Securities lending indemnification is mispriced, subsidised by the agent providers and not all it’s cracked up to be as a risk mitigant, according to Credit Benchmark’s co-founder Mark Faulkner. He talks to Bob Currie about how to reset the economics of the agency lending market
“Something's happening and it's happening right now, you're too blind to see it”, said UK punk band The Stranglers back in 1977. Perhaps this was a veiled reference to the extension of borrower default indemnification in the securities lending world and to the heavy regulated capital cost this would come to represent for the agent lender community.
Mark Faulkner liked the song enough to name a research paper after it — or so SFT would ask you believe — and his study, Something Better Change: Securities Lending Indemnification is Unsustainable in its Current Form (‘the report’), combines an ear for a good tune with a thought-provoking assessment of pricing, risk and competitive dynamics in the SBL marketplace.
An overarching thesis running through the report is that beneficial owners have become “overly dependent” upon securities lending indemnification, with many requiring it as a matter of course rather than after carefully evaluating its value as a genuine source of risk mitigation. This indemnification offers protection to the lender against two potential concurrent statistically low risk events — the default of the borrower counterparty and a shortfall in the collateral following liquidation.
This situation persists, in part, because securities lending indemnification remains poorly understood by those receiving the benefit. In many cases, Faulkner suggests, indemnification is mispriced — and artificially subsidised — by the service providers offering this protection. With the associated capital cost, this has become increasingly expensive for the agent bank to deliver (p 8).
Cost of protection
Running a more detailed cost-benefit analysis — and building on the groundwork done by State Street’s Glenn Horner in a 2013 paper named The Value and Cost of Borrower Default Indemnification — the report estimates that the economic benefit realised by the beneficial owner, the protected party, is low at approximately 0.2bps. This is well below the real-world cost of providing the indemnification, which it estimates to be closer to 0.9bps — a sizeable difference that, Faulkner suggests, is invariably absorbed by the custodial lending agent and almost never passed on to the lender or the borrower (p 22).
When the regulatory capital cost attached to providing indemnification is also factored in, the gap between the economic benefit to the lender and the aggregate cost of providing indemnification widens dramatically — roughly 13bps regulatory capital cost under the Basel III bank capital adequacy regime.
This regulatory cost, he notes, has grown significantly since 2015 and is likely to remain well above 2015 levels with enactment of forthcoming Basel IV regulation. “Although agent advocacy with regulators has gone some way to reducing this cost, the spreads between the benefit, economic cost and regulatory cost of indemnification remain material,” says Faulkner.
Recognising the economic drag, or disutility, presented by this cost overhead, the report argues that the regulatory capital costs have become decoupled from the economic reality. In practical terms, this has been punishing for agent lenders, with this heavy regulatory capital cost simply not an efficient use of capital for those organisations.
As one un-named agent lender told the report author: “We think about capital being fluid and moving around our organisation seeking returns. With the cost of capital across the street [being] circa 10 per cent and targeted post-tax returns of 15 per cent, using capital to support securities lending indemnification is not the best use of capital for any organisation — especially when significantly higher returns on capital are available elsewhere.” (p 9)
Responding to this dialogue, Mark Faulkner indicates that it is hard to understand why a large institutional lender (for example, a sovereign wealth fund, a large pension fund or endowment) would continue to demand this indemnification requirement from their agent when this provides such limited economic benefit for the recipient.
Faulkner understands that State Street has been in the securities lending business since 1974 and has never been required to pay out on an indemnification programme. Yet, the regulatory capital cost that agents bear in providing this indemnification remains punishingly high.
The train has long since left the station, however, and agents have missed earlier opportunities to explain the regulatory challenges that they were facing to their clients and to potentially reset the economics of the securities lending market.
This divergence between the cost, the benefit and the regulatory capital cost of indemnification is unsustainable in current market conditions, he notes, and this “will hopefully provide a catalyst for change”. The report urges interested parties to work with regulators to ensure that the securities lending industry can continue to deliver its important role in providing short-side liquidity for global capital markets. Failure to change could have significant ramifications for global capital markets, it concludes.
(Don’t you like the way) I seem to enjoy it
So how will the industry adapt from the current indemnified orthodoxy? For a lender that receives indemnified lending at heavily subsidised cost from the agent lender, there may be little motivation to change. And borrowers, who are unlikely to pick up this indemnification cost, will not be shouting loudly for reform. Somewhat caustically, the paper indicates that the largest borrowers have been “free riding” the provision of effectively free indemnification, which in practice has lowered the price of the GC lending market.
Understandably, many within the lender and borrower communities rather like the current arrangement. “For years, the agents have effectively been picking up the tab for facilitating business between principals while insuring one side of the trade against the default of the other,” says the report (p 28).
One reason for reform, Faulkner notes, is that this will potentially deliver a more cost efficient and innovative industry where lenders can explore more creative ways of lending either through intermediated or non-intermediated channels. In their cost-benefit analysis, lenders would do well to factor in the potentially significant benefits of being able to opt out of an indemnified programme and construct a bespoke lending solution for themselves.
A group of large global lenders — formed initially by the California Public Employees’ Retirement System (CALPERS), the Healthcare of Ontario Pension Plan (HOOPP), the State of Wisconsin Investment Board (SWIB) and the Ohio Public Employee Retirement System (OPERS) — have been collaborating for almost a decade (and considerably longer informally) to drive peer-to-peer securities lending and financing. For organisations that are not members of this Global Peer Financing Association (GPFA), attractive opportunities also exist to widen their lending opportunities through directed trading, by extending loan terms and employing wider collateral flexibility.
According to Faulkner, the benefits to asset owners of employing greater flexibility in their lending strategies is more than likely to outweigh the costs of foregoing this low value indemnification. He is encouraged that some forward-thinking asset owners are already moving in this direction and are active in the securities lending market without borrower default indemnification. He recounts a recent discussion with the head of risk at a global sovereign wealth fund, for example, who had already reached the same conclusion — that indemnification is unnecessary to operate their programme effectively and they may achieve better programme performance and more creative lending strategies without being indemnified.
Let me down easy
The genesis of this situation was that, at the time that indemnification was introduced, the economics involved in offering this service were not well understood. But with the way that margins have moved in the global securities services marketplace subsequently, Faulkner believes that the economics of indemnification are not sustainable for the agent lender as they are structured currently.
General collateral lending, in particular, consumes a high level of capital for the agent lender and provides relatively low returns for the lender — but, the report notes, this continues to be priced at unrealistic levels.
To sustain this GC lending activity, Faulkner believes that the market needs to embrace new business models, reduce the “historic and unnecessary dependency on indemnification” while potentially also managing down loan supply. The market, he feels, is saturated with GC supply and agents need to start managing some lenders — beneficial owners that generate limited supply of higher-value hard-to-borrow loan securities — out of the market. “It is well understood that managing an ‘unattractive’ lender out of a programme is fraught with difficulties,” notes Faulkner, “not least because there may be a material source of revenue for the agent in other business units such as custody and foreign exchange. However, it might make sense for agents to grasp this nettle.”
Several factors have worked in combination to reduce loan demand. Equity bull markets have suppressed short selling activity and reduced demand for hot and warm securities. Use of synthetics, particularly total return swaps, has lowered demand through “traditional” lending markets, with prime brokers and hedge funds meeting a greater share of their borrowing requirements through synthetic channels. Improved technology and process management at borrower organisations have enhanced their ability to manage internal inventory and to internalise more of their stock borrowing requirements. As noted, there is also a rising appetite from some large beneficial owners to transact directly between each other through a P2P network such as that extended by the GPFA.
With a decline in the volume of specials, which generate high lending fees, this has reduced the appetite for borrowers to maintain high GC balances with agent lenders to secure preferential access to these higher value specials.
This economic pressure has been exacerbated as lenders have negotiated more aggressive fee splits against their indemnified lending activities. The report notes that in the early 2000s, the custodial agent lender might receive 25-30 per cent of the lending fee, against 70-75 per cent paid to the beneficial owner. Subsequently, this split has narrowed over the past decade — often to just a 10-15 per cent fee share for the agent lender, or sometimes even lower.
Recounting a conversation with a large global custodial lender, Faulkner explains that with roughly 10bps revenue for GC lending and a 10-15 per cent fee split, the agent is receiving 1-1.5bps for GC lending prior to costs and tax. “Given the cost of indemnification of 10.3bps in a Basel III world, one needs to seriously ask whether it’s worth doing,” comments the agent lender (p 20).
Value and volume
Noting the aggressive fee splits that have developed within some agency lending relationships, SFT asked Faulkner whether this could work to the advantage of large custodian lenders, with huge lending books, high automation levels and powerful optimisation engines — thereby driving consolidation in an heavily populated agency lending market and enabling them to win further market share at the expense of smaller competitors.
Faulkner politely rejects this proposal, indicating that the economics do not add up even for the highest-volume players. “It is doubtful whether this will work to their advantage,” he says. “If the agent is being compensated 1 to 1.5bps for GC lending and faces a capital charge of 10 to 13bps per loan transaction in providing indemnification, a higher loan volume simply results in a higher capital overhead. For a GC volume-based lender, automation is likely to compound this problem – higher balances mean higher regulatory capital expense.”
By contrast, he believes that some of the smaller players that position themselves as value lenders may be better placed to take advantage. Brown Brothers Harriman (BBH), a self-professed value-based lender, looks likely to be purchased by State Street for example, potentially bringing wider value-lending opportunities to the Boston-based bank through the acquisition. But integrating this business may present its own organisational and operational challenges, recognising the complexity of running a value-based lending business and heavily automated GC flow alongside each other via a common architecture.
Other major global custodian banks have been developing enhanced custody solutions and stepping further onto the prime brokerage domain, offering an integrated custody, securities lending, financing, clearing and settlement solution, in some cases with directed lending and peer-to-peer repo services available as part of the package. With peer-to-peer loan value rising through the GPFA, large lending agents may be interested to explore opportunities to apply their sophisticated machines and workflow to support P2P transactions, along with associated collateral management and custody services, thereby facilitating principal-to-principal trades which will not typically require indemnification. The convergence of enhanced custody and agency prime seems to be gathering momentum in the marketplace.
For the agent bank, this offers a much more attractive business proposition which will potentially free up a large capital overhead that can be reinvested into product innovation and into delivering further service enhancements to the client.
More broadly, Faulkner predicts that at some point the scale of agency lending activity — with outstanding lending balances close to US$2 trillion according to recent estimates — will force change and a major lending agent will make the “prudent and brave” decision to change their lending business models and pricing.
No doubt, this will trigger resistance and will require awkward initial conversations with some clients. This may prompt a reaction from some large prime brokerage counterparties, for example, that may decide to withdraw their business and reposition GC balances with other providers. A problem, Faulkner notes, is that things that are given away for free will tend to be consumed infinitely — and perhaps indefinitely — and that is a hard cycle to break.
But, alluding to the Pareto Principle (think 80:20), Mark Faulkner indicates that it is not essential to change the whole market from day one. By convincing 20 per cent of the market that has the greatest impact and influence, this can have a major effect on the industry’s direction of travel.
For asset owners, Faulkner proposes that larger beneficial owners should ask themselves whether they really need an indemnification, given the risks it protects them from. If they do, they might consider sourcing this from a third-party insurer which, as a specialist insurance provider, may be better positioned to craft an indemnification package that is tailored to beneficial owners’ specific needs.
Significantly, the regulatory environment applied to insurance business may allow third-party insurers to indemnify this risk at a lower regulatory capital cost than is applicable to banks under the Basel regime. With this in mind, some banks are also exploring avenues to transfer significant risk associated with securities lending indemnification as a means of reducing the regulatory capital overhead.
Mark Faulkner, Something Better Change: Securities Lending Indemnification is Unsustainable in its Current Form, Credit Benchmark, June 2022, can be accessed here.
