Earlier this year, Overstock.com CEO Patrick Byrne bought a $500,000 bond issued by the online retailer.
But it wasn't the size of the bond that drew attention. It was the way it was sold: through Overstock's new digital trading platform T0, which records securities trades through the same shared ledger technology that backs cryptographic currencies like bitcoin.
Advocates of the digital ledger technology, known as the blockchain, say its use in traditional financial transactions could prove far more revolutionary than the alternative currencies for which it was first created.
Storing transactions in one automatically shared, tamper-proof database could eliminate the need for complicated procedures and clearinghouses now used to make sure banks have their records in sync, saving time and money and reducing the risk of error.
And that has startups and established companies lining up to design the trading platforms, programming tools, and other infrastructure that'll soon be needed if blockchain-based finance goes mainstream.
"The main event isn't bitcoin," says Byrne. "It's using the blockchain to disrupt other industries and Wall Street."
T0 takes its name from its ability to finalize transactions essentially instantaneously. Traditional stock and bond sales take three business days—or T+3 in industry jargon —to fully settle, before the assets are available in the buyer's account and the purchase value is available to the seller.
Those traditional transfers are generally logged by an industry body called the Depository Trust Company, created in the 1970s to track the ownership of securities. But with blockchain-based securities, records are automatically shared between all of the participants in the market in close to real time, so the transaction's effectively settled as soon as it's recorded to that collective ledger, without the need for an intermediary like the DTC.
"What blockchains do is allow for a transaction to occur between financial institutions or people but without the need of that intermediary, because you can reliably send the assets to each other very quickly and in basically near real time and without any chance of having that transaction be reversed or changed," says Peter Shiau, the cofounder and CEO of financial blockchain startup Blockstack.io.
The blockchain concept was first developed by bitcoin's pseudonymous (and mysterious) creator Satoshi Nakamoto as a way to track ownership of the digital currency. When bitcoin owners spend or transfer the currency, their digital wallet software publishes a cryptographically signed record of the transaction to the global network of bitcoin users.
Bitcoin miners are then rewarded with new bitcoin for bundling those individual records into groups called blocks and recording them to a permanent shared ledger called the blockchain. Only blocks satisfying certain mathematical properties in relation to the previous block on the chain are considered valid, making it effectively impossible to tamper with records of previous transactions.
But while blockchains were first created to track bitcoin, experts say there's no reason why they can't be used to track other types of assets, from securities to event tickets to old-fashioned U.S. currency.
"A blockchain's just a ledger—it just tells you who has what and that someone sent something to somebody else," says Shiau. "The ability to know with certainty that something happened is what's critical."
Overstock transferred the bond bought by Byrne, as well with a second "proof-of-concept"bond bought by financial firm First New York for $5 million, by recording the transaction to the actual public bitcoin blockchain, using a standard called Open Assets designed to record any kind of transfer to that shared ledger.
And other companies, such as Blockstack, are developing technology to let financial institutions create their own private blockchains, changing features from bitcoin and other cryptocurrencies to better suit their own purposes.
"If you take the software and install it in a private context, you can do whatever you want," says Shiau.
Banks might wish to adjust the size limits of transaction blocks, or the speed at which they're added to the network, security features, or other settings, or keep their blockchains only accessible to their trading partners over a secure network. They may also incorporate support for so-called smart contracts—essentially simple programs embedded within the blockchain that define rules for when assets get transferred, allowing procedures such as dividend and interest payments and escrow arrangements to be automated with logic stored in the same shared database as the assets themselves.
Some experts say the actual requirements banks have could be satisfied by data structures other than blockchains—once banks control who has access to a shared ledger, they could really rely on pre-bitcoin technology to achieve the same goals, argued Arvind Narayanan, an assistant professor of computer science at Princeton University, in a recent blog post.
Indeed, banks might be better off in many cases just sticking with traditional database technology, says Paul Chou, the CEO of LedgerX, a startup building trading infrastructure for bitcoin derivatives.
"If you restrict access to trusted parties, then an intelligently managed shared database is a fast and proven solution—there is no need for a blockchain," he wrote in an email to Fast Company.
And if banks find they won't benefit enough in terms of efficiency from blockchain-based systems, switching likely won't be worth the risk, he argues.
"Imagine how terrifying it is to design a whole new system from scratch with a datastore that has only been around for a few years," he wrote. "Only if banks find a technology that can improve things by an order of magnitude, say 10 times more efficient, will it make sense to totally abandon a slow, but proven platform to gain those efficiencies.
Still, Princeton's Narayanan wrote in his blog post, the excitement around bitcoin and the math that makes it work may have motivated banks to start thinking more seriously about updating some aging technology and processes.
"In my view, it's not the novelty of blockchain technology but rather its mindshare that has gotten Wall Street to converge on it, driven by the fear of missing out," wrote Narayanan. "It's acted as a focal point for standardization."
And banks may end up relying on different blockchains for different purposes, says Makoto Takemiya, the chief blockchain officer at Japanese startup Mijin, which plans next year to begin beta testing its private blockchain technology with industry partners and release open source code.
"I think there's going to be room for multiple blockchains that institutions can have," he says.
One London-based startup, called Everledger, is developing blockchain tools for the diamond industry, designed to log information on ethical sourcing, insurance claims, and theft investigations, associated with detailed digital descriptions of the valuable stones. Private information on police investigations or private transactions will be stored on a private blockchain, and information like diamonds' conflict-free certification will be logged to the public bitcoin blockchain, says CEO Leanne Kemp.
"You would be able to identify whether those diamonds are truly certified or not certified," she says.
Everledger participated earlier this year in an accelerator program backed by Barclays, the giant U.K.-based bank, as large financial institutions around the world are beginning to seriously investigate the benefits of blockchain technology.
Financial technology firm R3 announced last month that it's working with 22 major financial institutions, including Barclays, Goldman Sachs, Deutsche Bank, and Mitsubishi UFJ Financial Group, to develop and test software and standards the banks can use to build blockchain-based services. The firms have looked at a number of potential use cases—from issuing and settling the transfer of securities to transferring the rights to loans—but they're focusing first on devising lower-level protocols on which those services can be built, says R3 CEO David Rutter.
"We kind of think setting the foundation first is the right thing before we start worrying about the wallpaper in the bedroom," says Rutter.
R3 aims to develop, and potentially make open source, basic blockchain protocols analogous to telephone or Internet networking standards upon which financial services tools can be built. And, Rutter says, they're working closely with the banks to make sure those firms' needs are met.
"Let's pause the mad rush for a second, and make sure that we have the basic requirements that are needed to build this technology, especially if we think that it's going to be a solution that's going to provide benefits for the next 10, 20 or 30 years," he says.
That's not an unrealistic requirement, when the current system of securities settlement, revolving around the DTC, dates back to 1970s reforms led by Congress and the Securities and Exchange Commission. It, too, evolved from a push for efficiency, after increased transaction volume proved too much for the paper-based stock certificate system it replaced: A Wall Street "paperwork crisis" had led to increasing numbers of errors, canceled trades, and millions of dollars in hard-to-catch fraud.
Rutter says it's still too early for the company to have had detailed discussions with regulators like the SEC, though he says a move toward blockchain could make their lives easier as well, since it'll be easier to analyze transactions recorded in one shared ledger rather than looking at disparate databases across multiple financial firms.
"You can build software that would allow regulators to peek at that data in a way that would make sense for them in a more timely fashion and with absolute confidence that what they're looking at is a clean picture," he says.
As far as Overstock, which has said it plans to offer blockchain-recorded securities to a wider audience, Byrne says the company's had productive talks with the agency so far.
"I think," he says, "that they're gonna let the crypto-revolution occur."