This is an illustration from the report. All the money is swirling from outer space toward the Earth, possibly to land at a bank. Source: McKinsey & Company.
This is an illustration from the report. All the money is swirling from outer space toward the Earth, possibly to land at a bank. Source: McKinsey & Company.

One of my great skills as an investment banker was saying no. Clients or other bankers would come to me and say "oh, can we do this ridiculous thing?" and I would say "no" or "no you dummy" or "that's a really interesting idea, let me take it back to my team," or whatever the context-appropriate version of "no" was. And then the ridiculous idea would never darken my door again and I could get back to the important business of pitching people on my ridiculous ideas.1

This, it turns out, is not a universally valued skill in investment banking. There are good reasons for that. For one thing, most of the money in investment banking has historically been made by going and doing ridiculous-sounding things. "What if we packaged mortgages into bonds?," someone once said. The success rate is low, but the rewards to success are high, so you take every opportunity seriously even if it's pretty stupid.

For another thing, investment banking has a … culture. A culture of punching yourself in the face for client-service and looking-busy reasons, mostly.2 That's why bankers work 120 hours a week and are on call all the time: because if a client wants something, you drop everything and get it to her as quickly as possible and in a way that exceeds her expectations. "Where did my stock close today?," a CEO will ask, and two minutes later you'll email her back a 20-page presentation with charts of her stock's performance, comparisons to her peers, and an LBO analysis. And she'll be like "umm thanks I guess, I just wanted the price really, but I see it's in here."

Obviously, you will not charge her for this presentation. The idea is that by punching yourself in the face frequently and enthusiastically for the client, you will earn her respect and/or pity, and she will one day hire you to do a fairly simple thing like manage her initial public offering. Then she will pay you an amount of money that is (1) quite large and (2) entirely unrelated to the difficulty or effort of managing the IPO. You're not really getting paid for the IPO, though; you're getting paid for the relationship.

That's the model -- do a lot of pointless work and occasionally have it pay off massively. Anything you do for any client can be justified as relationship-building. "Well but they've literally never paid us a penny in the entire 10 years we've been calling on them." "Yes but that's because the relationship wasn't strong enough!" Etc.

So if you went to the typical senior investment banker and asked him, "would you prioritize providing high-value services to top-tier clients, or high-value services to second-tier clients, or low-value services to top-tier clients, or low-value services to second-tier clients," he would say "Yes!" and think himself pretty clever. And if you asked him, "should we do any sort of rigorous data-driven analysis of which of the services we're offering are actually profitable for us," he'd say "yes, you should absolutely do that, have it on my desk by morning, but don't expect us to actually stop doing any services just because they're unprofitable."

I have mixed feelings about this McKinsey report on global banking! McKinsey is after all McKinsey, not a bank, and so its people are pretty into things like efficiency and telling other companies how to run their businesses, and less into things like doing tons of uncompensated work. (Except this report I mean.) They are worried about the large global capital-markets banks, whose return on equity is below their cost of capital and getting lower. And their advice is in part: Maybe you should think about what you're doing instead of just, y'know, doing all of everything.

So the report says that "scarce resources" should be "allocated to the clients that have the potential to generate attractive returns," with less focus on clients who don't, y'know, pay. "Less focus" not just in the sense of "work for those clients between 1 and 4 a.m."; "less focus" in the sense of call centers:

This means migrating clients to full electronic platforms where possible, experimenting with in- and out-bound call centers in lower-cost near-shore locations for client-facing sales, and restricting RWA and balance-sheet allocation.

I will not opine on the feasibility of replacing bond salesmen with an 800 number to a "lower-cost near-shore location," but I suspect it'd be bad for morale.

They have a similar view on product proliferation:

For example, banks often offer multiple minor variations of the same product, designed to attract marginal income in a particular asset class or to cater to a specific client. These products often do not attract enough volume to justify their cost.

Banks must initially undertake a one-off simplification and rationalization exercise, with a view to cutting out dead wood and reducing costs. For example, the ongoing standardization of OTC derivatives provides a great opportunity to drastically simplify the product set in that space. This effort should start with liquid rates products in the United States. For bespoke and complex products, a better understanding is required of front-to-back economics through the lifecycle before a decision is made on whether the products make economic sense.

So what do you think? On the one hand, just yesterday I was defending the irrational in finance. Efficiency and profitability and ROE are fine for consultants, but global banks answer to a higher authority. That higher authority being "do crazy stuff because maybe it's good for the economy somehow," I don't know. But also: What is the point of working in finance if it isn't fun?3 Getting a better understanding of front-to-back economics through the life cycle before pitching a crazy new product does not seem fun.

On the other hand, it's hard to say that McKinsey is wrong here.4 In particular the last few years have seen a number of large global banks scaling back their focus on doing everything for everyone, with JPMorgan getting out of student loans, UBS scaling back investment banking, and Morgan Stanley shifting toward asset management. Presumably those decisions are driven by sober, serious, post-bubble reflection about what businesses at what banks can earn their cost of capital.

Most interesting, though, might be Goldman Sachs's decision -- which, sure, I have gently mocked -- to tell junior bankers to go home sometimes. Whatever this decision says about falling deal volumes or recruiting competition, it marks a big cultural shift simply by telling bankers that sometimes it's okay to say no: that the way to evaluate an assignment is "is it necessary?" or "is it profitable?" rather than just "did someone ask me to do it?" And that cultural shift seems to be in the air: Goldman is trying to answer those questions for its analysts and associates just as McKinsey is suggesting that the big banks need to start asking those questions for themselves.

1 Full disclosure, I may be misremembering and/or lying here. Actually I think I once had a performance review where I was told "you need to get better at saying no." By the end though I was AWESOME at it.

2 I can't really quote this Epicurean Dealmaker post here because this is a family newspaper but he gets the culture:

the core premise of my industry … is uncompromising customer service. The entire … point of working 24 hours a day, for days on end, and canceling weeknight dates with Kate Upton, Las Vegas bachelor parties with George Clooney and Brad Pitt, and Christmas holidays with the Pope of all Christendom at the last possible minute is that the entire investment bank works at the pleasure and whim of clients who pay us a small fortune to do so. That is why successful senior bankers never tell the client no. That is why we agree to impossible deadlines for ridiculous requests at the last minute. Why we therefore ruin our junior bankers’ lives with all-night and all-weekend work on a regular basis. Because that is the ... job.

Some of those ellipses cleverly disguise a bad word.

3 Is it money? Is the point money?

4 I mean, the Bloomberg News headline is "McKinsey Says Investment Banks Should Merge FICC, Stocks." McKinsey is probably wrong there.