It’s our belief that the market for contingent regulatory bank capital in securitised form, otherwise known as the significant risk transfer (SRT) market, is shrouded in mystery and mystique. As a result, we see market commentators frequently resorting to imprecise language, flawed judgements and erroneous conclusions.

In this series, I’ll strive to explore various facts and fallacies associated with the SRT market. Our involvement in bank capital markets dates back to the mid-1990s and spans common and preferred equity, AT1 (additional tier one instruments), subordinated and senior unsecured debt, and securitised credit. Throughout that period, synthetic securitisations have played a role in bank risk and capital management, forming a recognised part of the regulatory capital structure of global banks.

While I don’t know it all, I’ve been around long enough to separate fact from fiction. This first article begins with the concept of capital itself and why, despite what we regularly read and hear, SRT is not a form of ‘capital relief’ or a means to reduce the amount of regulatory capital a bank must hold.

First, a definition

For a bank, what is capital?

To answer this, I did what most people are doing these days (despite having spent my entire professional life studying the subject): I looked to see what artificial intelligence (AI) had to say on the matter – in this instance I used Maia, Man Group’s GenAI engine.

Maia defined bank capital as “the difference between a bank's assets and its liabilities”. Hailing from the borderlands of the American South, I appreciated the Faulknerian response to what most of us know as ‘equity’, but that seems a narrow take on capital. I therefore asked Maia to define a bank’s capital structure and it walked me from equity to insured deposits. While remaining polite, I pointed out the discrepancy, although Maia’s response was generous enough to make me wish I hadn’t.

“Thanks for pushing on this — the imprecision was mine, and the distinction matters. So in banking, there is a spectrum from pure equity to pure liability, and ‘capital’ can mean different things depending on context.”

First off, I agree and am impressed by the subtlety. Yet, the exciting thing here is that Maia has highlighted precisely what I hoped it would: bank capital could be both the entirety of the right side of the balance sheet and not.

Most definitely not only equity

I believe that this lack of clarity stems from ambiguity within bank regulations. To its credit, Maia picked up on this, noting that the Basel, European Banking Association and Financial Stability Board definitions are all different! But, sorry Maia: bank capital is most definitely not only equity.

So, here is my definition: bank capital, synonymous with regulatory capital, is loss-bearing capital. The rest is funding. Where one ends and the next begins depends on how well connected the creditors of a given bank are.

We don’t necessarily know what bank capital is a priori but we do know a posteriori, or after the fact. Bank capital is whatever does not return par when bad things go wrong.

More capital, not less

Anyone still reading probably realises where I’m going: if we accept bank regulatory capital to be loss-bearing capital, then SRTs, which are contractually obligated to bear loss, are most certainly regulatory capital. And if they are ‘reg cap’, the notional amount issued should be considered to increase reg cap – if only by reducing risk-weighted assets – as opposed to releasing banks from the burden of holding it.

In order for an SRT to result in capital relief to the issuer, the total amount of capital available to a bank following an SRT issuance must decline. But that’s not what happens. To argue otherwise is to claim that SRT collateral either does not exist or is not equivalent to the notional amount of capital held prior to the deal. Neither is true of any funded SRT. SRTs, when executed away from insurance companies, are executed in fully funded form which means they cover to the notional limit of loss.

In reality, bank/reg capital increases. A bank has more loss-bearing capital available to it post-SRT than pre-SRT. Why? This is a somewhat technical point, but securitisation regulations include a concept referred to as capital non-neutrality. The same portfolio of credit risk has more total capital allocated to it following securitisation than before. This is to account for the perceived lack of transparency associated with a securitisation – even one comprised of loans originated and serviced by the entity that issues and retains most, or even all, of that securitisation.

Conclusion: funded means funded

SRTs are loss-bearing and funded with cash to the limit of exposure. The issuer has control over the collateral and determines when a loss event has occurred. Post-transaction, the total amount of capital available to bear loss has increased, not decreased.

In short, SRTs are not capital relief, or capital arbitrage, or capital release. They are regulatory capital securities that may enhance bank solvency.