From an early age David Schwartz was obsessed with doorknobs and locks. At 5 he would use a screwdriver to dismantle them, removing them from household doors, so that family photos show gaping holes where the knobs should be.
“The function of a doorknob is to control the movement between two spaces. It’s a gateway, a barrier, and obstacle. It’s a control,” says Schwartz. “When that barrier goes away, you understand it. It sounds silly, but to me it might as well have been magic.” At age 48, Schwartz, bearded and balding with shoulder-length hair, is the Gandalfesque wizard of San Francisco’s Ripple and co-creator of the third-most-valuable cryptocurrency, XRP. As the company’s newly appointed chief technology officer, Schwartz is on a mission to dismantle and reassemble one of the biggest gateways on the planet—the one that connects nearly every bank in the world so that trillions of dollars can move from account to account.
In short, Schwartz wants to disrupt SWIFT, the Society for Worldwide Interbank Financial Telecommunication, a Belgian cooperative organization founded in 1973 that counts more than 10,000 financial institutions as its members and is the ultimate middleman in banking. Let’s say you want to transfer $5,000 from your account at JPMorgan Chase to your cousin in La Paz, Bolivia. Chase sends a secure message via SWIFT’s vast network to the receiving bank in Bolivia. Ultimately this results in foreign currency exchanges and a transfer of funds, often via numerous correspondent member banks. The process involves regulatory oversight, compliance checks and other layers of protection, and is facilitated via a complex series of service agreements. SWIFT handles approximately 25 million such messages per day, and the result is an estimated $6.74 trillion in transfers. However the SWIFT network is far from efficient. In an era when secure emails are transmitted instantly and the bitcoin and ethereum blockchains move millions in minutes, most international money transfers take at least three days to settle, with fees that are opaque and varied.
Under Schwartz’s technical leadership, Ripple would like to bring global money transfers into the 21st century. The company has recruited hundreds of the world’s largest financial institutions—from UBS and BBVA to American Express and the Bank of Indonesia—to test Ripple’s new tools designed to modernize the way they move money. But Schwartz, who just returned from a tour of Europe’s banks, faces many obstacles. There are detractors inside the cryptoverse who doubt that Ripple’s technology is what the company says it is — decentralized and not controlled by a single authority. And if persuading bankers to switch to the largely untested Ripple platform wasn’t difficult enough, the company also faces competition from a host of other startups, including Stellar, a blockchain money-transfer firm formed in 2014 by one of Ripple’s founders.
“We want to create a payment network like SWIFT. But one where the settlement, the actual movement of money, the actual plumbing underneath the surface, would be a decentralized, open network,” Schwartz says. “The endgame is just money moving invisibly, as easily as information.”
Schwartz’s background mirrors that of many of today’s technologists. As a precocious preschooler in suburban Long Island in the 1970s, he began programming his father’s Texas Instruments and Hewlett-Packard calculators, mostly to create pictures on their tape readouts.
In high school he was the prototypical nerd: Sports weren’t his thing, but chess was. In 1990 he received a degree in electrical engineering from the University of Houston. The following year he was awarded his first patent—20 years before Satoshi Nakamoto invented the bitcoin blockchain—for a distributed computer network he designed to lighten the burden of a central processor. In 1992, Schwartz and his father, a doctor of internal medicine, cofounded a medical technology company that developed a noninvasive device for recording data about heart murmurs. The product didn’t sell well, but demand for programmers rose during the dot-com era, and Schwartz pursued a number of network-related programming positions. Meawhile he was getting interested in cryptography. In 2001 he landed at a company in Santa Clara, California, called WebMaster Inc., where he helped design an early cloud-based storage system. During his decade there he also consulted for the National Security Agency (NSA), helping integrate the agency’s networking software with the existing security and public key infrastructure technology. In other words, Schwartz was gaining a working knowledge of cryptography at a high level. “It was a fantastic experience,” Schwartz says.
Along the way, Schwartz developed an online persona, JoelKatz (the name was inspired by Stimpson J. Cat from The Ren & Stimpy Show), to pseudonymously publish his philosophical meanderings. His widely read JoelKatz blog is subtitled “Democracy is vulnerable to a 51% attack,” an allusion to the tipping point at which one party could gain majority control of a cryptocurrency, no longer making it decentralized. @JoelKatz, his Twitter handle, has more than 100,000 followers.
In early 2011 Schwartz was looking for something new. Crypto-anarchists were beginning to explore the bitcoin blockchain as a way to avoid central oversight. While Schwartz doesn’t identify as a libertarian, he had purchased a few bitcoins and agreed with many of the movement’s ideals. He was particularly troubled by the centralized control of money.
“If no bank will do business with me, I don’t get a hearing in a court of law, I don’t get to read the law, I don’t get to confront my accusers. They’re enforcing the law in a way that doesn’t have any of the normal protections that law enforcement is supposed to have. And that really, philosophically, bothers me,” Schwartz says. “This idea of disintermediating these shadow regulators that are not democratically accountable and are not elected but act as policemen kind of resonated with me. That got me into the bitcoin community.”
Around this time Schwartz met Jed McCaleb, the founder of the Mt Gox bitcoin exchange and an early Napster competitor called eDonkey 2000. They met at a coffee shop, where McCaleb shared an idea he had called NewCoin. By the end of the conversation, the two had decided to see if they could build a financial infrastructure similar to bitcoin that would use far less energy and drastically reduce transaction time.
“My objective initially was just to see if it was true,” says Schwartz, who started working on the code that would eventually underpin the XRP cryptocurrency. “And then, probably a month and a half later, we reached the point where I proved to him that yes, this would work, it would be possible, but we didn’t know what it would be good for. It was like inventing a new material. Is it really light? Is it really strong? Is it manufacturable? Is it durable? Does it rust? And then, once you have all those properties, is there some use case for that?”
McCaleb and Schwartz joined forces and soon brought on another programmer, Arthur Britto, to help Schwartz finish XRP’s technical architecture. In 2012, while Schwartz and Britto wrote code, veteran tech executive Chris Larsen, who previously worked at Prosper Loans, joined Ripple as its first chief executive. Larsen quickly set out to recruit hundreds of global banks to test early versions of Ripple’s technology. In 2014, McCaleb became disgruntled and departed, “forking,” or copying, XRP’s code to start rival venture Stellar.
Schwartz went on to help build two financial tools that would become Ripple’s first core products: xVia, a payment interface designed to let users send payments globally with increased transparency, and xCurrent, enterprise software that lets banks initiate and settle transactions. By moving the transactions to a shared, distributed ledger that only permissioned users can access, Ripple claimed to be able to facilitate transactions in as little as a few seconds instead of days.
It didn’t take long for banks to see the potential in what Ripple was creating. In 2015 Ripple formed the RippleNet Committee, an advisory team made up of major banks including Bank of America Merrill Lynch, Japan’s MUFG Bank, Standard Chartered Bank, Westpac and Spain’s Banco Santander, which invested in Ripple and began experimenting with early versions of its payment technology.
Banks’ interest in Ripple coincided with the general excitement bubbling up around bitcoin in 2016 and 2017. Despite the fact that none of Ripple’s early banking partners were using XRP cryptocurrency, its penny stock price, coupled with confusion over its actual relationship to Ripple, caused XRP to soar on crypto exchanges to a peak of $3.65 in January 2018, up from $.006 only a year before. This gave the untested currency a market value of $140 billion. Larsen, who received 9 billion XRP tokens as its CEO, saw his crypto holdings rise to as much as $60 billion. ronically, Schwartz, Ripple’s most zealous ambassador, opted for a salary and a 2% stake in Ripple, instead of the XRP cryptocurrency he helped create. To this day Schwartz is not listed as a cofounder of Ripple, despite being employee number two and its chief architect. With Ripple’s value at $4.7 billion (and XRP’s market cap at $13 billion), Schwartz’s net worth is estimated to be about $90 million.
Forr a company on a mission to become the connective tissue for global banks, the location of Ripple’s ultramodern headquarters in San Francisco’s financial district couldn’t be more auspicious. In order to reach the offices, one must first pass through a grand archway that resembles something from Italy’s Medici era. On either side are the offices of Bank of America and U.S. Trust. Schwartz’s centrally located open-plan desk on the second floor is austere—two flat-panel displays and a black ergonomic keyboard are all you will find.
To date, Ripple has introduced three important products: xVia, xCurrent and the newer xRapid, which aims to address an age-old problem facing international banks. Most large banks are forced to maintain local currency accounts around the world for use during money transfers. xRapid frees up this capital and lowers costs by substituting the local currencies with the XRP cryptocurrency. Importantly, while only permissioned institutions can use Ripple’s products, xRapid is designed to make it easier for banks to use the XRP cryptocurrency, which anyone can buy, and the XRP platform on which anyone can build. In this way Ripple hopes to transform XRP into a decentralized reserve currency for international banks.
The adoption of Ripple products has been modest so far. Spain’s Banco Santander, which happens to have a seat on SWIFT’s board of directors, has launched a retail mobile app called One Pay FX using Ripple’s xCurrent payment product. The smartphone app allows customers to move money between four pilot countries—Spain, the United Kingdom, Brazil and Poland.
Ripple’s xRapid product has also won some converts. Mercury FX, a London-based foreign-exchange firm that offers clients an alternative to banks when sending and receiving international currencies, will soon move from its pilot xRapid program to live production.
“SWIFT has had the monopoly for so long,” says Mercury CEO Alastair Constance, “Why haven’t they been passing down the cost and time-saving? The answer is that inefficiency and laziness has made them a lot of money.”
Ripple isn’t the only blockchain startup that is after SWIFT. Banco Santander, for example, is working with two other financial infrastructure blockchain platforms, Hyperledger Fabric and Ion, to explore other applications. “Like a lot of banks, we have to place different bets in different areas,” says Santander managing director John Whelan.
The competition has jolted SWIFT into action. Last year it launched its Global Payment Innovation (GPI) initiative in an effort to make its payments real-time. As of June 2018, some 180 banks were using this encrypted, non-blockchain alternative to transfer about $100 billion in cross-border payments per day.
“GPI isn’t a rebuild of their entire back office,” says Harry Newman, SWIFT’s global head of banking. Fabian Vandenreydt, a former head of SWIFT Global Securities, adds that the issue isn’t whether Ripple’s technology works but whether it saves banks more money than it costs.
He mentions a project completed in 2015, when the European Central Bank launched a new platform for linking 20 central securities depositories (CSDs). The process took seven years and was estimated to cost $400 million.
“It’s a bit like heart surgery,” Vandenreydt says. “You need to move things so that the system still works during the transition, and that’s where the cost is.”
Schwartz dismisses such objections: “When you try to update a legacy system, you tend to be pressured to keep things as opposed to replacing them. I think the clean-sheet approach is almost always going to produce a better design at a lower cost.”
With more than 10,000 banks in SWIFT’s network, Newman is hopeful that he can transition all of SWIFT’s members to its new system GPI system by 2020. While GPI isn’t necessarily a Ripple-killer, it presents another serious hurdle for Ripple to clear.
In May, as if to acknowledge his being overlooked in Ripple’s original creation story and hierarchy, Schwartz was discreetly named chief technology officer of the company he helped build from scratch.
As CTO, Schwartz will report to CEO Brad Garlinghouse, but in terms of Ripple’s technical vision he’s in charge. “David’s not a guy who is going to require a lot of micromanagement,” Garlinghouse says. Chris Larsen, Ripple’s executive chairman, adds: “He’s, if not the soul, he’s a key part of the soul of what we’re trying to do here.”
Ironically, Schwartz’s number one priority is persuading the blockchain community and potential customers that his team at Ripple is losing control of the technology they built.
In a bizarre blockchain era twist, Ripple is accused of being as centrally controlled as SWIFT itself.
The debate centers around trust. While the XRP blockchain was designed to be open to anyone and therefore independent and trustworthy, Ripple has historically had disproportionate influence over its governance. On the bitcoin and ethereum blockchains, for example, validation comes from independent miners vying to confirm new blocks of transactions in exchange for the cryptocurrency. The founders of Ripple, by contrast, created all 100 billion of the XRP tokens at once in 2011. They sell the coins periodically and have distributed many to insiders. In fact, more than half the XRP that will ever exist are still owned by Ripple.
All the transactions recorded on the XRP blockchain are confirmed using a consensus system comprising groups of validators that analyze the network’s transactions. Those validators, or nodes, are in turn organized into groups that trust each other called UNLs, or “unique node lists.” While validators get to choose their own UNL, the list assembled by Ripple is the default, creating a possible area of centralization.
To help offset concerns that the company could flood the cryptocurrency market or manipulate prices, Ripple has locked up its XRP into smart contracts that hold the currency in escrow, temporarily releasing 1 billion tokens a month. But to truly decentralize the system, Schwartz is urging others to build on the XRP blockchain the way ethereum has. “You don’t need our permission, and we can’t stop you,” Schwartz says.
As for the validators, Schwartz claims that only 10 of the 150 currently supporting the network are managed by Ripple. For bitcoin, about 58% of transactions are processed by four mining pools, mostly in China. Some 57% of ether production is controlled by three mining pools. What remains unclear (by design) is how many of the “non-Ripple” validators are relying on the UNLs Ripple has actually deemed to be trustworthy. In other words, there is worry in the crypto community that, like SWIFT, Ripple’s system remains more centrally controlled than it appears. How many transactions rely on Ripple’s sanctioned validators? “We don’t know” is the official response from a company spokesperson.
Ironically, Schwartz himself may be Ripple’s best weapon to prove that the XRP blockchain is actually decentralized, and thus trustworthy and secure. Credibility is a precious asset in cryptoland these days, and Schwartz, little-known and passed over during the XRP windfall enjoyed by his peers, comes off as something of a cryptocurrency Eagle Scout. “My personal fortune is aligned with the success of the company and its products,” Schwartz says bluntly. “If it’s the best solution, use it. If not, why do I want to trick or force people to get a substandard outcome?”