[(6:40)] Jax: Hey, Sam. How’s it going? I’m Jax Jones.
[(6:48)] Sam: It’s going pretty well. How are you?
[(6:50)] Jax: Really good.
[(6:53)] Sam: Thanks for having me on.
[(6:53)] Jax: It looks like you’re in an offie at the moment.
[(6:56)] Sam: It’s true. I am in an office. We’ve been fortunate to be able to have offices open throughout the pandemic. It’s been super valuable. There are so many things that are just a little bit easier when you can talk in person.
[(7:10)] Jax: But you’re doing it in Hong Kong, right?
[(7:13)] Sam: Yeah.
[(7:14)] Jax: It looks like you’re in a gaming chair. Is that how you approach life?
[(7:17)] Sam: Yeah, we all changed to gaming chairs. We just tried out a bunch of chairs. Everyone who tried it just thought that the gaming chair is the best chair. So it’s like 50 of them.
[(7:27)] Jax: So amazing. Nice, good choice.
[(7:29)] Martin: Just for our audience, it’d be helpful just to ask you to reflect again just on how you arrived to be sitting in that chair in Hong Kong. Trading crypto, setting up an exchange and doing all these things we’re going to talk about today.
[(7:43)] Sam: Totally. I went to MIT where I had learned a lot about myself, including that I didn’t want to do physics. I could fill up enormous numbers of hours playing League of Legends if I tried. I didn’t know what to do with my life. I wandered into effective altruism, it was really wise, somewhere a direct path there, I guess. Which is basically a movement looking at if you’re trying to figure out how you can do the best with your life, what should you do? There are a lot of answers to that question.
I think some of the most compelling is to find a great organization and help spearhead it. One thing that you can try and do is figure out how you can donate as much as you can to effective organizations. Sometimes you’ll go to them and be like, “Do you want me as an employee? Do you want my money?” It’s like, “We don’t want you as an employee. I’m sorry, no offense. No, we’ll take your money.” I tried leafleting. No one wants me as a leafletter, I am not good.
Anyway, I thought about what that would mean. I ended up on Wall Street trading ETFs at Jane Street Capital. It’s a great place. I learned a ton there. He gave me a ton of respect and responsibility. I am grateful for that. After a few years, I started to get a bit restless and felt like it was time to start up my own thing. I left. This is late 2017. When you walk down the streets and you hear two people talking, you’re like, “That’s probably about Bitcoin.” I mean, I don’t know but pretty likely.
Basically, I just had a ton of priors that made me think that there might be a lot of money to be made trading crypto. It seemed very likely to be a field where there’s a ton of demand, a ton of volatility, a ton of wackiness, and craziness. There’s not much liquidity or infrastructure to support that. There might be a pretty big role to carve out, providing that. I started trading crypto, built out a crypto firm. There were a lot of good trades to do.
A lot of the difficulty wasn’t even finding the good trades, sometimes those are obvious. The difficulty sometimes was figuring out how to do them. How to do it without your bank account shutting you down. Bank compliance officers are not a fan of crypto trading firms. It checked all the boxes they didn’t want.
[(10:23)] Martin: What did they just call you? A money launderer, I would imagine.
[(10:26)] Sam: You’re not laundering money. They can’t call you one. What you do though, is you check a lot of the red flags that money launderers check. It’s in fact for an innocuous reason. You’re telling me, that every single fucking day, you send a $10 million wire transfer between two different countries, two different currencies, and two unrelated entities always in the same direction. What do you think that is? What business has only one-directional flow, huge size, every day?
The answer is, “I’m doing Bitcoin arbitrage between the Japanese and US exchanges.” That’s not usually the answer when you see that fact pattern. It’s basically just like the [inaudible] cycle, “I could get fired for allowing this account.” The bar, for me allowing this, has to be done so confident that nothing bad is going to happen here. That no one is ever going to give me scrutiny for having allowed this fact pattern to exist.
[(11:21)] Martin: There’s a lot of layers to you. You’re building that slowly?
[(11:25)] Sam: Yeah.
[(11:26)] Martin: I might ask, make it bloody far from 2017. I still don’t understand how you did it in 4 years. This is the genius of the story. How did you get to Hong Kong? Just tell us the progression between your different things right up to theorem.
[(11:40)] Sam: Totally. Allow me to say in crypto trading, for I mean, by and large, serve the motto as a proprietary trading firm, is “All press is bad press.” Why would you want anyone to know you exist? There’s nothing but bad reasons. That’s what we’ve been following. But that started to change in late 2018 for a number of reasons. One of which was, it started to spin up an “OTC desk”. We want to find institutional counterparties to provide liquidity there.
We started to do a little bit of Biz Dev, a little bit of reaching out, for the first time. Anyway, I went to a crypto conference in Macau. It’s my first time in Asia since living in Jane Street. I immediately realized, “Holy shit, there’s a lot of things I want to have.” There are 17 people I wanted to talk to. It’s a three-day trip. On day 3, I canceled my return flight and rented out WeWork in Hong Kong. I started having meetings. It was eye-opening. All of a sudden, I had this sense of, “Wow. All these things I thought I could never do seem a little less impossible now.”
I’m starting to meet some people, starting to build some networks and relationships. One of the things that had been bubbling up there was our frustration with the exchanges. Exchanges are everything in crypto. This is not true in traditional finance. No one’s like, “I got fucked over by the IC yesterday.” No one talks about it. What exchanges are, actually, in traditional finance, it’s a matching engine. It’s this back-end technology. But it’s just one piece of the puzzle.
[(13:33)] Jax: What’s the difference between exchange in the crypto space and traditional finance?
[(13:37)] Sam: In the crypto space, it’s every piece. So imagine, you buy Apple stock, right, you’re probably going to like Robin Hood, to Citadel, some clearing firms and custodians in the back end, some technology providers. Eventually, you go to Dark Pool. Then to another liquidity firm, then finally, you end up on the IC. Then you go through the whole thing on the other side. In crypto, there’s nothing but the buyer, the seller, and the exchange.
The exchange is everything. From the mobile app, the GUI, the API, the brand, the retail facing piece, the custodian, the clearing firm, the product design, the matching engine, the clearing, and the risk engine liquidation. Every part of the process between the buyer and seller is part of the exchange in crypto. It’s really all of those rolled up into one coherent product.
[(14:21)] Martin: It’s all in, right? If you build one, you take all the risk. But if you’re dealing with an exchange, again you take all the risk?
[(14:27)] Sam: That’s right. They’re huge in crypto. Most of the people’s crypto experiences are dominated by the exchanges that they interact with and the tokens that they trade or invest in, or whatever. So that’s really central piece of the ecosystem. Despite that though, they work kind of shit, especially the derivatives exchanges had really serious issues. You can enumerate them. But it’s almost too kind to enumerate the specific issues. What happened was, they are just bad. They had bad designs from the ground up.
That just led to all these issues, they’re losing a million dollars a day of customer funds to having incompetent risk engines. There is serious regulation during, and then one of the biggest names. They have this ridiculous margin system where every single different product you wanted to trade on an exchange, you had to completely separately collateralize. If you want to trade Ripple futures, you have to go buy Spot Ripple tokens and move into Ripple futures margin wallet to Ripple futures.
If you want to trade ETH futures, you take out your Ripple, you sell it for Bitcoin, use a Bitcoin to buy ETH, Move ETH into your ETH futures margin wallet, and then you can trade your ETH futures. So 80% of what you’re doing is moving Dongles around in order to just be able to then do the trade that you actually wanted to do. You have no flexibility on what your collateral is. You can liquidate it on your Ripple futures position, completely independent of however much Ethereum you have. So it’s a mess. It leads to all these downstream issues as well.
[(16:11)] Jax: I guess that’s very complicated for your average user as well, who is less technically able, isn’t it?
[(16:15)] Sam: It’s one of these rare things, which is bad in every way. It’s complicated for the average user. It’s terrifying for institutional users. You’re telling us when we can get liquidated despite having a billion dollars on the platform because we didn’t remember to collateralize 1 for 300 different wallets? They don’t want to hear that. The risk departments don’t want to hear that.
It’s also one of these things that is both complicating and reduces flexibility. It gives you fewer choices about what you can do, and simultaneously, also makes it more annoying to use. Those are one of the fundamental issues. But there’s just a ton of things and it’s been mounting over the course that year. We’ve been using these products like these. We can do better than that. We really want to.
Despite that, we thought it’s hopeless. We thought we could succeed at building a cool product. But we did not think that we could succeed at figuring out how to get any users. Our sense of the odds that we’d be able to get significant numbers of users went up from 2% to 15%.
At that point, just make a snap call, we’re in if I can go for this. Even though we think we’re going to fail to get users it’s so high upside that we would be negligent not to try. I called up my co-founder, CTO, and college roommate. I said, “Do you want to do this?” He is like, “Yeah.” So we started building out FTX.
[(17:51)] Jax: I love the blasé nature of it is. Like, “I called up my college roommate and there it was 10 billion later.”
[(17:58)] Martin: Just staying on FTX, right, I’ve got two quick questions. We’re going to keep on the surface of what you’re doing. Then we can start to drill down, I think most successfully. You’re trading $10 billion a day on FTX, is that true?
[(18:09)] Sam: Yeah. It’s averaged about $14 billion in the last month of daily volume on FTX.
[(18:15)] Jax: It’s insane.
[(18:17)] Martin: The second question, just to tee off some fault process down further into the conversation. You can’t trade in America for obvious reasons, but I have heard the rumors that you’re going to skinny down the product to have maybe a more vanilla version of FTX to get it into the US market? How you’re lobbying and all the rest of it doing, these guys wake up and give you a chance?
[(18:39)] Sam: Yeah. There is a US product. It’s “FTX.US”. It is a lame version of FTX certainly. It doesn’t have almost any of our innovative products on it. It’s first places to buy Spot Cryptocurrencies with fiat currencies so I think it’s pretty good within that limited design space. There’s a bunch of regulations. But the biggest is the CFTC, which is basically taking a reasonable position that this is a regulated financial activity that would require a license.
It is also taking the reasonable position that they have not built a license for it yet because it’s a complicated new field. Unfortunately, those positions don’t combine well. Because of that, it’s very difficult to find a pathway for offering crypto derivatives in the US right now. So the US platform doesn’t have any of the derivative products on it, which are the bulk of the volume on the main platform.
[(19:38)] Martin: But most Spac products have no issue at all, right?
[(19:42)] Sam: Spac products for cryptocurrencies, which the SEC doesn’t deem to be potentially unregistered securities.
[(19:48)] Martin: Okay. So, XRP?
[(19:50)] Sam: Yeah. Interesting case. We do not list XRP on the US exchange. That has been a borderline case for a while. There’s an active lawsuit going on right now between the SCC and Ripple about whether it is in fact an unregistered security or not.
[(20:07)] Martin: Do you have a view?
[(20:10)] Sam: I don’t think it’s an obvious answer. It gets to some interesting questions of what you care about as a regulator and what you’re trying to regulate in the first place because one of the weird properties of this particular area of law is, let’s say, you make a token and you say, “I’m going to give $10 every day to the token holders.” There are questions about whether that’s a security.
Let’s say, you made a token and said, “This token is a scam. It has nothing. We’re never going to do anything for this token. If you buy this, the joke’s on you. This is a Ponzi scheme.” That’s probably not a security, right? That’s actually more likely to be legal in the US. It’s not clear if that’s exactly where you want things to end up but that is where they do end up. What this is surrounding is, “Are you making claims or promises about what creates value for this token in specific ways?” And if so, then you would have to register it via a scheme that isn’t built out for digital assets yet.
[(21:18)] Martin: Well, I’m going to come back to XRP and these coins a bit later but on the stuff that you’re doing. You also have Alameda research, which I understand comes back to your roots in California. I understand that research is probably a stretch. There’s some research going on but it’s a trading firm. I’m going to give it to you, talk in your own terms what you do there.
[(21:42)] Sam: I’ve been fortunate to be able to hire a lot of great people there and be able to take a big step back and focus on building out FTX. What Alameda does came from originally was arbitrage. The core original trade is basically, “You find one exchange where bitcoins is trading for $10,000. Another worth trading for $11,000. You buy it for $10,000. Sell it for $11,000 and you’ve made $1000.”
That sounds like this two-way example couldn’t be true. But in crypto, it’s true sometimes. That thing does happen. But when you try and do it, you start to realize why it can be true. So much of the infrastructure is janky enough that it’s actually often quite difficult to execute that efficiently even though it sounds quite simple.
Any one of those exchanges could lock up your funds. You could have withdrawal limits that made it infeasible. Your bank could refuse to let you transfer dollars between them. There’s a lot of ways that this can go wrong. A lot of things you need to get going in order to make it work. It’s not as easy as it sounds to do this in crypto. The logistics of it end up often being the hardest part.
[(23:01)] Martin: It’s through Alameda that you did your arbitrage work, right?
[(23:04)] Sam: Yeah.
[(23:05)] Martin: Between a Japanese and US Bitcoin. I’m sure some other things along the way. I have two quick questions, again more of a context for later. What made you take some of these risks? I can only imagine entering the Japanese market and how they’ve got a big issue with trust and things being proven.
I can just imagine that you had a whole bunch of doors slammed on you just down to lawyers and bank accounts that you’ve already alluded to. That to me is a massive risk. You got to put sample trades down, then bring them up bit by bit. How do you ever get a chance to trade massive volume? How do you do it so quick?
[(23:42)] Sam: I think it’s one of these things where you can ask, “What’s the secret formula here? What’s the thing no one else understood?” That’s like, “Oh, man.” You got this one cute trick to make a billion dollars, whatever. I think that’s the answer in the end. What actually is pretty interesting is that there always wasn’t one trick. It was just doing a lot of things quickly and well and doggedly.
What do I mean by that? You could jump through 16 hoops, run into 16 brick walls, and break through each one and still have no ability to make money doing the trade and there are still barriers. Do number 17, you’ve still no ability to make money. Do number 18, and then all of a sudden the trades unlocked then you make a lot of money doing it. It’s one of these things where the answer in some sense is a straightforward obvious answer of “There’s some unknown number of barriers. Once you break through all of them, then you can do the trade.” But it doesn’t feel like that at that time. It feels like this whole Sisyphean feeling of “Oh, God, they found yet another way to stop us. This is hopeless. We’re never going to figure this out.”
Some of the answers is basically just having an instinct for easiest, directly doable, easier like at the light at the end of the tunnel, somewhere here. If it’s valuable enough to do it, just keep going. If someone throws in another roadblock, be creative. Don’t say, “Oh, well, our system can’t handle that.” Be like, “Fine, we’ll build a new system. You’re are making us jump through that hoop, we’ll learn how to do it. We’ll do that as fast as we can. We’ll drop everything we’re doing to figure out how to jump through that weird hoop that you just decided was important for us to know how to jump through.” Then, we’ll do it. Then, “Oh, God, there’s another hoop. All right, we’ll jump through that one too.”
Just not giving up. Being super dogged, and super creative about these. Being willing to just keep thinking of solutions that other people just wouldn’t do. Not because they’re wrong, bad, evil, or hard even. That’s just not what you do in some sort of like hard to define way. I was like “I don’t know. That’s not what you do in that case. You’re supposed to do this other thing.” Like, “What do you mean supposed to?”
[(25:55)] Jax: Just to touch on the price differences you mentioned earlier, Bitcoin costs more in Hong Kong than in America. Why is that at that point?
[(26:05)] Sam: Yeah. Why was there this big difference between Japanese and American Bitcoin? Is it racist? The basic answer is that it is late 2017 to early 2018 is the time of Japanese people want to buy a lot of bitcoins. They were. They’re buying you a billion dollars a day of crypto. That was a huge amount of inflows in Japan. It just overwhelmed the liquidity sources.
There were more buyers and sellers in Japan. Because of all of these banking difficulties and withdrawal limits and everything else existing in new space without lots of established big financial players. No one had the huge balance sheets to support this trade. The world wasn’t collectively able to provide as much liquidity as these customers were demanding by buying.
So they just lift all the offers. Then they lift more. Then they lift more and more and the price will go up and up. The only thing that stops this from happening, is if you have people who can do this arbitrage in big enough size, that they can close that spread. What happened was that the demand outstripped the supply of liquidity, so the prices started to diverge. The thing that was needed to bring it back in line was some trading firm to be able to buy as much Bitcoin as was needed in the United States and send it to Japan, and sell it there.
[(27:29)] Martin: It’s interesting. We’re both off-script, both passionate about finding answers to different parts of the puzzle. The last thing, on the surface and then we got a stream of questions that seem more logical. I think FTX is still centralized. You’ve got this “DeFi” initiative, Project Ethereum. Would you consider that your hobby project or your third real focus? You’ve got Alameda, you’ve got FTX, and then you’ve got this Project Ethereum.
[(27:57)] Sam: It’s interesting. One of the core things about this is that it is decentralized. There’s a limit to how involved I can be. It can’t just be the “Sam Bankman-Fried Variety Hour.” That’s not a decentralized protocol. I’m doing what I can to support it and happy to help in any way I can.
It was probably important at the beginning in building out some of the infrastructure and the philosophy behind it. Over time, it’s becoming more in more just in a bunch of random people’s hands. That’s how it has to be.
[(28:38)] Martin: Someone online labeled you as a “$10 billion man.” There could be worse titles, I guess, right?
[(28:43)] Sam: It’s not too bad, all things considered.
[(28:45)] Martin: Yeah. You might spend that money later on, but you can unwind that position and you know to allocate it out. But I’m interested in where did you make most of your net gain? Do you remember the emotions when you knew that this could scale up?
[(28:60)] Jax: The way I want to ask that is, do you remember your first big win? How did it feel?
[(29:03)] Sam: The first huge win was the Japan one. That was the first time that we put everything together for not nearly as big of sizes as we were hoping. We could have made 10 times as much if we’d had a better setup at the time but for big size. For size, it just clearly felt like, “Okay, if this is all it ever is, forever this is great.”
[(29:25)] Jax: What were you talking about at the time?
[(29:29)] Sam: So it was just giving ballparks. I think we’re doing 10 million, $15 million a day of volume on it. And the spreads were between 5 and 20%. You can do the math there. It’s a nontrivial amount of profit from that. It was a trade that you don’t see.
The rest of the world’s like, “The spreads are too big.” The sort of trade that you might just doubt could exist, it could really be real in crypto. At some point, we just realized, “No, it’s real.” This is not all just one long red herring. It’s a great feeling when we put all of that together. We’ve been toiling and trading in Alameda for a while, probably 2 months or so, before we made any money.
Basically, what happens is trade after trade that we’re not quite able to do. It would have two of the three necessary commodes. We have the two exchanges we learned half the night that we need or whatever. It felt like unlucky each time. I’m like, “We’re pretty close to being able to do this.” It kept not happening. I think this is one of these points where one of the conclusions you could draw was, “It’s never going to happen.” Another conclusion is just, “No. We’re two-thirds away there. It’s been two months. Let’s give it another month.”
It turns out the second conclusion was right. We started to get a taste of it with some trades where we had withdrawal limits and we didn’t have that much capital. But we’re able to do them and able to clearly make money doing them. Albeit, not as nearly as much as we could have made if we had no withdrawal from some of the exchanges and anything in capital at all. But at first, we were like, “Okay, no. We can make money doing these if we do them right.”
[(31:16)] Martin: Just to come back to it, it was a great feeling. But was it also a relief that you tested the cycle because of all the infrastructure that you needed?
[(31:24)] Sam: Yes, it was. Maybe for some other people, even more so than I, I think I’ve always been more willing to take wacky things without proof because I think it’ll work. There’s this really nice feeling of like, “Okay.” We’re worried that this is never going to work. Now, we’re not worried about that. Now, we know that this can work.
Not to say that it will always work. But there’s no sort of catch of like, “But this will never actually make money.” I think that was a pretty big relief. It was pretty motivating to keep trying to build things out. All of a sudden it seems like, “Yeah, clearly the sky is the limit.”
[(32:11)]Martin: Do you find that the real challenge is when you’re on the gravy train and you’ve got it to go to work? Give us an example of how you scaled up. How did you find a team of traders behind you? How do you get all the pieces in place?
[(32:28)] Jax: Good question.
[(32:30)] Martin: Yes, poorly.
[(32:30)] Sam: I mean, trial and error and a lot of error.
[(32:34)] Jax: You bought the gaming chairs and that was it, right? The people came.
[(32:37)] Sam: That’s right. It turns out that you needed people to sit and then the chairs themselves didn’t do a lot. But this is a huge problem for us. What happened was that we clearly needed to scale the team up at that time to be able to do what we want to do. You needed 10 times the manpower that we had. We’re doing everything we could to try and scale it up.
But the problem that we’ve run into was that when you’re a 5-person company, and then you send a market order for another 30 people in a month, it’s a lot of growth. It’s not easy to manage that. It’s not easy to hire that well. I never run a business before, of any kind. I had experienced mentoring people. But the thing about mentoring people, if it doesn’t work, you can just say, “Whatever. They screwed up.”
[(33:24)] Jax: Yes. Just move on.
[(33:24)] Sam: So be it.
[(33:25)] Martin: But that’s not a good answer, right? But it’s true.
[(33:28)] Sam: Well, it’s true.
[(33:29)] Martin: Your hands are off, right.
[(33:30)] Sam: That’s understood. When you’re a third-year employee at a company, you’re not expected to work miracles and save every possible situation. You’re expected to give a real fighting chance to anyone you’re mentoring. When you’re running a company, you’re not supposed to give a fighting chance to the team. That’s not how it works.
I didn’t have that context. When things got bad, when we lost a little bit of money, it’s just ways terrible in so many ways I didn’t predict. It wasn’t just that we lost some money, it was that we lost our will and our confidence that we could do good trades. All of a sudden, there’s tons of conflict on the team that didn’t exist before. It’s easy for everyone to agree when everything’s going well. Everyone’s just, “Yes, what we’re doing is making money. Let’s make money. Let’s keep doing it. This is good.”
[(34:26)] Jax: Is there a lack of a leap of faith for the people you brought into the team, basically?
[(34:30)] Sam: Yes, that’s exactly right. As soon as results start to not look good, everyone’s like, “Wait, why are we taking this leap of faith?” This seems too good to be true. Now it doesn’t seem true. Half of the people left. It was a huge nightmare.
[(34:48)] Jax: Wow.
[(34:50)] Sam: I had no idea what to do. I wasn’t actively making it worse. But it’s not fixing it. And it’s my job to fix it. I was confused and frustrated at why everyone was confused and frustrated. I said, “Hey, guys. We just screwed some things up. Let’s just fix that and keep going.” That didn’t work.
[(35:10)] Martin: That a good honest answer. How many people are in Alameda and FTX right now, like ballpark?
[(35:17)] Sam: Alameda is 15 or so. FTX has close to a hundred.
[(35:21)] Martin: There’s enough scaling lessons in resource there. I teach entrepreneurship for a living. I know the pain of building teams. If you lose a lot at one go, it destroys the business. That to me was a recovery. You came out of a loss.
[(35:35)] Jax: So how did you get through that?
[(35:37)] Sam: It was interesting. The day that people left, our total company productivity quadrupled.
[(35:43)] Jax: Why is that?
[(35:46)] Sam: It was really striking.
[(35:47)] Martin: Breaking news, yeah.
[(35:49)] Sam: Yes. Basically, the answer was that like the thing holding us back more so than anything else was conflict. That conflict between team members was coming from a conflict about whether the business was going to succeed. Whether it had a fighting chance to succeed.
When a lot of people working for a company don’t think it’s going to work, it’s not going to work. No one is good at working for a company they don’t believe in. It gets in the way of everything. Every project can’t get off the ground because a lot of people think it’s hopeless. They think the whole thing’s hopeless.
As soon as that changed, as soon as it was filtered down to the people who were excited about it.
[(36:34)]Martin: Believed, yeah.
[(36:36)]Sam: Basically, overnight, that solved the biggest problem that had been stopping us from being able to get our shit together. In really short order, we did get our shit together.
[(36:48)] Jax: Good for you.
[(36:48)]Martin: Good for you, man.
[(36:50)] Sam: It’s cool to see.
[(36:52)] Martin: There’s an analogy. Funny enough, this comes from your roots out in the valley. It’s called “avoiding cultural cancer.” The whole idea is early detection in culture. When you’re building teams, you go in with, “Look, guys. We’re not going to win, right? We could have some failures here and that’s going to be okay. But we want to learn from them, right?”
You set their expectations. But, at the same time, you’re looking for that early start of benign cancer. The cells that you’re going to have to eradicate. That’s a very aggressive way to think about building teams. But you’ve got to get them out.
[(37:27)]Sam: Yeah, for sure.
[(37:28)]Jax: Otherwise, it becomes cancer that kills the company.
[(37:30)] Sam: Completely agree. It just spreads incredibly quickly. It was just plausible enough that no one could quite fully dismiss it. It kills the whole tribe. It is absolutely right. The sad thing is, a lot of these were great, well-intentioned, brilliant people. Sometimes it’s not the right fit. Sometimes it’s just not what they believe in. Some random thing goes wrong. It’s not necessarily about who’s good or bad. It’s about who wants to be there and who’s on board with the mission of the company as it is.
[(38:08)] Jax: Is there a way you’d vet for that now? If someone wants to come and work for you, are there questions that you ask that get into that?
[(38:15)] Sam: Yes. It’s become an enormous part of our hiring process. It’s trying to figure out who is going to be excited here, and who’s going to be happy here. We try and make it sound bad. We try and talk a lot about all the worst parts of the company and the job when we’re hiring. They’re going to learn about those one way or the other in the end. We want them to learn about it before they decide whether they’re going to join.
We don’t want them to join if they’re not going to want to be here once they realize what it’s like. I think there are a lot of amazing parts that so I honestly think it’s fantastic. But there are pros and cons. I think it’s really important that people come in with their eyes open. I know people sometimes do have a sense of what sounds good and what sounds bad to them. You describe here the fast-moving, chaotic nature of the industry.
I think some people are like, “That sounds super exciting.” Other people like, “Oh, boy, that sounds so stressful and unpleasant.” I think, “Great, glad that we had that talk. That’s what it’s like. If that sounds stressful or unpleasant to you… You may be a great person and a great worker, and brilliant, this isn’t the company for you. There’s nothing wrong with that. There’s no point pretending that this is what you should do.”
[(39:34)] Martin: This show is all about finding insightful lessons. If we think it’s destructful for us, it’ll be great for us, this person, et cetera. So, I’ll throw you one out there. Over the years, I’ve invented a lot of the shit but I used to say to people. When I hire people, “Do you surf or swim?” I would say, “Hang on.” Actually, swimming is easier than swimming. I’d say, “Can you swim in 20-foot waves? Let me make that very clear.” That gets to your point that it’s actually hard. Let me tell them about what could be ahead of them. If you had to give a tip about your hiring practices, given you’ve now you’ve done a bit of tolling out, you scaled. Do you have one favorite question or approach you take for hiring?
[(40:15)] Sam: It totally depends on the role. For different roles, there are just different things that are important. There are some things are shared, but a lot of it is somewhat different. Frankly, I wish we had better questions. I try to think of some. I think I’ve underperformed but I was hoping in terms of that. I think that one of the things that I like doing is describing day-to-day here and just listening to what people’s responses are. They’re like, “That sounds really unpleasant.” Or they’re like, “That sounds so much cooler than what I’m doing right now.” That’s extremely telling but I think other things these are tricky. It’s really tricky to get them in a way that isn’t so context-dependent that you’re just filtering how people thought about a very particular situation. I think another side of this is filtering. When faced with a complex messy real-world situation with high stakes and a lot of uncertainty, use your instincts. Let’s try and solve this and figure out the right way to think about everything and not act until we get there. Or is your instinct like, “We’ll do what we can, and we’re probably going to do something wrong here but that’s okay.”
In different contexts, different lessons are important. But by and large in our company, we do a lot of the second thing. We do a lot of, “Look, we’re never going to know for sure. But we got to make a call here.” What’s your best guess? Do you understand the risks? Do you understand the trade-offs here? Can you make a call which is well-informed with low downside and high upside? Does that freak you out and feel I’m claiming like, “Look, I don’t want to do something here until you know the answer.”
I think the dev side, interestingly, is where we see some of this most clearly. There are a lot of really great developers who find it very important to their work that they write beautiful code. I think that there are advantages to it. Our industry moves so fast but you have to iterate. You have to be iterating quickly. You can’t get feedback until you launch something.
That’s been one of our most consistent pieces of feedback on developers. Given our business and our industry as it is, in order to fit in well, it’s important that you must be thinking hard about prioritizing the output on the product, rather than the code for its own sake. Except in some circumstances where the code is filling a fundamental role. It’s important that it really is.
[(43:03)] Martin: I talked to you about this, Jax. I think it’s a pretty much proven concept. In hyper-growth markets you can build out infrastructure on the fly, right? Build your own redundancy. It may not be the most efficient, but what you got to do is secure the market. You’ve got to secure the grab.
Then what you can do asynchronously is then go and build a better system. To build the second generation or next-generation FTX. But actually, you got to go and grab the market. For you, it’s perfect, right? You don’t need to have everything. Building braces, it’s going to last 10 years.
[(43:38)] Sam: You have to be extremely aware of what you can do eventually. You have to make sure that you’re not screwing yourself over. But if you are confident, you can get there. You don’t have to be there now. We think about that a lot when it comes to things like matching engine through put where we don’t need to have a million transactions per second capacity with our matching engines. People aren’t trying to send that many orders.
How many orders are people actually trying to send? They’re trying to send 50,000 transactions a second. So we need to have that much. But that’s going to keep going up. We need to be able to support the demand right now. Beyond that, we don’t want to be spending all of our time focusing on scaling up for demand that isn’t there, instead of building up a product people want.
But we also need to be confident that when demand does come up to 30,000 transactions a second, we can scale to meet that. We can then devote the time to get there. I think that’s exactly right. Just with the caveat of making sure that we understand that there is a medium-term roadmap to where we’re going to need to get.
[(44:43)] Jax: Well, I mean, Sam, I know Martin has some views on why cryptos exist and what their financial benefit is. For me, there’s still a lot of confusion as to exactly what cryptocurrencies are and where they get their value. You feel like the person to sum this up for the average listener, in the layman terms.
[(45:02)] Sam: Yes. There’s a lot of different types of crypto. They get value from different things as silver lane, but transfer? Maybe to give a few examples of that, one thing you could look at is Bitcoin and where it gets its value from. It’s similar to what is gold or fiat currency gets their value from. To some extent, it’s our collective imagination. It’s also agreed that this is a thing we’re going to ascribe value to and it being a good enough functional product to be able to hold that value. That serves as the answer for Bitcoin.
I think when you look at some smart contract chains like Ethereum, part of the answer is it’s the world’s global decentralized computer where you can submit your custom code to it and the blockchain will run that code. It will run your applications for you in a decentralized way. It gets its value to some extent from its ability to successfully run those programs and to scale for them, and provide the right guarantees and trust that you need.
You can look at Dogecoin and where it gets value from. I guess because I don’t know where GameStop gets value from. Where’s Tesla get some of its value from? You can call it a meme stock. You can call it an all client college Bitcoin. You can call it what you want, but it has value. If people give it value, then it has value. That’s what value means. Deny that at your own risk.
[(46:27)] Jax: Yes. The tricky part with something like that is, does the value evaporate once people are off of it if it’s not based on something tangible, isn’t it?
[(46:34)] Sam: Yes.
[(46:35)] Jax: That’s, I guess, what’s hard to get your head around.
[(46:38)] Sam: No, It totally could. Because of that, if it’s not based on something tangible, you have to have a sense of our people giving up on it. They’re same as gold, right? Gold’s value would evaporate if everyone got bored of gold.
[(46:49)] Martin: Totally, yes.
[(46:52)] Sam: People probably won’t get totally bored of it. They might get somewhat and I think that, yeah. With these things you have to decide like, “Is this a flash in the pan? Or is it something the world’s going to remain excited about? Maybe even get more excited about overtime?” It’s not always obvious. I think that’s a lot of what determines what the real value should be of a token.
[(47:11)] Martin: Can I add to that? Say, someone finds a rough diamond, right, and there’s a process to cut that diamond in a number of ways. Let’s just say if it takes six steps, it never changes. That’s it, that’s diamonds. The only other thing is that there’s a certain amount of supply and that’s it. If someone decides to sit down at a restaurant on the same day, every day because they think it’s great and they’ve got the most valued seat, they enjoyed a conversation, that’s value. They pay for it, right?
But actually, the best value is the value that has multiple applications and can scale. Think about Bitcoin and the multiple uses of Bitcoin, or XRP in terms of low commission costs really fast here, really fast settlement. The question is the application, right? You’re not just ascribing a value that you think, “You know what, I just love the idea of cryptocurrency. I like the idea of being able to work across the globe, and not have anyone fucking around with my money. However, I want to be able to use it in many different applications.” That scalability is probably the ultimate value, right?
[(48:14)] Sam: I tend to agree. I don’t know if I’d say it’s the ultimate value but it is a big ultimate value. I think that it’s one that people don’t think about very well. In particular, I think people are kind of short-sighted. With respect to, I think…
[(48:26)] Jax: Isn’t it that the most important factor in terms of the staying power of a coin?
[(48:31)] Sam: Well, it’s a good question. It does not empirically seem to correlate that closely with the market cap of tokens right now. I think it might continue to do so or more over time. I think that if you look at the biggest tokens right now, they are not the most scaling tokens. I think that one thing that’s going on here and I don’t know for sure but my best guess is that people just don’t think far enough in advance. That people think like, “Well, how many transactions do I want to send? How many transactions is each network has? Which have enough? Great, let’s go of these.”
What they don’t think is, in six years, specifically selecting for the cases that matter the most, specifically selecting for the cases where crypto gets huge, how many transactions am I going to want to send? What is able to scale up to that? That just gives you very different answers. Right now, the answer is that you need like a thousand transactions a second roughly, to run the application that existed in decentralized finance.
How many tweets do you think there are in a second? It’s more than a thousand. Let’s say 100,000, right? How many trades are there in New York Stock Exchange per second? How many Facebook likes are there per second? How many visa payments are there per second? You can look at all these things in the real world, and when you have a billion users, it turns out you often have 50,000 to 5 million transactions a second. It’s the sweet spot for global-scale programs and almost no blockchains. Even have it in their roadmap to eventually get up to anything close to that. Even if many of them can scale to what we have right now, most of them cannot scale to where things are supposed to go.
[(50:28)] Jax: Whatever, they can’t scale to that ever, you think?
[(50:31)] Sam: Ever.
[(50:32)]Jax: Oh, wow.
[(50:33)]Sam: That’s kind of a big deal. When you hear people talking about it, they don’t talk about it as if that’s going to be any important to get there. It’s just not part of their mental picture of things. I think that’s a big oversight a lot of people have had. It’s just thinking too short term, in terms of what the demand is going to be.
[(50:53)] Martin: This area fascinates me because having built a lot of technology over the years, I just loved the concept of centralized versus decentralized. What that means to infrastructure, what it means to bandwidth. I can tell you, being a math guy, being a guy that built computers, gone through the whole thing, 5 million transactions per second is not intellectually difficult to do. The problem is when you put it into a network, all of a sudden, it becomes huge.
Let’s assume the world’s not devoid of visionaries, right? If you can pick a coin that you think just got the ultimate underlying ecosystem and you say, “I’m going to build out that and make sure it can transmit 5 million transactions per second,” right? The question is, do you need a mix of a centralized and decentralized structure? We’re going to talk about DeFi. But, this idea that you might need a centralized coin or something in which you can ultimately distribute your server input. Think of Google.
[(51:51)] Jax: Can it be centralized if it can’t even handle the transactions? That’s what I hear from that. You can’t handle it.
[(51:58)] Martin: This the point. When it is completely decentralized, because reliant on the globality of the infrastructure, right? That’s got different transmission speeds. But if you can centralize part of the architecture, and then decentralize the rest, you could get it up.
[(52:10)]Jax: I got what you mean.
[(52:11)]Martin: I’m sure you’ve thought about this because this is the gold run, right? If you can find the ultimate currency, you’re a visionary, and you are clearly thinking about the future, this is a problem you got to solve. Is it sitting in the DeFi world? The CeFi world? What currency is it? How do you think about ultimately what that architecture looks like?
[(52:34)] Sam: Yes, it’s a really good question. The first thing I would do is to start with the things you’re most confident in. Then work out from there. What are the things you’re most confident in? What can some things never do? What can DeFi never do? Never is a strong word. But if you want to be globally decentralized, what does that imply?
Well, it implies that you probably have some nodes all across the world. In order for a transaction to happen, it has to go through a bunch of those nodes, which means light has to travel around the world. Anytime, anything happens in DeFi, every single tech has to be light traveling around the world. That takes like 100 milliseconds. That just gives a theoretical lower bound, basically, to how fast DeFi can be, right? DeFi’s clock speed cannot get faster than 100 milliseconds.
If you’ve any program that is deeply going to care about back-and-forth or interactivity on a scale faster than that, that interaction cannot happen in DeFi. It has to be more centralized. You can say, “Okay, do you care about that?” And the answer is, “I don’t know.” Some things do, some things so, right? When you’re tweeting, you don’t care. No one cares about a 100-millisecond delay in sending a tweet. That’s like 3 seconds from GUI for reasons no one understands. It’s totally fine to put Twitter on DeFi, right? Not relevant.
When you have two HFT firms trading against each other, do they care about a 100-milliseconds delay? Absolutely. Two HFT firms trading against each other, it’d be hard not to be in DeFi. You can go through task by task. Say like, “Can this fit in DeFi?” I think there are about half of the things that can and half can’t in terms of the world’s activity. So, you can cross off half of the things. The half that’s remained, that’s big. You get half the world on DeFi. That’s spectacular. That’s a $100 trillion value in DeFi, at least, probably more.
Say, we have that. Now, how about throughput? Since we talked about some sort of latency, how much in total can get through a decentralized network for a second, right? This one’s a little bit subtle. It’s not really, totally clear that there are fundamental constraints on this. There are constraints on computers. Our computers fast enough, right? Or, not fast enough. You can parallelize it. It’s part of this thing. If you have a bunch of independent shit going through the network, it’s not that hard. You can always rent more computers. The truth is that computer’s not the most scarce resource.
[(55:16)] Martin: Agree.
[(55:18)] Sam: The bottleneck is latency. The other big bottleneck here is the single points that lots of programs flow through. You can have, theoretically, 10 million transfers happening at once on the network. But if 10 million people all want to move the same token at once, from the same address, you can’t process those in parallel. They’re all drawing on the same resource.
Those have to go one by one. You’re limited by the computer’s clock speed. You can’t meaningfully have different computers processing those in parallel. They’re going to run into each other. The other fundamental constraint is, basically, that in the end no matter how big your decentralized network is, in terms of things which cannot be paralyzed, things which conflict in terms of what they spend, you can’t have more and think that can flow through one computer.
By the way, basically true for a centralized network as well. When we think about scaling FTX, that is the hard part. If we could just rent more boxes and scale up, it’ll just not be a problem. You’re like, “Why would we ever build a good system?” Like, “Yes, we’ll double our AWS bill. No big deal.” Right?
[(56:32)] Martin: Right.
[(56:33)] Sam: But the problem is that when you do that, you can’t. Like, two people are trying to look at the same offer in the same order book. You can’t have them go through different computers and not interacting, right? Like, “Only one of them can buy that.” You run into all these things where there are bottlenecks. Where it’s either impossible or, in many cases, possible but very tricky. To safely parallelize things like risk checks on an account. You don’t want the bills to spend the same risk check twice.
There are just fundamental constraints on it, but that leaves a lot of the world. HFT firms aren’t going to be trading with each other on DeFi. But Robinhood could be in DeFi. Robinhood’s traders, I mean, that’s a GUI that takes a second to use, right?
[(57:17)] Martin: True.
[(57:19)] Sam: A 100th of a millisecond is not important on that scale. The low-engagement retail financial applications can absolutely happen in DeFi. Long-timescale applications, some borrowing-lending can happen in DeFi because you’re not constantly wiggling things in and out. Then obviously, you can think about what’s the advantage. For which things, you get a ton of value out of putting them in DeFi, which things do you not want flowing to your centralized systems.
I tend to think we don’t have great instincts on that question. People think they know the answer, but when you drill into it, you can make comfortably compelling arguments for a lot of things. “It’s going to be super contingent. We’ll see how things play out.” But for a ton of the world, and a ton of the world’s infrastructure, there’s actually, decently compelling arguments for why it might make sense to put it in DeFi. If it’s not super late, it’s sensitive.
[(58:08)] Martin: What I sense, Sam, and this is a compliment because I see a lot of people that start from the “how”. They stumble on something they had. They continue going, and they don’t go back to the “why”. But you tend to be very rooted as I am. I think the correct place to start is, “Why are we doing this in the first place?” Do we continually go back to, “Does this make sense?” And along with that “why” question, just why bother with DeFi? Just make the argument for why DeFi is important or decentralized finance.
[(58:41)] Sam: I’m not going to make the argument for why it’s definitely important. I’m going to make the argument for why it’s plausibly important because it’s very hard to prove something here. In the end, this is going to come down to judgment calls.
[(58:53)] Martin: Fair point.
[(58:55)] Sam: My goal is to get people to a point where they think that a lot of people might think it would be important. That’s plausibly very important. What are some arguments here? Let’s think about what were some of the biggest shit shows that big companies have gone through over the last year? Really negative incidents that hurt their user base? What are the first ones that come to mind?
[(59:20)] Martin: Well, take Enron.
[(59:22)] Sam: Yes. Okay, great. Enron, not a good situation.
[(59:27)] Martin: I admit that was a bit of a curveball. But I think they all apply.
[(59:31)] Sam: They all apply. What happened with Enron? Well, they cooked their books. What could you do? Well, if you make books be on-chain, at least some pieces of it, some checksums for it, it makes it a lot harder to fabricate them. What else in the last year, that has had a total shit show or at least had a lot of controversy in pushback?
[(59:51)] Martin: How about price-fixing? Let’s take in one of the many that are out there right now in the last couple of years. Let’s take Elon’s $420.
[(59:58)] Sam: Yes. I think when you look at price-fixing, it is an interesting thing. I think there are different cases of it. You got some of them, you have to decide what the laws should be. On others though, when you look at things like the LIBOR scandal, part of the promise is a completely bespoke rate that was not calculated in a transparent or straightforward manner, that’s it’s very hard to vet, very hard to see if it was biased. That’s something DeFi could help with. You make the process transparent. People can’t just systematically bias it in their direction, silently.
Right, what else? Think about social media around the election. There are huge controversies. 2016 social media did not block anything. They got destroyed for not blocking Russian propaganda. Then in 2020, social media banned a bunch of propaganda including Trump’s account. It got destroyed for intervening in a partisan way. It’s a little bit of, “damned if you do, damned if you don’t”. But what it exposes is that no one knows what the right answer is here. It’s a little weird if the answer is Jack Dorsey. If the answer is like, “That’s the guy who decides what the world gets to see.” Nothing against him. I think that’s a compelling case. If you build DeFi social media, what can you do? Well, again, the underlying blockchains are censorship-resistant. People can post whatever they want and upload whatever they want there. Anyone can host a GUI that draws from blockchain data.
You can use your own GUI that censors or don’t censor. It does whatever you want. Anyone can access the data if they want to host a platform or GUI. That does it, but the data is all free. All the platforms can access the same underlying data. I don’t know if it’s the right answer but it’s a pretty plausible solution to this question. You can let people choose whether they want to be on a censored or censorship-resistant platform.
I don’t mean that in a leading way. I think there are compelling reasons I want to be on a censored platform. There are a lot of things I don’t want to see on my feed. I want to moderate or filtering those out before the mods go crazy. I’m like, “Fuck. There’s another GUI. All my tweets are still there. I’ll just go somewhere that doesn’t have shitty mods.”
Here’s another one, how about Robinhood? Robinhood has had some pretty bad days. One in particular. What happened was a completely nightmarishly complex and intricate financial system where no one knew where anyone’s money was for sure. Robinhood was in custody bonds. There were separate custodians and clearing firms. But the dollars and the stocks are all over the place. No one knew for sure if anyone was going to deliver anything, anywhere.
The SEC was like, “You got to put 5 billion dollars of collateral.” Despite having no positions because your customers might fall. The money they sent in might not make it, or have made it, or might be recalled. That’s the thing where if you have actual straightforward transparency about what is where, you don’t need to worry about that. If you don’t, then you just have this massive system relying on trust. It just breaks down sometimes. I think that is another case where there is at least a compelling reason that DeFi would not have that problem.
[(63:09)] Martin: It might also reduce capital adequacy. We could talk forever on decentralized finance because there are so many angles, I mean, the argument. I think the biggest one, and I love the fact that you didn’t say “You fully commit, this could be successful.” Because a lot of the current system works because it has checks and balances and they are not involved, but they’re always exposed in the end something bad, that’s off balance sheet or someone fucking around..
But the “we” are smarter than me. That’s the idea of an open decentralized transparency system. The collective masses say, “I think this is good for humanity.” That’s kind of a pretty cool argument. But where that ends, I’m fascinated to find out.
[(63:50)] Jax: Thinking about the future, for you, Sam, what are the currencies to watch? Bitcoin’s had a boom. Ethereum was rocketed in the last year. What crypto should everyone know the name of?
[(64:00)] Sam: Dogecoin. Everyone knows that name, but you should. I don’t know how things go in the end. Do you know how Time has a person of the year thing? If there’s like an asset of the year for 2020 and 2021, it would be Dogecoin. That is the asset that accurately reflects our current economic climate. That is the asset that we have all chosen and that we deserve, for better or for worse.
[(64:25)] Jax: Well, you heard it.
[(64:27)] Sam: I don’t say that as an endorsement or an anti-endorsement of it. I’m not trying to take a position on it. I’m just saying, everything we as society has decided over the last year has been dragging us closer and closer to the Dogecoin financial standard.
[(64:41)] Martin: I got to press here. Just to make sure, before I start moving some positions.
[(64:44)] Jax: That’s what I was going to say, you heard it here.
[(64:47)] Martin: Hold on. Just for transparency, I’m not in Dogecoin, right, at least not yet. But why did you say that we’d all chosen? I would love to know what is your early evidence that other than Elon saying, “Choose Dogecoin.”
[(65:05)] Sam: One thing I want to say is it’s up a $100. That doesn’t necessarily mean it’s a good buy or sell here. You have to decide whether we’ve chosen more or less than the amount that is already appreciated. But what have we done as a society? When did we swipe right on Dogecoin? There’s just been a ton of things.
Elon is a powerful piece of this. He is the most influential man in the world when it comes to financial assets right now. His tweets move markets. He chose Dogecoin and no one was surprised. Everyone knew he’d choose Dogecoin. It’s not like “That’s why he chose it.” Like, “Okay. No, obviously, acting in retrospect.” That’s what he’s going to…
We chose Elon. We signed up for this. We signed up for some amount of self-aware mockery, for some amount of sense of humor, and for some amount of challenging norms. Saying, “Okay, but seriously, maybe, well, makes your thing better than Dogecoin.” No, really answered me that. Yes, I realized it’s a meme. My question still stands.
When you look at what happened on GameStop day, the only surprising thing was that it wasn’t a cryptocurrency. Because apparently that happens all the time in crypto, it’s not unusual. The moment that Robinhood banned the buying of GameStop, do you what happened? Hundreds of millions of dollars flowed instantly into Dogecoin. Directly out of GameStop, into Dogecoin.
Those are the same type of asset. People see those very similarly. If you drill into, “Why were people buying GameStop?” It was a, “Fuck you.” It was a “Fuck you” combined with a little bit of, “Look, you think that this thing is cheap?” I don’t think you understand markets. I think we understand something you don’t. If we all get together on social media and decide with our collective imagination that this thing has value and you decide it doesn’t, why are you so sure you’re right? Maybe you’re not quite as right as you think you are. Not as clearly as you think you are. What if we collectively, on social media, accept more capital than you do?”
[(67:22)]Jax: Yeah.
[(67:23)] Martin: It’s a super interesting question. Do you think that that’s sustainable in Dogecoin? Do you think the wider the application, I mean, what pressure wanted you to apply?
[(67:32)] Jax: How does the application of Dogecoin work?
[(67:34)]Martin: I was just going to say, maybe that’s the question.
[(67:38)]Jax: What separates it from other coins?
[(67:39)] Sam: It is, as one of my colleagues and friends put it, objectively the funniest cryptocurrency. That’s what separates it. It’s not the tech. No one thinks it’s the tech. No one even tries to argue with the tech. No one who is buying it buys it for the tech, or the product, or the road map. They buy it because it’s a funny-looking dog and the fact that it’s pronounced “doge.” A kind dog but it’s not. That’s why they buy it.
[(68:05)] Jax: Are you serious right now?
[(68:08)] Martin: No. Sam has answered this exactly how I would answer it. There’s nothing else. We’re ascribing value to it.
[(68:16)] Jax: I need the Dogecoin marketers to come and work for me.
[(68:20)] Sam: You know what the cool thing is, there’s no Dogecoin company. They don’t hire anyone to market it. The founder gave up on the cryptocurrency a decade ago. No, that came about five years ago. Who’s our market here? It’s Elon Musk, and it’s Reddit, and Twitter, right? It’s just a bunch of people getting together. It’s memes. What’s their most powerful brand? It’s that picture, of like a cloud coming over the city with the Dogecoin face.
[(68:53)] Martin: It reminds me of Scooby-Doo, Sam.
[(68:55)] Sam: Completely right, it could be Scooby Coin.
[(68:57)] Jax: I got to look at this. Hold on. I’m going to look at Dogecoin now.
[(69:00)] Martin: This is the question. Why can’t it not be, the currency?.
[(69:03)] Jax: To be fair, I would buy this coin. It looks cool.
[(69:06)] Martin: It’s super sick.
[(69:07)]Jax: That’s great. The coin is cool.
[(69:07)]Martin: But do we need anything else for this to be a permanent currency? If we all ascribe that this is the means.
[(69:09)] Sam: How about Bitcoin? What makes Bitcoin worth more than Dogecoin?
[(69:17)] Martin: I agree.
[(69:19)] Jax: I always thought Bitcoin, in my basic view of crypto, is the most stable. That’s why everyone goes for Bitcoin.
[(69:27)] Sam: Absolutely. A huge built-in advantage. But if enough people get together and decide that it’s Dogecoin now, could that shift?
[(69:34)] Martin: Yes.
[(69:34)] Sam: Maybe.
[(69:36)] Jax: Once people move on to another coin, it gets its 50 minutes of fame, and then it drops off. It’s like an Instagram coin. Do you know what I mean?
[(69:44)] Sam: You need loyalty. If you don’t have loyalty…
[(69:45)] Jax: Yeah. So how do you build loyalty for Dogecoin, outside of…
[(69:48)] Sam: It’s a really interesting question. That is the collective experiment the world is undergoing right now, right?
[(69:54)] Jax: I don’t like this experiment, man.
[(69:55)] Sam: It’s not just Dogecoin. What is Elon Musk’s best product? What is the most valuable product that he has built?
[(70:04)] Martin: His voice.
[(70:05)] Sam: His voice.
[(70:06)] Jax: Him. His brand.
[(70:09)] Sam: The “ticker TSLA” was so much more than electric cars.
[(70:17)] Martin: For sure.
[(70:20)] Jax: What’s the “ticker TSLA”?
[(70:21)] Martin: Tesla.
[(70:22)] Jax: I got you.
[(70:25)] Sam: You see this all over the place and it’s becoming a bigger and bigger part of our financial ecosystem. Because for the first time, you look, historically, about what could cause something to massively appreciate in price? They are two things: either, it was just the company that makes so much money is paying huge dividends, or the global wealth management platforms had to get together and decide that it was a great company. They’re going to put all their fund’s money into it.
Those are not going to buy Dogecoin. There’s nothing else big enough to make something be worth a ton. What’s happened in the last decade is social media. All of a sudden, for the first time, 10 million people all across the world can get together and tweet at each other and agree via shit-posting memes on Reddit and Twitter about what they are into. They can coordinate, not explicitly even, just implicitly.
You put those 10 million people together, even if they only have a thousand dollars each, that’s 10 billion dollars of inflows. That rivals the inflows you get from the biggest institutions in the world. All of a sudden, you have this global decentralized network of bored people at home on their computers who, collectively, are the biggest pension fund, the biggest mutual fund, the biggest ATF in the world. They don’t have the same fiduciary duty to shareholders that fidelity has. They don’t have the same institutional history. They can buy whatever they want. If they want to buy a Dogecoin, they are going to buy a Dogecoin.
[(72:27)] Martin: I love this.
[(72:27)]Jax: This is crazy
[(72:28)]Martin: All right. So we’re going to ask questions on the “What to watch” because this is probably the most interesting segment for the average person. Sam can put his flavor on it. But let’s hit Bitcoin. Is it going up and how much further? In your opinion, what’s happening to Bitcoin?
[(72:42)] Sam: It’s one of the things I have the weakest opinions on because it just has the most forces on it. It’s one of the more efficient.
[(72:50)] Martin: That’s a physics answer, by the way. That’s a very clever answer.
[(72:53)] Sam: Let me give you two opposing forces here which angle are going to play as usual. One of them is leverage. The crypto industry is leverage to Long Crypto right now . That’s why when crypto crashes, it crashes more because it gets liquidated. That causes you more selling. If you want to tell a story where Bitcoin goes down to $15,000 in next months. It’s $7 billion of liquidations. Stringing together with some bad news and some institutions dumping. It’s just a bloody dump, right?
All of a sudden, the narrative changes. It’s 2018 all over again and people are like, “What were we thinking? Obviously, it was a bubble.” In the last year, we’ve gone from 5% of financial institutions to something 80% that have decided that they will, eventually, do something in crypto.
[(73:50)]Jax: This is huge.
[(73:51)]Sam: They haven’t really done it yet. You talk to almost all of them, they don’t even know what they’re going to do. They’re trying to figure that out.
[(73:58)] Jax: Do you think they won’t adopt Dogecoin purely because they will view it differently compared to how…
[(74:04)] Sam: That’s a really good question. It’s not going to be the first thing they buy.
[(74:06)]Jax: Yeah.
[(74:06)] Martin: It’s not in their culture.
[(74:08)] Jax: That’s what I’m saying. It’s so alien to them.
[(74:12)] Sam: The world is getting more competitive and faster phase. People who were used to follow financial incentives are getting caught. They’re just blowing out. The world’s becoming more connected. Information is getting freer. It’s easier for everyone to see who’s losing money because of this. The world is changing. Even the staunchest places are changing. They’re changing more slowly but they are changing.
I think Tesla taught a lesson to a lot of hedge funds. There are a lot of hedge funds that were short Tesla, and they all lost enormous amounts of money. I think they’re all going to look at Dogecoin and they’re going to have a little bit of PTSD. They’re going to be like, “We’ve been there before.” We don’t want to be a melting capital.
Every bone in our body is telling us to short this. But also we kind of blew out the last time we listened to every bone in our body. I think people in Bell boom were hesitant to bet against it. You’re going to get the renegades. The iconoclasts who are going to say, “Fuck all you guys. I’m just gonna buy it and make a lot of money and I’m going to be right. You’re going to be wrong.” “Fine, don’t invite me to your dinner party for three months. In six months, I’ll be the most popular guest in town.” Grab and call it correctly.
I do think we’re seeing some of that change. It’s not all or nothing. It’s bit by bit. But the institutions are not going to be as Dogecoin unfriendly as they used to be. We forget how much we’ve grown. Things that just seem obvious now, one example, what is a boring, safe, blue-chip stock, as a stodgy firm that you’d put your capital in, if you just wanted no one to be able to object to it?
[(75:55)] Jax: General Motors has got a thing.
[(75:56)] Martin: It used to be IBM, right?
[(75:57)]Jax: The IBM, yeah.
[(75:59)] Sam: It used to be IBM. You’re putting General Motors right now. You kind of worry that it’s going to go bankrupt, right? I don’t know if that’s true but that doesn’t. It’s not the thing that the most conservative asset managers currently think is necessary, fundamentally. I think Facebook is way up there, right? Talk about a booming business with a long track record that has huge revenue, huge demand. It’s to reflect has ascended to that length. But where did it start?
You watch the dramatization on self… I’m not agreeing there’s some truth and some pieces of this as like untouchable for most VCs. Only the “Fuck you” VCs were going to invest. It was so different. It’s so wacky. It was so much like, “You’re telling me some random 20-year-old dropout, with his weird internet idea about making friends, is going to be able to build a huge company. Also, his company looks like a mess to me.” The VCs investing are a mess. This whole thing is a mess. Now it’s not worth a trillion dollars.
That was not quite the accurate description of it. It was a mess in some ways. But it’s extremely sharp, detail-oriented, and strategic. A lot of extremely important ways to the business. The world has internalized now, that paradigm, if you see that pattern, that is no longer weird and untouchable. That is now what you’re supposed to invest in as it grows fund.
It hasn’t quite gone to the Dogecoin level. But on a decade time scale, funds learn what works and what doesn’t. They respond to those incentives. Eventually, they just erase any history of that not being what you’re supposed to invest in the first place.
[(77:44)] Martin: Do you think XRP, if they get passed the SEC, is going to go to the moon? For sure, in terms of its process, it’s promising.
[(77:53)] Sam: I don’t know what’s going to happen to its price. Its price might go when. I don’t know and I don’t want to say it won’t. I’m not supper bullish on its product long term. I say this without supreme confidence, but I think part of this is weird and hard to explain exactly how things got where they did. They just did get there. Nothing is building on XRP. No businesses, no applications, no products, no DeFi, no users are building up the XRP network. It’s a ghost town.
[(78:24)] Martin: But that’s because of the SCC, right now. Right?
[(78:27)] Sam: It was true before the SCC. That never been people do thing right. It’s interesting. You ask why, I can give you what is the answer most people in crypto would give, which is that it’s not sufficiently decentralized. It’ll say that it’s validated or set. It’s much more permissioned and much less decentralized than most. There’s a real point there, I don’t want to dispute that.
I assume that people are a little bit overconfident about that as the reason for it. Some of this is weirdly contingent that they just didn’t put a lot of marketing. They put a lot of marketing effort into the product. But that effort wasn’t into getting people to build on it exactly. It was into marketing the token and the institutionality of it, they never built a decentralized ecosystem on it. It just never came. The fact that it is a bit more centralized is a black mark for a lot of the people currently building on it. The SCC is a cloud that’ll pass one way or another. It’s not where we have seen it…
[(79:29)] Jax: Martin’s out here. He’s sweating listening to your explanation.
[(79:33)] Martin: I’m sweating. Sam, can you move my position out of XRP? I’m no longer comfortable.
[(79:40)] Jax: I’ve got to ask, Sam, it’s so interesting, the different value systems in the decentralized crypto world. But, for listeners that want to get into this space, what is it they should be looking out for?
[(79:57)] Martin: The better question is, what are you investing in right now, right?
[(80:01)] Jax: But why? I think the answer is “why” because it is so different. If you were to talk about blue chips, like I was saying to you earlier, you’d say, “Do Bitcoin, it’s the most stable.” Do you know what I mean? It’s going to be around. Do you know what I mean? But the “why” is important in terms of what you’re looking at and why.
[(80:19)] Sam: If you want, you just think of everything I’m caveated as saying, “Crypto’s up a ton. I invested a lot of things when they were lower.” Who knows? Maybe it’s a great investment, maybe it’s not. I don’t know. Maybe I’ve thoughts on it. I’m not going to address my thoughts on it. This isn’t about me talking about the price. What the price is now or what it will be.
[(80:36)] Jax: It’s more about the principle that you got. Yeah.
[(80:38)] Sam: Right. The principle of the network and the tokens. If I was just to ignore market forces and just be in it for the tech. Ethereum has done an enormous service to the space. I think that it really introduced the notion in a huge way of you can use the blockchain not just to transfer tokens, but to run any program. It’s basis try and complete. You can run applications on the blockchain and then have decentralized companies. That’s super cool.
It really changed crypto forever. It’s not the most scaling cryptocurrency. It’s hilariously un-scaling right now, it’s 10 transactions a second. You just don’t need to think for a very long. Ten is not enough, for the entire world to split between the entire system. It had a four-year-long plan to transition to a new network, still years away.
When you think about which networks are the most ambitious in scaling, I will say that I’ve done a survey of them and it wasn’t exhaustive, but I did a decent amount of legwork and I decide there’s one that was substantially more ambitious about scaling plans than any of other ones. At this point of financial interest as well, take everything I say with a grain of salt here. That’s the Solana blockchain. It just gives a high level of stats. Right now it can handle about 50,000 transactions a second, probably the fastest in the crypto ecosystem.
Transaction costs about hundreds of a penny. It served the highest throughput. Right now, transactions are settling in less than a second. Also somewhat uniquely and like almost any other network. The team cares about scaling and is doing everything that they can to increase that number. This is not the final number. This is where we are now. First of all, it naturally scales with Moore’s Law. Anytime that computers get faster, Solana gets faster. Who knows how long that will continue but certainly, historically that’s been pretty big.
It has just a lot of concrete plans to get another few orders of magnitude. Over the next few years, it’s actively working. I feel pretty good about its prospects of getting up to about a million transactions a second for the next couple of years. That’s one of the reasons that Solana is way more likely than most, to get there because it’s going to try.
[(83:10)] Martin: Is Solana being built in centralized applications as well? Is the technology been used for other coins?
[(83:17)] Sam: So, the basic answer to that is what’s Solana? I was not involved at all in developing the Solana blockchain although I’ve been involved in building some protocols on it. Basically what it is, is it’s a super high throughput blockchain super awesome team. What we did, we’ve built a bunch of things on it, including a DEX. It’s an exchange. It’s fully on-chain and it’s actually full-exchange. It has an order book and a matching engine. Almost no DEX’s have that. It’s almost no DEX’s can but Serum can.
It’s been live for almost a year. It gets up to thousands of transactions a second on Serum DEX alone. It works. It has been super cool to watch. It’s a nice proof of concept. Now, there’s a lot left to do, that’s not the be-all, end-all. A lot of other people are now building applications on it. There’s been a ton of growth over the last few months in the Solana ecosystem. It’s had its time in the sun and more and more people are starting to build on it which has been super cool to see as well. I think, basically, it’s just going into hyper-growth in terms of the ecosystem.
[(84:33)] Jax: You’re still looking into technology and the application as being the ground on why you should jump into currency?
[(84:40)] Sam: Right. Obviously, a fine response is, “Okay, sure. But that doesn’t explain why Dogecoin has value.” I just get done talking about why it has value. So why am I here talking about the tech? Does the tech matter? I think my phase, I don’t know. I wasn’t talking about necessarily all the things that would drive value to a currency. I was talking about what drives value to the technology.
[(85:02)]Jax: Got you.
[(85:20)]Sam: You can decide that other currencies will be the valuable ones. They very well might be. I’m not saying Solana’s going to be more valuable than Bitcoins. What I’m saying is to the extent that you think that that value will be driven by the ultimate utility of the technology, then that is the one blockchain that I feel the most excited about partly in large, yeah.