There’s so much fuzzy thinking around crypto-currencies and their enabling blockchain protocols that it was easy to for someone with a bulldozer mouth and outsized ego to rattle the industry.
That someone is Nouriel Roubini, the U.S. economist famous for predicting the 2008 financial crisis. Last week he gave testimony to the U.S. Senate decrying bitcoin as a scam and blockchain as irrelevant; he calls the industry a cesspool, its proponents bloodsuckers.
Timed with the ongoing collapse of much of the post-bubble crypto market, Roubini is trying to destroy the industry by scaring off institutional money.
Whether the industry deserves this shiv in the ribs depends on whether decentralized computing is going to have a major, positive impact on the world. If it’s not, then Roubini’s attack is warranted and everybody should move on.
Many of his arguments are valid, and are often acknowledged by blockchain proponents. But that doesn’t mean he's right.
The Roubini critique
Roubini’s most important attacks are as follows:
- Bitcoin is too volatile to be a means of payment or win widespread adoption;
- The world cannot accept a crypto reserve currency because it is not backstopped by a central bank, not tied to a tax base, and there is no state insurance against losses (unlike with bank deposits);
- As noted by Vitalik Buterin, the inventor of Ethereum, crypto so far can’t be scalable, decentralized and secure at the same time;
- Crypto is de-facto centralized in the hands of miners (located in countries lacking rule of law), and smaller coins are routinely hacked;
- Arguments comparing crypto today to the internet in the early 1990s are false because a decade on from bitcoin’s creation, no other use cases have emerged and instead of incorporating hundreds of millions of users, bitcoin now has a mere 22 million wallet holders;
- Banks, corporations and governments will never accept using public, permissionless blockchains, although they will accept niche-focused distributed ledgers that are private and permissioned, such as Corda;
- Bitcoin and other proof-of-work consensus mechanisms are an environmental disaster, but other consensus mechanisms haven’t made it off the drawing board;
- ICOs are nothing but scams.
Roubini’s flawed assumption
Many of the problems he identifies can be addressed by technological improvements. He misrepresents Buterin’s trilemma (scale, decentralization, security) as an impossibility, rather than a tech challenge. He mixes hacks of exchanges with the security of the bitcoin protocol itself. He overstates the threat of Chinese miners hacking the system.
Roubini’s attacks would be more reasonable if decentralized computing was on the cusp of widespread deployment. That’s what’s behind Roubini’s biggest complaint, that the idea that it took the internet a mere decade to scale.
Blockchain aficionados have themselves to blame for this one, by constantly parroting the line that the technology is akin to the internet in the early 1990s. If that were true, Roubini’s arguments would hold more weight.
But Peter Van Valkenburgh, director of research at Coin Center in the U.S., who gave opposing testimony alongside Roubini, notes that email’s TCP/IP protocol was invented in 1972. It took two decades for the underlying tech to get to the point of adoption just as the internet was taking off. Bitcoin, Ethereum and other blockchain protocols are vying to play a similar role as TCP/IP.
That suggests to me that today isn’t like 1994 for blockchain...it’s more like 1984. Mainstream deployment is further off than its proponents would like.
The fact that blockchain’s first use case was bitcoin, that is, digital cash, has attracted speculation as well as the interest of financial institutions, far too soon for the healthy development of the tech. That’s why we’re seeing so many incompatible chains being created. This compatibility issue is both a tech and a business-model challenge.
Tim Berners-Lee faced a similar challenge once he invented the World Wide Web; he had to spend years thereafter lobbying big tech companies not to create walled gardens but to let people seamlessly travel among their sites; that is, to allow an “inter-net”. But the Web remained in the domain of tech-nerds long before the late-1990s dot-com boom.
Decentralized computing, on the other hand, because of its financial use case, went through its Pets.com debacle long before the technology was ready.
Why decentralization matters
Even so, it’s simply not true that the tech is being shelved. I speak with banks using variants of decentralized computing to reach new customers; with fintechs using it to serve people who are bypassed by orthodox financial service providers; and with licensed financial investment groups looking for ways to enter the space. When I speak with large-scale financial institutions, the tone of conversation has shifted from suspicion to practical questions.
What decentralized computing does is enable apps to run without needing to trust a single operator or administrator. Roubini assailed the anarcho-libertarian thought leaders who seem to think the Fed is evil. I agree: I also don’t think the Fed can be or should be disrupted by digital money.
But this misses the point that any centralized point of failure is a risk. Just as the central bank of Bangladesh, whose SWIFT transmissions were used by hackers to defraud it of over $100 million, or the 143 million Americans whose data was lost in the Equifax hack.
It also misses the vision for crypto to allow anyone with a computer and an internet connection to send money as if it were cash, without the need to have a bank account – to allow paper cash, which we accept in person in complete faith, to be replicated digitally.
And it misses the vision of combining the security of decentralization with the fungibility of electronic money for the coming explosions of payments to power the Internet of Things and the rapid growth of e-commerce.
Moreover I don’t see any other technology that has the potential to put data sovereignty in the hands of individuals, and claw back the dysfunctional model for Facebook and Google of mining user data for advertizing (or the Chinese variant with no real privacy protections).
In 1998, Paul Krugman predicted the growth of email and internet-related communications would slow down, because he couldn’t imagine what people would have to say to one another. “By 2005 it will become clear that the internet’s impact on the economy has been no greater than the fax machine’s.”
This doesn’t make Krugman a bad guy or an idiot, and it doesn’t invalidate his vast body of work. But it does mean he was wrong about the internet as a network of information.
My guess is the same will go for Roubini: he is right about ICOs, but wrong about decentralized computing’s potential as a network of value. He’s not a bad guy or an idiot, and honest proponents of blockchain will recognize they need to answer his attacks. Many people are working on trying to get the industry regulated, but that’s not enough. It’s incumbent upon blockchain developers, investors, and users to prove Roubini wrong.