Guest User

July 1, 2022

Understanding Crypto

Understanding Crypto 5: Stephen Diehl: The Case Against Crypto

Guest User

July 1, 2022

Understanding Crypto

Stephen Diehl is a software engineer and writer based in London.

Welcome back to another limited series of Rational Reminder Podcast, focused on learning about cryptocurrencies. Our journey about cryptocurrencies has led us to speak to various experts on the subject, all of whom see some benefits to cryptocurrencies and the underlying blockchain technology. However, what does a skeptic think about cryptocurrencies and the benefits to the current financial system? In today’s episode, we speak to Stephen Diehl, a software engineer who works with financial technology within the finance sector and is an outspoken cryptocurrency skeptic. His engineering background, coupled with his experience working with financial technology, provides a unique perspective on the future of cryptocurrencies. We move through the episode learning about public blockchain technology, different consensus mechanisms, what potential problems blockchain technology can solve, whether crypto can improve the current financial system, if Bitcoin really is decentralized, what drives crypto prices, reasons why crypto will not work, what makes it similar to gambling, and more. Tune in today to hear a unique opposing view of cryptocurrencies and DeFi technology with expert and skeptic, Stephen Diehl!

Key Points From This Episode:

A brief breakdown of public blockchain technology. [0:03:28]
The current problems that public blockchain technology is trying to solve. [0:04:16]
Proof of work consensus and how it tries to eliminate the need for a trusted third party. [0:05:44]
Some of the downsides associated with the proof of work concept. [0:07:41]
How other consensus mechanisms have improved the proof of work concept. [0:09:21]
What the costs associated with proof of stake relative to proof of work are. [0:11:09]
Problems that both consensus methods have regarding recentralization. [0:12:07]
What other problems blockchain technologies can be used to solve. [0:12:50]
The problems in the financial technological system that public blockchains solve. [0:14:29]
Why finality of payments associated with cryptocurrencies is not a good thing. [0:15:42]
What limitations can blockchain technology remove regarding international money transfers. [0:17:06]
How to prevent double-spending under the current financial system. [0:20:34]
What Stephen thinks drives the value of cryptocurrencies. [0:21:15]
Whether Bitcoin is decentralized in Stephen’s opinion. [0:23:19]
Reasons why concentrated mining power does not cultivate decentralization. [0:24:46]
How permissioned blockchains can improve on the pitfalls of public blockchains. [0:25:40]
A discussion about the potential benefits of private blockchains. [0:27:15]
We learn what a smart contract is. [0:29:49]
Outline of other useful applications for smart contracts. [0:31:25]
Examples of illicit activities associated with cryptocurrencies. [0:32:08]
Code is law: deferring to code for implementing law. [0:33:17]
What Stephen thinks is the value of the underlying blockchain technology. [0:34:32]
Stephen explains what Web3 is and if it improves the financial system [0:37:05]
We find out if there is anything about crypto technologies that excite him. [0:41:06]
The most compelling argument for crypto that Stephen has heard. [0:43:08]
He explains what he means by suffering stemming from cryptocurrencies. [0:44:40]
Stephen shares his experiences as an outspoken crypto skeptic. [0:45:26]
How he began working with researchers from the London School of Economics. [0:47:03]
Discussion about the narrative of cryptocurrencies and why Stephen is outspoken on the subject. [0:50:33]

Read the Transcript:

Stephen, to start, can you describe public blockchain technology?

So public blockchain technology is a technology that's rooted in what I would call a large financial reconfiguration project based on a great deal of myth making about the value of reinventing private money from first principles. So this includes technology such as Bitcoin, Ethereum. These are large networks that people have deployed with the aim of creating new forms of private money that can exist outside of the remits of nation-states and/or internet native money. That's the original aim of the project.

What the project has morphed into over the last 13 years has actually opened to some large amount of debate, but public blockchain technology is largely a project to reinvent money from first principles.

What problem are public blockchains trying to solve?

Primarily, the ideations of the original movement came out of a belief that the private sector should be the source of money, that the financial system itself had become too corrupt and calcified, and the events of the financial crisis gave us reason to think about new foundations for a new financial system that was detached from the legacy institutions of banks and central banks and nation-states.

So it's really much rooted in a very anti-state libertarian ideology. Since then, public blockchains have largely become, I wouldn't say they become, they started off as an inhibition to build a new payment rail system, but what's happened over the last I'd say 13 years is that largely the narrative around this as an asset has been changed from a payment system into creating some new speculative asset class, and the exact properties of those things are going to vary between the different types of products that people issue on these public blockchains, but the project now has become largely about creating new assets that people can invest in, and those assets have some rather strange properties that we can get into in later details, but that's the essence where they're at today.

Can you describe proof of work consensus and how it tries to eliminate the need for a trusted third party in transactions?

Yeah. So the first public blockchain project is called Bitcoin, and Bitcoin is an attempt to create a completely censorship resistance money that exists outside of the regulatory perimeter and outside of the remit of nation-states. It does this by applying a very clever trick, which basically, we create a lottery system that runs on computers across the world, that all race to solve mathematical puzzles, and the probability of you winning the lottery is proportional to the amount of compute time that you spend trying to solve the problem. So you could think of it like the more lottery tickets you buy, the more likely you are to win.

So proof of work basically does that on a very large scale using this clever trick to create this completely fair, basically, function that assigns a winner to one computer every 10 minutes. Since the process is basically completely random as to who wins, this means that no one party is privileged in their ability to confirm the transactions that go on to the payment network.

Once transactions are confirmed by the winner of this what's called a proof of work lottery, that creates basically a new financial history in that 10-minute window about the transactions are committed to this global ledger of transactions. So proof of work is an extremely controversial topic because by nature, it's wasteful by design. It's about basically expending large amounts of compute time and, therefore, energy and, therefore, largely fossil fuels to basically power large data centers to run this lottery system for the purpose of basically guaranteeing that no one party has control over the entire network.

That's really the clever trick that Bitcoin invented and people have modified variations of this algorithm, different projects throughout time. That's the essence of how it gets its very, very special property that makes it resistant to temper from interference from large actors.

Are there any other downsides to proof of work?

It's extremely slow. So if you look at like traditional payment rails that we have, like CPA, like ACH, and like Visa, these are all run by central parties. They're run by a company like Visa and they batch up large transactions from payment terminals and they commit them to some global source of truth on a much more frequent basis. Those can process hundreds of thousands of transactions a minute. We're very good at building really, really fast payment networks now, at least in most Western countries.

So Bitcoin tries to solve this in a very strange way because it wants to be censorship resistant. It wants to basically resist outside actors from having any kind of say about what transactions get committed. Whereas a Visa is a regulated entity that basically has to comply with laws and consumer demands and fraud detection and all that.

So the proof of work system is extremely slow. It can process about three to four transactions a second. So that's not at the scale of what most domestic payment rails can do in the world. So that fundamentally limits the scalability of the entire system to basically be at the level of enough transactions to run a small supermarket, but not to run in a nation-state.

That's just the trade off, right? I think Bitcoin proponents may not even see that as a downside. It's just like, "We've decided we want censorship resistance and, therefore, we have slower or fewer transactions over time."

It's a trade off, yeah, exactly.

Does proof of stake or any of the other consensus mechanisms that have developed improve on the problems that proof of work has?

So proof of stake is a very intelligent algorithm by which all of these networks issue their own native token, which you can think of as their flavor of private money. So in proof of stake algorithm, participants participate in a lottery, but instead of it being proportional to the amount of servers and compute time, you allocate toward winning the lottery. It's proportional to how many of the virtual asset you stake in the system, and then that basically runs same lottery system, and then whoever happens to win the lottery gets to be able to commit the block of transactions to the network.

So the problem with both proof of work and both proof of stake is that they both arise out of the economics of this structure is that they both end up having the property that in proof of work, basically, you have all these people that are running large data centers and they have large electricity bills and you can't pay your electricity bills in Bitcoin. So you're forced to basically cash out of the system to pay your electricity bills to maintain the system. So that means the entire system is basically in negative sum.

In proof of stake, you have the same kind of process by which these tokens that are locked up in the staking mechanism means that the economics of the entire thing are completely deflationary. So the amount of money available basically has to monotonically go down. So both of them have this baked in, fixed supply, deflationary economics, which is very different than your average national currency, which tends to be inflationary.

So part of the, I guess, appeal of things like Bitcoin and proof of stake coins is that they are deflationary assets. The supply will only go down over time. To its proponents, that's actually a feature, and to its critics, that's the flaw.

Do you give up anything else with proof of stake relative to proof of work?

Proof of stake doesn't have the environmental footprint, but it's basically a form of plutocracy, basically ruled by the wealthy, who literally get to influence the commitments of transactions to the network. So if you look at the ideological framework that the Bitcoin network was created out of, we want to replace the big banks or we want to replace payment processors or money services businesses.

So with the proof of stake network, you basically just replace one set of stakeholders for another set of stakeholders in a process that some people call recentralization, where you've not created something new, you've just recentralized like we had Visa before, now we have a bunch of large token holders that control the entire network. The decentralization narrative falls apart in that kind of situation.

I've heard that one. Is it worse in proof of stake than it is in proof of work? Because in both cases, you end up with we have a big concentration of minors in proof of work. Is there a reason proof of stake is worse from the recentralization perspective or are they similar?

Well, in the proof of stake, the rewards that you get for winning the lottery are more tokens. So by design, you're basically incentivizing people to stay tokens. At which point, then they can get more tokens for basically confirming and staking their transactions. So it results in consolidation by design. That's how the economics of it just work, basically stay tokens to get more tokens. This is only ever going to lead to, if you look at the genie coefficient of what these things are going to look like over time, it could only result in plutocracy, and I guess crypto-aristocracy, if you will.

As a software engineer, under what circumstances would you use a byzantine fault tolerant distributed system like a public blockchain to solve a problem.

Okay. So let's define some terms for the audience. A byzantine fault tolerant system is basically a network that's resistant to collusions of different parties to manipulate the system. So you want basically a consensus algorithm. It's a means of having everybody agree about a specific piece of data or database. So byzantine tolerance basically means that its resistance against collusion attacks on the network.

So this is a very niche technology. You'd only use this in environments in which the people who are participating in your network are not trusted. So you want to create this game theoretic mechanism by which there's no incentive to collude. That's a really niche thing that doesn't occur all of that much because normally in traditional financial services, financial parties, they engage in contracts with each other that limit their behavior and you have a governing law and says that if some party does something malicious, then there's a recourse in the court or something.

So it's very rare that we have completely malicious environments and people want to transact with each other, and it needs to use technology to solve the problem. Typically, we just use legal framework in the course and contracts.

So coming from somebody who's worked in market infrastructure and payment support, these don't occur that naturally. I can't think of a single use case where I would use something like a public blockchain because this is just not something that occurs in normal B2B software.

So you mentioned you work in financial technology. What problems with the current financial system do these public blockchain solve?

I don't think that the benefits of public blockchain technologies really amount to much. They become a speculative asset that people will trade, that appeals to a certain number of people because they like the asset class and they want to add to their portfolio for various reasons, but it's ultimately and fundamentally a negative sum asset class. So every single person that makes money off of it, ultimately corresponds to multiple people who lost money off of it. So it's not productive. It's negative sum game.

From an investment perspective, it's not really solving a problem. It's not really great at the payment rail system because it's too slow and doesn't scale, and there are better payment mechanisms. So as a speculative asset, it's completely uninteresting and it doesn't really provide payment infrastructure. So I think the value it's adding to the world is almost entirely elusory. It's becoming a speculative gambling product and that appeals to a certain class of people, but fundamentally, it's not a big value add to society or to my industry, really.

One of the defining features of cryptocurrencies is finality of payments. That's like if I give you a $20 bill, you take it and that payment is final, and you can recreate that to an extent with digital currencies, which is interesting, I guess. Is that considered a good thing in a payment system, though?

Not generally. You want transaction reversal in the case of fraud. You want to do mitigation in the case if there's some externality to the payment that needs to be reversed. You want to have trusted third party that basically mediates the transaction so that if something goes wrong with it, that you can reverse it, and there's some entity that you have legal recourse to in the case of either fraud or reversal.

I don't fundamentally see what creating a digital cash-like system really buys. I mean, cash serves a purpose in our society for doing very low volume transactions privately. For most transactions that you would do, you'd probably want to use a credit card and then have a trusted third party like Visa or Amex to mediate the whole thing because then you have somebody to go to when there's problems.

So I don't see much benefit to this digital cash as an acceptance, some very extreme externalities where it's like, "I need to send money to some dissident in some country where the government has capital controls or something," but these are so extreme to be almost completely irrelevant to most people in day-to-day lives.

That leads to my next question. So we've heard that cryptocurrencies are supposed to improve the efficiency of sending money internationally. What are the limitations on the international money transfers now in the current traditional banking system that it's looking to fix or improve on?

So I think this goes to a big misnomer about the technology is that somehow this is going to provide a means for doing faster international payments, and the limitations in international payments these days are not technical problems. The reason that money takes a while to transfer is usually due to compliance or regulatory overhead that the bank has to approve.

Ultimately, when the money actually does move, it basically just changes two numbers in two banks' databases, and that happens almost as fast as you can up in database. So it's not a technology problem to be solved. If you talk about the remittance industry, that's a big industry, those massive amounts of money that people send to family abroad. So say you're sitting on dollars in the United States and I'm sitting on pound Sterling here in Britain, it's unclear to me that if you want to send Sterling to me from dollars that introducing a third intermediate currency in this entire process actually has a value add. It's better just to swap dollars for Sterling rather than go from dollars to Bitcoin to Sterling because in the middle, you have this kind of hyper volatile speculative asset that you're using to the exchange.

Why not just go straight from dollars to Sterling? It doesn't add any value. Then if you're running that money services business for the remittance, all you have added is basically just the price risk of something that sits in the middle of that transaction for no reason. So the remittance use case for crypto doesn't seem to really match my lived experience of seeing how those things actually work in practice.

I guess what it does give you is because it can operate outside of the regulatory and legal system, all of the checks and balances normally required for sending money, which might cause delays or fees or whatever, you can bypass that, but you're bypassing it by bypassing the legal system, which maybe has other problems, and maybe that part doesn't last forever either.

I think the best example of what you're saying there is the Canadian convoy that's happening up in Canada. They solicited a lot of international donations from people abroad and they were able to get all of the Bitcoin pulled in one address, but then when they go to actually take that Bitcoin app into Canadian dollars, they find the government's basically blocked the final transaction. So they're stuck with a bunch of Bitcoin that they can't actually cash out and they can't actually spend it on petrol or on food or actually doing anything.

So if you want to send money outside of the regulatory perimeter to a country that's hostile, you still can't use the Bitcoin to actually buy anything because the government's going to restrict the withdrawal into the national currency at the endpoint. So fundamentally, all of these money movement and remittance problems that operate outside the regulatory perimeter, you still need to go from national currency or to national currency, and that's not really a means to bypass that because the state always has control over domestic money service businesses.

Yeah, that one was super interesting because I think that they also filed an injunction where separate from cashing out or leaving the crypto sphere and going into the regular government money, they made it so that even interacting with that wallet was punishable by prison based on the injunction that was filed. So it's like even if you find a way to spend it in Bitcoin, if you do that, you're breaking the law, and the law still matters to you because you still live in Canada.

Exactly.

Yeah. I thought that one was pretty interesting. How do you prevent double spending, which is one of the problems that proof of work consensus solved in a system without a consensus mechanism?

You have a trusted third party, creates a single source of truth about transactions. If there's any kind of problems, then you solve it with some dispute resolution process. So the double spend problem is only a thing if you want to basically create a distributed system of computers that need to agree about some data that can't be duplicated. This isn't a problem in centralized, trusted third party systems that run payment rails. It's a nonissue over there by design.

The blockchains and the cryptocurrencies that serve them have exploded clearly a number and value since Bitcoin was created. What do you think is the core innovation that's really driving this?

Well, as a skeptic, I think there's no innovation. I think it's similar to what we've seen throughout history. There's a lot of market medias, and the defining feature of market media is when people start trading things detached from their fundamental value based purely on the fact that they think they can offload it on a greater fool. That's just part of market structure. That's happened since the invention of capitalism, the early markets back in the 1700s in London and tulip manias and beanie babies.

If I look at the opinion of a lot of people that I respect in the industry, I think eight of the last Nobel prize winners in economics have described the crypto phenomenon as a speculative bubble. Warren Buffett, Charlie Munger, George Soros, they all say this thing looks like a bubble, and I'm inclined to agree. I can't quite figure out how to actually value these things as an asset, and it seems to be a lot of speculative hype that's detached from underlying realities. So my thesis is that, largely, it's a market bubble.

It does have bubble-like characteristic. The interesting thing about it is it created a mechanism where financial assets could be created outside of the regulatory system. That is maybe not a net positive innovation, but it's still an innovation like, "Hey, we can mint these assets now that can't be regulated or at least that are resistant to regulation and censorship."

Yeah. I mean the purpose of public blockchains is to create asset bubbles. Now, every teenager can go in their basement and create some dog money from scratch and the whole thing can get up to a very large size where it can draft billions of dollars in investment. That's a very strange phenomenon. I guess it is an innovation of sorts, but it's a very frosty, bubbly bash of things that we don't know how to value that's popped out of nowhere and it's a very strange thing. We've never seen this before in markets.

So we've talked about public blockchains being distributed systems, and decentralization, of course, being one of the core attractions of Bitcoin since it was created. Practically speaking, is Bitcoin decentralized?

This goes to the fact that decentralize is a buzzword in a computer science perspective. Yes, the network is decentralized. It has a distributed topology. The code actually does function in a centralized manner. In practice, if you layer on the incentives that fall out of the network, it's that most of the mining that's actually done on Bitcoin is done in these large consolidated mining pools that are basically owned by three Chinese entrepreneurs, and they control basically 90% of the total mining capacity.

So they run these massive data centers because economies of scale means that it's not worth it for your average person to run their own Bitcoin mining. We're getting more where you have these massive data centers to consume vast amounts of power and position them very cheap sources of petrol. So that happens in places like Kazakhstan and Siberia, where these companies basically just run these things.

From an outside perspective, that means that you basically have three folks that are running a giant company that controls 90% of the mining. So that's not very decentralized. That's basically just this ambassador Russian money services business that basically processes most of the transactions, and that doesn't look very centralized to me. That looks very much like you've just recreated Visa but worse.

We've touched on bits and pieces of this, but I want to just key in for a second on the mining power. Can you just talk about what the miners are doing and why a concentration in mining power is a negative thing from the perspective of decentralization?

The concentration in mining power is from the fact that it behooves people to basically run massive amounts of these data centers, where all of the returns from winning the lottery get pulled into one entity where they can use those winnings from winning the block lottery to basically fuel the data center, which they can use purchasing power. So it's just a pure economies of scale argument. Why? It's just more efficient to consolidate all of your servers in one room and just pull the money around a common enterprise rather than having this massively centralized network of individuals doing this, and that's just a basic argument from economics. So that's what's happening is mining pools are massively centralized now.

Some companies are talking about using a variation of public blockchains run by a set of trusted parties. So do these permissioned blockchain systems improve on the shortcomings of these public blockchains?

So there are these things called permission blockchains and they exist, but they exist in a whole different ontological category of software because while the public blockchains are trying to reinvent money from first principles, these permission blockchains primarily exist to basically do enterprise data management problems, where you have a bunch of people that need to have a single source of truth about some set of data because they're business partners or they're trading partners. So they basically set up the system that acts like a quorum that basically collects a bunch of data and creates a single golden source of truth out of it.

Typically in the permission blockchains, there's no token, there's no speculation. It's basically just a way of reconciling data between a bunch of companies. So that's a very fundamentally different beast altogether. So permission blockchains don't have necessarily a great track record. They are a solution that some people have used, but they're controversial because to first approximation, they're basically the same thing as running like a Microsoft SQL server, which is like a regular database that you can buy from Microsoft today and just replicating that a bunch of parties that would also work as a solution. So the permissioned blockchain solution seems to be a solution in search of a problem in most cases.

That's interesting. So you could solve what a private blockchain is trying to solve using existing technology. Are there any other differences? Is a private blockchain more efficient or is it more reliable? Anything like that?

Typically, it's worse. It's actually slower and harder to use. So there's not a lot of success stories around permissioned blockchains. IBM used to have a big division that used to do this stuff. They mostly spun it down. Oracle still serves a few of these prototypes that they've stood up, but to first approximation, it's just strictly worse than using traditional solutions, and there's not a lot of success stories about permissioned blockchains being terribly useful.

I mean, I still see media releases from large companies bragging about their use or their intended use of permissioned blockchains. Why do you think that's happening?

Well, okay, let's talk about this media enterprise software a little bit. Companies that do these kind of things, typically it's innovation department inside of some corporate, and innovations inside of corporates do very strange things. Sometimes they basically bring in external parties to explore new and developing solutions and software and see if they can apply them to some aspect of the business to improve efficiency. They will do this for all sorts of bizarre, strange reasons, mostly to appear innovative to their shareholders.

So you'll see a lot of these things that are called press release where corporate will do a project on a blockchain, not for the purpose of adding value to the company, but for the purpose of doing the blockchain project.

So in Britain, there's this supermarket called Tesco. They sell food. They did a blockchain prototype where they would track produce that would come from the European Union on a blockchain, and they're replacing this very large old SAP system that they used to do it. So they did this blockchain prototype that ran for a month and they spun it down, but the purpose of doing it was to basically produce a press release to say that Tesco is using a blockchain, and that's really, I can't understand this. That's a lot of what you see in these press releases. They're not really very real. They're theater.

I heard the exact same example for Walmart, I think, where they could track the lettuce going to Walmart stores in North America.

Yeah. These projects don't really have a great track record. Maybe they're not as bad as the old thing they replaced or maybe they're semi-functional, but when we're talking about tracking lettuce for some back office process inside of Walmart's logistics chain, this is not a big transformative paradigm shift in technology. This is a way of tracking billables, those lading for lettuce. This is not going to transform the global financial system. This is basically just doing more back office data management. I think it's really important to not overstate that permissioned blockchains are not a big paradigm shift in our industry at all. They're just maybe a partial solution for some very small problems.

Let's shift a bit. Can you describe what a smart contract is?

So a smart contract is a unit of logic, which you can think of is code that lives on a blockchain that mediates the transfer of the native crypto asset of that blockchain between different counterparties that use the blockchain. The name itself is a bit problematic because smart contracts aren't either smart nor contracts. They're generally very dumb pieces of logic that have almost nothing to do with contracts as defined by common law.

So you can think of like, "I want to release some Bitcoin to you at some point in the future," say 10 days later. So you can encode that logic in a smart contract and say, "You should release these funds only after this date." So a lot of these things are really that simple.

Then there's more complicated ones that run on things like the Ethereum blockchain, which allow people to do things like create an equity crowdfunding structure in which basically a certain amount of the native token of the Ethereum blockchain, which is called Ether, is exchange for a positional value in the contract itself, which basically maps a percentage of ownership in the contract just like you would a share in a company, and that's primarily most of these things are being used for. They're used for crowdfunding solutions where you basically allocate a certain amount of cryptocurrency to a position in a common enterprise controlled by a bunch of other people.

In some sense, it's a way of doing regulatory arbitrage, basically creating a cap table for a corporate entity that exists outside of the regulatory perimeter. It can pool money in this slush fund that the smart contract oversees.

Are there any other useful applications for smart contracts?

The use cases that people would do are the ones where they want to transact in crypto assets. Like I mentioned before, crypto assets aren't only a great payment mechanism and they're purely speculative assets. So the use cases are limited to things that can only exist outside of the regulatory perimeter or things that are straight up illegal like Ponzi schemes or like gambling products that people want to do on the blockchain that have to exist outside the regulatory perimeter. I haven't seen a really compelling use case of smart contracts that was simultaneously compelling and legal.

Yeah. You've been watching this space pretty closely. Are there a lot of on the other side, a lot of the Ponzi schemes and, yeah, well, stuff that's maybe not so good for the world?

Yeah. So I encourage everybody to go to this crypto skeptic, her name is Molly White. She does this really great site. It's called web3isgoinggreat.com, and it basically enumerates the day-by-day catastrophes that happen in the crypto world. Every single day, one of these smart contracts or new schemes or high yield investment fraud programs basically just implodes, and this happens every other day now.

At the bottom of the page, there's this giant number and it's how much money has been lit on fire. No, mainly, this is notional value. It's basically pretend money that people are losing these schemes. It's some up to 15 billion equivalent to the US dollars in cryptocurrency. Really, just a lot of bad stuff going on. It's the Wild West out there where there's a lot of scams, a lot of fraud, and it looks like the 1920s again. I think we all know how that ended.

Yes, we do. From your perspective as a software engineer, and this relates back to the smart contracts, well, to crypto in general, this idea that code is law. You obviously live in software writing code. Does it make sense to you to defer to code as law?

So this is the term that people use to describe smart contracts, where if a smart contract does something and we all just agree that that's the law of the contract, then you really can't claim that there's any fraud or anything involved. It would be the equivalent of if I do something on a smart contract and spits out all of its funds to me, it'd be the equivalent of me walking up to somebody on the street and basically being like, "Mate, give me your wallet," and he just does, and just gives me his wallet. Is that really theft anymore? It's basically the equivalent of that button code.

So if the smart contract happens to deposit $25 million into my account, that's what the logic said it should do and, therefore, there's no crime or no fraud involved anymore. That doesn't really match with our conception because a programmer basically just made an error in the smart contract or that a hacker basically figures out an exploit in the logic, and that happens basically every day now, then the contract is just doing what it's supposed to do. This is not necessarily a great way to run a financial system because code and expectations don't always match.

That's really interesting. So one narrative out there is that while crypto might be in an asset bubble, the underlying blockchain technology truly is an innovation and will be here to stay. What do you think about that? Do you agree?

From my perspective working in the software industry, I don't see a lot of compelling use cases for the so-called underlying blockchain technology that would probably refer to things like permissioned blockchains, which to be frank, I have not seen an enormous amount of use cases for. You have things like the Walmart prototype, but it's not really a resounding success, and every single company I know that's been in the permissioned blockchain space is either folded or just done really bizarre consulting work building standup prototypes for making press releases.

So the underlying blockchain is really a solution in search of a problem. I don't see crypto assets being terribly useful because they suck as investments. They suck as a payment mechanism. They're basically just a way to spin up asset bubbles on 10 attending fundamentals or exposure to the broader market. So to me, the entire thing seems like a loosery in its value add to society.

The economy just made this interesting because an uncorrelated asset might not be such a bad thing, but I think Bitcoin's actually become increasingly correlated with the US stock market over the last years. It's got a very high beta with US stocks. So it's not even behaving like an uncorrelated asset, at least not at the moment.

Yeah. It's an open question about how does the price formation of Bitcoin actually come from because it's a pure greater fool asset, but it's not a store of value because the variance on the price is just too volatile. It doesn't act like gold. So it does not store value. It's not like anti-correlated with the NASDAQ. If anything, that seems to be basically correlated with the NASDAQ. When the COVID crisis hit, it basically felt like everything else in the market.

So what is this doing if you're an institutional investor? You're just adding beta to your portfolio for no reason. As an asset, I don't have a valuation model for this thing. The value of it is whatever the next fool will pay for it is. There's no fundamentals, there's no income. So to me, this does not seem like something I really want in my portfolio. I don't know why people would want it from a financial analyst perspective. It seems very risky and not a whole lot of value add or diversification for that matter.

I agree with you. One of the scary things that I've seen, believe it or not, is back tests including Bitcoin, but of course, Bitcoin has a relatively short and extremely positive, for the most part, price history that early on was uncorrelated the stocks. So it looks great in the back test and you can sell back tests.

That is true.

That's a little scary. We talked briefly about Molly White's website, Web3 Is Going Great. Can you describe what web3 is?

Oh, God. What is web3? I mean, nobody has a coherent definition that I've heard yet. If I were to back up and try to give the most broad definition possible, it's the broad notion that crypto assets, NFTs, DAOs, and smart contracts need to be integrated into existing web infrastructure to give rise to new types of businesses that can exist, that utilize blockchain technology.

Now, I fundamentally think blockchain technology is basically useless. So I think this entire thing is an empty buzzword, but that's what people in the industry refer to broadly, but it has no universal agreed upon definition in the software industry. It's considered a controversial buzzword because it doesn't seem to be tied to any. There's no company you can point to that's the Google of web3 or the Facebook of web3. It's a term that people use around without a whole lot of thought to what it actually means.

Does anything improve or get better with web3?

Well, the burden of proof is on them to build something that shows some improvement over existing technologies, but since the underlying technology, like I said, crypto assets are rubbish investments that have no fundamentals, they're not great to invest in. They don't serve as a payment rail. Blockchain technology doesn't really have an application. It seems like it's building on a foundation of sand and it seems to be a solution in search of a problem.

So it's not impossible that some broad umbrella of web3 could have some value, but I think the fundamental technology it's based on has yet to prove itself. I think there's strong technical and economic reasons why building on top of this foundation is problematic.

What are proponents of web3 saying will get better with web3?

If you go to the Wikipedia for web3, I think there's a lot of grand aspirations about either technical or financial reconfiguration that basically web3 is going to bring an end to the consolidation of the American tech in gemini, so the Googles, the Apples, the Facebooks of the world are going to be undercut by web3 companies that are going to do everything they do and better.

This is a really great story because we all recognize that some of these companies are just simply too powerful in the United States at the moment, and maybe they should be broken up or something. That's going to be a debate. So web3 has said, "We used the market to that instead of antitrust," and web3 is basically, "Oh, what if we could build a decentralized version of YouTube in which we pay out people not in dollars anymore, but in some of native crypto asset?"

The technical imaginaries of the entire project, I guess they imagine a world in which there's a hundred thousand different types of private money, one for every single site you would possibly visit. So there's a YouTube coin. There's a Facebook coin. There's the Uber coin. There's a Deliveroo coin, and everybody instead of transacting in pounds or dollars anymore, now you have to go through a gatekeeper in an exchange to buy the native token to basically purchase goods and services on the platform.

It's also the reverse operation that the payout structures can actually be that can pay out users inside of the native currency of the platform. So it's the fracturing of a coherent unified financial system into a hundred thousand different mini financial systems all controlled by different gatekeepers and different rent seekers. Some people think that's actually a utopian vision of the future. Other people think that's very dystopian vision of the future because it introduces a lot of exchange risk and rent seeking and artificial barriers to entry and transaction costs, but that's probably great if you're the one collecting the transactions, but probably not someone if you're paying them.

Yeah. I've heard, too, that the decentralization of platforms like YouTube, it gives power to the creators and all that stuff. That's the grand vision. So you've made your skeptical position pretty clear throughout our conversation, which has been great. As a technologist, have you been impressed with or excited by anything related to these technologies?

Some of the ideas, the technical ideas on a pure theoretical basis are quite clever. There are some interesting advances in cryptography and some computer science stuff that has occurred as a result of this stuff. Fundamentally, I think when you apply these technologies to actual real world problems, problems that businesses actually have, they tend not to really hold up because in most places like completely trustless environments, it simply don't exist in capitalism. High trust societies are a good thing. Basically, capitalism works best when there's low barriers to entry, when markets have lots of public information, when there's a unified payment platform that we can all agree upon like the dollar is a good thing. We don't want a hundred variants of the dollar. We did that back in the 1800s. It was called the Wildcat banking era, where every single bank would issue their own version of their own bank notes and everything.

It was a nightmare for most people. I don't think recreating the Wildcat banking era on top of the modern platform capitalism is a value add for either consumers or for capitalism itself because it introduces all this artificial friction to markets now. As somebody that wants to see payments and money flow and value being created in companies, I want to see more rent seeking for people that basically issue their own forms of private money, basically, just to do the businesses we're already doing.

So to me, this seems like it's a very regressive project that seems to be proposing that these new speculative assets simply should just exist for no reason other than people like to gamble on them. So the entire crypto project, I have a lot of problems with both from just this technologist perspective. From somebody that's worked on market infrastructure and somebody that's just interested in making capitalism work better, to me, crypto, the entire project, seems antithetical to all those goals. So I don't see much value.

You're pretty vocal about that. So I'm sure you've heard many counterarguments, maybe all the counterarguments. So I'm curious. What's the most compelling argument for crypto that you've heard?

So I think the most compelling one is that there are ... So I live in Great Britain, and the laws of this country I probably agree with 98% of them because it's basically just liberal parliamentary democracy, right? Most people in the world are not as blessed to live in stable democracies across the world. So there are countries that are authoritarian regimes that have very, very strict capital controls. There are people that genuinely need to flee countries because they're either dissidents or they're being hunted by authoritarians.

Perhaps a crypto asset could possibly be of use for somebody that needed to flee an authoritarian regime and then set up a new life somewhere. That might be a use case. That might have a very, very niche application for people that want to basically do the equivalent of going on an airplane with a barrel of gold bricks to basically just start a new life in a new country. That's probably the most compelling one I've heard, but if you factor in all of the externalities of this entire system existing, the value add to a very, very small pool of distance versus the externalities and the cost board by the rest of the world and society as a whole in terms of just the amount of illicit activity it enables all the terrorists in financing, all of the money laundering, all of the environmental wastes, they don't sum to a net positive on that.

So helping a few dissidents versus having this massively destructive program that seems to be creating massive amounts of suffering versus enabling a few other people to thrive, it just doesn't sum to be a net good for the world.

What do you mean by the suffering stemming from crypto?

Oh, because it's largely a vehicle for lots of fraud. So yes, it helps maybe a few 20, 30 distance, but a hundred thousand people lose all of their savings. So it's not sum to a net positive for the world or net negative. I already said net negative because it's just more suffering for more people versus wellbeing for a small number of people. It's a utilitarian calculus I'm doing here, but it's one that I think is substantiative.

You were, from what I could gather anyway, one of the earlier people speaking out skeptically about crypto and Bitcoin, at least writing about it. I saw one of your blog posts maybe a year ago and I think you'd started writing much earlier than that. What has the experience been like being one of the skeptics, one of the lightning rods for crypto skepticism? What's that been like?

There's been a few other people that were before me, actually, and the people that laid the foundations for ... because at some point, I was more on the edge about the thing, but then once I really dug into it and figured out what's actually going on, why the economics of it don't work from it being deflationary, and I sat down with a bunch of economists from the London School of Economics. We had this big discussion and I brought my knowledge about the tech and they brought their knowledge about the economics. We walked away from the evening being like, "Oh, this is really bad. This doesn't actually work. This will never work."

So I came into a long tradition of people that have already been speaking out about this from computer scientists to other activists, and the community of skeptics is actually very, very supportive, people like the Financial Times journalist. The Financial Times has been talking about the downsides of crypto for longer than I can even imagine. Their expertise on the subject has largely been much shaped my perspective and their skepticism.

So when I started speaking out about crypto, it was a conversation between five to 10 people basically thinking, "Is this good or not? No, it's not good," and then now, ever since I've been speaking out like others have been speaking out, it's a much larger conversation. I dare say within my industry it's now one of the most controversial topics. There's very much the pro and anti camp, and there's not a whole lot of in between. You either love it or you hate it. There's not a whole lot in between because the ideas appeal to a very certain set of political imaginaries about the world and markets in the state that tends to divide people along ideological lines.

Can you talk more about the genesis of the meeting with the economists at London School of Economics? Is that where it was?

Yeah, it was at LSC. Yeah. So I had a friend of mine who introduced me some folks in the econ department there. Then I had been thinking about Bitcoin and Ethereum, all this stuff, and seeing if there was actually any use case from a computer science perspective. My knowledge of monetary theory and just the whole central banking infrastructure and how that whole system works was very much incomplete. So I really sat down with some guys who were really deep in the monetary theory and central banking and stuff and we talked about what are the desirable properties of money, what are the desirable properties about inflation targeting, and how does central bank work and everything.

So it filled my gaps in my knowledge of the stuff. Then we realized that basically as a currency, it's never worked because it's fixed supply and there's variable demand. So there's no mechanism. I wish there will be equilibrium on this thing. So if there's no equilibrium, then you have no price stability, then you can't do things like ever write loans. So you can never run a financial system on this thing. So currencies out by design because it's broken by design.

Then as a speculative asset, I know enough background in quantitative finance to be able to look at financial products and just look at the cashflows of this thing and figure out where its demand curve is generated from. The demand curve is purely speculative and there's no income. So as a speculative asset, it seems dead on arrival. So it looks to me like a bubble at that point then. It's just some tulip, which happens, but that's the essence of the conclusion that we came to together with our computer science and finance backgrounds.

That sounds like a really interesting conversation. Yeah, we didn't cover it much in this conversation and we won't, but the whole misunderstandings about how banking works, the role of central banks, the role of commercial and private banks, it's horribly misunderstood. From what I can gather, that is a big reason for or it's a big part of the narrative behind crypto. So that's, in some of our other episodes in this podcast series, that's some of the things we're going to try and debunk, but yeah, I think that's a key. It makes sense that after having those knowledge gaps filled in, plus your technical knowledge, I can see how that would've changed your perspective.

I think you're right. It's a very narrative-driven asset. It's just the problem is that the narrative it's based on is this phony populism against the banks, and don't get me wrong. There's plenty of problems with the financial system that I could go into for days and days and talk about inefficiencies and corruption, everything. Some of those are real and some of those aren't, but people have to realize that the financial system that we have today works. It works very, very well, and trying to replace that with this thing that seems fundamentally broken by design but wraps itself in this phony populous narrative about, "Oh, we can overthrow the banks, and we're just going to burn down the financial system," like a Phoenix, something new will arise that will create this new utopia to me seems absurd. If that's what's driving the demand curve for the asset, then I think there's a hard reality that people are going to meet at some point in the future, and the narrative around stuff seems very problematic.

These are old narratives. They've been around for decades. I mean, since the creation of the federal reserve, there have been people who have spoken out very skeptically about what it is. I mean, well, yeah, I don't want to get too deep into this stuff now, but there's a book by, I'm sure you've read it, by David Columbia that goes into the right wing politics that drive most of the Bitcoin narrative. That's a whole other very interesting rabbit hole to go down.

So you're clearly passionate about this. I mean, you've written a lot on a bunch of different mediums. You've been on other podcasts. You're joining us for this podcast. Why is speaking out about this important to you?

I would say a lot of the narrative around these assets doesn't seem to jive with my understanding about how the financial system works. It seems to be untethered to reality, and it seems to be creating a lot of suffering in the world. My fear is that there's some parallels that I see right here right now with what we saw, not exactly, but what happened with subprime back in 2008. I don't want to see this bubble pop and then a lot of retail investors basically just get completely wiped out because that would just be very, very socially corrosive just like 2008 was. I'm young enough to have lived and come of age during that time period. I just don't want to see another financial crisis again and look at this bubble, and I don't see any reason why it's not a bubble and I don't see any reason why it won't come to a very violent, catastrophic end.

So I'm passionate about the fact that if I was a banker back in 2007, would've had an obligation to speak out then, yeah, probably. I see the same manias and corruption, some forces that gave rise to that manifesting themselves in a new form now, and anything I can do to slow the bubble down and mitigate some of the disasters that innocent people don't get wiped out in this whole thing, I think I have an obligation to do that, to speak out about it.

That makes a lot of sense. One of the other people that we're speaking with for this series has done research on investors in Denmark, I believe, who in the financial crisis lost everything in Danish bank stocks. I believe that was their sample, but they found major impacts on future risk taking for people that were burned by those investments. So I think it's a real valid concern not only that retail investors will lose wealth, but they'll be intimidated from taking good risk, diversified, positive, expected, return type risk with their investments in the future, which can have real long lasting societal and wealth inequality impacts.

Absolutely. The biggest slide in finances at this time it's different. Everybody thinks that they're going to time the bubble. Timing bubbles is a negative something. Most people get wiped out. People need to recognize that the difference between gambling and investing, and investing is not sexy. It's about time in the market, not timing the market, Warren Buffett. I think there's a new generation of investors that have some very, very pathological ideas about returns and massive growth in their portfolios that are not tethered to reality because they're trading these super hyper volatile, negative sum products that are mostly based on gambling.

My fear is that there's a younger generation that's not ultimately going to hurt them when they go get into their 30s and start buying a house and everything when all their money is tied up in images of apes or dog money or something. I don't see how this ends well for most of them.

Yeah. I don't know if I came up with this or have I heard somebody else say it, but if you're excited about an investment, it's probably not a good investment.

Yeah. Slow and steady growth over time and diversification and understanding the products that you're buying, all the things that Warren Buffetts says about value investing, take those to heart because, clearly, he's done well for himself.

That is true. He has. All right, Stephen. Well, that's the end of our conversation. This has been great. We really appreciate you giving us the time.

Take care, gentlemen. Later.