Zumo's response to OECD's call for consultation on the environmental impact of crypto-asset and broader digital asset activity.
Zumo answered’s the OECD call for consultation on the environmental impact of crypto-asset and broader digital asset activity. You can read the contribution below, the final paper by the OECD is downloadable here. For more information on our work visit our sustainability hub here.
The monetary arguments for crypto-assets are by now familiar and speak to the ‘S’ & ‘G’ components of the ESG acronym: faster, cheaper, disintermediated settlements; borderless, universal, permissionless access; a citizen-empowered counterweight to financial gatekeeping, corruption & unchallenged state monopolies on money and monetary policy.
More broadly, however, as it evolves, the blockchain-facilitated digital asset is also emerging as the rail and unit for new models of commerce and organisation, ranging from the likes of real-world asset tokenisation to blockchain-verified digital identity and smart contract governance. While such applications remain in their infancy, the argument remains that blockchain-enabled digital asset infrastructure could do what the internet did before it in reimagining the delivery of solutions across every industry and vertical.
Interestingly, environmentally speaking, the increased transparency, streamlining and supply chain simplification of such a blockchain transition could be viewed as a net climate positive: if the main input of such systems is electricity consumption, and that consumption can over time be transparently and verifiably tracked, then we may focus our attention productively on making that consumption renewable in the knowledge that there is a clearly definable and addressable problem to solve.
It is why we believe the crypto sector is uniquely well placed to deliver decarbonisation at speed, as we have explored with reference to the specific case of Bitcoin in our recent benchmarking and solutions reports.
There are a few nuances to take into account here:
(a) The concept of fair launch – how to publicly distribute the asset in the first place without the PoW mining process? New PoS chains will generally do a pre-mine that allocates tokens upfront to team members/seed investors etc – the distribution mechanism is therefore not the same as scaled public issuance over time through mining rewards.
(b) Concentration of wealth in the network over time. Given that the PoS staking rewards are generally awarded based on % stake in the network, there will be a general concentration of wealth towards larger stake holders over time. The degree to which this poses an issue depends on the initial allocation of the token among the user base – and clearly interacts with the fair launch/distribution mechanic mentioned in (a). While concentration of wealth is undoubtedly a problem area of PoW networks as well, PoS arguably has more capacity, through its directly financial resource-based consensus mechanism, to rapidly compound any inequalities that are written into coin issuance from the outset.
(c) PoS blockchains must find answers to the ‘nothing at stake’ problem. Bear in mind, blockchain consensus must definitively answer not only (a) whether an individual block is valid but also (b) the commonly agreed history of the chain to which that block is added. In PoW, there is an easy answer: the chain with the greatest cumulative proof of work is the universally valid one. Because adding to the chain is a competition among miners requiring both time and real-world-bound energy, there is little incentive to divert that resource to anything other than the chain with the greatest cumulative proof of work. Further, to go back and try to alter that history means rewriting all the blocks from the point of attack (and all the repeat physical energy and time that requires, which rapidly increases the further back in time you go). In theory, the act of validation in PoS is trivial – it relies purely on financial stake, and not, per PoW, on financial resource + physical hardware obtained + energy put in + time taken. What then, presented with two potential forks of a PoS blockchain, stops validators from using their stake to validate on two competing forks of the blockchain at once or indeed to seek to recreate the entire history of the blockchain from scratch? While PoS blockchains have generally addressed these problems through design of stake slashing mechanisms, it nevertheless presents a whole problem area that PoW chains are able to answer in a simple, if deliberately highly inefficient, way.
(d) The current generation of PoS-based blockchains are often observed to be faster with larger block size and more transaction throughput. The trade-off to this is the size of the blockchain created, which, in contrast to Bitcoin PoW, for instance, and its deliberately limited capacity, may prevent the computers connected to the network from practically storing full transaction histories. Depending on design priorities, this might be viewed as a compromise on security and data integrity.
(e) Numbers of total validators in PoS networks may be comparatively low, and therefore the degree of decentralisation achieved compared to the most established PoW protocols correspondingly lessened. The security track record of PoS in practice is also less tested.
This is not to represent one or the other position in the PoW/PoS debate: indeed we think there is a place for both depending on the use case and application-specific needs. However, there is an alarming trend to frame PoS as a direct (only more energy-efficient) substitute of PoW, presenting this to the public as an open-shut case without considering any of the surrounding context or nuance.
It is striking that utility/transaction becomes a much larger part of the proposition in the emerging economy context, and investment more prominent in developed economies – as per this recent Block research. The potential usage of assets subject to crypto-mining is therefore different in the emerging economy context. Generally, there will also be a difference in attitudes to the environmental impact of crypto-mining in the first place, dependent on geography: environmental concerns are unlikely to play a similar role for a miner based in Kazakhstan as they are for a miner based in the US.
We would consider that the visibility is linked to the enormously increased adoption and capitalisation of cryptocurrency networks. As we observed in our most recent report on the topic (p. 42), BTC energy usage in 2009 was estimated at 0.003 TWh; in 2021 it was some 102 TWh. This almost exponential increase in electricity consumption has naturally drawn attention.
This leads into the other point: the environmental impact conversation has been almost entirely a Bitcoin electricity consumption conversation – as we have pointed out, it is crucially important at this time of public debate that any reporting is clear on the diversity in the crypto asset taxonomy, and the fact that ‘crypto’ is more than Bitcoin.
With that said, Bitcoin is where progress needs to be made fast, which is why we have focused our most recent attention on the practical solutions that can be applied to the Bitcoin instance.
Mostly because the electricity consumption statistic is viewed in isolation, with an implicit assumption that there is no utility received in return for that consumption (cf. also answer to Q1).
Policymakers should be aware of:
– the landscape beyond Bitcoin, and the differences in energy consumption outside of Bitcoin
– the action being taken by the industry e.g. Crypto Climate Accord to define common standards and ambitious industry goals (cf. pp. 29-30 of report here for references)
– the interaction of crypto mining and renewable energy & procurement (cf. pp 20-29, 36-38)
– the industry’s potential to achieve rapid decarbonisation.
In our view, there needs to be more emphasis on the practical solutions and steps that can be taken now to address the urgency of the challenge. As a wallet/exchange platform, we are exploring REC-based compensation models as one example, and have summarised the findings of our pilot project here for reference.
We have also compiled relevant applicable methodologies, summarised in our initial survey report here.
We recognise e-waste as an issue that requires further consideration. However, we would make the observation that what is profitable/not-profitable is not an absolute. Because of the dynamic mining difficulty algorithm in Bitcoin, for instance, equipment that was not efficient enough to be profitable at times of peak hash rate may become profitable in periods of decreased competition, when the PoW difficulty algorithm adjusts lower to compensate. Miners should be encouraged to maximise the lifetime of all their hardware (which they surely do anyway for purely economic reasons).
This is an area that needs further research, as currently there is very little publicly available data.
As we have stated in past reviews on this topic, it is a key advantage for the industry to have electricity as its main input, as work can be done to decarbonise the sources that generate that electricity, and the industry is already making encouraging progress in this regard. In this sense, it is not like other industries where the fundamental tech may be lacking to progress with decarbonisation, or where data is opaque and the supply chains complex and tangled.
In our view, it is time to see the opportunities and not just the challenges in what is an innovative and dynamic space.
Fundamentally, this should be viewed as an energy issue and not simply a sector-specific issue. The biggest single action that policy makers could take on every front is to remove subsidies on fossil fuels. In doing so, it would become cheaper to both purchase and produce renewable energy and the energy profile of the crypto sector would change rapidly as a result.
We encourage policy-makers to see crypto as part of the solution, not the problem, and pursue policy that can use the industry as a lever for the use of renewables – miners can set up anywhere globally that there are cheap, abundant renewables – and make use of crypto as a means of balancing the grid.
The worst possible solution environmentally would be to implement bans on crypto mining. Implementing blanket bans will not stop crypto mining, it will send miners elsewhere, and normally to dirtier grids.
Further information on the background context may be found in our most recent report.
We have written in some detail about the Bitcoin mining landscape and the impacts of changing geographies, commercialisation, innovation, regulation and public scrutiny (see our guide here pp. 32-41).
The trend for miners to go public (in the sense of publicly traded company) means they increasingly need to comply with certain ESG standards as an issue of business necessity.
And industry organisations (e.g. CCA, RMI) are actively working on verified ‘proof of green’/individual certification for renewable operations.
Over time, the nuance of this probably shifts from competitive advantage to existential business risk if neglected – and this dial is shifting fast.
None of the issues seems insurmountable – indeed, an aggregate analysis would suggest more opportunity than threat.
While we think there is very little chance PoW will cease to exist in its current format, the chances of decarbonisation for the sector are extremely good, and the Crypto Climate Accord is committed to net zero for the entire industry by 2030.
Evidently, the more data we can obtain directly from miners the better from a quantification standpoint. However, unless a crypto exchange integrates a mining operation of its own, it is removed from the mining data as much as anyone else.
Voluntary disclosure through industry bodies such as the BMC seems to be making headway, with the most recent BMC report claiming to report on 50%+ of network hash. Given the current miner commercialisation trends, we could expect this to continue to increase in transparency.
In the meantime, standards are important – that the industry agrees on the interpretations applied and the assumptions made in quantification. We and industry organisations are working hard in this area.
Yes, this is an interesting area of application – as one example, one blockchain in the space is experimenting with a smart contract implementation that can automatically divert a portion of each transaction fee to carbon credit purchases to offset the blockchain’s footprint as it goes.
The overall transparency and verifiability of blockchain systems is also attracting interest in terms of developing more efficient and more liquid marketplaces for renewable energy certificates or carbon credits.
Perhaps naturally, industry reports have a tendency to cherry pick and make the most of what could be rather than things as they are. On the other hand, academic reports are lacking in practical industry focus, and dwell too much on problems over opportunities. Both are needed to find a balanced way forwards.
We have mainly focused our attention on renewable energy certificates RECs (please see our report for our overall views on crypto & renewables).
The point to make here is that the size of the electricity consumption of the crypto sector as opposed to the size of REC marketplaces (or any marketplace for renewable power) is significant.
If crypto goes 100% renewable, it instantly adds a significant factor to the size of renewable energy markets worldwide. In aggregate, it acts as a structurally important demand signal for renewable energy and, if it can collaborate effectively on this, will drive demand for renewables and stimulate renewable capacity and cost effectiveness.
This argument would hold best if we were currently maximising renewables usage effectively, however there is evidence that a significant amount of renewably generated energy is curtailed (i.e. goes to waste) in the absence of grid demand or effective storage solutions (see, for instance IEA stats on this topic). In a grid environment in which we will only rely more on fluctuating renewable sources, we need the load balancing, demand flexibility and – in the absence of any other demand – productive monetisation of excess energy that mining can bring.
It also assumes (in our view, misguidedly) that there is some sort of hard cap on renewable power – that crypto will ‘take away’ from something else. In our view, the big advantage of crypto is that, as a huge, electricity-based industry with the power to make environmentally conscious decisions, it incentivises new supply and in fact grows the renewable market as a whole rather than eat into any existing resource.
We would agree that miners are entirely profit-driven – competition is high and they will not choose more expensive renewable energy over fossil fuel energy unless the commercial implications within an ESG context prevent them from doing so. That is why we need to remove fossil fuel subsidies and focus on incentivising demand for renewable energy – so that fossil fuel is not the cheaper option.
In our view, it ties into geographical miner reshuffles towards North America and a transition to a more commercialised, corporate public entity miner model (see our Bitcoin mining guide for further context).
Realistically, it seems likely that this progress will hit an upper bound in the near term: mining is still distributed worldwide, including in jurisdictions where there is cheap fossil fuel power and little interest in the environmental impact of energy sources. These ‘hard cases’ will only shift at the point at which renewable energy in the region becomes more economically viable than fossil fuels, and it is hard to determine (or influence) when that will be.
Most practically, renewable energy needs to become cheaper, which means signalling demand for new renewable capacity.
There also needs to be a shift in education and mindset, at policy and society level, that is more supportive of miners and the innovative ways in which they could contribute to climate solutions rather than dismissing them out of hand as an unwanted waste of energy.
The handy thing about crypto mining is that it is a 24/7/365 operation – but can power up or down at any time. There is no operational need to be online at a particular time and miners can perfectly well deal with a highly fluctuating supply, run only at certain times of day and still run profitably.
We consider it unlikely that PoW would be the preferred mechanism for a CBDC implementation. For example, it is more likely that a CBDC would be account-based rather than token-based – a point against the standard PoW implementation. Other considerations such as programmability (smart contract functionality) also argue for an implementation with wider flexibility than PoW. Overall, the decentralised security model of PoW with its public, permissionless focus does not fit most obviously with the needs of a CBDC, which may consider other forms of DLT that provide greater degrees of control and oversight.
Most likely, asset tokenisation projects will use one of the emerging L1 smart contract blockchains (generally PoS in consensus), if not a specially designed enterprise DLT system. Providing some overlap with the CBDC instance, the priorities are likely to be interoperability; programmability and security – again with little suggestion that PoW would be considered as the natural choice for such applications.
As elsewhere, the opportunity is to provide clarity in the framework being applied and how it is to be interpreted – and demonstrating through that framework that policy makers have consulted with the industry and understand the dynamics at play.
The China example is a timely reminder that bans don’t work but rather displace or conceal. Future bans in individual jurisdictions will cause further relocation – and opportunities for those who think positively rather than negatively about this issue.
Some select points:
– Policymakers don’t have to start from scratch. Policy should aim to join up with, consult opinion of, and collaborate with existing projects and initiatives.
– Policy should be sensitive to nuance i.e. differentiate by use case and avoid looking at energy consumption in isolation. There is a need to pair the technology employed and its use case.
– Policy doesn’t need to implement sweeping regulation at industry-level. Rather, focus efforts according to where the impact is largest. Policymakers should focus on the comparatively well-delineated pockets of the ecosystem where they can make the biggest difference, as opposed to going after crypto as a homogenous mass.
– Operating by ban or regulatory burden should be avoided, as it merely displaces operations elsewhere.
We are pleased to input into this consultation. For any follow-up, please reach out to:
Kirsteen Harrison, Sustainability Adviser to the Zumo Board, email@example.com
Daniel Taylor, Research and Policy Lead, firstname.lastname@example.org