"Ethereum Has a Funding Problem - Here's the Fix "/ Devansh Mehta x Ethereum Foundation

[music]

Hi, my name is Devansh and I work at the Ethereum Foundation in a team called funding coordination. Which maybe in a broad way is looking at succession planning. That like how can the ecosystem itself come together and fund what we all rely on.

[bell]

So we are doing like various ways of trying to increase funding at the top of the funnel. I think like as a whole the public good space in Ethereum has focused too much on allocation that once we have a pot of money how do we distribute it amongst different projects and too little on where the money comes from in the first place in a sustainable and sustained way which doesn't keep relying on periodic donations. So we are basically like there's so much value flow going on in Ethereum. How do we uh How do we like route a percentage of that into funding the uh entire like the commons the Ethereum commons. Um so that there's very low overhead and it's programmatic.

So I guess like really trying to solve Ethereum's public goods funding problem is what our team looks at. I think the main problem is that a lot of it in the past relied on tokens and so when some team is launching a token they keep a percentage for uh protocol guild or for grants programs. And I think what we actually need is a percentage of fees which actually gives like I mean When I speak to a lot of teams their biggest issue is reliability of funding. That they feel that um they have to keep applying for periodic grants which are very intermittent. So it's hard to do forward planning with that.

But when it actually becomes a percentage of revenue gets routed to you then you can actually do some financial modeling projections hire team hire team members while being able to guarantee them employment longer than the period of the grant. So I think that's the real switch we need to make from one-time big donations, which you can't really do planning on, into embedded value flows, which you can then hire people with longer guarantees on. So, for me at least a public goods have a very specific mathematical definition, which is value creation minus value capture. So, there are some like Uniswap, which have created value, but they also managed to capture a lot of the value. So, that's not exactly a public good.

Whereas another example might be um like the Solidity team. So, so like it creates a lot of value. Every smart contract or most smart contracts are built on Solidity or on Viper, but did but they don't manage to capture any of that value. So, that then does become a public goods problem because um if you don't fund them, it results in something called tread weight loss. And what this is is uh people want to make a transaction, but they can't coordinate amongst each other to actually make that happen.

So, I think we are seeing that with Solidity, where a lot of teams do need a smart contract language and they need it to be updated and modern, but they can't coordinate amongst each other to actually get the funding to Solidity. And as a result, the value it creates is way higher than the value it manages to capture. So, privacy like for me is like selective disclosure. Um so, there are some things I don't mind if they are public. It's like it actually reminds me of this very funny story about this privacy activist talking about Facebook in the early days, 2011 to 15 times, where he was initially very opposed to Facebook and like like protesting about its privacy implications.

And then one day he changed his mind, and he said, "I actually figured it out. What I want private stays over here. What I want public gets put on Facebook." Which was actually a neat way of thinking about it, that um not everything needs to be private, but uh the right to decide what should be private and shouldn't be private, that should belong to us. And so, that's how I see privacy as selective disclosure.

Privacy also suffers, I think, in some ways from the public goods problem that a lot of the tooling is not easily monetizable. And if you do monetize privacy tooling, then it becomes less useful. Um so, it's like what Vitalik referred to as a revenue evil curve where uh the more you try to monetize, the less value creation is done. Um so, I think that's the kind of dilemma with privacy as well, that do I build a nice solution and make it open to everyone or do I make it closed so that I can capture more of the value that a privacy pool that that a privacy solution creates? So, it's a similar thing about value creation and value capture and whether privacy is um by trying to capture value with your privacy tool, are you resulting in less value creation, which is one end of the spectrum, and the other is that you make it open to everyone, but as a result, you can't capture any of that value?

So, both of those are issues that do afflict the privacy space. So, I think it's not even a question of like outsized returns or like making a massive uh kind of being able to retire the next three generations. I don't think that's the right motivation. It's just being able to act to actually like just live a decent life uh while still building things for the world and not focusing on value capture. Um I think that's the dream of many of us in the space that we want to just build and like make sure it creates the maximum possible value, but at the same time, it allows us to pay our rent and it allows us to feed our families.

Um and that has proven to be an extremely difficult uh balancing act in practice. Yeah, so I guess like how would uh projects sustain themselves when they are only focused on the value creation aspect of it. And I think we've seen a couple of models emerge in practice. So, one is maybe what I'd call the Renaissance model or the Medici model, where you have some benefactor who funds their entire repo, and they get reputational benefits from it, and the repo itself gets that stability of being housed within a larger organization. So, a good example is like uh Prism, which is a consensus client on Ethereum, is fully funded by uh the Offchain Labs and Arbitrum team.

So, that's maybe a good example of like how the Renaissance or the Medici model was, where these rich patrons would fund Leonardo da Vinci and others to create their art. So, that's one model. The second model is um like existing pots of public goods money. So, like things like um like quadratic funding rounds. That's very new to our space, where a lot of people, like rich people, I've noticed, are willing to give money, but they don't want any headache, and they don't want any reputational risk.

So, if the rich person also has to decide on who gets the money, A, there's a lot of headache in that and time, which they don't like. And B, there's reputational risk that oh, you gave the money and it went to a scammer or something like that. So, they want to avoid both of those. So, as a solution to that, uh we've come up with like a category called funding mechanisms, where uh you you just put money into a pool, and then some algorithm decides how that money gets distributed. So, that de-risks that that solves both those pain points.

There's no headache, you just put money into one public key, and there's also no reputational loss, because uh it's the algorithm that then decides it, and it's not your personal decision that decides I'm going to give this much money to this person. Um so, we've seen that play out with quadratic funding. Like most recently, there's a Dow security round, uh which is funding all of the different security like making it helium more secure and there's a pool of about a million dollars and that's allocated according to how people it's like a donation matching tool like I'll give money to a project and then some amount of money from the matching pool gets given to it in a formula that rewards more number of contributors rather than just how much you give. Um so that's another one. So we have the Renaissance model of a patron.

We have the algorithmic model. And um and then we have just the regular grants that we see. Uh which is what the world relies on. So even like governments give things through procurement frameworks. Um and we have different grant funders in the ecosystem like ENS and and EF and like many other L2 foundations.

So that's maybe the third source of funding that I've seen. But overall yeah, it's like not easy and it's a hard life to choose. So I think the easiest way is to just like participate in these sort of mechanisms and actually tweet about them. So for example, in the Dow security round, I donated what like I donated $40 like not any life earth-shattering amount of money. But then I also tweeted that these are the projects I supported and this is why I supported them.

And then that encourages others to give and that also helps the matching pool be given more accurately. So take part in these rounds when they happen these quadratic rounds. Take part in Octant. Like go there's a pro- there's a project called Octant where you can lock up a token called GLM and then you get voting power to decide who should get funding in the ecosystem in these various rounds that they hold. Um so lock up GLM, uh take part in these quadratic funding rounds, or even just like go to these regular Dows and like be a contributor there, make proposals.

So if you're willing to give time but not money, there are enough ways that we've devised as an ecosystem of doing that. On a rational actor basis, like we should all be supporting it because it reduces the deadweight loss in the Ethereum economy. And I'll explain what the concept of a deadweight loss is. It is that imagine that there's some project which is delivering um uh $10 of value to six projects and it costs $50 to maintain. So now there are the total amount of value it's delivering is $60.

$10 to six projects each. The cost of its upkeep is $50. So no single project finds it economically viable to give that $50 on their own, but they each find it profitable to give uh like $8 each because they're still making a $2 profit. But they don't do that in the absence of coordination. So that's referred So this loss of earnings, that $50 is what the project was what the public good costs.

$60 is the benefit it delivers, but we aren't able to coordinate and as a result we lose out on that benefit of $10. So that $10 is called the deadweight loss in an economy. So maybe the most easiest example is like a road. A road costs some amount of money to build, but then it delivers value in the form of the shopkeepers around the road who make money in like the extra transactions it facilitates having a road. So not having the road increases the deadweight loss in that society.

So I think we should view it in a similar lens that if we actually want to outcompete the regular world and make Ethereum the new economy, then we have to reduce the deadweight loss and be more efficient than the regular world. Which we are not right now. Like the regular world has taxation, which for all of its drawbacks is still like a good solution to the free rider problem uh because everyone has to give taxes. And they use that to reduce the deadweight loss. But Ethereum currently relies entirely on altruism, so we need to figure out a better way of doing it.

Interesting.

Right, my favorite public goods project. So I am inspired a lot by Protocol Guild, which started as a collection of all the Ethereum developers, and they have very tight eligibility requirement that you have to have contributed to the core Ethereum protocol, and then anyone can give money to one address, and it gets distributed to all of the developers. So, I think that's like a very neat and simple model. Um some of the other projects I like are like Nouns. Nouns DAO is cool, and like you hold an NFT, and then you can vote on how the treasury should be allocated into different projects.

So, Nouns and Public Nouns are another cool one. Um and um I like also like a lot of the public goods tooling, like how can we actually allocate the public goods that we have much better. So, things like uh like Gardens DAO, for example, which like, you know, has a conviction voting sort of model for how you allocate money. Um Hypercerts, which is like impact certificates, where you create an impact, and similar to how you buy a t-shirt, you can now buy that impact that was created. Um I think those are some other cool projects in the space.

Uh I think the 1993 era of cypherpunks were looking at information that like, I should be able to communicate information that I want privately and securely. And today's uh era of cypherpunks, at least within the blockchain space, is more focused on financial value and transferring that securely. And so, I think that's the biggest difference. Like, you know, we've seen a bunch of rulings saying say saying that money is free speech. So, now that's a natural extension that if me giving money is like a vote, like if I buy a t-shirt, it's a vote that the t-shirt is good, so that is free speech.

So, if we accept if we accept the proposition that money is free speech, then the new era of cypherpunks are trying to extend the same 1993 era of cryptography to be able to talk freely towards the realm of money. So, I think like as a rule of thumb, like we should try to stick to the figure of 10% that 10% of the funding is public goods and 90% are private goods. I think that's like uh that's like the realistic one, but I hope it can become 20% like, you know, that's personally what I think is the right amount that 20% of the economy is public goods in nature and 80% is based on market mechanism, supply, demand, all of those dynamics. Because I think right now the biggest issue is that we have no way of knowing whether money is wasted or not in a public good. With a private good, like we can know that it's utilized well because if it doesn't earn revenue, and they've not made something customers want, then they go out of business because they have to sell it.

But, it's not as clean for public goods like now. If you can convince funders, then you can survive, and if you can't convince them, even if you're useful, you go out of business. Um so, I think that's the biggest issue that how can we get good account- like as good of an accountability mechanism in public goods as we have in private goods? I think that's like the first question to try answering. And in the failure or in the absence of being able to answer that, at least 10 to 20% we should guess.

Yes, I think the mandate is super useful in like finally clarifying our scope. So, um I think there's like an argument to be made that you can't boil the ocean. And like under the new mandate, we have like it was actually written as an employee handbook, which is like very funny, and then they made it public. Um so, as an employee handbook, I find it incredibly valuable because it tells me which type of teams I should work with. So, what are some actionable points from it?

One, like only work with teams that are actively trying to become independent of the EF. So, if there's a team which is which doesn't have a plan to be independent from us, then we will not work with them. Like that is super useful. That's also increases my position when I'm engaging with the ecosystem that I can be like, "Hey, the mandate is saying I need to work with teams that actively try to be independent from the EF. Like what is your plan to do that?"

And so that's one thing I like a lot about the mandate. Um and the other is just like it's very narrow focus. So like for example, we have to choose between a stable coin run by a big institution versus but it's not at all friendly towards censorship resistance. It has a freeze function. It's not private.

It's not secure. And I have to choose between another one which is a startup which doesn't have a lot of traction, but it's building it the right way, that stable coin. So now earlier we were always confused like you know even in the realm of privacy like should we try to be building privacy solutions that adhere to the overarching like laws and frameworks and KYC regime or should we be supporting like the privacy in just in the code? So the mandate has helped provide like some guidance on these difficult questions that we confront when we do our jobs. Um but okay, we will support the startup stable coin.

We will support the uh privacy solutions which are in the code and we will like you know focus less on making KYC easier for example. Like that is something which I would personally now spend less time on. Like incremental solutions which uh rather than like the radical like this is how privacy should be done from a ground up level. I think the ideal world if we succeed is like this becomes like the future of work. Like that's what we should all be working on in the blockchain industry that like how we work together and how we get paid and everything.

It's all like global, permissionless, distributed. It's like how the market should operate. But currently we are like broken up by a patchwork of different laws and regulations and regimes um which have the which are useful in like squashing out bad actors, but the problem is that we bear the cost of that. So, I think like more accurate targeting, like which you can do with software, is like what I what I hope the world becomes that like you know we all get paid in um either directly for working for a company and we can like program laws and like okay this much revenue should be distributed in this way of the company or we use things like quadratic funding to get paid for some public goods that we are working on. Um yeah, like more software paying humans and less bosses paying humans is how I hope we evolve if we are successful.