As we observed the other day, there was a time when investment banks were among the first adopters of new technologies, and when top technologists queued up to work for them. That time was decades ago, before the birth of FAANG, but it could make a comeback.
In the mid-2000s, a group of quants at Credit Suisse were among the first adopters of the F# programming language invented by Don Syme, a principal researcher at Microsoft Research in Cambridge. Syme has just immortalized their efforts in a paper on the early history of F#.
Credit Suisse's global modelling and analytics group trialed and adopted early versions of F# in 2006-2007, says Syme: it was used to author models in financial instruments and to orchestrate existing Windows COM components.
John Elder, a former Credit Suisse quant and head of market risk research methodology, who now works for BlueCrest Capital Management, wrote on LinkedIn that the early adoption of the language at Credit Suisse took place despite the upfront cost of training the bank's C++ programmers in "a brand new, functional programming language (when most of us had never experienced FP before)." The use of F# enabled Credit Suisse to build out a, "library of several hundred, very substantial “valuation models” in around two years, a level of productivity that would have been unthinkable in C++," added Elder.
F# is now widely used in model development jobs, especially in credit and fixed income, but in 2006 the language was still very new. "The development of the core features of F# 1.0 happened from 2004-2007," says Syme. It was only 'productionized' by Microsoft in 2010. Credit Suisse was therefore one of the earliest of the early adopters.
It wasn't the only bank at the party. Syme says Morgan Stanley also, "initiated a large project to convert large portions of their analytics to Windows and F#, to replace a legacy APL codebase" well before F# was widely adopted.
Banks' enthusiasm helped encourage Microsoft to back the language: "In the summer of 2007 the influence of Wall St was at its peak worldwide," says Symes. - If banks used a new language, so would everyone else.
Thirteen years later, banks' threshold for novelty is greatly diminished. Languages like Rust and Julia are barely used in finance as risk averse banks stick to tried and tested 'stadium languages' like Python, Java and C++. Credit Suisse's experience with F# suggests this is their loss.
The good news is that there are some flickers of innovation appearing: Goldman recently helped design a new programming language for data science. More of this would be beneficial: banks would help regain their position as technology leaders rather than followers, and technology talent would be more inclined to knock on their doors.
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.)