Hedge Funds is The New "Big Tech"
How is it possible that a single pod consisting of a PM and 3-4 analysts can manage a $3B+ book?
Hedge Funds is the new “Big Tech”? Or… it has always been “big tech”, but people have only recently started to realize it.
Normally, when someone says they want to work in big tech or tech more generally, they refer to Amazon, Google, Meta, Microsoft or other Silicon Valley tech firms. Working at a large hedge fund, especially in fundamental long/short equity, has traditionally not been considered a tech job.
But today, successful hedge funds are technology companies.
If you have read The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution, you might have noticed that, in its early days, Renaissance Technologies was doing what we now call machine learning, even before it became a formal field of study, and arguably invented some of the sophisticated ML and AI techniques that became mainstream decades later.
This is an extreme example, but it illustrates the point well. You might say this is quite obvious for the most quantitatively advanced hedge fund of all time.
Let’s now think about fundamental long/short equity. There are a number of pods consisting of a PM and 3-4 analysts at the big pod shops that manage multi-billion-dollar portfolios (think $3B+). A team of only 4 people managing $3B+?
How is this possible?
We can use Formula 1 as a strong analogy here. The true power and success lie not just in the skill of the driver but in the car they drive. The PM is the driver, and the car - with its finely-tuned engine, precision hydraulics and support team - represents the hedge fund and the resources it provides to the PM such as state-of-the-art risk management infrastructure, cutting-edge software, alternative data and more.
Risk Management and Portfolio Construction Infrastructure
We have talked at length about the extremely robust and sophisticated risk management and portfolio construction these funds employ (more on this to come):
Reading the above gives you a sense of how technologically advanced the infrastructure is. They hire quantitative researchers and risk managers from top PhD programs to work on these systems. Pod shops manage risk both at the individual pod level and at the centralized level. What is even more incredible is that they don’t just do this for a single strategy like fundamental long/short; they do it across strategies (commodities, global macro, fixed income, etc).
The risk teams also build proprietary equity factor risk models that are superior to the commercial ones and consistently seek ways to get as granular with the portfolio exposures as possible, so they can identify even the smallest risks (more on this soon). There are some very advanced statistics and engineering in that process. This is a key ingredient to minimizing risk and continually minimizing the chance of losing money.
This is why the barriers to entry in the multi-manager world are so incredibly high. Building up to that level of sophistication in risk management requires top talent, time and resources.
AI labs
Many large funds now have AI labs and are racing to figure out how to best leverage the rapid advancement in large language models (LLMs). A quick search on LinkedIn reveals numerous AI researchers, AI product managers, AI engineers working at these firms.
There is this whole debate about having an AI analyst that handles the time-consuming manual tasks that research analysts do like updating models, parsing through transcripts, summarizing notes, pulling out numbers from filings, eventually even fully replacing the majority of an analyst’s work.
A similar concept is seen in software engineering, where companies are striving to develop fully autonomous AI software engineers, far beyond the capabilities of current coding co-pilots (Devin is an early version of that). JP Morgan recently announced the rollout of an AI-based agent in their asset and wealth management division that “can do the work of a research analyst”.
There is a growing number of AI tools being developed for analysts that specialize in doing some of the tasks mentioned above (updating models, summarizing and reading transcripts). However, as with alternative data, firms that can develop these tools internally and use them more quickly and effectively than others will gain a significant competitive edge.
Data
The big pod shops have internal data teams consisting of data/software engineers and PhD data scientists and fundamental and data analysts who ingest, analyze, clean and use the data to make predictions and actionable conclusions about various companies. They also create the entire infrastructure around accessing the data quicker and getting more accurate insights from it than everybody else.
PM and 3-4 analysts manage 3B+? Scaling processes.
Returning to the question of how a single pod can scale to a multi-billion dollar book. It’s all about taking the PM’s process and providing all the necessary tools and technology to make it faster, automate it, amplify it and even enhance it where possible. That might involve automating parts of their screening process, getting more operational (compliance, back office, administrative work) and trading execution support. PMs might also get increased support from risk management teams as they look to diversify their portfolio, add new sectors, geographies and even asset classes.
The goal is to ensure the F1 driver has the best car out there, one that fits their driving style and sets them up for success. You want to the driver to focus on driving and as little else as possible. PMs who are world-class stock-pickers and want to scale should be focusing primarily on strategy and decision-making.
You sometimes see great PMs start smaller $300-400m funds but not always perform well or fail to scale their AUM. It’s often not because they suddenly aren’t great stock-pickers but because their new “car” is quite limited.
The point is that the big funds have massive economies of scale in risk management, data, research infrastructure and technology. That is because they are great technology companies.