Every trading house runs the same graduate specification, against the same shortlists, competing with investment banks and technology firms that carry a stronger campus brand and a bigger budget. The pool everyone announced they were opening is the one nobody is actually covering.
A note from Matt. Last month I argued that the base of the trading org chart is being removed, and that almost nobody has framed it as a decision. The diamond that Pete Hirsch described is arriving on desks faster than anywhere else. This month, the harder question I promised. Not how many people we bring in, but whether they should all be the same person. That is a desk design question long before it is a recruitment one.
One specification
Read the graduate specification at almost any trading house and you find the same three things. A degree in finance, economics or engineering. A strong quantitative result. And a list of universities that has barely moved in fifteen years.
Take each on its own and it is defensible. Numeracy on a desk is not a preference. Someone who cannot hold a curve in their head, or spot a wrong number in a position report before it becomes a problem, is a liability, and no amount of character makes up for it. The filter is not illogical.
The problem is what a decade of applying it produces. A room of people taught the same models by the same faculties, who learned to be wrong in the same direction. Every desk head can tell you what is wrong with a book that is all one way. Far fewer ask whether the thinking on a bench is all one way too.
The route that closed
It was not always like this. For most of the history of this business the trade was learned somewhere other than a lecture theatre. The post room. The shipping desk. The pit. The back office, where someone spent three years reconciling other people’s mistakes and came out understanding exactly how a deal settles.
Those doors closed over roughly twenty years, and mostly by accident. Operations professionalised. Graduate schemes gave HR a clean and defensible process. Compliance made informal promotion harder to justify. Nobody sat down and abolished the route in. It just stopped being anybody’s job to keep it open.
What went with it is the part people underrate. Someone who came up through demurrage claims, or credit, or scheduling, arrives on a desk knowing where the money actually leaks. That is a different education from knowing how to price the option, and on a bad day it is the more valuable one.
The rhetoric is a long way ahead of the hiring
The wider corporate world has been noisy about this for five years, and the data suggests that it is indeed mostly noise.
A Lumina Foundation and Gallup survey of 2,000 US employers, published in March 2026, found 23 per cent had removed degree requirements from some roles in the previous three years, with a further 20 per cent in the process. Read that on its own and you would think the filter is coming down everywhere.
Then read what happened next. Joseph Fuller at Harvard Business School, working with the Burning Glass Institute, examined more than 11,000 roles at US firms between 2014 and 2023 and found that fewer than one new hire in 700 actually benefited. Announcing it and doing it are clearly two different projects.
I read that gap as an opportunity rather than a disappointment. The graduate milkround is the most crowded hiring space in this industry and the same shortlists are being worked by investment banks, hedge funds and the large technology firms, many of them carrying a stronger brand on campus and a budget most energy & commodities trading businesses are not going to outbid. That is a fight we do not win on brand or on budget. The pool everyone announced they were opening is the one nobody is covering at all, and for anyone building a desk, that asymmetry is worth more than another year of losing the same names to firms that were always going to pay more.
Geography is doing more of the work than people think
This is where one global early-careers template quietly exports somebody else’s constraint.
In Singapore, where we sit, the graduate pool is small and a degree carries a social weight that makes the alternative route awkward for the candidate as much as for the employer. In the United Kingdom and the United States the pools are bigger and messier, and the back-office-to-desk route survived longer in places. In parts of the Middle East and Africa, nationalisation targets and a much younger population change the calculation again.
A house running one specification across all of that is not being consistent. It is applying the tightest constraint it has to every market it operates in and calling the result a standard.
What I would actually do
Not scrap the graduate scheme. It works, and the case for it is stronger now the intake is smaller.
Run a second track alongside it. Smaller, on its own timetable, off campus entirely, with its own specification. If the graduate scheme is engineered for consistency, this one is engineered for difference, and you judge it on whether it produces people who see what the first track does not.
Then open the third door, which is the cheapest of the three and the most neglected. Mid-career moves into the front office from operations, scheduling, chartering, credit and the middle office. These people are already in the building. They know the counterparties, they know the products, and they have been assessed by four or five years of work rather than by an interview. Most houses have one or two who made that jump though very few have a route.
The objection is that this is expensive and slow at a moment when everyone is trying to run leaner. Half true. A second track costs real money and it will produce some misses. But it is a fraction of the scheme it sits next to, and the alternative is not free either. You are paying for a bench that agrees with itself, and that cost does not arrive gradually. It arrives all at once.
The point
The number of ways into this industry has been falling for twenty years, and the number of people we take through each one is falling too. When you have fewer entry points, the design of each one matters more. Most houses have answered a smaller intake by running a smaller version of the same scheme, which is the decision I would go back and look at.
On my radar
One for leaders to actually do something about. Take the last ten people who joined your front office desks in any seat and write down where they came from. Not the employer, the route: graduate scheme, lateral hire, internal move from operations or the middle office. If nine came through one door, that is a concentration in your hiring process, and it will not appear in any report you currently read.
Whether the smaller graduate intakes now being run are being redesigned or simply cut. Cutting is a budget decision. Redesigning is a strategic one, and at most houses only one of the two is being made deliberately.
Pay in the two to five year band. If the intake below is thinner, the people already three years in become much harder to replace, and that usually shows up in comp before it shows up in anyone’s plan.
Next month I want to take the other side of the argument I have been making for two editions. There are desks where building is the wrong answer and buying a team outright is the right one, and I want to be specific about which is which.
Sign-off
If you are hiring anywhere across commodities trading and the commercial, functional and leadership roles that surround the desk, that is exactly where I spend my time, and I am easy to reach.