Can Decentralized Finance Concepts Improve Traditional Exchange-Traded Funds?
Dr. Zach Feinstein and his Stevens School of Business colleagues are exploring how automated market makers can be adapted to work within more traditional investments
The emergence of decentralized finance and automated market makers (AMM) such as Uniswap and Curve revolutionized investing by increasing market opportunities and market trust over centralized exchanges such as the now notorious FTX.
Zach Feinstein, assistant professor and director of the Fintech Certificate Program, and his co-primary investigators, Ionut Florescu, a research professor and the director of both the Stevens Financial Technology and Analytics program and the Hanlon Financial Systems Labs, adjunct professor Ivan Bakrack, the Head of Treasury at ConsenSys, and a graduate of the Stevens Financial Analytics & Data Science master’s program, are a year into their project, “Extending, Simulating and Scaling Decentralized Exchanges Made by Automated Market Makers.”
The project is funded by the Center for Research Toward Advancing Financial Technologies (CRAFT), the first fintech-focused Industry University Cooperative Research Center funded by the National Science Foundation.
“The idea of automated market makers is something that came out of decentralized finance (DeFi) and blockchain,” Dr. Feinstein said. “The overarching idea of this project is to see what concepts from DeFi and blockchain might be applicable to traditional financial markets. There are interesting innovations happening right now, but we also need to understand how these new constructions should be regulated especially if you were to apply them to traditional markets.”
Defining Key Terms
AMM: A system that allows people to trade assets directly with a pool of assets provided by individual investors, without needing a traditional middleman like a stock exchange or broker.
Exchange-Traded Funds (ETF): A collection of various assets, such as stocks, bonds or commodities, that you can buy or sell on a stock exchange just like a regular stock.
Frequent Batch Auction: A trading method where buy and sell orders are collected and executed at regular, predetermined intervals, with all trades processed together at a single price determined by supply and demand during each interval.
Liquidity Token: A digital asset, similar to a voucher, that represents a share of value in a pool of funds or assets.
Why CRAFT was Interested
Liquidity concerns led CME Group, formerly known as the Chicago Mercantile Exchange, to become a key sponsor of the project.
“From our perspective this is a new type of market maker that can sit on some of these exchanges; DeFi can help to shore up these markets,” Dr. Feinstein explained. “And if you're a large financial institution, you also want to understand what novelties are coming from this decentralized space.”
Stevens Infrastructure Powers Research
The research is being done using a two-pronged approach. The first is, “purely mathematical,” creating models to decipher how the systems should theoretically behave. The second part is utilizing SHIFT, a platform that realistically simulates high-frequency trading. The system was developed by Dr. Florescu and colleagues in Stevens’ Hanlon Financial Systems Lab.
“Over the course of this project, we've been looking at these decentralized exchanges to see what requires the blockchain, what can exist in traditional finance, and how these products actually affect financial markets through simulations,” Dr. Feinstein said. “Building up a true simulation system is a heavy task, so having the SHIFT system saved us years of development time. It meant that it took months instead of years to get the simulations up and running.”
Project Progress
The first six months of the project were spent exploring the DeFi landscape for what products existed and how they function. Those findings actually shifted the original goals, which were to construct new decentralized market structures and simulate the dynamics of these decentralized exchanges. The areas of focus now include ways in which decentralized finance can be improved and how some of the AMM characteristics can be applied to more traditional ETFs.
The first idea is the use of a generalized liquidity token.
“Right now, if you're the first dollar investing or you're the millionth dollar investing in an AMM, you're getting the same returns, but there should be some idea of supply and demand,” Dr. Feinstein said. “If I already have $1 million invested with me, that next dollar doesn't provide too much value. That leads to other implications about how people will invest, and what the value of liquidity is. That idea lends itself to some comparisons between automated market makers and ETFs.”
“We're looking at the key differences in pricing,” he continued. “ETFs will usually rebalance once a year or once a quarter. If Apple goes up in value, its market cap expands. An ETF tracking the S&P 500 should have more value in Apple at the next rebalancing. An AMM is doing this automatically through a kind of passive investing. Instead of saying it's going to update every quarter, it can update in real time. We're exploring if this might be a new way to construct and rebalance ETFs.”
Another idea being investigated is how fees are charged.
“If you're a market maker, you're charging fees whenever anyone trades against you,” Dr. Feinstein said. “That's the traditional way it's done. We are looking at trying different fee structures. What if we use the traditional fee structure? What if we use different formulas, what does that mean for the resulting market? In general, we're looking at different structures that we can place the AMM in.”
“In DeFi, automated market makers sit on their own and there's a bunch of different fragmented markets. In the United States, when you trade, your broker is required to provide the best price available across all exchanges. If you have the same stock issued on listed on the Nasdaq, the New York Stock Exchange and others, when you trade, you have to be given the best price that's listed on any of these exchanges. We’re asking how we fit AMMs into that system so that it's not just some external entity, but is incorporated into the National Best Bid and Offer system. We're also looking at ideas like frequent batch auctions and other matching engines which can also change how the market is running.”
Potential Applications
Expanding the number of market makers and more widely distributing the liquidity could help avoid widespread financial market downturns like the one in 2008.
“The core idea is to take this market-making business, which is big business for some of banks, and say that any individual can invest in that,” Dr. Feinstein said. “This project asks how far can we push these ideas and how they might alter traditional financial markets?”
“If you look at the data on who are the market makers for most ETFs, or markets more generally, the top three often comprise more than 90% of market liquidity. You have very few players, which means if you have a 2008-like scenario where a market-maker goes down, there can be outsized implications for the rest of the system. We’re looking at what happens if we augment this system with individuals working in aggregate to act as market makers; by providing additional liquidity, does this make markets more robust during bad times as well?”
What's Next?
While the research continues, Dr. Feinstein and the team will begin the work of writing academic papers and potentially obtaining intellectual property rights.
“The ETF application is one very interesting, clearly useful, concept because most investors are really invested in ETFs and not in the underlying assets,” he said. “If I'm going to buy the S&P 500, I would just buy this one share of SPY instead of purchasing 500 different stocks in the right proportions. If using AMMs makes it easier for people to do that, to essentially buy an ETF, but earn fees from holding that ETF, that seems like a worthwhile innovation. At the same time, we’re trying to make sure that there are no unintended consequences of doing this.”