Feb 23
Central Bank Digital Currencies and CO2: Digital currency as a tool for reducing greenhouse gas emissions
By Timothy Higgins (Winner of the Monetary Economic Competition)
This paper addresses the question of whether central bank digital currencies (CBDCs) could be a tool for combating climate change by reducing carbon emissions from producing and managing physical cash. Several studies have shown that the manufacture, circulation, and destruction of currency does contribute materially to countries’ carbon footprints. We assess the extent to which banknotes specifically contribute to greenhouse gas emissions, and how those emissions may vary across two similar economies with very different geographic features: the UK and Canada. We then discuss how the design of a CBDC could be tailored to a country’s unique features to most effectively reduce emissions related to banknotes.
Carbon emissions and the cash lifecycle in the UK and Canada
In the past ten years, both the Bank of England (BoE) and the Bank of Canada (BoC) have conducted banknote life cycle assessments (LCAs) to evaluate the environmental impact of cash in their respective economies. We focus on these two studies to develop an understanding of what elements of the cash lifecycle contribute most to carbon dioxide equivalent (CO2e) emissions. These reports also assess two very physically different countries: Canada being much larger and less dense than the United Kingdom (4.3 people/sq. km vs. 278/sq. km in the UK), with greater travel distances between communities. This allows us to evaluate the energy consumed in distributing cash, compared to the production and use stages.
Energy use and carbon emissions from banknotes in England
In 2013 the BoE conducted a cradle-to-grave evaluation of banknotes’ environmental impact. It includes a breakdown, by denomination, of the percent of total energy consumed at each stage in a banknote’s life cycle. I have combined these data with BoE statistics on notes currently in circulation to construct a weighted-average composite showing the proportion of energy consumed by each lifecycle stage across all GBP notes. Figure 1 below presents those calculations:
We can see that ATM use is the key source of energy demand from banknotes in the UK, responsible for almost 65% of total consumption. Because £10 and £20 notes account for a large majority of notes in circulation (34%, 49%, respectively), the aggregate figure reflects the heavy weighting toward ATM use by these denominations. By contrast, the £50 note is rarely stocked in ATMs, primarily sitting in cash centres and banks. Interestingly, the “transport” categories represent a comparatively small amount of the total energy demand — less than 5%. This could be due to the UK’s high population density (82nd percentile, globally), requiring shorter travel distances between cash centres, ATMs, banks, and businesses.
To understand the significance of these findings, we next examine aggregate CO2e emissions from banknotes beginning with a closer study of the top energy demand category: ATMs. The BoE LCA provides an overview of kilowatt-hours (kWh) of energy used by ATMs per banknote in both vending and stand-by mode, and for machines operating indoors, outdoors, and outdoors with a heater (for when the temperature is <0oC). Those figures are in Table 1 below.
While these data are relatively uninformative on their own, when combined with the UK Government’s 2021 guidance on emissions per kWh of energy consumed, we can extrapolate the CO2e impact of these ATMs on a per-note basis. We can then use BoE data on the number of notes in circulation to estimate total annual emissions from the UK ATM network:
These calculations show that, using a weighted average of energy demand across ATM features and functionality, ATM use for the 4.541 billion banknotes currently circulating in the UK results in approximately 350 million kilograms of CO2e emissions per year, or 350,000 tons. As we have seen, these ATMs account for approximately 65% of energy demand in the banknote lifecycle, meaning that (assuming the same CO2e emissions ratio) annual CO2e emissions directly attributable to the energy consumed by banknotes is approximately 550,000 tons. Figure 2 provides a breakdown of these emissions mapped to lifecycle stage:
Of the total 550,000 tons of CO2e emissions from the UK banknote system, nearly 350,000 are attributable to ATM use, with the next largest category, substrate production, contributing over 100,000 tons. The use stage of a banknote’s lifecycle in the UK is by far the most impactful to global warming, with the manufacturing stage (135,000 tons from substrate production, raw material production, and printing) a very distant second. Emissions from distribution (cash centres and transport) and end of life (EoL) are significantly lower, at 62,000 tons and 5,000 tons respectively.
Energy use and carbon emissions from banknotes in Canada
Next, we review banknote emissions in Canada. The BoC LCA provides a similar breakdown of energy demand by lifecycle stage. To allow for comparison, I have reclassified the original UK lifecycle stages with the BoC categories and presented them along with the Canadian data in Figure 3 below:
The striking difference in these figures is the relatively outsized energy requirements of distributing banknotes in Canada compared to the UK — 52% of total energy demand to just 11%. This results in 20% proportional decreases in energy demand for both the manufacturing and use lifecycle stages (5% vs. 25%, and 43% vs. 63%, respectively). End-of-life accounts for a de minimis proportion of 1% or less in both cases.
The relatively higher distribution requirements could be due to the greater geographic area and lower population density of Canada, necessitating longer trips for the armoured cars and aeroplanes carrying notes from BoC printing centres. The BoC and BoE LCAs provide travel distances within the distribution stage. In figure 3 below I have combined those figures across sub-stage (e.g., transport from printing to cash centre, cash centre to bank branch) and presented the average distance travelled by a banknote in each country:
A significant difference is clear: during distribution, the average banknote in Canada travels almost 1,120 km (about the distance from Vancouver to Calgary) to just 480 km in the UK (about the distance from London to Newcastle) — or 133% farther. This reflects the much greater distance notes travel in the initial stages after printing to regional distribution centres: approximately 380 km in the UK but nearly 800 km in Canada. These travel differences account for the greater share of energy demand attributable to the distribution phase of the banknote lifecycle in Canada compared to the UK.
Before compiling aggregate emissions statistics for the Canadian banknote system and making comparisons to the UK, we must make several caveats. First, the size of the UK banknote system is 1.62x that of Canada: there are 4.5 billion GBP banknotes in circulation compared to 2.8 billion CAD notes. This partly reflects the larger UK economy, which has a GDP of roughly 1.75x Canada’s ($2.8T to $1.6T). Additionally, the BoC LCA admits that its ATM energy consumption model is oversimplified and, unlike the BoE analysis, does not weigh energy use by ATM functionality and features — likely resulting in underestimation. When the more detailed BoE statistics on ATM energy use are applied to Canadian banknote figures, along with a Canadian emissions coefficient, annual ATM CO2e emissions for Canada come to 141,500 tons.
With these caveats in mind, Figure 5 below presents a comparison of UK and Canadian CO2e emissions. It includes an aggregate for Canadian banknote emissions using BoE ATM data but original BoC manufacturing and distribution figures; an increase of these figures by 70% to reflect the larger UK economy and currency pool and allow a more like-for-like comparison, and; the UK emission statistics originally presented in Figure 2:
Even before adjusting for differences in economy size, distribution of banknotes in Canada accounted for nearly 10,000 more tons of CO2e emissions than in the UK. After adjustment Canadian ATM emissions are still significantly lower than those of the UK., reflecting less carbon-intensive power generation in Canada. Given that the original BoC figures on ATM emissions (60,500 tons) likely underestimated the actual amount significantly, it seems possible that its reported figures for manufacturing emissions — not even 10% of those in the UK — could be low as well.
Conclusions: comparing Canadian and UK banknote greenhouse gas emissions
These data imply that both geographic and power-generation features of an economy make a significant difference to greenhouse gas emissions from currency management. In the larger country where the average banknote travelled roughly 133% farther, CO2e emissions from distribution were nearly double (when adjusted for economic size). Emissions from ATM use were also much higher in the country with a greater ratio of greenhouse gas production per kWh of energy consumed (.212 kg CO2e in the UK to .14 kg CO2e in Canada). While not revelatory, this confirms the intuition that the use stage of banknotes will produce fewer greenhouse gases where the power grid upon which ATMs draw is more CO2e efficient. Unfortunately, the manufacturing data seems too unreliable for comparative analysis, and indeed, quantifying its environmental impact is a much more complicated process, requiring difficult judgements on which upstream processes should and should not be considered. For this reason, we will contain our assessment in the following section — on how to apply these lessons to the design of a CBDC — to the distribution and use stages of banknotes.
CBDC design considerations
Minimizing distribution emissions: essential CBDC design features
As we’ve seen, greenhouse gas emissions from distributing hard currency can be a meaningful component of a country’s carbon footprint. The greater distances banknotes travel in Canada lead directly to higher relative CO2e emissions compared with the UK, highlighting the importance of a country’s geographic features in its management of currency. By extension, countries with the following traits could reduce cash-distribution-related emissions by the adoption of a CBDC:
a) Low population density with high smartphone penetration (Russia, Australia).
b) Significant transport challenges, e.g.: archipelagos (Indonesia), mountainous regions (Nepal), countries with many remote communities and poor transport infrastructure (Brazil).
c) Outdated vehicles with poor mileage and high emissions (Iran).
For these cases, a CBDC could be effective if designed with several key features. First, it would need to be a retail CBDC, with digital accounts or wallets for individuals at the central bank, accessed via smartphone at the point-of-sale — in practice, effectively the same as current contactless payment through debit or credit card. One proposed alternative for countries with low smartphone use or poor network coverage is pre-loaded central bank “debit cards”. However, these would still require distribution to end-users, and plastic cards would be heavier than paper banknotes, requiring more energy to transport. Transactions via smartphone would remove the need for physical distribution of CBDC cards, and individuals could have balances “saved” onto devices that could be used to transact in areas with poor service or during network outages.
A second necessary CBDC feature would be seamless and free peer-to-peer payments. A core function of cash in many countries is the ability to facilitate transactions in the informal economy. While user-to-business smartphone transactions are well established with solid technological foundations, any successful CBDC designed to replace cash and reduce distribution emissions must be capable of instantaneous, free, peer-to-peer payment. There are examples of such services currently operating on existing digital payment infrastructure (e.g. Venmo in the U.S., M-Pesa in Kenya), however, they rely on network access to verify balances and execute transfers. A key technological challenge will be maintaining functionality while users are offline.
Minimizing use emissions: essential CBDC design features
We have seen that ATM energy consumption is the core source of CO2e emissions during the cash use lifecycle stage. While a more direct method for making these machines less environmentally damaging would be to modify how they consume electricity (at both the grid level and the unit level), minimizing emissions through a CBDC would require decreasing the number of ATMs in use by lowering the demand for hard currency in favour of a digital alternative. In a 2020 report by the Bank for International Settlements on design features of a successful CBDC, the bank lists prerequisites for wide CBDC adoption:
“[A] CBDC must be convertible, convenient, accessible and low cost.
The underlying system should be resilient, available 24/7, flexible,
interoperable, private and secure for the general public.”
Convertibility here means that any CBDC could be converted into cash at parity. A convenient, accessible and low-cost CBDC would be available to all citizens through a commonly used medium (i.e. smartphone) at no or negligible expense. And the technology or “underlying system” must be infrastructure as reliable as physical cash — no outages, usable anytime for any type of transaction, and interoperable with practically all systems within the economic and financial environment. To the degree that a CBDC can achieve these goals, it can function effectively as a cash substitute, reducing the demand for hard currency and the need for a CO2e-intensive ATM network.
Conclusion
Greenhouse gas emissions related to physical currency vary by the features of the country within which that currency circulates. Geographically large currency systems expend more carbon-based energy transporting cash than compact ones. Economies which can generate more of their energy from cleaner sources will see fewer emissions from the use of cash — specifically ATMs.
Designing a CBDC to accommodate these features can lower aggregate CO2e emissions, but will such a reduction be a powerful tool for combating climate change? The most recent BoE figures on greenhouse gas emissions show that the UK produced 326 million tons of CO2e in 2020. Our calculations estimate that the cash management system accounted for only 550,000 tons. Likewise, Canadian emissions were 730 million tons for 2019,17 with cash management emissions estimated at only 222,285 tons. Emissions reduction should be pursued in all areas of the economy and shrinking cash-related emission figures would mean material decreases in CO2e (potentially by hundreds of thousands of tons), but CBDCs will not be a shortcut to solving the urgent problem of global warming.
This article was written by Timothy Higgins as his submission in the Monetary Economic Competition 2021/22.
This article was judged by Sir Charles Bean, the Former Chief Economist and Former Deputy Governor for Monetary Policy at the Bank of England.
“I thought this was a really enterprising attempt to look at an underexplored dimension of the CBDC debate and learnt a lot from reading it. The essay is well constructed, with clear narrative and well written. One issue that is not really addressed, is whether the introduction of a retail CBDC is really necessary to get the emissions gains from reducing the use of banknotes. The Covid pandemic has already led to a massive expansion in the use of contactless payments and this seems unlikely to unwind as the pandemic recedes.”
A short excerpt of comments by Professor Charles Bean
This article was reviewed by Anna Clarey (President of the Central Banking Society).
References
Bank for International Settlements. “Central bank digital currencies: foundational principles and core features”. BIS website. 9 October 2020. https://www.bis.org/publ/othp33.htm
Da Silva, Nuno; Binder, Marc; Marincovic, Coppelia; Pritchard, Kathy. “Life Cycle Assessment of Polymer and Cotton Paper Bank Notes”. Ottawa: Bank of Canada, 2011.
Smalldridge, Georgina. “2020 UK greenhouse gas emissions, provisional figures”. London: UK Government, Department for Business, Energy & Industrial Strategy, 25 March 2021.
Shonfield, Paul. “LCA of Paper and Polymer Bank Notes”. London: Bank of England, 2013.
Appendix A: Core CBDC Features
Source: Bank for International Settlements, “Central Bank Digital Currencies Foundational Principles”
