Author | Hayden Adams, Founder @Uniswap

Compiled by | Deep Tide TechFlow

Original link:

https://www.techflowpost.com/article/33433

Editor’s note:

Uniswap founder Hayden Adams believes that as traditional assets such as stocks become tokenized, “correlation trading pairs” like NVDA/SPY will reduce inventory and hedging costs, giving passive AMMs an opportunity to enter the core markets long dominated by traditional market makers. Crypto KOL Cody, meanwhile, conducted further analysis of the feasibility of this idea in today’s on-chain market by combining historical backtesting with real LP experience.

The following is the full text:

This is my first blog post since 2019!

Correlation trading pairs: how AMMs win the biggest markets

I’ve worked on the DeFi frontier for 9 years. It’s a fascinating space with infinite depth and the ability to change capital markets.

I’ve always believed in the huge potential of AMMs, but for the past decade there’s been one big question that has bothered me: can this new market structure truly become the core engine for all financial markets?

After years of evolution and growth, a path toward AMM’s global dominance is becoming increasingly clear. And the best way to explain it starts in 1976.

Tokenization changes who makes markets

Index funds mark their 50th anniversary this month. When Jack Bogle launched them in 1976, he wanted to raise $150 million. He only raised $11.3 million. Competitors called it “Bogle’s folly” and printed posters saying index funds are “non-American.” They believed a fund that makes no decisions can never beat professionals who get paid to make decisions. Today, most U.S. fund assets are stored in passive instruments.

I’ve been thinking about this recently, because the “silly moments” of tokenization are ending. The SEC has approved tokenized stocks trading on Nasdaq and the New York Stock Exchange. The DTCC, which clears almost every security in the U.S., conducted a live test of tokenized trading in July. Almost all of these activities have been described as the same thing: tokenization as an infrastructure upgrade.

Same market, faster, cheaper, running all day. It’s all true—but I think the “upgrade” framework hides a bigger story. Tokenization makes markets programmable: changing which markets exist, who makes markets, and what these markets trade against.

In 2018 I built Uniswap. Anyone can deposit two assets into a shared pool and earn trading fees on every trade, while the price adjusts along a curve as people buy and sell. Uniswap has been running autonomously since day one, settling over $4.6 trillion in trading volume and helping decentralized exchanges grow from less than 1% of centralized spot trading volume to more than 20% over the past few years.

As AMMs like Uniswap grow, their liquidity has formed a pattern that most of finance hasn’t noticed yet: correlation trading pairs.

The easiest markets to win

To win everything, you have to win something first. AMMs found product-market fit in the long-tail markets first, where most assets can’t even get the attention of professional market makers. On Uniswap, anyone can create a market in a single transaction—issuers and early supporters become the first LPs.

Stablecoin pairs follow closely: on pairs like USDC/USDT, a good passive strategy is already close to optimal, and lower capital costs are enough to make up the gap. That’s why professional trading firms don’t bother today to market-make for stablecoin swaps—passive AMMs race to the bottom and have outcompeted them.

High profit, no competition

Traditional financial markets belong entirely to market-making firms. They bundle capital, trading strategies, execution technology, settlement, and distribution into a vertically integrated business. There’s a reason this architecture evolved: assets live in isolated systems, settlement is slow, and each function needs someone to execute—so one company ends up doing everything.

At a sufficiently large scale, all these fixed costs can cover themselves. Citadel Securities handles about 25% of U.S. stock trading volume and produced a record $12.2 billion in net trading revenue on about $21 billion in trading capital last year.

Most people see these numbers as proof that the system is working well. I see it as self-justification.

Break the lock-in

Blockchains create competition at every layer and break lock-in. Execution is handled by code. Custody and settlement become shared services anyone can access. What once required proprietary infrastructure is now open-source software.

In AMMs, capital is a scarce input. The advantage belongs to whoever can hold inventory at the lowest cost. Trading companies need high returns to justify their daily expenses, so LPs willing to accept lower returns can race to the bottom on price. Most market makers hedge away all price exposure, but hedging costs money—so investors who already hold the asset can absorb those exposures for free. Meanwhile, the asset issuer has a negative cost of funds, because issuers typically have to pay professional market makers to make markets in their newly issued assets.

In short, DeFi and AMMs lower the barrier to market making and open up the field to many new participants. Their advantages can come from many different sources: lower cost of capital, willingness to hold inventory exposure that professional firms would normally hedge out, and even the fact that they themselves are the issuers.

But all of this depends on one question: can automated strategies perform well enough for it to work?

Liquidity follows correlation

Recently, I had a call with one of the largest institutions in finance. They asked me what the most common benchmark trading pair is in DeFi. I explained that based on Ethereum-based assets, people tend to trade against ETH; Solana assets trade against SOL; stablecoins pair with each other; and a small number of highly liquid pairs are used to build bridges between these clusters.

Nobody designed this pattern. It emerged naturally, partly because LPs perform best when the assets they hold fluctuate synchronously. Correlation means liquidity providers have lower inventory risk, which deepens liquidity. With tokenization of assets, the world’s largest markets will reorganize in the same way.

They can’t do that today. Traditional markets settle in dollars by necessity. Assets exist in isolated systems, and fiat rails like SWIFT and Fedwire are the glue that holds everything together. But blockchains are a much more expressive kind of glue. When you tokenize assets, they share a settlement layer, so any asset can directly trade against any other asset.

NVDA/USD can become NVDA/SPY, using SPY/USD as the bridge back to dollars. Oil companies can trade oil ETFs or tokenized oil. Private credit can trade tokenized treasury funds. Tokenization also makes markets spanning different types of assets possible—something extremely difficult and even infeasible in traditional financial infrastructure.

Delta neutrality is inefficient

Traditional market-making firms typically try to maintain “delta neutral”—a trader slang meaning valued in dollars and hoping to minimize any risk that isn’t in dollars. When making markets for a volatile asset, they spend money to reduce their non-dollar risk (i.e., hedge), usually through options. This is one of the costliest parts of traditional market making.

Pair assets into low-volatility “correlation pairs,” then connect them with a small number of high-volatility “bridged pairs.” This brings many efficiency gains, but most importantly: if market makers actually want to hold the underlying assets, market making becomes cheaper and more efficient.

And the higher the correlation between pairs, the smaller the gap today between passive AMM strategies and even the most complex active strategies—and the easier it is to “race to the bottom” on inventory costs.

Here’s a concrete example: if someone goes long Nvidia, you might also go long SPY. The efficiency gap between passive AMMs and active strategies on NVDA/SPY is far smaller than on NVDA/USD.

interconnected liquidity

If stocks are changed to trade against SPY, then every trade that starts or ends in dollars routes through the same pair: SPY/USD. These bridged pairs still require professional capabilities, but there are far fewer of them—and they carry so much traffic that professional attention is worth it.

DeFi has already proven this. ETH/USDC is one of the deepest markets on-chain because every cluster routes through it. Passive LPs provide the relevant trading pairs, while active LPs compete on the bridged pairs.

Investors can still buy and sell everything with dollars, because cross-pool routing is automatic. Liquidity will concentrate where risk is lowest, not where legacy plumbing requires it to sit. This pushes the deepest markets toward correlation pairs—AMMs’ strongest position.

Correlation RWA trading pairs already exist

On-chain correlation liquidity starts with crypto-native assets. But tokenized stock correlation markets already exist: within a Uniswap pool on the Robinhood chain there are ten tokenized stock pairs traded against SPY.

In the first 12 days, these pools completed over 33 million dollars of trading volume involving more than 11,000 traders—most of it occurring during the U.S. market close window. Some trades moved directly from one stock to another without touching dollars at all.

Worth mentioning is that we’re also starting to see memecoin paired with “related” stocks: an Elon-themed meme coin paired with Tesla stock, and a hot-dog themed meme coin paired with Costco stock. I’m not sure how correlated these are in terms of price, but I suppose “vibes” can be a kind of correlation too.

AMMs will win

Correlation trading pairs are only part of the puzzle. The other part is AMM design and customization.

Uniswap v4 hooks enable comprehensive market customization, which can significantly improve LP returns. For example, our recently released DualPool hook—when passive AMM funds aren’t used for swaps, they’re put to work earning lending yields.

Although Uniswap has about $4.6 trillion in trading volume, I believe AMMs are still in the early stages, with many other paths to improve their competitiveness. There are also many other promising ways to increase LP returns being built both within Labs and by our partners and the wider ecosystem. More is coming!

The reason to oppose index funds in 1976 was: a fund that makes no decisions can never beat professionals who are paid to make decisions. Fifty years later, funds that make no decisions beat about 90% of professionals. Even more importantly, index funds democratized investing and improved the lives of ordinary people. I believe passive liquidity will win with similar strategies and create an even greater impact—by dramatically lowering the barriers to creating and participating in markets.

Crypto KOL perspective

Hayden Adams’s article also sparked discussion in the Chinese community. Later, crypto KOL Cody (@Cody_DeFi) further analyzed the idea of “correlation trading pairs” from the perspective of real market making and historical backtests. The author feels it’s a pretty good interpretation and analysis of Uniswap founder’s original design thinking; the original post is also linked below:

It feels like no one in the Chinese community is paying attention to Uniswap’s founder’s big post (how to replace traditional market makers for on-chain stocks with AMMs). Is it that no one is paying attention to DeFi?

The founder @haydenzadams actually mentioned a core point: if you’re willing to actively hold an asset—actively exposing yourself to a long position in an asset—then your cost of making markets is far lower than that of traditional market makers, because you don’t need to hedge against price volatility. Traditional market makers do need to: they have to stay in a so-called delta-neutral state, which comes with very high hedging costs.

Because of the AMM mechanism, it can almost be seen as an asymmetric fee-based grid trade. If you hold a long position—say NVDA—and you use an AMM to make markets in the NVDA–USDC pair, you face the risk of getting undercut (sell-off risk). As NVDA’s price rises, you end up with more USDC and less NVDA; normally these losses are called “impermanent loss.”

To solve this problem of getting undercut, hayden introduced the so-called concept of “paired AMM market making.” In simple terms: instead of deploying an NVDA/USDC AMM pair, normal people deploy correlation pairs like SPY/NVDA. This reduces impermanent loss while maintaining long exposure to a correlated set of assets, earning fees.

This idea comes from how he observes the current mainstream deep trading pairs in crypto: ETH\SOL are used as the primary trading assets, and then mapped to tokens in their respective chain ecosystems, just like the cover art of this article.

Finally, he believes that this market-making strategy—meant for those willing to hold long spot—will eventually take over an increasing share of stock-token market-making, just like after Uniswap’s AMM arrived, the on-chain spot trading volume relative to traditional CEXs rose from 0 to around 20% over the past few years.

Coming back to my own view, the part I agree with most is that AMM really does offer regular people who are willing to hold inventory an opportunity to make markets. Because if you’re willing to hold a certain kind of asset, you can indeed do market making without hedging.

Beyond that, hayden proposed a third advantage: the advantage of paired-stock-token LPs would be greater than simply holding the stock. In that context, I ran a simple backtest using NVDA and SPY based on the past 3 years of data. The rough impermanent loss came out to about -10.8% of end-of-period principal. Without reinvesting fees, you’d need about 11.5% per year in order to make up the fee amount at the end of the period.

I took a look at it. Given current on-chain liquidity and fee conditions, it’s hard—because when stocks aren’t in their trading hours, there are basically no fees. The corresponding pairs for SPY/NVDA are also rare, and even when they exist, the APY is very low. This probably only can hope to be resolved over time.

Finally, compared to the paired LP strategy, what I do on-chain myself is USDC/NVDA, and strategies like USDC/CRCL. The core idea is to do LP strategies in the form of a fee grid, combined with traditional finance valuation models—turning traditional holding-based returns into a hybrid of holding + trading fees + grid-style buying low and selling high.

This yield setup is still being validated in live trading, so the timeline may be relatively long. But I believe the AMM market-making strategy truly fills a gap in today’s market making for tokenized stocks. The core is still that theory: if you want to hold a position, then your market-making cost doesn’t need to be hedged. But in practice you still have to account for impermanent loss—this is a gap and an area still to be explored.