Decentralized Finance

Defi in 2024: No Free Lunch

The future’s bright… if you know where to look

Written by Vitali, edited and polished by Jim Hawkins.

This year will feature a bull run in crypto for sure. But not in the way that you expect.

Some only draw hope for the future from the success of BTC ETFs or the imminent BTC halving—or both.

I think it’s better if nothing changes after either of these. And by that I mean it’s better that those trying to get rich easily fail to do so.

Why? Because this will force people to do some actual work. To deploy the right hemisphere of their brains, get creative, and build things that are useful for a user base larger than their Twitter following.

This is unimaginable for some degens in the space, who cannot conceive of making money except by buying a token, getting a pump, and then dumping it on retail.

But another path is possible. A richer path. It just requires work.

Crypto was built for DeFi. This destiny has yet to be realized — and it’s up to us to manifest it. In the manifesto below, I will show you how.

The Current State of DeFi

The very word ‘DeFi’ elicits all manner of opinions — most of them negative these days.

• VCs hate it. Most DeFi tokens are shit and have mostly been inflationary, slow-motion exercises in rugpulling thus far.
• Institutional adoption has simply not taken place. With the exception of certain adventurous family offices, ‘big money’ sticks to the CEXs.
• Even degenerate adoption is mixed. Many degens only use DEXs to farm airdrops.
• “One Uniswap to rule them all”: Uni remains the king of DeFi, with little immediate competition.

Hats off to the DeFi OGs: Curve, AAVE, Uni, Compound, Balancer, MakerDAO. These guys made it.

But let us explain how so much of the space has failed to succeed.

The story so far: A tale of ‘Bob’ and ‘Alice’

Meet Bob. Bob is a buidler. Bob attends EthDenver and many hackathons. Bob builds a DeFi protocol and launches a ‘governance’ token.

Bob is looking for liquidity to bring to his protocol (VCs told him he has to do so).
Bob also plans to raise capital at some point

Bob revealed his DEX idea to Alice, whom he met at a hackathon. Alice told him she was a ‘Market Maker’ (a term that is effectively a euphemism in crypto at this point).

Bob knows he needs Alice.

Bob sells Alice some tokens for $250k. Alice goes to a (Chinese) CEX and dumps the entire token allocation there. Bob watches his token price graph paint an L-shaped red/green diagram.

Bob isn’t happy with this situation, but he has little room to object. While Bob is smart, Alice is a smartass.

Everyone is (kind-of) happy… Except Bob’s VCs. They wanted a 10x from Bob’s tokens. They won’t get anything but heavy bags unless Bob saves the token price from a death spiral. They tell Bob they won’t invest in Bob’s next project if he doesn’t make the token grow in price.

Bob asks Alice to help. She tells him to fuck off says ‘it’s all good’.

— — YOU ARE HERE — —

That’s what DeFi 2020–2023 has been so far.

VCs basically stopped investing in DeFi in 2023. You will read a lot of news to the contrary but this is mostly copium.You won’t see $500M rounds of investments in DeFi protocols anymore. This was only ever going to happen once.

What we may see is a divide forming between two types of projects: Those that worked and those that haven’t.

What’s common about the projects that worked? Binance, OKX, Kucoin, on the other side of this (de)centralization range are Curve, Uni, Comp?

These guys serve well defined customer segments, offering the products for them. The product is the prize. They don’t have the airdrops anymore and they don’t need them.

Basically, CEXs offer the products everyone needs:

• Institutions (hedge funds and trading firms where Alice is employed) buy DeFi token allocations from the developers and dump them on the CEXes.
• Retail traders that do DEX-CEX arbitrage, gamblers retail investors that buy tiny token allocations. There’s also a bunch of sketchy schemes with P2P and exchange.
• DeFi OGs offer steady revenue strategies to the same, mostly institutional customers.

Their tokens are fine, they aren’t getting hacked (that Curve + mansion situation was just weird), and they are doing just fine.

So why can’t we just leave things like this?

The most ‘innocent’ reason is that Uniswap, AAVE, and Comp are slow and expensive and can’t be associated with good trader experience.

But more importantly…

(Imagine Gary’s photo). This guy doesn’t want it. He has been busy. He has raised $324B. And he’s coming for you.

He wants cryptos to be securities, not currencies/money.

What DeFi ‘should’ look like

If you made it here and you’re actually interested in DeFi, behold: This is what the future looks like.

2024 will forever be known as the year when we started trading on-chain.

Because in 2024, on-chain order books finally start to work.

Let me explain. The world of modern finance is built on the back of what is called HFT (High Frequency Trading), and it comes with a set of tools:

• (Central Limit) Order Books aka CLOBs. These are the backbone of the modern financial system. It’s where all the economic bubbles are inflated and popped. That’s where the 2008 crisis started, that’s where it will end. That’s where Uber, Wework, dot-coms and other horrible events occurred. You need an order book to make a modern economy.
• Algorithmic strategies, aka bots. These are the set of tools used by the institutions to trade and move markets to one direction or another on the order books which run the global financial system.
• Websocket APIs and data interfaces. This is how you plug your algorithm into the CEX.

Crypto doesn’t have any of these. CEXs have done something similar, but these aren’t crypto products as they are mostly offchain.

Why doesn’t DeFi have any of this you might ask?

Because it wasn’t technically possible… Until now.

Arguably Binance and other CEXes have achieved a great thing. They made millions of people excited about crypto. They have formed huge cohorts of traders and influenced the world economy in a major way. They have changed the behaviour of people forever.

We couldn’t put it all on-chain before — but we can now.

Pushed to the on-chain frontier

No-one is going on-chain voluntarily. Innovation is the mother of necessity.

In this case, big Gary G is pushing people into the world of DeFi by forcing us to abandon CEXes and other off-chain tools.
Do most degens care about the difference between on-chain or off-chain? Gary cares.

TL;DR: As the SEC is regulating CEXes, DEXes become a Zion for crypto traders, a promised land, a secure ground to run your ponzi schemes algo trading.

So Bob from before finally seizes the initiative and starts building what his customers want him to build (making tiny breaks for a beer and a couple joints).

AMMs aren’t the solution

Capital should start migrating on-chain once the chains allow for the order books and HFT.

To be clear, fully-fledged HFT likely isn’t happening this year. The blockchain industry isn’t ready for it.

HFT requires at least 20k TPS (two traders will kill the entire bandwidth with this throughput). And yes no matter what you read on ‘Twittuh’, Arbitrum is the best you get in the Eth ecosystem with its 40 tps, while Solana will allow you to run something like 2500 tps securely (it might crash right after though).

On-chain and off-chain order books have only just started to appear. Anything else before has been a so-called AMM: an Automated Market Maker. There are awesome examples of AMM based DeFi products: Uniswap, CoW, Curve, and a lot of other OGs have laid the foundations of this industry.

But AMMs are not secure, as they require users to lock their funds in smart contracts. A lot of funds in just a single contract are a glaring target for hackers. Remember the movies where they rob banks/casinos? The smart-contracts in AMMs are digital vaults asking to be robbed. So AMMs are suited for a different product — using them for trading is a workaround.

CLOBs are essential to the survival of DeFi

TradFi is CLOB-addicted. It’s been the default way to trade since forever. It’s not the tech per se, it’s a set of standard practices for trading. Especially important are APIs allowing a trader to build their strategies on top of the product, bypassing the UI (speed matters).

Very soon there will be a couple hundred order books, both on-chain and off-chain, appchains, order book chains etc, offering all sorts of advantages and somehow promising more on-chain stuff.

Why so? Ok so there are 4 things traders are looking for:

• Tokens
• More tokens
• Volumes
• Pay less to the regulators

These things come in a sequence one after another during the lifetime of a DeFi project.

DeFi projects are looking for ways to provide all of these.

I mentioned tokens twice partially as a joke, but also because most of the institutions are actually looking to accumulate both generic project tokens and DeFi protocol tokens as a prize for providing liquidity to both. What happens to those tokens stays outside this article. What you need to know is that the tokens are normally used to bootstrap market liquidity (part of the liquidity mining programs of both a blockchain and a DeFi protocol).

Then comes the volume. TL;DR a DeFi protocol should maintain solid liquidity within it to allow everyone to profit from engaging with the DeFi protocol.

I’m providing this context to make the following argument: DeFi is slowly going to drift from the AMMs to the CLOBs. This will initially be an off-chain process (whoever calls Cosmos SDK + Tendermint stack ‘on-chain’ is wrong. But it’s still the only one working at the time I write this).

We’ll also see all sorts of DeFi infra: Sequencers, relayers, Gas rebates, batch processors, meta transactions, All sorts of L2s and L3s, with or without ZK stuff in there. Everything is built to allow all counterparties involved to profit and then profit even more from dealing with the DeFi protocol, blockchain or ecosystem. It’s also done to minimize or ideally get rid of the fees.

Traders hate fees. Small protocol fees, or better, the absence of fees, directly influences the profit of a trader.

We at sprk.fi also build a CLOB. We like it because the idea is to stay on Ethereum, not withdraw anything off-chain, and still maintain decent volumes.

Now institutions and even mid size trading teams automate their strategies by building algorithms (I like calling those the bots). They plug their bots into a DeFi protocol via an API, automating their work to capture more volumes.

(Funnily enough, only the CEXes were good at providing decent API sets until now.)

The key to DeFi in 2024

Whoever starts onboarding traders from the CEXes seamlessly, will win the DeFi game. This is the endgame for 2024.

Which brings us to the next important question: Wen capital on-chain?

This is tricky. Let me explain:

Blockchains are still slow. Yeah I know, Twitter says there are hundreds of super fast chains with hundreds of TPS. Why then, is everyone either building off-chain execution (via Cosmos, 0x, or Seaport) or sticking to the AMMs? Because nothing else works. A couple dozen companies are working to fix this problem (including us at Compolabs), but the time to market is around 6 months, and the time to adoption is more. We’ll start as Perps + Spot, introduce a CLOB to build on as the next step, and scale into an Appchain next year.

Yeah, that’s all. Some ask Okay what about all those RWAs, pools, 450 various types of Perps, decentralized treasuries etc? Patience frens. I’m the infra guy so I say: Infra first, liquidity next. First, we need to lay the basement.

Exciting times ahead!

Wishing everyone a great year, wagmi!