When Wirecard went belly up a few years ago, Deutsche Bank ended up with a loss of just €18mn — miraculously little for a bank that had up until then made a habit of ambling into nearly every major financial cow pie in the world.
And this had been a giant pile of manure right on its own doorstep. Deutsche had previously underwritten Wirecard bonds, arranged loans for the company, and handed its chief executive a giant margin loan. Fellow German lender Commerzbank took a €175mn hit.
How did Deutsche manage to avoid this doo-doo? FT Alphaville gathers that it was probably at least partly thanks to something known as a “synthetic risk transfer” — one of the hottest bits of high-octane financial engineering these days. Deutsche Bank declined to comment.
In SRTs, a bank offloads some or all of the risks of some of its loans to ease how much capital it has to set aside for regulatory purposes. The loans remain on the bank’s balance sheet, but the buyer of an SRT typically promises to cover a chunk of the losses if the loans go bad. The buyers are investors such as insurance companies, hedge funds and (increasingly) private credit funds, which take on the risk in exchange for a fee.
In a sign that SRT has finally arrived in the big leagues, the exotic form of securitisation received the IMF Global Financial Stability Report treatment in October. In a breakout box the fund’s analysts said (FTAV’s emphasis below):
An increasing number of banks around the world have begun using synthetic risk transfers (SRTs) to manage credit risk and lower capital requirements. SRTs move the credit risks associated with a pool of assets from banks to investors through a financial guarantee or credit-linked notes while keeping the loans on banks’ balance sheets. Through this credit protection, banks can effectively claim capital relief and reduce regulatory capital charges.
However, the transactions can generate risks to financial stability that need to be assessed and monitored. Globally, more than $1.1 trillion in assets have been synthetically securitised since 2016, of which almost two-thirds were in Europe.
In the United States, activity picked up in 2023 and is expected to accelerate further because the regulatory landscape has become clearer. In Europe, corporate and small- and medium-enterprise lending, a well known and stable loan category for investors, backs up most of the issuance; recent transactions in the United States have centred on retail loans, particularly automobile loans. In Europe, issuers of SRTs include global systemically important banks and large banks, whereas in the United States, regional banks issue SRTs as well.
As the IMF report hints at, SRTs sit at the heart of the burgeoning private credit-bank nexus. In fact, while they are on some levels competitors, SRTs are central to how they can often work together, and help explain why we’re seeing more tie-ups between the $3tn private credit industry and the traditional banking industry.
But what exactly does this financial engineering entail, how long has it been going on, why has it recently exploded, and just how dangerous is this phenomenon — if at all? FTAV thought we’d poke around to find out a bit more about the latest booming three-letter acronym.
The evolution of SRTs
There’s an old saying in Norwegian that “a dear child has many names”. In reality, synthetic risk transfers are not quite as novel as the fairly new moniker may imply.
Synthetic risk transfers are often also called “significant risk transfers”, but both terms are iterations of what used to be called CRT — for “credit risk transfer” or “capital relief transactions”. Back in the halcyon early 2010s they were usually called “regulatory capital trades”, or “capital relief trades”.
Even before that, the phenomenon manifested itself as balance sheet securitisations or synthetic collateralised loan obligations, or even arguably JPMorgan’s pioneering Bistro deals, as Tracy Alloway’s brilliant exploration of the etymology of SRTs noted:
It’s a truism of Wall Street that you can’t sell something without first figuring out what to call it. Making money requires catchy titles and acronyms that are ideally vague enough to convey importance and desirability without necessarily explaining exactly what’s being sold.
. . . Perhaps a better name would be “synthetic risk management for the purposes of regulatory relief and sometimes arbitrage, the true value of which must be determined by premiums paid, with real risk transfer only guaranteed if it doesn’t come back to the banks in some way”, or SRMPRRSATVMBDPPRRTOGDCBSW for short.
Here’s how the basic structure of a modern synthetic risk transfer works. This is a hypothetical example so the details will vary, but hopefully this will give a general idea of how they function.
Let’s say Banque Alphaville has tonnes of corporate loans, mortgages and credit card debts on its balance sheet, but the regulatory capital that it needs to set aside to insure against losses is a pain in the arse. Most importantly for the good people that run Banque Alphaville, they affect the bank’s profitability metrics, and thus our their bonuses.
The bank therefore selects €1bn of reasonably high-quality corporate loans it carries on its book as a “reference pool”. It then finds a club of investors led by Generic Greek God Capital that will for a fee insure against the first 10 per cent of any losses on that pool, known as the “junior tranche”.
In other words, GGG Capital will cover the first €100mn of losses in return for periodic payments of, say, 3 to 10 percentage points over a benchmark interest rate (with the level decided by the riskiness of the underlying loans).
A fund manager will typically have to pony up the full amount insured to a third-party custodian — the deposit account in the diagram above — but bigger and stronger counterparties (such as an actual insurance company) might do an “unfunded” deal.
The advantage for GGG Capital is that it can harvest returns of typically 10 to 15 per cent without much work (beyond the initial due diligence on the loan pool) The loans remain on the Banque Alphaville balance sheet, so it does the ongoing work of monitoring the borrowers. And if they go bad, Banque Alphaville has to handle the actual clean-up, since they’re still on its balance sheet. GGG Capital is just there to reimburse the bank for losses (up to a point).
For Banque Alphaville, the advantage is (if the structure passes muster as a “true” risk transfer) that regulators will then require less capital to be set aside for the loans.
To comply with risk retention rules, Banque Alphaville will have to keep a slice of its own SRT, but for simplicity let’s set that aside for now. A €1bn loan pool might require capital of €105mn before any SRT deal — assuming full risk-weighting and a tier one capital requirement of 10.5 per cent — but with GGG Capital on the hook for the first €100mn of losses it might only need to set aside about €15mn of capital.
That allows Banque Alphaville to make more loans, return money to investors or simply improve its capital ratios and de-risk itself. In other words, it looks like a leaner, meaner and better bank, thanks to some clever behind-the-scenes engineering.
Europe leads the way
Mario Draghi might bemoan Europe’s woeful securitisation markets, but SRTs is one area where the old continent has clearly led the way (yes, despite the “Wall Street” headline).
In fact, the concept of such deals was enshrined in the EU’s regulatory framework through its implementation of the Basel II banking rule book in 2006, and in 2014 the European Banking Authority laid out detailed guidelines for them. As the EBA said at the time:
The Basel II capital framework recognises that credit risk transfer techniques can significantly reduce credit risk to which institutions are exposed and recognises that the credit risk transfer can be an effective risk management tool. The framework establishes that where credit risk transfers are direct, explicit, irrevocable and unconditional, and supervisors are satisfied that banks fulfil certain minimum operational conditions relating to risk management processes, banks may take account of such credit risk transfer in calculating own funds requirements.
As a result, European banks now account for roughly two-thirds of the $1.1tn of SRT deals tallied by the IMF, with heavy users including the likes of Deutsche Bank, Barclays, BNP Paribas, Santander and Credit Suisse (before it went kaput), plus a host of smaller lenders across the continent.
Most of the loans referenced in SRT pools are fairly solid loans to smaller and mid-sized companies, but personal loans and dicier corporate loans are also getting packaged up more and more often. A recent S&P Global report measured activity by the size and type of retained SRT slices kept by various banks on their balance sheets:
In fact, this has now become such a large, established and vibrant market — and pricing so cheap because of the money flooding into the deals — that SRT specialists argue that the rare refuseniks stand out from the crowd. As one told FTAV:
It’s gotten to the stage where it’s weird for banks not to do it . . . It’s a great technology that allows banks to free up capital.
Even the European Systemic Risk Board now says that “the importance of the SRT securitisation market for European banks and the European economy cannot be overstated”. But some people reckon the days of this rare area of European financial dominance will soon fade.
The Americans are coming
On September 28 2023, the Federal Reserve posted a seemingly innocuous legal interpretation titled “Frequently Asked Questions about Regulation Q”, with the even more off-putting subtitle “Capital Adequacy of Bank Holding Companies, Savings and Loan Holding Companies, and State Member Banks”.
It was an unlikely banger. The guidance constituted a de facto Fed blessing for the capital relief that SRTs can offer, and gave the US market a massive jolt.
Since then the likes of JPMorgan, Goldman Sachs, and Morgan Stanley have all jumped in, and Pimco predicts that by the end of next year US issuance of SRTs will probably match the volume of issuance in Europe.
The American SRT market will look a bit different, however. While there has been a spate of high-profile deals from blue-chip banks (well, as high-profile as anything is in this fairly opaque market), most people expect US issuance to be dominated by regional banks, and thus evolve differently.
Here’s KKR’s explanation:
In the United States, banks have tactically mobilised SRTs to free up capital and liquidity on the heels of the regional banking crisis in 2023, shedding exposure to long duration fixed rate assets, to manage the risk of deposit flight. These SRTs are typically in the form of cash securitizations.
In Europe, most European banks are using SRTs strategically to optimise their use of capital in the face of ongoing capital constraints that constrain their ability to provide credit to core customer segments. Contrary to the United States, which has thousands of regional banks, the banking sector in Europe is smaller, more concentrated, and predominantly served by large banks. This makes it worthwhile for banks as well as investors to treat European bank-sponsored SRTs as a sizeable opportunity.
The X factor in the US is what happens with the Basel III “Endgame” rules.
As Dan Davies noted last month on FTAV, Donald Trump winning the US presidency casts a big fat shadow over the planned implementation of the final details of post-crisis bank regulations. If it gets scrapped altogether, a lot of the capital pressures on many US banks will dissipate and lessen the need for SRTs.
However, most people still expect the Endgame rules to be implemented in some form or fashion. Even if they are dramatically watered down then many American lenders will still need to improve their capital ratios. As KKR wrote in May, long before the US election:
We believe banks will continue to seek out ways to de-risk and optimise their balance sheets even if the final capital requirements are “looser” than previously anticipated. Banks have been under continued pressure to reduce capital associated with consumer, mortgage, and commercial and industrial loans, and we do not see that pressure abating. Based on estimated bank exposures as of year-end 2023, even a 1% reduction in exposure to these areas could result in some $7 billion of SRTs through CLN issuance.
We also note that raising equity, another option banks could evaluate to improve capital ratios, is both dilutive and expensive in an elevated-rate environment. SRTs allow banks to obtain regulatory capital relief and improve capital ratios without diluting shareholders.
Private credit barrels in
Historically, the main buyers of SRTs were pension funds and a smattering of credit-focused hedge funds and asset managers. Lately, insurance companies have become more meaningful, but the biggest change is the swelling involvement of private credit funds.
The IMF estimates that private credit funds are now the “dominant buyers” with a market share of more than 60 per cent, but this seems to be a misreading of the ESRB’s earlier report (HT MainFT’s Cat Rutter Pooley for the spot). Here are its rough estimates for the SRT buyer base as of June 2023:
As the ESRB noted:
The profile of the demand side is clearly composed of professional and sophisticated credit portfolio managers. The survey also indicates that the number of investors is steadily growing in almost all categories. Market participation is limited to professional investors who appear prepared to understand and sustain the risks associated with these structures.
However, people close to the market say it has become noticeable that private credit funds — overflowing with cash thanks to the hotness of the asset class but finding it tougher to source attractive deals — are becoming increasingly major SRT players. As one private credit executive told FTAV: “People just have to put the money to work.” Another admitted that it was akin to a “private credit arms race”.
That’s usually not a great recipe for diligent credit work or appropriate risk compensation.
Alphaville gathers that investors could help shape the reference portfolio being insured when the SRT market was still developing — stipulating the bank chuck out certain dicier exposures, or limiting exposures to entire sectors — but as it has grown hotter it has become more “take it or leave it”.
However, the real worry obviously isn’t that some private credit funds might not see the returns they’re hoping for. They’re staffed by adults, and their investors are overwhelmingly big institutional investors. Caveat emptor etc.
The danger that some people fret about is that SRTs are actually making the financial system shakier — the opposite of the intended purpose.
Recycling risk
There are many facets to these worries. The IMF has a good summary in its last Global Financial Stability Report:
First, SRTs may elevate interconnectedness and create negative feedback loops during stress. For instance, there is anecdotal evidence that banks are providing leverage for credit funds to buy credit-linked notes issued by other banks. From a financial system perspective, such structures retain substantial risk within the banking system but with lower capital coverage.
The magnitude of the interconnections is difficult to assess because the market remains opaque, with only a fraction of deals being made public and no centralised repository for data on SRTs.
Second, SRTs may mask banks’ degree of resilience because they may increase a bank’s regulatory capital ratio while its overall capital level remains unchanged. Increased use of SRTs may reflect inability to build capital organically because of weaker fundamentals and profitability performance.
Furthermore, overreliance on SRTs exposes banks to business challenges should liquidity from the SRT market dry up. Currently, the asset pools being securitised seem to be of higher quality; however, there are signs of increased concerns regarding deterioration of asset quality.
Financial innovation may lead to securitisation of riskier asset pools, challenging banks with less sophisticated tools for risk management, because some more complex products make the identity of the ultimate risk holder less clear.
Finally, although lower capital charges at a bank level are reasonable, given the risk transfer, cross-sector regulatory arbitrage may reduce capital buffers in the broad financial system while overall risks remain largely unchanged. Financial sector supervisors need to closely monitor these risks and ensure the necessary transparency regarding the SRTs and their impact on banks’ regulatory capital.
This isn’t just vague hand-wringing by an institution whose job it is to semi-annually wring hands over lots of random things that could go wrong. There are credit specialists who think things are getting a bit frothy.
Just last week, Pimco’s Kristofer Kraus, co-head of the asset manager’s asset-backed finance business, said that “it is important to approach certain sectors with caution given significant capital formation or hidden risks that have yet to be tested”, specifically citing SRT as an example.
SRT people are adamant that the mechanism does actually make the financial system safer, by shifting risks out of banks and on to investors. They point to situations like Wirecard-Deutsche Bank as a prime example of the benefits. Even if many SRT investors nowadays use leverage to juice returns, they’re still a lot less leveraged than banks are. And if they go bust then the fallout is much smaller.
That’s all true. Most of all, it’s still a modest market, so it’s hard to get overly worked up about it. For now, the right way for regulators to handle this is probably mostly to sternly remind banks of existing guidelines around wrong-way risks.
However, it’s not inconceivable that standards slip and some structures end up not working entirely as envisaged. At some point, sloppiness always creeps into a hot market. Eventually, sloppiness morphs into outright stupidity.
Most of all, if the risks just keep getting recycled back into banks through new loans to private credit firms, hedge funds and asset managers then it just becomes leverage upon leverage, all the way down.
Source link
