A former supervisory leader at the Office of the Comptroller of the Currency doesn’t see the flood of new charter activity threatening the banking system.
Rather, “the threats to the banking system really come from within,” and bank failures are usually tied to credit or lack of capital, said Mike Kirk, who joined financial services advisory and investing firm Klaros Group as a senior director in June.
“Often, competition can lead to a chase to a bomb,” Kirk said in a recent interview. “People trying to win the deal lower their underwriting standards.” Banks that fare best through cycles tend not to lower their standards for competitive reasons, he said.
Kirk spent about 20 years at the OCC, during which he was team lead for supervisory staff at JPMorgan Chase and Citi, where he was closely involved with the 2012 London Whale trading incident at JPMorgan and Citi’s 2020 consent order addressing risk and control issues.
Prior to that, Kirk spent about two decades in the industry, mostly in markets businesses.
He spoke with Banking Dive about staffing shifts at the OCC and how regulators think about progress when it comes to satisfying consent orders.
Editor’s note: This interview has been edited for clarity and brevity.
BANKING DIVE: What was your involvement with Citi’s consent order?
MIKE KIRK: Consent orders don't happen overnight in regulatory agencies. They take time to build, and the groundwork to that consent order began mid-2018. That started within the capital markets business, the supervision of which was run by me at the time. My team identified a large number of issues that spanned the organization, which is why eventually other teams got involved. We identified technology issues, governance issues, data sharing, data integrity issues, risk management concerns. I remember writing a supervisory letter that I brought to the examiner-in-charge, and he looked at me and said, “My God, we're trying to boil the ocean here. This is just too big. How could all this be?” At the time, their CAMELS ratings were satisfactory. I said, “Well, one of two things have to be true here: either all my work is wrong, or our CAMELS ratings are wrong. I'm going to guess it's the CAMELS ratings. And I would bet my reputation on it.” I was relatively new there. I took over the team in the second half of 2016.
We didn't have enough expertise in some areas, in the right seats, and we missed a lot of things for a long period of time, and the bank was in worse shape than we believed it to be. There was a lot of disbelief amongst a lot of people at the agency. We did a whole bunch of exam work on the back of this because of the evidence presented, other teams got involved. I was spearheading a lot of that effort, and eventually came to the epiphany, “Oh my God, this is much worse than we ever thought.”
We went step by step through the process and gave Citibank a number of opportunities to correct deficiencies, and when they failed to do so, we issued the consent order.
Do you think the bank is close to completing its required work?
There's always that tension between regulators and bankers as to progress, and what one party believes to be true versus what the other party believes to be true. To some extent, there is some negotiation involved – what is good enough? When you have something as large as a consent order, a bank may issue thousands of pages of plans to execute upon that, for regulatory approval. Despite all that, what you think would be a lot of granularity, there's a lot of open-ended questions that remain within there. Regulators don't necessarily approve a plan, but we may not object to it.
So the question then becomes, as they go down that path: Did they meet the spirit of that agreement? What were the unwritten aspects that we thought they were going to do versus what they thought they were going to do, and where we land on those things? There's no doubt Citibank is a much stronger institution from a risk management perspective, from what it was in 2018. They've made significant progress.
Whether they've made enough progress to close the consent order at this point in time remains to be seen. That depends on the facts on the ground, and I have confidence the agency will make the right decision.
What’s your take on regulators’ proposed overhaul of CAMELS?
I'm not a big fan of the change. The comptroller is correct that there is some overlap in the ratings for the individual components of CAMELS. I think that was done purposely. Let's say, for example, you're looking at the sensitivity factor, and the same group of management are not necessarily running every aspect of the bank. So those responsible for managing the sensitivity of the bank to interest rates are at a different level in the organization than your CEO and top bench level. They may have the best intentions to make changes to their risk management systems, procedures and processes, but can't get buy-in from the C-suite for the budget they need to do it. You could see a circumstance where you may want to ding management for that in the management rating, but not in the sensitivity rating, even though it's a risk management issue that you would think largely belongs in sensitivity.
By bifurcating things, you lose the dynamic that management can influence many of the other pillars within, and then who's responsible for what and how that should be reflected in the ratings.
With the Citibank example, one could argue that one of the reasons they ended up with that consent order is that certain executives didn't want to spend money on enhancements at certain points in time. Today, under the CAMELS proposal, that may not be represented properly or accurately in the M rating. I think it’s a mistake to go down that path.
What do you make of recent staffing shifts at the OCC, with some supervision staff working in chartering?
Licensing was sleepy for so many years, and now they've changed their public posture toward licensing. They've had a significant increase in volumes, which led to the increase in staffing that would be required to handle that. The Trump administration has looked to downsize the federal government and bank regulators in particular, and pursued changes to the supervisory cycle, so you need fewer examiners. So a lot of examiners are looking at alternatives and saying, “Where do I want to be right now in the OCC, if I want to not only have an interesting job, but also a safer job?”
A fair amount of people decided they wanted to work in licensing. Some came from supervision. When you're filling out an application now and submitting it to the OCC, the likelihood of having someone who actually does supervision work reading aspects, if not the entirety of that application, is so much higher now than it was under previous regimes. Bank examiners who have a penchant for detail are looking at these applications saying, “Yeah, so the legal requirements might be met here, but prudential risk management items are not being properly addressed.”
It’s a good thing that people are mobile within the agency in this environment. Anytime you move around within any organization, you bring a change in perspective. It adds a diversity of thought.
But the agency has lost experienced staffers, too.
That's the greatest travesty of the changes that have occurred over the last 18 months. A flood of talent has left the agency. A lot of people chose to retire earlier than they might have otherwise because they just don't like the environment they're working in. When you have that amount of intellectual capital leaving an agency, it's going to leave a void.
What I'm concerned about most is, who's going to lead that agency in a handful of years? Because so many people in the middle of the structure have walked out the door, and a lot of people at the top walked out the door, and those are usually people who mentor people.
I'm concerned about the agency’s ability to respond to a significant financial crisis. People like JPMorgan Chase CEO Jamie Dimon are 100% correct: There will be another financial crisis, and we’re probably due for one. They occur with some regularity, and it's been close to 20 years since the last one. You don't want to have a place where you don't have well-skilled people to respond to that.
