The Blockchain Debate Podcast

#3. Motion: DeFi is DeFicient (Udi Wertheimer vs. Haseeb Qureshi)

February 05, 2020 Episode 3
The Blockchain Debate Podcast
#3. Motion: DeFi is DeFicient (Udi Wertheimer vs. Haseeb Qureshi)
Chapters
The Blockchain Debate Podcast
#3. Motion: DeFi is DeFicient (Udi Wertheimer vs. Haseeb Qureshi)
Feb 05, 2020 Episode 3
Richard Yan

Guests:

Udi Wertheimer (@UdiWertheimer) - debating FOR the motion
Haseeb Qureshi (@Hosseeb) - debating AGAINST the motion

Host:

Richard Yan (@gentso09)


Today’s motion concerns DeFi, or decentralized finance. The vision of decentralized finance is financial markets such as borrowing/lending and trading without intermediaries. The hope is this will improve efficiency, transparency and cost-effectiveness. MakerDAO, Compound, Uniswap are examples of DeFi apps w/ the most traction today.

Our two guests, Udi Wertheimer and Haseeb Qureshi are active members of the crypto community, to say the least. In fact, many consider them to be OG’s in the space.

No summary does justice to the rich exchange of the debate; Nevertheless here is a teaser:

Udi sees DeFi as purporting but failing to provide true self-custody and censorship-resistance. He also sees DeFi's composability as a non-innovative feature already found in non-DeFi products. And regulations will come for DeFi if and when it gets sufficient traction.

Haseeb believes that DeFi provides much stronger form of permissionless-ness than centralized counterparts (e.g., Bitfinex), and DeFi entrepreneurs can build on top of each other's work with tools and freedom unavailable in centralized finance. He also thinks some of the DeFi systems (e.g., MakerDAO) are regulation-robust for reasons like strong team decentralization and core clientele being outside of potentially hostile jurisdictions (e.g., U.S.).

If you are interested in debating or want to nominate someone to debate, DM me at @blockdebate.

Note that nothing in our podcast should be construed as financial advice.

Source of select items discussed in the debate:

Show Notes Transcript

Guests:

Udi Wertheimer (@UdiWertheimer) - debating FOR the motion
Haseeb Qureshi (@Hosseeb) - debating AGAINST the motion

Host:

Richard Yan (@gentso09)


Today’s motion concerns DeFi, or decentralized finance. The vision of decentralized finance is financial markets such as borrowing/lending and trading without intermediaries. The hope is this will improve efficiency, transparency and cost-effectiveness. MakerDAO, Compound, Uniswap are examples of DeFi apps w/ the most traction today.

Our two guests, Udi Wertheimer and Haseeb Qureshi are active members of the crypto community, to say the least. In fact, many consider them to be OG’s in the space.

No summary does justice to the rich exchange of the debate; Nevertheless here is a teaser:

Udi sees DeFi as purporting but failing to provide true self-custody and censorship-resistance. He also sees DeFi's composability as a non-innovative feature already found in non-DeFi products. And regulations will come for DeFi if and when it gets sufficient traction.

Haseeb believes that DeFi provides much stronger form of permissionless-ness than centralized counterparts (e.g., Bitfinex), and DeFi entrepreneurs can build on top of each other's work with tools and freedom unavailable in centralized finance. He also thinks some of the DeFi systems (e.g., MakerDAO) are regulation-robust for reasons like strong team decentralization and core clientele being outside of potentially hostile jurisdictions (e.g., U.S.).

If you are interested in debating or want to nominate someone to debate, DM me at @blockdebate.

Note that nothing in our podcast should be construed as financial advice.

Source of select items discussed in the debate:

Richard:
0:00
Welcome to another episode of The Blockchain Debate Podcast, where consensus is optional but proof-of-thought is required. I'm your host, Richard Yan. Today's motion concerns DeFi, or Decentralized Finance. The vision of Decentralized Finance is financial markets, such as borrowing, lending, and trading, without intermediaries. The hope is this will improve efficiency, transparency and cost effectiveness. MakerDAO, Compound, Uniswap are examples of DeFi apps with the most traction today.
Richard:
0:38
Our two guests, Udi Wertheimer and Haseeb Qureshi are active members of the crypto community, to say the least. In fact, many consider them to be OGs in the space. A lot of the debate is around comparison between DeFi and its centralized counterparts. Is the former truly permissionless? Does it have less attack surface? How is it different in dealing with regulations? From the debate, you will learn things like why composability of smart contract platforms may be overhyped, non-binary nature of censorship resistance, DeFi's permissionless-ness for entrepreneurs, and many more.
Richard:
1:16
Also, we had previously recorded two other episodes. Our first motion was "Bitcoin will attain Global Store of Value status by 2040" with Jimmy Song vs. David Gerard; and our second motion was "Tokenization of real world assets and smart contracts are useful ideas," featuring Emin Gün Sirer vs. Edmund Schuster. So definitely check those out if you're interested in those topics. If you would like to debate, or want to nominate someone, please DM me at @blockdebate on Twitter.
Richard:
1:47
Please note that nothing in our podcast should be construed as financial advice. I hope you enjoy listening to this debate as much as I loved hosting it. Let's dive right in!
Richard:
2:03
Welcome to the debate. Consensus optional, proof of thought required. I'm your host, Richard Yan.
Richard:
2:08
Today's motion: DeFi is DeFicient. Or: Decentralized Finance is DeFicient. To my left, is Udi Wertheimer who will be debating FOR the motion. He agrees that DeFi is deficient for a lack of even stronger words. He thinks that the way DeFi currently works adds almost nothing new to the world of finance, and he has yet to see recent development in DeFi to change his mind.
Richard:
2:32
To my right, is Haseeb Qureshi who will be debating against the motion. He disagrees that DeFi is DeFicient. He sees future growth in DeFi volume, and he also sees development of practical use cases beyond mechanisms that allow getting high leverage in crypto price speculation. Gentlemen, I'm excited to welcome you both. Welcome!
Udi:
2:52
Hey, thank you!
Haseeb:
2:52
Thanks for having us!
Richard:
2:55
No problem. Here's a bio for the two debaters. Udi is an independent developer and consultant, Bitcoin activist and blockchain skeptic. He is also a self-described, and I suspect many in the Crypto Twitter would agree, an adversarial thinker and troll. He co-hosts a "toxic" podcast called "Reckless Review," which I can attest to, befitting Udi's style, is irreverent and entertaining. Previously, Udi led blockchain research at Colu, a colored coin protocol for Bitcoin. He was also a software developer at PayPal and eBay, which he believes, back then, still offers better UX than many crypto wallets today.
Richard:
3:36
Haseeb is a managing partner at Dragonfly Capital, a cryptocurrency venture fund. Before this role, he was a General Partner at another crypto fund named Metastable Capital. He is also a crypto educator and has published a comprehensive blockchain course on Nakamoto.com. He previously developed software for Earn.com and Airbnb. Prior to his journey in tech and crypto, he was a Full Tilt Poker sponsored professional poker player and he conveyed his learnings in the book, "The philosophy of Poker." Haseeb subscribes to the philosophy of Effective Altruism, where one undertakes high paying career to donate his earned money to charity.
Richard:
4:18
As usual, the debate has three parts: An opening statement from both sides starting with Udi. Then second round is a bona fide debate, with me shooting off questions to the debaters. Frequent follow-ups from both sides are highly encouraged. The last round is audience questions selected from Twitter followed by concluding remarks from both debaters.
Richard:
4:40
Currently our Twitter poll shows roughly 60% against DeFi and roughly 30% liking it. We will have a post-debate poll and whoever tips the ratio more to their side, wins the debate. So without further ado, let's get started with the opening statement. Udi, please go ahead.
Udi:
5:01
Right. So I guess if you want to to figure out if DeFi is efficient or not, we first have to probably define what DeFi is. I mean that, you know, DeFi people say that it's decentralized finance. I'm never sure what falls under different decentralized finance. Like is Bitcoin itself DeFi? It sounds like it probably should be. If it is, then what about USDT/Tether? Is that also DeFi? Is USDC also DeFi? So it's hard to tell. I think that the three properties that people usually bring up as, as the defining features of, of DeFi are self custody, censorship resistance, and composability.
Udi:
5:46
So with regards to self custody, you would say that, you know, user holds, the user would hold their own keys and they would control their own funds no matter what happens the activity that they have. So if they, you know, they take a loan, they give it a loan they take some leverage, whatever they do, they control the funds. And they're guaranteed. That's the specific terms of the DeFi app will be executed. But it seems like this is almost never true. So this is almost never the case that the user is actually having self custody over their funds. Almost all of the projects that consider themselves a DeFi have usually two types of distinct controls over the funds. One of them is the Oracle. So it's the known Oracle problem. When most of these apps need to know what's the price of ether or what's the price of Bitcoin or what's, you know, what's the current market price of some assets and they can't just figure it out on their own. So someone must have the permission to tell the app what the correct coin price is.
Udi:
6:50
And if you can, you know, usually it would be the operator of the app or some other permissioned party. And this party has basically, you know, they can manipulate the contract to do whatever they want. So that's one problem that almost all of these apps have. And another one is the kind of, most of them will kind of have this emergency switch, a way to shut down or upgrade and replace the contract that is being used. And it's for good reason because if there's some bug that is found, they want to be able to fix it. But thing is if you can replace the smart contract, you can replace it to do whatever. You could, for example, replace the smart contract with a new one that says the company can take all of the funds, for example. So usually those two ways kind of control the way that users have custody on the funds.
Udi:
7:43
So the other thing is censorship resistance. And for the same reasons because of the Oracle problem, because of the upgrade problem it's hard for these apps to be censorship resistant actually, because if some state agency comes to the operators of say Compound or MakerDAO or any of the other ones, they could just say, listen, we know you can upgrade the contracts, so please upgrade the contracts. Please make sure that you do KYC checks to all your users. Otherwise we don't approve them or just shut down entirely. I don't know. So there's definitely a, you know, a point that you can press.
Udi:
8:18
The last property that people talk about is composability. We hear terms like money Legos that people sometimes use. In regard to DeFi apps. And I guess what they mean is that you can use an, you know, like two apps together, you can use MakerDAO to create DAI tokens and then you can take the DAI tokens, and maybe make markets on Uniswap or maybe they give a loan on Compound and so on. The thing is that, you know, traditional finance has composability too. Most financial services today, are digital and they have APIs that you can access and use. Now, it's not always true that everyone has access to the API - sometimes the APIs are public, sometimes they're not. But obviously you can have a service that has public API, and that way you can compose multiple services with multiple APIs. So I guess the first two things, self custody and censorship resistance are just not, not really applicable currently to DeFi apps. And the third thing is applicable to all apps, not just DeFi apps.
Udi:
9:23
So I guess what I'm saying is, you know, it's true for everything. By this definition, BitMEX and Bitfinex and such Bitcoin exchanges should be considered DeFi as well. And if we don't consider them DeFi, then I guess DeFi doesn't have a real advantage over there, because it's pretty much the same. So I guess that's my, my main view on this.
Richard:
9:47
Yeah, very cool. Strong words, strong words. I keep picturing if this was a rap battle, how that would unfold. But yeah, those were very cogent points. I really looking forward to Haseeb's response. So Haseeb, your opening statement. And we'll have time for you guys to have ping-pong follow-ups with each other's points later.
Haseeb:
10:05
Yes. Great. So my opening statement, so there are some points that he made that I, that I would like to respond to at some point. But to start, I'd like to sort of set the stage for what I see as the vision of DeFi, and what makes it unique relative to the other things going on in crypto. So I think, you know, Udi is right to point out that the definition of DeFi as decentralized finance could well be applied to Bitcoin, right? You can think about it as Satoshi Nakamoto was trying to solve this problem of creating decentralized money. And Bitcoin was the way that he saw in, you know, the predecessors to Bitcoin of things like, what is it called - DigiCash, eGold, Liberty reserve - all these attempts that predated Bitcoin to try to create decentralized money, they all failed.
Haseeb:
10:56
And Satoshi saw that in order to create an architecture that would satisfy all the constraints that he wanted to. To create internet native digital money, he had to pursue a different architecture. And so he came up with Bitcoin. And so Bitcoin you can think of as being this sort of base numeraire, this kind of digital gold is the way that we now tend to look at it. And that is the sort of the, that is the first step in the evolution of money is coming up with a sort of base numeraire that people can use for day-to-day transactions. Now on top of that, eventually somebody came up with the idea. I mean people started adding newer opcodes to Bitcoin and all these different altcoins they came up with, and eventually Vitalik abstracting over all of these into Ethereum, which you can think of it in a way as instantiating arbitrary contracts, right.
Haseeb:
11:42
And so arbitrary contracts, that's another stage in the evolution of finance. In a way you can almost describe that as a kind of decentralized finance. You can describe that as part of the march of progress of more and more financial primitives. And after smart contracts, what do you get? And you get almost immediately ICO's. ICO's are decentralized fundraising. It's a joint stock corporation, right. And then you have DAO's, which arguably you could, you could say is even closer to what it means to create a joint stock corporation where you have multiple people who have collective ownership in running or the economic profits of, and some kind of on-chain organization. After that, we get what we now call or at least have inherited the slogan of decentralized finance, which are things like loans, credit derivatives, synthetic assets.
Haseeb:
12:33
All of these things seem to me like following the same fundamental progression that we've been seeing in the evolution of crypto, which is developing more and more financial primitives to let us do things that previously people were not able to do in a permissionless way. And it's clear that people want this. You know, it's one of the dominant narratives of 2019 has been the rapid growth and interest in this space across the field. And you know, you have people now want to build DeFi primitive on top of Bitcoin - Things like RSK and Echo. Ethereum and all the other smart contract platforms have been racing to try to get DeFi adoption on their own platforms.
Haseeb:
13:13
And the overall way that I look at DeFi is basically this, you know, when we saw the internet really start to disrupt different fields, it became really clear that the internet was going to eat everything. And almost every single industry today is massively disrupted due to what the internet has done. It's changed the way that industries function. It's changed the leaders and the winners in a lot of different spaces. It is fundamentally changed our relationship with many of the companies that we do business with. But the one place that did not happen was in finance. Finance is basically the same as it was before the internet.
Haseeb:
13:56
The only difference is that now you can access your checking account with your mobile phone, right? But for the most part financial services are basically undisrupted by technology. And I think the story of DeFi is blockchain doing to finance what the internet did to every other industry. It is the opening-up and the, and the sort of, you can almost think about it as the un-capturing of, of stagnant industries by a technology that fundamentally lets anybody get access to these products, lets anybody build in a permissionless way. In the same way that, you know, any entrepreneur can go out and build themselves a new way to book a doctor's appointment. Right? I mean, that's literally ZocDoc or you know, many other products that people can use to innovate in different parts of the economy. They've never been able to do that in finance. You cannot create a new interface for your checking account. You cannot create a new way to do lending. You can, but it's extremely difficult, extremely high regulatory burdens. You can't do it internationally. DeFi gives you the first platform from which that is now possible. And, to be clear, DeFi is tiny today. DeFi of MakerDAO, and DyDx, and Compound and all these platforms that we all have to talk about. They're, they're pretty small.
Haseeb:
15:13
They're much more smaller than Bitcoin, much smaller than Ethereum, much smaller than anything going on in the real world. But they're all very, very new. These things have really only been around about a year and a half in earnest. And I think we are, we are already seeing the growth, the influx of entrepreneurs, the rapid pace of ideation and new approaches and new UX and new ideas and all of this stuff I think is going to result in a pretty rapid germination of a new way of thinking about finance and a new way of approaching the way that people get access to financial products around the world.
Haseeb:
15:51
The last point that I want to make, and then I'll cede my time is: It's pretty clear to me - I mean Udi was talking about, you know, is USDC DeFi, is a MakerDAO DeFi, I think it's pretty clear to say MakerDAO is DeFi, whether USDC is or not, is just depending on where you draw the boundaries.
Haseeb:
16:06
But it's clear to me today looking at something like MakerDAO that you know, if you, if you fast forward 20 years in the future, the way that countries are going to dollarize. Dollarization, for those who don't know, is when a country that has a weak local currency basically decides to adopt a another currency from a country that has stronger monetary policy. So, you know, for example, a place like Iran where they're, you know, they're facing a big currency crisis right now, they might well adopt from the bottoms up a different currency. They might adopt the US Dollar. That's the most common thing that tends to happen in a lot of countries in the future when countries dollarize I have a hard time believing that they're going to dollarize by importing, you know, hundreds of millions of tiny little green pieces of paper. I don't think that's what's going to happen in the year 2040. I think that the way that they dollarize will be using a digital dollar, and that digital dollar most likely, given the world that we know today, the biggest bet that I would make is that they will use something intermediated by crypto to get access to that digital US Dollar. And there's a good chance if another competitor doesn't rise in that time, that it might be something quite a lot like MakerDAO. So that's my time.
Richard:
17:19
Okay, great. It always makes the host's job easier when the two sides both come prepared with lots of arguments. So let's move on to round two. This will be a little bit more free-for-all, where each side will have the ability to respond to the other's previous arguments or bring up new arguments. But there will also be guided questions from me. So let me just maybe bring up the first question for Udi.
Richard:
17:40
Udi, feel free to respond directly to this question, or feel free to respond to the points that Haseeb has raised. So, some, and this includes my previous colleagues from the traditional finance world, think access to yield-y investments for Main Street is few and far in between, partially because of intermediaries rent-seeking the system. So DeFi is meant as a way to build an entirely new way of investment, totally divorced from the old world, that cuts out the middleman and provides decent returns to retail investors. Your thoughts on DeFi from this angle?
Udi:
18:15
Right. So yeah, I think I can try to combine my response to Haseeb and to answer this question hopefully together. So yeah, it's true. You can, you know, you can use DeFi apps right now to take, to take loans or to lend out money at rates that are probably favorable to, let's say what you can get at a commercial bank. And the thing is, is it really the only place you can do that? And I would argue that the answer is no because you can go to Bitfinex right now, which is a, a very large and very well known cryptocurrency exchange. And you can sign up. You don't have to provide any KYC data if you sign up. And even if you know, let's say some of these exchanges, even if you are not in the allowed list of countries, you can just install like a VPN or something and still sign up, right?
Udi:
19:15
They're not doing actual checks. So you can use those exchanges and lend out money through the exchange. So you can take a USDT or USDC or sometimes even DAI, and lend them out and, actually, seems to be that you consistently would get better rates through these platforms. You can also find much more volume there. So it's going to be much easier to move large sums of money there. And so, you know, you might ask, well, but, but there's a middleman there and we want something that doesn't have the middleman. I would argue, you know, it's just one middleman and there are a lot of competitors, you know, can replace. But most importantly, it's all transparent, so you can see, you know, they're not hiding anything. You can see all of the loans, you can see what's on offer, you can see the order books, you can see whatever you want.
Udi:
20:02
You can decide, you can see how orders were matched in the past. So, you know, it's pretty open. And comparing that to something like Maker. So, you know, Haseeb said you can do loans and you can do leverage and you can do synthetics in a way that is permissionless on DeFi. And again, coming back to my opening statement, I'm not really sure it's permissionless. It seems pretty permission because again, you have these, these these companies or these foundations that stand in the middle and they could, you know, they could decide to say, okay, we're gonna blacklist. Some of the participants, they probably don't want to do that, right? No one wants to get rid of customers, but they might be forced to do that. And the thing is they can comply because again, they have these controls that allow them to kind of change things and you know, do basically arbitrarily whatever they want.
Udi:
21:00
So it seems to me, again that it's not different from something like Bitfinex or something like BitMEX when you can create synthetics for example. And you seem to be able to get better rates there. So again, why not, you know. And, I'm, again, I'm giving these examples of, of Bitfinex and BitMEX, but there are, as we know, there are like dozens of exchanges. And they seem to have the same kind of properties where they're, you know, they are willing to accept every user they want them. They're trying to be very open and transparent about at least the way that the markets work. Now you know, there, of course there are downsides. Like, you have some exchanges that had major hacks, lost a lot of funds, but a lot of the times when that happens, they actually managed to make customers whole and just based on their profits and revenues.
Udi:
21:51
And the point is that the same can happen to something like MakerDAO or Compound, because again, they control keys that control the funds. So these funds could be stolen by someone that could be some hacker who somehow gets hold of the keys and take the funds. Usually what DeFi proponents would say at that point is that, well, this might change in the future, you know, that someday we get to a place where we don't need any control. But you know, I can say the same about BitMEX, right. So maybe BitMEX and Bitfinex in the future will not have control over the funds. But if we discuss what's happening right now they both seem very similar in, in, in, you know, in security, in censorship resistance. And actually the more that the non-DeFi options seem to offer better rates. So, you know, I asked myself, why not use them? Why would anyone prefer the DeFi apps over the exchanges that are willing to accept them.
Richard:
22:49
Great. Haseeb - Would you like to respond?
Haseeb:
22:51
Yeah. so you made a few points there that I'd like to, I'd like to go through. So you mentioned, first off, that Bitfinex and other centralized exchanges also offer lending services and oftentimes their rates are even competitive. So the first thing that I would say is you're absolutely right. And it's very clear actually, the sort of existence proof that people want this and that's very valuable is the fact that all of these unregulated exchanges offer this stuff and people love it. People use it very widely. I don't think that's any evidence against DeFi being valuable any less than it would be evidenced that showing that commercial banks also offer the service means that, you know, why use DeFi, commercial banks offer this. I completely agree. Of course commercial banks offer this it's sort of like the argument against, you know like saying, Like, well, okay, central bank is offering interest rates. Why do I care that my commercial bank offers me even better interest rates? Yes, that's true. That's the whole point of having, you know, once you have something people want, a lot of people will compete to offer it.
Haseeb:
23:48
The point of having DeFi is that it is actually a genuinely permissionless substrate on which anybody, regardless of whether they can KYC, regardless of whether they're backlisted, regardless of whether they're from a country that is on the OFAC list. Or you know, alternatively, just regardless of whether they have even know about Bitfinex or whether Bitfinex is willing to take them on as a customer. So the second part of that is you're arguing that non-DeFi offers better rates on many of these products. And I agree with that and that's the sign of a competitive, healthy market. I don't see why any argument about DeFi being valuable or powerful, would depend on the claim that DeFi offers the best rates for anything. It's again, sort of like the, you know, there will always be higher rates for any lending product provided that you move up risk, right? Risk is the ultimate lever that that should result in higher return in any lending market. So if I lend out to like some, some crazy degenerate, I'm going to charge very high rates, that I can get better rates there that can get somewhere else. So claiming that there are higher rates elsewhere doesn't make any, it doesn't really carry the argument of, okay, well clearly you should prefer to lend on here as opposed to lending through a base platform [inaudible].
Haseeb:
25:12
And of course you can't get rates for everything, right? So like the rates for lending ETH on DeFi are really, really low right now, because there's an oversupply - there's just a huge overhang of ETH being supplied these platforms essentially as collateral. And so you'll get better rates lending out your ETH pretty much anywhere else in a centralized exchange. However, you can't really lend DAI, right? So if you want a US Dollar denominated savings account you know, you're not gonna get that anywhere else. I mean, one, it would just be kind of radioactive for any centralized exchange to be offering that kind of service. Just because I had imagined the regulatory perception of that would just be so negative that they wouldn't want to touch it. If you want that, you can only really get that through DeFi today and there's no real counterpart for that in other places unless you want to do it in Tether or something else.
Haseeb:
26:02
And the second point that you made was that, well, you know are these things really permissionless? Are they really censorship resistant? Because MakerDAO has these admin hooks or you know, it has a governance and this governance could decide to like, you know, give all the money away or to steal all the money or, you know, any of these kind of sort of nightmare scenarios. These are all in principle possible. So, so two things that I want to say to that. So the first is that censorship resistance is not a binary. And I think Udi, you should, you should be sort of you should know that. And I think you do - that censorship resistance is not a yes or no question. It is a, it is depending on your threat model, and it is always parameterized by what kind of censorship are you resistant to, and what kind of censorship attacks could succeed, if coordinated.
Haseeb:
26:49
So for something like MakerDAO, if your attack model is that, I am worried that the MakerDAO team and their governance and all this stuff will collude to steal all the money. Yes. It is not resistant to that, that will, that will kill you. In the same way in Bitcoin you know, if my censorship model is that, you know, a miner with 34% of the hash rate will not selfish mine, or do a feather-forking attack or whatever that, that is not covered by the Bitcoin censorship resistance model either. Right? All these things are parameters that you as a user choose. What censorship do I care about? What am I actually worried about and how much of it do I want? And I think for most people using DAI today, and you'd imagine people in the future using DAI they're not that worried about what if the MakerDAO team decides to run me over, because I dunno, the MakerDAO team doesn't really seem like they want to run people over, there doesn't seem like they have any interest in doing that.
Haseeb:
27:43
It seems like predominantly their risk model is, I am worried about my government taking my money or I am worried about a local warlord or a gangster or, you know a some kind of corrupt local official taking my money. That's what I'm worried about. I'm worried about my bank getting basically going under. That's what I'm worried about. In that threat model, yes, DAI is massively more censorship resistant than anything else they have. And I would argue, so is USDC and so is Tether, but I don't think those things are going to last as being censorship resistant. Whereas DAI at least has a plausible story about how it will decentralize further. And of course the fact is DAI has only been around for about a year and a half, slightly longer than that. And this thing is, is very new. It's an experiment. There's a lot of work that is yet to be done on this thing.
Haseeb:
28:31
So again, I'd sort of look at it as being analogous to Bitcoin in year one, or Ethereum in year one. It is quite centralized today relative to where it could be. And that's by design. It should be that way. It should be the case that all of the people who are responsible for running this thing can get together on an IRC channel and talk about what to do and initiate a hard fork if necessary. This is early stage technology and I don't recommend that anybody actually put their life savings into DAI today, because it is new and it's novel tech. But the point is, I think this is an architecture that shows us how we get there. So I'm not making the argument today that MakerDAO is, you know, the future of the dollar and that's what we're all, you know, it's already here. You know, we just have to, we just have to get on it. I am saying that this shows us a glimpse of what is going to happen in the future and that is the argument that I make today in favor of DeFi.
Richard:
29:22
Okay, great. Udi - you can respond directly to it.
Udi:
29:28
Right, yeah. Okay, great. So yeah, I, you know, I definitely agree with a lot of your points. I see. I think that, you know, let's, let's start with you know, you said that the existence of, of current centralized changes and the fact that they have, you know, competitive rates doesn't mean that DeFi isn't valuable. It's the other way around. It kind of validates the market. And of course I agree with that. But then you said we should prefer DeFi because it's genuinely permissionless, and you know, like we mentioned before, I really don't see how it's more permissionless than something like Bitfinex. So the thing is, you know, I would argue that the reason that MakerDAO didn't attract regulatory oversight, or at least not too much, is probably because it's just too small right now.
Udi:
30:17
When Bitfinex was the size of MakerDAO, no regulator who cared about it, either it was too small, it didn't, you know, we didn't see any pushback. Pushback started much later when it was, you know, already massive. So I don't see anything in MakerDAO. It makes it more resistant, this kind of oversight, and this kind of pushback that we might see from regulators. So, but I would say, you know, to your point, it's still possibly OK, right. just because, okay, so we have this market and we have products like MakerDAO and like Bitfinex, both are somewhat exposed to regulatory oversight. That's fine. And sure you could argue why not use MakerDAO - if they're both the same, why not? And I agree. You know, I just think they don't have any special advantage.
Udi:
31:10
It seems like people believe that they have any, some unique advantage over traditional exchanges and I just don't, I just fail to see that, that advantage. And it does have some disadvantages though, you know, for example -- at least currently, blockchains are slow. We all know that, and we probably shouldn't get into a scaling debate. Now what, you know, at least right now it seems to make the product inferior, maybe it will change in the future, but that, that seems to be the case right now. And again, we don't get any advantages.
Udi:
31:43
Another disadvantage is that you see, you kind of broadcast your entire transactional activity to the entire world. When you use something like MakerDAO or Compound while on Bitfinex, you only broadcast the important things that people need to know. Like someone took an order, but most people don't know who took the order. So that's better. Of course, Bitfinex would know and Bitfinex have this large data chest now, which isn't optimal, but it's, in my opinion, it's better than everyone having this data. So it seems to be, you know, it seems that these DeFi apps have like strict disadvantages over the traditional apps and I fail to see the real advantage.
Udi:
32:26
You also mentioned, you know, you can have a dollar-denominated lending account. That's, that's unfortunately, that's not true. So you know, you can have on Bitfinex USDT and lend them out, you will get better rates than you get currently on most of these apps. When you lend out to DAI you can go to BitMEX and create a synthetic dollar by shorting BTC/USD or ETH/USD. And then you have basically synthetic dollars in your account and you will also get paid the funding rate of BitMEX, which is very similar to lending. And it's usually even higher in percentages. So you know, you have to manage that, but it's definitely something you can do. You don't have to use DeFi for that. There are also, you know, there are companies like I guess Celsius and BlockFi, others that do lending, although they do they do demand like KYC checks so they're not exactly the same. So yeah, you have multiple ways to, to do a dollar denominated saving account.
Udi:
33:29
The last thing was, you know, the censorship resistance isn't binary and I fully agree with you. You know, I think it's completely right. There are levels to this and there's also different threats that you are going to want to protect yourself from. And, I definitely agree that there is no reason for anyone to expect that MakerDAO or the other stakeholders in MakerDAO would steal someone's funds. I'm pretty sure no one has any interest in doing that. But one, we have to remember that there's the option of someone compromising one of the keys that control the funds. You know, it's not necessarily MakerDAO wanting to do it. It might be someone gaining access to those keys and doing it for them, which is basically what you're scared of when you have funds deposited at an exchange. I mean, I don't expect Bitfinex to steal my funds either, but I think that it's possible that someone would hack Bitfinex, like happened in the past. So that's a very real risk that, you know, that applies to both types of apps.
Udi:
34:36
And you mentioned, you know, you mentioned that you might want to, you know, kind of make sure that, you know, the government where you live isn't going to confiscate your funds, that some warlord isn't going to come in and confiscate your funds. And I think you get the same with Bitfinex, with BitMEX and so on. Right. So it's, again, it seems to be very similar. Haseeb mentioned that he's concerned about regulatory oversight with something like USDT or USDC or so on and, or definitely for Bitfinex and BitMEX and the other exchanges as well. And yeah, I fully agree. I worry about that too. It could happen to any of them. But the thing is it could happen to MakerDAO and Compound as well if they become large enough. The thing is that they do have the capability to shut things down. And because they do, it's definitely possible that some agency will require them to shut down, or add KYC checks, or both or you know, some, some form of that. So I see, I think it could happen to both equally. The thing is that the DeFi apps are just smaller right now. So it doesn't happen there because they don't attract attention yet. But if they become larger it seems like, you know, they probably will attract the attention of this kind.
Haseeb:
35:49
So let me respond to a few of those points. Yeah, so you raise, a few there that I want to go through in order. So the first is the claim that Bitfinex is just as permissionless as MakerDAO. I don't think that's correct. And I imagine most of our listeners will not think that that's a fair statement to make. It may be true that Bitfinex does not necessarily require KYC in some jurisdictions. And it may well be true that Bitfinex has API's. But I don't think those two things are sufficient to claim that Bitfinex is just as permissionless as MakerDAO. There are levels to permissionless and maybe there's a disagreement in how we define whether something is permissionless, but the idea that anybody in the world, anywhere without needing anyone to know who they are, or what they're up to, can create can basically do whatever they want on MakerDAO. And that's not true on Bitfinex. It's not true on any centralized platform that exists in any meaningful scale today.
Haseeb:
36:44
The second point you made was that DeFi has only strict disadvantages over other forms.
Richard:
36:52
Sorry Haseeb - Are you able to clarify what you mean by doing anything they want on a system?
Haseeb:
37:00
Yes, so I mean that you can, you basically can decide to take out a loan. You can decide to, you know, create a CDP, if I'm speaking of MakerDAO specifically, to create a CDP, to transfer your DAI to, to basically do whatever you want without any oversight, without anybody collecting information on you, any more than what's leaked directly through the blockchain. And you know, again, we talked about things like OFAC list and we talked about things like, you know, obviously all these exchanges have blacklists, that are independent from what are imposed by regulators, of people who they believe are hackers, or people who they believe are associated with some kind of criminal activity. All exchanges do this, all exchanges that are at any meaningful scale. And, you know, anything along these lines of just, you know, look, I want to do something and I don't want to ask for anybody's permission.
Haseeb:
37:48
I don't want to get held up and have my account frozen. I don't want to have a problem where I deposit some money and I'm going to do something and then I can't withdraw because you're not solvent. None of that stuff. Like Bitfinex had all these things happen. Literally Bitfinex has had massive, massive hacks occur to it. That's in large part where Tether came from. So the idea that Bitfinex is just as permissionless as a Defi product like MakerDAO, I just don't think that's correct. And I don't think anybody listening to that, who knows the history of these products, thinks that's correct. I think Bitfinex obviously is pretty far on the permissionless spectrum because that's the product they're offering. They want to be this kind of cowboy, wild West exchange. And that's great. And people love that and obviously there's a lot of demand for that in crypto today, but I don't think it is, I don't think it is fair to say that those are the same thing or that they are perfect substitutes for each other.
Haseeb:
38:37
The second point I want to make is the claim that DeFi only has strict disadvantages over other products. And, to this end, you claim that there are other ways to do dollars nominated savings account or, or to get dollars nominated yield. I agree with that. And I, and I granted that earlier. If you want to do that through Tether, that is doable and you're right - going long and short, you can get sort of a flat dollar exposure on a BitMEX. However, the claim that I was intrinsically making, there is obviously pretty much everything that you want today, there are alternatives outside of DeFi -- Other than synthetics, right? So we have seen something such as getting exposure to gold, getting exposure to the S&P 500 that you can do through DeFi products today. They don't have very high market share or very high demand today.
Haseeb:
39:27
But again, those are things that I think are very, very obviously things people want around the world. There's one thing that we know is tremendously valuable and has critical product market fit is access to US Equities and, you know, sort of mini-financial products around the world that people don't have access to, if they're not in the US. It's extremely, extremely hard to get access to investing in a US stock companies. But you can do that through DeFi. You can create synthetic assets that anybody in the world can get access to and can invest in. And today it's very hard to do. There's not a lot of liquidity that, you know, these things haven't really gotten jump-started yet. Most of the people on DeFi today are ETH-heads and very early adopters to these products, but there are things that you cannot get on other exchanges that you can get on DeFi, right?
Haseeb:
40:09
DeFi literally lets you remix any product. That's what I mean by permissionless innovation. You can build anything you want provided that you provide a price feed and that is very different than what we see today in the crypto exchanges that you see around the world. So but I, but I grant you that if you want higher leverage, you'll get it on, you'll get it better on on BitMEX and you're going to get it through DyDx. Right? That is absolutely true. And I expect that to continue because like you said, centralized exchanges are more efficient in just a strict scalability and economic sense. But that ratio, that Delta will be going down over time as the tech gets better and as scalability tech improves, it'll never go to zero, but it will, it will get smaller over time.
Haseeb:
40:52
And the other thing that you mentioned was the risk of compromise and you claimed that, well, you know, it's possible in principle for MakerDAO to get compromised, just like so it is for Bitfinex to get compromised, and you know, we should be equally concerned about both. I think just the, just the record of it tells us that the likelihood of compromise in something like a Bitfinex is probably much higher, if not an order of magnitude higher than it is for any of these DeFi protocols. Literally in their existence, we have not seen a major compromise of a DeFi protocol, which I don't mean to say that it'll never happen. I expect that it will. But DeFi protocols for the most part have been fairly secure and have had a pretty good track record. And the hacks of centralized exchanges, like they just, they just don't stop.
Haseeb:
41:41
And it just, it's just a sort of background radiation that we experienced that at centralized exchanges, whether it's insiders, whether it's you know, external people attacking them, their attack surfaces are just much larger and they're just much harder to secure than DeFi products. DeFi products - just by definition - we know enough now to know how to make these things relatively safe. There are enough best practices that people have learned of just, you know, always allow withdrawals, always allow people to be able to globally freeze the protocol, have a strict sort of alarm bells if things are getting withdrawn too quickly or that, you know, there are certain checks that don't pass. So I grant you that I expect there will be hacks on both sides of the table. But I don't think it's fair to say that the risk of compromising DeFi is just as high as it is in centralized exchanges. I don't think the history record has attested to that. Even while these protocols have been their least battle-tested that they've ever been meaning in their first year of operation.
Udi:
42:36
Okay. I guess I'll jump in and kind of address those. So, well, I mean it's definitely true that I'll try to just do it by order. So it's definitely true that like centralized exchanges such as Bitfinex and others have these black lists and they would, you know, they would reject some users. They would reject some countries and so on. That's definitely true. But the thing is that, and you can see it, you know, if you go to like people who are actually using those exchanges a lot of them, if not most of them will tell you that they're using various VPN services. Sometimes they even use mixing services to make sure that their coins don't have any residue of past actions. And they manage to use these exchanges. I don't want to, you know, I'm not going to name any names, but they definitely managed to, to use these exchanges. So it's, it seems like it's,
Haseeb:
43:32
Well, Udi - you saw the wasabi story with Binance, didn't you?
Udi:
43:36
Yeah. So that was Binance Singapore, which is a regulated exchange. It's not the same as Binance Global - Binance Global never did that as far as we know. Binance Singapore is highly regulated in Singapore, so yeah, you definitely can't expect that from, from this kind of an exchange, but a lot of other non-regulated exchanges actually or, you know, I shouldn't say non-regulated but, but a lot of the exchanges that are regulated in other territories don't tend to do that at least so far now. It's definitely possible this will change. I agree with you that it's possible that, you know, may be all of them add KYC checks in the next year. I think that if that happens, two things will happen. One, we'll see a lot of smaller exchanges pop up and compete and try to grab all of the people that are, that are abandoned and it will be risky for them, but I think they'd be happy to take the risk. So that's one thing that's going to happen if, if we see this kind of clamp-down, and because it's always possible because there's no connection to the kindest current our financial system, you just need to accept cryptocurrencies. Anyone can do it right. You don't need permission to, to start a an exchange. So you are likely to see new ones pop up. That's one thing.
Udi:
44:48
The second thing that will probably happen is that the same regulators will go to MakerDAO and say, okay guys, please add KYC checks as well. Because again, they can, they have the possibility to do that. They can just replace the smart contract completely to one that is white listed and only whitelisted addresses are allowed to participate in. They can do that. So it seems like, again, the permissionless-ness is, is really a, a function of just not being big enough right now to attract attention. Which is great - We need options like that. Right. It's great that we have them. But I don't think it's unique to the structure or unique to the architecture that they use. You also mentioned investing in US stocks. It really sounds like a great use case. Of course there are, I was trying to look for them. I don't remember the actual names, but of course there are centralized exchanges that will allow you to trade synthetic US stocks when you deposit Bitcoin or Ether. They're centralized. Again, obviously they don't do KYC checks because they aren't regulated anyways. And you have the same risk as there of course, but you'll have that risk with a synthetic DeFi app as well.
Udi:
46:12
And I guess, there's the issue of the likelihood of, of Bitfinex and those centralized exchanges to become compromised. And you know, it's true that in the past we had a lot of compromises to those centralized exchanges. I think that again, one of the reasons is because they were really big and other reason was because there used to be a lot of bad practices that have kind of seemed to improve. But I guess, you know, I just don't see why a DeFi app would be less vulnerable to that. So let's say, you know, MakerDAO right now, the foundation holds a key, holds a private key that can change the contract, right. And the same for, well as same for a lot of the other DeFi apps. So..
Haseeb:
46:53
To be clear, a multisig, not a single key.
Udi:
46:57
Well, yeah, probably, yeah, probably. But you know, essentially these exchanges probably use multisig too.
Haseeb:
47:02
Yeah. Of course.
Udi:
47:03
Yeah. So, but they have this ability to do that. And, the thing is, it just doesn't seem like this is easier for DeFi apps to protect against. Now, I'm not talking about bugs in the code, those happen as well. So for example, Bancor which was it's pretty similar to Uniswap, actually. I know that you know it well, Haseeb, but maybe the listeners don't know it as well. Bancor is a version very similar to Uniswap, that was earlier. I think they had a hack that pretty much emptied the entire funds in the, in the contract. It was about $15 million, if I remember correctly at a time. So, you know, it definitely happens.
Udi:
47:53
It could happen and I don't see why, you know, DeFi apps are smaller and they've existed for less time, but it definitely can happen. I don't see why they would be more secure than something like Bitfinex. The nice thing about the centralized exchanges is that they have large experienced teams that have, you know, they're battle- tested, that they, they, they faced a lot of, of mistakes in the past and learned from them. And they have, you know, best practices in place. Just like the best practices you mentioned for DeFi apps, they have the same best practices in place, but with really massively experienced teams who are protecting these funds. I'm not saying that there's never going to be any theft again. There probably will be, but I just don't see the advantage of DeFi over that. And the really great thing which is of course based on trust is that all of those large hacks you know, that happened to Bitfinex, to Poloniex, to Binance, and other exchanges that are still with us. They actually made customers whole. So it was a big mess. It was terrible. It was painful, but eventually they made customers whole and they could do that because they had profits that we could use to, to make customers whole. Of course some exchanges didn't do that. Some exchanges disappeared after being hacked. That's, that's possible too. It just, I would argue that that can happen to any DeFi app as well.
Haseeb:
49:18
So, okay. Last couple of points. And then I know that we have some questions that Richard wanted to ask us. I think I mostly agree with you on your points and I do think I want to emphasize, emphasize just for, just for the sake of the audience, that one to be clear, like there is not one person with one key that can change the contracts in MakerDAO - that's not how it works. There's a multisig and there's a governance process and all this stuff. And in the future you know, kind of kind of along the lines of something like Tezos, the goal for all of these platforms, Compound, MakerDAO, etc, is that while they're early, there's going to be a lot of development and the founding teams are very actively involved later on. It basically gets entirely delegated to governance. So that's not the case today. And I think it's correct for it not to be the case today because these things are early stage experimental technology and they need to innovate and move quickly. But once they figure out the right designs and the right architectures, then I think there can be more that's delegated to governance.
Haseeb:
50:16
But the bigger point that I wanted to make, and maybe this is sort of a good place to, to sort of just stick the landing, is that I think one place where we fundamentally disagree is what does it mean to be permissionless? You claimed earlier that Bitfinex is permissionless, because you don't need to KYC necessarily and you can set up an account and use their APIs, and they'll mostly just let you do whatever because they're just in it to make money and they don't care.
Haseeb:
50:41
And I agree with you, that's a certain degree of permissionless-ness and that's great. But it is not what I think is sufficient to categorize something as permissionless because permissionless-ness is not just about users. And yes, there are blacklists. And yes, at some point regulators may put pressure on these guys and they may, you know eventually get them to start also blacklisting, piracy technology or forcing them to do other things if they want to do business and not get sanctioned by different regulators. All that is obviously possible and on the table and should be part of your threat model when using centralized exchanges, but even more directly, one aspect to permissionless-ness that does not exist with centralized exchanges, is the permissionless-ness to entrepreneurs. The permissionless-ness to people building new things. I cannot go and build a new exchange on top of Bitfinex. Bitfinex won't let me do that. If I try it, they will block my API access, right? They will. They will blacklist me from doing that because Bitfinex does not want me to create new businesses. Bitfinex does not want me to compete with it. Bitfinex does not want me to create a new UI that has new products along with a Bitfinex product. I can't do that. In MakerDAO. I can create any business I want, right? I can do anything that I want on top of MakerDAO. I don't have to ask anybody for permission. Permissionless innovation is not just anybody can get access and it's a wild West. Permissionless-ness is the idea that when I have a new idea for a business or for a product that could be valuable to people, I can just go build it. I don't need to go around making sure that I'm not stepping on anybody else's toes.
Haseeb:
52:12
And that is exactly what is essential to what made the internet so powerful. It's the innovations that the people were able to build on top of it - is that random people from anyone in the world, in their dorm room were able to come up with an idea, launch it and see if people wanted it. And I can't do that on top of Bitfinex, but I can do that on top of MakerDAO. And that I claim over the long run is the much bigger difference that is going to resonate in the way that DeFi evolves.
Udi:
52:41
Right. yeah, I generally agree with the last point. I think that the thing is that with something like Bitfinex, you know, or any centralized exchange we can see, I mean there is an existence of some variety of apps that are built on top of Bitfinex. Obviously those exist - people use them both like consumer apps and both apps that they use internally. Those exist. I would agree with you that it's possible that Bitfinex would find one of those apps as you know, as a competitor and would decide to shut it down because they can't, it's definitely possible. I don't know if it actually happened, but it's definitely possible that it would. But again, I would argue that the same is unfortunately possible with MakerDAO as well. So if, you know, if, and at this point in time, obviously MakerDAO would want multiple people to build on the platform, like Bitfinex for sure wanted people to build on their platform when it was younger.
Udi:
53:42
But you know, it's possible that if MakerDAO becomes larger and more popular and maybe they will see someone building on top of the platform and see that as a threat. And you know, shareholders or token holders in, in, in MakerDAO make the governance decision to block that. That's, that's definitely possible. You could argue that it wouldn't be in their best interest and I would generally agree with that. But the same could be said on Bitfinex. Bitfinex has shareholders too. They can vote on things. Usually they let the CEO probably decide most of the things, but they could vote him out. So, you know, I think that that, I don't think it's really different. it's pretty similar. So yeah, that, that mostly, but I agree it could happen with Bitfinex for sure.
Richard:
54:26
Yeah, it feels like in that sense, there's a bit of experimentation in terms of governance model with DeFi, where you don't allow just one person to make the decision, right? Anyway, my next question is for Haseeb. And this is built upon some of the discussion earlier. So imagine that MakerDAO or one of these DeFi platforms get really big, and there has been an incident where a massive amount of investors, say retail investors, lost their money as a result of participating in that ecosystem, be it borrowing, borrowing to lever up and say putting their money in some kind of decentralized exchange, ultimately getting their money drained and so on. And the point I'm trying to get here is the involvement of regulators. They typically come in when their constituents have experienced massive problems, right? This is where a lot of the past regulations have been drafted, all in response to massive issues, be it securities fraud or just people losing money for all kinds of reasons.
Richard:
55:27
So can you play out for us - what the scenario would look like if a regulator comes in and says, MakerDAO, we need to shut it down. How can they shut it down? Can they shut it down? And how would the community respond as a result? And I think that would be a true test as censorship resistance of these DeFi platforms as a result. Because we all know how that could happen with something like Bitfinex if they're under the full jurisdiction of a regulator, and the regulator is dogged enough to try to shut them down. But I'm really curious to hear how something like this could happen to a DeFi platform when the regulator pays sufficient interest.
Haseeb:
56:04
Yeah. So I don't know that I can, I don't know that I'm necessarily that good at predicting one, exactly how a regulator would go about trying to shut down a system like MakerDAO. I mean, the most obvious thing they would do is just ask them, Hey, we think that what you're doing is breaking the law and we want you to shut this down. The nice thing about MakerDAO relative to pretty much almost all other DeFi platforms is that the team is already very decentralized, right? So, Rune, the CEO and founder of MakerDAO is you know, he's a citizen of the Netherlands, no, Denmark. And he spends most of his time in Chengdu in China. And so although I hope he's not there right now because of the coronavirus stuff, but otherwise, he, he's basically a citizen of the world.
Haseeb:
56:50
And if the U.S. government didn't like it, they wouldn't have a lot of direct ways to affect him. They have, I think some employees in the U.S. But not that many. And so there's, there's not a lot of obvious ways in which they are subject to U.S. Governance or to U.S. Regulators. That's part of the reason why I think MakerDAO being, you know, when we say regulators I think it's very easy to sort of be vague about it. But you have to actually ask yourself the question of which regulatory body is going to do what and what can, what actual hooks do they have? What actual repercussions can they, can they give someone? So it could be that they say, Hey, Rune, what you're doing we don't like, and we think that you're in violation of, you know, such and such international laws.
Haseeb:
57:29
So please, you know, don't come to the U.S. ever. And if you do, we will arrest you - then okay - then Rune doesn't come into the U.S. or let's say just Rune resigns. Rune resigns, and the rest of the MakerDAO team, which is fairly decentralized and again, people are mostly remote around the world. They say, okay, well we're, you know, at some point it's possible, you know, in the short term soon enough that there are enough sort of weaknesses in and there's enough centralization in maker now that if a regulator felt strongly enough to literally look, we are going to entreat the international cooperation of the other countries where these people are citizens and extradite them and then prosecute them. That is probably enough to get all these people to fold. Totally agree with that. But I want, I think that's very unlikely to happen because like that's just one, it's extremely hard, extremely slow for that level of international cooperation too.
Haseeb:
58:18
It's like very non-obvious what laws these people are breaking. And even if, even if it were, that would have to be argued and agreed upon by multiple regulatory bodies around the world. Third that also gives them a lot of time to just decentralize. Right. And so I would argue that even if today, all of the MakerDAO team basically folded and decided they were going to pack up shop and go home, it would be fairly straight forward to just freeze MakerDAO the way it is today. Decentralize the keys. Nobody knows who's controlling the keys, and the maker token holders do all the rest of the governance and then basically it just goes dark. They just go entirely Satoshi. So again it would be fairly quick to try to do that today. I don't think that they would be ready within three months. Within six months, probably, if they knew this was happening, this could be done.
Haseeb:
59:06
And if you push it out even farther than that, which I think is vastly more likely given how small MakerDAO is today. Then yeah, it would be, I think all of this would be fairly plausibly done. But the thing is, well to recognize is that MakerDAO is not actually much of a threat to U.S. Regulators, in large part because people in the U.S. don't use MakerDAO. Why would they? They already have dollars. They have no interest in getting access to a synthetic dollar. It's mostly people outside of the U.S. who would care about using something like MakerDAO. So for that reason as well, I think the U.S. government probably doesn't care that much about what's going on with MakerDAO, especially given that most of the people who use it, especially today are ETH-heads right there. There's sort of these very sophisticated technical users. They're not, you know, random mom and pop, grannies and grandmas on Main Street, who are getting defrauded by big bad investors. I think anybody using DeFi today has caveated them towards, or slapped on their face every time they try to use a product. So I think it's unlikely that in the short term they would be under that kind of regulatory pressure, but, you know, "never say never" is I guess how I'll end that claim.
Haseeb:
60:12
Okay, great. Udi, you want to add?
Udi:
60:13
Right. So, yeah, I'll try to jump in. So it's probably true that a lot of those teams are very decentralized. Maybe, you know, some of them may be even not exactly incorporated in any way. And you know, the founders and the employees could be all over the place. That's definitely true. Of course a traditional exchange can do the same. You know, Binance claims to, to be, to have a similar structure. Of course, if they are licensed in some jurisdictions, but you know, they, they jump between jurisdictions and they jump between physical offices quite a lot. There was, there was a lot of noise about that in the past. You know, they, they seem to be very efficient at it. They seem to be very well practiced to doing this.
Udi:
61:01
So, you know, obviously a DeFi app can do that. I would argue that the traditional app can do the same. And actually Haseeb has a background in poker for example. So an interesting thing that happened lately was Isai Scheinberg, who was the founder or co founder of PokerStars, which is in, was the biggest poker online poker site. He recently turned himself in to the, to the FBI or, or, well, I'm not sure to who exactly, but someone in the U.S. after, you know, he was indicted in 2011, so that was a decade. So, you know, these people, they can wait. They're fine with waiting. They're fine with taking their time. They don't forget apparently. And he's, he's like an old man. He's, he's over 70 years old and he turned himself in.
Udi:
61:56
You know, some people speculated that he'll get some light sentencing, but, but whatever the point is, you know, they keep at it and they, even if the person isn't necessarily U.S. Citizen, they keep at it. Although - I'm not sure - he might be U.S. citizen, I don't know. So you know, it could still happen if, if they're, you know, if they're motivated enough and I guess it's true, maybe they're not that motivated because again, maybe you know, maybe U.S. citizens don't even use the services as much. They definitely use poker sites, but maybe they don't use those, those DeFi apps as much. Again, maybe, but then, you know, they also don't use Binance as much, so why not, you know, why would they go after Binance and not after MakerDAO? It seems to be the same kind of, you know, the same kind of audience I guess.
Udi:
62:46
And the last thing is about MakerDAO being able to decentralizing in say six months. I will, I will say this, I used to be, you know, in 2015 I used to be very bullish on Ethereum, you know, on, on the potential ecosystem that Ethereum would have, I remember you know, I remember a very early blog post by Vitalik where he kind of described a stablecoin that will act pretty much the same way as DAI acts today. And he mentioned back then two problems, which if I remember correctly, you know, he said those are, you know, solvable problems. One is the Oracle problem and the other is that someone must be able to upgrade the contract. And you know, it's been five years since then and we don't seem to have a solution to both of these problems.
Udi:
63:39
Now a lot of people are researching, a lot of people are looking into solutions. I can't say about anything that it won't be solved; but it seems like, you know, for five years we've known this is a challenge, and there doesn't seem to be a solution right now to how to do that. In a way that is not, that is, you know, completely decentralized or, or more decentralized than, than what it is right now. So I find it highly unlikely that they'll be able to just solve it in six months. If they had to, it's possible. Sure. But, but you know, I find it unlikely.
Richard:
64:10
Okay, great. So we could have this ping-pong going on for a while, so I prefer that we move on to some other questions maybe to address a different aspect of the debate. So I'm going to throw out the last one here before moving on to audience questions. What is the social impact of DeFi beyond funding speculation? So maybe this question is more fit for Haseeb and that Udi can jump in.
Haseeb:
64:34
Hmm. What is the social impact of DeFi?
Richard:
64:38
So, yeah. So what net societal benefits would you say DeFi is bringing beyond enabling some kind of zero-sum game? I suppose you could make the same argument with centralized finance, right? But how is DeFi adding to that?
Haseeb:
64:53
Right. So I think I can give a very simple illustration of just one case in which it kind of captures both the idea of permissionless innovation and financial access. Right. So let's say that you are, I'm trying to build some kind of FinTech product to give, you know, we talk a lot about bank, the unbanked in crypto and I think it's to the point where it's kind of lame to even continue to bring it up, but I'm going to bring it up anyway. You know, there, there are a lot of FinTech companies that are basically contending to do the same thing that many people in crypto claim to do. Banking the unbanked, give them a payment system, something that's more efficient than what they do today, which in many parts of the world is still cash, cash based. So as mobile phone penetration increases in a lot of the poorest parts of the world, people can start doing online banking or mobile banking.
Haseeb:
65:39
And problem of course with mobile banking is that it is, it is very, very, very financially regulated. And of course the regulations, you know, we think about regulations in first world countries where all of us live. Regulations are, are tough, but they're fairly reasonable. The worst parts of the world are even more regulated than the places where we live. And the regulations are much worse and they're much more rent-seeking and they're much more just random and expensive to comply with. And that's a big drain on anybody trying to be a financial entrepreneur. That's one of the reasons why it's so hard to bank the unbanked is, that where they live, it is actually so expensive and it's, there's so much regulatory capture in the financial system. The one thing that's so beautiful about something like MakerDAO is that it allows you to say, you know what, instead of trying to go through traditional payment rails of the banking system that's local to whatever country that we have to get that I have to pay bribes, I have to get local lobbyists and I have to like figure out ways to navigate the local financial system and not get my accounts frozen and not get all this stuff happen when I'm trying to do business, and you know rural India or someplace in Africa.
Haseeb:
66:43
Instead I can use DAI and, again, this is what it means to be permissionless - I can create a business, I can use DAI, I can have them transfer to each other. And through that I can again create a financial gateway to anybody in the world without having to apply for a license, without having to pay any bribes. And I can do that anywhere in the world. And that allows anybody to get access to products they wouldn't otherwise have access to, which is of course the majority of people around the world who, who don't have access to you know, first world country's financial system. So that I think is a good illustration of why DeFi like products. While today I completely grant, they are mostly used by speculators. You know, the UX's are pretty bad, the fees are fairly high, the complexity is fairly high, but all of this stuff, I think we are seeing the first inklings of what will be used to power financial products of the future that are a real serious competitor to using traditional payment rails.
Haseeb:
67:43
That is what these things offer. So I don't think the most interesting comparison to make between DeFi is centralized exchanges. Central exchanges are great for speculating and DeFi is great for speculating if you don't have access to centralized exchanges or you don't want to KYC. But I don't think that's primarily what they will be used for. And I don't think that's primarily the value proposition of DeFi in the future. That is today. 'cause it's the only thing in crypto that people do today is they speculate. But I don't think that will be true going forward.
Richard:
68:10
Okay, great. Do you have anything to add? Otherwise we move on to audience questions?
Udi:
68:16
Yeah, I'll just say that, I love that vision. I'm super supportive. I want to see exactly the same. You know, with those things about DeFi that I say, usually people think, Oh, you don't want this vision. Of course I do. I'm happy to be proved wrong and I'm happy for these things to succeed. I will say that I think that the innovation that enables this, so you know, MakerDAO is basically a way for you to take a cryptocurrency in as a, as as collateral and use that collateral to create a synthetic like a synthetic U.S. dollar. That's what we have here. And again, what I claim is you can do that with traditional architecture. You don't need smart contracts for that. You don't need any of those things.
Udi:
69:01
The innovation is the availability of something like Bitcoin. Bitcoin is the innovation because it can be collateral that you can use for things. And if you don't like Bitcoin, you prefer ether or any of the other cryptocurrencies. That's fine. But I believe the innovation is there in the way to, to use collateral in a way that no one else can control. And as long as you have this you know, business owners can create platforms to do those synthetics because they have collateral that they can accept and they can start those companies anywhere in the world. They can hide if they have to hide or they can decentralize if they have to decentralize, but they don't need to use smart contracts necessarily. They can use any traditional architecture to achieve the same thing basically, as long as they have some cryptocurrency that they can use as collateral. That's the big innovation here, I think.
Richard:
69:47
Okay, great. So let's move on to questions from audience members. So the first question is actually for both, this is from The Gryphon, @thisisindeed on Twitter: What would need to happen for you to concede that your fellow debater was right? Defining failure conditions lets the public assess how grounded their opinion are, as opposed to blind faith/hate. Whoever wants to answer this first. This is in reference to how you think the other person would be right and you would be wrong. What needs to materialize in the world for you to agree?
Haseeb:
70:20
I think I would concede my side of the argument if if I saw or were convinced that centralized exchanges really could offer the same properties in total that I claim that that DeFi's offering. And I think that there's a possible world where that's true, where exchanges just get so kind of crazy and so innovative in the way that they offer different kinds of products that they basically supersede all of the stuff that I'm talking about with respect to DAI, and synthetic assets and you know, synthetic U.S. Equities and so on. I can see a possible world where that happens.
Haseeb:
71:02
I think most likely U.S. regulators would just shut down anything. We were talking before about the PokerStars former CEO and I remember very well in 2010 there was an event in poker known as black Friday when the us of justice shut down a bunch of overseas poker sites, most of which were based on the BVI. They were basically doing what centralized exchanges are doing today. And almost every single one of them were shut down. All the money was confiscated. And they were basically, I mean, it just kind of shows the power of the U.S. regulatory authority that if they want something badly enough, they can make it happen pretty much anywhere in the world. And of course all this stuff, these poker sites taking money from from U.S. Customers, and been going on kind of underneath the table for almost like seven or eight years. And it took a while for the U.S. government to actually to actually get around to shutting everything down. the legal apparatus moves slowly. But I think if you look at things like Tether today, I think it's very clear that Tether's days are numbered and eventually the U.S. is going to shut it down because it depends on U.S. financial rails. So if you, if I could be shown that actually, you know, there is a system that provides the exact same stuff that is not dependent on U.S. Financial rails, which I'm not convinced that there is. Then I think I'd probably can see that. Okay. Maybe this stuff, I think the market need is clearly there, but it will get satisfied through other means than through DeFi.
Udi:
72:24
Right. yeah, I think, I think that's a great question. I really love this question. I would love to say something like you know, what if, if I see that, that a project becomes, you know, truly decentralized, then I would be able to concede. But, you know, there's probably not this, there's probably not a point of full decentralization or total decentralization. You know, there's always going to be something. And, probably the real question is, are we going to find a sweet-spot that is, that is better than what we have with the alternative. So I would say because it's hard to predict what that sweet-spot would be then I would say just probably if we see that there's more activity, more volume, more trading volume on those DeFi apps, which I'm not sure what again, what falls under DeFi and what not, but let's say, you know, those smart contract powered solutions. So if we see more volume there than on the centralized solutions, like Binance and Bitfinex and BitMEX and so on, or whoever replaces them in the future, then yeah, I think I would be able to concede and say that, okay, this seems like, you know, the market believes this is a better solution.
Haseeb:
73:39
Hold on, hold on. So you think that I would win this debate if literally DeFi has more volume than all the centralized exchanges?
Udi:
73:48
It seems like a good measure.
Haseeb:
73:49
That's the only way DeFi is not deficient? That seems like a pretty high bar.
Udi:
73:53
I mean, in what way it succeeds, if those centralized exchanges still exist and still operate and are still, you know, still serving the customers.
Haseeb:
74:04
Oh, I mean - you're describing DEX's, I'm not claiming that DEX's are why DeFi is going to be great.
Udi:
74:10
No, I mean, I mean, you know, let's say, okay, so let's say for lending, right? So you have lending on, on Compound and you have a lending on Bitfinex. So if we see more lending on Compound than on Bitfinex for a sustained period of time and that seems like, you know, a good measure to say Compound is better. If it's not better than then, that's kind of my point. I'm not saying that it can't exist. Obviously it exists and it can continue to exist. I'm just saying that it's inferior. At least currently. So, and I believe it will continue to be inferior. And I would say I, I'm willing to concede that it is not inferior, if we see that it's, you know, that it attracts more activity. I think that's the best thing you can do.
Haseeb:
74:55
Yeah, that's fair. I think I disagree with the way that you adjudicate that, but I, but I think it makes sense.
Richard:
75:02
I think Udi's point is, at the end of the day, if DeFi is meant to be as revolutionary as they should be, then in some ways should topple the status quo. And so somehow it needs to achieve the same order of magnitude in terms of trade volume if we're using DEX's, and lending volume for using lending services.
Udi:
75:19
Yeah, I mean Haseeb mentions things like you know in the future, countries would want to dollarize. They would maybe use MakerDAO or something like MakerDAO. So obviously that should exceed the current volumes we see right now. And if that happens or even smaller things that that happen, then I would be willing to concede. Before that I would just not, not know that I'm wrong, unfortunately. It's definitely possible that I'm wrong, but I just don't know it yet.
Richard:
75:48
Okay. Very cool. So next question, and this is actually for both of you as well. This is from, Oh man, @c477bfef..., it's a really long string. So anyway, the question is: is deFi better on more stable ecosystems like ETC or RSK (which is built on top of bitcoin)? By stable, I mean in relation to monetary policies and code updates/forks frequency. So I think that this question also hits at the heart of whether, the most popular chain to be building DeFi on right now, does it have sufficient stability measured by monetary policy and code updates/forks frequency, to justify it being the carrier for future DeFi applications. And this is for both of you, so feel free to jump in.
Haseeb:
76:39
So if I understand the question, it's asking, would DeFi be better if it were built on top of Bitcoin because Bitcoin has all these superior qualities? Is that essentially ...
Richard:
76:50
Ethereum Classic. Well, is it better to be built on a more stable chain? So something like ETC. Or RSK, etc.
Haseeb:
76:57
ETC? As in Ethereum Classic is more stable than Ethereum? Hasn't it been 51% attacked?
Richard:
77:10
So I think that this person is referring to the fact that ETC stays true to the ethos of decentralization - the whole reorg from the DAO attack. So I don't think he literally means let's do everything on ETC instead, but this person is arguing that you need a chain that's more stable in terms of monetary policy and code updates/forks frequency.
Haseeb:
77:30
Yeah. that's, I mean that's a, that's a fair point that obviously, you know, it's better. All the things equal that a chain that is used as infrastructure for financial technology not change very frequently. That said, you know I think the reality is that if you can't do this, you can't do this on layer one on Bitcoin. It just, you know, the script language is very constrained. So if you want to do it on Bitcoin, you'd do some kind of side chain. And I can imagine the Bitcoin side chains also would probably undergo forks and upgrades and so on because it's just new technology. So I don't think you're gonna get out of that, unfortunately. If you want something that's like fully stateful and complicated, you probably also want it to get better. And if you want it to get better, that means that you're going to have upgrades and you're going to have forks. So I think that's just, unfortunately, if you're doing smart contracts, you're doing general computation you're, you're also signing up for the cycle of the forks.
Udi:
78:22
Right, Yeah. I think we finally found a point that, that both me and Haseeb agree on. I mean, yeah, if you're going to do a DeFi app, I think it's almost obvious that it should be on Ethereum, and not anything else. You know, people say, Oh, you should use RSK or Liquid or whatever because it uses Bitcoin. I mean, I don't see what's the point. It's not, you know, it's not the main chain. It's a side chain that someone controls if you're going to do this, if you want to do this, it seems obvious that Ethereum is the easiest choice. Or Tron.
Haseeb:
78:55
Or Tron!
Udi:
78:55
It's pretty, yeah. I mean Tron is pretty similar to, to Ethereum. It's a fork of Ethereum just like ETC. But it's bigger. So you know it's faster too, it's scalable.
Haseeb:
79:05
Well, I mean, the obvious answer of why you want to build on Ethereum is because that's where all the other assets are, and a financial system wants to interact with other assets.
Richard:
79:12
Yeah. And this question is for Udi, and actually for myself. You know, it's funny you bring up Tron, but do you genuinely believe that DeFi, be it decentralized exchanges or lending services can be built on something like Tron, and that would ultimately exceed the theory in terms of traction?
Udi:
79:28
So I guess the thing is I, because I don't believe that this so-called decentralization actually exists on Ethereum not, not on Ethereum itself, not on Ethereum itself, but on, on, on those DeFi apps. So because I don't think that those DeFi apps are actually decentralized in any meaningful way, then I think they could just as well be built on top of Tron. And you know, you can go to the Tron foundation and ask for funding, and they will probably give it to you. So maybe it's a better, you know, better approach for, for an entrepreneur. But, you know, I just don't see the big difference between both of them. Tron is actually faster. Obviously it has very significant centralization points, but I'm not sure it matters in, you know, in that use case.
Richard:
80:20
Right. Right.
Udi:
80:21
For now, eventually it will matter, but it will matter for all of them, so.
Richard:
80:25
Okay, cool. Well, thank you very much. It's now time for a concluding remark. So this is your chance to sort of summarize what you have been saying, and perhaps acknowledge something you have learned through debating with your opponent. And also tell us where we'd find you.
Udi:
80:40
Yeah. Okay. So, so I'll start. So, so I guess, yeah, so I guess my, my points again are, you know, I love DeFi, I hope for, you know, whatever succeeds, I'm happy for it. You know, I want to see, you know, I think that the visions that DeFi people have and centralized exchange people have, are pretty similar and I'm happy for either of them to win. I think that's a win for everyone. And I, I, you know, I want this, I'm skeptical on DeFi having an advantage because I don't believe it actually has an advantage on, on self custody, on censorship resistance or on composability. But you know, maybe things will change and I'm happy to see that. I, you know, I was, I was happy to learn, like things like MakerDAO being very decentralized. And actually, I didn't know that, that the team is like all over the place, which is probably very good, and you know, kind of reassuring about its success. But you know, in the end I'm happy for whoever wins this. This is, you know, I think, you know, I think obviously all everyone should be happy with this.
Udi:
81:47
So, Oh, and where we can find me, that's probably, I can't spell my name. It's too long, but, but I'm on Twitter and also there's the the pro-Ethereum alerts channel on telegram, which I recommend everyone joins. It's very pro-Ethereum. So if you're into Ethereum, you should join the channel.
Haseeb:
82:07
You run a pro-Ethereum channel?
Udi:
82:09
Yup. It's called, not pro-Ethereum, pro-Ethereum alerts. That's the name of the channel. It has just, just has great alerts.
Haseeb:
82:17
That's surprising.
Udi:
82:17
Yeah. I'm pro-Ethereum guy.
Haseeb:
82:19
Okay. Okay. I learned, so I definitely learned something new in our interaction today, that you're pro-Ethereum.
Richard:
82:27
You want to spell the link for us, for your telegram?
Udi:
82:30
The link for the channel. So it's, I believe it's https://t.me/ProEthereumAlerts, it's just no spaces, no dashes.
Haseeb:
82:40
Great. so I guess to wrap up on my side I think you guys, I guess the audience must at this point have gotten a gist of what I believe. And I just think that DeFi is a, is a fundamental phase transition in the way that in the degree of permissionless-ness that people will have in the future. And I think it's a sign of what's to come. I think all of this stuff is very new. It's very nascent and it's still super janky. And I expect that to continue for at least a couple of years before things start to really shore up and get better. So as this, as a, the base layers get more scalable, as layer two start to get built out, I think we're going to see more and more sophistication and better competitiveness with centralized exchanges.
Haseeb:
83:25
But today it's going to be worse. And the only way that DeFi is really going to get better is purely in ways where it actually does offer something that centralized exchanges can't. And so I think where Udi and I most differ is just in, in one believing that that delta is going to increase. They're gonna be more and more things that DeFi can do that centralized exchanges can't. And second, there are some things today which is proven by the fact that people are using DeFi. I don't think the people are using DeFi today are doing it just because they think it's cool and because they're ideologues. I think it's because they're doing things that they would not be able to do or would not be allowed to do on centralized exchanges. So as for I think Udi made some great points and I think he made a very compelling case that a centralized exchanges do today offer many of the same things that DeFi does in the future. And it makes a compelling case that they may continue to do that and they may even out-compete DeFi in even in, in, in many of the things that I laid out. I don't think they will. But I think we have to take the possibility seriously that centralized companies/centralized exchanges will basically sort of lead the vanguard of where I claim that DeFi is going and the new grounds that, that it will lay out. So that's it for me.
Haseeb:
84:39
Where you can find me online. I'm on Twitter at @Hosseeb. I'd also recommend if you, if you like the cut of my jib, you can check out some of our writing at Dragonfly at https://dcp.capital. And we do a lot of research and post a lot of stuff there.
Richard:
84:55
Perfect. And @hosseeb, H O S S E E B, correct?
Haseeb:
85:00
Correct. Yeah. Or just Google my name and you can find my Twitter handle.
Richard:
85:02
Perfect. Perfect. Okay. Well, thank you both very much, Udi and Haseeb. It's been an absolute pleasure to have you come on the show and debate. I have been personally looking forward to this and you do not disappoint. Listeners, we would love to hear from you and have you join the debate via Twitter. We're @blockdebate, definitely vote in the post-debate poll, which will be released shortly, and feel free to leave your comments. We look forward to seeing you in future episodes of the blockchain debate podcast. Consensus optional, proof of thought required.
Udi:
85:34
Thanks a lot.
Haseeb:
85:35
This is fun. Yeah, thanks. Thanks Udi. Thanks. Thanks Richard. This was, I expected this to be more adversarial, and it was actually very polite and friendly, so I liked that.
Udi:
85:44
Yeah, it's not actually me. I hired someone to do this debate.
Haseeb:
85:47
All right, that makes sense. That makes sense. All right. I wouldn't want, I wouldn't want your rep to be ruined by being too nice to someone.
Richard:
85:55
Okay, well it's been fun guys.
Haseeb:
85:56
Take it easy. Bye bye.
Udi:
85:59
Yeah, it's been great. Thank you.
Richard:
86:02
I would like to express my appreciation again to our debaters, Udi and Haseeb. We discussed the possibility of having them come back for round two when things develop further in the DeFi front, in terms of technology, market and regulations. I'm your host, Richard Yan, and my Twitter is @gentso09. Our show's Twitter is @blockdebate. Thanks for listening and see you at our next debate!
×

Listen to this podcast on