Max Resnick - Solana's plan to disrupt 500T Securities Market - TOKEN2049 Singapore 2025

Hi everyone. How's everybody doing today? The global securities industry is $500 trillion in market cap. So it's a huge opportunity size and today I'm going to tell you all about everything that teams and Salana are doing to disrupt that and bring it on chain on Salana. So there's kind of two big segments of the global securities industry by market cap.

You have equities and fixed income. I'm going to talk about equities today because I think they're a lot more exciting. Um the big thing about equities is that the concentration of equity market cap is really really focused in the US. So the Forbes 2000 that's kind of a global equities index with the top 2,000 companies in the world by market cap. If you look at that and you kind of sum up all the market caps, 54% of all global equities market cap is from companies headquartered in the United States.

So as a little agenda, we're going to be focusing on how can we get those massive pile of assets, those securities trading on chain. The first step is we have to make sure that these securities can be issued natively on chain. That means not wrapped, not uh some special acquisition vehicle, something like this. The actual security itself issued on chain, the ledger of record that carries all of the same rights of a class A share issued on chain. That's step one.

Step two, we need clear rules around what constitutes an exchange and whether trading of securities on chain requires the same regulatory rules as trading on say the NASDAQ does. It's a pretty different environment. Rules are probably going to have to be different. And uh so I'll talk about that a little bit as well. And then third, and this is kind of what I work on in particular, is trying to make the product of Salana, the technical product, good enough that it rivals NASDAQ and even surpasses NASDAQ in terms of execution.

So make the product good enough that people actually want to use it, not only to issue securities, but also trade them on chain. So I'm not a lawyer. I'm an economist. Um, but these are kind of just to orient you around securities laws in the US. There's a lot of action going on right now, but kind of all goes back to these three big bills.

The 1933 Securities Act was kind of what defined a lot of the rules around issuance and transparency. 1934 act is about exchanges. And 1940, which we're not going to talk about too much today, is kind of about investment companies and investment advisory. There was a recent no action letter yesterday kind of touching on some of those things, but we're not going to talk about 1940 too much today. Mostly focusing on issues around 1933 issuance and 1934 exchange.

The good news is the first problem of getting actual class A securities listed on on the chain like just existing is already solved. So already this year we had the first NASDAQ listed security issued as the ledger record on on the Salana blockchain already. This happened a month ago. So Galaxy tokenized their shares. It carries all the same rights as class A common stock.

That's uh the ledger is something else. So this problem is already solved. Here's the smart contract the program as we call it in Salana land. You can see uh it's there. So I looked a little bit at the data.

There's actually only 70 holders of Galaxy Digital Class A common stock so far, which is a good number, but kind of we'd like to get that way up. It's only about 0.02% of the total Galaxy stock that's currently held by people who hold it on chain. And the reason for that is you can't really do anything with it. You can buy it.

You can look at it in your wallet. It looks kind of nice, but you can't do anything with it. You can't trade it. You can't really use it uh to as collateral. So, what's the next step is now that we have it on chain, first of all, let's get some more stuff listed and uh forward, which is a slant of also NASDAQ listed is is kind of working on that as well.

But we need to make it so people can actually do stuff with this, not just hold it. So step one's already all set, but now step two, we need clear rules of the road around trading securities on chain and that's where the Salana policy institute is doing a lot of work on project open. So what is project open? It's a new framework that's been submitted to the SEC which details all things around issuance, trading of securities and also some investment advisory stuff as well. So all of that is part of project open which Salana policy institute came up with and they've been working very closely with regulators in DC to try and make that happen.

And that's part of why we were already able to see some stocks already on chain. So the first part of their goal is already done. Now they're working on getting clear rulem around trading of securities on chain. The good news is with the new administration, Washington is 100% ready and focused on crypto. They're willing to talk.

They're very reasonable and they really want to get uh rules passed. So two days ago, uh, Commissioner Paul Atkins, he's the chair chairman of the SEC actually, he's, uh, saying basically crypto is our number one focus right now. And the same is heard from a bunch of lawmakers and all sorts of people who are kind of relevant to this conversation. So that's part two. And again, that's work that a bunch of people on the legal side are doing.

Now I'm going to talk about some more of the details of what I work on which is the product itself. Salana as a technology, Salana as a product. So we need to make Salana good enough that people want to issue their securities on chain. They're already doing that and we need to make it good enough that they want to trade those securities on chain. So part of that is a market structure problem.

Part of that is uptime. Part of that is uh latency. We'll talk about all of that in a second. The good news is we have a new type of market structure for DeFi that's evolving right now on Salon. It's not happening anywhere else as far as I know and it's called proprietary AMMs.

What it is is basically a trading firm makes a program that's only for them to provide liquidity and they put their proprietary trading strategies on that AMM. So this is already we're already starting to see a lot of effects of this uh this development which is about 10 months old now in the market structure of Salana and in what people are doing on chain but mostly for native crypto assets that are not considered security by the SEC. So how do these things actually work? Basically one trading firm makes one program and that is their order book that they put out. But the real innovation in this market structure is the separation of the two concerns in trading.

So in in high frequency trading usually you'll have two kind of pretty different things that different types of people work on. You'll have pricing which is uh how do we know what the price of Salana is at this millisecond and that involves a lot of linear regression a lot of fancier math models. Usually these are the kind of quants that work on the pricing side. Then you have your order execution side and that's like given the price let me set some curves let me set some orders out on the book and get ready for people to take my orders and the prop amms are the core innovation is just separating these two concerns. So you actually write your order execution logic in the program itself and that means that there's zero latency between what happens on chain in terms of orders getting filled and the response of the program on chain.

So classical kind of person who wants to FUD crypto as as a medium for a securities exchange would say well it's crypto you have decentralized consensus you have nodes all over the world there's no possible way that you can be faster or fast enough to trade securities but with this structure the program lives on chain and so there's zero roundtrip time the chain isn't even part of the loop the consensus is not part of the loop it's all happening on chain so it's actually faster to react act because it's living on chain. And then the pricing stuff comes in because you're basically running all your models on your local machines, getting all the data from around the world, turning that into your view of what the prices are, and then you put it in a very small Oracle transaction, often like less than 100 compute units. For comparison, the average Jupiter swap is 200,000 compute units. So you've got these really tiny and therefore very easy to land price updates that go in to the order execution program. That's kind of how these propm work.

And the nice thing is they work really really well. So what we've seen so far this year as these prop AMMs have gained dominance and now one of the prop AMMs is the number one exchange on Salana by volume now is that they're starting to dominate and they're starting to bring a new functionality that no other chain has. So right now every other major chain, obviously Bitcoin has no trading on chain, but even something like ETH that has a lot of DeFi, still Binance has higher volumes in its native token. So it's about double for ETH. Twice as much ETH traded on Binance as on Ethereum itself.

Whereas for Salana, we actually have three times as much volume in the Salana USD pair on chain than Binance does. So people are starting to use it. It's actually getting quite good prices and in many cases better than Binance. And so the volume naturally has moved on chain which is great. We want to use that same structure for securities.

But that's kind of the good side. Now let me give you a little bit of caveats and some some of the kind of thorns that we're working on. And this is really what I work on. So this is kind of a lot of stuff on the page, but this is a view of the three types of Salana transactions in four slots, one right after the other. So on the bottom is the TPU transaction.

So those these are basically normal transactions that don't go through JDO. Jetto is an MEV similar to kind of PBS if you're familiar with that. That's what JTO is. The red ones, those are JTO and the green ones are actually votes. So you can kind of ignore those.

Those are for consensus purposes. They don't really affect the market structure. But the basic idea is we've got streaming going on. You have 50 millisecond judo batches which you can see here based on the kind of spacing of those red batches. But I want to point a few things out to you.

So first of all, I want to draw your attention to the blue plots. And looking just right at the beginning in the first tick of the blue plots, you can see there's often a little bit of a buildup there. And in fact, the buildup is even bigger in the first slot. Here we have four leader windows in a row. So like you have 400 millisecond slot, four in a row per liter.

And so the first one happens after a leader handoff. Leader handoff. Right now we stop processing transactions for about 70 milliseconds. Not a great thing to do if you really want to take on this 500 trillion dollar securities industry. You don't really want to stop processing trades for 70 milliseconds.

NASDAQ doesn't do that during the trading day. So, we want to fix that. And then even in the space where you're the same leader, we have about a 20 millisecond drop off. And that's kind of why you're seeing some of the transactions piling up is that there's a a 20 millisecond stoppage when we're kind of transferring from building one block to the next block even when you're the same leader. So these are the kind of problems that are extremely technical to fix but we spend a lot of time working on them at Anza basically writing more performant code.

Uh we have d0ero coming soon which is now running on 16% of the network which is uh a private fiber network that will allow some of that 70 millisecond leader handoff to come down. And then we have some more complicated kind of deep tech crypto uh consensus work that's going on to try and bring that down as well. Another thing that we're starting to notice is more and more Salana nodes, of which we have about a thousand, are realizing that MEV is a thing and that it also exists on Salana, not just on ETH. So, here's uh 16 liter slots and I've plotted on the x-axis is the non-vote transactions. On the y-axis, sorry, the y axis is the number of nonvote transactions cumulative.

The x axis is time. What you can see is like this is the standard agaveuler that runs about 80% on the network right now and it's pretty linear which is probably a pretty good thing for transactions to be coming in and coming out around the same pace, right? But as we start to zoom out and look at a few more uh leader windows, you can start to see that there's something else going on here. And what this is is a revenueuler. And so the nodes have realized me is a thing.

And one way to make more money is to stop processing transactions until the last 50 milliseconds of your slot. And so if I zoom out a little bit more, you can see, hey, this is getting to be quite common. It's actually over 10% of the network now running these revenue schedulers. And the the real question is like what does that do to the actual market? Is it a problem or is it just something that we can look at on the picture?

Um, here's kind of some of the prop amm trades plotted over a two-minute period. And I've shaded in gray the trades that are happening during these re these um revenueuler slots. So the question we want to ask is does this actually change the profit and loss of the market maker and does that mean that they have to in turn widen out and quote wider spreads and in fact the answer is yes. You can see here the kind of magenta bar the pink bar is the is revenue boolean. So that's if theuler is a revenueuler and the red one is vanilla agave which like I said runs 80% of the network.

So the red one here corresponds to the profit and loss if we're in a normal slot and the the purple one is if we're in a revenueuler slot. And you can see there's actually a noticeable difference. The revenueuler does reduce the performance of these proprietary AMMs just from the empirical data. And so it's a little bit unclear exactly what will happen because we don't think that the prop amms have done too much adjustment yet to deal with the change in structure. So that kind of second order we haven't been able to observe yet what will happen.

But it's pretty obvious to me at least that if you stop processing transactions for a while then your exchange is going to be worse. And that's exactly one of the things that we'd like to fix. So how do we fix it? Well, the core problem is that there's this leader monopoly for four consecutive slots in a row. The leader gets all the power, all the control, chooses which transactions go in and in which order they go in.

And this is obviously not conducive to disrupting the $500 trillion global securities industry and and winning against our competitors who are currently doing all of the securities trades, right? So, how do we disrupt that? You need to basically build crypto, build the blockchain in a way that is designed for trading. And that means removing the leader monopoly by design. So multiple concurrent leaders is something I've been working on for a long time.

Uh you can check out this paper we just put on archive. It's basically how do we do it and why are we doing it. And it's pretty dense. So, uh, don't get mad at me if you see a lot of Greek letters in here, but the core idea is pretty simple. It just says, hey, instead of having one leader who gets to include transactions, let's have multiple leaders who get to include transactions.

And then let's take the ordering power away from the leaders. All they are going to get is inclusion. And we're going to do the reordering at the protocol level. So, that's one idea of how to deal with this leader monopoly. Another idea that we're thinking about is something called Gatling, which is super fast leader rotation.

And this is uh kind of a depiction of how it works. There's actually kind of a speed limit in the consensus literature that says you can't really go below around 300 millisecond liter slots if you want to do one right after the other because we know that it kind of takes some time to come to consensus on the previous block before we can start the next one. And so to get around that, we're going to have multiple chains going in parallel and then kind of merge them, squash them all down into one chain. So that's another idea. And this doesn't get rid of the leader monopoly, but it dramatically allows us to shorten it.

So those are two ideas. We're probably going to do them both to be honest. So we're going to have multiple leaders and super super fast slots. Uh that's at least my goal and it's something that we're working on. We're prototyping these out.

You can follow along. Everything we do is open source and public. So if you want to follow along with some of the prototyping work that we're doing, it's on GitHub in the Anza GitHub. This is my last slide, but I just wanted to shout out. We also have a bunch of technical content on our blog on anza.

xyz. So you can check out kind of stuff about running a Salana node, uh about kind of the the future roadmap of Salana and then about just hey, you know, something like what are the transaction costs, how how can I be more effective at landing? All of that's available on the blog. Thank you guys.