[Repost] Three major issues affecting blockchain.

In the past, people often used centralized systems to store and manage records. With the emergence of Distributed Ledger Technology (DLT), there has been a significant change in how we store records, with the level of innovation comparable to the double-entry bookkeeping method first used in 14th century Italy. The emergence of DLT has completely changed the way financial transactions and ownership data are recorded.

Blockchain is a typical application of Distributed Ledger Technology. As a decentralized system, the members of the blockchain are usually anonymous, and there is no central entity. Each member can view and verify every data entry, which is the core operation of a distributed ledger. Ideally, a distributed ledger should have the following characteristics: accurately record every piece of information; low cost; fully decentralized to avoid centralization of power.

In 2008, Satoshi Nakamoto invented blockchain technology. Since then, a large number of related innovative applications have emerged one after another, aiming to release power from centralized government institutions or monopolistic enterprises. However, the cost of achieving a decentralized blockchain community is extremely high. Research by Abadi and Brunnermeier in 2018 pointed out that the development of blockchain faces the 'trilemma' - where no ledger can simultaneously meet the following three requirements. Specifically, decentralized systems face three major drawbacks: resource waste, low measurability and inefficiency of network externalities.

Next, I will first discuss the application scope of blockchain, and then explain in detail the 'trilemma', which is the relationship between accuracy, cost-effectiveness, and decentralization.

Currently, blockchain technology is mainly used for transaction records of cryptocurrencies, with the most well-known cryptocurrency being Bitcoin. However, the application scope of blockchain goes far beyond this. Some companies are using blockchain to track complex supply chain conditions. Consumer credit histories and merchant ratings can also be stored using blockchain technology, while major e-commerce platforms (such as Alibaba and Amazon) mainly rely on centralized systems. In the financial markets, blockchain can be used to record stock ownership, complete transfers of collateral, and conduct interbank payments. For governments, although blockchain may weaken their authority, there is still enormous potential for blockchain-based technology in real estate registration.

To better understand the "trilemma" of blockchain, we need to first understand why decentralized, anonymous blockchain systems require so much computational resources, which is one of the most important costs of blockchain. For public blockchain systems, each member can add data. Therefore, in order to filter out non-compliant or false data, a consensus algorithm must be designed.

Satoshi Nakamoto proposed a solution in 2008: each member writing data into the blockchain must perform a step with the purpose of entering the information into the system, passing it to each member, and verifying the reliability of the information (i.e., 'voting' for the information). This process requires a large amount of computational resources. The most commonly used consensus algorithm is Proof of Work (PoW). In PoW, "miners" must verify each piece of information through brute force. The computational power of each miner determines their 'voting weight'. As a result, the entire blockchain system is highly democratized. Using the PoW algorithm does not require a centralized entity to allocate voting weights; anyone can freely join the system.

This 'freely join' characteristic significantly encourages people to use blockchain systems. In the past, the motivation for centralized intermediaries to prevent fraudulent behavior was that if their misconduct was exposed, the value of their franchise would be severely damaged. In other words, the behavioral motivation of centralized institutions comes from expected benefits. If the profits from tampering with the ledger exceed the losses from expected benefits, they may choose to commit fraud. In a blockchain system, there is no franchise value because every member can freely enter the system. Blockchain members need to weigh the benefits of a single attack against the cost of writing information to the blockchain. For example, cryptocurrency miners compare the cost and benefits of implementing a 'double spend' attack. Therefore, achieving decentralization through 'freely join' leads to significant waste of computational resources, which is part of the blockchain 'trilemma'.

The 'freely join' feature is closely related to the second cost of blockchain - measurability. If users do not trust other members in the blockchain to truthfully report content, they must store the entire blockchain. Even for blockchain systems with limited throughput, storage costs are very high. For example, the processing speed of the Bitcoin blockchain is only 7 transactions per second, while the entire system's capacity has reached 250GB. In short, the higher the frequency of blockchain usage, the higher the cost of achieving fully decentralized records.

In addition to 'freely join', blockchain also allows another form of competition - 'forking'. When some members wish to change the rules of the blockchain system, they can add information to the old blockchain according to the new rules, which may be illegitimate information under the old rules. At this point, the blockchain will fork into a new chain and an old chain. For example, the birth of Bitcoin Cash is a typical case. Some members of the Bitcoin blockchain wanted to increase the transaction limit for each block, so they created the new chain, Bitcoin Cash.

'Fork competition' results in an obvious outcome: data from an existing ledger may move to a new ledger. If information can freely move, competition between different ledgers will tend towards perfect competition. In contrast, users of centralized ledgers are often tied to the ledger itself because the cost of switching to a new ledger with different information is too high. Network externality encourages this 'binding' behavior, making it difficult for new ledgers to emerge.

While 'forking competition' promotes competition among members, it also has significant drawbacks. Forking competition directly leads to the third cost of the blockchain, that is, the inefficiency of network externality. If the transition between forks is too frequent, the stability of the blockchain will be threatened, thereby unable to fully leverage the network effect. In fact, both Bitcoin and Ethereum, the two major cryptocurrency blockchains, have experienced forks - a considerable number of members abandoned the original blockchain.

One issue often overlooked for blockchain is: while blockchain can well record the transfer of ownership, it cannot guarantee the transfer of possession. For example, in the real estate market, the owner of the house is the name on the property certificate, but the actual resident is the possessor. After the buyer obtains the property certificate, they still need to ensure that they can physically possess the property. Broadly speaking, blockchain can record obligations, but cannot penalize members who fail to fulfill obligations. Blockchain cannot provide incentives for users to comply with rules, while centralized intermediary institutions can effectively perform this function because non-compliance will damage their reputation.

In some application scenarios (such as cryptocurrencies), enforcement may not be a necessary consideration, but in other applications, enforcement is crucial, especially in the financial sector. For example, in real estate registration, countries like Sweden and India have started using blockchain-based systems for management. Government involvement is necessary anytime, anywhere. When enforcement becomes necessary, compliance with rules is the best choice for reputable entities.

The 'trilemma' of blockchain reveals some economic considerations when designing decentralized ledgers. Traditional ledgers managed by a single entity cannot achieve decentralization. Managers of centralized ledgers have the incentive to record truthfully because they do not want to harm their expected benefits and franchise value. Blockchain reduces the cost of centralized intermediaries through 'free entry' and 'forking competition', but the cost of achieving decentralization is sacrificing efficiency. 'Free entry' ignores the franchise value of members; the accuracy of the ledger depends on the incentive of members through the PoW algorithm; and forking competition, while promoting competition, also leads to the instability and even chaos of the blockchain. In addition, effective blockchain operations may still require legal supervision and enforcement by the government.