The OCC And I Work Hard To Guarantee Investment Returns


This is shaping up to be CFA week for me, and with my impending triumph/humiliation I've pretty much stopped thinking about much else. I've also stopped reading about much else, putting aside Trotsky temporarily to focus on those six stupid books. Yesterday was corporate finance - I can now unlever and relever betas like a champ - and portfolio management, which I got about halfway through before falling asleep. Today is equity and fixed income. The end is in sight!

But there's still occasionally time to think about blast-from-the-past favorite topics, like the slow-motion disaster that is the US regulatory effort to end official reliance on ratings agencies. The latest is the OCC, which released a proposed rule today that will change the definition of "investment grade" securities, which banks can invest in, from "rated in one of the four highest rating categories by two or more NRSROs" to this:

Investment grade means the issuer of a security has an adequate capacity to meet financial commitments under the security for the projected life of the asset or exposure. An issuer has an adequate capacity to meet financial commitments if the risk of default by the obligor is low and the full and timely repayment of principal and interest is expected.

How would you know? Well, the OCC offers some guidance:

The OCC expects national banks and Federal savings associations to conduct an appropriate level of due diligence to determine that an investment security is a permissible investment. This may include consideration of internal analyses, third party research and analytics including external credit ratings, internal risk ratings, default statistics, and other sources of information as appropriate for the particular security. The depth of the due diligence should be a function of the security’s credit quality, the complexity of the structure, and the size of the investment. The more complex a security’s structure is, the more credit-related due diligence an institution should perform, even when the credit quality is perceived to be very high. Bank management should ensure they understand the security’s structure and how the security will perform in different default environments, and should be particularly diligent when purchasing structured securities. The OCC expects national banks and Federal savings associations to consider a variety of factors relevant to the particular security when determining whether a security is a permissible and sound investment. The range and type of specific factors an institution should consider will vary depending on the particular type and nature of the securities. As a general matter, a national bank or Federal savings association will have a greater burden to support its determination if one factor is contradicted by a finding under another factor.

There's even a handy matrix of factors to consider for various types of bonds, like "Confirm risk of default is low and consistent with bonds of similar credit quality" (applicable to all bonds, and also, what?) and "Evaluate and understand the quality of the underwriting and of the underlying collateral as well as any risk concentrations" (applicable only to structured products and now you tell them). "External credit ratings," which I bolded up there, doesn't make the checklist.

Now a thing that I find charming about the CFA because it exists in an ideal world where everyone thinks about investment decisions from a variety of complementary perspectives, does a ton of research, recommends an approach to a client while describing its risks and tradeoffs clearly and appropriately, and then sits back and smiles wisely while the client, who in this scenario is some guy who just won the lottery and is wearing a bowling shirt and hasn't paid attention to anything you said, says "yeah put it all in those things you said were good, equities was it?"

In my experience, the advisory banking business actually involves a fair amount of exactly that. Investment bankers aren't particularly paid to be right. Your job really is to bring data and analysis and intelligence to bear in helping clients with their decisionmaking process. You listen to their problems, tell them a lot of facts, run numbers, make pretty charts, format the pitchbook beautifully and generally provide a range of alternatives while thoughtfully laying out the pros and cons of each. Then the client makes a decision, because that's their job.

I'm not sure that managing assets, at a hedge fund or mutual fund or, say, an OCC regulated bank, works that way. Being balanced and thoughtful and wise is not as good as making good decisions that get outsize returns. Investors do not like getting a thoughtful letter saying "here's why we were wrong" nearly as much as they like getting a postcard saying "Up 50% vs. S&P down 10%, see ya bitches."

In any case, in both advisory and decisionmaking businesses, there are times and places to think deep thoughts, and there are times and places for quick rules of thumb, bright-line cutoffs, and vaguely understood benchmarks that more or less work even though you don't understand them. Listing a bunch of factors, and saying that there is "a greater burden to support its determination if one factor is contradicted by a finding under another factor," is a pleasant way to sound responsible, but it doesn't provide much in the way of guidance. It washes out to "use your best judgment to find securities that won't default," which mostly works fine, until it doesn't. Though I guess the same is true of ratings.

But when you want to stand for quality and integrity - whether you're the CFA Institute or the regulatory agencies - the need for bright lines and rules of thumb can be hard to acknowledge. The CFA books don't mention rational ignorance. There's no chapter on "just use 12% for the WACC, good enough." There is a chapter making the wholesomely nerdy pitch for using NPV to evaluate projects, and then a sheepish admission that the less-delightful IRR (underdetermined with alternating inflows and outflows!) is more popular among actual managers and that the utterly dopey undiscounted payback period is pretty close.

Similarly, it is of course entirely sensible to ask banks to do credit diligence, think about whether the bonds they buy are likely to be repaid, and then only buy the ones where they can say with heart and mind "yes, they're good." And in a good and just world, they would do exactly that rather than just rely on ratings from agencies who hate them anyway. And, if they failed to do so, the perfect-world regulators would find out and yell at them and make them do better diligence.

But in the actual world, why on earth would you count on banks to confine themselves to the safest assets without any external metrics? And why would the OCC want to wade into a bank's books without any sort of handy, external, imperfect but roughly useful rule of thumb to evaluate their credit risk?

Probably because they wouldn't. I highlighted that "including external credit ratings" for a reason: something tells me that that's gonna get a lot of weight from both the banks and the OCC. I feel their pain. Like them, I'm going to spend the next week talking a good game about subjecting every investing decision to careful diligence and deep from-the-ground-up analysis. And, for me and for them, that impulse will eventually fade and I'll go back to bumbling along simplistically and hoping things work out okay.

OCC Proposes Rule to Remove References to Credit Ratings and Guidance on Due Diligence Requirements in Determining Whether Investment Securities Are Eligible for Investment [OCC]

Are Rating Firms Getting a Free Pass? [BW]