Bankers eagerly await the Hong Kong IPO of Bitmain, the dominant provider of computer hardware for cryptocurrency miners which had $2.8 billion of revenue in the first half of 2018. But as they wonder what valuation public investors will eventually put on a company that has already raised $872 million in private funding – with a $422 million venture round completed in August that drew in Temasek Holding among others – let’s not forget Argo Blockchain.
Crypto analysts have at various times in recent months talked up a valuation for Bitmain, which was founded in 2013 and is headquartered in Beijing, of anywhere from $15 billion to $50 billion.
Founded in December 2017, headquartered in London but with most operations in Canada, Argo is a much smaller company, with a market cap a mere fraction of that. But it has already listed on the London Stock Exchange main market, raising £25 million through a placement in August 2018, to build out a new platform.
We’ve had software-as-a-service, banking-as-a-service and blockchain-as-a-service.
Argo now offers mining-as-a-service.
It is in the process of leasing state-of-the-art blockchain mining hardware to build up racks of servers while selling subscription packages for retail users to rent this industrial-scale mining capacity and gain payment in mined cryptocurrencies.
These subscription packages range from $49 per month to $499 per month and allow for mining initially in Bitcoin Gold, Ethereum, Ethereum Classic, Zcash, Komodo and Horizen, with Bitcoin itself due to come in from the start of November.
Two things strike Euromoney immediately. First, the people behind Argo Blockchain are obviously making a bet on the growth of cryptocurrency, but they are not mining for their own account and this is not an investment scheme for crypto assets. Instead, they prefer to let subscribers take the risk of volatility in crypto while themselves being paid in hard, fiat currency, thank you very much.
Although any business only succeeds if its customers believe they derive benefit from using it, that looks a little cute.
That successful public listing, coming just eight months after the company was founded, is as much a sign of investors’ faith in the serial entrepreneurs behind Argo Blockchain – particularly executive chairman Jonathan Bixby and executive director and president Mike Edwards – and their track record of repeatedly backing and successfully exiting tech start-ups, as it is a punt on the business model itself.
When Argo Blockchain floated in August, Bitcoin had fallen by 64% from its December 2017 high and Ethereum by 70%.
Ethereum has kept on falling since.
“The enormous amount of volatility around cryptocurrency and associated technology is not new to us,” Edwards tells Euromoney. “One thing we have learned is the enormous advantage of building a company at the end of a bull run, after the massive initial hype has died down. Everything is on sale and you can build strategically.”
The swift dash to raise capital was crucial to this. Edwards says: “We were investing personally in crypto from 2012, advising certain exchanges, while also holding, selling and trading crypto. And in 2014 we first began looking at the opportunity in mining. What became clear is that to succeed, you have to mine on an industrial scale. And that in turn raises three further requirements: access to substantial capital to acquire the technology; access to intellectual property to deploy it in a highly efficient way; and access to cheap sources of power.” He adds: “All of that is impossible for the individual.”
Argo Blockchain claims to be democratizing crypto mining, opening it up to the likes of you and me. “Your average person could not light up a mining rig,” says Edwards. “We consider ourselves fairly technical people and acquired a Linux-based rig but realized we couldn’t make significant progress without cheap power.”
This is where being Canadian helped. Edwards says: “The Canadian government has installed substantial hydro-electrical power generation – as have other governments, for example in Sweden – in the hope of attracting industrial users close to those dams. But that hasn’t happened. A lot of power is now being transmitted from Quebec to New York, for example. We can be local users of that green power, taking advantage of its low cost.”
On the hardware front, Argo intends to be flexible. Edwards says: “You see companies expending huge amounts of capital in one big go on new technology and then being stuck with hardware that is quickly out of date. Bitcoin mining is a very fast-moving space. We started with graphics processing units (GPUs), which are almost the Swiss army knife of crypto mining. We’re now predominantly using application-specific integrated circuits (ASICs).”
Crypto nerds tell Euromoney that these are small, fast and efficient crypto mining computers but that they lack flexibility to adapt to new algorithms for new currencies with a serious risk of redundancy if new cryptos really take off.
Edwards thinks Argo Blockchain has an answer. “New machines are coming out every six to eight months and we are seeing big shifts in processing power. Because we are an on-ramp for crypto miners with subscription demand ahead of our current capacity, we’re constantly adding new equipment. So, we’re not stuck with old kit. We tend to lease machines rather than buy them and aim to develop relationships with the leading chip makers to gain access to new machines early when they first appear.”
He adds: “We do believe that we are still in blockhain 1.0 but that we will be able to adapt quickly to new protocols.”
A big part of the business risk for Argo Blockchain, rather like for magazine publishers, comes down to the cost of acquiring new subscribers, their churn rate and the lifetime value of subscriptions. So, while the timing of Argo’s launch – when everything is on sale – may not be a great signal for Bitmain if it hopes to float this year, its founders are all-in on crypto.
The business plan is to go from the 400 or so test subscribers when Argo first listed on the London Stock Exchange back in August to 30,000 by the end of its first year of operations, 140,000 at the end of year two and more than a million after year three.
In October, Argo published a regulatory news release that it had exceeded its initial target of 3,800 subscribers after its first month.
“That was a Herculean effort but it has left us very confident,” says Edwards. He adds: “We are trying to define our demographic. It’s largely millennial, urban and it’s not in Asia, being mainly North America, Europe, South Africa and Australasia. There are a lot of studies and statistics of how many users have access to cryptos, some of which may also be interested in mining. Ethereum has close to 40 million addresses. There are projections that eventually as many as one billion millennials will use cryptos – though not all will mine, of course.
“We only need a fraction of that total. Coinbase has 20 million users. We could be the Coinbase of crypto mining.”
Plenty could go wrong, of course. The crypto fad could pass. While the theory persists that inefficiencies in the payments system add significantly to costs for merchants that they then pass on to consumers, the cohort of early adopters of crypto has not been followed by the mass of populations into broad adoption.
Money works well enough for most people, who would rather cope with inefficiencies than have their wallets hacked and wealth stolen with no bank or central authority to appeal to in a world where only code is law.
The existing financial system could improve its offering. Regulators could seriously clamp down on fake money.
Perhaps most pertinently, crypto could develop away from mining.
Mining is a function of the requirement for a proof of work on the bitcoin blockchain: blockchain 1.0. That’s what raises the requirement for huge computing power and energy consumption.
Ethereum has been working for a while now on proof of stake, requiring lower computer capacity and power with blocks confirmed by market participants with a sufficient stake in the system already that they will not try and double spend for fear of undermining the whole thing and erasing their own wealth.
This is blockchain 2.0, a more centralized version of the first, although huge computer power required for proof of work already hands effective power to a few big players in blockchain 1.0.
But even before blockchain 2.0 is fully up and running, already coming into view is blockchain 3.0, which depends on proof of trust and works on a new kind of distributed ledger that doesn’t involve blocks at all and so doesn’t require mining either.
Welcome to COTI, an Israeli company founded in May 2017, which is pushing what it calls a Trustchain, based on a directed acyclic graph (DAG) data structure instead of a blockchain. It hopes this will deliver the initial promise of cryptocurrency as a new means of payment that is scalable, low cost and so cheap for merchants and other users, especially for large numbers of low-value payments.
COTI Pay is the first internally developed application on the COTI platform to utilize a fiat-pegged stablecoin, the COTI-DIME. It targets both e-commerce and point of sale (POS) payments and is equipped to initially process tens of thousands of transactions per second, unlike bitcoin which can only process a handful of transactions per second.
COTI Pay provides fully customized white label wallets for consumers and merchants and can be integrated with online checkout processes through POS add-ons and debit cards.
Shahaf Bar-Geffen, chief executive of COTI, tells Euromoney: “The whole notion of distributed payments systems goes against everything that has been instilled into us. Even if we don’t think about it much, our intuition is that financial transactions will be recorded in a central register or server whose operator we can appeal to if something goes wrong. Yes, if you take out the middlemen in payments you can save substantial cost, but when you pay a merchant for a product and he doesn’t deliver, that’s when you remember why you want a middleman.”
Shahaf Bar-Geffen, COTI
COTI offers a buyer-seller protection service. Users can charge back a transaction in cases of fraud or any other dispute related to deals settled through the COTI payment system.
But the real breakthrough is the process by which initial trust scores are assigned to node operators and then how the system works off new payments transactions being confirmed through a cluster of chains of these system users rather than on blocks.
Bar-Geffen says: “Traditional blockchains remove the incentive to cheat by making it hugely expensive to infiltrate all blocks simultaneously and technically impossible. But as a result, blocks are difficult to approve in the first place.
“The directed acyclic graph is very different. The beautiful thing is that new transactions are not bundled together into one new block being confirmed every ten minutes or so. Instead, each new transaction coming onto the DAG needs to participate in validating two previous transactions already awaiting confirmation. It works a little like file-sharing – which I’m sure neither you nor anyone at Euromoney has ever done – but it means that more people adding new transactions makes the whole system faster, whereas the blockchain gets slower.”
Proof of trust is the new consensus mechanism behind this. (For more detail, read the white paper here.)
Not all customers or merchants using the system need to engage in confirmation.
COTI does use proof of work for spam protection and network participant incentivization for node operators. But this is a short computational operation, which node operators can complete from a laptop and should not be confused with the expensive proof of work employed in miner-based ledgers like Bitcoin, with racks of ASIC servers sucking their power from that nearby hydro-electric power station.
There are modest incentives for those who participate in the system fairly. Users, which include merchants, receive an initial trust score, which over time rises in recognition of good behaviour, including contribution of capacity to the system, but can also be cut on evidence of attempts to double spend or not sending goods in return for payment.
For this to work, these participants cannot be anonymous. The initial trust score, calculated by an underwriting algorithm, is in large part based on a stringent KYC/AML process, including previous records of fiat transactions and identity factors such as your age and where you live.
Bar-Geffen says: “What we are building solves for the lack of scalability and high transactions costs on blockchain. In addition, I have to say that I don’t believe payments in future will stay anonymous or pseudonymous. Most people don’t want to break the law and most merchants don’t want to deal with people who do. This trustchain consensus mechanism is a breakthrough in crypto. Payments and trust are interconnected and we believe that the payments systems of the future will need this.”
The development of such a consensus mechanism offers a new signpost for crypto technology moving into the regulated financial system.
When a new transaction is added to the network it connects to two previous transactions with similar trust scores. If these are low trust scores, more nodes must come into the chain to rise above a pre-specified aggregate threshold trust score and so approve a transaction, thus taking longer, raising the cost of confirmation and reducing the reward for doing so.
All this adds an incentive to build the trust score so as to gain credit for more transactions in shorter chains with fewer participants approved more quickly.
Look. No-one said any of this was going to be simple.
But if this version does take off, what need is there of mining?
Bar-Geffen says: “As a consumer or a merchant, you need no computing power. For node operators, we have vastly reduced the power required. I actually struggle to believe that in a time when we are developing hybrid and battery-operated cars to help the environment, we are proceeding with bitcoin mining which is so destructive.”
It’s not what Edwards wants to hear, presumably. What the founders of Argo and Blockchain share, however, is a strong sense that new technology will change how money and payments work.
“Our philosophy is that in 2019 the name of the game will not be the technology but its adoption. Yes, distributed ledger looks like a technology that people will use. But who is using it in the real world yet? To get adoption it needs to be simple and to solve an actual problem that people have. We are helping enterprises right now who might benefit from a new low-cost way to process micro-payments, for example publishers who may ask consumers to pay a small sum for a single article instead of a large sum for an annual subscription. We think we are close to having the digital technology to white label payments solutions for the largest companies in the world to process such micro-payments.”
The many doomsayers on cryptocurrency often make the point that governments won’t like it and will close it down. Whoever prints the money we use holds considerable power.
How does power balance today between governments and giant corporations who choose their tax domiciles in return for bringing jobs and get outraged when asked to pay their dues.
What if Amazon printed its own digital stablecoin currency: would you use it?