Exchanges Are Open Finance

The largest profit centers of crypto—the exchanges—are generally under analyzed from a strategic lens, especially given their systemic importance to the space as a whole. This is especially true given that the public can invest in the success of the exchanges through each exchange’s respective token (with the exceptions of the US-based exchanges, BitMEX, and Deribit).

Crypto commentators like to discuss what can cause the exchanges to stumble and fall. But the questions that almost no one asks are: What can the market reasonably expect exchanges to do over the coming years to consolidate even more power? What if the exchanges haven’t peaked, but instead are only in the beginning stages of their ascent?

Note: for the remainder of this essay, when I use the term “exchanges,” I’m referring to non-US exchanges, with Binance, Bitfinex, BitMEX, CoinFLEX, Deribit, FTX, KuCoin, Huobi, and OKEx being the top-of-mind examples. Given the regulatory regime in the US, it’s difficult to see US-based exchanges executing on the vision outlined below.

When Capital Aggregates

Exchanges are naturally subject to network effects because traders want to trade on the most liquid venues. While there are some limits to the network effects of liquidity as arbitrageurs can bridge liquidity pools, the network effects are real (hence why so many upstart exchanges have failed to gain traction). As a result of the network effect of liquidity, capital aggregates on exchanges. It’s estimated that 4.73% of BTC and 8.66% of ETH are held on the top 7 exchanges.

Crypto exchanges have been steadily expanding their purview. Most of them started by trading spot assets. Then they started adding more advanced trading features such as lending, margin trading, and futures. While these products are very much in the purview of traditional exchange businesses, the crypto exchanges have also launched new business lines that are outside the scope of traditional exchange businesses. A few examples:

1. Bitfinex acquired Tether in 2014, which has become a vital part of global crypto capital flows. Today, Tether is the dominant stablecoin by virtually every measure: market cap, liquidity, daily transaction volume, and addresses holding it.
2. After the Chinese government banned crypto trading in September 2017, Huobi and OkEx didn’t shut down. Although they did stop direct fiat-crypto trading, Huobi launched an OTC trading app, which is analogous to Local Bitcoins. This app allows Chinese consumers to buy and sell crypto in a truly peer-to-peer way. They can even settle trades using Alipay and WeChat Pay. More recently, Binance introduced a directly competitive offering. These OTC venues facilitate huge volumes, and are generally not reported in any of the official trading statistics.
3. In the last year, Binance launched Binance Labs, which acts as both a venture fund and accelerator. Just as Y Combinator started with an early-stage accelerator program that has since gone downstream with a larger venture fund, Binance Labs has expanded downstream with Launchpad, an Initial Exchange Offering (IEO) platform. By underwriting IEOs, Binance has started competing as an investment bank. In the last six months, most of the major exchanges have followed Binance’s lead and launched an IEO product.

More recently, we’ve started seeing exchanges offer crypto-native financial services:

1. A few exchanges have launched or acquired decentralized exchanges (DEX): Coinbase acquired Paradex, and Bitfinex built Ethfinex in house. A few months ago, Binance launched Binance DEX as its own chain (rather than on top of an existing chain). Huobi and OKEx have announced their own chains, but haven’t launched them yet.
2. Non-custodial wallets. Coinbase built Coinbase Wallet, a non-custodial crypto wallet. Huobi did the same with Huobi Wallet. Binance acquired Trust Wallet.
3. Earning crypto-denominated yield. Poloniex began offering ATOM-staking rewards about 6 months ago, and Binance has quickly followed suit by rolling out an Earn tab that includes staking for XLM, NEO, TRX, STRAT, ONT, VET, KMD, ERD, FET, ONE and ALGO. Binance’s Earn tab also includes lending.

With this positioning, it’s clear that Binance is trying to help their customers use crypto as a way to generate income. Note that this is basically the same positioning that virtually all of the non-custodial wallet aggregators are using (e.g. crypto.com, Instadapp, Nuo, Argent, etc.).

The trend is clear: Crypto exchanges are (re)building every major financial service from 100- year-old services like lending to crypto-native services like staking. What’s more, the pace at which exchanges are rolling out new services is accelerating.

Given their position as aggregators of capital and their interest in building both traditional and crypto-native financial services, the exchanges are perfectly positioned to catalyze the adoption of Open Finance.

Catalyzing Open Finance Adoption

Some of the initiatives highlighted below have already been announced. Others are conjectures. While I can’t know with certainty what each exchange will do, given the trends highlighted above, the salient question is: Why won’t exchanges continue to expand their financial services offerings? The question then is not if the exchanges will support these services, but the relative order of operations for each exchange.

1. In the LEO white paper, Bitfinex announced a few upcoming products, including a regulated securities exchange, and Betfinex, a prediction market platform (likely to be directly competitive with Augur).
2. The major crypto exchanges maintain custody of billions of dollars’ worth of assets. Someone will create a payment network that connects debit cards to the assets users store on exchanges, including stablecoins like USDT, USDC, BUSD, and DAI (crypto.com already offers a debit card for assets custodied with them). This is strategic for the exchanges as it creates more liquidity for their stablecoin/fiat pairs. Taking this a bit further, It’s pretty easy to imagine Binance offering discounts for customers who pay merchants with BUSD to catalyze adoption.
3. With billions of dollars of assets in hand, it’s only natural for exchanges to offer interest-bearing accounts. The exchanges have a number of ways to enable this. They can stake assets on behalf of users (where applicable—some are already doing this), run internal money markets, or leverage open finance money market protocols. I’ll come back to money markets in the next section.
4. Coupled with a lending market that automatically accrues interest to a customer’s Binance account, Binance will feel like a checking account to many (especially outside the developed world, where FDIC-like insurance is much less prevalent). Moreover, with P2P fiat-crypto on ramps, the crypto exchanges offer consumers a clear path to exit their existing fiat money systems—both in terms of technical infrastructure and monetary policy—and opt into the crypto financial system with permissionless and censorship-resistant payment rails and transparent and auditable monetary policy.
5. If we think further into the future, once the exchanges offer interest-bearing accounts at large, they’ll start to think of themselves as banks. And if they think of themselves as banks, they’ll start to offer credit. While there are many forms of credit they can offer, they’ll likely start with small lines of credit—in the form of secured credit cards—before eventually moving into unsecured credit and larger lines of credit.

I can keep going, but I think the point is clear: The exchanges aggregate capital, and they can strongly influence where and how that capital flows. Moreover, because most of the exchanges have native tokens that reward users for staying in their respective ecosystems, both the exchanges and their users are incentivized to… build in and stay in their respective ecosystems!

Now, let’s revisit money markets and consider forking in open systems.

To Fork or Not To Fork

Rather than hosting internal money markets, it’s in the best interest of the exchanges to leverage open finance protocols so they can aggregate external liquidity to offer better rates for their clients. However, it’s less clear if exchanges will provide liquidity to existing protocols like Compound, or whether they’ll fork Compound and bootstrap competitive networks with their own liquidity pools.

Given how much larger the exchanges are than standalone liquidity pools like Compound, I suspect the exchanges will fork Compound and create their own pools so they can exercise more control, and create more value for holders of their respective tokens (e.g. by funneling profits from the money market to holders of their respective exchange tokens instead of a third party like Compound equity holders).

To make this more tangible, let’s consider Binance, which has already released the centerpiece of its open finance strategy: Binance Chain. On top of the base chain, one can start to see the rest of the strategy come into focus: Binance will leverage Trust Wallet as the preferred (though not exclusive) way to interact with Binance Chain. They’ll probably want to leverage key management systems like Torus (from Binance Labs Season 1 accelerator class) for seamless key management.

They’ll either fork Compound and Maker themselves—or incentivize others to do so via Binance X—to bring these protocols into the Binance Chain ecosystem, and direct the natural profits of those protocols to BNB holders. This in turn incentivizes more people to hold BNB and trade on Binance (Chain), increasing liquidity for all ecosystem participants, and creating more value for BNB holders in a virtuous cycle.

Note: a valid counter to this is that Ethereum is more trust-minimized than Binance Chain, which both 1) requires trusted gateways to bridge non-native assets, and 2) still has relatively permissioned consensus. However, there are pretty clear solutions to both of those problems in the form of Cosmos’s Inter Blockchain Communication (IBC) protocol, and permissionless Tendermint consensus. I expect other exchanges that don’t launch their own ecosystem chains to fork protocols like Compound and keep the forks on Ethereum, preserving the same trust-minimization guarantees, while leveraging their existing liquidity to bootstrap network effects.

What I’ve described above in many respects sounds unfeasible. How can it be possible for a single company to offer all of these services, when each of the markets above is comprised of dozens of disparate companies in the legacy financial system?

There are, broadly speaking, two reasons why crypto is different.

First, it’s crypto! When all forms of value transfer—currencies, commodities, bonds, equities, etc.—are powered by a permissionless and censorship-resistant API on an open ledger with generic, universal, reusable, and composable smart contracts, it becomes at least 10x easier to build financial services. This is a major reason why Binance has been able to move so much more quickly than legacy financial services companies.

Second, it’s unlikely that a single company will build all of these services. The exchanges are going to push development out to the community, and encourage developers to build in their respective ecosystems that are powered by their respective ecosystem tokens (as noted above, this can be on a separate chain, or on neutral chains). Moreover, they will aggressively fork other people’s open source code (e.g. Compound and Maker) because the source of network effect in all of these systems is capital, not code.

This will happen more quickly than people expect: Binance is offering developers grants to port Open Finance protocols to Binance Chain today. This is happening everywhere in crypto: Open Libra forking Libra, Solana implementing Libra’s Move VM, Tezos implementing Cosmos’s Tendermint consensus, Tezos implementing Zcash’s SNARK circuits, Solana leveraging Filecoin’s PoReps, and more. The primary beneficiaries of this recombinant forking will be entities and ecosystems that are already subject to network effects, like Binance’s.

From Companies to DAOs

To Open Finance believers, everything I’ve described above sounds… like the antithesis of Open Finance. Indeed, I felt similarly uncomfortable the first time I reasoned through all of this: What if the centralized exchanges aggregate all the capital and don’t decentralize?

However, due to regulations I increasingly find it unlikely that centralized exchanges can aggregate all the capital and financial services under one economic system. As the exchanges accrue more power, they’ll increasingly become subject to an array of contradictory regulations. Politicians will rip them apart. See Libra.

The best way for the exchanges to survive in the long run is to decentralize themselves. Tokens are the only way to achieve the political and architectural decentralization necessary to be maximally resilient. Tokens achieve this by unifying disparate and distrusting parties with shared economic alignment.

Today, the exchanges are centralized companies. But they’re slowly evolving into DAOs.

Decentralization is not binary. Decentralization is a spectrum. Therefore decentralization does not occur at a point in time, but as a process. As of now, Binance is the most evolved in this transition, but it has only barely started the process.

Today, exchanges are the best-positioned entities to onboard billions of users into crypto. As they morph into DAOs, the early participants in their respective ecosystems are likely to be rewarded.

Hat tip to Jesse Walden, Micky Malka, Nick Grossman, and Nick Shalek for providing feedback on this essay.