Less Is More: Underestimated Contract Features for Institutional Adoption | ETHTaipei 2026

So, hi everybody. Um my name is Anton and today the topic of the talk is less is more and what I learned as a solutions engineer at Morpho when we are trying to like approach institutional clients um you know, from like a contract perspective contract design perspective. And um yeah, let's get started. So, a little bit about myself. I've been I've been in the DeFi space since 2020 and in the past as I've been uh a protocol engineer for a few DeFi option protocols and I've been building uh spent the last 2 years uh building a independent front end on top of Morpho uh until a few months ago that I joined and I'm also a former E-Taipei co-host.

So, feel free to talk to me about anything any idea you have uh that's around about this conference. But basically um what I want to cover today is like a very brief introduction about Morpho, how things work, um the different components and we're going to talk about walk you through like a story that I found um interesting and how we solve it and how we learn uh from the story.

[snorts]

So, a little bit about Morpho. Uh Morpho is now one of the biggest lending protocol on EVM. Uh we have about 14 billion dollar um locked in the contract across uh dozen of networks and we have like uh 5 billion active loans taken out and the prime example of integration we have in the last 2 3 years were uh Robinhood and Coinbase. So, you might heard that you know, they offer some lending or borrowing um feature. And those are powered by Morpho underneath.

So, if you look at this diagram on the right, you can kind of you can kind of imagine Morpho as a two-layer architecture, right? On the top layer, there's Morpho Vault that has lots of great feature to improve the UI UX, um, and underneath it's the Morpho Blue layer, which are markets. And that we can we can dive into that. Okay, so let's talk about the markets. So, the markets are the immutable layers at uh, which is sitting at the core of the whole system.

That's also the contract that's holding your funds and then actually managing all the lending logic. There are more than two, three thousands of markets on Morpho. Anyone can come in at any time and then create a token pair. You just have to choose a collateral asset and a loan asset and some risk parameters. So, anyone can spin off any markets.

That means anyone else can also come here to either lend your asset to earn some interest or to use the collateral asset, um, as collateral to borrow, uh, the loan asset. So, because of the nature of permissionless and immutability, we now see, you know, like thousands of markets on Morpho. And also, there's no concept of ownership at the market level. So, market level is like totally static. And the vault layer on top, you can think of it as a delegation or management layer.

So, our thesis is that no user is actually going to go on chain and manage, uh, where the interest is coming from. This is probably like a professional job or some some strategy you you, you know, you you trust Robinhood or you trust someone else to do it.

[snorts]

So, it's very natural to like pool people's fund together and then ask like a role what we call curators to do the uh heavy lifting of uh managing liquidity, managing risk, and they can also charge a fee on top. So, this is kind of like high-level how Morpho work. Morpho midnight is something um I just want to mention briefly about uh not just Morpho blue, which is a variable rate lending market. Morpho today also offer fixed rate lending. So, a vault here, if we dive a little bit more into it, you can see on the diagram on the right, you can see a Morpho vault can go through different adapters to deposit into different type of markets, which is fixed rate market, uh variable rate market, and also it can also put money in another vault.

So, we can kind of like piece all the Legos together. And on the left, let's focus on all the roles we have in a Morpho vault. So, what we define here is basically a owner who can set uh whatever he wants or like appoint uh roles to to the infrastructure. But, the most important role here is the curator, which I highlighted. And also, the curator is the one that decides where the funds go, right?

So, curator is basically what we imagine to be like asset manager uh kind of kind of position. And there are other roles that, you know, like our contract designs, like allocators. It could be like a key that can execute certain trades. Uh also, there's like a sentinel, which is like a guardian kind of role that help you de-risk. So, they can only pull assets out from the the adapter or markets, but they cannot add the risk, which is putting money back in.

So, these are like the basic structure and also like the golden path of what we imagine like a business would do, which is a curator running a vault and then anyone deposit trusting the curator about how how it does things. But here comes a problem. Because when we design the contracts, we didn't exactly know how will people actually run the business, right? So, what we learned recently is that there are actually lots of uh when when we approach traditional finance, they love to split things up or, you know, maybe have a contractor or maybe have a third party um uh company that they may they maybe have a business agreement with to work with them together to deliver this product, right? So, let's take Think of Alice as example.

Let's imagine Alice has a bank, um but she doesn't really know how to run all the on-chain lending stuff, but they want to offer this as a, you know, yield source for all the users users. What Alice will control is she will want to control the fee and monetization, and they will want to have another agreement with someone Let's call Bob, who's like a DeFi expert and do all the risk management. So, how can we set this as set this up on-chain, right? Because like I said before, in a Morpho Vault, uh we only have one role, which is the creator, but creator have all the finance-related or strategy-related um authority. So, they have lots of power to do lots of things.

What we are trying to do uh like at the vault In this scenario, what we are trying to do is we're eventually trying to split the role, right? We want uh to let Alice control maybe the monetization and the fee, but we also want to make sure Bob, which is the creator who's going to operate the vault, cannot change the fee. So, very intuitive solution you may think of is What if we build like a third like a wrapper contract, something in the middle that handle these kind of logic? But this will become very very um tiresome when you when when someone come to you asking for a solution and you ask them to build something new. So, it turns out what we have uh the the best solution we have so far is what we call like a fee wrapper or I want I like to refer it to as a dual vault setup, which is we can spin up two vaults.

And in one vault, we let the curator be bank of Alice, right? So, the banking the Alice can control all the fees. And we set up the vault in a way that we direct all the liquidity to underlying vault, which is controlled by Bob. So, Bob can take the money and then allocate to markets and everything. But, here comes another problem, right?

Because now we have another trust assumption because what if Alice pull the money out and then put the money into another vault? Or what if the Bob, uh who who is supposed to not charge any fee, um maybe like jack up the fee and then influence uh the monetization on on the parent vault? Right? So, there are more um we kind of raise more issue by this by this setup. But, thankfully, um we have a very cool feature on on the vault, which is called abnegation.

So, we have a function on the Morpho vault contract that is basically allowing any owner to permanently disable a certain feature on chain. So, what after you call the function, let's say uh Alice, you call the abnegate function on the adapter-related uh functions, uh you can no longer use this function at all. So, what does it enable us to do is we can coordinate a setup on chain which these two parties, they all abnegate or disable a certain set of functions. And by so they can kind of like fix the setup on chain, right? So uh there will no trust there will be no trust assumption between the two parties after after the setup.

Yeah, so with this uh with this uh structure, we successfully coordinate the two parties that, you know, fit the fit the scenario. Another extra thing on top is that also it's a Now it's I think it's a very common pattern um that's called gates. So it's quite common now that protocol have this gates feature that they can control who can enter or exit a a certain protocol. What we can do now here is that it's very often that when we talk to institutions or uh fintech companies, they will certainly focus on um compliance, right? So maybe like only compliance user can deposit into a vault.

We can do this like perfectly fine with with all the gates, right? So Alice controlling the distribution in the front end will then maybe work with a KYC owner and then sorry, KYC vendor and then say only KYC user can deposit into the vault. So if we combine this with the previous setup we have, we end up having a very very natural setup, which is you can have multiple different um you can have one one to many uh entities, maybe across jurisdiction. So for example, maybe Alice is like have two regulated exchange in two different jurisdiction, Taiwan and Japan. She can then have two different vaults and then set up uh the gate with a different set of KYC vendor based on regulations, everything, but eventually all their funds can be um routed into the underlying vault that is like actually running the strategy.

So, with this kind of setup, what we can achieve is that uh, we can coordinate lots of different entities and lots of different parties on chain, even though you know, we without adding any contract dependency and also just like lower their trust uh, trust assumption among each other. Yeah. So, one last uh, point I want to make is that uh, when it comes to regulatory and compliance issue um, I think lots of protocol doesn't really consider how can we help people get out? With a Morpho infrastructure, because we have Remember the vault layer is strategy, right? Like the strategy layer, but underneath uh, layer is immutable and and static.

What we can do is we can allow user to use flash loan to like um, convert their vault share uh, vault position directly into like an underlying market position. So, what would this do? Is that uh, from a compliant perspective, we can offer very, very clear two-layer architecture. So, it's very clear that if you put money in a vault, there's some kind of delegation happening. Um, there are some clear regulations around how to do this.

This is and also uh, sorry, just the second point. Uh, we can uh, enforce self-custody. Because at any time a user can revoke the access. So, it's not tied to liquidity, even though there's no liquidity in the pool in the market, you can still opt out. So, after you opt out, no more um, no one else can mess up your position or move your funds around.

Um, which is like a very cool feature we get to argue like, you know, we're like neutral vault or like a infra provider that turns out to be like a good argument Uh, um, for to to to regulators. So, yeah. That's basically the topic today. I think the takeaway is gates integrated with the application function, which I think is the most underrated feature, because I think what I learned so far is that lots of times when we talk to picking institutions or just like other parties, it's not about what you can do, is it's what it's about what you cannot do. Cuz sometimes we give people too much power or any rule too much power, but it turns out to be like a regulatory issue when it comes like real world adoption.

So, it's a bit of a little bit counterintuitive, but sometimes restricting the restricting the power can actually help us a lot. So, I hope that's a good takeaway. And um

Okay, thank you, Anton. We actually have a few minutes for the cute questions. So, anyone has questions?

Uh so, as you may have have a conversation with some Taiwan financial institutions, what's their one is there a one talk Do they want to have some restriction to those like SM ledgers? Uh accounting to for those like yield strategies like those SM SM ledgers could only use like lending and no staking or do they have some restriction to those SM ledgers yield strategies? And what is their um attitude towards entering to the DeFi space like vault something like that?

Yeah, honestly, I'm not the one that's talking to all the institutions, and especially Taiwan is like very strict currently. We don't have that much direct information about Taiwan's regulatory thing, but I guess like the overall sentiment that we have in Asia, uh Singapore, Japan, um I think honestly it's like a far it's a it's a far road ahead of us that I think probably what we can expect very soon is just stable coin. Uh I think lending is actually like the easiest, right? But what we have to package it is like very similar to, you know, like maybe like a credit uh private credit fund, right? Cuz private credit funds is like similar in a way that they pool people's money, and they let some manager and some regulated uh people to to like lend out the asset to someone else.

So, the closest example we have is those private credit funds. So, probably So far, my knowledge is that similar regulation would apply to these kind of rules.

Thank you.

Thank you. That's all the time we have for the questions. So, thank you, Anton.

[music]