The SEC's press release touting its triumph over rebel-without-a-cause rating agency Egan-Jones gives just the slightest impression that it was written in embarrassment. A trope of SEC press releases is "[thing we are enforcing] is among the most important things in the whole wide universe"; this is hard to say with a straight face when the defendant is guilty "essentially, of filling out forms wrong," as Jesse Eisinger put it last year. But two SEC enforcement bigwigs give it their best shot:
“Accuracy and transparency in the registration process are essential to the Commission’s oversight of credit rating agencies,” said Robert Khuzami, Director of the SEC’s Division of Enforcement. “EJR and Egan’s misrepresentation of the firm’s actual experience rating issuers of asset-backed and government securities is a serious violation that undercuts the integrity of the SEC’s NRSRO registration process.”
Antonia Chion, Associate Director of the SEC’s Division of Enforcement, added, “Provisions requiring NRSROs to retain certain records and address conflicts of interest are central to the SEC’s oversight of credit rating agencies. EJR’s violations of these provisions were significant and recurring.”
To be clear what happened in the Egan-Jones case was, as we've discussed before:
- In 2008, Egan-Jones told the SEC “we have issued 200 ratings and they are on the internet.”
- A few months later, Egan-Jones corrected the number of ABS and muni ratings from 200 to 23.
- The correct number was actually zero, as you could tell by looking at E-J's website.
- Four years later, the SEC noticed.
"'Accuracy and transparency in the registration process are essential to the Commission’s oversight of credit rating agencies,' said Robert Khuzami, Director of the SEC’s Division of Enforcement," though not so essential that the SEC would get around to noticing admitted inaccuracy inside of four years.
So, I mean: don't fill out forms wrong! It's bad. Everything bad is bad. Some bad things are more bad than other bad things. Only in SEC press releases are all bad things equally "essential" to this and "central" to that. Egan-Jones look like dopes here, and arguably they look like dopes elsewhere, but their dopiness here is not only very, very, very small-time,1 it is also pretty unrelated to the job that they did - rating ABS and muni issues - and that the SEC has sort-of banned them from doing for 18 months.2
Wait didn't someone else run into some trouble with ABS ratings, once? It's worth contrasting Egan-Jones's problem - wrote wrong thing on form - with, say, S&P's behavior regarding CPDOs. I wrote about eight thousand words about that a while back, and I was barely scratching the surface of a four million page Australian court ruling, so don't expect me to summarize it here, but it had sentences like "Anyway, S&P used a long-term value of 80 for the credit spread – which made the product look better than would a lower spread, despite the fact that ABN Amro was apparently arguing for a lower spread."
Which is to say it was a little more complicated than "they wrote on a form that they'd done 200 ratings but actually they'd done zero, as you could tell if you, like, looked."
In financial markets, regulation, life, etc., it's useful to keep in mind that the smarter often prey on the dumber, but it's also useful to keep in mind that there are many such unequal relationships and being the smarter predator in one of them doesn't stop you from being the dumber prey in another. The CPDO case is among other things - and there's a lot going on - but among other things it's about bankers, some of them former S&P employees, ruthlessly exploiting stupidity in S&P's structured credit ratings process. "S&P rates based solely on probability of loss, and not severity? Okay then, we will build them a martingale with a 0.7% probability of 100% loss." It seems clear - again from the lengthy court opinion - that S&P had no real idea what was going on or how they were being exploited; it seems equally clear that the product was ultimately sold to people who understood even less.
You might usefully question how much badness flows the other way: smart, or adequately smart, people being victimized by their reliance on idiots. Some, surely: Ezra Merkin appears to be a genuine moron whom smarter people trusted with their money, and you could shoehorn Dragon Systems' bankers into that category though I probably wouldn't. As a wild guess, though, I'd say that's less important economically - and it's surely a less important enforcement priority for an anti-fraud-ish agency than is the victimizing of the dumb by the smart.
So you might imagine that the SEC would spend its time trying to catch very smart people who use very complex processes to exploit less smart - though perhaps still quite smart - people.
Do you see the problem with that?
It's hard! You have to be really smart! Smarter than the smart people who are in turn smarter than other smart people! (Just because you are the regulator doesn't mean you are immune from the smartness-and-exploitation food chain; if you are dumb, you get exploited.) You have to sift through the complex processes to, as it were, get the joke, and then figure out how to unravel it. You have to synthesize immense amounts of information and apply market knowledge and judgment. You can't look for a smoking gun, because there is no smoking gun; smoking guns are for amateurs. There's certainly not a smoking gun of the form "we wrote the wrong number of ratings on our application, as you could tell from our website."
With hindsight, would you rather have bought a muni bond rated AAA by Egan-Jones, or a CPDO rated AAA by S&P?
We talked last week about the SEC's laser-like focus on insider trading, the millions of documents and phone calls it will review to catch an insider trader. I worried then that the skill set developed and rewarded by the SEC is not necessarily well suited to financial regulation; that finding needles in haystacks develops skills - data-mining, smoking-gun-hunting - that are sort of the opposite of the big-picture synthesizing that would characterize an ideal financial regulator. The application of that thesis to the SEC's ratings-agency enforcement actions - four years investigating whether Egan-Jones filled out a form wrong, nothing on the big agencies' structured-credit ratings - is left as an extremely easy exercise for the reader.
Though maybe the explanation is even simpler. Insider trading is dumb.3 Putting obvious wild inaccuracies on SEC forms is dumb. Selling CPDOs is smart. (Buying them is dumb.) You can build a pretty good model of SEC enforcement from those facts alone.
Egan-Jones Barred for 18 Months on Some Ratings [DealBook]
Taking On the Little Guy, but Missing the Bigger Ones [Dealbook / ProPublica, May 2012]
Egan-Jones and Founder Sean Egan Agree to 18-Month Bars from Rating Asset-Backed and Government Securities Issuers as NRSRO [SEC]
1.To me the most telling thing about the SEC's order here is the lengthy discussion of the mechanics of how Egan-Jones has to pay its $30,000 fine:
Payment shall be made in the following installments: (1) Respondents shall pay $15,000 within fourteen (14) days of the entry of this Order; and (2) Respondents shall pay $15,000 within sixty (60) days of the entry of the Order, as well as post-judgment interest on the second installment. If timely payment is not made, additional interest shall accrue pursuant to 31 U.S.C. § 3717. Payment must be made in one of the following ways: [etc., but let's just say no personal checks].
When the bad guy you've caught can't afford to pay a $30,000 fine without splitting it into two easy installments, it's safe to say that he's not that big a bad guy.
Egan-Jones, which also agreed to correct any deficiencies and submit a report describing those steps, can still rate asset-backed and government securities issuers during the ban — just not with the government blessing that confers special authority on its opinions.
The SEC order says they can keep rating ABS and muni securities as long as those ratings are on a separate web page prominently disclosing that "such ratings are not issued or maintained by a registered NRSRO," though of course that is literally false as they are issued by Egan-Jones, which remains a registered NRSRO, but whatever. The point is: various stuff - Basel LCR, money market fund rules, bank capital, etc. - depend on getting at least [some number of] NRSRO ratings of at least [___]. Do a lot of people count (after all investor-paid) Egan-Jones ratings toward those requirements? In other words, will this cost Egan-Jones business? My structural guess would be no - if most investors in Issue X require at least Y NRSRO ratings, then the issuer is gonna go get at least Y ratings from issuer-paid NRSROs so that every investor can buy - so I'm not too worried? Presumably people mostly get the E-J ratings extracurricularly, as credit advice rather than regulatory box-checking? But I guess at least some ABS is pretty bespoke so maybe a bilateral deal would count an E-J rating that now no longer counts? Anyway, more informed speculation would be welcome.
3.Oh it just is! Do I really need proof of that? Go fit a curve to this chart and compare it to your utility curve; I for one would not do 2.5 years in jail for $1 million. Plus you gotta give back the million. There's a probability weighting but still.
Photo of cow from Brian Furbush, http://www.brianfurbush.com/blog