Brickwork has to shut shop

Or how (not) to run a credit rating agency

A brick wall being cleaned up (via Stable Diffusion)

If you’re reading this in your inbox, do forward this email to your friends who you think might enjoy it. If you’re reading this on Substack, do subscribe using the button below.

---

A classic, very well known conflict of interest in finance is that of a credit rating agency with their client. Here is roughly how it works:

1. A company needs money which it chooses to get by issuing a bond which is just a type of loan

2. The company can get this loan from investors. But before it does, there is a legal mandate for it to go to a credit rating agency. Said company will pay this agency and in return the agency will give its bond a rating after asking questions like “what is this money for” and “how will you repay it”

3. There are multiple credit rating agencies and all of them compete against each other to get the company’s business

The conflict here is obvious—a company issuing a bond wants a good rating. A rating agency that gives the company what it wants (a good rating) is more likely to get its business. This is an inherent conflict of interest that everyone inside finance has just learned to live with. Investors who buy the bonds know about this conflict. So do companies and rating agencies. As does SEBI, who’s the regulator that sets rules for everyone playing this game.

No regulator can get rid of this inherent conflict of interest. The goal instead is to set rules to get any stupidly obvious conflicts out of the way. A way to avoid one such stupidly obvious conflict if you’re a credit rating agency is that you have one person handle the “ratings” part, which is your actual job, and another person handle the “selling” part, which is essential to making more money. Also make sure they don’t talk to each other. More importantly—make sure that they meet the client separately. Once to discuss business and the other time to discuss ratings. They should never be in the same room together with the client.

Anyway, last month, SEBI asked credit rating agency Brickwork Ratings to please shut shop. The regulator has a nice table with 36 reasons in its order¹. Individually, they attracted warnings and fines. Together, SEBI decided that they were serious enough to warrant curtains for Brickwork.

I can’t obviously write about all the 36 Reasons Why SEBI asked Brickwork to wind up. And really most of them are similar to each other just with different clients. So I’m going to write about 2 of the reasons that I thought were the most fun.

The first one, of course, was that Brickwork’s Head of Ratings (ratings person) and Head of Business Development (selling person) were found in the same room with IDFC First bank (the client whose bonds Brickwork was rating)

Here’s what happened:

1. Some years ago Brickwork rated some bonds of Capital First, an NBFC

2. Capital First merged with IDFC Bank. The combined entity became IDFC First Bank

3. Now, Capital First was Brickwork’s client and it had rated its bonds or other securities. But Capital First no longer existed. IDFC First Bank did

4. So there was “ratings” work to do (do the Capital First ratings apply to this new entity?) as well as some “selling” work to do (does the new entity even hire Brickwork anymore?)

5. 4 people were in a meeting to sort things out. From IDFC First Bank: the CEO. From Brickwork: the Head of Ratings, Head of Business Development, and the Founder.

This is bad! I can imagine them in this meeting with the Business Development guy being the best salesperson trying to woo IDFC First’s CEO with something over the top—maybe an ML-infused rating on a blockchain?—but the CEO interested in only what the Ratings guy has to say. And what was the founder of Brickwork even doing there? I don’t even know. Going by Brickwork’s response, the founder was there for a “courtesy call” with the new CEO.

One aspect of corporate finance that’s often forgotten by those outside the industry is that a large part of finance is relationship-building. If your client’s company has just merged with a bank you’d probably want to meet them once to congratulate them or whatever. It’s just good business. SEBI doesn’t disallow it. But it's important to make sure that these “courtesy” visits don’t happen alongside the heads of Ratings and Biz Development².

Prediction but also recording
As a credit rating agency, one part of your business is that of predicting. This is hard. The world is volatile, life is volatile, and predicting the future is hard. When a rating agency gives a bond an “AAA” rating the agency is essentially making a prediction that the company won’t default on its loans. It justifies this prediction by saying that this company is financially healthy, has a decent business model, whatever they need the money for makes sense, etc. but the rating is still a prediction.

But another part of your business is that of recording events in real time as they happen. What do you do if you give a company an AAA rating and it defaults? Then you’re no longer in the business of predicting (you failed) but in the business of recording. You quickly go back to your rating and revise it and give it a nice “D”. Your prediction may have been crap but at least you’ve learned from it. This isn’t so hard, unlike predicting.

Well.. Brickwork didn’t do well at recording. On 14 August 2019, Sintex Plastics informed the world that it had defaulted on its loans. Brickwork, who had rated Sintex with an “A”³ , took 7 days to revise its rating. According to SEBI's rules, it had to do this within 2 days.

Something to be clear here is that from an investor standpoint it wouldn’t have mattered a single bit had Brickwork revised its rating within the mandated 2 days. Once a piece of information is public, sophisticated investors aren’t going to wait for a rating agency to update its rating, they have their own software and algorithms and Bloomberg Terminals to deal with precisely such situations.

But a bond default is huge! Being in the ratings business and failing to quickly recognise a bond default is the equivalent of being a sports journalist and forgetting to write about the World Cup. Everyone knows the results as they happen, but it’s still important to write about them!

Brickwork wasn’t late just with Sintex. According to SEBI, it did this with at least six other companies. One might imagine that if you were in the ratings business you’d likely have software of your own that would alert you if one of your clients defaults (investors certainly do!). Maybe you’re bad with tech. In which case you’d set up a Google Alert. But if you really wanted to do your job well, you’d not just wait for defaults to become news. You’d probably put on your calendar something like “check if Sintex has paid its bond interest which was due today” and you’d be able to revise your credit rating before this was public. Then your rating revision might even be useful for investors!

Back from the dead?
Brickwork might still survive though. As one would expect, the rating agency contested SEBI’s order in the SAT (Securities Appellate Tribunal or the place where companies fight it out with SEBI) and has some time to make its case and hope that the SAT rules in its favour.

I don’t really want to get into the merits of SEBI’s order and Brickwork’s argument that the punishment is disproportionate. Brickwork messed up, but the mess ups cited are also at least 3 years old now, but then everything is slow in India so who knows what’s appropriate.

But let’s assume that Brickwork wins against SEBI and survives. There is no way things can be business-as-usual.

If Brickwork ends up surviving I think the utility of its rating would be reversed. Usually, a favourable rating by a credit rating agency reflects well on the company that has been rated. But if you’re an investor looking to invest in a bond that has been rated by a rating agency that almost shut shop because the regulator thought it was incompetent and had 36 reasons for it—that rating wouldn’t help you pick which bonds to invest in but would certainly help you pick which ones not to.

---

If you enjoyed reading this post share it with your friends on WhatsApp! Here’s an easy button to do it:

1. If you’re really interested in checking out the table go to page #8.

2. A fun aspect of this whole thing is that SEBI found out about this meeting because Brickwork gave IDFC First Bank a discount on its agreed pricing and there was no documentation of any meeting of a price-related discussion. That’s how it turned out that the meeting had happened, the problem was that it happened alongside the ratings meeting.

3. “A” being the third-highest rating on Brickwork’s ratings scale.