DeFi Weekly #51

Dai cap raised to $100m, Crypto lending markets lose their shine, dYdX moves away from 0x

Latest this week

  • Hello $100M DAI ceiling cap! It's taken almost 4-5 years for the Maker Foundation to reach this point. Since the stability fee has dropped to 5.5% we've seen more whales take out more leverage on cheap ETH. It's unfortunate to see how lending rates have dropped from 15% all the way to 5%. Suddenly lending your money on DeFi doesn't really make sense as you're only beating the banks by 3% maximum but also taking on a lot of unquantifiable risk. Compounding this problem if you live outside the US then the fiat currency conversion better not be more than 3% otherwise you lose out on any interest that you may have accrued. High interest rates are a great use cases, provided that they can be high over a 5-10 year time horizon. As a space we got a glimpse of what practical blockchain applications could look like but like most things, I think it's still in the prototype stage. Dharma, Linen, Outlet and Juno (relayer on Nuo) will be interesting to see play out. Dharma's $ locked up has decreased to $800k in the past few days. Juno has a better chance as Indian lending rates are close to 0 however we'll need liquidity in INR stable coin lending pools - which the space is still too young to cater for. I've covered this in previous issues of this newsletter but the biggest limiting factor to crypto markets at this point in time is liquidity. Without it all the amazing ground breaking use cases we come up with aren't worth much. A useful mental model I find to think about this stuff is as follows:

    • Objective 1 - build a global and secure base layer network that enables composability (Ethereum)

    • Objective 2 - have some behaviour/objective that drives liquidity from the real world into crypto

    • Objective 3 - create financial products from liquidity in new interesting ways (DeFi industry today)

  • For the past few months I think a few of us in crypto were excited about the fact that objective 2 could be fulfilled by high interest rates and savy retail investors buying in. However with sliding interest rates I don't see this thesis playing out as strongly in the short term. The only counter argument would be that 3rd world countries could benefit from even 4% however for that to work: either USD needs to be the primary currency they care about and can get in with minimal slippage (big shot with current regulatory environment) or native currency money markets (ETH/DAI/USDC) are king at the moment. The only unexplored thesis for objective 2 could be NFTs...

  • The second interesting point I found this week was dYdX rejecting support for 0x v3 moving forward. Why? Because their native ETH/DAI market is better than 0x's and they don't want to value to accrue to ZRX token holders. On the surface this may not seem like a big deal but I think it gives us hints as to what the future of value capture may look like. Unpacking this slowly, first of all dYdX can still support and most likely will support v2 since a large amount of liquidity still exists there. The reason why they can choose to reject v3 is because infrastructure can't be undeployed and v2 will continue to exist forever. Effectively this means that open source infrastructure requires new network effects to be re-established whenever a new update comes out. ZeroEx being ZeroEx has a large relayer community that it supports, and by being part of that community they're also locked in to ZeroEx's utilities. For example, if you ever integrate into the ecosystem there's a lot of helper libraries and abstractions that help non-blockchain heavy teams establish and plug into relayer networks. If tomorrow all of this code only supports v3 and support for v2 is slowly dropped, more teams might just think it's worth biting the bullet rather than investing the amount of engineering to ensure backwards compatibility. dYdX was able to bootstrap their initial application through 0x, ETH2DAI and Kyber liquidity pools, however once they build up sufficient network effects they moved away from 0x (the only off-chain orderbook) to their own orderbook and existing on-chain liquidity pools. Long story short, applications that grow large enough can effectively build the underlying stacks that they sit on top of. Forks aren't the real threat in open-source infrastructure, retaining large players is. To make this point clear, DDEX also followed in a similar suite. Although they claimed to fork, what they really did was move off 0x once they got large enough and built their own infrastructure that suited them. Extrapolating forward could it be that once interoperability finally works (5-10 years from now) everyone builds on open permissionless networks until they can actually create their own chain and cannibalise the layers below them? DapperLabs is attempting to do this with Flow but I think the time is too early for base layers to be "forked" out from the brand's user experience.

On Chain Statistics

  • Total Locked in DeFi: $650M (ETH ATH).

  • Biggest Gainer: Synthetix up by another $10m reaching $100m locked up!

  • Interesting Stat: Augur continues to fall down the ranks to a mere $600k locked up.