The Economics Driving Bitcoin’s Inevitable Centralization
Bitcoin’s innovations are numerous and undeniable. However, as a functioning network, Bitcoin suffers from economic limitations that cannot be eliminated through technical improvement. Validating with miners requires fee compensation, which will always incentivize consolidation regardless of technical developments because mining exhibits strong economies of scale.
Commodity good + economies of scale = centralized production
Bitcoin (or its progeny) will transform the world, but not as the instant, free, and decentralized system it promised to be. Building a network that performs on these three vectors simultaneously requires a paradigm shift in how networks are constructed.
Capacity: Bitcoin’s Inevitable Fees
Congestion is a fundamental challenge not only to Bitcoin, but to any network that moves information or physical things. Any network with growing demand requires continuous investment in throughput just to maintain service usability.
In a perfect world, networks scale capacity at the same rate as usage. However, new network capacity requires upfront investment and has deployment lag time. To meet new demand with corresponding supply requires either 1) the ability to perfectly predict the future or 2) the existence of pre-built excess capacity.
In a centrally planned network, overbuilt capacity enables the illusion of infinite bandwidth. For instance, Visa maintains redundant data centers that each can handle three times more than Visa’s highest use cases. This ensures that transactions are always instant.
In a network where supply is provided by market participants, price is required to signal when excess capacity should be enabled. For instance, in an open energy market, less efficient generators sit idly until the cost of electricity matches their economic cost of production (see: dispatch curves).
As a decentralized network, Bitcoin, by definition, relies on market participants for network capacity. Much like an energy market, Bitcoin requires a reward mechanism (i.e. price) to create capacity. A fundamental innovation of Bitcoin and other tokenized networks is the ability to bootstrap large global networks by rewarding capacity suppliers (miners) with the network’s own token (block rewards). It’s zero cost, zero risk capex. However, because Bitcoin is deflationary, the network can’t provide block rewards indefinitely, and must use other means to incentivize miner capacity.
Fees: Bitcoin’s Inevitable Centralization
With block rewards decreasing, Bitcoin increasingly uses fees to incentivize miner participation in the network. Without these fees, miners would not create blocks, and transactions would not be validated. No fees → no Bitcoin.
The service miners provide, ledger validation, is a commodity. Bitcoin is trustless and decentralized, which means, again by definition, Bitcoin users can have no preference for specific miners — they will always include the lowest fee possible in their transactions. This dynamic is problematic for decentralization because large scale mining operations exhibit economies of scale.
Commodity good + economies of scale = centralized production
Miners with lower cost structures can accept transactions with lower fees, and will expand their capacity until the market clearing fee is equal to their economic costs. The expansion in capacity will cause fees to fall. As individuals and smaller miners become cash flow negative due to lower fee revenue, they’ll stop mining, leading to a consolidation of hashing power*.
Bitcoin users are quite aware of this issue, and while individually they may seek lowest price, collectively, their preference is for decentralization. The collective voice of Bitcoin maximalists broke up mining pools and prevented miner-friendly changes to the protocol. However, as Bitcoin becomes a mainstream asset, it’s difficult to believe the average user, who already trades privacy for free services every day, won’t value fast and cheap transactions significantly more than they’ll value decentralization.
Quite simply, if miners on a blockchain exhibit economies of scales, there will be strong incentives for centralization.
Consolidation
The question for Bitcoin is really not if there will be consolidation, but rather, how much. Economies of scale are not infinite and it’s possible that the minimum efficient scale for Bitcoin mining is achievable by individuals. However, if the size of the data centers in Iowa and Nebraska are any indication of minimum efficient scale, individuals are unlikely to be competitive.
Ultimately, the level of consolidation in the Bitcoin and broader blockchain markets will be driven by consumer preference for maximalist style decentralization. It seems more likely than not that a preference for free and fast will drive the most popular blockchains to just a handful of large scale miners. This doesn’t mean Bitcoin will cease to be innovative. With consolidated, but not completely centralized mining, coin holders will have an international, functionally (though not theoretically) decentralized network with low fees and minimal congestion. That’s a step change improvement over existing networks, but is it ideal?
If miners, by virtue of basic economics truths, necessitate congestion, fees, and consolidation, wouldn’t a better network be one that could validate transactions without them?