In October 2008 Satoshi Nakamoto, the inventor of the Bitcoin, published a white paper called: "Bitcoin: A Peer-to-Peer Electronic Cash System" describing how to implement a public, secure and decentralized electronic system to transfer money (value) over the internet without going through any financial institution.
If you already know a bit about cryptocurrencies and did not read yet the paper, I highly recommend having a look at it (here).
Online Payments, the trusted third party issue
In its introduction Satoshi Nakamoto clearly explains the issues merchants are facing when trying to sell goods online:
"Commerce on the Internet has come to rely almost exclusively on financial institutions serving as trusted third parties to process electronic payments. While the system works well enough for most transactions, it still suffers from the inherent weaknesses of the trust based model. Completely non-reversible transactions are not really possible, since financial institutions cannot avoid mediating disputes. The cost of mediation increases transaction costs, limiting the minimum practical transaction size and cutting off the possibility for small casual transactions, and there is a broader cost in the loss of ability to make non-reversible payments for non-reversible services."
On the P2P Foundation Forum [here], Mr Nakamoto also mentioned central banks as another third party that need to be trusted as they can debase the currency. For sure this statement participated to the demoniac libertarian reputation of the virtual currency, but that will be a topic for another article. Today, I would like to be more pragmatic and focus on the issue quoted here above.
Concretely, existing third parties are not efficient, biased and costly
A trusted third party is an intermediary that people rely on to perform a valuable service, in our case financial institutions are trusted third parties enabling electronic payments. Where in the physical world a merchant can avoid intermediaries by accepting cash, online, sellers and buyers can almost exclusively count on credit card companies to handle payments. Moreover, where everyone would trust Visa or Master Card to secure, validate and process transactions, we would be much more reluctant to mandate an unknown company to do it. Hence the notion of trust.
Having intermediaries induces cost and constraints. One of the issues Mr Nakamoto is mentioning is that online credit card payments are reversible. A credit card holder can claim a reimbursement up to 4 months after a transaction is made if:
- his card has been stolen and used for fraudulent payment or,
- he bought articles from a merchant that did not deliver the goods
A dispute is then opened by the credit card company which has the possibility to cancel the payment and, if so, charge the seller fees (chargebacks) ranging from 15 to 100 USD. Disputes are expensive and time consuming for both merchants and credit card institutions, but where in the physical world credit card companies take the whole responsibility in case of a fraud, in the virtual world it is the merchant who is liable.
For an honest vendor, a transaction being cancelled when the products had been shipped or the service delivered is synonym of loss of revenues, inventories and time. In North America, merchants revenue losses due to fraud was estimated to $3.4B in 2011 [source]. This number combine both real fraud (from stolen credit cards) and "friendly fraud" (i.e. when a buyer claims a reimbursement after receiving the products).
Furthermore, credit card institutions has an incentive to keep their clients (the card holder) happy to encourage online payments. Disputes are closed most of the time in favour of the buyer, even in case of friendly frauds, leading to a bias in the model.
The system is hence inefficient, biased, costly and particularly disadvantageous for merchants.
Bitcoin, designed to solve those issues
The main idea of the Bitcoin is to remove the trusted third party enabling direct payments from the buyer to the seller. For that, Mr Nakamoto imagined an infrastructure combining 3 fields of the Mathematics and Computer Sciences:
- Databases, to keep a record of all transactions
- Peer to Peer networks, to avoid centralizing the data and hence having someone able to control them, and
- Cryptography, to protect the data
In each of those fields, nothing was revolutionary. But what was totally new and clever was the way of combining them together to build a robust model. This model is called the Blockchain.
The main features of the Bitcoin ecosystem are:
- Payments are directly made from one user to another, validated, processed and secured by the whole network
- Transactions are non-reversible
- A user has to effectively own the funds to do a transfer (credit being made impossible, there is no risk of default of payment)
- A Bitcoin cannot be double spent (a malicious user can't spend a Bitcoin he already spent)
- Transaction fees are, in theory, freely decided by the buyer
- Everyone participates in securing the system
- There is neither central bank nor regulator. The value of the money is based on demand and supply and not backed by a physical asset.
Any change in the system needs to be approved by the majority
A collaborative system with no intermediaries, non-reversible transactions, no default of payment, no fraud, low transaction fees, no central organization taking arbitrary decisions... this solution seems to be the answer to the problems identified by Mr Nakamoto.
Which problems it does not solve?
Ostensibly the blockchain is not a "trust based model", but where some issues might be solved, others are emerging. Among them:
- To have access to his money on the network, a user needs to authenticate itself through credentials that are stored on his computer. In the event of a hacker accessing the credentials, nothing can prevent him siphoning the account. There is no insurance or protection against theft unlike with credit cards. Moreover, if credentials are lost (hardware issues, inadvertently deleted, virus...), the money stored in the electronic wallet will be forever blocked in the blockchain, which means lost.
To solve those critical issues that are preventing a large adoption of the Bitcoin by non-experts, companies started to offer ready-to-use accounts using electronic vaults. In exchange they are collecting account creation fees, account maintenance fees, transaction fees, withdrawal fees... It sounds familiar, doesn't it? The Bitcoin economy is replicating the banking system, introducing trusted third parties that are not foolproof (as demonstrated by the bankruptcy of Mt. Gox in February 2014 or more recently the hack against Bitfinex in August 2016).
- Also, the fact that transactions are non-reversible does effectively protect sellers, but generates a risk for buyers. What if a buyer purchase a service via Bitcoin, but that service never get delivered? There is no possibility for him to recover his money. In addition, as Bitcoin accounts are totally anonymous it is (almost) impossible to find out who is hiding behind that account.
To tackle this issue, the Bitcoin community implemented what is called an escrow mechanism having a third party that both protagonists trust to validate the transaction. Once more, it is replicating what exists in the common economy introducing another trusted third party. Here the problem is not a trust issue toward a third party but rather a trust issue between the two actors of the transaction, the escrow agent serving as a guarantor. However, here we can agree that the blockchain dissociates the service provider transferring the money and the escrow agent, lowering the bias mentioned earlier.
At its core level, the Bitcoin model might be considered as a "trustless model" (i.e. not requiring trust), but in practice it is not the case. For a currency to be adopted by a large population, trust is the cornerstone. Third parties do help to build trust by offering valuable services such as insurances, vaults, escrows and so on.
What is the real breakthrough of the Bitcoin system?
Despite all this, what Mr Nakamoto and the Bitcoin community achieved is quite impressive:
- The market capitalization of the Bitcoin is over $10 Billion
- with an estimated average transaction volume of $150 Million daily in 2016
- and 8 million accounts created (approximation of number of users)
This would not have been made possible if the system implemented was not highly reliable and secure. Even though it is an open source project (i.e. the code is freely accessible and readable by everyone), as of today the blockchain had never been hacked*. Moreover the blockchain (here, the database) which handled millions of transactions over the last 8 years weight less than 80 Go, which can be considered as very light if we think about the amount of information it is storing. Of course, the model has its limits and is not perfect (it will be discussed in coming articles), but there is a very active community working on improving it and developing other usages based on its specificities that could disrupted some industries.
Mr Nakamoto explains the failure of previous cryptocurrencies by the fact that there were centralized and hence depending on a third party people needed to trust. He had the idea to implement a model based on Peer to Peer networks to overcome the issue. We can argue for a long time rather the model is not relying on trust or well design enough to be trusted, but as a matter of fact an economy has emerged around the Bitcoin and 8 years later it is still existing and is gaining popularity.
The model behind the Bitcoin, the blockchain, is a real breakthrough and there are opportunities to apply it to other fields than cryptocurrencies. It would need to be adapted and improved, and is of course not relevant to any kind of challenges, but the Tech industry seriously believes in its disruptive potential.
In later articles we will dive deeper into the blockchain world, trying to understand how it works, what are the technical challenges, what are the different solutions existing, and most importantly in which particular situation the blockchain is more relevant than other models (demystifying far-fetched promises creating the blockchain hype).
About the Author: Romain Rouquier worked for financial institutions as market analyst and risk manager before creating its own company. Graduated from a master of Computer Engineering and a master of Financial Engineering, he is passionate about disruptive innovations linked to Computer Science - among them Data Science, Business Intelligence and Blockchain - and their applications to the industry.
* Bitcoins had been stolen in the past but not because of a security breach on the blockchain, but because credentials to access the blockchain had been stolen, as if a credit card was robbed.
This article originally appeared on LinkedIn, click here to see the original post and follow Romain Rouquier!