Op Ed: In Defense of Bitcoin: A Response to Nouriel Roubini
Rhetorical style should never distract from the content of someone’s arguments. So when I heard that Nouriel Roubini testified to the U.S. Senate Committee on Banking, Housing, and Urban Affairs on the cryptocurrency and blockchain ecosystem, I decided to peruse his testimony and reflect on it.
His testimony makes an extensive number of critical points on the Bitcoin and cryptocurrency ecosystem. In this article, I want to set out a defense of Bitcoin against them. Importantly, I have not attempted to critically assess each and every single point he makes. Instead, I will focus my discussion on those areas of his testimony which I think are most important to defending Bitcoin.
I offer no extensive defense for the cryptocurrency ecosystem in general because I am, to a substantial extent, in agreement with Roubini. Though there are certainly some projects of technical and economic value outside of Bitcoin, in my view, it is mostly Bitcoin that is of value and, therefore, what specifically needs to be defended against his claims.
With regard to the “enterprise blockchain” ecosystem, as Roubini rightly notes, the word “blockchain” in this context is usually somewhat of a misuse of the term, and the initiatives in the enterprise ecosystem currently labeled as such generally have little to do with Bitcoin and cryptocurrencies. This is not to say that these enterprise initiatives have no value and are complete fads that are now “fading and imploding,” as he notes. There is currently a substantial wave within financial institutions and elsewhere in creating and reshaping digital platforms that integrate the actors and activities in ecosystems to enable value exchange. Using what are generally called “permissioned ledgers” can sometimes be useful in building these platforms. But such permissioned ledger systems should have entirely different designs, properties and purposes than Bitcoin and cryptocurrencies. The common slogan that these permissioned ledger systems use “the technology behind Bitcoin” is misguided, and discussion of their merits should be separated from the discussion on Bitcoin and cryptocurrencies.
Overall, Roubini makes a fair number of points on which he and I are largely in agreement. But there are also a substantial number of Roubini’s arguments which I think are flat-out wrong, misguided or at least lacking in proper nuance and context, specifically as they apply to Bitcoin. And this matters. As someone with a skeptical perspective on this ecosystem as a whole, I am left with a substantially more positive disposition toward that ecosystem, particularly where the future evolution and legitimacy of Bitcoin are concerned.
Bitcoin’s Main Value
From the discussion, it is clear that Roubini does not believe there is much of a positive case to be made for Bitcoin having any societal value, let alone any other cryptocurrency. That is obvious from a number of his remarks. He notes, for instance, that “until now, Bitcoin’s only real use has been to facilitate illegal activities such as drug transactions, tax evasion, avoidance of capital controls, or money laundering.” In addition, he attacks many of the core propositions that Bitcoin supporters commonly tout, such as its potential to be a store of value.
A positive case for Bitcoin, however, can certainly be made, and I don’t think that any of his more legitimate criticisms really significantly undermine it. In my view and that of many others, the fundamental value proposition that Bitcoin could potentially offer to the world is to be sound money for the digital age.
As can be expected, there are some different ideas of what exactly this means within the Bitcoin community. But I see sound digital money as a money with two aspects that would make it qualitatively different than modern fiat currencies: (1) it is a better store of value, and (2) it better enables financial sovereignty. Sound digital money may not be optimal in every single possible respect, but it would have notable advantages over fiat currencies in the aforementioned respects.
Bitcoin already has some aspects to it which we would expect from a good store of value: there is a hard guarantee on the scarcity of the number of bitcoin (about 21 million); it has a number of advantages over gold in terms of divisibility, transportation and storage; and, finally, the system has shown great resilience in the face of attacks over the last decade, as we would expect from a resilient money that holds its value in the long run.
In addition, Bitcoin transactions are currently practically uncensorable, bitcoin is a digital bearer asset that is difficult to confiscate if stored properly, and the system offers an opt-out to the government fiat and the banking system. These are the types of properties to be expected from a currency that wishes to promote financial sovereignty.
Clearly, however, bitcoin has gone only some way toward becoming sound money for the digital age, and it is far too early to claim victory. To focus on an obvious point: although bitcoin has, for many, been a good investment that has displayed a generally upward line in its market value over the years (despite Roubini’s persistent warnings at much, and I mean much, lower prices), bitcoin price volatility is still way too high to reasonably claim the store-of-value status. Given the small market for bitcoin, this price volatility is not surprising. Even though this volatility has been decreasing substantially over the years, it would certainly have to decrease further in the long run for bitcoin to be considered a reasonable store of value.
I would argue, however, that we should value bitcoin not so much because it is fully formed, sound digital money, but rather because it seems to be in the process of becoming sound digital money. The value argument is, therefore, one primarily concerning its potential.
Why should anyone care about financial sovereignty and the store-of-value proposition? Let’s begin by looking at financial sovereignty and deal with the store-of-value proposition later. It strikes me that there is a plethora of possible examples that attest to the value of financial sovereignty, but Roubini has made no attempt to engage with any of them.
Sweden, for example, has almost completely eliminated cash and made its population, to a large extent, dependent on the servers of their banks for access to their money and the ability to make or receive payments. That gives these banks an enormous amount of power over the population. It also creates highly centralized points of failure, which increases overall risks with regard to system failures and pernicious forces.
There are also many people around the world who depend on banks that are much less reliable than those in Sweden. And Bitcoin could be a good medium to hold some of our value, particularly for people who find themselves in those circumstances. Roubini’s claim that Bitcoin offers no value to the world’s poor and that only “fintech” matters in this regard is not very convincing. The “black mirror” usage of modern technology by the Chinese government that is intertwined with applications such as WeChat and Alipay, which Roubini mentions as apparently good examples of such “fintech,” is hardly a beacon of light if we care to advance freedom and democracy in the digital age.
All this is not to say, of course, that banking and banks have no value or are going to disappear entirely from the face of the planet. Intermediaries such as banks can address knowledge as well as risk problems for us. Such intermediaries will continue to perform these tasks in the future, and it makes sense that a lot of our value is, therefore, managed by them.
But it also seems wise to hold on to some of our value not merely via the protection of the law (as is the case with our bank deposits), but also in a technical sense. Before Bitcoin, it was not possible to have a bearer asset in a digital form — which is why Bitcoin was technologically revolutionary and a break with our technological past — and one would have had to resort to cash, gold, paintings or something similar. Bitcoin now makes it possible to have a bearer asset in digital form that can be easily divided and transported, and, with some expertise, well and safely stored. In addition, Bitcoin may offer new types of possibilities for “custodian” solutions that decrease risks for the user without giving up control entirely. An obvious example here is a basic multisignature scheme.
In conclusion, Bitcoin is the product of decades of research on digital bearer assets and digital scarcity, and to me it’s quite likely that the world will come to value it, once they understand it better and once it has become more practical to use for everyone, not just those who are technologically savvy.
Bitcoin’s Decentralized Character
The key property of the Bitcoin system that could allow it to play the role of sound digital money is its decentralized character. The term “decentralization” is thrown around a lot, sometimes with different meanings. But in my view, the key way in which Bitcoin seems to be decentralized is that power over the network appears to be dispersed: that is, power over the activities on the network and the rules that govern it seems to be distributed among a community, rather than concentrated in the hands of a single entity or a small group of entities.
This does not mean Bitcoin is a perfect democracy. Many philosophers have tried to come up with conceptions of what a perfect democratic community would be, whether it consists of everyone in the community having an equal say in decision-making, its members having power in decision-making relative to their interest in the decision being made, and so on. However we might conceive of that ideal, it is certainly not the case with Bitcoin that everyone in the community has an equally powerful voice (nor is that the case for any other community on the face of the planet). Instead, the point is merely that power over Bitcoin seems to be very spread out and non-concentrated, much in the way that decision-making power is spread out in parliamentary democracies over the courts, bureaucracy, executive, media, various representative bodies, the public and so on, but also to a greater extent and with a more popular form.
Decentralization is fundamental to Bitcoin because without it, there are no guarantees on the main properties that underlie the store-of-value and financial-sovereignty propositions. If the network were controlled by a central party, for instance, the limit of 21 million bitcoin or lack of censorship in transactions would largely depend on its whims. To put it bluntly, without Bitcoin’s decentralized character, bitcoin (the currency) would not be much more than the equivalent of digital Beanie Babies (an expression which I believe I owe to Joshua Unseth and Chris DeRose).
It is common for Bitcoin detractors to claim that the system is not really very decentralized, usually with the argument that power is concentrated in the hands of a few mining companies. Indeed, there is a limited number of mining pools and companies that produce mining equipment, and a lot of the network’s hash rate is controlled by a few mining companies. Mining also still seems to be relatively concentrated in China. However, there are potentially some positive changes in these respects. It seems, for example, that mining is shifting away from China to countries such as Canada, the United States and Iceland. In addition, there is technological progress that could help decentralize the mining ecosystem further (see, for instance, Matt Corallo’s recent work). Nevertheless, it is certainly fair to conclude that the mining market is largely oligopolistic in character, and we can presume that it will probably always retain that character to some extent.
We should not, however, make the mistake of concluding that oligopolistic tendencies in the mining market equate to a system that is centralized overall. That simply does not appear to be the case. The decentralized character of the system as a whole was significantly attested to, for example, by the failure of both the Bitcoin Cash fork and the New York Agreement in 2017 — basically both initiatives instigated by a small number of actors predominantly from the mining sector and a few other types of big companies to push through scaling solutions that were against the will of the majority of the market. So even with rather oligopolistic tendencies in the mining market, Bitcoin has apparently managed to stay relatively decentralized as a system — a sign of Bitcoin’s strength, I would say.
In the end, the claim that Bitcoin is decentralized is a thesis based on empirical observations and theoretical understandings of how Bitcoin works. There is much to be said for it. Those such as Roubini who claim that Bitcoin is largely controlled by miners or other centralized industries would have to be able to explain these empirical facts such as mentioned above and what is wrong with the checks within the system that currently curb their influence. But I have yet to see a really convincing story by him or anyone else along these lines.
As a final thought on this matter, the history of political systems does show that maintaining a community with a highly popular character is very challenging. Over time, such systems tend to be subverted in various ways. So skepticism in this regard is certainly warranted, and maintaining its decentralized character is, therefore, in my view, certainly the largest challenge for Bitcoin going forward. But in order for any concerns to be legitimate in this regard, they need to move beyond myopically pointing out the oligopolistic tendencies in the mining market or other types of Bitcoin industries, particularly given the fact that Bitcoin seems to be becoming more decentralized over time, not less.
Bitcoin’s Further Value
All the previous discussion is not to say that the only potential value to be gained from the Bitcoin system is sound digital money, buttressed by its decentralized character. It may be possible to scale the Bitcoin system to make it more suitable for payments and not just a system for a censorship-resistant store of value. Such scaling will undoubtedly have to come from a complex combination of initiatives that are difficult to foresee completely at this point, just as it would have been difficult in 1995 to envision exactly how the modern version of Netflix could have operated on the internet 20 years later, as Andreas Antonopoulos recently quipped.
The exponential growth of the Lightning Network this year reveals one of the ways in which such scaling is possible, and it is telling that Roubini makes no mention of it and other efforts in his discussion of Vitalik Buterin’s “inconsistent trinity,” which claims that it is impossible to have a system that is decentralized, secure and scalable.
There may also be further applications for Bitcoin. It may also be the case, for instance, that eventually Bitcoin can offer new ways to raise capital, such as through initial coin offerings (ICOs) — though they would have to be executed much better than the ways we have seen so far (there are some recent initiatives in this direction such as RGB and Drivechain, which would at least provide a sounder technological basis for them than the Ethereum platform). There already exist digital art and game applications on top of Bitcoin (e.g., Spells of Genesis, Rare Pepe), even if these initiatives are at an early stage. The open timestamping protocol or the possibility for one-time cryptographic seals could also have utility for financial institutions, government and other types of organizations in a number of applications (see, for instance, the work of Peter Todd on these topics).
But all of these possibilities rely on Bitcoin being sound digital money, as this is what supports the security and most important properties of the system. As money is one of our most important societal institutions, it is also by far the most important value proposition of Bitcoin from a societal impact perspective. If Bitcoin continues to grow, it could potentially have an impact on capital controls, time preferences and consumption patterns, monetary policy, financial privacy and so on. It could, for instance, make the system of excessive government debt, supported largely by the Federal Reserve in the United States, of which Roubini speaks so highly, a somewhat more complicated affair.
And in this way, the whole scaling discussion is somewhat misguided. Even if Bitcoin cannot scale to become a good digital payment system or to have any other applications, it would still have substantial value and could create a substantial impact on the world by offering a qualitatively different form of digital money, one that certainly scales better than the physical gold market. This system is certainly costly. And as subsidies for mining will continue to decrease over time, costs will noticeably be displaced directly to the users through transaction fees. The security of Bitcoin requires a price to be paid. But it’s hardly a bug in the system as Roubini makes it out to be: the system is designed to be that way. You cannot compare the costs of buying a cup of coffee with the ability to move censorship-resistant digital gold almost instantaneously around the world.
Importantly, lest I be misunderstood, it is most certainly not my contention that we are working toward a future with a “libertarian decentralization of all economic activity, transactions and human interactions,” where “everything will end up on a public decentralized distributed permissionless trustless ledger.” Bitcoin has the potential to enable sound digital money. There may be additional useful applications. But centralized systems work fine in a range of contexts. In fact, I would contend they work fine in most contexts. I certainly do not feel very comfortable with the idea of having my patient data, my identity or my house registered on the Bitcoin blockchain, to name but some of the more questionable ideas that I have seen circulating in the ecosystem.
Bitcoin as Money
Bitcoin is also not really “money” according to Roubini. In order to be money, something has to fulfill three purposes, he contends: a unit of account, a means of payment and a stable store of value. In his view, “it is none of those things.”
This is an argument frequently made by many other economists, but it strikes me as somewhat akin to saying that early gasoline-powered cars were not really forms of transportation because they constantly had technical issues and did not move very fast. I find it reminiscent, in other words, of the claim made by Henri Studebaker in the 19th century that gasoline-powered vehicles “are clumsy, dangerous, noisy brutes which stink to high heaven, break down at the worst possible moment, and are a public nuisance.” Just because Bitcoin does not yet completely have the properties that we normally associate with money does not mean that it cannot come to have those properties, or at least some of those properties, notably the store-of-value aspect.
And to some extent, it already does. Bitcoin certainly functions as a unit of account within the cryptocurrency ecosystem as a whole. A number of people with a long-term time frame already use it as a store of value. Although not frequently used as a medium of exchange, in my view, that can only come in the long term as the ecosystem grows and becomes more stable, and as layer-two solutions such as the Lightning Network are developed. It is just unrealistic to expect that Bitcoin could have become a fully developed form of money within a decade, and it may still take another decade or even more to fully develop.
And there is certainly historical precedent that attests to the challenge of introducing new forms of money. As Edin Mujagic, for example, shows in his wonderful exposition of Dutch monetary history from the 19th century onward (Boeiend en Geboeid: Een Monetaire Geschiedenis van Nederland sinds 1814/16), the introduction of paper money in the early 19th century by the Dutch Central Bank was hardly smooth sailing. He notes, “From day one, the Dutch Central Bank faced substantial resistance. Distrust of the bank was enormous. No one wanted to use their paper money. The average Dutchman refused to use it. … Whoever received paper money wanted to get rid of it as quickly as possible. The refusal of paper notes was not a temporary phenomenon: decades after the creation of the Dutch Central Bank, this was still the case” (p. 26, my somewhat liberal translation and emphasis). The idea that Bitcoin should have already been a successful form of money by now, given its time span of 10 years, therefore, strikes me as very shortsighted.
Roubini is partially correct when he contends, “The idea that hundreds of cryptocurrencies could viably operate together not only contradicts the very concept of money with a single numeraire that can be used for the price discovery of the relative price of thousands of good; it is utterly idiotic as the use of multiple numeraires is like the stone age of barter before money was created.” This is one reason why I am skeptical of the profuse number of cryptocurrencies other than Bitcoin.
But it is certainly not unrealistic to think that a number of currencies may operate alongside one another. In fact, that was the case for much of our monetary history, and the centralization of money issuance has only been a relatively recent phenomenon, spurred on by central banks, primarily since the 19th century.
Roubini’s main argument for a single numeraire is efficiency. He notes, “In the U.S., the reason we do not use euros or yen in addition to dollars is obvious: doing so would be pointless, and it would make the economy far less efficient.” If Bitcoin offered the same value proposition as fiat currencies, then his statement would indeed make some sense. But the point is that Bitcoin offers a qualitatively different value proposition. At the end of the day, it may very well be that there is not one form of money that offers all the kinds of value we might want from it, so a multicurrency system could certainly be valuable in this regard, despite some sacrifices in efficiency, which, given modern technology, would probably not be all that great in any case. In addition, there are also potentially benefits of competition that may be generated by a multicurrency system. It has always struck me as odd that economists who generally tout the benefits of competition in every aspect of economic life fail to see any room for those types of benefits when it comes to currencies, even if there are advantages to having a single currency.
I am generally skeptical of most alternative currencies — meaning here those that have their own blockchain system, not higher-level tokens — precisely because Bitcoin seems to have already captured the main value proposition that can be proffered in the ecosystem and has taken what appears to be an insurmountable lead. Nevertheless, it may indeed be the case that there are additional value propositions that Bitcoin cannot entirely capture, or that some alternative cryptocurrencies can carve out their own existence around a similar value proposition. This would provide room for perhaps a few more additional cryptocurrencies that operate outside the confines of just a small community. So far, however, I am not strongly convinced of that possibility.
The Intrinsic-Value Discussion on Bitcoin
Roubini contends that “cryptocurrencies have no intrinsic value, whereas fiat currencies certainly do, because they can be used to pay taxes. Fiat currencies are legal tender and can be used and are used to buy any good or service; and they can be used to pay for tax liabilities. They are also protected from value debasement by central banks committed to price stability; and if a fiat currency loses credibility, as in some weak monetary systems with high inflation, it will be swapped out for more stable foreign fiat currencies — like the dollar or the euro — or real assets such as real estate, equities and possibly gold.”
This point on intrinsic value is commonly made, both by gold bugs and fiat bugs alike, but usually for different reasons. For gold bugs, the intrinsic value lies in the fact that gold can be used for other purposes, such as jewelry, satellites and so on. This is really a misuse of the term “intrinsic value,” which, on the face of it, would have to refer to the idea that something has value regardless of its utilitarian value, such as the value of beauty that a painting may have regardless of its price. It’s also misleading because the primary value of gold throughout history has been as a monetary asset, due to properties such as transportability, divisibility, scarcity and its stock-to-flow ratio, not its utility for other purposes.
Roubini makes this same linguistic error. Paying taxes or serving as legal tender is clearly also a utilitarian, not an intrinsic, value of fiat money. A government backing it may also give it additional utilitarian value but cannot really be said to be adding intrinsic value.
Putting linguistic quibbles aside, the chartalist argument regarding tax payments and government backing is not entirely invalid (though much of it depends on which government that is). That certainly seems to be one way in which currencies can acquire value. And it would certainly be a boon for bitcoin if it became widely used for tax payments, as it already is in various Swiss localities.
The argument is merely that tax payments and government backing are not the only aspects of a currency that can give it value, as I’ve made the case for earlier in this piece. Gold here is the historical precedent. It certainly did not have value as a currency only because of the ability to pay taxes on it or the fact that governments valued it. Instead, however, gold has had substantial value, including as a currency historically, because of a number of beneficial properties. And there are at least many reasons why gold does not function well as a currency in the digital age, which are to a large extent addressed by Bitcoin. In addition, given that all major currencies are controlled by states, there is room to argue that Bitcoin is valuable precisely because it is not controlled by a state.
An Inflation-Resistant Asset
Roubini argues that “the usual crypto critique of fiat currencies that can be debased via inflation is nonsense: for the last 30 years commitment to inflation targeting in advanced economies and most emerging markets has led to price stability (the 2% inflation target of most central banks) and for the last decade the biggest problem of central banks has been that achieving the inflation target of 2% after the GFC has become extremely difficult as, in spite of unconventional monetary policies, the inflation rate has systematically undershot its 2% target.”
At least one big problem with this argument is that current definitions of inflation usually do not consider properly all the basic living expenses for an average individual, most notably housing prices. There have been drastic swings in housing prices in many Western countries in recent years, and a big role here has been the policies of central banks. The worrying low interest rates we’ve seen since the early 2000s, as well as quantitative easing policies more recently, have caused two periods of substantial inflation in housing prices with a deflationary period in between in the form of the Great Recession. (In some places such as the Netherlands, well-intentioned but misguided policies by the government to stimulate home ownership have also played a role.) These types of monetary policies are also likely to lead to a similar kind of crisis in the coming years.
Additionally, these recent central bank policies have had a massive impact on the valuations of financial assets, notably stocks. I would not be surprised to see a massive decrease in these valuations somewhere in the coming years. It may be valid not to include those considerations in our inflation calculations, but then inflation calculations do not capture all the negative aspects that may result from our central bank policies.
Even if we ignore the problems with current definitions of inflation and were to accept that central banks have indeed managed to keep inflation below 2 percent for a substantial period of time in a number of countries (certainly not Argentina, Turkey, Venezuela or Zimbabwe, to give some recent examples), it is still the case that just holding on to my cash is not exactly a boon to my wealth. Even the U.S. dollar has lost much of its value in the last 30 years.
In addition, fiat currencies purely based on trust can never manage to give the same guarantees of inflation-resistance as Bitcoin. To close, if gold is any indication, then an inflation-resistant asset could certainly also become a store of value in the modern day (as I have also argued for already), and, having solved some of the difficulties of having gold as a currency in the digital age, could also become a more fully developed currency, if history is any indicator.
Roubini also argues that bitcoin is subject to inflation. He notes that “cryptocurrencies in general are based on a false premise. According to its promoters, Bitcoin has a steady-state supply of 21 million units, so it cannot be debased like fiat currencies. But that claim is clearly fraudulent, considering that it has already forked off into several branches and spin-offs: Bitcoin Cash and Bitcoin Gold. … Moreover, hundreds of other cryptocurrencies are invented every day, alongside scams known as ‘initial coin offerings,’ which are mostly designed to skirt securities laws. And their supply is created and debased every day by pure fiat and in the most arbitrary way. So cryptocurrencies are creating crypto money supply and debasing it at a much faster pace than any major central bank ever has. No wonder that the average cryptocurrency has lost 95% of its value in a matter of a year.”
But this argument just reflects the misguided idea that all cryptocurrencies are somehow interchangeable. Importantly, the Bitcoin system is much more than a design and code alone. Its innovativeness and value as a system also stems from its infrastructure, products, community, history, brand-name recognition and distributed governance. We might as well argue that the U.S. dollar has been subject to extremely high inflation since 1935, because of the success of the board game Monopoly.
The Early Days of the Internet
I have now addressed many of Roubini’s comments on bitcoin as money, so let me turn to some of his other criticisms.
To start, at some point, Roubini argues that blockchain technology does not resemble the early days of the internet: “Blockchain’s boosters would argue that its early days resemble the early days of the internet, before it had commercial applications. But that comparison is simply false. Whereas the internet quickly gave rise to email, the World Wide Web, and millions of viable commercial ventures used by billions of people in less than a decade, cryptocurrencies such as Bitcoin do not even fulfill their own stated purpose.”
According to Roubini, the start date of the internet was the launch of the World Wide Web in 1991. Taking this date as a starting point for the internet is really misleading. It is true that the internet quickly gave rise to users and applications after the invention of web browsers (the first, Netscape, in 1994). But before that time, the groundwork of the internet had been laid for several decades by a confined group of generally highly technical users and builders. In other words, it took several decades of laying the groundwork for the internet to finally make it usable for the average person, primarily due to the invention of web browsers.
And this long period toward adoption by the mainstream for new technologies is hardly uncommon. The world’s first credit card, the Diners Club card, was invented in 1950, but it took several decades before credit cards came into more widespread use. The groundwork for pagers was laid in the 1960s, but by the mid-1980s they were still only used primarily by doctors and drug dealers. The first combustion-engine cars were created at the end of the 19th century. Yet, when I was a small child, my grandfather still used to brag to me about how his family was only one of three to own a car in his mid-sized Dutch town of several thousand inhabitants in the 1920s.
So while it’s true that successful new technologies will generally, at some point, have an “exponential increase of the number of users,” it is misleading to state that this generally happens very quickly. In fact, I would say it is quite the opposite for innovations that are truly revolutionary.
Why the Protocol Analogy Makes Some Sense
Roubini also takes aim at the comparison that is frequently made between blockchains and universal protocols such as TCP/IP.
My interpretation of that comparison is basically the following: that competition in the blockchain space much more resembles competition between alternative conceptions of basic internet protocols like TCP/IP than it does competition between software applications for such things as email or social networking. And that, additionally, due to existent network effects and its decentralized nature, it is much more likely that the Bitcoin protocol will be the basic building block for facilitating interactions that we might eventually want to take place on a blockchain system, rather than, say, the Litecoin or Ethereum protocols.
This is not to say that Bitcoin will necessarily be the only blockchain system in existence, but merely that it will probably be the dominant blockchain system. We saw these types of “protocol wars” in the early days of the internet, and I suspect they will play out in a similar way in the blockchain space.
This position is not meant to imply, as Roubini suggests, that Bitcoin is intended to replace TCP/IP; that would be an absurd claim. It is also not meant to convey, as he suggests, that every digital transaction or interaction we have with each other should be registered on the Bitcoin blockchain. That is not possible even for things that you might actually want to use Bitcoin for, such as payments. The only way a large number of people could ever buy all their cups of coffee with bitcoin would be through second-layer solutions.
Furthermore, there are many ways in which we would want to continue to interact in a Bitcoin world that doesn’t require Bitcoin. Applications for Facebook, email, Uber and a great long list of other things do not need to be built on top of Bitcoin or any other type of blockchain system, despite claims of the new internet, web 3.0 and so on that have frequently been made in the space in recent years, predominantly by people trying to launch pie-in-the-sky schemes. At most, such applications would partially leverage Bitcoin, such as for integrated payments.
Should investing in one of the cryptocurrencies be seen as investing in one of these protocols? Roubini claims it does not, and I would say that this is correct for most cryptocurrency systems. As most of them have little to no economic value, any technical value of the protocol can largely be detached from their performance within those systems.
This is not the case with Bitcoin, however. Investing in bitcoin, the currency, is, in fact, taking a stake in the adoption of the Bitcoin protocol. This is because Bitcoin is a system innovation in which the protocol acquires its value largely from sitting within that larger system. This is not to say that the Bitcoin protocol could not be decoupled from the Bitcoin system to some other fruitful end at all. And, indeed, a bitcoin investment does not really benefit much from such new applications. It’s just to say that the most important instance of the existence of the Bitcoin protocol is within the Bitcoin system, so that any investment in the currency is, at least to a significant degree, about investing in its protocol in this regard.
Bitcoin and Illegal Activities
According to Roubini, “Until now, Bitcoin’s only real use has been to facilitate illegal activities such as drug transactions, tax evasion, avoidance of capital controls, or money laundering.” Next to the energy consumption argument, this is probably the most common and long-standing argument that we see against Bitcoin. There is an element of truth in the objection here, but there is also substantial room for nuance and reflection.
To start, the association between Bitcoin and crime has been a prominent narrative in recent years, and a lot of that has to do, I suspect, with the dynamics underlying journalism. My thesis here basically is that regardless of the extent of crime in the Bitcoin ecosystem, mainstream journalists would have naturally tended to focus on such narratives. Why? It basically relies on three propositions:
- The Bitcoin ecosystem is difficult to understand.
- Journalists look for easy, relatable topics for their audiences.
- Journalists focus on conflict and drama.
Hence, as narratives about Bitcoin and crime are easier to report on, are easily relatable to by the public at large and are full of conflict and drama, it is only to be expected that stories about Bitcoin and crime would have circulated profusely. My point here is not that there are no mainstream journalists at all that produce quality content on Bitcoin. There are. Instead, it is merely to say that for many, stories about crime would naturally have exerted a lot of pull to write about. I think that this situation is continuously changing, however, and compared to just a few years ago, mainstream reporting has substantially improved, even if it is not yet where I would like it to be.
Is that all there is to it? Certainly not.
Early adopters to technologies tend to come in two groups: those members of the general public who see the value of an innovation early on and enjoy experimenting with it; and criminals, who operate in a high-risk environment and try to stay ahead of the law enforcement agencies. As the social psychologist Robert Merton once quipped, “Criminals are innovators”: they tend to adopt new technologies quickly, and they tend to be innovators themselves.
It’s not so hard to find examples of this in the historical record. For instance, early users of pagers in the 1980s generally consisted of doctors and drug dealers. Why would drug dealers value pagers? It helped them have better communication lines to suppliers and buyers. In addition, pagers were used as part of warning systems when dealing drugs on the streets. Certain people would be on the lookout for patrolling police cars and notify street dealers whenever they were in the area.
Many of the debates we saw around pagers at the time are reminiscent of the discussions we have seen circulate about Bitcoin in recent years. Take for instance this quote from an associate superintendent in the United States on pagers in the 1980s: “‘How can we expect students to “just say no to drugs” when we allow them to wear the most dominant symbol of the drug trade on their belts,’ said James Fleming, associate superintendent for the Dade County Public Schools in Florida, which bans beepers and other such disruptive items as radios. ‘I can think of no reason outside of special medical emergencies for students to carry beepers’” (New York Times, Sept 25, 1988, my emphasis).
You can find many other examples of this quick adoption of new technologies by criminals throughout history. And to some extent this probably happened to Bitcoin as well. One of the earliest use cases for cryptocurrencies was the purchase of primarily drugs on the Silk Road. In any case, most recent studies on the use of cryptocurrencies for illicit transactions tend to show a relative decrease of this type of usage over time, as is to be expected given the history of new technologies. An additional important discussion point here, of course, is the nature of the types of crimes that are committed with Bitcoin. The overwhelming majority is associated with the illegal drug industry, as is the case with illicit financial activities in cash or the financial system in general. In my personal view, the war on drugs does much more harm than good. But regardless of whether you share that view, such crimes are certainly of a qualitatively different nature than, say, terrorism. And to the best of my knowledge, neither Bitcoin nor any other cryptocurrency is hardly a boon for such activities at the moment. Investigation by the House Financial Services Committee attested to that fact recently.
There is one point that Roubini makes with which I am largely in agreement. The cryptocurrency ecosystem is afflicted with many scams and fishy schemes. By scam projects, I mean projects that intentionally have the purpose to fleece people from their money, such as pyramid schemes, Ponzi schemes and pump-and-dump schemes orchestrated by the founders. By fishy projects, I mean pie-in-the-sky schemes in which the dreamers who orchestrate these schemes do not intentionally set out to fleece the participants but in which greed has warped their critical thinking to a blameworthy extent.
It is often difficult to distinguish between whether something is truly a scam or whether a project is just a fishy scheme (or a project in which, as many people in the industry like to say, the participants are “scamming themselves”). Similarly, it can be difficult to assess whether a project is just misguided regarding its value proposition or whether there is any real blame to go around as in fishy schemes. In any case, I would say there are certainly a substantial number of scams and an even greater number of fishy schemes. I would say that the majority of the ICOs that I’ve seen fall into either of these two categories. And there is certainly a great number of alternative cryptocurrency projects that also fit that description.
But part of that is also a growing pain for new technologies and ideas. The ecosystem is, in a way, not very different than previous social movements that had the same kinds of factors in place, such as the railroad craze in the mid-19th century or the rise of the internet and the IPO craze of the late 1990s. William Aytoun’s classic “How We Got Up the Glenmutchkin Railway and How We Got Out of It” might as well have been called “How We Got Up to ICOs and How We Got Out of Them.” So like with many new technologies, it seems that Bitcoin — and, to be fair, a number of other valid projects in the ecosystem, some that I am probably not aware of — will have to suffer through the hype boom that spurs all these scams and fishy schemes.
Perhaps it is true, as Roubini contends, that the extent of scams and fishy schemes in the cryptocurrency ecosystem exceeds that seen in other historical bubbles, as perpetrating scams and fishy schemes has been incredibly easy. Perhaps that is also why the cryptocurrency ecosystem’s crash from the 2017 run-up has been more substantial than for other historical bubbles, as Roubini notes. But none of that proves that somehow Bitcoin will not shed these growing pains, just like as we have seen in the past.
Importantly, this is also an area in which governments can help support the industry. Established securities laws in Western countries, despite their flaws, generally do offer some degree of protection for the consumer, and I look favorably on these laws now starting to be applied. Not everyone who is reasonably disposed to the classical liberal visions of small government — in line with some of the greatest thinkers in our historical canon, such as John Locke or Adam Smith — believes that there should be no government at all. It is certainly not the case that everyone in the ecosystem believes that “all governments, central banks, moneys, institutions, banks, corporations, entities with reputation and credibility buil[t] over centuries are evil centralized concentrations of power that literally need to be destroyed” (my emphasis). In fact, I would go so far as to say that it is likely most don’t hold such extreme views — certainly not the almost 10 percent of the European population that, according to a recent study by the ING, holds cryptocurrencies.
Bitcoin and Energy Usage
The main general criticisms on Bitcoin, besides those related to crime, tend to focus on its energy usage. Roubini fully embraces them and claims that “the environmental costs of the energy use of Bitcoin and other cryptocurrencies is so vast that [it] has been correctly and repeatedly compared to an environmental disaster.” It’s true that Bitcoin consumes a substantial amount of energy, even if it is only a very small fraction of the world’s total energy consumption. But as with the crime argument, there is substantial room for nuance, reflection and further research here.
A balanced discussion on the energy usage of Bitcoin should focus on three separate issues:
- the current environmental impact;
- the expected evolution of this environmental impact;
- the value generated by energy consumption.
To start with the first issue, if we want to have a sensible discussion on Bitcoin’s current environmental impact, then how much energy is used by Bitcoin is not all that relevant. Much more relevant is where that energy comes from.
For instance, it seems that in recent years Bitcoin mining in China has concentrated particularly in areas where a lot of surplus hydroelectric power is created. That energy would otherwise have not been used for any other purpose. Furthermore, recently, it seems that mining operations have started to shift toward other countries, such as Iceland, Sweden, Canada and the United States, toward renewable and cleaner sources of energy.
If we want to have a sensible discussion around Bitcoin’s environmental impact, where the energy comes from is, therefore, a key issue. To the best of my knowledge, no one has actually created a comprehensive view on Bitcoin’s environmental impact in this way and shown any conclusive proof that Bitcoin is indeed an “environmental disaster.” Until someone does, I remain highly skeptical of that claim.
We have also seen some pretty bold predictions about the growth of Bitcoin energy usage recently. To get a sense of how mining will develop as an industry, we need to consider the main incentives of miners:
- Bitcoin miners search for cheap and reliable energy sources.
- Bitcoin miners search for cheap, reliable and efficient Bitcoin mining equipment.
- Bitcoin miners tend to concentrate in colder climates.
- Bitcoin miners search for socially stable operating environments.
- The costs that miners are willing to incur on mining will increase as the mining reward increases and will decrease as the mining reward decreases (the mining reward consisting of the addition of transaction fees and the block subsidy).
- Miner calculations and incentives are not so much influenced by network security, which is arguably already sufficient at the moment (Bitcoin mining offers security to the network only as a byproduct).
- It is, of course, very difficult to make exact predictions as to how these incentives will impact the evolution of total energy consumption of Bitcoin mining in the future. But broadly speaking, we really should not expect energy consumption to increase at the same rate as in the past.
Primarily due to decreasing block subsidies (in BTC terms), higher-layer value exchange solutions and a slower rate of price increases, the market will more and more start to converge on the price Bitcoin users are willing to pay for ultimate security. It strikes me as patently absurd, for example, that Bitcoin would use the same amount of energy as the United States by 2019 and the world by 2020 (as was estimated by the website PowerCompare in late 2017 on the basis of data from a website called the Digiconomist).
Notably, it seems common for society to become anxious about the extent of energy usage when new technologies are introduced. There was, for example, substantial worry about electricity usage of the internet in the 1990s, and many studies at the time grossly overestimated how much energy the internet would consume in the future.
The key question in all of this is, then, of course where the energy will come from in the future. Perhaps even more difficult to predict than energy consumption. But there are reasons to think that there will be substantial downward pressures on the environmental impact of Bitcoin mining in this regard. Renewable energy will continue to become more financially attractive relative to fossil fuels (consistently cheaper by 2020, according to the International Renewable Energy Agency), which will exert incentives on Bitcoin miners to use such energy sources. Additionally, the need for social stability and the decreasing rate of efficiency improvements in mining equipment will probably continue to push miners toward cheap, cleaner sources of energy in Western countries such as in the United States, Canada, Australia, Sweden and Iceland.
When assessing the environmental impact of Bitcoin mining, we should also not forget that there are potentially various types of benefits of having this industry. Bitcoin mining might, for example, help make alternative energy projects more financially viable as they can monetize surplus production that cannot easily be controlled with hydroelectric, wind and solar power. Also, the quest for cheap and reliable energy sources and other efficiencies will probably lead to further innovations such as in energy generation and heat conduction.
Importantly, I can only share my impression that concerns regarding the environmental impact of Bitcoin mining seem to be far overblown. I am hardly an expert on making estimations regarding the current and future impact on the environment of a system like Bitcoin. But, to the best of my knowledge, neither is Roubini. So there seems to be substantial further empirical and theoretical research that can best be done here by people who have real expertise in such matters, of which there is, again to the best of my knowledge, currently very little.
Clearly, none of my discussion on energy so far would matter if Bitcoin offered little of value to the public. But as I have tried to show above in various places, such a view can be defended. If that view does not hold up, one would expect that the whole “energy problem” would, in any case, resolve on its own, as Bitcoin would not likely stick around if it does not actually offer anything of value to anyone.
Another way in which none of my points on the energy discussion would be valid is if there were an energy-friendly alternative to creating a decentralized digital currency. Proof of stake is commonly touted as such an alternative. But in contrast to what Roubini suggests with his claim that “supporters of crypto have been promising forever — Buterin spoke of Proof of Stake (PoS) in 2013 — systems that are vastly scalable,” I and many others do not think there is any better way to create a decentralized digital currency than via proof-of-work mining, and that the whole idea of proof of stake is just fundamentally flawed. Many in the Bitcoin ecosystem, such as Andrew Poelstra and Paul Sztorc, have argued why that is the case over the years.
Finally, the argument about energy usage often suffers from improper framing. Roubini falls victim to this when he claims that “using millions of computers to do useless cryptographic games to secure the verification of crypto transactions is a useless waste of energy — as the same transactions could be reported at near zero energy costs on a single Excel spreadsheet.”
Consider placing a potential future form of Bitcoin next to our current global clearing and settlement system. Transactions on the Bitcoin Block Chain, regardless of their origins and destinations, are cleared and settled usually within an hour or so. It may be possible to increase the transaction efficiency in this system in a number of ways, such as through the Lightning Network, Drivechain, Liquid, Schnorr signatures and so on. Perhaps Bitcoin could eventually even compete with Visa through such innovations as a payment system. If we imagine the way that a future Bitcoin system could look and compare that to the current global clearing and settlement system for financial institutions, would it then really be so inefficient in this regard? I’m not trying to argue one way or the other, but merely pointing out that the argument of inefficiency commonly made seems to suffer from improper framing, as the current global clearing and settlement system certainly does not run on a “single Excel spreadsheet” and hopefully never will. (Note: even if Bitcoin can only achieve the sound-digital-money proposition, it would still be improper to compare its energy usage to that of an Excel spreadsheet.)
With his usual zeal, Roubini argues that this reframing of the energy discussion is invalid: “The mining of gold or the provision of financial services produces value added and output to the economy that is 1000X than the pseudo value added of crypto mining. And financial services provide payment and other services to billions of people daily in hundreds of billions of daily transactions. So of course their use of energy will be larger than crypto. Crypto is used by 22 million folks globally — less than 5 million active ones today — and its entire market cap is 200 billion — not the 300 trillion of global financial and real assets — and is producing value added that is a few billions a year — new crypto mining. But its energy use cost is already about $5 billion a year. So comparing the energy use of useless, inefficient and tiny crypto to the services of financial institutions serving daily billions of people is utter nonsense of comparing apples and oranges or, better, crypto parasites with useful financial services (payments, credit, insurance, asset management, capital market services) used by billions.’”
I don’t find this reasoning very convincing. The argument for Bitcoin, as I’ve explained, is very much about its potential for realizing the sound digital money proposition. There are certainly more than 22 million people (which in any case seems to be too low of an estimate for the people who currently hold cryptocurrencies, given the recent ING report) which could benefit from it given sufficient time. In fact, a substantial portion of the world could probably come to find value in sound digital money. And while there is substantial room for growth in Bitcoin’s value to the world, its energy consumption is unlikely to increase at the same rate in the future for the reasons I have stated, and likely to become more sustainable in its sourcing. If we can realize next to sound digital money a number of other valuable higher-layer applications for Bitcoin which take advantage of its secure base layer, then that is only the icing on the cake.
In conclusion, Roubini makes a number of valid criticisms in his testimony of the Bitcoin and cryptocurrency ecosystem. And I welcome his skepticism, as it can help sharpen the thoughts and arguments of those of us who have a more positive disposition toward it.
But in the end, his testimony fails to be sufficiently convincing against Bitcoin. The primary reasons are that he does not undermine the value proposition of sound digital money in a significant way, and that he does not do enough to dissuade us of the notion that Bitcoin could fully achieve that value proposition given sufficient time. Additionally, his criticisms on energy usage and crime seem to have fallen victim to the kind of fear-mongering that is common around the introduction of new technologies.
According to the recent rumors, Roubini is writing a book on the Bitcoin and cryptocurrency ecosystem. If he can truly undermine the value proposition of sound digital money and/or Bitcoin’s potential for achieving it given sufficient time in this book, I, for one, would consider parting with my bitcoin. But until that time, I think I’ll just hodl on to them with the rest of my fellow “con artists, self-serving peddlers, scammers, carnival barkers, charlatans, and outright criminals.”
With thanks to “the Janitor,” not an actual janitor but the Bitcoin expert who goes by that pseudonym, for providing thoughtful feedback on this opinion piece.
This is a guest post by Dr. Jan-Willem Burgers. Opinions expressed are his own and do not necessarily reflect those of Bitcoin Magazine or BTC Inc.