If there’s one part of the investment banking industry that always seems to be on the brink of extinction, but never seems to go away, sell side research is it. It’s an attractive job for the right kind of person, because it combines some of the joy of gambling on financial markets with a degree of insulation from the financial consequences getting it wrong. Historically, (and this is said with the voice of fifteen years’ in this role) it has been the perfect role for the kind of people who begin to feel uncomfortable if they haven’t been giving someone else their opinions for more than five minutes.
In general, though, as the proverb goes, opinions are like dreams and intestines – everyone has them, and nobody is very interested in the content of anyone else’s. So, understandably, as price pressure for the sell side has become extreme under MiFID II, the Financial Times notes that heads of research have been trying to shift their talkative charges in the direction of something that people might be prepared to pay a premium price for. “Analysis used to be very opinion-oriented. Now everyone wants the raw info. It’s about analysing it better,” says Lou Pirenc, global head of research data at Morgan Stanley.
The most extreme examples of this are things like the UBS Evidence Lab, which effectively produces and retails alternative data to the mass fund management market, and which has now been spun off as a separate business unit. But even everyday stock analysts are being encouraged to use satellite images and credit card data, and to crunch (or get a junior to crunch) big datasets with machine learning packages in order to provide some backing for their hunches. According to JP Morgan’s chair of research, Joyce Chang, “if all you’re doing is traditional fundamental analysis, it’s not enough any more”.
The trouble for the industry is, though, that data is fundamentally a commodity, and when you’re doing the same kind of thing as everyone else with the same kind of software as everyone else, you’re unlikely to get results which are sufficiently differentiated. One of the good things about educated analyst opinions is that, also like dreams and intestines, they’re unique and difficult to replicate. For this reason, few banks are getting out of the analyst game entirely; a lot of the tech input into research departments has gone into creating customised delivery systems, boiling a report down into a short “charticle” so you can get the key insights without having to listen to quite so many of the author’s general bloviations.
This is valuable, except that for all that they claim to be short of time, there are plenty of investors who also like to talk. And a lot of the value in the research department has always resided in the fact that it’s one of the highest-touch parts of the bank in terms of everyday client contact with investors. While a good regular chat about which stocks you like is the stuff that lifelong friendships are made of, it’s hard to build a personal relationship on the output of a machine learning package.
Elsewhere, something pretty wild seems to have gone down at Morgan Stanley, with at least four trades fired or on administrative leave and a concealed loss of between $100m and $140m, apparently. The source is the usual “people with knowledge of the matter”, but it doesn’t seem like the sort of thing that it would be possible to be unsure or ambiguous about. The individuals named all seem to have worked in and around the macro trading desks, and the issue appears to relate to mis-marking of currency options.
This is really the sort of thing that wasn’t meant to be possible any more. We have straight-through processing, automatic pricing and data validation; the days when traders priced their own books at close of business, particularly in FX markets, ought to be far in the past. It just goes to show the dangerous gap between the perception of banking IT as streamlined and technically sophisticated and the reality of legacy systems stitched together with a whole lot of dependence on fallible human beings.
This happened at Brevan Howard and now it’s happened at CQS with Michael Hintze. A turnaround in overall firm profitability and assets, led by a strong period of performance in the flagship fund run by the boss. Nothing like leading by example. (Financial News)
As well as landing a role where the succession to Mike Corbat is probably hers to lose, Jane Fraser’s promotion at Citi comes with a $12.5m bonus (Bloomberg)
A series asks millennial workers what it’s like to live on the current average entry-level graduate salary of $50,000 in various cities of the USA. In New York and San Francisco the answer is presumably going to be pretty frightening; in Detroit it doesn’t look too bad. (WSJ)
An unsual sighting – a Softbank investment that’s profitable. The new CEO of British lending startup OakNorth is Sunil Chandra who is a former investment banker, although since he calls it “Barclays Capital” in interviews, not a recent one. (Business Insider)
Setting up a code of conduct for crypto exchanges was always going to be something of a thankless task (Financial News)
“You know what’s cooler than a billion dollars? A million dollars!”. Not exactly what Mark Zuckerberg said in the film, but UBS has now joined Credit Suisse in setting up a team to concentrate on the lower levels of affluence between $500k and $2m. The billionaires market is expensive in terms of personnel and overhead, and if you can use technology cleverly, you can serve 1000 millionaires at a better all in profit margin. (Bloomberg)
Have a confidential story, tip, or comment you’d like to share? Contact: firstname.lastname@example.org in the first instance. Whatsapp/Signal/Telegram also available.
Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)
Photo by Scott Umstattd on Unsplash