RWA Capital Inefficiency โ and How Shared Inventory Solves for it
2026-09-04
A PancakeSwap case study experiment (Part 1: The Setup)
This is the first part in a two-part case study. In this part, we lay out the real capital efficiency problem RWA asset issuers face, introduce how the Shared Inventory Hook (SIH) by PancakeSwap is meant to solve it, and introduce a live case study experiment to test it with onchain data.
Over the next few weeks, we publish Part 2: The Results.
๐งฉ What is the Capital Efficiency Problem?
The RWA industry loves one number: how much has been tokenized. It's the headline on almost every report - a new asset class arriving, billions issued, 100s of percentage growth.
There's a second number that gets buried though, which might be a more prominent signal: how much of it can actually be traded?
The gap between those two is where itโs at. On the canonical tracker, rwa.xyz, the amount of value represented onchain dwarfs the amount that can be transacted against at any given moment โ the data shows locked, reference, and distributed value running more than an order of magnitude larger than what sits in tradable books.
Source: app.rwa.xyz/stocks
For an RWA issuer, this is the trap. Tokenizing a stock, an ETF, or a treasury is the easy part - getting it to trade tightly and reliably is the challenge, and it's where capital efficiency lives or dies. The ultimate goal is to let traders actually trade the tokenized asset easy and cheap - thatโs ultimately where the volume and fees come from.
And here's why it's specifically a capital-efficiency problem, not just a liquidity one.
Every pool quoted demands its own exclusive USDT stack
To make one RWA pair tradable in a traditional concentrated liquidity (CLAMM) pool, issuers need dedicated quote-asset โ USDT โ sitting there for buyers to sell into. List ten pairs, and the CLAMM model requires funding of ten separate USDT stacks, one stack per pair, none of them able to help each other. So quote capital gets shredded into exclusive silos.
We measured it on our own book. We took 7 tokenized-equity pools live on PancakeSwap (each running a traditional V3 CLAMM pool) and pulled the exclusive USDT.
Snapshot: 2 September 2026.
CRCLB / USDT
$352,836
$654,170
SPYB / USDT
$170,644
$355,677
SNDKB / USDT
$123,262
$227,068
GMEB / USDT
$64,323
$135,665
MUB / USDT
$19,146
$35,733
DRAMB / USDT
$11,041
$22,032
SOXLB / USDT
$4,082
$8,276
Total
$745,334
$1,438,621
That's ~$745k of USDT exclusively allocated.
And this is only the names that have a book.
SNXXB, KORUB, SOXSB, and DJTB have no CLAMM USDT book at all โ tokenized, but hard to trade.
๐ฏ What should an Issuer be Optimizing for?
Strip away the mechanics and an RWA issuer must solve for three things at the core:
Their asset to be truly and efficiently tradable โ tight quotes on every asset, not just flagship tokens.
The least capital possible tied up doing it โ Quote-asset is expensive; funding ten stacks of USDT exclusively is not very efficient.
Not to run a trading desk โ To issue assets without needing to operate liquidity infrastructure every day.
So the question is whether liquidity can be structured differently โ and that's what the Shared Inventory Hook by PancakeSwap is designed to solve.
๐ฅ Shared Inventory Hook (SIH)
SIH is a PancakeSwap Infinity hook that lets an RWA issuer back many pairs from one shared USDT pot instead of one exclusive stack per pair.
SIH does this by providing liquidity just-in-time, enabling:
Deeper quotes on pairs
Lower price impact, especially on large ticket trades
Freed up capital for RWA issuers since less capital deployed quotes more pairs
Letโs look at these claims in practice.
As of the same 2 September 2026 snapshot, SIH holds $635,028 of inventory in a single hook, quoting 11 pairs from $180,467 of shared USDT, running at 87.7% utilization. With this, SIH has processed $3.56M across 17,299 transactions.
So that is one shared USDT balance of ~$700k quoting across ~11 pairs versus ~$745k quoting only 7 pairs.
Same job, only designed to stop duplicating capital.
Mapped back to what issuers care about:
N exclusive USDT balances traditionally โ โ one shared pot with SIH
Issued but not tradable โ โ tradable Shared inventory injecting just-in-time around the live price means tokens that have no V3 USDT book can still be quoted, not just minted.
Operating a desk โ โ a permissioned operator The issuer supplies inventory; the hook and its operator handle the quoting.
๐ฌ Try the Shared Inventory Simulator
Build intuition for how SIH behaves with our interactive comparison tool โ https://pancakeswap-shared-inventory.netlify.app/
This is a high-level, normalised comparison of SIH against a traditional CLAMM book โ a number of real-world factors are held constant on both sides so the capital-efficiency mechanic can be isolated cleanly. It's a model for understanding, not a P&L.
Three tabs:
Simulator tab - play with inventory, band width, and flow to see how SIH would perform against a comparable CLAMM pool under different conditions.
Comparison tab - SIH set directly against a PancakeSwap V3 CLAMM book on the same pools, so you can see the shared USDT pot-vs-exclusive USDT difference in capital committed and coverage side by side.
Live data tab โ measure how SIH is actually performing right now, pulled from the onchain queries powering this study.
The simulator is aimed to help demonstrate how the Shared Inventory Hook operates. Try it, then measure it against the real results we publish in the next couple of weeks.
๐งช The Experiment
Over the next few weeks we run SIH live and measure it against the capital-efficiency claims above, using onchain sources. Every week we will drop new info-bites on SIH performance and its architecture.
Live RWA pairs in the study: CRCLB, MUB, GMEB, SNDKB, SPYB, SOXLB, SOXSB, DRAMB, SNXXB, KORUB, DJTB.
Note: SNXXB, KORUB, SOXSB, and DJTB have no traditional CLAMM book at all.
We will publish the results over the next few weeks on PancakeSwapโs Twitter.
The results published will cover:
The full performance dataset - All SIH data over next few weeks, combined into one picture.
The capital allocation question - How much capital SIH actually saved an RWA issuer.
Beat / miss - Where SIH outperformed expectations, and where it needs work.
The verdict - A grounded conclusion on how effective SIH really is for RWA capital efficiency.
A note on data: Figures here are drawn from live onchain queries and public pair data, cited inline. Some comparisons use different windows and we've flagged those โ the study is designed to measure like-for-like where this intro could only frame the question.
Want to enable SIH on your assets?
If youโre a RWA issuer and you're tired of funding a separate USDT stack for every pool you quote, this is for you.
We're onboarding RWA issuers now. Fill out this google form and we'll walk you through getting your asset quoted on SIH - https://forms.gle/LV24KsoDfJdu2RQU9
Thanks for reading! Follow us on X for the latest updates, and join the conversation on Telegram and Discord.
Letโs build, The Chefs ๐ฅ
Disclaimer: This content is for general information and educational purposes only. It should not be construed as financial, legal, or other professional advice. Digital asset prices can be volatile. The value of your investment may go down or up.
Last updated
Was this helpful?