You (Probably) Don't Need Blockchain For It
By now you’ve heard about all the amazing ways Blockchain is going to change our lives for better - it will make the supply chain more reliable, banking settlement faster, commodities trading cheaper and government more transparent and efficient. Walmart wants to trace mangoes from farmers all the way to its customers and World Economic Forum implied here, that there is a causal relationship between implementing blockchain into government services and attracting investments and creating jobs. Soon enough I’m sure blockchain will be solving climate change and the world’s hunger.
Ironically, some of the biggest proponents of these applications are those who were the first in the firing line of the attack on the system launched by Bitcoin, the first application of blockchain. As Leo Weese puts it in his brilliant piece about blockchain as a cargo cult:
“In an attempt to dismiss the threat posed by Bitcoin and hopefully retain the upper hand, at least in conversations, the buzzword ‘Blockchain’ was born. The narrative is that Bitcoin can be safely ignored while its technology can be co-opted to enhance the existing financial system. Blockchains make banks stronger, increase profits and make it easier for governments to collect taxes and monitor its citizens.”
There are several issues with the approach to blockchain we’re seeing around. First is not really related to the technology. It’s with the approach to problems and their solutions. Too few people ask the fundamental question of what is really their problem and why do they have it in the first place. They don’t ask “why” more than once. If they did that, most of the time they would probably find out the real problem they have is an unnecessary complexity of their integration architecture, their own corporate bureaucracy, or a broader legislation within which they need to operate.
I sat with an executive of one of the world’s biggest system integrators’ proserve division and challenged their approach of private and consortium blockchains in supply chains. The point we kept getting back to was, that the whole integration architecture was either too complex or non-existent. The solution to that, I argued, was not blockchain, but a service-oriented architecture (I helped to build a pretty big one back in the day). I was told, that they were just testing it everywhere to see where it fits. To me, that is an equivalent of testing if you can render web pages using SQL queries. My impression from the whole approach was, that the word blockchain made it easier to get more parties on board, to break through some bureaucratic barriers or simply to sell more consulting mandays.
Further issues are directly related to the understanding of the context within which the technology operates. Cryptographic signatures, elliptic curves, and merkle trees we’d seen and used before. What makes blockchain revolutionary is the way it uses them to create the first truly scarce digital assets, to achieve consensus in an open and distributed manner, and to introduce a cryptocurrency not only as a product in itself but as an incentive mechanism to achieve that distribution. That, in turn, removes the necessity of trust and provides immutability and censorship resistance. Which makes it greatly applicable to domains which are outlawed or heavily regulated (private moneys, prediction markets), or where trust in a centralized institution is not desirable.
Immutability is blockchain’s main value proposition, but there is no immutability without openness. I couldn’t care less if something is dug deep into a blockchain, that has been kept private. In private blockchain, it’s easy to go back, change the desired record and re-create the whole ledger again from there.
It becomes more difficult in consortium blockchains where more parties are involved, but than there’s still someone maintaining that consortium and a permission to be part of it. There’s still a central authority, which gives access, which means all the parties trust that central authority and transfer that trust to each other. Which means, they’d be still better off with a solution built on a centralized database, exposed via a permissioned API. Why?
Because blockchain is very, very expensive and slow. It was designed and intended that way. The whole consensus mechanism is supposed to be expensive so that anyone who would want to manipulate it would find it prohibitively difficult. As a result, everyone involved can trust that consensus without trusting anyone else in particular. Blockchains, which are not open, permissionless and distributed are then just slow and expensive databases and APIs.
Another major limitation of blockchain’s application should be obvious to all but is missed by too many and that is its (in)ability to cross a tangible/digital chasm. Blockchain provides trustless immutability thanks to open and permissionless consensus distribution (sorry for hammering that point repeatedly). Tracking mangoes, which are physically handed over between two parties, on blockchain makes no sense because there is no consensus to be made. Not a single miner, let alone a majority of them had a chance to physically examine the transaction. If that transaction isn’t supposed to be settled using a cryptocurrency running on that blockchain, there is no reason for it to be stored there (the whole mango example is bizarre from other perspectives too, but I’ll get to that in a separate piece later).
It should be obvious then, that where blockchain truly makes sense is in exchange of purely digital assets. Bitcoin as blockchain’s first application was such a case, where the goods were the first scarce digital asset supposed to serve as a medium of exchange. I’ll get to further examples in more details in the following weeks.
When someone says blockchain, but not open, they are clueless about where the real value proposition of the technology lies. If they add “no cryptocurrency”, they don’t understand incentives, often even incentives of their own organization.
“When in the history of governance, banking, and journalism have we dedicated so much of our attention and efforts to a database? When has the world of finance and technology ever obsessed over the technicalities of wire transfers? The answer is that blockchains (…) are a result of failing to understand, or refusing to recognize, the technological and social relationships of a cryptocurrency like Bitcoin, coupled with the fear and rejection of being ‘colonized’ by a group of people whose culture and values appear foreign.”
So this brings us to cryptocurrencies.
“If someone offers you blockchain without cryptocurrencies being absolutely inseparable part of the project, make sure to run away and to not give them any money. It’s an IT equivalent of a multi-level dishwashing machine tablets sales representative (...) Cryptocurrencies are not just ‘better fiat’, it’s something completely new and I truly believe it will change the world. An important concept in addition to decentralization is ‘trustless’ and ‘permissionless’ operation. Trustless means, that I don’t need to trust anyone to use the innovation and permissionless means, that I can join the network without anyone giving me permission as long as I follow the protocol rules. The fact that cryptocurrencies use blockchain to achieve something, that wasn’t possible before this innovation. Order and payment tracking had been possible to build on a centralized database. Bitcoin was not.”
Think about it this way: cryptocurrencies are the main product of a solution built on blockchain technology. An analogy to that would be if Uber have built the data architecture to store all the information about drivers and riders and integration between them, but would not release an app to connect them or built a payment solution to incentivize the service provision between them.
Bitcoin was the first cryptocurrency and the solution of that cryptocurrency was decentralized, permissionless, private money where governments outlawed and shut down all previous centralized solutions. Bitcoin was the product of that solution and was used the wider and traded the higher the more people believed that solution was working as designed. In that sense, it is not too different from the old private moneys, which were products of trusted issuers. In the digital realm, cryptocurrencies can have more specific use-cases to serve the marketplaces they’re built for.
On top of being a product in itself, cryptocurrencies also serve two critical functions in the stakeholder map of any blockchain solution.
First, they incentivize miners to secure the network. Miners compete with each other and get rewarded for contributing their computing power to grow and maintain the trustless consensus. They’re some of the most important stakeholders in the whole incentive map. The bigger their competition, the more expensive it becomes for anyone to manipulate the consensus. The reward is in the form of cryptocurrency running on the blockchain.
The second is means of smart contract enforcement. The limitation of blockchain’s capability in crossing the tangible/digital chasm applies to enforcement of the contracts too. It’s all fine and dandy that two parties exchange digital goods via cryptographic signatures on the blockchain. But if the settlement of that transaction happens off the chain and its enforcement is dependant on a particular jurisdiction’s law enforcement, the blockchain doesn’t really provide any added value. As long as they need to rely on the physical service (police, a judicial system, etc.) to enforce the contract, all they need is any form of contract that that service respects (which can be signed e-mail for that matter).
Cryptocurrencies, however, would allow for automatic settlement of the specified transaction and more sophisticated smart contracts could bring auditing and dispute resolution capabilities.
For companies considering implementing blockchain-based solutions, the currency part (in a non-legislative sense of that word) can serve as an objective evaluation of the success of the solution and its return on investments. Here’s Juraj’s brilliant insight:
“If you use cryptocurrencies, it’s sufficient to look at your balance sheet. How much of the revenue goes through crypto? Do you have a percentage from transactions? Have you grown your network? If you use cryptocurrencies in your business model, do you use the volatility to your benefit? How much did you earn in the first quarter, year, how much in five years”
Speaking of companies, one important concept to mention are stablecoins. It is a topic worth a separate piece, but let me summarize below. But first off, I have to prevent (at least some) slack: I don’t hate fixed supply cryptocurrencies like Bitcoin (Cash), Litecoin etc.. In fact I love them. I want everyone to use them. But I also live in the real world where companies have risk departments, accountants, balance sheets, and fiduciary responsibility to their shareholders.
Stablecoins are cryptocurrencies pegged to a certain asset, typically USD. They should (at least in theory) deliver all the benefits of cryptocurrencies like censorship resistance and immutability, but have the same unit of account and not have to be a separate asset class in the balance sheet. They should allow participants to enter a cryptocurrency marketplace without being exposed to the volatility or without introducing new accounting rules.
There are several types, from centralized, fiat-collateralized (and critically vulnerable) Tether to completely decentralized, crypto-collateralized and market- and incentive-based Cryptopeg. Technically, the type of stablecoin doesn’t really matter for its application, but when it comes to delivering the benefits of crypto in the real world, I’m interested only in the market-based Cryptopeg-like stablecoins.
In the follow-up posts, I will explore various publicly presented use-cases in different domains and try to debunk some myths and narrow down to those cases where cryptocurrencies and perhaps blockchain can, in fact, be valuable and useful. Hope I will save you a lot of money and navigate you to proper uses of these concepts and technologies.