Ecological impacts of NFTs
Exploring the current & future effects that Non-fungible Tokens (NFTs) have on the health of the natural world.
Since the vast majority of NFT-related content on the internet is different explanations of what NFTs are, I won’t attempt to do that. This post assumes a basic understanding of them. Even so, some of the environmental impacts require a little context about how blockchains and NFTs function.
If you do not have any prior knowledge on cryptocurrencies, blockchains or NFTs, I recommend watching this easy-to-understand video by Johnny Harris (5:40 onwards).
How do NFTs cause emissions?
NFTs create direct and measurable emissions on the expense of our biosphere by being a part of the energy consumption of blockchains. Blockchains are often referred to as “cryptocurrencies” and the online ecosystem of blockchains is called “web3”. Physically these blockchains are computers around the world that are using electric energy in the form of computing power to secure & implement transactions. Another indirect source of emissions is the life-cycle of the hardware running the electricity-consuming blockchain with NFTs on it.
When saying that these NFTs are “on” a blockchain, it means NFT are attached to and saved onto it. NFTs are inseparable from blockchains, so the energy consumption of NFTs is based on that of the blockchain hosting it. The differences in blockchains’ carbon footprints is very significant for reasons discussed below.
How do blockchains cause emissions?
Regardless of what financial technology is used to maintain records & transfer value over the internet, it consumes electricity. In addition to these tasks, energy plays a key role in securing the most significant blockchains. Their energy-intensity varies greatly depending on the consensus mechanism used by the blockchain. There are Proof-of-Work (POW) blockchains like Bitcoin & Ethereum and then there are Proof-of-Stake (POS) blockchains like Cardano, Solana & Tezos.
It is notable that in most cases there is a possibility to change or improve the consensus mechanism over time. This is the case with the chain with most NFTs on it: Ethereum. The non-profit organization ‘Ethereum Foundation’ has estimated that moving from it’s current consensus mechanism of Proof-of-Work (PoW) to Proof-of-Stake (PoS) will cut its emissions by 99.95%.
The sensational headlines and fear-mongering about the energy use of cryptocurrencies have been blown out of proportion for the entire technology’s short history. According to analyst Lyn Alden, this is something that “often comes down to them not understanding the scaling process that the network is going through.” [10]
Why do blockchains cause emissions?
First I will have to clarify that some blockchains use next to nothing in energy (like Tezos that I have minted my NFTs on), but since they are not the issue, I will focus on explaining the ones that do present an issue.
The graphic below features some of these low-energy chains with VisaNet Electronic payment network & Bitcoin for comparison. Note that the graph is exponential along both X & Y axis.
The purpose of these proof-of-anythings is to establish security and a consistent record among a peer-to-peer network. The consensus mechanism can be thought of as a voting mechanism according to which the nodes settle on which transactions are valid and in which order the transactions are saved to the blockchain [3].
In the more energy-intensive POW model each network participant “verifies” themselves by expending energy to compute trivial mathematics problems and in exchange is given the right to calculate actual traffic on the network and to earn crypto in the process. This way, energy use functions as a collective safety precaution granting a kind of “voting right” to the blockchain.
Proof-of-Work prevents a single person from making a 1000 accounts to appear as 1000 miners because that would require 1000x more energy, computing power and hardware. If a malicious actor wanted to manipulate or in any how alter the blockchain, they would need to control over 51% of all the active resources of the network (this is referred to as a “51% attack”) – the complexity and cost of organising such a simultaneous global hack is why blockchains are so secure.
In the Proof-of-Stake model energy is replaced by a “staked” (locked, deposited) amount of cryptocurrency. Each miner needs to deposit f.ex. Ethereum in exchange for the right to calculate Ethereum transactions and thereby earn more of it. Significant preparations and steps for the implementation of POS have already been taken and the energy consumption related part of the upgrade towards ‘Ethereum 2.0’ is estimated to be issued in Q1 of 2022 [3].
Altogether the emissions of cryptocurrencies have been sensationalised in the press over the years. Analyst Lyn Alden believes this has been because “it often comes down to them not understanding the scaling process that the network is going through.” [10]
Blockchain emissions VS. existing technology
With traditional credit cards this issue is not discussed, because credit card transactions only use a tiny amounts of energy. The difference with crypto is that what it represents is not the transaction, but the entire financial infrastructure. The tiny amount of energy per transaction of credit cards is true, but does not take into account the energy used by the banking ecosystem that every transaction relies upon: data centers, bank branches, card networks etc.
Even though the apparent emissions are higher, the reality is more complicated as blockchains don’t need banks or payment processors to rely on and externalise the energy consumption upon. There is only the public log of activity (ledger) and it is completely maintained by a peer-to-peer network. The comparison of credit cards vs. Bitcoin is in this sense highly misleading.
It is often stated that crypto uses “lots of energy” and this implies, that the same could be achieved with less. In the case of BTC’s role as “digital gold”, a fitting comparison would therefore be the gold market – including the mining and trade of it.
Context for energy consumption of blockchains
What are Ethereum’s actual current emissions? How much is this supposed 100% from which 99.95% is going to be deducted and how does this amount compare to energy use in daily life?
At the time of writing (22nd of October 2021) Bitcoin’s blockchain is using electricity at a pace of 104.44 TWh per year [1]. It is certainly a substantial amount, but still much less than the “legacy” industries it is most closely compared to. Most importantly moving forward, Bitcoin and other cryptocurrencies have many key advantages when it comes to the sourcing of this energy.
In Ethereum’s current Proof-of-Work (PoW) -model one (1) transaction requires the same amount of electricity as powering a house for days. However, this is not a very good metric and estimates of the amount of days at Ethereum’s current level of usage vary between 2.8 – 6.1 days.In case of the future’s updated Proof-of-Stake (PoS) Ethereum, the equivalent amount of energy of one transaction is going to be “about 20 minutes of TV” [5].
For comparison, the same transaction on the mammoth chain of Bitcoin uses something in the ballpark of “38 house-days worth” [5]. This figure is staggering and on a global scale Bitcoin accounts for 0.40% of global electricity consumption (note: not total energy consumption, but total electricity consumption) [9].
The energy use of Proof-of-Work blockchains (below) is not as straight-forward as many environmentally concerned reports on the matter indicate. A similar dynamic can also be seen in the decreasing wattage/transaction of POS chains compared in the first graph above (UCL CBT).
Energy per transaction is not fixed
The energy consumption of POW chains is not directly correlated with the amount of transactions processed (transaction throughput). Regardless of if there are more transactions to calculate than the blockchain can fit (full blocks) or if there are none at all (empty blocks), a large amount of energy is expended because this is necessary to keep the network safe from a 51% attack. For the technically inclined, all nodes are kept busy regardless of transaction throughput by way of a parameter called mining difficulty. Usually this is just some arbitrary, useless guessing game like in the case of Bitcoin & Ethereum, but other alternatives such as “Proof-of-Useful-Work” have emerged and can be expected to prevail in the long run. The idea of POUW is to crowd-source computation for very complex problems like Folding@Home is doing: “Together, we have created the most powerful supercomputer on the planet and are using it to help understand SARS-CoV-2/COVID-19 and to develop new therapies.”
In conclusion of the question of electricity consumption I quote the research of J. Sedmeir et al.: “While their energy consumption is, indeed, massive, particularly when compared to the number of transactions they can operate, we found that they do not pose a large threat to the climate, mainly because the energy consumption of PoW blockchains does not increase substantially when they process more transactions.” [14]
Solutions & innovations to blockchains’ electricity use
These amounts of energy are quite significant, but they are also being addressed. In addition to Ethereum (Eth) network updating it’s consensus mechanism, Bitcoin’s high-energy use and resulting reputation as a polluter has incentivised miners to adopt renewable energy. Current estimates of the share of renewable energy used in its mining vary between 40% and 75% [7]. On the bright side the energy needed for mining accounts for 60-70% of the total expenses and this gives miners a clear motive to collect their energy themselves by using renewables [4].
Another positive aspect in the adoption of more sustainable energy sources for mining is the lack of surrounding infrastructure needed. Traditionally industries require access to raw ingredients, water ways, logistics hubs like railways, harbours, airports and finally distribution channels like cities. Blockchain miners don’t need those, so the industry can take advantage of energy sources that aren’t viable for others. As an example, the Icelandic Blockchain Foundation says that 8% of all Bitcoins are mined there thanks to the geothermal energy and decreased cooling costs in the cold environment (continually running mining computers need cooling). Another positive case of untapped energy being used to mine cryptocurrencies is El Salvador’s president Nayib Bukele’s succesfully tested plan to use geothermal from El Salvador’s volcanoes to mine Bitcoin [6]. Other visions include renewable energy producers mining with their excess energy in times of low electricity demand.
Artwork by NFT pioneer Alotta Money celebrating El Salvador’s adoption of Bitcoin as legal tender (click the image to see full artwork)
What about emissions related to hardware?
In addition to electrical energy used for upkeeping the network, cryptocurrencies cause collateral damage on the environment by contributing to the demand of hardware. The ecological impact of cryptocurrency mining is the e-waste generated by updating the mining computers as the difficulty of mining increases. As more miners come online, the difficulty of mining increases and new machines are necessary to maintain steady revenue.
Manufacturing IT-devices consumes massive amounts of resources: from geologic extraction to chemical refining to building different components and to finally assembling end-use devices. Between the computers used at each end of the network there is also the entire hardware of our global IT-infrastructure that causes emissions. This is shared by any and all things online – the host environment of all online content is not only your computer and the service provider’s servers, but also the planet-wide hardware ecosystem known as “the internet”.
With regards to NFTs, like any other computer activity, they lead to use of energy to power devices and to make new devices. As all things are increasingly digitised, people update their hardware to better access resources, function in society and interact with each other. Blaming this on blockchains or NFTs is misguided, but they too are surely contributing towards this development.
What about specifically NFTs on the blockchains?
Fundamentally NFTs are nothing but smart-contracts on a blockchain like Ethereum with somekind of content (media) attached to them. The process of creating an NFT is called “minting”. This means adding the NFT’s code to the (most often Ethereum) blockchain. Generally speaking NFTs and their smart contracts are more complicated (=include more data) than those of regular cryptocurrency tokens. Because NFTs contain more data, employing them and engaging in market activity with them involves higher transaction fees and higher electricity consumption than regular tokens – however this difference is negligible as most of the electricity consumption is because of the POW, not the network activity itself.
When assessing the sustainability or un-sustainability of these technologies, it is agreed that they use a large amount of energy. Statements like ”a large amount” contain a comparison to something comparable that uses less energy.
In the case of NFTs, they are such a new invention that there are no directly comparable companies or services to which the carbon footprint of NFTs could be compared to. NFTs cover a wide spectrum of applications and roles within the digital world, but even as NFTs would win the comparison, it is not useful to compare the carbon footprint of the art market, internet memes or digital music streaming to that of NFTs. We really don’t know yet what NFTs will turn into and what their ecological impact will be. Neither do we know what they might displace as existing technologies or industries.
Environmental impact of NFTs as natively digital objects
In general, our daily lives depend on a wide internet infrastructure and the mining of materials from the Earth to maintain and update the network of machines. Most of the assets circulating in the internet are only using internet as a carrier medium in order to distribute the data globally. This meaning, that a PDF or a photo is only uploaded to the internet because it needs to reach someone else through that channel – not because we would not be able to read papers or look at photos without the internet. However, in the case of NFTs, they are part of a paradigm shift in how assets exist digitally.
NFTs are digitally native. They don’t have a place or way to exist outside the digital space. They belong online and they can’t be offline without compromise – just like physical things can’t be online without compromise.
NFTs don’t exist as passive physical objects, but as active objects that are digitally native packets of data tied to their owners. NFTs are then never here nor there, but rather exist on the peer-to-peer blockchain network in every node simultaneously. They are stored in a long chain of data that interacts with thousands of end-user’s machines and travels through routers, cellular towers, internet hubs, a global web of fiberoptic glass cables under the oceans and satellites. This level of connectivity is unprecedented and has ecological consequences as they require some computation to exist in the first place.
In understanding the nature and “being” of these assets we have come to call NFTs, it might be useful to think about the spectrum of how “hot or cold” something is in cybernetic terms. In daily life our phones are very hot – they are connected via cellular, wifi and bluetooth. Another extreme would be a simple mp3-player that is cold – it is not connected to anything.
NFTs are very hot.
For somebody else than the creator of an NFT to show an NFT requires the functioning of at least one computer and in most cases an internet connection. Seeing as we are by now accustomed to storing everything on cloud services and an internet connection is a pre-requisite to daily life, this is not an issue in practical, nor environmental terms. This is just an aspect that is distinct to NFTs and it will cause a small increase in emissions in the future as digital frames become more popular and start replacing paintings on walls.
Discussing the current state of affairs
In this fuzzy front-end of innovation we can only contribute towards directions that we estimate will yield positive effects in terms of the environmental consequences. This is the primary reason why I have dedicated much of my time to researching and contributing in the NFT space. As a technology user and artistic researcher I see massive cultural and social potential, but as an environmentalist I also see potential for harm. If everyone worried about the ecological impact of these technologies decided not to participate in the development, then how can we assume ecological considerations to be taken into account?
In addition to the figures, I personally have no doubt that these technologies present us tools for building an overall more sustainable and egalitarian society.
My own stance on technology in general used to be closer to the anarcho-primitivist (i.e. Pentti Linkola) than the technology-optimist (i.e. Elon Musk), but for the first time in my life I truly believe we have at our hands the ingredients of a reasonably advanced, but fairly distributed and socially inclusive society. Many of the interpersonal arrangements enabled by web3 & NFTs bring new financial freedom and egalitarian opportunities to an internet currently owned, bought and sold by legacy technology companies to maximise profit from their users.
In terms of the environmental question, I believe the security and decentralised nature of these networks and NFT-assets can lead to a decreased environmental impact of human activity. Might sound contradictory and I explain this below, but this is the inevitable position of a technology optimist: trusting that the onwards march of scientific discoveries will solve more problems that it will create.
Envisioning a more sustainable future with NFTs
Lastly I would like to speculate how (on a practical level) blockchains and NFTs will create positive environmental change. Firstly and quite inevitably the decentralization of legacy governing institutions will decrease inefficiencies that arise from middlemen and overlapping mechanisms of administration.
Secondly the short history of the NFT space has already proven to offer a perfect substitute for conspicuous consumption – consumption of luxury goods to “flex” (show off). This desire to use resources on non-essential things must not be underestimated and has played a big role in the triumphant march of capitalism and consumerism in the past century. The core idea of this form of extremely wasteful “splurging” is to buy things that have a high symbolic value, but zero to little use value: like a Picasso, Lamborghini or a gold chain. The requirements are to be instantly recognisable and to carry a large amount of “cultural premium” or “prestige” that can ideally be appreciated by everyone else within the same cultural context.
NFTs are ideal for this and they solve problems associated with traditional conspicuous consumption: their ownership is instantly verifiable (unlike a rented Lambo), the authenticity is instantly verifiable (unlike a painting estimated to by Rembrandt), everyone can see it in your wallet (unlike a stack of bills) and as natively digital objects they can be bragged about online seamlessly. The last is really not the least as sharing trophies of consumption has until now always required an imperfect and often arduous process of translating the object of desire into a much lamer digital approximation like a picture. The optimistic take on this would naturally then be, that some of the most unnecessarily emissive activities on Earth (being the most extreme manifestations of the luxury segment) will decrease in amount and degree of currency as more flexing becomes natively digital.
Remote work & socialising is normalised
+ everything is getting digitised
+ people like to own things
= the metaverse
One of the legitimately most concerning sources of global emissions is transport as it accounts for “7.0 GtCO2eq of direct GHG emissions (including non-CO2 gases) in 2010 and hence was responsible for approximately 23 % of total energy-related CO2 emissions (6.7 GtCO2).” [15] The advent of the metaverse as an entirely remote environment for socializing will decrease the need and ambition of travel. As our capabilities to interact and meaningfully spend time together across physical distances become greater, the need to travel decreases. Although this may sound dystopian to some, certain countries already have significant sub-populations who primarily invest their time in non-base reality activities. So if this trend has existed before, why would NFTs make it any different? NFTs’ functionalities mean people have a way to truly own digital things – not within a game or website, but on the blockchain so they can be exchanged for real money. In other words, NFTs are not just consumption, but investment and this creates a logical reason to invest resources (inc. not just money, but time and social capital) in interactive online locations.
NFTs enable a whole new format of culture in a more secure and democratic network owned by the people and used by the people. What we, the people, decide to make of it depends on us. Not as passive consumer base voting by being customers in a corporate landscape, but as a self-governing creative hivemind.
References:
The University of Cambridge. Cambridge Bitcoin Electricity Consumption Index Cambridge Centre for Alternative Finance, https://cbeci.org/index
Galaxy Digital. On Bitcoin’s Energy Consumption: A Quantitative Approach to a Subjective Question https://docsend.com/view/adwmdeeyfvqwecj2
Oyinloye,D.P.;Teh,J.S..; Alawida, M.; Jamil, N. Blockchain Consensus: An Overview of Alternative Protocols. Symmetry 2021, 13,1363. https://doi.org/10.3390/ sym13081363
Zou, J.; Ye, B.; Qu, L.; Wang, Y.; Orgun, M.A.; Li, L. A Proof-of-Trust Consensus Protocol for Enhancing Accountability in Crowdsourcing Services. IEEE Trans. Serv. Comput. 2019, 12, 429–445.
Beekhuizen, Carl. A country’s worth of power, no more! 18.5.2021 https://blog.ethereum.org/2021/05/18/country-power-no-more/
Feng, Alice. Is cryptomining harming the environment? 27.2.2021 https://psci.princeton.edu/tips/2021/2/27/is-cryptomining-harming-the-environment
Jon Huang, Claire O’Neill and Hiroko Tabuchi. Bitcoin Uses More Electricity Than Many Countries. How Is That Possible? 9.3.2021 https://www.nytimes.com/interactive/2021/09/03/climate/bitcoin-carbon-footprint-electricity.html
Anthony Esposito Does money grow on volcanoes? El Salvador explores bitcoin mining 10.6.2021 https://www.reuters.com/technology/el-savador-exploring-volcanic-bitcoin-mining-bukele-says-2021-06-09/
Crispin, Sterling. NFTs and Crypto Art: The Sky is not Falling. 22.2.2021. https://sterlingcrispin.blogspot.com/2021/02/crypto-art-sky-is-not-falling.html
Alden, Lyn. Bitcoin’s Energy Usage Isn’t a Problem. Here’s Why.
Manca, Adriano. NFT Gas Fees Explained (Thank God!) https://visionaryart.io/mag/nft/nft-gas-fees-explained-thank-god/
Molitchu-Hou, Michael. 3D Printed NFT Art: Is It Worth the Ecological Impact? https://3dprint.com/281662/3d-printed-nft-art-is-it-worth-the-ecological-impact/
The Energy Consumption of Blockchain Technology: Beyond Myth
J. Sedlmeir et al.: The Energy Consumption of Blockchain Technology, Business & Information Systems Engineering, volume 62(6):599–608 (2020)
Intergovernmental Panel on Climate Change. (2015). Transport. In Climate Change 2014: Mitigation of Climate Change: Working Group III Contribution to the IPCC Fifth Assessment Report (pp. 599-670). Cambridge: Cambridge University Press. doi:10.1017/CBO9781107415416.014