The Red Pill: Automated Market Makers for Capital Markets

If you are a capital markets "boomer" like I am, you grew up knowing a fundamental truth about markets: There is someone on the other side of every trade. Buying a stock means someone else sold a stock and the orders were matched by a market maker. If you believe this fundamental truth, like I did for most of my career, then this article might be for you, but first, you have a choice:

"You take the blue pill... the story ends, you wake up in your bed and believe whatever you want to believe. You take the red pill... you stay in Wonderland, and I show you how deep the rabbit hole goes."

If you choose to keep reading, welcome to the land of automated market makers.

From Order Book to Liquidity Book

The recent introduction of practical automated market makers (AMMs) represents a profound innovation born out of decentralized finance (DeFi) that has the potential to turn all of finance on its head whether decentralized or not.

To understand AMMs, it helps to contrast them to traditional market makers (TMMs). Price discovery via TMMs is achieved via a central limit order book. The price of a security represents an equilibrium point among buy and sell orders. In contrast, with an AMM, the price of a security represents an equilibrium point among pools of staked assets.

Liquidity of a TMM is measured by how deep the order book is, i.e. how many buy and sell orders there are in the book and what value do those orders represent. In contrast, the liquidity of an AMM is measured simply by how much of an asset has been staked into a given pool. This shift from order book to what I have been calling "liquidity book" requires a major shift in how we think about capital markets and price discovery in general.

Swapping Apples for Oranges

Every transaction is a swap. When you buy a stock on an exchange, you are swapping the currency of the exchange for the stock. When you convert a currency, you are swapping one currency for another. When you buy an apple, you are swapping the currency accepted by the shop for the apple.

Consider a thought experiment where I own an apple orchard and you own an orange orchard, and we want to create a marketplace for people who want to buy / sell apples and oranges as well as swap apples and oranges. I have a certain number of apples (call it #Apples) and you have a certain number of oranges (call it #Oranges). We agree to put all our produce into a fruit shop ("liquidity pool") and agree that the total value of apples is equal to the total value of oranges in the shop. This gives an implied exchange rate between apples and oranges:

Apple price (in oranges) = #Oranges / #Apples        

and obviously

Orange price (in apples) = #Apples / #Oranges        

This seems weird at first. The price is just the ratio of the number of each fruit in the fruit shop. When the supply of apples in the shop decreases relative to the supply of oranges in the shop, the price of apples increases. Taken to an extreme, if the demand for apples is so great relative to oranges that there is just one apple left, the one apple is equal in value to all remaining oranges.

That doesn't seem right. How can these potential price distortions resulting from our arrangement be corrected?

Arbitraging Apples and Oranges

The answer is simple. If the price of an apple in our shop exceeds a reasonable market price, fruit traders are incentivized to sell apples to the shop in exchange for oranges, i.e., they swap apples for oranges. This swap will both increase the supply of apples in the shop as well as decrease the number of oranges in the shop. If enough fruit traders take advantage of this price discrepancy, the supply ratio will adjust itself until apples and oranges are priced reasonably again. Fruit traders make profit from these arbitrage opportunities while ensuring correct prices for apples and oranges in the process. It's beautiful.

Fruit Tokens

It is hopefully clear from the above how fruit traders can profit from arbitrage opportunities stemming from our arrangement, but why should you and I create this arrangement and provide, i.e., "stake", our apples and oranges in the first place?

We need to receive something in exchange for our apples and oranges, so we create a new kind of token, call it "Fruit", and issue ourselves equal amounts of the Fruit token so that

Apple price (in Fruit) = 1/2 x #Fruits / #Apples        

and

Orange price (in Fruit) = 1/2 x #Fruits / #Oranges        

where #Fruits is the total number of Fruit tokens issued by our shop.

Now, fruit traders do not need to own Fruit tokens to swap apples and oranges. However, anyone, including fruit traders, can stake apples or oranges in exchange for Fruit tokens. Conversely, anyone who owns fruit tokens can exchange them for apples or oranges.

In essence, Fruit tokens are a liability against the apple and orange assets of the shop and the total value of the shop (assets plus liabilities) is always zero.

Transaction Fees and Yield

Functioning decentralized AMMs in the wild are not much more complicated than the above thought experiment involving a marketplace for apples and oranges. The price of an asset is determined by its relative supply in a liquidity pool and arbitrageurs profit from taking advantage of price mismatches while ensuring reasonable prices for all assets in the pool.

An important addition is that liquidity providers earn yield on their staked assets via small transaction fees on every swap. This creates beautiful symbioses among liquidity providers who own assets and wish to earn yields on those assets, arbitrageurs who wish to profit from price discrepancies and those who simply require swapping two assets at reasonable prices for some business purpose.

Conclusions

Although practical AMMs were born out of DeFi, the symbiotic relationships that exist among the various parties would still exist even in a more traditional offchain setting. I, as many others have before me, envision a future where insurance risks of all kinds are liquid and tradable digital assets. The benefits that securitization brought to lending should bring the same or more benefits to insurance improving capital efficiency. I see insurance securitization as creating an entire new fixed-income asset class. The challenge with this idea is the lack of depth in the market to support a traditional order book for price discovery and market efficiency. AMMs solve this problem. No longer does there need to be someone on the other side of a trade. There simply needs to be a pool of staked assets on the other side.

Personally, I have been on a mission to transform the insurance industry for the past 10 years and I feel like we are at a turning point in history. If this article and / or decentralized capital markets in general including insurance capital management is of interest to you, then I invite you to join me on my new Discord server:

• The CavalRe

It is still early days so you might find me alone in there for a while, but I'd love to start bringing together a community of like-minded professionals interested in DeFi with a particular focus on automated market makers, (re)insurance capital management, securitization, and other crypto quant-related subjects.

PS: New Technical Foundations

While thinking through how to do DeFi insurance "right" back in early October, I had the proverbial "Eureka!" moment and am quite excited about what I found. I discovered how to build an entirely new kind of DeFi fixed-income market based on AMM principles. I quickly wrote up a foundational paper with my coauthor, Leo Liu , from BTX Capital :

• A Family of Multi-Asset Automated Market Makers

This paper represents a new kind of AMM for spot markets. In future papers (or possibly a white paper), I'll write up the details for building a crypto yield curve.

I am currently looking for partners interested in helping to realize these visions so please reach out (join Discord! ) and I'd be happy to explore possible synergies.