Bitcoin’s transaction speed, often criticized, is a deliberate and essential feature. Its difficulty adjustment mechanism links the decentralized ledger to real-world time, enabling secure coordination and maintaining the integrity of the system.
Economist Walter Williams once wrote that,
We can think of dollars as certificates of performance. With these certificates of performance in hand, I go to my grocer and ask him to give me a pound of steak and a six-pack of beer that my fellow man produced. In effect the grocer says, “You’re making a claim on something your fellow man produced. You’re asking him to serve you – but did you serve him?” I say, “Yes I did.” The grocer responds, “Prove it!” That’s when I show him my certificates of performance – namely, the money my fellow man paid me to mow his lawn.
This idea has been formalized elsewhere in the economics literature. Economist Narayana Kocherlakota’s paper “Money is Memory” highlights the role that money serves in conventional monetary theory: Money is a record-keeping device.
This point is important when we consider the impetus for the creation of Bitcoin. The Bitcoin white paper describes bitcoin as electronic cash, which is significant. Even prior to the introduction of Bitcoin, people concerned about privacy in an increasingly digital age sought to find a way to create a digital form of cash that could potentially provide the privacy features of physical cash. Along the way, a number of valuable lessons were learned. One of the most notable lessons is the role of third-parties in any attempt at creating an electronic alternative to cash. Often, this is framed as a matter of trust. However, it is more than just trust. Even a trustworthy third-party might go out of business or be targeted by less trust-worthy actors. It is also important to consider the nature of record-keeping.
Physical cash is a substitute for record-keeping. Electronic cash requires record-keeping. It is not hard to understand why. Ownership of something tangible can be determined by possession. Ownership of something intangible, however, requires keeping some record of ownership and transfers thereof.
Most people are familiar with the nature of electronic transactions. One can deposit cash into the bank in exchange for a claim to that cash (along with an option to exercise that claim at any time). However, one need not exercise that option to spend the money deposited at the bank. It is possible to send payments electronically. If this transfer occurs between customers of the same bank, the bank simply updates its own electronic ledger, debiting money from one account and crediting money to the other account. The bank is the trusted third-party maintaining the ledger. When the transfer is made between two parties who use different banks, this requires a clearing agent. However, the nature of the transaction remains the same, trusted third-parties simply update their ledgers.
Although these types of banking transactions are electronic transfers of money, these bank account balances are not electronic cash. The nature of cash is peer-to-peer. There are no third-parties involved in cash transactions. Herein lies the challenge of developing an electronic form of cash. To have an electronic form of cash, (a) there must be a ledger, and (b) the transactions must be peer-to-peer. However, if there is a ledger, someone must maintain the ledger. Yet, if someone maintains the ledger, then the transactions aren’t peer-to-peer.
Electronic cash therefore requires a ledger that is updated in a decentralized manner. If maintenance of the ledger is sufficiently decentralized, then one doesn’t have to rely on a single third-party. Of course, creating a decentralized ledger is easier said than done.
The benefit of having a bank update a ledger is that there is no disagreement about what is on the ledger. Decentralized maintenance of a ledger requires that everyone maintaining a copy of the ledger has the same copy. This requires updating the ledger in the exact same way. But how is that possible? Again, this is often framed in terms of trust. If the people maintaining the ledger can communicate, then subsets of people could collude to update the ledger in a way that is beneficial to the group. However, even if everyone is trustworthy, there remains the issue of honest coordination.
Bitcoin solves this problem through its proof-of-work consensus mechanism. Bitcoin miners compete to produce a cryptographic hash that begins with a certain number of zeros. Miners must commit computing power to find a valid hash. Since a valid hash is hard to find, but easy to verify, this creates a lottery-like process in which generating more hashes gives miners more opportunities to “win,” but is not guaranteed of success. Once a miner does find a valid hash, it is easy to verify, and thus other miners and nodes can update their ledger accordingly.
However, another aspect of the proof-of-work mechanism is critical to the functioning of Bitcoin, even though it often gets less attention in popular discussions. That feature is the difficulty adjustment. The number of leading zeros in a valid hash depends on how quickly new blocks of transactions are being added to the ledger. Every 2,016 blocks, the Bitcoin protocol updates the difficulty of mining blocks. If transactions are occurring faster than every 10 minutes, then a valid hash must start with a greater number of zeros. This makes finding a valid hash more difficult and pushes the pace of block additions back towards 10-minute intervals. If transactions are occurring more slowly than every 10 minutes, then a valid hash must start with a fewer number of zeros. This makes finding a valid hash easier and therefore speeds up the pace of transactions to get back toward 10-minute intervals.
By operating in this way, the difficulty adjustment ensures that as more computing power is added to the network, the expected time until the next block of transactions is added to the ledger is 10 minutes. In other words, more computing power gives a miner a better chance of finding a valid hash, but has no effect on the expected time with which blocks of transactions are processed.
Critics often view this 10-minute interval of time as a flaw. They argue that Bitcoin needs to process transactions faster than this. However, those critics are wrong because they fundamentally misunderstand the role of time in maintaining a decentralized ledger.
A ledger keeps track of ownership and transactions that transfer ownership. The crucial aspect of such transfers are when they occur and thus are a question of time. The problem of determining when something happened is largely trivial in a centralized system. The question of when something occurred is determined by the entity that controls the ledger. It is not trivial in a decentralized system.
The difficulty adjustment is crucial to solving this problem. In the absence of the difficulty adjustment, the proof-of-work mechanism is incapable of solving the coordination problem. Without the difficulty adjustment, greater computing power on the network would result in shorter and shorter intervals of time between blocks of transactions. This additional computing power could come from greater numbers of miners on the network or simply from technological improvements. Regardless, this would eventually result in blocks being added at such a short interval of time that it would be impossible for all of the nodes on the network to coordinate and share a common transaction ledger. Instead, the difficulty adjustment explicitly links the timing of transactions in Bitcoin with time in the physical world. As such, and by maintaining a consistent expected time interval until the next block, it allows for the coordination among all those maintaining the ledger.
Contrary to critics, the speed of Bitcoin transactions is a feature and not a bug. The difficulty adjustment mechanism is what ties Bitcoin to the physical world and its measure of time. In doing so, it allows for the coordination of a decentralized network capable of maintaining a common, shared ledger. Without that decentralization, one cannot have electronic cash. This is the innovation, and arguably the genius, of the design of Bitcoin.
