David Andolfatto

The Impact of Central Bank Digital Currencies on the Banking Sector

Since the rise of digital currencies and cryptocurrencies, central banks are considering the role these new forms of money may play in our evolving digital economy. One of the ideas studied is the notion of a central bank digital currency. While people and companies can hold central bank liabilities in the form of cash, only licensed banks have access to digital cash accounts with central banks.

We’re joined by David Andolfatto, VP of Research at the Federal Reserve Bank of St. Louis. David was previously on the podcast to discuss his idea for Fedcoin, a central bank issued cryptocurrency. In a recent paper, he explores the impact central bank digital currencies may have on the monopolistic banking sector.

Topics we discussed in this episode
  • The state of central bank research on digital currencies and cryptocurrencies
  • The idea that central banks may hold Bitcoin reserves
  • David’s new paper on the impact of central bank digital currencies (CBDC)
  • The potential impacts of CBDC’s on the banking sector and our economy
  • The role of fractional reserve banking in our economy
  • How fractional reserve banking applies to cryptocurrencies
  • The Debreu model and the need for money in an entirely liquid market
  • David’s outlook for the future of Bitcoin and cryptocurrencies
Sponsored by
  • Toptal: Simplify your hiring process & access the best blockchain talent
. Get a $1,000 credit on your first hire at toptal.com/epicenter.
Transcript

Sebastien Couture:  Hi, welcome to Epicenter.  My name is Sebastian Couture.

Brian Fabian Crain:  And my name is Brian Fabian Crain.

Sebastien:  How’s is it going Brian?

Brian:  Yeah, it’s very good.

Sebastien:  Yeah, cool. Yeah, so actually as we’re recording this, I’m in your hometown.  I’m here in Basel and apparently not too far from where you grew up, which is interesting and just want to mention that over the course of the next few weeks, I’m going to be releasing some content that I reported here some interviews with some of the key members of the Hyperledger community and also Brian Behlendorf who we had on the show a couple weeks ago.  So look for those interviews on our YouTube channel and also, we will be tweeting about of course and sharing them on the social channels. But yeah, it’s a great conference and I’m really happy that Hyperledger invited me or the Linux Foundation sorry invited me to come out here and do these interviews.

It’s a little welcome change from the other side of things, which is like the Devcon and permissionless systems.  Here we’re seeing a lot of enterprise all the guys in suits, very different from Devcon I must say, but yeah so today we’re speaking with David Andolfatto, who is a researcher at the St. Louis Federal Reserve and we had David on years ago to talk about Fedcoin and so today we are fortunate enough to be able to speak with him again and we talk about a number of things including, the space and how its evolved, since we last spoke, the idea of Central Bank Digital Currencies.  We also talked about his recent paper that he wrote about the impact of Central Bank Digital Currencies on the banking sector and touch on some other interesting high-level economic theorizing about the tokenization of liquidity.

So here’s our interview with David Andolfatto.  Hi, welcome to Epicenter. My name is Sebastian Couture.

Brian:  And my name is Brian Crain.  We are today with David Andolfatto.  He was on the podcast before and actually we just checked before and it’s been three and a half years.  So long time, last time we talked about Fedcoin, which was basically the idea of how could the Central Bank like the Federal Reserve issue its own cryptocurrency?  I think there are a lot of people were like, what is this horrendous idea. So yeah today, of course in the meantime, lots of things have happened and we’re really excited to have David back on to speak a little bit about some of the – he said years back then and the activities around Central Banking and Federal and cryptocurrencies and some of the other research topic that he’s been working on. So, thanks so much for coming on David.

David Andolfatto:  That’s my pleasure, thanks.

Brian:  So yeah you before mentioned well, fortunately not much has happened.  So let’s come to that question. Since the last time you spoke, and I think at the time, it was hardly anybody from a central bank writing about Bitcoin and writing about crypto.  It’s become lots of activities. So what do you think are the most important development in the last three years when it comes to Central Banking and cryptocurrencies?

David:  Well, I think in terms of Central Banking, I’d probably say the most important development from my perspective is just a greater awareness of the possibility of the Central Bank, a Central Bank becoming more directly involved in issuing digital money to regular people and not just limiting the privilege to a set number of banks.

Sebastien:  Give us a bit of an overview, I guess from your perspective, what is the state of research and maybe specific projects being developed in the area of Central Bank Digital Currencies?

David:  Well, I guess a lot of that answer will depend on exactly how we define digital currencies.  I think that the research is multifaceted. I mean there is a lot more to digital currencies and just currencies as you know.  It basically has to do with database management. So, I’m aware of efforts throughout Central Banks in studying broader applications of distributed ledger or blockchain technologies.  There’s considerable amount of research being done trying to study the financial market implications of this emergent technology. What impact that might have for a regulatory policy going forward or Central Bank policy going forward and then there’s some work actually asking the question of whether or not the Central Bank itself might enter into the space and issue its own form of digital currency in this.  Primarily is in the form of what I would call just digital money, not necessarily a crypto asset like a digital token, but more of like just a regular account-based money that is available to non-banked individuals.

Sebastien:  So I guess on that side, we’re talking about like the realm of digital currencies from like all across spectrum, right so from having a digital ledger, which is based on a centralized database model or more when account based system like the ones that we see in cryptocurrencies and blockchain.  And on the other hand, we also have like a central bank’s positions on decentralized cryptocurrency is like Bitcoin and Ethereum with regards to the latter and you did say that there has been a broadening of awareness. How are Central Banks now looking at the public permissionless cryptocurrencies like Bitcoin and Ethereum.  Has there been a change of opinions with regards to how things were three or four years ago?

David:  A change in opinion.  I would characterize it more as a – not a change.  There was really no opinion before and that’s not just in Central Banking, but almost everywhere, because I would say as of three years ago the last time we spoke, not many people really understood what the underlying technology was and what it was delivering, relative to existing protocols in database management.  Central Bank researchers have become more aware of the underlying technology as has everybody else. And so, I would say that the way the opinion, it’s hard to generalize because what we’re talking about here are individual researcher sprinkled throughout the research divisions of Central Banks around the world and there could be a wide variety of opinions out there.  But I would say for the most part Central Bankers, being trained economists for the most part are highly skeptical of the endeavor in terms of providing at least a digital currency.

There maybe 10% I mean from what I’ve seen are less skeptical, but by and large there’s a skepticism of the role that our private money might play to displace a well-managed state money.  That’s how my measure of what the attitude is of at this stage.

Brian:  So the scepticism, what are the biggest things that make Central Bankers so skeptical of cryptocurrencies?

David:  Well, I would say that a combination of things to be generous, I think that one way to think about it is, the skepticism is based on our historical experience says with private money issues.  They haven’t often turned out particularly well. Now in fairness, it’s really hard to assess the historical data, because there’s so much going on, the interaction of regulations with the private money provision.  So it’s very hard to tell sometimes whether it’s the private provision of the money supply that should be criticized vis-à-vis the underlying regulation that prevents it from operating well, but by and large there’s just this intuitive feel based on the data.  If we look at the so called free banking era in the United States ran from about 1836 to 1863 and what you saw was just literally thousands of different moneys issued by thousands of different banks and the system functioned and it by some measures it functioned relatively well, but there’s just a great inconvenience of having these different private monies coexisting circulating this counterfeiting problems etcetera.

So rightly or wrongly I think that a large amount of the skepticism just comes from based on what we have seen in the history of private money provision.

Brian:  That’s very interesting, because we have this technology wherein and having people in the Blockchain space, like forward-looking, tailored to something totally new, its revolutionary, it’s never been seen before and the technology is so much at the center and then you have maybe like people looking at it from a central bank in this historical perspective and saying like oh, but it’s similar to the thing in the 1800’s in the private banking area I think applying that framework on it.  It feels like a very big difference now in terms of the perspective brought to the topic.

David:  Yeah. Well, it’s almost a universal truth that there’s almost nothing new when you look around.  There’s some example or examples we can draw on in history that are very close and this is no different.  Fintech Financial Innovations have been with us all the time from the beginning of recorded history. The invention of the checking account for example, that largely replaced small denomination bank paper notes.  This is a database management technology and it was facilitated with the use of wire communications and techniques for storing the data. Basically, what we see today is more or less the same thing as I’ve written about before, but on a much broader scale just because of the innovations we have in communications, storage capacity and also, cryptography and things like that.

Brian:  That’s interesting.  I find it a bit puzzling.  So you think really that generally the best way to predict the future is to say, okay how is it similar to some episode in the past and then use that?

David:  Well, let me put it to you this way.  Think about what Bitcoin is fundamentally and blockchain more generally.  What it is fundamentally and I think to my view, it is a database management system.  Money is one example of a ledger system. You have its information that is stored in accounts and you need a way to manage those accounts, keep the data secure and to manage the information flow across accounts, the debits and credits of money across accounts.  This is database management. It’s very important, but it’s nothing new and so in that sense we can look to history, because this database management problem is around us everywhere, not just in monetary systems, but as you know supply chains for example, or the relationships between suppliers and their customers and so on.

Big big issues, the perennial issues, so in that sense nothing new.  What is new? What is new are the emergent technologies that facilitate communications.  We can now speak by phone or communicate over the internet. We can now store data more securely and in larger quantities.  These are the innovations that lead to innovations in these database management systems. These are the innovations that permit wonderful technology like Bitcoin to emerge and no one can really predict exactly what’s going to happen, but at the same time one can appreciate that the fundamental problem is always the same.  It’s database management. How do you keep the data secure? And how do you read and write to the database? Keep the data secure and make sure that there’s easy and widespread access to the database or to the constituents that you’re trying to target.

Brian:  So I mean, I certainly agree with you.  Like you can look at Bitcoin and this is a perfectly fair description of Bitcoin, but I feel like you can also have a different way of looking at Bitcoin, where you could say Bitcoin is like this term decentralized autonomous organization, right?  So you have in a way all of these different players and the miners and the developers and the users and the exchanges and there’s no central organism, yet they somehow all coordinate this like massive system and it evolves in some way, right? So that’s something that feels pretty awful to me.  Do you see ….

David:  No, it doesn’t sound awful to me at all.  In fact, I’ve written on this saying it sounds like the most ancient thing that I can think of.  It’s called decentralized communal behaviour, little communes, little hunter-gatherer societies operate on this principle of that having any centralized authority, people getting together and communally deciding on how to manage the societal history for the general prosperity of the tribe.  This principle is not new at all. Again what is new is the technology that permits this idea of communal record-keeping to scale, on a global scale that is new, but the underlying principle is not new at all.

Sebastien:  I was just going to get to that, is that the way that I perceive blockchain in this context and whether that be cryptocurrencies or permission networks or whatever is that there are essentially governance mechanisms that attempt to apply governance to human interaction at scale and governance of human interaction at scale sometimes gets complex if you look at like democracy as a governance mechanism.  There are definitely advantages to it, but there’s also issues with democracy once you get to a certain point. Corporate structures also have their advantages and like their fail points and so to me blockchain is a type of governance mechanism that allows you to scale, that also has its advantages and points, where it might fail and it’s still finding the balance between what do you most care about? You care about user experience or privacy or the ability to have a governance that’s more decentralized or less decentralizing.  So we have to pick some of these and find the right balance for your application. Bitcoin has one certain way and a theory might go another way or a permission network, where you’re dealing with known set of actors might interact in a different way, much like our societies interact in different ways when thinking about governance.

David:  Yeah, I think that’s exactly the right way to think about it.  The database management as a problem. There’s many problems, many different, there’s no one solution fits all the problems.  So it depends on what the constituency is looking for in terms of the properties of the database management system. As you mentioned, if the decentralized record-keeping is something that the constituents find desirable, then a blockchain structure sounds like the type of database management system that could better serve their needs, but for a lot of people that’s just not important.  For a lot of people they are very willing to trust delegated recordkeepers, trusted historians if you like to manage the data, but by and large I see room for coexistence here. I don’t see why it necessarily has to be one thing or the other. I think that as always, we see coexistence to fit the various niche demands the constituents.

Brian:  Okay, so then I’d like to maybe bring a different way or different blockchain direction and hear your take on that.  So if you look at Ethereum this is idea of the world computer where anybody can write their application on it and it is interoperable with other applications and I can create my application and interrupt with yours, but we don’t know each other.  We don’t trust each other, anyone can deploy it on there. It’s a global censorship resistant. So people are now doing whether that’s lending or derivatives or these kind of organizational structures or issuing all kinds of assets to games. Does that not feel like something like really radically revolutionary novel, that just is not comparable to something we’ve had in the past?

David:  You’re really searching for something truly novel.  So let’s think about it here. This notion of a decentralized autonomous organization sounds pretty cool and I don’t mean cool necessarily in a good way.  It’s either really exciting or could be very frightened too if you start thinking about it. Can I off the top of my head draw some analogy from history? I mean what do we got here?  Decentralized autonomous organization is just basically there’s no central authority. So that’s okay. Yeah, where you got me on this one. Maybe there is something we basically have the prospect of robots governing how contractual terms are executed.  Some sort of contract that we could write up and code and share. We write a contract. I don’t know you, you don’t know me. We can write the terms of the contract, the terms of the contract can be executed on publicly available information and the terms are going to be contracted whether they’re going to be executed, whether we like it or not.  I mean, that’s I guess something that sounds very new. And in any case, it certainly is something that’s feasible with this new technology and I think it’s exactly that area that I think is the decentralized autonomous organization that I think will be seen policy makers are going to be starting to open their eyes a lot more to that dimension of this endeavor, because I think that actually potentially that part of it is likely to have the most profound implications for society.

Brian:  Yeah that I very much agree with and of course you’re absolutely right that with any powerful new thing, there is the upsides and the downsides and certainly those with dollars it’s very absolutely possible to imagine quite horrendous ways this could be used as well as wonderful ways it could be used.  So there’s no and I think it could be a perfectly reasonable stance in my view to say okay, I see those possibilities and I see the good possibilities and the bad one and for me I think the bad ones are maybe too likely or too bad so that I’m overall against this. I think that’s a reasonable position to have.

David:  I’m glad to hear that you think I’m reasonable.

Sebastien:  Before you move onto the next topic, which is this paper that was recently released.  I wanted to ask you your thoughts about it. So in the Bitcoin space, there are some people that throw around the idea that Central Banks should and government entities should start holding crypto in their reserves.  I would like to get your thoughts on this, if you think that’s a good idea or something desirable or not at all.

David:  Yeah. Well, that’s – it’s interesting.  I’d like to know yeah, I personally think that Central Banks around the world should start stocking up on reserves on my own personal IO use and I guess the point of that really terrible joke is to ask what is motivating people saying things like that?  It sounds like they’re just trying to generate some sort of additional demand for something that they’re heavily invested in. The idea of a central bank holding reserves in some security like crypto or Bitcoin to be specific to me seems strange. I’m not sure why.

Sebastien:  I mean maybe not Central Banks, but perhaps government entities that typically would hold currencies of other countries in reserves or gold or perhaps some sort of security to hedge against risk or to, I mean I’m not sure exactly, but I think you get the idea.

David:  Yeah, yeah.  I guess I mean, I have to say I don’t think that that’s a particularly good idea, but – but I mean we live in second-best world or maybe even third best world.  So could I imagine scenarios where that might make sense? And I guess I can, I mean one scenario would be for example, if you’re living in a country that permitted say banks to issue liabilities denominated in bitcoin, it’s perfectly feasible banks could potentially denominate their deposits or make their loans denominated in whatever they want legislation permitting.  Now I would actually argue for legislation that would prevent that from happening at least for banks that had access to deposit insurance or the lender of last resort facility to Central Bank, but I imagine you’re living in a country where for whatever reason banks are permitted to write loans and issue deposits denominated in Bitcoin, then we have a lot of historical experience of what happens with these types of structures.  Just replace Bitcoin with gold, what you get is the possibility of these bank runs and in fractional reserve banking systems, the banks collectively are not going to have enough gold or Bitcoin to make good on their obligations and so you could imagine a government wanting or instructing banks to hold sufficient Bitcoin reserves, if they’re going to get into the business of issuing Bitcoin denominated liabilities.

So those are a lot of ifs, I mean if, if, if, but whether government should get involved in this as a matter of principle, I don’t know I could argue against I mean you could say listen, if you’re a bank and you want to get into this business get into this business, but make sure that either you hold a 100%. Bitcoin back make, hold sufficient reserves of Bitcoin or – then the threat is if you get into trouble, we’re not going to bail you out, but the problem with governments is that they can never commit to those promises and so I just think it’s a bad idea.

Brian:  Yeah, I mean, I guess let me try to articulate what I think is the position and yeah, I think basically the position isn’t and you would probably disagree with this that okay Bitcoin is going to grow and more and more people will use it and it will become this digital gold thing, right that’s held by many individuals and that so over time that you may have an erosion of the value and the trust that some people have in fiat currencies that will increase the shift to bitcoin and said that then it will be important for Central Banks to have in their thing backing their currency have, bitcoin reserves just like they have maybe gold reserves today and then there could be some sort of mechanism. where this – if let’s just say that the different Central Banks believe that there is some good chance of that happening and if it happens it will have some effect on Bitcoin price, they then have the incentive to be basically first mover and say okay, I’m going to go ahead, because then I’m on a relative basis, better off than the other Central Banks and so yeah.

David:  Yeah, maybe but my view is if a central bank can’t be trusted to manage the domestic money supply in a responsible manner, they should get out of the business and just let Bitcoin take over, just let private banking take over.  So why should a Central Bank intervene by buying up a Bitcoin when it has the power to create its own money and manage it responsibly on its own. Buying up Bitcoin doesn’t necessarily endow the bank with any magical management prowess.  So I just think it’s wrong. Either the central bank can be trusted to manage the money supply responsibly and this is often times a very complicated thing, because the way it interacts with the fiscal authority, but this should be just all ultimately to the voters of the jurisdiction and if it turns out that the Central Banks and fiscal authorities cannot be trusted to manage the money supply in this manner, then I think naturally these other competitors like Bitcoin or gold or whatever will provide the substitutes that people want, that are better managed, but I don’t see any reason for why the Central Bank should dabble in Bitcoin Reserves.

Sebastien:  So moving onto the next topic, we want to talk about your paper, which came out in October and is titled “assessing the impact of Central Bank digital currency on private banks by digital currency”.  I presume hear you’re talking about little currencies of the crypto flavour and so why – how does this paper build on your previous research topics with regards to Central Bank cryptocurrencies?

David:  Well, so cryptocurrencies again, this is more of, I make the distinction between what I call it digital money and digital cash, I guess.  Digital money, we already have digital money. We use digital money all the time, debit card transactions. Most of our transactions are done with digital money.  We have bank accounts and the idea of Central Bank Digital Currency is the idea of instead of having a digital money account with a regular chartered bank, why not permit people to just open up bank accounts with the Central Bank directly.  That’s what I mean by Central Bank digital money. One could go one step further and have the central bank or some other entity issue what I call digital cash, which has more of the token aspect to it, but that would be distinguished simply by, I think the analog would be like imagine the Central Bank permitting users to open up anonymous Swiss style bank account.  So in that scenario people would be sending money from account to account and would be relatively anonymous, as opposed to the system we have today, where you’d have to identify yourself, you identify your account number and you know that I’m sending money to you.

So my paper was really about Central Bank digital currency, which I say I think probably the empirically more relevant case, because I don’t think a lot of Central Banks will be willing to experiment with the notion of issuing, the equivalent of digital cash, just for a lot of regulatory concerns, like know your customer requirements or anti-money laundering rules.  So the purpose of this paper was to say look, if you take a look at the way the payment system is structured at the United States today, there seems to be two classes of individuals, here agencies I guess. On the one hand you have these big powerful depository institutions, these private banks that have access that can hold accounts directly with the Federal Reserve Bank.  These accounts are interest-bearing accounts. They presently earn something like 2% interest. The banks can send money between themselves using Fedwire, which is a real time gross settlement system.

So you can send money instantaneously and there’s like trillions of dollars that flow through this system every day and moreover the banks, it costs almost nothing to operate this database management system.  It’s basically free as far as the big banks are concerned. So this is the banks on the one hand and on the other hand you have the retail experience. The one that I have to go through or some small business, for example.  They either have to use cash which is very costly. I mean, it’s filthy. It takes a lot of resources to keep it secure, to transport it securely to a bank, to deposit it. It’s subject to theft for example. If instead you want to use a digital money you have all these interchange fees you have to pay.  I mean and this can often be a significant fraction of the sale. So in contrast to these big banks that pay basically nothing for instantaneous payments, you got the small business person who’s working on very small margins, they have to pay these big interchange fees and the payments don’t clear for two or three days.  It’s on a completely separate the ACH system, that was basically developed in the 1970s.

So this proposal in this paper is to say listen, what if we combine / permit these regular retail people to have access to interest-bearing Central Bank digital money, where the payments clear instantaneously on a real-time gross settlement system.  Why don’t we do that and the answer one often gets is that there are some I guess constituencies in the economy that are very opposed to that system and typically they are the incumbent banks. People argue that these banks are going to be worried about whether or not they’re going to be losing business.  Somehow it’s going to disintermediate them in some manner. What if everybody moves their money to the central bank for example, what’s that going to do for banks who are interested in funding their Investments? And so the purpose of my paper was to study that question formally in the context of an economic model, where you had basically monopoly banks and that’s very different than what’s commonly done.  What’s commonly done in this literature is to assume that the banking sector is competitive, which is I mean everybody knows it’s not quite competitive, but the other extreme is to think of it as being more oligopolistic or more monopolistic and that’s what I do in my paper.

I argue that that’s a much better approximation of the current US banking system.  And then I ask the question of what happens if the Central Bank introduces an interest-bearing digital money instrument for the masses.  What effect does this have on on the lending activities of monopolistic banks and also and their profits basically and what does it do in terms of financial inclusion?  What I find in the paper is that not surprisingly the introduction of this Central Bank digital currency increases financial inclusion, because it makes it more attractive to hold interest-bearing digital money at the Central Bank instead of the zero interest bearing paper stuff that relatively poor people are now forced to hold.  So one, it increases financial inclusion. It makes the regular depositor better off. It has no effect on the ability of banks to fund their Investments. I think that’s a striking result. So for banks who are arguing that’s going to impinge on their ability to fund investments, I think that that’s a very questionable claim that they’re making and then finally what I find is not surprisingly that it cuts into bank profits, but it cuts into bank profits, but it makes everybody else better off and it doesn’t impinge on the bank’s lending activity.  So that’s basically the upshot of the paper was tight how to identify at least in the context of a simple model what the likely impacts, repercussions of introducing a central bank currency would be on the incumbent banking system.

The conclusion is basically don’t worry about it.  For us the banks will have lower profits, but Central Bank and governments are not put in power to maximize the well-being of incumbent banks.  They are put into place to maximize the well-being of the broader population. So I say on those grounds this paper justifies that type of intervention.

Sebastien:  There’s a lot to unpack here and it’s a really interesting paper.  It does also describe this model that you’re talked about in a very mathematical form.  So if for those who are not very comfortable with reading lots of mathematical formulas and mathematical models, this might not be the paper for you, but you described it and summed it up very nicely.  I would like to come back to the notion of digital currencies at Central Bank and the spectrum that we have there. So on one hand, we have something that’s quite familiar I guess which is akin to having a bank account, right?  So there’s a there’s a ledger, it’s privately owned and privately managed and that ledger has a set of accounts and those accounts have balances. Instead of having that ledger at a bank right, like Chase or Wells Fargo or BNP or whatever, it’s with the central bank.  So essentially rather than holding cash and money that you have on this private ledger, this digital money is held with the central bank. And then there maybe gradients of that and these are theoretical of course, where you get to a point where perhaps a central bank has some sort of a DLT style or blockchain style digital currency, where essentially the central bank would be the only note and what’s interesting with this idea and I think this is probably what we discussed in our last conversation, is it here you have permissionless innovation on top.  So the wallets, the applications that you can build on on top of these, on top of this theoretical Central Bank cryptocurrency, the Innovation is opened up for anyone to partake in.

David:  You clarify what you said.  You said there’s a single node?

Sebastien:  Right, where the central bank essentially is using a little bit.

David:  Does the clearing?

Sebastien:  Right, does the clearing, but it is an open stack of technology so that one can build applications on top.  So regardless of where you were one thinks of on the spectrum, the paper addresses, so the more general idea of okay, a central bank has accounts where individuals, companies, small and large in whole balances there.  And in fact since our last conversation, this is something that I thought about quite a bit and talked with people about it and it opened my mind to like the role of the central bank’s play in our in our economy and also the role what banks play and one thing that I find really interesting about this is that opening accounts to individuals and businesses at Central Banks effectively renders some of the activities of a bank obsolete [indiscernible] [00:41:08] from the landing part I believe.  So if this was the case, if you if you could have an account with a Central Bank rather as an individual and that account was generating 2% interest, what is the scenario? Like, what does that look like in terms of the more like broader economy? What is the effect that has on the economy as a whole?

David:  Yeah, yeah, this is something that ongoing research is trying to answer.  Okay, so the short answer we don’t know for sure and of course, my paper was just an example of trying to chip away at that and try to understand what the likely effects might be.  Some of the likely effects might be, if I had a guess …..

Sebastien:  Yeah, let’s hypothesize it.  All of a sudden people don’t hold savings in banks anymore. Savings are no longer held in banks.

David:  Okay well, back up a second.  I mean that’s not necessarily the assumption you want to make. The assumption you want to make is suppose that the Central Bank offers a competing product that banks do.  Now first of all, we have to think about this. I have in mind something modeled more like the old U.S. Postal Savings System and this was an old idea.  Many countries have postal Savings Banks and the idea is that the government is getting involved in providing a very basic utility service. It’s not a full-service account.  It’s just a basic plain vanilla bank account that could potentially pay interest, very low fees, fully insured and it just that you make payments. You just pay people using your phone or whatever.  Now you’re charging interest on that, so banks are going to have to compete against that product and the way they’re going to compete against it is one of two ways.

They’re going to like start raising the interest rate on the deposit accounts they offer you to attract your business, but there they are more likely that could try to compete on the basis of non-interest bearing benefits.  There’s a whole bunch of other services that banks offer. So they’ll get into the full service bank account. They might offer insurance on the side. There’s all sorts of things that the bank’s might still profitably be able to do even though this basic utility service Central Bank digital currency is available.  So I don’t think you’re necessarily going to see deposits flocking from the private banking sector into the Central Bank. You might see some and the fact that the central bank is paying interest on these deposits is likely to force the private banks to offer better terms to depositors, because if you take a look right now, take a look at right now since 2015 when the FED started raising its interest rate and take a look at what’s happened to deposit rates in the United States.  They’ve basically remained zero. Even though the interest on reserves, the interest rate that the Federal Reserve pays banks is 200 basis points. At the same time, you see the lending rates that banks are charging have gone up appreciably. I think the lending rate on home secured loans is something like 600 basis points. So they’re making that big spread. So my view is that if nothing else even if nobody comes over to the central bank, the very threat of the central bank offering a 2% interest-bearing account would be sufficient to induce the banks to compete and so they’ll be forced to compete to keep the deposits and this will show up as higher interest rate on our accounts and better service ultimately as they compete to keep our accounts.

Brian:  So I want to ask related scenario which may seem outlandish you, but maybe not right?  I think we talked about it a little bit last time maybe. So I mean, let’s say now because right today we have this Fractional Reserve System, right?  So where you have banks can basically create more dollars, right and you have this deposit and Bitcoin anyway is like a full Reserve System, right? So there’s you have a Bitcoin, you have a Bitcoin like I can’t give you or nobody can really issue Fractional Reserve Bitcoin or when it happens it’s considered a fraud right?  Like let’s say on Mt. Gox you could say there was some sort of a Fractional Reserve Bitcoin system at one point. So why not create full reserve US dollars, right? So, there is no fractional like let’s say you made a cryptocurrency and it’s the US dollar cryptocurrency and anybody can check the ledger, right? You sent me some US dollar, I can check on the ledger and say okay this is real US dollar, so it wouldn’t make any sense for anybody to accept some sort of fractionally back US dollar.  What do you think of that? Is that a good idea?

David:  Well, in fact the Central Bank digital currency that I propose is essential a variant of that.  Right, the accounts that you’re holding are 100% backed. There’s no fractional reserve banking there, but the basic idea that you’re espousing right now is an old idea.  It’s just called narrow banking, the narrow banking proposal or full reserve banking proposal. It’s called the Chicago plan. If you take a look at the literature, there was a big proposal I think in the 1930s from the University of Chicago making exactly this point.  The idea that fractional reserve banking should be separated from the payments system, what that basically means is that if you’re a bank and you’re in the business of processing payments, you better make sure that these liabilities that you’ve issued, these deposit liabilities are fully back with reserves, either in the form of gold, could be Bitcoin government treasuries, whatever.  They have to be fully backed and that the credit market should operate separately from the banking system. I honestly don’t know where I come down on this idea, because I’m still exploring the history and also the theory here, the idea of fractional reserve banks is not inherently a bad idea. The question is what are banks doing when they’re issuing fractional reserves. When they go to you and make you alone, what are they doing?  They’re creating money out of thin air. And they’re giving it to you. What they’re in effect doing as monetizing your human capital. They’re giving you credit, they’re saying you know what Brian, you know what Sebastian, we believe in you. Here’s the money and by the way, it happens to be redeemable for cash or we’re going to have to keep some cash on hand, but we know that by and large you can just pay this money account to account.  It doesn’t necessarily have to be redeemed in cash, except under special circumstances and so what the banks are effectively doing is monetizing the debtors. If I have an investment, I want finance, the banks are in a position to translate your dream, your investment into monetary purchasing power. Everything hinges on the ability of the bank to make good decisions in assessing credit risk. By conditional on assessing, making the correct assessments, the idea of issuing and creating money to facilitate positive Net Present Value projects like yourself, like your education and holding only a little bit of cash on the side just to satisfy the occasional redemptions that come whenever people have to like pull out a $10 bill to pay for a cab ride.

There’s nothing inherently wrong with that business model in my view.  So we have to ask ourselves. What is it that you’re trying to like? Why would people you asked that?  You said something, I’m sorry that if people had access to a full reserve account that they would never deposit their money in the fractional reserve account.  I think that’s false. The difference between these two accounts is that the fractional reserve account is going to offer you a higher interest. So there’s going to be a cost-benefit analysis.  This high interest rate that the fractional reserve bank is able to charge is a by product of a profitable investments its financing and you and your friends and other positive Net Present Value projects.  And so it’s not clear to me that that depositors are going to go flocking to the narrow banks. So narrow bank is going to be paying a very low interest rate. It’ll be safe, but a very low rate of return and so there’s going to be a trade-off.

Brian:  Okay, this is a great point, but I guess where in many people in the Bitcoin space, right if they look at the history of banking, right?  They see fractional reserve banking and correct me if this history’s wrong, but it seems like it basically started as a bit of a fraud, right?  So people would put in gold, let’s say in a bank and then you give up a receipt for the gold and then the bank could give out more receipts, then your gold without people necessarily knowing and it does have a bigger business, which is basically what you just described.  I guess that I think you’re probably right that if you have this full reserve like the crypto US dollar, probably still people will create these fractional systems on top of it, but in today’s system you don’t like in that system, it will be transparent whether you have a real US dollar, which you don’t have some kind of – doesn’t have a dependency on how well a bank has managed its credit portfolio versus JPMorgan US dollar.  Wage JPMorgan promises you if you go down and bring JPMorgan US dollar they’re going to give you like a real US dollar, but I mean you’re aware that there is an additional risk there, because JP Morgan doesn’t have full backing, because they issue more JPMorgan US dollars than they have real dollars.

David:  Correct, while they issue more JP Morgan dollars than they have real dollars, but they don’t necessarily issue more JP Morgan dollars and they have assets.

Brian:  Right, right.

David:  So the whole question, the whole issue of fragility in the banking sector is the liquidity of these assets, but that’s the whole business of banking is to render those illiquid assets liquid.  If those assets were already liquid, if your human capital was already liquid, you wouldn’t have to use it as collateral for a loan. You could just sell off pieces of yourself, tokenize yourself.  Little bits of human capital of Brian is Sebastian claims on your future wages and you wouldn’t have to go and raise money at a bank. So the whole reason our banking is there to overcome these frictions.  You said one other thing, you mentioned that the history of banking is a history of fraud. You have to be careful. You go and look in history, all sorts of activities are fraudulent, why pick on banking. There oil and mining is with fraud as well.  We don’t band mining, I mean we don’t mining exploration. So the goldsmith story that you told is a one that’s commonly told in textbooks and I guess there’s some truth to that, but it’s an anecdote. You take a look at monetary historians, people like George Selgin or Warren Webber who study these episodes, go take a look at the Scottish Free banking episode in the 1700s.  There the banks had I think double liability. I mean the bankers if their operations went under, they were personally liable to make good on the losses for example.

So it is not – and that system arguably by my reading work pretty well.  So you have to be careful at reading select anecdotes from your favorite economics textbook and generalize as to what the historical experience has been for fractional reserve banking.  The story is mixed, was there a fraud of course, there was fraud. There’s fraud in every endeavor in history. That’s not the issue. The issue is just how did it work. Did it work relatively well or not and if not, why not.  Was it because there’s something intrinsically wrong with fractional reserve banking? Or did it interact some way with the existing legislation that prevented the banks from otherwise operating fractional reserve Bank from operating in a more stable manner.  Sometimes that’s a very key part of the equation is the legislation often prevents banks from operating in a way that you might consider in a more prudent manner.

Sebastien:  Yeah, this falls into the next topic, which is the tokenization of liquidity and you made a great point that a bank’s job is to create value out of things that are illiquid and I think that to Brian’s point and I think this is what he’s alluding to is and perhaps since the last banking crisis in 2008, as people have become aware that banks were backing debt with illiquid assets, but giving way too much value to these assets with regards to how much they were actually worth.  So there was a great gap between the underlying value of an asset and what banks were lending out and I think this is what has caused a lot of the skepticism around banks generally and a lot of the skepticism I guess in the Bitcoin space perhaps, because Bitcoin came out around the same time.

David:  So after every financial crisis, people suddenly recognize the problems with banking.  The Federal Reserve was born out of the financial panic of 1907. The US Postal savings system was formed in 1911.  The Postal Savings Bank they offered small deposits that were fully insured by the government. It was a tremendously popular program and it came in the aftermath of the panic of 1907.  So what you say is true. It’s just in some sense remarkable that the same sort of mistakes keep on getting repeated although in some defense, I mean the crisis, the most recent crisis did not happen at the retail level.  The recent crisis did not really happen in the regulated banking sector. People have deposit insurance, nobody lost their money in the bank. There were no bank runs, deposit insurance rose to $250,000 in the United States.  The crisis this time happened in the so-called shadow banking sector, the less regulated more opaque sector of the economy, but the principle is exactly like what you said. It’s liquidity mismatch and also some very bad or questionable investments were made and some questionable ratings were attached to these investments and this all came crashing down not in the retail sector this time, but in the wholesale sector, but the principles are the same.  Now I think what you want to ask is like what do we do about it? Like what did these new technologies? How can they mitigate these problems?

Sebastien:  Well, your blog post and will look to this post in the show notes that so describes a scenario where one can tokenize his own assets.  So basically, the job of a bank or one of the jobs of bank is to say, okay these are illiquid assets and we’re going to lend money with those assets as collateral, but if one could say okay, this is you know, I’m meek.  I consider my time to be valuable, my experience and I have these abilities and so therefore I’m going to use that as collateral in some transaction or perhaps it’s physical assets, like I’ve got a farm with some cattle or something, some basic example.  So maybe yeah you walk us down this route and with this blog post in mind.

David:  One thing I love about this whole endeavor cryptocurrencies, Bitcoin, blockchain, it has brought to attention what the fundamental issues are in monetary theory.  This idea of money being a ledger for example was nobody knew what I was talking about, what I used to. I’ve been lecturing on this for almost 30 years. I preface every one of my monetary theory courses with, why is it when I go to buy a cup of coffee?  Why can’t I just pay with a personal IOU that represents a claim to the lecture that I’m giving you guys? I’m a professor. I give economics lectures, people value that. Why can’t I just issue tokens? I didn’t use the word tokens. I used personal IOUs, but it’s the same thing.  I should in principle be, if it’s some famous person probably they could do it in fact, but I’m just anonymous and if I attempted to do so, or to issue a claim against my house for example, suppose I want to buy a cup of coffee in the morning, a dollar cup of coffee with a one one-millionth share of the bathroom of my house.  Why can’t I do that? The answer is in economic theory you can do that and it’s exactly in a frictionless world that’s exactly what you can do. There’s no liquidity issues.

I mean, if you can use your imagination and just imagine a world, where these frictions are absent, you should be exactly able to do what I described.  And so then the question is what are the real world frictions that prevent this from happening. Why is it that when I go to the coffee shop and I offer a 1-1 millionth slice of my salary as payment, why doesn’t that work?  And the answer is well, first of all, they don’t know who I am. They don’t know if I am a professor or not. They don’t know what claims I’m making. It’s very easy to fabricate information. I could be lied and almost surely I might be lying.  Second of all, even if I’m not lying what makes anybody think that I would honor the claims of these personal tokens or IOUs. I mean, I claim that my house can generate rental income of $10,000 a year and I issue a token against that revenue stream.  Well who’s going to enforce the payment of that token? Who’s going to enforce the tenants of my house to pay the rent? Me just because I’m recording the ownership on a blockchain or some sort of ledger doesn’t magically solve the problem of enforcing payment of the tenant or of enforcing my payment of my set part of my salary to the obligations I’ve issued.  So it doesn’t necessarily solve a whole pile of informational frictions that render certain types of capital like your human capital, my human capital illiquid. That’s exactly the job of banks. Banks have credit officers that are specialized to interview you, to go and make an assessment of your home. We trust these third parties to make accurate assessments of the underlying collateral.  Now that they often fail, of course. They don’t operate perfectly, but my question is how does a decentralized database management system or the datas recorded by in some Merkle tree structure, how is that supposed to solve these underlying frictions of basically enforcing property rights? And these kind of informational issues that are associated with the asset that makes the asset illiquid.

So I think these are the issues that need to be addressed in this kind of tokenization endeavour and I’m still studying it.  So maybe there are some good answers, but I don’t know what they are yet.

Brian:  So this is a very interesting point.  I mean I think it is commonly held view in the blockchain space that exactly one of the things that is being created is that all of those things are becoming much more liquid and that we would go to a world where you have almost perfect liquid market and to give an example, I once years ago worked a little bit in commodities trading and was given the execution of these trade and it’s like a complete nightmare and super tedious and one of the issues is that you have then let’s say I have a promise of payment, if I deliver these goods there, but this is a highly liquid.  So let’s say it’s a commodities trading company. It has all of these promises of payment, but it can’t actually sell the individual promises of payments on an open market. It has instead to go to the bank and the bank looks at all of it and it gives you a loan which is probably much less than you could get in a liquid market. And now there’s many projects, to give one example, there’s a project called centrifuge, where there are trying to take all – give anybody the ability to put like purchase orders or invoices and stuff like that and immediately sell them and then at same time you have things like prediction markets right, so you could have maybe liquid markets on information.

There’s a lot of work that’s being done on having exchanges right, that maybe can half liquidity and narrow prices even on a large number of markets, where there’s not big order books and stuff like that.  So it does feel to me like we are going to a place where okay, it’s not going to be perfect liquidity and perfect market everywhere, but get massive step in this direction. So did you see that as well and let’s just assume this actually did end up happening.  What would be the consequences of that?

David:  Well, do I see it happening?  Yes, I do see it happening, but I see it happen in even independent of blockchain technology.  I mean, it’s what you just said was true, things are becoming more liquid, because of the what renders objects liquid though.  It’s enhanced communications, enhanced security of information, making information more symmetric renders things liquid. And database management systems can be designed to enhance communication between different parties to render what’s on the books more transparent.  These are statements that we can make, independent of blockchain, but these are just the properties of a database management system in general. The question I have is, when people say tokenization they are specifically meaning using a decentralized database structure, where the data is recorded in some Merkle tree form and that the clearing is undertaken in a decentralized manner by some sort of consensus protocol.  That’s the part that I have some issue with. I have a hard time seeing where apart from the application in terms of like DAOs, I’m not entirely clear how decentralized consensus is supposed to enhance the liquidity of an object, independent of the other innovations that I’m speaking of, that are occurring even independently of blockchain considerations, and I’m sorry you had one more question. What is the limit? I mean what happens if things become more and more liquid?  I think that what you’ll see is basically a gradual disintermediation. The whole reason we have intermediaries like firms or banks or whatever is to deal with these issues. If we go to the limit in a world where everything is fungible that there’s no questions about when I make a promise. It’s a good promise that I’m good for it or at least that it’s risk-adjusted prices, is priced correctly in that world a lot of the institutions that we see around us, the very existence of corporations and firms would cease to exist.  There would be no need to organize activity through any intermediary. We could just go about our daily business and be trading with this very futuristic world, but we could in principle just undertake all of our economic activities independent of belonging to any particular firm or having a particular bank account at any trusted intermediary. That’s what the limit would look like.

Brian:  Yeah, absolutely.  I mean, I think the big difference is if you have let’s say these assets that are issued on this open system, then anybody can build things that like interact with each other, right?  So for example, you have this purchase order and somebody can use it on Ethereum to back alone and that doesn’t really work with normal Fintech companies right, because ….

David:  Well, that backup, the fact that it’s open is something that you don’t need blockchain to make a database open.  Do you?

Brian:  I think I would disagree with that.

David:  So if I publish my diary online, I can publish my diary online and everybody in the world can download it, that database to their computer.  That’s effectively a distributed ledger of my life.

Brian:  Yeah, but when like who has then the authoritative copy or like if somebody changes it, you change it retroactively or like how am I going to rely on that?

David:  I have the master copy.  I print out a new version every day.  Only I can write to it, but I make it open to everybody can read it.  You can’t go on my blog site right now and change my blog, can you? But everybody in the world can read it.  It’s a distributed open database. So I don’t need blockchain to make a database distributed in open.

Brian:  I think I would disagree with that.  I mean in this example, like you’re okay, you’re publishing your database, right and I can copy it locally, but now you fiddle with it.  You retroactively change something. Okay, I have the evidence but someone else has a different record now of the – blockchain massively gets rid of the coordination problem by okay, there’s one authoritative copy we all know the sequence in which it has been created.  Nobody can mess with it. Nobody can change it like you can go and change it back. Like you can go on your website, right? You can pretend you have some blog posts that you wrote a year ago that you didn’t actually write a year ago, but in a blockchain you would be able to, right.  So I can easily build an application that builds on your data feed without having to worry about like those kind of risks.

Sebastien:  I think if I could just interject here.  I think David is right on the point that this is possible through blockchain, but it doesn’t scale and this is coming back to what I was saying earlier about governance and I guess similarly one thing that’s similar in this sense were of course, we can achieve it, but it doesn’t scale is like PGP.  So in the 80s and 90s and even today I mean cypherpunks and people that are conscious about their security use PGP to send encrypted email and there is this like this trust system and everybody thought that we were going to move to the system to send secure data communications, but that system simply doesn’t scale, because you have to meet people in person and exchange keys on paper or like verify that your key is in fact the one that I have and blockchain solve some of the issues here is with the scaling aspect.  So like we can now do this at scale and we’re finding solutions, where this big secure exchange of information that is anchored in time and where I can go back and look at a record and know that that record hasn’t been tampered with. This is what blockchain enables.

David:  Yeah, so totally agree with what you said.  I think that we can come to a mutual understanding here by understanding that what we’re arguing here is the conceptual distinction between the write privilege to a database and the read privilege.  I’m talking about the read privilege. So I’m saying suppose that you trust the writer, me for example, suppose you trust that I’m not going to go back and fabricate old blog posts. I’m just assuming that, then there’s the question about the read privilege.  Can we make it open? And my answer is, yes. I don’t need a blockchain to make the database open and in fact, it could work perfectly function and I can be transparent. I can be honest. Of course, it requires you to trust the writer. If you don’t trust the writer, then what you say is absolutely true, what the blockchain enables the decentralized consensus protocol permits us to not having to trust the writer either and that permits the system to scale in a system where you want open, read privileges and you don’t trust any single writer to the database.

So I think we can reconcile these two.  I just wanted to push back that you don’t necessarily need blockchain to make a database open and transparent was my point.

Sebastien:  Yeah, I think you’re right, but what blockchain also does to a large extent is it allows you to forego having to even think about whether or not you trust someone and so as we enter and as we live in a society where our data is free flowing on all types of systems and we have interactions whether financial or purely on a communication basis with dozens, hundreds of entities every day, the blockchain allows you to forget about the trust issue.  You don’t have to think about whether or not I want to trust someone or whether or not that exchange is genuine and it takes up that out of the equation to some extent.

David:  Trust. I want to push back a little bit.  I know where you’re coming from and I agree, but on the other hand I don’t think that’s exactly true.  You still have to – if I’m using blockchain for the first time, I have to enter it with some degree of trust.  I can write C++ code. I don’t know the cryptography behind anything. How do I trust?

Sebastien:  This is true.  This is true.

David:  How do I trust that my automobile is going to get me to work safely.  I don’t know how the internal combustion engine works in detail. The way the trust happens is we on the basis of experience and so if it turns out the blockchain renders very good user experience, people will come to trust it, but the exact same thing is true of my bank.  So the question is really where do you want to place your trust, I think as opposed to not having to think about it.

Sebastien:  Of course, yeah, so this is all very fascinating,  I think we could go on for hours here, but we’re very conscious of your time and so before we wrap up, I just maybe want to get your final thoughts on – if you look forward 10 years from now and so at the rate at which the space is growing and at the rate at which government entities, Central Banks are now changing their – not changing, but observing these technologies and past experimenting with them, from your perspective where can we expect things to be in the next 10 years?

David:  Wow. This is tough, right?  I mean, you’re not going to hold me to any of this, but …..

Sebastien:  No, we could use a prediction market to come up with it.

David:  Right, I see basically more of the same in terms of the innovations in Fintech database management and I think that Central Banks will have to be increasingly on guard I think to make sure that they are managing their policies in a socially responsible manner.  I actually view this emergent class of technologies is useful in terms of like disciplining banks, Central banks in incentivizing them to really look for the use value that they can deliver society. If at the end of the day, they can’t pass that market test, they should probably disappear.  I don’t know if that’s not going to happen in the next ten years, but the other thing I think I would look for is again going back to what we alluded to earlier is this issue of decentralized autonomous organizations. I can envisage very very rapid spontaneous growth in that dimension along various aspects and I think this is going to give regulators and policymakers headaches, endless headaches, because I don’t think it’s something they can necessarily – they’re not going to be able to control and that makes regulators very very nervous and maybe it’s good, but I can also see some dangers and we could spend another episode I guess discussing what those might be, but it’s going to be very interesting I think.

I think it’s just going to see that DEO space develop very rapidly and with it the discussions by policymakers wringing their hands about what can be done about it?  What’s the proper public policy to take? Public policy view to take on these issues?

Sebastien:  Great. Well, I mean I guess we can say that we’ll have you back on again.

David:  I hope in less than three years though.

Sebastien:  With pleasure.  So do come back anytime.  We’d love to talk to you again about all these things and follow the trends with regards to Central Banks and how they’re experimenting or thinking at least where the research is going in this topic, it’s a fascinating topic.

David:  Yeah, there has been a lot of fun.  Always fun talking to you guys.

Sebastien:  Likewise. Thanks very much for coming out here.

Brian:  Thanks so much.

David:  My pleasure.  Thank you.