Back to basics – What is credit?

I work as a corporate credit analyst at M&G and my job is to assess the credit risk of a borrower, as measured by its expected loss. OK, that is the theory; but in practice, what makes credit possible to begin with? The answer is in the word itself: credit. And although I know the answer, I decided to test my new Copilot assistant.

“The English word ‘credit’ comes from the Latin word creditum, which means ‘a loan, or something entrusted to another’. It is the past participle of the Latin verb credere, meaning ‘to believe and to trust’. Therefore, the root concept of credit is based on trust and belief in someone’s ability to fulfil an obligation, often in a financial context.”

Well done, Copilot! Not only is it an accurate answer, but it is better worded than mine. So credit is possible when there is trust; without trust, there is no credit.

If this is the case, which strategies are deployed by issuers to build trust with bond investors? I see two main approaches: one that is more transactional and aims to get the best terms for the current transaction, potentially through selective disclosure; and one that is more relational and focused on building a track record of credibility.

For example, an issuer with a transactional approach might disclose a quarterly EBITDA of x, but it might omit that it was achieved thanks to a positive one-off factor contributing 20% of it. Investors might only discover this a year later when quarterly EBITDA is down year-on-year, because it has not been repeated and a narrative is required to justify the decline.

An issuer with a relational approach would instead disclose such one-off. Admittedly, issuers with a relational approach make the credit analyst’s job much easier.

As issuers can adopt different approaches to disclose information and build trust, credit analysts are faced with the challenge of discriminating between them and then adapting their analysis accordingly.

So how can we discriminate between the a transactional and a relational issuer? Some couple of examples below (not an exhaustive list):

1. adjusting for positive one-offs is a clear sign that the issuer is adopting a relational approach.
2. when results are particularly good, a relational issuer would stress that this is due to particularly favourable conditions that are unlikely to repeat, while a transactional issuer might try to convince you that this is the new normal.
3. a relational issuer, in my experience, makes very limited adjustments to EBITDA.

A legitimate, but difficult to answer question is whether there are more transactional or relational issuers in the market. Judging by the fact that EBITDA adjustments are invariably positive and very rarely negative, I’m tempted to conclude that most issuers adopt a transactional approach. If positive and negative adjustments are added back using similar criteria, one would expect to see both negative and positive adjustments, unless bad luck consistently beats good luck.

In addition, the temptation to adopt a transactional approach increases as credit quality decreases. Transparency is a luxury that not all issuers can afford.  

Can technology help distinguish between the two approaches? I asked Copilot if it could help and its honest answer was:

“I can’t make judgments about whether a borrower is being truthful, especially if it involves evaluating someone’s behaviour or intentions.”

While Copilot is currently unhelpful, technology might fill this gap in the future. In the meantime, the lack of a reliable and deployable lie detector makes the credit analyst job all the more interesting.