As the spotlight returns to Bitcoin (BTC), helped by some tweets by a certain mercurial celebrity, the raging debate over its energy use has once again reignited. It centers around one, seemingly clear-cut question: Does Bitcoin use too much energy?
The basic contours of the issue are clear. Bitcoin secures its network from a hostile takeover using proof-of-work (PoW), a process that expends significant quantities of electricity due to the computing power required. Every time we have this discussion, the all-too-familiar battle lines are redrawn.
Critics argue that Bitcoin’s energy use simply cannot be justified. At various stages in recent years, reports have estimated that the network uses as much electricity as entire states such as Denmark or Ireland, for instance.
On the other side of the fence, Bitcoin’s proponents contend that the network might spur the greater use of renewables. Furthermore, they point out that we are not accounting for the energy use of the alternative. We cannot gauge the relative efficiency of Bitcoin as a means to secure and exchange value unless we compare it with the total energy use of the traditional banking system. Just as we should move beyond the narrow metric of tailpipe emissions to measure the environmental impact of vehicles, Bitcoin advocates assert that we need a comprehensive audit of the environmental impact of traditional finance, including all the infrastructure, brick-and-mortar buildings, travel and hardware that support it. In addition, lurking in the background, are the other alternatives — what about consensus mechanisms such as proof-of-stake (PoS), the approach that Ethereum is transitioning to?
The standard battle lines
It is a fact, and noncontroversial, that mining technology for blockchain consumes vast amounts of energy. This is particularly stark when comparing the cost of producing and circulating currency.
Bitcoin, for instance, is estimated to consume 123.77 billion kilowatt-hours of energy annually, compared with 2.64 million kWh for cash. According to Digiconomist founder Alex de Vries, if Bitcoin became the world’s reserve currency, global energy production would need to double.
Others claim that miners will eventually gravitate to wherever energy costs are lowest, or become the green energy buyers of last resort. Whether the argument stands up over the long haul remains to be seen, given the degree of regulation in energy markets, the physical costs of relocation and the potential security implications of concentrating miners.
Framing the opportunity costs
Of all these arguments, comparing the energy use of cryptocurrencies with the traditional banking sector — or fiat, in particular — is relatively new. Comparisons with legacy payment systems, however, overlook the difference in transaction volume: While the Visa network completed over 185 billion transactions in 2019 alone, Bitcoin has facilitated 643 million since its inception. Furthermore, commercial entities like Visa are well integrated with energy markets, which are highly regulated in many countries. In the mental models where miners move en masse into new energy markets, it is highly likely that transition costs (as well as the resistance of incumbents) are being discounted. Again, these tendencies are not surprising, as cryptocurrency advocates tend to look optimistically to the future, imagining that markets work more efficiently than they actually do.
Setting aside the non-trivial, highly complicated implications of energy use for the security of blockchains, the idea that miners will follow cheaper electricity prices does not necessarily mean cleaner energy, as cheaper is often dirtier. But even more importantly, the idea that miners will eventually just switch to renewables ignores the opportunity cost of energy. According to the United States Energy Information Administration, global energy usage will grow 50% by 2050. The emergence of unforeseen computational requirements posed by smart cities or integrated supply chains, for instance, will require blockchain to be more energy efficient — all while humanity needs to keep an eye on climate goals.
So, while Bitcoin maximalists are undeterred in their belief that Bitcoin is the first best use for energy, and while proponents of Ethereum — which is moving to a different model, in part apparently due to energy use concerns — may think they have a long-term solution, the general public may not be convinced that cryptocurrency (and nonfungible tokens built on technologies like that of Ethereum) have a sustainable answer to the question: What will be best for society?
Blockchain is now receiving mainstream attention, which gives those of us in the industry a chance to restate the problem in a way that speaks to all of us. Do we think the benefits of blockchain will be worth the opportunity cost? When it comes to memecoins built on mining chains — which are fads, peaking and waning in price with popular sentiment (and new memes) — and the many scams and imitations that have popped up (to the continual embarrassment of serious projects in this space), blockchain technologists are rightfully afraid the public will decide it is not.
However, if we are discussing the benefits of new blockchain technologies that take resource use seriously and open new markets as the internet did, that is an entirely different matter. In that case, the correct comparison is not merely with the opportunity cost of staying with the status quo in finance but with the intermediated economy as a whole.
More to the issue than just mining
While the debate about the efficiency of cryptocurrency tends to be dominated by the discussion of mining, less attention is given to the alternatives. PoS protocols sidestep the need for mining by changing what bad actors stand to lose if they try to falsify transactions. While with PoW such actors could potentially lose the energy they invested, on a PoS network they would forfeit cryptocurrency staked in advance. But this solution also comes with energy considerations.
Suppose that some of these stakers are centralized exchanges: Their first incentive will be profitable trading, not monitoring the energy efficiency of the underlying blockchain. In this respect, we need to consider how information is disseminated among nodes. Mainstream blockchains typically use peer-to-peer gossip networks to communicate. Put simply, such networks pass transaction data from node to node until it is known by all participants. As a result, however, the same message may be repeatedly sent to peers who have already received it from others, wasting resources. And the protocols that assume that security and transaction volume will be able to attract a sufficiently large number of nodes to maintain accuracy in some fashion — whether they are new delegated PoS protocols, directed acyclic graphs, layer-two solutions or cross-chain bridges — are similar to PoW in their assumption that the correct transactions will be confirmed and propagated wherever the network needs that information to be.
However, if we manage to overcome the limitations of gossip networks, a whole new world opens up. For instance, nodes on Geeq blockchains use a hub-and-spoke structure to communicate, in order to transmit a minimal set of messages without defaulting to a centralized power structure. Any honest (and potentially anonymous) node may serve as a hub for one block and communicate with the nodes on that blockchain’s active node list (the nodes that happen to be on active spokes).
Unlike a gossip network, where each node sends messages (gossips) to every node around it, meaning that a particular node could receive the same message redundantly from all of its gossiping buddies, this structure results in messaging that is parsimonious, predictable and verifiable. As a result, the use of resources is lower by magnitudes compared with PoW or PoS based protocols, bringing computation, bandwidth and storage costs per transaction as low as a hundredth of a cent, making micropayments feasible.
Furthermore, future blockchain architecture will need to be multichain and flexible, providing a set of parameters that can be adjusted according to the specific requirements — such as speed, transaction throughput or security — of a given use case. A more lightweight, “smarter” blockchain would certainly have a smaller environmental footprint, but it would also be easier to adopt, and could even provide the underlying infrastructure for more sustainable societies.
Small is beautiful
One promising application in this regard is P2P energy trading. Currently, large utility companies supply entire cities with electricity through centralized networks. However, smart cities in the future could rely on a more flexible web of microgrids instead. To satisfy local consumption, these localized, autonomous electricity networks would use mainly local sources like power generators or photovoltaic panels.
Related: Talking digital future: Smart cities
Blockchain technology has always been a promising way to execute, validate and record P2P energy transactions, letting anyone on a local microgrid become either a producer or a buyer of electricity. However, up until now, the technology has not been up to the challenge. In order for the market to work well, units of energy as low as a few kilowatts would need to be traded, which would equate to a monetary value of just a few cents. Such transactions are infeasible given current blockchain transaction fees. When transaction costs are fractions of a cent, however, this hurdle would be eliminated. In turn, this would allow blockchain to serve as the technological bedrock of smart cities, allowing millions of Internet of Things devices like smart meters or solar panels to seamlessly interact and interface with digital wallets, often without human intervention.
For example, before going to work in the morning, you could charge your electric vehicle from the energy gathered from photovoltaic panels installed on your roof. Later you may decide to sell unused electricity to your neighbors before going on a vacation. It would also be possible to set up networkwide demand response rules, written in smart contracts. According to the Natural Resources Defense Council, for instance, the cost of “vampire electricity” consumed by plugged-in but unused devices is circa $165 per household, amounting to 4.6% of the total residential electricity production in the United States. Hence, an electric toothbrush left on the charger would be turned off during certain time periods automatically. To override network rules, you would need to pay a small compensatory fee, incentivizing producers to offset extra demand while discouraging users from wasting energy.
In addition, blockchain-based applications — decentralized applications, or DApps — may be built to ensure the traceability of clean energy. Thus, when purchasing electricity, you could check via an app whether it came from a sustainable source. Empowering the consumer to make these decisions is only possible with decentralized technology; otherwise, intermediaries will be able to distort markets to their own tastes. With the rise of global environmental consciousness, traceability may become a key tool to incentivize the production of renewable energy.
New horizons ahead
With such a drastic growth in global energy consumption predicted, it is easy to see why blockchain’s environmental footprint is coming under scrutiny. However, it is also important not to throw out the baby with the bathwater.
As well as taking a holistic view of the relative energy consumption of blockchain compared with traditional finance, we should begin a wider discussion about the net positives and negatives of the technology for society more broadly. In order for blockchain to fulfill its transformative potential, underpin smart cities and support low-carbon economies, we need to focus on developing smarter, more affordable blockchain architecture.
The views, thoughts and opinions expressed here are the author’s alone and do not necessarily reflect or represent the views and opinions of Cointelegraph.
Stephanie So is an economist, policy analyst and co-founder of Geeq, a blockchain security company. Throughout her career, she has applied technology within her specialist disciplines. In 2001, she was the first to use machine learning on social science data at the National Center for Supercomputing Applications. More recently, she researched the use of distributed networking processes in healthcare and patient safety in her role as a senior lecturer at Vanderbilt University. Stephanie is a graduate of Princeton University and the University of Rochester.