Central banks around the world are reportedly exploring the idea of issuing a variant of fiat currency that bridges the functional divide between physical cash and account-based systems. While the timing of the efforts can largely be attributed to the flourishing space of private and open-source experiments in the cryptocurrency space, the motivations cited by discussion and design papers of these efforts most often center around common goods arguments, such as the inclusion of unbanked citizens. Economic considerations are largely focused on the technology’s effect on the relationship between commercial and central banks and general statements of global competitiveness, as well as cross-border payments. With economic powerhouses such as China officially piloting a central bank digital currency, US officials are mainly paying lip service to the support of a digital version of the US Dollar.
According to Bloomberg, Treasury Secretary Janet Yellen pronounced the administration's endorsement of a U.S.-backed digital currency, saying it could result in "faster, safer and cheaper payments" for U.S. residents who lack bank accounts.
The statement echoes many of the earlier more timid voices and sentiments around the world’s reserve currency. The following analysis offers a thesis-driven financial technology investor’s perspective on digital currencies, with particular considerations of variants discussed by central banks and their shared focal point of use as payment instruments by consumers.
Investors quite often chose the term thesis in an effort to lend credence to their investment philosophy, without indeed engaging in the rigorous scientific method the term implies. The following deviates from this practice in several important ways, in that the discussed thesis is indeed the result of a refined and falsifiable problem statement, built on a multi-year effort of qualified financial engineers, data scientists, and peer reviews from hundreds of development teams conducting experiments around the discussed subject matter. The latter has led to a broader comprehension of the position legacy money systems occupy within a global financial topology, while observing the inevitable progress of technology.
Technology cannot be un-invented.
As thesis-driven investors with a focus on buying equity in technology companies addressing value transfer systems, we are using the total addressable market (TAM) size as a starting point to evaluate the investability of specific technologies. We further segment these markets based on inefficiencies that can be addressed via innovative (technology) solutions. Introduction of the latter should mathematically result in the reduction of - often manual - processing times, while improving transparency, leading to an overall reduction of costs to provide a product or service. The combined total of these savings can be viewed as the basis for a TAM in a given segment. This is not the entire equation, but this summary provides a framework and backdrop for how we approach the viability of a solution within our area of expertise and focus.
Approaching technologies from first principles has led us to the thesis that money can be best understood as a contract between two or more parties, covering the two main use cases of lending and spending (payment).
Most of the aforementioned agreements are currently recorded in databases, using a customary unit of account. Setting physical cash aside, all currencies - including fiat and cryptocurrencies - share two of the three historic definitions of money: the medium of exchange (bytes), and the unit of account (fiat). The unit of account for bitcoin is $ (more here).
As for the respective store of value functions within the various categories, these are in essence expressions of the lenders' or borrowers' portfolio distribution and time preferences. While the aforementioned portfolio construction may include new and/or evolving asset classes such as digitally native assets, the functional overlap of these instruments (use as payment) can currently only be observed where government-issued currencies fail to maintain their pricing function.
As such, this analysis excludes anecdotal references, such as the use of bitcoin in Venezuela, and El Salvador to pay for goods and services (the discussion of bitcoin as 'money' is summarized here).
The difference between the various categories of money is mostly rooted in the complexity of the agreement, or what might be referred to as meta-data associated with the value shown as account balance. The higher complexity paired with often unstructured meta-data frequently requires manual intervention, batching, and other procedures performed by the financial service providers maintaining these contracts via databases. The costs of these inefficiencies are currently borne by the account holder in form of fees that include margins for the financial service providers which control these legacy systems.
Decentralized software systems - such as blockchains and graphs - can address these frictions via standardization of metadata into digital bearer instruments (“tokens”) and automated transfer mechanisms via smart contracts.
Our evaluation framework further consists of horizontal, and vertical metrics. The former references the different contract categories (money), while the latter generally refers to the velocity within the category.
Horizontally these functions are being implemented in digital wallets with more than one hundred projects either live or in development. Vertically it can be observed that several projects are taking market share within categories such as mortgages, municipal bonds, and payments.
A crucial question investor seeks to answer is that of product-market fit. Put simply: Does the offer fulfill a need or solve a problem for a proposed customer segment willing to adopt the product?
Ostensibly, fiat currencies address challenges arising from the desired value exchange between two or more parties through funds that are readily accessible for spending.
The latter are grouped under the label M1 by the Federal Reserve and, aside from physical cash, include funds recorded in demand deposit accounts, commonly referred to as "checking accounts".
While many merchants prefer cash over digital payments, this is largely a result of costs and time delays inherent to current electronic payment systems and their purveyors, such as debit and credit card companies. Ultimately, these payment options overwhelmingly make use of the aforementioned money classified by central banks as M1.
Assuming a digital currency addresses these problems for sellers, it is a reasonable thesis that merchants would be motivated to adopt the new solution while encouraging counterparties to use it. Thus, the product-market fit could generally be achieved. However, because consumers frequently have myriad other digital payment solutions at their disposal, we will next examine these options.
Investors carefully review market incumbents impacted by new solutions. According to a 2020 McKinsey report, physical cash accounted for 28% of all transactions in the U.S., representing a significant share of payment options.
The significant curtailing of in-person transactions due to the COVID-19 crisis might be seen as an opportunity for digital currencies to gain market share from cash users.
However, the 2019 payments study conducted by the Federal Reserve shows that debit cards and credit cards top buyers’ noncash payment methods. Aside from the inconvenience of carrying cash, many debit and credit cards offer additional features, such as loyalty points and lines of credit. Aside from one IMF Working Paper, there is little indication in current design papers that digital currencies (including central bank digital currencies, or CBDCs) will offer these or other attributes, so investors will heavily discount the segment of debit and credit card users from the total addressable market.
Competition may also arise from digital currencies issued by other nation-states. While foreign exchange markets are generally not designed to cater to the payment needs of consumers, digital currencies issued by central banks may remove friction introduced by intermediaries in the form of time delays and fees.
Depending on interoperability and design choices, citizens exposed to currencies with rapidly decreasing purchasing power could opt to store value in a currency of their choosing and only convert to the local fiat when necessary for payment functions.
As outlined in a previous article, a significant metric for the investor is the total addressable market. A report published by McKinsey in October projects that 2020 global revenues from payments will fall below their 2019 all-time high of $2 trillion.
Setting further classifications into consumer, commercial, and more granular segmentations aside, investors may be tempted to consider the final cumulative value for 2020 as the total addressable market. However, the serviceable market for digital currencies, including CBDCs, might be significantly smaller, depending on the adoption of digital currency-like features by incumbents, such as the seemingly instant settlement function offered by payment solutions such as Zelle. Financial service providers participating in the Zelle network agree to take on the settlement risk, so while it does not technically fulfill the instant settlement component, users generally will not perceive a difference to digital cash.
Digital currencies are frequently described as possessing the same positive attributes as cash. As such, fee extraction by third parties for payment functions must be considered a quickly diminishing opportunity, trending toward that of physical cash, the usage of which is generally feeless.
Thoughtful investors will seek digital currency projects with business models addressing new revenue opportunities rather than those that rely on extracting value from a payment transaction.
Practical design considerations are largely absent from the current discussion. As such the following design considerations are based on investors’ desires to measure the impact of the new technology.
CBDCs across the world should follow protocols, and consequently, form new standards across these categories - i.e., each new unit could carry a prefix such as C-BOFA-202003206-L-M-I-5000000.
This particular example qualifies each unit with the prefix range as having been created by a commercial bank (B), in this case Bank of America (BOFA), on March 6th, 2021 (20200723) representing a Loan (L), specifically a mortgage (M) for and Individual (I).
Aside from providing transparency in this way, will provide future utility as the desired “programmable money”, as divided into three main sections: creation, moving, storing – each section with their own subsections of who/why/how.
As payments instruments for consumers, digital currencies in general and variations issued by central banks in particular will continue to face a number of hurdles that must be factored into a go-to-market strategy. Aside from technophiles, early adopters can potentially be found within the segment of the unbanked. However, merchants with existing payment provider agreements might not readily be willing or able to accept digital currency issued by central banks. Because of this, additional incentive models may be required to ensure that digital currencies — including those issued by central banks — are adopted by a wide selection of merchants.
The current payment focus of CBDC is somewhat surprising given that payments are a secondary use case of fiat currencies today, and that the main concern of fiat currency users is its continued loss of purchasing power.
CBDC design must consider the creation process of all money across a number of categories which in step one could greatly improve the overall transparency, and manageability of the movement of money, and related values – i.e., seigniorage and demurrage. Our current thoughts on these categories include mortgages, certificates of deposits, corporate debt, student loans, auto loans, credit card debt, checking accounts, and HELOC.
CBDC design papers largely omit discussions around the instantiation of fiat currency. It is necessary to examine the current birth process of fiat currencies which are largely brought into existence by the process of the collateralized lending activities by commercial banks (e.g., mortgages), a business model not readily available to competing private currencies.
It can readily be observed that a small but flourishing lending market is evolving in the decentralized finance space. Although these solutions are today largely limited to digitally native assets.
Central bank digital currencies might open access to these solutions to a broader audience, should they be interoperable with these innovative solutions. However, residual frictions, and - most importantly: continued debasement of fiat currency - will mean that fiat currencies may be relegated to their unit of account function, while interest-bearing digital assets will be the new status quo.
Listen to an audio interview on this topic here.
*I published a shorter version of these thoughts in March 2021 in Forbes.