Compartmentalizing Risk

Oh shit, he used a big word in the title - this one's gotta be good

Introduction

I’m gonna rip the band-aid off early here and let you know there will be no explicit alpha in this writeup. It’s there in between the lines, but if you’re the type of person who reads long form in the search of your next trade, it’s probably not for you. I wanted to try my hand at some good old-fashioned thought leadership, and I’ve noticed this concept being applied creatively in a number of ways across defi. Here I’ll unpack the different ways compartmentalization is being applied, why it’s been such a game-changer, and why it will continue to be.

Compartmentalized risk (or risk/LP tranching, call it whatever you want) is a broad concept that is exactly what it sounds like. It’s when risk providers (liquidity providers) are given a variety of different options for the risk/reward parameters they’d like to take on in order to have more control over their strategy. Through this process, counterparties subsequently receive greater capital efficiency and overall flexibility. With how prevalent modularity is within blockchain infrastructure across multiple different layers (separating consensus/execution, defi legos, layers 1-3, data availability, etc.) I hink that this concept doesn’t get enough credit when designing incentivization structures, sustainable flywheels, and efficient money markets.

A quick and dirty mental model I like to use splits this concept into 2 approaches - horizontal and vertical compartmentalization. Compartmentalizing horizontally means creating more segmentations so users have greater freedom of choice between profiles to best fit their strategy. Compartmentalizing vertically means making said segments more distinct, so as to better distinguish specific user groups and create more appropriate products for each one.

Let’s look at a couple of examples before getting into some larger implications/ideas.

Subscribe to keep reading

This content is free, but you must be subscribed to Ali's Research to continue reading.

Already a subscriber?Sign In.Not now