do what the US BHCs are allowed to do, which is issue external TLAC to the market, but they are not able to do this under the current proposal. That completely disrupts the global Mpoe strategy, meaning the strategy approved by the home-country authority is sacrificed without recognition that the Mpoe resolution strategy contemplates the local operations will be protected,” explains Sally Miller, chief executive of the Institute of International Bankers in New York.

Whether IHCs could proceed to issue external debt in the hope that it will ultimately be eligible as TLAC is unclear. The Fed’s proposed TLAC rule states only that “US regulators are cognisant of the need to prepare for other plausible contingencies, including the Mpoe resolution of a G-Sib. While this proposal is primarily focused on implementing the Spoe resolution strategy, it would also substantially improve the prospects for a successful Mpoe resolution of a G-Sib by requiring US G-Sibs and the IHCs of foreign G-Sibs to maintain substantially more loss-absorbing capacity.”

With little clarity on the matter, market participants have requested the regulator make a definitive pronouncement on whether Mpoe banks will be able to go their own way in issuing external TLAC instruments.

“The key difference between Spoe and Mpoe banks is that, in the Spoe, all the losses are absorbed at the BHC level, while in the Mpoe, banks can be resolved locally and they should have loss-absorption capacity in the subsidiaries. That requires internal TLAC for Spoe banks and external TLAC for subsidiaries of Mpoe banks that are resolution entities. In this regard, subsidiaries of Mpoe banks in the US should be required to hold external TLAC and not internal TLAC, because they are resolution entities in the US,” says Santiago Fernández de Lis, chief economist of financial systems and regulation at BBVA Research in Madrid.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here