As the summer of 2016 turned to fall, the DAO fiasco and the hard-fork drama receded, albeit slowly and not without controversy, into the background. Ethereum's annual developer conference approached, and this time it would be held in Shanghai.
Corporate support for blockchain as a platform, which had started a year before with Microsoft coming aboard as a lead sponsor of DevCon 1 in London, only grew at DevCon 2. The likes of IBM and R3, a consortium of all the world's largest banks that were now experimenting with blockchain, were major presences in Shanghai. The problem was, they were slagging off Ethereum, saying it couldn't be trusted for commercial applications.
Corporations and the financial world were now fully in hype mode about the benefits of blockchain to their businesses. The longstanding refusal of regulated companies to have anything to do with Bitcoin had passed now that blockchain was a buzzword. If a supply chain–dependent business like a car manufacturer could get all its subsidiaries and suppliers on the same network, on the same blockchain, it could save them millions of dollars a year. Imagine being able to pinpoint a breakdown in the supply chain with a blockchain, being able to reroute orders and supplies instantly. That kind of real-time analysis and communication was incredibly difficult to do with a disparate network of firms that communicated over different time zones with various people in charge of keeping the assembly line moving. Similarly, bankers were intrigued by the potential of creating a private network for all the participants in a market like credit derivatives. That could cut the time it took to settle trades from days to minutes and save the banks millions of dollars in the bargain.
To form these private networks, banks and corporations didn't need to use the public blockchain systems that had made Bitcoin and Ethereum successful. There was no need for JPMorgan and Bank of America to use a proof-of-work system to mine blockchain transactions because they already knew each other. A proof-of-work system is only needed when strangers are interacting. It injects trust into a transaction where the parties don't trust each other. JPMorgan and Bank of America, on the other hand, already trade billions of dollars' worth of financial products between themselves every day, both for the bank's own account and on behalf of their customers. People began applying the term distributed ledger instead of blockchain to this kind of transaction system.
Distributed ledger projects were being created by IBM and R3 in the fall of 2016, and their executives had come to Shanghai to pitch them. Jeremy Millar had been involved with ConsenSys since April of 2016, and he became convinced that Ethereum needed a commercial enterprise arm to complement the foundation.
John Wolpert from IBM and Richard Gendal Brown from R3 presented at DevCon 2, “both of which had slides in it that basically said companies can't trust Ethereum, it's a fringe open-source project that can't be trusted for commercial work,” Millar said.
Microsoft's Marley Gray was in Shanghai and remembered the IBM and R3 presentations. “IBM was particularly heavy on the FUD,” he said, referring to the acronym for “fear, uncertainty, and doubt” that crypto people use as shorthand for anyone criticizing their work. “I still give Jerry some grief about that,” he said, referring to Jerry Cuomo, IBM's VP of blockchain technologies.
It was clear in Shanghai that Ethereum needed to create its own commercial development arm aimed directly at banks and corporations, or someone else would steal that market right from under them.
Joe Lubin, Vitalik, Jeremy Millar, Marley Gray, Alex Batlin, and Andrew Keys were among the people in Shanghai who had the first conversations about what would become the Enterprise Ethereum Alliance. About 10–12 people initially joined the group, which Joe bankrolled until membership dues were enough to pay the bills. Marley Gray offered the Microsoft offices near Times Square for EEA meetings.
As always when Ethereum is involved there were political minefields to navigate, and the early enterprise team discovered one right away. The first name it came up with for itself was the Enterprise Ethereum Foundation, which went over like a ton of bricks with Ethereum Foundation leadership. There was already bad blood between Ming Chan and Joe Lubin, and here Lubin was backing a commercial project that seemed to be trying to take the very name of the Ethereum Foundation for itself.
The Ethereum codebase would need some work as well if it was going to appeal to businesses. This was the early advantage IBM's Hyperledger project and R3's Corda blockchain had over Ethereum.
Marley Gray said Ethereum was under pressure from other enterprise blockchains like Hyperledger that had better privacy controls and performance. “Corda was starting to make some noise. We felt like if we didn't do something…,” Gray said.
An organization that was doing something was JPMorgan. Its Quorum enterprise blockchain project was in its infancy, but to Marley and others, the fact that JPMorgan was willing to throw its lot in with Ethereum only strengthened their belief that the EEA would receive broad support.
One of the earliest JPMorgan employees to work on Quorum (back before it was called Quorum) was Amber Baldet. She'd always liked to build things, to create: Back when she was a kid she'd watch and sometimes help as her dad worked on their car engine or replaced their roof with the help of Amber's brother. She spent a lot of time in the garage with power tools and made jewelry out of spent bullet casings. The home improvement network HGTV used to show This Old House and Amber couldn't get enough of it. “I learned how to install a lot of plumbing,” she said.
Her love of technology goes back to when she was 13 and persuaded her parents to let her use a 9600-baud modem left behind by a deceased relative. The experience was life changing. Baldet connected to university message boards, whose members didn't know they were interacting with a wee teenager. The respect she was able to garner from them was a huge confidence boost.
Baldet first encountered finance while working at a small arbitrage brokerage in her home state of Florida. She'd started to code when she was 10 years old and rewrote part of the brokerage's middle-office software to let the salespeople and research people know what was working for their clients. She thought a career as a secret agent, like in the CIA, was in the cards, but it didn't seem to be a good fit. What she really wanted to do was move to New York City and work on Wall Street.
She found her way to JPMorgan after working at a consulting gig and being hired by Vikram Pandit's hedge fund, Old Lane, in 2008, just as the financial world was about to implode. In 2009, she discovered Bitcoin, so a few years later when the banking world was starting to talk about blockchain she felt she had a good background to help JPMorgan sort through this new technology.
The team Baldet joined was known at first as Gemini, which oversaw several avenues the bank was pursuing. One area was strategic partnerships, like the investments JPM had made in startups Digital Asset Holdings and Axoni. Another was the issue of using public blockchains for business, which is problematic because public blockchains reveal too much information for businesses to feel comfortable using them. To address the latter issue JPM could try to use Ethereum – if Ethereum could be tweaked to be more private – or go with R3 and its Corda blockchain or build its own internal blockchain from scratch.
Since no one at JPM knew what the best path to take was, the company decided both to build its own blockchain version, known as Juno, and to tinker with Ethereum to see if it could be made suitable for the regulated banking world. “We had an internal bake-off,” Amber said.
The problem that Baldet had to solve to appease her bosses was that Ethereum, like Bitcoin, is public. Every transaction is viewable on a blockchain explorer like Etherscan, and banks can't have that level of transparency because it could expose their customers' private information. There had to be a way to get the benefit of a network like Ethereum but to also shield the details of those customers' transactions.
Christine Moy was Amber's first hire in the blockchain unit, and she helped during the bake-off. Patrick Nielsen came in to help the team with a crucial idea about how to get the needed level of privacy. “He and Amber and a couple others on the team architected this private contract store node that would sit in parallel with the Geth node,” Christine said. The private node would store all the private transactions, which would be encrypted and then hashed so they could be uploaded to the public version of the bank's blockchain.
This was huge and solved the privacy issue. In essence, all you could see was an event log. It would show that something had taken place, but unless you had the private keys to that transaction you couldn't access any more detail. The name Gemini came from the twin servers, one private, one public; JPM eventually changed the name and introduced the project to the world as Quorum.
Baldet and her team worked with Jeff Wilcke and some other Ethereum Foundation members to get Quorum up and running. Whereas support from Microsoft had helped establish Ethereum as a serious player, JPMorgan basing its entire blockchain model on it was a vote of confidence at a whole new level. The financial world was beginning to pay serious attention to Ethereum and what smart contracts could do for business.
Alex Batlin played an important role in the early adoption of Ethereum by the financial world through his position running UBS Labs, a fintech-focused unit at the Swiss bank. His first foray into Ethereum involved creating a “smart bond,” where a token on the blockchain represents corporate debt. Batlin named the blockchain project at UBS “Pathfinder,” because in the early days they had no idea where this would lead.
“The public Ethereum blockchain absolutely makes a lot of sense, but if you're going to be trading security tokens between regulated banks then you didn't need to have the burden of proof-of-work,” Alex said. As each member of the bank-trading network would be known to each other, UBS used a system called proof-of-authority, which doesn't require an ungodly amount of electricity to maintain.
There were three big areas where Ethereum, or some version of it, could specifically help the financial world, Batlin believed. The first was in doing away with what's called issuer risk – in this case, let's use the case of the issuer of government-backed currency. In the China-US trade relationship, for example, most debt is held by China in the form of US dollars. “They're now not just subject to credit risk, i.e., the US going bankrupt, they also have an issuer risk,” Batlin said. “If the US were to suddenly print a huge amount of new money, effectively they would devalue any loans a third party has against them which hold them in dollars.”
He brought up the fact that gold has been used as a substitute for fiat currency when there is a lack of trust in the government issuing the currency. But on a practical level gold is hard to deal with, and you might not get it back from the US government if it happens to be holding it for you. “Where do you sell the gold?” Alex said. Cryptocurrency sidesteps both of these concerns.
“A zero physical-settlement risk currency, with no issuer risk, that's a pretty good idea,” Batlin said. “It's more applicable in international settings rather than national.”
A second point is that cryptocurrency provides a way to pay people more directly and to incentivize them. The incentive can build on itself if a certain blockchain network rewards its users and developers. That way the cohesion grows and the network effect can be multiplied.
Lastly, there is the thorny problem of national interests in securities markets. For understandable reasons, perhaps, most countries have centralized control over their own domestic stock markets and the associated back-office settlement procedures that are arguably more important. That makes it difficult to sell shares across the world because business in London has to be reconciled with US-based business, and Asian share purchases have to be reconciled with sales of shares in the Middle East. You get the idea.
“A distributed ledger technology, or blockchain, is perfect because it's both local and global, so you no longer need to reconcile between nations,” Batlin said.
“My view is this is a natural evolution,” he said. Paper money worked well when most trade was regional – within the Eastern Seaboard of the US, for example. When trade went global, however, paper money slowed down transaction times.
Or, as the EEA's Jeremy Millar put it, “Global business needs global networks.”
JPMorgan took this idea seriously and soon realized that just sticking a blockchain into an existing financial market only adds another layer of complexity, often without improving efficiency. “But what if we built a new debt instrument from scratch on a blockchain?” Christine Moy said. “That's where the cash token was born, or JPMCoin was born.”
This was the banks' way of moving money around its global locations: from Frankfurt to Auckland in less than a second, using a digital representation of the bank's deposits. While cash has been electronic for a long time, what's different here is that all the recipients of JPM's money are on the same network. In the traditional model, my bank sends another bank money electronically, yet this has to be recorded in each set of books at the bank. This takes time, up to three days in many markets right now, and can be error prone. In a digital system, the sender and receiver are connected via the same payment rail, so the debit and credit are instant on each set of books. This is not only faster, it saves the banks money as they are required to set aside some of their funds for the duration of a trade in case it fails. The syndicated loan market can take more than 15 days to settle. The money to be saved there is immense, and only one small example.
Christine Moy had an already varied and probably exciting career for a banker under her belt when she joined the bank's blockchain effort. During the financial crisis of 2008 she'd helped to renew credit lines for the big three US automakers so they wouldn't go bankrupt. She traded commodities under Blythe Masters, helping North American mining companies hedge their risk. Then she went to JPMorgan's private bank, where the idea of digitization was beginning to take hold. She'd been told that the bank's loss given default calculator, which is used to assess loans given to private bank clients, couldn't be automated. So she automated it.
“At that time, it felt like I'd done all the things at JPMorgan,” she said. Then she saw the job posting Amber Baldet put out where she asked for a Twitter handle, and Christine knew this was not just any other position within JPM. “It was the most mysterious job post,” Moy said.
The focus on blockchain was all happening at the bank because of Sanoke Viswanathan, who was then JPMorgan's chief administrative officer of the corporate and investment bank. Moy referred to him as the blockchain godfather at JPMorgan. He set the permissive tone needed to experiment with emerging technology at the staid and bureaucratic enterprise that is JPMorgan. “It was really important because you need the culture to change to be open to new technologies,” Moy said. We were in a conference room in the bank's headquarters on Park Avenue, and as I took notes Moy realized I was having trouble with the spelling of Sanoke's last name. She grabbed my notebook and wrote it out for me. I remembered Vitalik looking over my notes once, too, and helping me correct the spelling of Mihai Alisie's name. But he'd just spelled it out for me; Christine took matters into her own hands. That's Wall Street, baby.
And even though this was the first time I'd met Christine Moy, our paths had kind of crossed before. She mentioned a 2015 cover story in Bloomberg Markets titled “Blythe Masters Tells Banks the Blockchain Changes Everything.” That story had a big effect on the street and helped change the philosophy of banks on blockchain, Moy said. I'd written that story with my colleague Edward Robinson. It was one of those small, neat times like in When Harry Met Sally when a writer can tell someone, I wrote that!
JPMorgan is thinking about this paradigm shift in a 5- to 10-year time frame. After I spoke to Christine in the JPMorgan headquarters in Manhattan, I noticed something in the lobby of their building. The Starbucks there no longer accepts cash.
●●●
In late February of 2017, about 200 executives, coders, and developers gathered in the downtown Brooklyn office of JPMorgan. This was the official launch of the EEA and it was to be an all-day affair. The founding members of the group showed the breadth that had been achieved in just a few months from the initial idea for the EEA in Shanghai. Tech giants Microsoft and Intel were among the founding members, as was CME Group, the world's largest futures exchange; BNY Mellon, Banco Santander, Credit Suisse, UBS, and JPMorgan represented the financial sector, with accounting firm Accenture among the founders as well. Oil behemoth BP was in the mix, as was Thomson Reuters.
I attended this event for Bloomberg News and remember a glitch at the beginning that seemed a perfect metaphor for Ethereum at the time. Vitalik wasn't in Brooklyn that day but had recorded a video message that the EEA wanted to play to kick the meeting off. Except they couldn't get it to work: the computer hosting the message had crashed. A frozen image of Vitalik's face hung on the screen in the conference room, then disappeared. As people tried to get the video to play, Amber Baldet stood up and took control of the room. She was funny and forceful in a way that quieted everyone immediately. After about 30 minutes they got the video to play.
As the world was waking up to Ethereum, it still had a long way to go. It was slow, first of all, in a performance sense. Then too it had to convince regulators that it was safe and capable of replacing critical infrastructure in the corporate and financial worlds. These were big hurdles, but there were also signs that once stuffy and staid corporations were willing to give this a shot.
One example of this was how Baldet and her team convinced the JPMorgan higher-ups to allow Quorum to be open source. JPMorgan had never done an open-source project before: Why would a bank work hard to create a software product and then give it away to the rest of the world? It was about as antithetical to banking as publicly declaring your customers' positions in the middle of Broad Street.
“The good thing about approaching this with a blockchain project was, it's not an argument, it's an easier narrative to build that we're giving away the operating system, but charging for the apps,” Amber said. “We'll be able to introduce a number of products to market that are monetized that sit on top of it. So that resonated and made sense to people.”
It also led to interactions like the Central Bank of Brazil getting in touch with Amber to let her know it had downloaded Quorum and was experimenting with it for the bank's real-time gross settlement processes.
In a larger sense, though, while the EEA was helping establish Ethereum as fit for business, Amber and a lot of other people involved with the group wanted enterprise blockchains to lead to a better public blockchain system. The hope was that, like in the early days of the Internet, private intranets would one day merge with the public Internet. If in business or on the public chain, many in the Ethereum community wanted to move the ball in the same direction.
I asked Amber about these ambitions at the EEA launch in Brooklyn.
“I don't want to build AOL,” she said. “I want to build the world wide web.”