Is Crypto Ruining the Environment? Depends Where You Look
There has been a debate in the media and at a regulatory level around the environmental impacts of cryptocurrency, however, a full and fair discussion of the energy use placed in the larger social, economic and environmental context is required. The amount of media and regulatory focus on these impacts far outweighs the actual levels of electricity consumption when compared to other high-consumption industrial and technological power usage. Such discussion also generally fails to take many factors into account, including the real world utility of this technology and the potential efficiency gains it offers by replacing or reducing far more energy inefficient systems such as the global finance and banking systems. It also often ignores the actual and potential amounts of crypto energy use being derived by utilizing renewable sources and otherwise wasted energy. This more complex picture suggests that perhaps the negative narratives are in-fact motivated by other aspects of this technology unrelated to energy use that threaten the interests of established industries. This essay attempts to provide a detailed overview of the true energy usage of different types of cryptocurrency and how this usage will likely change over time.
Why It Matters
There is no question we are facing global climate issues that are directly related to carbon and similar emissions into the atmosphere. In addition to vehicle emissions, global electricity production is responsible for a significant amount of global emissions. Globally, the last estimate according to ourworldindata.org from 2020, is nearly 35 billion metric tonnes of CO2. While only 25% of the global CO2 production is a byproduct of electricity production according to the EPA, it’s estimated that 60% of global energy production is created by burning fossil fuels, which directly contributes to CO2 emissions.
The high power demand for Proof of Work (PoW) cryptocurrencies stems from the fact that these cryptocurrencies require computing power to essentially guess the correct hash for the next block of the blockchain. This is why the output of miners for PoW cryptocurrencies is rated in hashes per second, as this number is how many “guesses” the miner can make each second, and the faster it can compute possible outcomes the more likely it is to discover the correct hash. Once correctly guessed, this hash is used to create the next block, which is then verified by other miners through a distributed ledger and either accepted or rejected (uncle blocks).
This computing power is what gives blockchain its immutable, trusted, and decentralized nature and removes the need for centralized intermediaries and human oversight like in banking and other systems. It is certainly true that blockchain technology uses significant electricity like any advanced computing task. However, the exact amounts consumed vary across blockchains and consensus solutions.
Any industry with direct or indirect emissions should rightly be scrutinized carefully, which could be why “crypto’’ or “cryptocurrencies” are often criticized for being energy intensive or inefficient. However, as we will discuss, the electricity use of cryptocurrency is mostly linked to Bitcoin mining — as the largest digital asset by market cap with over half of the market cap of all cryptocurrencies.
However, in return for this energy use, the Bitcoin network is the most fully decentralized, most resilient to attack or censorship by either private or state actors, and considered the most secure cryptocurrency. This is a key factor in its popularity and the belief of many proponents (and even some nations) in Bitcoin’s enduring place as a long-term digital asset and immutable store of value. Any discussion of Bitcoin’s energy usage must take into account this unique use case which to many cannot even be quantified in a financial calculation.
While Bitcoin mining does currently consume over 200 tera-watts per hour and emits an estimated 41–72 metric tonnes of CO2 per year according to Messari and digiconomist, there are some myths or misinformation about Bitcoin’s environmental impact and the cryptocurrency ecosystem as a whole. The graph below shows Bitcoin’s carbon production compared to similar industries like Gold and the global banking system, in addition to various other industries and a variety of household appliances. These estimates for carbon emissions or overall environmental impacts also largely depend on where the industry’s energy is produced, in addition to the usage of renewable energy.
Despite the large amount of energy consumed by Bitcoin miners, renewables power a substantial amount of the network and the Bitcoin mining industry continues to move rapidly towards utilizing renewables more heavily. It is simply more cost-efficient for miners to do so with major incentives for this to happen as the income for miners reduces and electricity costs increase over time.
Background to Cryptocurrency Energy Concerns
The vast majority of energy concerns around blockchain technology stem from Bitcoin, as it is a Proof of Work (PoW) algorithm that uses high-power Application-Specific Integrated Circuit (ASIC) miners. These types of miners are needed as the difficulty of mining Bitcoin has drastically increased since its creation in 2009, but a single ASIC miner typically uses 2–5 kilowatts per hour. This is what the Bitcoin or crypto-related energy concerns stem from since large-scale miners or mining corporations often have mining warehouses similar to the image below which can have 1000s or 10s of thousands of Bitcoin ASIC miners.
Mining Background
Proof of Work cryptocurrencies use blockchain technology to essentially guess the correct cryptographic hashes for each block that is created within a blockchain. All the miners share a distributed ledger that is composed of all previous blocks, and all miners compete to guess the correct hash for the next block. This is why a miner’s performance is rated in hashes per second, as this metric is how many guesses or hashes a miner can process in a second. Once correct, the block is verified by other miners on the network via the ledger before being accepted or rejected (uncle blocks). Accepted blocks are then added to the blockchain, appended to the network’s ledger, and the process starts again for the next block. With each new block, thousands of transactions can be processed, but the overall throughput of the blockchain depends on both the number of transactions per block and the block time.
Bitcoin Background
When Bitcoin was originally created in 2009, it could be mined with consumer graphical processing units (GPU) also referred to as graphic cards, similar to how Ethereum is mined. However, as the mining difficulty has increased throughout the years, GPUs are no longer powerful or efficient enough to provide good profits for those who mine bitcoin, so ASICs are used to provide higher hashrate density as ASICs are about half the size of a standard office computer and can provide over a hundred tera-hashes per second, while a GPU on Ethereum’s algorithm (Ethash) might average 60–100 mega-hashes per second depending on the specific type of card. However, with this improvement in algorithm-specific efficiency and hashrate density, comes an increase in power. https://whattomine.com/miners
Proof of Work Algorithms and Energy
When looking at PoW algorithms, the largest networks are Bitcoin and Ethereum, with Bitcoin dominating 47% of the crypto space and Ethereum at almost 18% as of June 2022. The market caps for each are almost 600 billion USD for Bitcoin and 220 billion USD for Ethereum. Other notable PoW coins in descending order include DogeCoin, Litecoin, Monero, Bitcoin Cash, and Ethereum Classic. However, these coins combined have a market cap of only 25 billion and combined market dominance of 2% as of June 2022. Top PoW Tokens by Market Capitalization | CoinMarketCap
Although Ethereum is in the process of transitioning to Proof of Stake (PoS) which uses less energy, Bitcoin and smaller PoW coins still collectively make up about half or more of the total crypto market cap, and because of the large difference in market values, Bitcoin will be the primary focus when discussing PoW algorithms.
The current (6/2022) estimate of Bitcoin’s energy consumption by digiconomist is 204.5 tera-watts per hour (TWh), which is almost double the estimate just a year ago of 125 TWh in June 2021, and almost triple the estimate in the beginning of 2021 of 77.8 TWh as shown in the chart below.
Although the energy consumption may seem to be increasing at an exponential rate over the last few years, Bitcoin is using renewable energy sources as a surprisingly substantial amount of its energy consumption, estimated as anywhere between 58% to 74%. It’s also worth noting that the estimated 200+ TWh/Year is less than 0.2% of the world’s energy consumption.
While Bitcoin can’t pivot its algorithm to Proof of Stake like Ethereum is aiming to, its energy consumption can be further shifted towards renewable energy sources, although it is already one of the greenest industries in the world as illustrated throughout the images below.
These images highlight how Bitcoin’s energy usage compares to other industries, although the estimate for Bitcoin’s energy at 79 TWh is now outdated with new estimates over 200 TWh, similar to the gold and jewelry industries, but with the potential to impact a larger population than the gold and jewelry industries do. It’s also important to note that the central banking systems use approximately 260 TWh of energy. While central banking is widely available in first-world countries, third-world countries often don’t have extensive access, if any, to banking infrastructure, but cryptocurrencies like Bitcoin or stablecoins can offer banking to anyone with an Internet connection.
Excess Energy Potential for Bitcoin
To provide insight into Bitcoin’s renewable energy sources, it’s important to understand the incentive for PoW miners to use renewable energy sources. When a new generation or model of ASICs are released, they typically have the most hashrate density and best efficiency compared to older miners, therefore more expensive (grid) power is reasonable to use as the increased profits are worth the higher energy costs.
However, older miners don’t bring much in profits and use roughly the same energy, sometimes more, so to keep these miners running it’s important for miners to find super cheap power, which is typically renewable sources for mining long-term. According to statista.com, the price per Kilowatt-hour (KWh) for renewable energy was anywhere from $0.039 to $0.108 in 2020, which drastically dropped to an average of $0.039 per KWh in 2021 as compared to an average of $0.07–0.08 for 2020. However, these numbers are likely lower in 2022, and any of these numbers are cheaper than the global averages for grid power, which is $0.133 per KWh for residential power and $0.124 per KWh for businesses.
Miners could also have a hybrid approach to mining by utilizing excess renewable energy during the day, such as solar/wind/hydro, that can’t be used by normal grid usage and then use the grid’s power at night while power companies may have cheaper rates due to less grid demand.
Many will argue that “excess renewable energy” doesn’t exist and that miners using renewable energy are taking energy that could be used to power the grid. Realistically, renewable energy like solar, wind, and hydro can have large amounts of energy output during the day or during certain seasons and if the grid doesn’t have an immediate need for the energy or a reliable place to store it, the energy is therefore wasted or unused. However, if this excess energy can be used to power mining operations, it’s less demand miners would take from the grid. Since it’s using energy that would be otherwise wasted, it could instead be sold at a cheap rate to miners so it’s not a complete loss for the utility companies or owners of the renewable source.
This can be illustrated by a few articles, the first by dw.com which tells the story of a hydroelectric power farm that’s been running for 30 years in Costa Rica that had to shut down in 2020 due to the country having a “surplus” of energy and refused to buy the power the owner was generating. Exporting the excess power to neighboring countries would typically be an option, but the neighboring country of Nicaragua doesn’t have the infrastructure to import the energy. Therefore, the owner decided to convert the hydropower into a bitcoin mining farm after 3 months of inactivity. He now says they’re improving their profitability and thinks they won’t go back to the plant’s original purpose, powering the grid.
Another example of using “excess” energy for Bitcoin mining is a couple of graduate students from Texas A&M who harnessed the excess flare gas from oil mining to power Bitcoin mining operations and mined $4 million according to the article from engineering.tamu.edu. In short, oil mining is a huge market in Texas but when drilling encounters any natural gas pockets or similar highly pressurized gasses, they pipe it above the surface and burn it to relieve pressure from mining equipment. This excess gas that is burned is known as flare gas and is shown in the image below.
This flare gas is essentially wasted, but instead of burning it, the natural gas can be piped to a generator and used to create power which was used for Bitcoin mining.
The image above is the two students with their Bitcoin ASICs in a storage container, but other large companies are also using similar “wasted” energy to power Bitcoin mining, as outlined in an article by coindesk.com.
Additionally, the article by coindesk describes how Bitcoin mining could be used as a flexible load to accommodate more renewable energy sources and their fluctuating power generation. Because Bitcoin miners don’t have any negative impacts when power is cut, unlike other industries, they can be described as an “interruptible load” that can run during times of excess power generation and can then be powered off during times of higher power demand or lower generation. This ensures the excess energy isn’t being wasted, while still providing stable renewable power to the grid. While this may seem like a profit cut to miners when their miners are turned off, it could be worth the power incentive or potential deals that could be made with utility companies for a fixed payout, agreed-upon minimum runtime, or similar factors.
A great example of Bitcoin miners working with utility companies is illustrated in an article by CoinDesk that shows Core Scientific’s mining facility in Kentucky. The mining facility works with utility companies by being able to quickly power off miners during peak demand or when requested, and the mining facility benefits from the utility companies offering cheaper rates. This situation is mutually beneficial for both parties as it’s easier to power down the mining facility than to ramp up energy production for the utility companies, and the savings in electricity costs likely outweighs the loss in profits for the mining facility during peak demand from the grid. According to their website, Core Scientific has several similar mining facilities, similar to the one pictured below, with a total of 85k+ Bitcoin miners that make up roughly 10% of Bitcoin’s network.
Energy Efficient vs Environment-Friendly
Although it has been established that Bitcoin consumes large amounts of energy, it can’t be assumed that Bitcoin is bad for the environment. Like any industry that uses electricity, the environmental impact to produce the electricity is entirely dependent on where and how the energy is being produced. According to eia.gov, the energy production of the United States is only 12% renewable as of 2021, but an article by climatecouncil.org shows that other countries like Costa Rica, Uruguay and Nicaragua produce over 90% of their energy from renewables.
This proves the point that although Bitcoin mining uses lots of power, if the energy source is clean/renewable then the environmental impact is much lower than energy production utilizing fossil fuels like coal and natural gas. As discussed, there’s a large incentive for Bitcoin miners to use renewable energy sources, as they are often much cheaper and often have excess production that can’t be utilized by the grid’s demand. This incentive combined with a global effort to invest in renewable energy production is likely the reason why Bitcoin has become more “green” and has relied more on renewable sources throughout recent years.
Compared to other industries, cryptocurrency power demands are put into a different perspective. Bitcoin is often referred as “Digital Gold,” however, Gold is destructive to the environment and utilizes significant amounts of energy, fossil fuels, and water, often leaving water sources polluted after extraction according to an article by Brilliant Earth, which speaks volumes as Brilliant Earth is a jeweler who sells gold. While the Gold industry’s energy consumption isn’t much more than current Bitcoin consumption, Messari estimates the Gold industry produces 122 metric tonnes of carbon emissions per year, roughly 3x the amount of Bitcoin.
The Banking Sector also uses significantly more power across its supply chain and infrastructure than cryptocurrency. If as many believe, crypto can eventually replace or supplement this sector, this would actually represent a global net energy gain. It’s also important to note that Messari estimates the banking industry produces 130 metric tonnes of carbon emissions per year, significantly more than Bitcoin and even slightly more than the gold industry. Every person that switches from traditional banking to digital alternatives as cryptocurrencies become more viable, would represent more energy conservation overall based on the current consumption rates of each industry.
Many major nations already plan to adopt Central Bank Digital Currencies (CBDCs), essentially government-issued digital currencies utilizing blockchain technology similar to China’s initiative within the sector. The International Monetary Fund (IMF) has even stated that digital assets can be a more effective payment solution than current credit and debit card systems as discussed in an article by finbold.com.
Proof of Stake
To put Proof of Stake (PoS) vs Proof of Work (PoW) energy usage into perspective, an article by the Ethereum Foundation at ethereum.org, says the Ethereum 2.0 merge will reduce the network’s overall energy usage by roughly 99.95%. The Ethereum 2.0 merge provides a good example of how current PoW blockchains could pivot towards PoS with active development and will be the primary example when discussing PoW vs. PoS.
Within the article by the Ethereum Foundation, they confirm that PoS has been their goal since the beginning but has used PoW as a crutch because the Foundation didn’t want to sacrifice network security and decentralization. While this may be true, Ethereum is the second largest Proof of Work blockchain by market share (after Bitcoin) and has delayed the 2.0 merge several times, likely to ensure a smooth transition for the blockchain and its investors/users.
Ethereum 2.0 is certainly not the first blockchain to implement Proof of Stake, the concept and technology have been around for roughly a decade or longer, but changing a blockchain’s consensus mechanism is arguably harder than just designing a PoS network, especially as Ethereum has got such a large market share. Ethereum has currently switched many of its test nets to PoS and is in the process of transitioning some of its oldest test nets to PoS by merging them with the Beacon Chain that has been running alongside the PoW mainnet since late 2020.
In general, Proof of Stake is a consensus mechanism similar to Proof of Work, but instead of powerful miners fighting for hashrate and competing to mine blocks and receive network rewards, Proof of Stake has “validators” that are closer to a typical computer or server that uses staked equity to “vote” on the blockchain or process transactions on a more central ledger that the blockchain is comprised of.
The amount of staked equity depends on the blockchain, but for a network like ETH 2.0, the current minimum is 32 ETH to become a validator and is used as collateral and incentive to be a fair validator. If a validator begins to act maliciously by frequently missing transactions or attacking the network in other ways such as double-signing transactions, then their staked equity can be “slashed” which involves a partial or complete loss of the validator’s equity. The Ethereum Foundation claims the network can be run on low-power devices like Raspberry Pi’s, but more powerful hardware may be desired to ensure validator reliability/up-time on the network.
The current estimated usage of Ethereum is roughly 50 Terawatt-hours per year (TWh/yr) with the peak of 2022 at around 100 TWh/yr. After the Ethereum 2.0 merge, it is estimated that the overall energy consumption of the blockchain will be closer to 0.01 TWh/yr, a several thousand-time decrease in power, but these estimates might change post-merge as hard data is calculated and as more validators come online. This highlights one of the many reasons why PoS is sought-after as the future consensus mechanism.
Proof of Work vs Proof of Stake
Proof of Work (PoW) and Proof of Stake (PoS) are two of the most popular consensus solutions for blockchain. Proof of Work, as discussed, is processed by power-hungry hardware like Bitcoin ASICs or GPUs for blockchains like Ethereum and other smaller blockchains. This consensus mechanism has worked well for Bitcoin, but as it consumes a substantial amount of energy and is the center of blockchain’s energy concerns, Proof of Stake has become more popular with coins like Solana, Cardano, and Ethereum 2.0. However, both PoS and PoW have inherent security flaws like 51% attacks for PoW where if 51%+ of the blockchain’s hashrate becomes centralized in a single mining pool, blocks could be falsified, or how PoS relies on more centralized nodes over decentralized miners.
There’s a good debate about PoW vs PoS on youtube, called The PoW vs. PoS Debate, and one of the first topics discussed was a blockchain’s “hardness” to recover from attacks, and which consensus mechanism is harder to attack, or requires more capital to attack. Lyn from the debate argued that a PoW chain like Bitcoin can more easily recover after a power or Internet outage simply by restarting miners, whereas many PoS validator nodes must be manually restarted or be re-validated/res-synced with the blockchain.
Some PoS blockchain validators are only rewarded when they have a certain percentage of up-time, and PoS validators can also partially or completely lose their stake when offline for extended periods of time. While this could seem like a good security measure and incentive to increase/maintain node up-time, it could be hard to maintain network security if several large validators go down at once and validators with a smaller stake are left to run or manipulate the network.
Additionally, if these large validators are custodial stakes, hundreds of people may lose funds if the stake is lost, as custodial stake users are at mercy of the validator’s performance. While something similar is possible with PoW coins through a substantial decrease in hashrate, PoW miners tend to be more spread out globally, as many PoW blockchains like Ethereum and smaller coins have a low barrier to entry. A lower barrier to entry is good for decentralization as more people can buy miners and actively mine with PoW coins, but one downside that Justin from the debate pointed out is that a lower barrier to entry requires less capital for a 51% attack or similar attack.
However, the barrier to entry depends on the blockchain. For instance, the barrier to entry for custodial PoS can be very low, it’s typically just a set minimum amount of coins to stake, but when looking at a PoS coin like Ethereum, the minimum to stake is 32 ETH which is a pretty steep barrier for most people. Running a proper validator node can also have varying barriers to entry. For smaller blockchains, a validator node might just be a Raspberry Pi and a few dollars worth of crypto (Energi Blockchain) depending on price and minimum staking amount, but a validator node on a blockchain like Solana can cost thousands in hardware costs, requires fast/reliable internet, and still has some power costs associated, in addition to needing 1 SOL per day to vote on the blockchain (~$50 daily). Because of high PoS barriers to entry, one could argue that many PoS tend to be more centralized as the blockchain is run by 100s or 1000s of validators (1000+ for Solana), while a network like Bitcoin is estimated to have over a million miners around the world.
Proof of Stake Security
As discussed in a medium.com article, Proof of Stake’s security is typically misunderstood and compared directly to Proof of Work. In reality, buying more hardware and attempting a 51% attack on a PoW blockchain is different than acquiring enough staked tokens/coins for a similar attack on a PoS blockchain. This is because the majority of any PoS token or cryptocurrency is likely already staked, and therefore the only tokens an attacker could acquire would be the tokens not already staked, likely on various exchanges or in private wallets. This means that a blockchain with a large percentage of its tokens staked (A larger “Staking Ratio”) tends to be more secure, as there is less liquidity that attackers could quickly acquire.
For example, Binance’s BNB chain has over 80% of its tokens staked, and Algorand has almost 90% staked as of July 2022. Therefore, the only liquidity that attackers could buy would be the remaining 10–20% which is likely spread across different exchanges in smaller amounts. However, a blockchain like Solana which only has 25% staked could be more susceptible to an attack if the attacker managed to acquire 33%+ of the crypto’s market cap. With at least 33% of the market cap an attacker could slow or halt consensus, but 66%+ is generally needed for a double-spend or more extensive attack. However, if an attacker did manage to acquire 66%+ of a PoS blockchain’s market cap, there are still risks for the attacker, typically in the form of “slashing” where a percentage of a perceived attacker’s tokens are burned or removed from their account. This percentage varies from 0.01% to 100% depending on the blockchain, attacker’s history, and type of attack. Depending on the slashing percentage, this is the risk for attackers. They will typically try to profit much more than this percentage to cover potential losses, whether financial or social, and to ensure an acceptable profit to risk ratio.
Despite Proof of Stake’s seemingly secure protocol, there’s a difference between blockchain attacks and hacks on centralized nodes. As cryptocurrencies have continued to gain popularity, lots of hacks have occurred within the crypto space.
The most recent and one of the largest was in March of 2022 with the hack of Axie Infinity’s Ronin Bridge Hack. Axie Infinity is a blockchain-based game where you buy/sell and train “Axie” creatures (Based on Axolotls), which are treated as assets on a blockchain, similar to Non-Fungible Tokens (NFTs). While the individual Axie’s and the game’s currency are held in a player’s Ronin Wallet after being bought or sold, the Ronin blockchain (Side chain of Ethereum) processing all of the Axie buying/selling and other game transactions is quite centralized. The entire blockchain is run by 9 validator nodes with 50%+ of nodes needed to perform hacks/attacks. The hacker got access to a system that operates four of the nodes and exploited a bug to access a fifth node.
This is a perfect example of why decentralization is important because if the Ronin blockchain had 100s or 1000s or validators it would at least be much harder for a hacker to obtain access to 50%+ of the nodes. The hacker stole roughly 173,600 Ethereum and 25.5 million dollars of USDC (USD Stablecoin), a net value of 615–625 million at the time of the hack. Despite this, it took a week for the development team to discover the hack and they pledged to increase the required nodes needed for an attack from 50%+ (5/9) to 8/9 (88.8%+), meaning a future hacker would need to have access to all but one node to perform a hack.
However, this change likely isn’t enough to prevent future hacks and the development team seems to be sacrificing security to keep up with user demand and functionality. Having more nodes would require more maintenance and could result in a delay in consensus between nodes, but ultimately security should be a priority for the development team as the player base and Ronin network continues to scale.
Proof of Work Security
While centralized PoS validators could (and have) been liable to hacks, PoW coins like Bitcoin could fall victim to 51% attacks. To counter this point, Lyn claims that the largest Bitcoin mining farm in the world only has 1% of the total network hashrate, which may be even less currently as the total Bitcoin hashrate is almost 213 ExaHash per second as of June 2022. However, the more likely attacker for a 51% attack on Bitcoin would be large mining pools, as smaller miners typically join mining pools and it’s a more centralized amount of hashrate that splits rewards between individual miners. The current hashrate distribution is shown below, with the largest pool accounting for 22.5% of Bitcoin’s hashrate.
According to bitcoinist.com, the ASIC manufacturer BitMain is approaching the 51% mark, at around 45% if they moved all of their hashrate to just Bitcoin, but the only time a company has crossed the 51% hashrate for Bitcoin was in 2014 by a company named Ghash who provided mining pools for smaller miners. However, the company encouraged miners to move their hashrate to a different pool and a 51% attack didn’t occur. The only 51% attack to occur on Bitcoin’s SHA256 algorithm was on hard fork chains such as Bitcoin SV (BSV) and Bitcoin Gold (BTG), as they share the same algorithm as Bitcoin, and Bitcoin’s hashrate was likely moved to these smaller chains to perform the attacks.
Scalability
Despite the energy Bitcoin consumes, Bitcoin’s overall throughput is still quite low when compared to Centralized Finance (CeFi), Proof of Stake, or similar consensus mechanisms. As discussed in an article by towardsdatascience.com, Visa’s throughput is roughly 1,700 Transactions per Second (TPS), and in comparison Bitcoin’s average is 4.6 TPS. Although Bitcoin processes an average of almost 2,800 transactions per block, Bitcoin’s block time is 10 minutes which greatly reduces its throughput. However, Bitcoin likely wasn’t designed for high throughput, but it has proven to be quite secure and immutable while other blockchains or CeFi equivalents have been susceptible to various hacks. This article then gives various scenarios to how Bitcoin could match Visa’s throughput by either increasing its average transactions per block from 2,800 to over 1 billion transactions per block or by decreasing its block time from 10 minutes to under 1.6 seconds.
However, both of these scenarios are cause for propagation delays, as a block needs to be sent to over 99% of the network’s miners to ensure proper security. Increasing Bitcoin’s average transactions per block to meet Visa’s throughput would result in a block size increase from 1 megabyte of data to 377.5 megabytes of data, which would greatly hinder the network’s ability to propagate new blocks to miners with slow internet, and could cause bandwidth congestion for others.
Additionally, the size increase is near impossible as even Bitcoin’s current Segwit protocol can only scale to around 4 megabytes per block. Decreasing Bitcoin’s block time also has a similar effect since propagating 1 megabyte of data will take the same time to propagate, but if a new block is made every 1.6 seconds then a new block could be made before the previous block was received by 99%+ of the network.
Essentially, the security of Bitcoin hinges on the fact that new blocks can propagate easily and longer block times means the newest block can fully propagate and be verified before a newer block is made. The visual for these various scenarios are shown below but are unlikely as Bitcoin is intentionally hard-coded to function as-is by its creator to maintain the network’s security.
Despite Bitcoin’s possible shortcomings in scalability, Proof of Stake was designed with scalability in mind, and according to data by analyticsinsight.net, many PoS blockchains have met or surpassed Visa’s throughput. These blockchains include Cosmos with 10,000 TPS, Avalanche with 5,000 TPS, EOS with 4,000 TPS, and Solana with 2,825 TPS. Additionally, Ethereum 2.0 is estimated at 10,000 TPS with a current PoW transaction throughput of 12–25 TPS. Runner-Ups to Visa’s transaction throughput standard include PoS blockchains like XRP and Polkadot with 1,500 and 1,000 TPS respectively.
With various PoS blockchains, confirmation time is also an important metric. Confirmation time is essentially how long it takes for a transaction to be included within a verified block on the blockchain. For most networks this metric’s maximum could be the same as the block time, however, some networks require several blocks after a transaction’s block to confirm it’s a valid transaction. This metric can range anywhere from 0.4 seconds for Solana, 0.5 seconds for EOS, 1–2 seconds for Avalanche, 7 seconds for Cosmos, up to 6 minutes for Ethereum as PoW, or up to 10 minutes for Bitcoin.
In short, scalability will continue to improve for PoS or similar/newer consensus mechanisms but Bitcoin still has a purpose in the crypto space, despite its limited scalability. Bitcoin’s slow block time could even be considered a good security measure as potential attackers would have 10 minutes between blocks, whereas on a faster network an attacker has the potential to produce invalid blocks at a much faster rate which could cause more extensive damage to the network if the blocks were accepted.
Future of Bitcoin
Despite the multitude of PoW coins, the main topic of blockchain’s energy concerns is centered around Bitcoin, as it alone dominates roughly half the crypto market cap. Even the second largest PoW coin, Ethereum, aims to drastically reduce energy usage by becoming Proof of Stake (PoS) with its deployment of Ethereum 2.0. This transition to lower energy usage and scalability has always been the aim of Ethereum with PoW as a temporary solution, but this is only possible as Ethereum has an active development team that can implement this change. In comparison, Bitcoin is hard-coded to be used and operated as-is by its anonymous creator with the pseudonym of Satoshi Nakamoto. This is a feature of the system, as it makes Bitcoin highly decentralized and entirely secure from external attack even if it were to come from a nation-state. This decentralization means Bitcoin can’t just change its algorithm or infrastructure, but instead the energy used to mine and process transactions on Bitcoin’s network needs to shift to renewable sources.
The Bottom Line
Both Proof of Work and Proof of Stake have viable use cases and various tradeoffs. While Proof of Stake is more energy efficient and therefore considered more scalable (and sustainable), Proof of Stake networks haven’t been proven to be efficient at large scales yet, although this could change after a successful ETH 2.0 merge.
While lots of Proof of Stake blockchains exist and function, the throughput of these systems hasn’t been extensively tested. Mny PoS networks like Solana are growing as Web3 and Non-Fungible Tokens (NFTs) continue to become popular and users attempt to find alternatives to expensive Ethereum network fees during congestion. While PoW algorithms may not have the same scalability of a massive PoS blockchain as ETH 2.0 seeks to be, PoW blockchains are simple, secure and are proven to work. Bitcoin may use massive amounts of energy in comparison but it’s generally fast enough, with low transaction fees, and an unmatched level of decentralization around the globe.
Bitcoin does need to become more green, but estimates show this is happening as miners are incentivized to find cheap energy in the form of renewable sources or excess energy production that can be sold to Bitcoin miners instead of wasted. Despite energy concerns, Bitcoin still has a solid position as a hardened and timeless store of value, often referred to as “digital gold.” The beauty of Bitcoin is that it just works, it’s simple, elegant, and secure.
Proof of Stake systems and various derivations can and likely will be the dominant consensus mechanisms in the future, but Proof of Stake systems still have a ways to go with providing hardened decentralization and proven scalability. After years of development, the ETH 2.0 merge is make-or-break for PoS and the overall crypto ecosystem, a successful merge could be a turning point for the crypto ecosystem but a failed merge could be a substantial setback for PoS adoption, not only for Ethereum.
