Goldman Regulatory Advisers Were Too Friendly With Regulators

Being too close to regulators is the new way for banks to get in regulatory trouble.

The New York Fed looks secure, anyway.

Photographer: Daniel Acker/Bloomberg

The Federal Reserve Board is trying to bar a former Goldman Sachs banker named Joseph Jiampietro from the banking industry for misusing confidential information. Jiampietro is fighting this effort. "The Fed has the law wrong and the facts wrong," says his lawyer. But even if he wins, this case can't be great for his future in banking, not only because it's awkward to fight the Fed generally but also because the Fed published Jiampietro's bad performance reviews:

While Jiampietro’s regulatory practice was well regarded at Goldman Sachs, his supervisors had also identified issues with his performance. Jiampietro’s review for the performance year ending in 2013 indicated that his ratings were in the bottom 10% of managing directors, both regionally and globally. Jiampietro’s performance evaluations for that year also reflected that Jiampietro’s colleagues had concerns regarding Jiampietro’s handling of confidential client information. In addition, Jiampietro’s managers advised him that he needed increase the amount of revenue generating business he brought into the firm.

That is kind of cold. I feel like the bad performance-review ratings were not strictly necessary, from a legal perspective. (Disclosure: I used to work at Goldman, where my managers may have once or twice advised me that I needed to increase the amount of revenue-generating business I brought into the firm.) 

Alongside its case against Jiampietro, the Fed fined Goldman Sachs $36 million Wednesday for this misuse of confidential supervisory information. This comes on top of a $50 million fine that Goldman paid to the New York State Department of Financial Services last year. The two fines are for the same thing: A Goldman associate named Rohit Bansal, who used to work at the Federal Reserve Bank of New York, got confidential information about other banks from his friend who was still at the Fed, and he used it in his work of pitching and advising banks on their regulatory requirements. The Fed banned Bansal from banking last year, though it waited until now to fine Goldman and to go after Jiampietro, who was Bansal's supervisor.

Jiampietro and Bansal worked in the part of Goldman Sachs responsible for advising other banks on how to deal with regulatory issues, and the alleged misuse of confidential supervisory information ("CSI") was stuff like this:

In August 2014, Goldman Sachs advised Bank A on a presentation Bank A planned to make to regulators seeking approval to engage in an acquisition of another financial institution. In connection with that advice, Jiampietro requested and Bansal provided to him in hard copy CSI materials regarding Bank A’s past examinations, including first-day letters from examiners, supervisory assessments, and reports of examination prepared by examiners. Bank A decided to engage Goldman Sachs to assist with additional regulatory advisory work related to satisfying regulators’ recommendations from recent examinations. 

That is: Goldman learned what the New York Fed thought about Bank A by reading the New York Fed's secret files about Bank A, and then used that information to convince Bank A that it knew a lot about what the New York Fed thought about Bank A, and that Bank A should hire Goldman to make the New York Fed think nicer thoughts about Bank A. Bank A was, unsurprisingly, convinced. 1  And so, in the Fed's telling, Jiampietro was able to satisfy his bosses' demands that he bring in more paying business. 

We talked about this case back when the DFS fined Goldman. It is a weird one; it seems simultaneously terrible and trivial. Obviously you are not supposed to take the New York Fed's secret files about banks and use them for your personal gain. (There is a rule that says that, sort of. 2 ) But it's not like the banks were harmed: Goldman didn't use the information to undermine them, or to trade against them, or give it to their competitors. (Which are presumably the reasons that the Fed wants to keep its supervisory information confidential.) Instead, Goldman used the information to convince banks that it could give them good regulatory advice, and then to give them good regulatory advice. Informed regulatory advice seems like a good thing, no?

Now, informed regulatory advice can sometimes be a bad thing. The Fed keeps banks on their toes in part by not sharing its stress test models, which makes the tests harder to game; a regulator with a few tricks up its sleeve is a more effective regulator than one that is completely open with banks. And, sure, arguably Goldman's competitors were harmed by this, since they couldn't compete fairly to give regulatory advice when Goldman had the Fed's secret documents.

But those seem like pretty abstract concerns. The main harm that Goldman did in this case -- the justification for the $86 million in fines -- is that it embarrassed the New York Fed. When Goldman Sachs gets secret Fed documents that it's not supposed to have, and uses those documents to help other banks deal with the Fed, that sort of ... sends the wrong message about how bank regulation works. The regulators are not really supposed to be slipping the answers to the test to the banks that they regulate. And they're really not supposed to be slipping the answers to Goldman, so that Goldman can turn around and, in effect, re-sell them to those banks.

A lot of people worry that the regulators and the banks -- particularly the Fed and Goldman -- are too cozy with each other; it doesn't help when Goldman turns out to be making a profit off of Fed regulation. 3  (Also embarrassing: The problem was ultimately caught by Goldman, not the Fed. 4 ) There is a sort of meta-regulatory quality to this enforcement action: The point of all these regulators ganging up to fine Goldman is to wash away the bad taste of a regulator that showed Goldman illicit favoritism.

Another embarrassment here is that, unlike the frequent cases of half-literate traders manipulating Libor rates or foreign-exchange fixes because it never occurred to them not to, these regulatory violations were committed by the bankers whose job it was to help banks deal with regulators. (And who were themselves ex-regulators: Bansal was freshly arrived from the New York Fed, while Jiampietro was a former senior adviser to Federal Deposit Insurance Corporation Chairman Sheila Bair.) You'd kind of expect them to know better! There is some narrow sense in which they were really well positioned to advise their clients on regulatory matters: They had privileged knowledge about those matters, because of their secret Fed documents. But in a broader sense, the people stealing secret Fed documents are really not the people you want advising you on how to deal with the Fed. Like, if were advising you on how to deal with the Fed, my first piece of advice would be "don't steal secret Fed documents, come on." 5

But there is something fitting about that. We've talked recently about the argument that Wall Street has been "tamed," that higher capital requirements and lower compensation and increasing investments in compliance have truly and permanently forced the big banks out of their riskiest businesses and left them safer and nicer and better-behaved. A lot of the traditional ways for banks to get in trouble for being too aggressive -- by taking outsize proprietary trading risks, by building up huge positions without much capital against them, by lying to customers -- have been closed off. Banks are less dangerous and more utility-like, compliant creatures of regulation rather than masters of risk-taking id.

But the old habits of aggressiveness die hard. It's just a matter of applying them to compliance and regulation rather than securitization and bond trading. If the regulators have made it hard for banks to get in trouble in the old ways, it's up to the regulatory-relations bankers to find new ways to get in trouble. Goldman's regulatory-relations bankers were just a little ahead of everyone else.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
  1. Or:

    In August 2014, Jiampietro and Bansal worked on a PowerPoint presentation to Bank B to pitch regulatory advisory services related to stress testing, including ERM and MRM. In connection with that presentation, Jiampietro requested and Bansal provided to him CSI materials obtained from the Reserve Bank, including a 2013 MRM survey the Reserve Bank conducted of Bank A, and a 2013 stress testing survey the Reserve Bank conducted of Bank A. After receiving Bansal’s email containing the 2013 MRM Survey, Jiampietro replied to the email, directing Bansal and others to use it “as a guide for [Bank B].” The CSI materials were copied or paraphrased in the Bank B presentation, giving Jiampietro and Goldman Sachs a competitive advantage in soliciting Bank B’s business.

    That is: Goldman learned how the New York Fed was thinking about enterprise risk management and model risk management ("ERM" and "MRM") for Bank A, and used that information to convince Bank B that it knew a lot about how the Fed would think about ERM and MRM at Bank B. Which seems plausible enough -- you know, banks are banks -- though the Fed does not mention whether it worked. 

  2. Is there? The case against Jiampietro cites 12 C.F.R. 261.22(e), which says:

    All confidential supervisory information made available under this section shall remain the property of the Board. Any person in possession of such information shall not use or disclose such information for any purpose other than that authorized by the General Counsel of the Board without his or her prior written approval.

    But that applies to confidential supervisory information made available under that section, which is about disclosure made by the Federal Reserve Board in response to formal requests in litigation, etc. That's not what happened here. No one made a formal request for the information here; it was disclosed over a dinner at Peter Luger's. So that rule doesn't seem to apply. There's another rule -- 12 C.F.R. 261.20(g) -- prohibiting disclosure of information provided to banks, but again that doesn't seem quite right here. 

    Obviously there is something wrong with taking confidential information from the Fed and using it for your own purposes, but that's not quite the same as saying that there's a rule against it.

  3. By the way: not that much of a profit. Eamon Javers said on Twitter that the "unknowable but important question as always is how much did they make doing this? And what percentage of that is this settlement?" He's right that it's hard to know, but my guess is that Goldman's $86 million in fines represents more than 1,000 percent of the profits from its misconduct. Advising banks on regulatory issues just isn't that lucrative; it's not like a merger or underwriting assignment where the numbers involved are so huge that a multi-million-dollar banking fee is rounding error. In any case, the misconduct here seems to have involved about three years of work (from 2012 through 2014) by one apparently underperforming managing director, when revenue per managing director at many merger-focused firms was in the $5 million to $10 million range. Even if Jiampietro was producing at the high end of that range (unlikely, given his reviews and his focus on regulatory advice), and even if 100 percent of his revenues came from cheating (unlikely, given the few anecdotes in the enforcement actions), he probably brought in less over three years than Goldman paid in its Federal Reserve fine, and certainly less than it paid to the New York DFS.

    This is a more general point. People really believe that bank fines are just a cost of doing business, and normally far smaller than the profits that banks make from their misconduct. That may be true about crisis-era mortgage fraud -- it is hard to quantify -- but it's not true anymore. Now the fines seem to be an order of magnitude larger than the misconduct, as here, or as in the foreign-exchange-rigging cases, where some banks' fines for manipulation were bigger than their total foreign-exchange trading profits during the time of the manipulation.

  4. From the New York DFS case:

    On September 26, 2014, Goldman had an internal call regarding the calculation of certain asset ratios, during which there was disagreement over the appropriate method. During the call, the Associate circulated an internal New York Fed document – which the Associate had recently obtained from the New York Fed Employee – relating to the calculation, to the call participants, writing, “Pls keep confidential?” Following the group call, the Partner called the Associate to discuss the document, including where he had obtained it, and the Associate told him that he had obtained it from the New York Fed. The Partner then called the Global Head of IBD Compliance to report the matter and forwarded the document.

  5. I am also not the person you want advising you on how to deal with the Fed, to be fair.

To contact the author of this story:
Matt Levine at

To contact the editor responsible for this story:
Brooke Sample at

Before it's here, it's on the Bloomberg Terminal.