Tag Archives: reversible transactions

Multi-signature Transactions: The Future of Bitcoin

One of the criticisms of Bitcoin is that a Bitcoin transaction is one way and irreversible. However, within the Bitcoin protocol, resides a powerful functionality called “m-of-n” or “multi-signature” transaction. A multi-signature transaction is one that is associated to more than one private key. The ‘n’ is for number of private keys and ‘m’ is the minimum number of signatures needed for that transaction to be successful. For example, in a 2-of-3 transaction, two signatures are needed to send the transaction to blockchain. This functionality has many advantages, especially in wallet security, arbitration and smart contracts.

Wallet Security

Most wallets currently available are single key. If a user looses his/her laptop, hard drive, forgets the password or gets hacked, there is no way to retrieve the coins. In a multi-signature wallet, there are typically three private keys. One key is stored on a web server, second is the user’s main key and third is a backup that can be stored on a paper wallet or USB drive. The wallet can receive funds as normal but when the user wants to withdraw any coins, signatures from atleast 2 of the 3 keys are needed to send the transaction to the blockchain. When the user initiates a withdrawal from his main key, a partially signed signature shows up on other two wallets for approval. Only when one of the other two keys has been signed, the transaction is sent to the blockchain. With a 2-of-3 type wallet, a hacker would have to steal at least 2 keys to withdraw funds. If the user forgets or looses one key, he/she can still withdraw funds using the other two keys.


One of the benefits of making a purchase with a credit card is that in the event of a dispute, the customer can file a chargeback with the credit card company. There are two primary issues with Mastercard and Visa arbitration systems: (1) the system is open to abuse (2) the cost of fraud is split across all consumers.

Card Not Present (CNP) affects all e-commerce transactions, as the card holder is never present at merchant’s premises. All a customer has to do is sign his/her name differently on the courier delivery note. The customer can then file a chargeback and win; the merchant will loose both the money and the product. Credit card companies not only charge merchants an admin fee for processing chargebacks but also factor in a certain percentage in their fees for dispute resolution; hence, due to network affect, everyone ends up paying for fraud. The arbitration service is costly and inefficient. On some occasions, a specialised arbitrator is needed but Mastercard and Visa can only offer a generic standardised service. For some transactions, such as sending money to charities, arbitration is not needed at all.

In a multi-signature Bitcoin transaction, there will be three players: customer, merchant and a third party arbitrator. Customer will transfer funds to the arbitrator and the merchant will ship the goods. If customer is happy with the goods, he/she will sign the transaction and merchant will receive the funds. In case of a dispute, either customer or merchant can appeal to the arbitrator, who like a credit card company, will review evidence and mediate the dispute. The arbitrator can not withdraw the funds and is paid a pre-determined market rate only if he/she is involved. Arbitration service being developed by Bitrated will work in a very similar way. Bitrated’s platform will also allow arbitrators to build reputation over time. Multi-signature makes arbitration cost effective; if there is no dispute, no fee is charged. In the long run, due to competition between merchants, these cost savings will be passed onto customers. The system is also fairer, as unlike Paypal and credit card companies, the arbitrator has no incentive to rule in the favour of the customer.

Smart Contracts

This is when things start to get really interesting. Multi-signature transactions can also be used to draw automated smart contracts. The idea of smart contracts was first discussed in 1997 by George Washington University law professor and computer scientist Nick Szabo. Szabo’s ideas were revolutionary, he argued that smart contracts must be enforced algorithmically, using cryptography and other security principals; not by law. Bitcoin and multi-signature transactions make smart contracts possible.

Consider the example of a will – one of the parties in the will would be a computer server. The computer will have its own unique key, with which it will be able to sign Bitcoin transactions. The computer will determine whether a person is alive by checking a central death certificate database periodically. If the computer finds a certificate, it will sign the transaction and transfer the funds to the beneficiary. Contracts can also be split ‘n’ ways (‘n’ grandchildren or business partners) and to authorise transfer of funds at least ‘m’ signatures would be needed. The functionality can be used to automate mortgages and leases. Automation and verification of contracts by the blockchain will reduce, if not eliminate, brokerage and banking fees. The system can be used to verify and transfer ownership of assets. Szabo described how cars could be made to read the blockchain and disable themselves if a loan payment was not made on time. Szabo explains:

consider a hypothetical digital security system for automobiles. The smart contract design strategy suggests that we successively refine security protocols to more fully embed in a property the contractual terms which deal with it. These protocols would give control of the cryptographic keys for operating the property to the person who rightfully owns that property, based on the terms of the contract. In the most straightforward implementation, the car can be rendered inoperable unless the proper challenge-response protocol is completed with its rightful owner, preventing theft.

The system could be designed to automate ownership of physical assets such as phones and houses (smart property) and non-physical assets such as shares in a company. The ability to put up assets as collateral and give away shares in a company will allow businesses to raise money over the internet. Many of these features are yet to be explored and implemented. As the Bitcoin eco-system grows, entrepreneurs and investors will flock to build new companies to explore full potential of the Bitcoin protocol.