What We Know and Do Not Know About CBDCs (So Far)
- Future of Finance
- Dec 18, 2023
- 19 min read
Updated: Jul 10
A Future of Finance interview with Alisa DiCaprio, Chief Economist at R3.
A year and a half may have elapsed since the last central bank digital currency (CBDC) was issued. But the work at and by central banks, banks, supranational organisations and technology vendors has continued. The accompanying output of policy statements, academic papers, discussion documents and accounts of experiments is a vast but rich source of experience and information. Which is why a team at R3, a leading force in the digitisation of financial markets and a participant in multiple CBDC projects, embarked on a review of the literature that has accumulated about CBDCs. Dominic Hobson, co-founder of Future of Finance, spoke to Alisa DiCaprio, the Chief Economist at R3 who lead the investigation, about what the review unveiled about how CBDCs are being designed, issued and operated, and learned what CBDC designers everywhere still lack: a trove of empirical data about the impact of CBDCs on the behaviour of money, markets and especially consumers.
Key Insights
A review of the literature about CBDCs finds a wealth of interesting facts, figures and analyses, but the preponderance of policy notes over academic papers and the lack of data (thanks to lack of issuances) imparts a bias to abstract design issues over practical experience.
Although every CBDC has some issues in common with every other CBDC – such as the impact on bank funding and the conduct of monetary policy and the possibility of disintermediation – every CBDC is also sui generis, in that every economy and financial economy is different.
All CBDC designs assume commercial banks will persist but what form commercial bank money will take is driven less by current forms (Stablecoins and tokenised deposits) than the settlement needs of tokenised markets and the payments needs of consumers.
The universal validity of current CBDC designs is inhibited by the limited applicability of general equilibrium modelling, the lack of empirical data from CBDCs in issue and a reliance on data from a small group of developed but atypical economies in North America and Europe.
Although much of the work that has been done on CBDCs is too academic to be immediately useful to central banks pondering, say, how to encourage take-up by merchants and consumers, it anticipates issues which may arise later, such as illiquidity and cyber-attacks.
It would be helpful to CBDC design if neutral third parties such as the BIS and the IMF intermediated data collection from countries that have issued or are planning to issue a CBDC because it could help solve questions such as the impact of CBDCs on the velocity of money.
The most solid and practically useful of the theoretical work undertaken on CBDCs lies in the area of financial stability, where the differences between economies are not as large, and there is a consensus on ways in which sources of instability can be managed.
The principal gap in knowledge about the impact of CBDCs on economies lies in the area of consumer behaviour, rather than banks or merchants, because take-up of existing CBDCs is low and studies of how consumers would react to and use a CBDC are not extant.
The pioneers in CBDC issuance are already having a profound influence over the design of future CBDCs but no central bank can assume that issuing a CBDC is a one-off exercise: every CBDC must be revisited constantly and adapted to new information continuously.
Transcript
00:15 Dominic Hobson: Hello, I’m Dominic Hobson, Co-founder of Future of Finance. My guest today is Alisa DiCaprio, chief economist at R3, where she recently completed an extensive review of the vast and growing literature on central bank digital currencies, or CBDCs, which threw a rather fascinating light on not only what central banks and banks are doing right now, but also what they might do in the future. Alisa, thanks very much for joining us.
00:45 Alisa DiCaprio: Thank you. I’m excited to be here.
00:48 Dominic Hobson: Having conducted this literature review, do you feel able now to characterise what the current state of play in the CBDC universe might be? By which I mean things like how many live CBDCs we have versus how many proofs of concepts and pilot schemes. We have the whole retail versus wholesale debate. And of course, what’s happening in developed markets as opposed to what’s happening in developing markets. How clear a picture do you have of what’s happening?
01:18 Alisa DiCaprio: Well, I think those are actually two different questions because in terms of what’s happening in the state of play, that is absolutely … You cannot learn that from the literature. The literature has a lot of very interesting facts and figures and really does good analysis on the data. But it takes six months to a year to complete a paper and to publish it. Maybe longer if it’s an academic paper, not just a policy note. So I feel like the things that we learn from the literature are more about the potential ways you could design a central bank digital currency rather than who’s issuing it, who’s been successful. And I think a second part of this answer is also that there’s so few issuances that there’s very little data. So a lot of the literature tends to be theoretical rather than actually based on what’s been issued.
02:10 Dominic Hobson: Your point about the learning how to design a CBDC is well made. On that very point, there seem to be a number of things we’ve been talking about ever since CBDC’s first emerged on the scene, like how it would affect the funding of the private commercial banks, how it would either enable or disable the way central banks conduct monetary policy and, of course, the entire possibility of banks being disintermediated in some way. Do we now, or do you now have, do you think a better understanding of what the impact of a CBDC is likely to be on the current structures, both of central banking and of commercial banking?
02:53 Alisa DiCaprio: I would say after the experience that I’ve had over the past maybe year and a half, actually working on deployments, that I do. But it’s also what I’ve realised is this is not something you can read about. This is something that you actually have to go and do because every central bank works a little bit differently. Every country is sovereign and has slightly different ways of doing monetary policy. They have different capital regimes. They have different monetary regimes. So I think very much it’s bespoke, at least at this stage and probably forever, in terms of kind of how this is being issued. But that given, I think that there are certain things that are common across countries. So you mentioned bank disintermediation, commercial bank disintermediation. It’s in the interest of every single country that’s doing an issuance to not disintermediate commercial banks, because they’re really the backbone of financial stability. And we’ve seen that in every case, in every single live issuance, distribution is done through banks. It’s not done directly by the central bank. And that’s to avoid this disintermediation potential.
04:10 Dominic Hobson: As you point out, a CBDC which disintermediated commercial banks is not going to be very successful. Commercial banks have got to survive. If we look at the way economies and financial economies work today, most of the money is created by commercial banks. So do you, as a result of the literature review, have a better idea now of exactly what forms commercial bank money is likely to take and how CBDCs will interact with that commercial bank money? The sort of things I’m thinking of here are Stable coins and, more and more, tokenised deposits as forms of commercial bank money.
04:51 Alisa DiCaprio: I think right now – what is happening today – is they’re all coexisting. Although if you actually look at it carefully, there are very few countries which have issued CBDCs which also have sort of active Stablecoins. It tends to be one or the other. And the second point about that is increasingly what we’re seeing, particularly around digital assets, is that CBDC is being sort of included in the vision to underpin these different financial networks. Because right now you can certainly settle without a central bank digital currency. That’s fine. You can do that with private money. But if you could do it with CBDC, then you do the transaction, the execution and the settlement altogether. So that’s a net new way of doing things. And it’s pretty interesting. So I think what we’re seeing is in the future, this vision where CBDC would underpin a lot of these digital assets and kind of financial networks. Today we see them acting very separately. And the third point is about consumers. Because one thing that you see in a lot of the literature is this discussion of sovereign money. So there’s this assumption that sovereign money, which right now is money that’s issued by the central bank, so cash, is fundamentally different from private sector money, so credit what banks are issuing. But it’s unclear that consumers actually make a distinction between those. The literature certainly does, and certainly we’ve done it in reports as well, but there’s very little evidence about how consumers think about money. Is it important for them to use sovereign currency? It’s unclear. And the existing issuances that we have, which are live, have very low adoption, below 0.2%. And so we can’t really draw any data from that because the usage is so low. So that’s, I think, where we are right now in terms of what we know about how these different types of money will interact.
07:10 Dominic Hobson: You’re making it pretty clear that there are lots of things we don’t know, despite the fluctuations … The large volume, if you like, of academic papers and commercial papers being published in this area. But are there specific gaps in the literature that you reviewed where you feel that there’s an opportunity for more work to be done or you would like to know more?
07:38 Alisa DiCaprio: Well, there’s a couple of different things that we found in the literature review, which people tend to pass over, because you look at it and you say, oh, there’s hundreds of papers, but those hundreds of papers actually are not that different. The first thing is, of course, what we talked about before, it’s not based on empirical data. The papers are theoretical papers because there is no good empirical data yet. The second is that a lot of these papers are partial equilibrium models. And that matters a lot, because kind of the best way to do this would be a general equilibrium model. But general equilibrium models are also very difficult to put together. We do see a movement in that direction, but the large majority is partial equilibrium. And then I think the final thing that we see in the literature is, even though it often isn’t based on a specific country, if you actually look at the kind of way the papers are framed, they’re based on an economy very similar to the US or the EU or the UK. And when there is data, it’s based on the UK and the US. And those economies are very unique, right? So if you’re thinking about how is a CBDC going to perform in a country with a fixed exchange rate or an open capital account, things that are not exactly the same, the literature becomes a lot less informative for those types of countries. And I think this gets back to the data issue. Where you see literature, it’s going to be based in countries where there is a great deal of data, and the largest volumes of data that we have are US, EU and UK. But that’s certainly not typical of every country. And in fact, it’s very atypical of countries that have already issued CBDCs.
09:44 Dominic Hobson: You said a few minutes ago that CBDCs are something which you have to do rather than read about. And however disappointing that is for those of us who like to read about things. You’ve also pointed out that a lot of the papers are, of course, academic, somewhat detached, somewhat abstract. So in this area, as in so many other areas of life, there is theory and then there’s practice. Are you able to say how far you think the practice – and you have pointed out there’s not actually been many real issues out there – are you able to say how far practice has strayed away from theory that you read about in the papers you have read?
10:23 Alisa DiCaprio: I don’t think that it’s strayed so much that it’s separate. In terms of the papers, there’s actually two types. So when we did this literature review, there’s academic papers. And so those are the ones that I was talking about that take a year or two or more to publish. So they’re never at the cutting edge of what is happening. But on the other hand, there’s policy papers. And so the policy papers are those that you see coming out of the IMF [International Monetary Fund] or the BIS [Bank for International Settlements], which have a much faster turnaround and are, to the extent possible, based on their interactions with countries that are thinking about issuing or that are participating in proofs of concept or beginning the process of producing a central bank digital currency. So I think the policy papers are closer to what’s happening. But also the whole point in having this literature is to think about things that we haven’t seen yet. So what happens if there’s a giant cyber-attack? The papers would have looked at that. What happens if we do FX using automated market makers [AMMs]? We don’t see that in practice, but also the practice is just at the beginning. And so the papers are kind of a little bit further afield, kind of thinking about, okay, five years from now or ten years from now, what choices might we have to make? Whereas practice is, okay, we’re going to issue this next month, how do we account for it in the central banking system? How do we make sure that it interacts with existing payment rails so that merchants can accept it, for example? So the problems that are being dealt with when you’re on the ground issuing a central bank digital currency are slightly different than what the policy papers or the academic papers are looking at, because those are like a little bit more forward-looking and kind of taking a more long term view.
12:26 Dominic Hobson: What if I put the question to you the other way around? You’ve pointed out that we don’t actually have a general economic equilibrium model to apply to this problem yet. It’d be a very difficult thing to do. You’ve also pointed out we don’t have many working examples of CBDCs, and those that we do have not experienced exactly high levels of take-up. But to put the question to the other way, has practice uncovered flaws or shortcomings in theory of CBDCs that’s described in the many papers that you’ve read?
12:56 Alisa DiCaprio: Well, okay, so first I want to make a little bit of a correction here. There is a general equilibrium model that is increasingly being used. So this is a DSGE (Dynamic Stochastic General equilibrium) model, a dynamic stochastic general equilibrium model. It’s been used for a good amount of time in cross-border country studies or, sorry, in country studies. But increasingly it’s being used in the CBDC literature to kind of consider what impact a CBDC might have. So there is a model, it’s not widely used, but increasingly is. There is a recent paper on Bahrain that used it as an example. So there is a model that exists. But back to your question in terms of sort of practice uncovering shortcomings in theory, I think until there’s a lot of data, we’re not going to see a big difference between those. Because of this disconnect between the kind of long-term visionary stuff that you see coming out of the academic literature and the more shorter-term policy decisions that are being made on the ground. I would love to see countries releasing more data. I mean, they have it. It’s just proprietary and it’s generally not shared. So to the extent that maybe neutral third parties like the IMF or the BIS can gather that data or can work together with those issuing countries, that’s going to be huge. I mean, that would be so great for the literature to just get a really clear idea of what happens to the velocity of money, for example. We can kind of model it, but it’s not that easy. Having good data would be really important.
14:53 Dominic Hobson: You use the term the long-term visionary thing. Is there a risk here? And this is probably a bigger question than it appears. Is there a vision here? We’ve been talking about CBDCs now for, in my case, seven or eight years now, possibly longer for other people. Is there a danger here that theorising about it, the vision thing, is running so far ahead of practice that it’s actually becoming a problem? We’re talking about something which has no practical presence in the world at all. And the more we talk about it, the more difficult that practical presence comes to achieve. I mean, in short, is theory run too far ahead of practice?
15:33 Alisa DiCaprio: That’s a very interesting way to phrase that question. I think theory must run ahead of practice. It’s important that it does and it’s good that it does, but certainly some of it is going to be wrong. I mean, take the example of AMMs that I mentioned earlier. Last year, I think there were maybe three or four papers about it. Is it something we’re going to see anytime soon? Certainly not, because it’s kind of taking an instrument that you see in DeFi [Decentralised Finance] and moving it into traditional finance and there’s a number of different adaptations that would have to happen that would be fairly complicated. So, yes, I think to answer your question directly, theory has run ahead of practise, but I think that’s its role. It absolutely should be doing that because five or ten years from now when central banks are thinking about, okay, well, how do we do FX with low liquidity currencies? Maybe AMMs becomes the answer and we already have a solid foundation of beginning discussions that we can then take forward. So I think there will be ebbs and flows in terms of theory and practice working together as we go forward.
16:55 Dominic Hobson: If I was running a central bank, I’d be wanting to ask you what have you learned that will actually be useful to me? And what I mean by that is, has this literature review enabled you to distil all this theory and all this practice into a sort of set of, I was about to say maxims, but I probably don’t mean that. I probably mean some sort of empirically sound assumptions which could be used by any central bank in any marketplace as they begin work on a CBDC project. Here are the eight or ten things which you can absolutely bet will influence how your CBDC project unfolds. Do those sound assumptions exist or not?
17:37 Alisa DiCaprio: I would say sort of, which I know is not the answer that you’re looking for, but it gets a very.
17:45 Dominic Hobson: It’s a very economist’s answer that one, I think.
17:48 Alisa DiCaprio: On the one hand, and on the other hand. I think certainly there are some kind of benchmarks that we see in most of the issuing economies and that the literature supports. So, for example, one would be interest rate on CBDCs. Every issuance has a 0% interest rate. Doesn’t have to. I mean, this is an instrument you could put an interest rate on. But what we know from the literature, though not really the practice, because in practice we only have zero, is that the higher the interest rate, the higher the adoption. And that’s kind of interesting, given what we talked about earlier, where adoption is incredibly low. So maybe the solution is to put on an interest rate. But then, of course, interest rate we know from the literature has a number of other issues, which is that if you put an interest rate on it, you’re definitely going to disintermediate banks, because nobody is going to want to keep their money in banks if you can get a higher interest rate on your CBDC and you can keep it in a wallet. So I think there are certainly things like that. Or commercial bank disintermediation – there’s been an enormous amount of work on that. I would say in the literature where we know the most about is financial stability. The literature has gone through every single potential problem with financial stability and how that could be adjusted by central bank digital currency, or what we could do to attenuate any impacts of central bank digital currency on creating financial instability. So I think in that area we know a lot. The literature is incredibly helpful. There’s not too much more that needs to be done now. Where there does need to be much more research, which is difficult to do, is in monetary policy. So how is the central bank going to deal with the issuance of a CBDC, and how will that impact monetary operations? How will it impact liquidity management, liquidity forecasting? The literature has much less information on that, but I think that’s also because that is something that’s much more specific to each economy. Whereas financial stability is sort of something we understand. It’s not so different between every economy, although there are certainly differences. So just to kind of tie it off, I think there’s no single blueprint, though the literature gives us a very good sort of path forward on financial stability.
20:17 Dominic Hobson: This risks asking the same question all over again. But does the literature also give you a sense of the paths you shouldn’t go down? In other words, is it possible to discern a CBDC project at quite an early stage where it’s going down paths or doing things or making assumptions which mean it’s bound to fail?
20:37 Alisa DiCaprio: I have not seen that, because I think it’s unclear at this stage if there’s anything that would definitely lead it to fail. I mean, we can say certain things like if there’s no merchant adoption, that would probably be terrible. If there’s no reason for consumers to use it, that would probably also be not great. So I think to some extent, the answer to your question is targeted use-cases. So it’s pretty clear that there need to be obvious use-cases for why you would issue this. So, for example, a lot of the current issuances focus on financial inclusion. So if there’s a large part of your population that doesn’t have access to bank accounts, this is one potential way to give them access. Now, whether that actually bears fruit in practice is unclear. Given the adoption rates, it seems that that is not really happening, but it could certainly be the timing of the issuances. Most of them happened during COVID. It could be educational campaigns around this. So it’s clear where the problems are, but it’s less clear what the solutions are. And I guess I would go back to the one thing that I am fascinated by that the literature doesn’t tell us, which is what is consumer behaviour around central bank digital currencies? I mean, this is the critical thing we don’t know. How do people feel about them? How do people use them? To have that kind of data … That would really give us a much clearer idea of whether a project is going to fail or not. Because when you ask it that way, it can fail because of the central bank. It can fail because of commercial banks, wallet issuers, or consumers or merchants. And the large entities, like the entities in the financial sector or merchants, we have a fairly clear idea of what is going on with them. But consumers, we have almost no insight into consumer behaviour.
22:44 Dominic Hobson: This is my final question, and in a way it’s just asking you where we’ve got to. You’ve pointed out there have been very few CBDCs in issue. Those that have been issued have experienced low take-up. It’s been difficult to make even those use-cases which appear to be obvious, difficult to make them work in practice. You’ve said a number of times there’s the shortage of data to work out what’s going on here. And we have very little experience, or indeed data, about how consumers and merchants are actually going to interact with these CBDCs. So my question is, – we’re in the pioneering stage here, clearly on CBDCs – and maybe the pioneers as always have got the arrows here, because they were prepared to not only start work on something, but actually put something into the marketplace possibly prematurely before everything had been thought through. Is that where we’ve got to? We’ve had some pioneers who’ve just gone a bit too early on assumptions which haven’t proved to be valid, or what? What do you think? Where have we got to?
23:52 Alisa DiCaprio: I am so proud of the pioneers. I think somebody had to be a first mover and I’m delighted that it was developing countries because I think this is absolutely the correct way to do this because, if you think about any kind of project, you want to start in the simplest way. And what do you have in the markets that have issued central bank digital currencies? You have relatively uncomplicated financial sectors. You have a limited number of commercial banks. You have a limited number of payment service providers. And so this is exactly where you want to start because you want to not have to deal with the complexities that you would have in the US market, which is one reason, I think, that it’s taking much longer for markets like the US and like the EU to move forward with issuance. So I don’t think the pioneers moved too fast. You can’t wait forever. I mean, you can, I guess, but also what happens if you’re a pioneer is that you’re defining this field. So the way that central bank digital currencies are going to be designed going forward very much are based on these countries that have already issued them. So we see this all the time when we work with clients and they say, well, what is the benchmark? And the benchmark is everybody that’s already issued. So those pioneer countries are actually forming the future of central bank digital currencies. I guess, to your point, they do have the arrows, right? I mean, they’re the ones that are kind of defining the direction early on. But certainly, and I would kind of close this off by saying every country is constantly revisiting their central bank digital currency. They’re not like issuing it and going like, okay, best of luck, market, take it up. They are going back and seeing like, okay, well, how can we improve merchant uptake? Or – I think it’s Jamaica maybe, where they have kind of like a cash back to increase adoption. So there’s different methods that we see countries testing out and trying out and it’ll be really interesting to have that going forward, particularly when we have more complex economies that have much more complicated problems that will be based on the design that was done by these emerging economies in the beginning. I’m so delighted about that.
26:21 Dominic Hobson: It’s a good point to end on there. This is never going to be a fire-and-forget. This is going to be something which you have to keep going back to. It’s a continuous process where you have to adapt to what is happening in the marketplace. Alisa DiCaprio, thank you very much for taking the time to share with the members of Future of Finance what you found out from your extensive review of the literature about CBDCs.
26:43 Alisa DiCaprio: Great. Well, thank you. I really enjoyed it. I’m glad somebody was interested in hearing about it.
