Towards a theory of co-headship

A detailed taxonomy of management company rationales for this weird Wall Street anomaly

“Are you equal co-heads?”

Dan Davies is a managing director at Frontline Analysts. He is the author of the upcoming book The Unaccountability Machine, and writes an annoyingly good Substack when he should be writing more for Alphaville.

Of all the many curious things about investment banking, the titles it gives people are not the least. “Managing directors” who are not directors of any company and don’t manage anything. “Vice-presidents” who carry bags and take minutes. “Executive directors” who are mid-level bureaucrats. But perhaps the weirdest of all is the profusion of “co-heads”.

Since the MD and VP titles are almost meaningless, you might think that global co-head is also either a vanity title or a piece of marketing. That’s not quite true. “Head of X”, in investment banking contexts, is often a vanity title; it just indicates there’s nobody more senior in the X business for that bank, which might not mean much. However, “co-head” at least implies that being head of X is sufficiently important that more than one person wanted it. And banks love their co-head structures. Or at least, they love to hate them; Citigroup is currently in the middle of a big project to remove layers of management, and Goldman Sachs recently had quite a nasty internal conflict partly driven by a situation in which one co-head was appointed to an important committee and the other wasn’t. Craig Coben has detailed the carefully calibrated decorum required for it to function (ish)

In general, everyone knows that split responsibility is an incipient managerial problem waiting to happen. But they keep being recreated. Why? The answer is that there are several different reasons, because there are a variety of problems which a bank might hope to solve with a co-head structure.

Below FT Alphaville presents a taxonomy of the most common ones, each named after the phrase which someone has undoubtedly spoken at the meeting where the co-head structure was agreed, and each rated for its potential to cause a massive meltdown at some point in the future…

“I absolutely will not report to that little squirt”

The division has a big revenue producer with an out of control ego. He or she has to be kept happy because of the money they bring in, but nobody in their right mind would put someone like this in a management role. There’s another more logical promotion to be overall head, but it’s someone who is younger or less experienced than the Big Dog. Hence, a titular co-head role.

Meltdown potential: Quite high.

The Big Dog doesn’t really want to get involved in management, but that won’t stop them from second-guessing and undermining the other co-head when the fancy strikes. Other rainmakers will follow the example, and their juniors will take the cultural lead from the seniors, meaning that the person doing the work of running the division gets little or no respect and is stereotyped as a boring admin drone, probably a drag on the division. This is a hardship posting for professional managers; all they can do is wait it out and hope for that a compliance scandal or major cardiac event gets rid of the Big Dog.

“Don’t worry, the job’s yours, we just need to do this for internal reasons”

The bank has recruited a high flyer from somewhere else to be the division head. But they’re a little bit worried about cultural fit, plus there’s a lingering suspicion from the management committee that if the new recruit was so great, they would have been promoted at their old place. So they set up a co-head structure; the high[1]flyer is put alongside a safe pair of hands, usually a “lifer” from inside the firm who is quite near retirement age.

Meltdown potential: Low.

This model has a limited lifespan built into its design; either the high-flyer will demonstrate capability to be sole head or they won’t. If they don’t they’ll get fired; if they do, then the incumbent co-head can gracefully retire or move into a less demanding job. When you see an investment banking division that has multiple “chairmen”, the reason is often that they’re the natural outcome of this kind of co[1]head structure.

“You’ll have a mutiny on your hands!”

A disproportionate amount of the revenue is generated by an overseas franchise that’s not in the same timezone as head office. The bank wants to appoint a manager that’s close to the C-suite, but the over-mighty subjects aren’t going to accept this; they want self-government. So a co-head structure is agreed pretty much explicitly along colonial lines; one co-head is the voice of the bank in the business and the other is the voice of the business in the bank.

Meltdown potential: Wholly cyclical.

A structure like this will rub along while the money is rolling in. In the good times, the co-head at head office will be feeling good, as they generously allow some of their division’s wealth to be used to subsidise other people’s bonus pools. When things turn down a little bit, it has the potential to get very aggressive, as the co[1]head tries to call in favours and discovers that there’s no gratitude in investment banking. And when the market turns down a lot, the bank realises that the overseas office isn’t so mighty any more and what they need is someone to cut costs; the co[1]head at head office gets made sole head and the overseas co-head is made the scapegoat and fired.

“When the call comes at 3am, will you be ready?”

This can be seen as a version of the above, but in this case the important business unit is right on the other side of the world, and is in a volatile market which requires emergency decisions to be made. The co-head structure exists so that these decisions can be made during the domestic trading day.

Meltdown potential: Low.

This is one of the few cases where the structure serves an actual purpose, so unless there’s genuine interpersonal hatred or a severe case of childishness erupts — neither of which is wholly unknown — it ought to last. Even if it does melt down, you often find that domestic regulators prefer to have someone on the ground to yell at, so it might be replaced with another co-head structure.

“I guess we could make them co-heads?”

Sadly, probably the most common structure of its kind. The committee is deadlocked; there is a strong case for either option, and a high probability that the
disappointed party will leave. So — let’s split the difference and make them both co-heads! It only costs twice as much!

Meltdown potential: High to certainty.

“When you see a fork in the road, take it” is not good personnel advice. Delaying the inevitable can be the right thing to do — particularly if the alternative is losing a key employee at the wrong point in the cycle — but without truly extraordinary inputs of maturity and common sense, this structure is doomed.

https://www.ft.com/content/2ba20e5d-c8ea-4b7c-8768-88b3cc18615d