How do the best PMs trade L/S consumer and other sectors?
There are three core approaches. But first we need to understand how market structure has evolved over the past 10-15 years. It has fundamentally changed both the investment processes top PMs employ and who the incremental price setters in these stocks are.
The consumer sector has undergone a major transformation in the past decade with the rise of alternative data. It has been especially impacted as the nature of these datasets makes them highly predictive of core business metrics. Credit card and email receipt data generally allow for a granular view into all kinds of company KPIs, product and customer level analyses. The intuition is quite simple. When you purchase something using your credit card, especially in the US, it gets recorded as a transaction (you can see its official description on your credit card statement), anonymously aggregated and sold to hedge funds. Think through all the purchases you’ve made: Amazon Prime, clothes, subscriptions, coffee, etc.
Beyond tracking KPIs, you can answer all kinds of interesting questions about consumer behaviour with data. If someone starts using ChatGPT, how does that affect their Google search usage or their engagement with AI application layer products like Chegg or Duolingo? If someone is shopping at Alo, are they also shopping at Lululemon? How is that behaviour changing over time? What’s the lifetime value of customers acquired during periods of aggressive discounting?
How is total consumer spending trending? Is there any impact from tariffs or tax refunds and are there meaningful differences across income cohorts? In a consumer-driven economy like the US, questions like these are critical.
Read previous write-up: How to Win Against the Pods
Some history: the 2015-2020 period was the golden age for alt data. Fundamental long/short investors who were well-versed in the data and had the resources to access it before the rest of the market were able to generate an enormous amount of alpha from it. Full control and mastery of the data stack require owning the entire raw data → insight process. While I would argue this process is still far from commoditized, back then adoption was only a tiny fraction of where it is today. Bigger funds with the budgets to buy datasets and hire the talent needed to extract alpha from them were significantly ahead of everyone else.
Fundamental PMs at the large multi-managers who were early in developing the data science skills, teams and infrastructure were major beneficiaries of this trend.
Some of the bigger funds, including the large multi-managers, were even directly monetizing the data through central portfolios, separately from the individual PM discretionary books. In practice, this meant the decision to long or short a name was based on a wide range of conditions and signals from a wide variety of data sources. A key use case was forecasting core KPIs, e.g., whether a company was likely to beat and raise revenue guidance when reporting earnings.
Some funds even ran semi and fully-systematic strategies to profit from it. Coatue notably was able to monetize it early in its public equities business after starting a data science team early in 2014, eventually evolving into a systematic strategy.
It was all roses and sunshine for early discretionary alternative data adopters until Covid hit.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.