WEBVTT 00:00:00.000 --> 00:00:02.468 align:middle line:90% [SQUEAKING] 00:00:02.468 --> 00:00:03.944 align:middle line:90% [RUSTLING] 00:00:03.944 --> 00:00:05.912 align:middle line:90% [CLICKING] 00:00:05.912 --> 00:00:12.620 align:middle line:90% 00:00:12.620 --> 00:00:15.400 align:middle line:84% ROBERT TOWNSEND: So this is the second lecture, 00:00:15.400 --> 00:00:19.400 align:middle line:84% although the first full presentation 00:00:19.400 --> 00:00:23.960 align:middle line:84% of a body of material rather than just a summary. 00:00:23.960 --> 00:00:28.080 align:middle line:84% And in particular, we're going to talk about blockchains 00:00:28.080 --> 00:00:30.320 align:middle line:90% as a database. 00:00:30.320 --> 00:00:35.880 align:middle line:84% And the underlying themes here are a unified view 00:00:35.880 --> 00:00:38.580 align:middle line:84% of distributed ledgers and financial accounts, 00:00:38.580 --> 00:00:41.080 align:middle line:84% putting them in one bin, so to speak. 00:00:41.080 --> 00:00:45.440 align:middle line:84% Distinct views of money and the larger community perspective. 00:00:45.440 --> 00:00:48.840 align:middle line:84% So this is an outline of what we're going to do. 00:00:48.840 --> 00:00:50.900 align:middle line:84% Talk about blockchain as a database. 00:00:50.900 --> 00:00:55.160 align:middle line:84% Talk about the financial accounts. 00:00:55.160 --> 00:00:58.440 align:middle line:84% Take a broader view of the blockchain. 00:00:58.440 --> 00:01:01.780 align:middle line:84% This tension I mentioned last time between individual 00:01:01.780 --> 00:01:04.220 align:middle line:90% and community perspectives. 00:01:04.220 --> 00:01:09.700 align:middle line:84% Then some technical limitations of blockchains 00:01:09.700 --> 00:01:14.140 align:middle line:84% also reflected in current difficulties and some progress 00:01:14.140 --> 00:01:17.140 align:middle line:84% in mapping economies and next steps. 00:01:17.140 --> 00:01:22.900 align:middle line:84% So that's the overall outline of the lecture. 00:01:22.900 --> 00:01:26.820 align:middle line:84% And then we have the individual sections within. 00:01:26.820 --> 00:01:29.940 align:middle line:84% Blockchain is a special form of the database. 00:01:29.940 --> 00:01:32.620 align:middle line:84% We'll talk about the creation of Bitcoin, 00:01:32.620 --> 00:01:36.940 align:middle line:84% an incorrect association of blockchain with money, 00:01:36.940 --> 00:01:40.020 align:middle line:84% peer-to-peer exchange certificates, 00:01:40.020 --> 00:01:47.180 align:middle line:84% and central bank reactions to money and ledgers. 00:01:47.180 --> 00:01:51.020 align:middle line:90% OK, so here's a definition. 00:01:51.020 --> 00:01:54.860 align:middle line:84% Any data structure used to store information 00:01:54.860 --> 00:01:59.040 align:middle line:90% can be considered a database. 00:01:59.040 --> 00:02:02.920 align:middle line:84% A blockchain technology is a ledger 00:02:02.920 --> 00:02:06.760 align:middle line:84% to store information about transactions. 00:02:06.760 --> 00:02:09.100 align:middle line:84% So those are the objects in the database. 00:02:09.100 --> 00:02:11.280 align:middle line:84% And hence a blockchain can be considered 00:02:11.280 --> 00:02:16.240 align:middle line:84% to be a database and nothing more or less. 00:02:16.240 --> 00:02:19.920 align:middle line:90% Data is stored in signed blocks. 00:02:19.920 --> 00:02:22.600 align:middle line:84% The blocks are linked to each other, 00:02:22.600 --> 00:02:28.160 align:middle line:84% creating a chain of immutable, interconnected data entries. 00:02:28.160 --> 00:02:32.040 align:middle line:84% So that is an important mechanical aspect 00:02:32.040 --> 00:02:35.700 align:middle line:84% of how the blockchain database works. 00:02:35.700 --> 00:02:38.280 align:middle line:90% 00:02:38.280 --> 00:02:45.040 align:middle line:84% In particular for Bitcoin, for example, each node 00:02:45.040 --> 00:02:50.640 align:middle line:84% is trying out a bundle of the set of incoming transactions 00:02:50.640 --> 00:02:55.320 align:middle line:84% since the last committed block in order to form a new block. 00:02:55.320 --> 00:02:59.140 align:middle line:84% And these blocks consist of a time stamp 00:02:59.140 --> 00:03:04.180 align:middle line:84% when the block is created, a nonce used for proof of work 00:03:04.180 --> 00:03:07.800 align:middle line:84% as a reference to a previous block, the hash of it, 00:03:07.800 --> 00:03:11.180 align:middle line:90% and a list of transactions. 00:03:11.180 --> 00:03:12.940 align:middle line:90% Pause a minute. 00:03:12.940 --> 00:03:19.300 align:middle line:84% We're going to develop the material to go back and describe 00:03:19.300 --> 00:03:22.660 align:middle line:90% nonces, and hashes, and so on. 00:03:22.660 --> 00:03:27.100 align:middle line:84% But still I wanted to get the ingredients on the table here. 00:03:27.100 --> 00:03:31.360 align:middle line:84% So this is how you create new blocks from previous blocks. 00:03:31.360 --> 00:03:34.020 align:middle line:90% And this is a curious item. 00:03:34.020 --> 00:03:35.780 align:middle line:84% The blockchain for Bitcoin can be 00:03:35.780 --> 00:03:38.700 align:middle line:90% thought of as a state machine. 00:03:38.700 --> 00:03:44.580 align:middle line:84% But actually it's not giving you the current state automatically. 00:03:44.580 --> 00:03:47.700 align:middle line:84% You have to go back to the whole history of transactions 00:03:47.700 --> 00:03:53.740 align:middle line:84% since the genesis block in order to get the current state, which 00:03:53.740 --> 00:03:59.840 align:middle line:84% is a bit odd, but that's the way Bitcoin was constructed. 00:03:59.840 --> 00:04:01.880 align:middle line:90% OK. 00:04:01.880 --> 00:04:05.640 align:middle line:84% Oh, and then there's even more detail up here. 00:04:05.640 --> 00:04:11.320 align:middle line:84% Just anticipating more material on encryption and so on. 00:04:11.320 --> 00:04:18.600 align:middle line:84% So these messages are encoded with public and private keys. 00:04:18.600 --> 00:04:23.560 align:middle line:84% The keys ensure no one else can transact on someone else's ID, 00:04:23.560 --> 00:04:25.220 align:middle line:90% impersonating a node. 00:04:25.220 --> 00:04:33.220 align:middle line:84% I guess I should add the node is like a computer account. 00:04:33.220 --> 00:04:36.480 align:middle line:90% It's not a personal identifier. 00:04:36.480 --> 00:04:41.000 align:middle line:84% So you know who the node is but not who's operating the node. 00:04:41.000 --> 00:04:45.000 align:middle line:84% The keys ensure no one can transact on someone else's ID. 00:04:45.000 --> 00:04:47.500 align:middle line:84% There's a commitment to the transaction, 00:04:47.500 --> 00:04:49.880 align:middle line:90% so it can't be undone later. 00:04:49.880 --> 00:04:51.540 align:middle line:84% And these transactions are, again, 00:04:51.540 --> 00:04:54.420 align:middle line:84% just transfers of, quote, the coin 00:04:54.420 --> 00:04:56.540 align:middle line:90% from one account to another. 00:04:56.540 --> 00:04:58.940 align:middle line:84% And it's the transactions that are being recorded 00:04:58.940 --> 00:05:01.420 align:middle line:90% on the ledger in groups. 00:05:01.420 --> 00:05:02.500 align:middle line:90% OK? 00:05:02.500 --> 00:05:05.840 align:middle line:84% So back to the creation of Bitcoin. 00:05:05.840 --> 00:05:11.620 align:middle line:84% It was registered August the 18th, 2008. 00:05:11.620 --> 00:05:15.020 align:middle line:84% A white paper came out by someone 00:05:15.020 --> 00:05:19.180 align:middle line:84% under the pen name of Satoshi Nakamoto. 00:05:19.180 --> 00:05:22.620 align:middle line:84% The title of the white paper is revealing. 00:05:22.620 --> 00:05:27.900 align:middle line:84% A Peer-to-Peer Electronic Cash System. 00:05:27.900 --> 00:05:31.300 align:middle line:84% So remember the peer-to-peer aspect. 00:05:31.300 --> 00:05:37.440 align:middle line:84% And open-source code was released in January of 2009. 00:05:37.440 --> 00:05:41.980 align:middle line:84% We still don't know who Nakamoto is or was, although there's 00:05:41.980 --> 00:05:45.300 align:middle line:90% a lot of speculation about it. 00:05:45.300 --> 00:05:52.600 align:middle line:84% January 3, 2009, Bitcoin network was created 00:05:52.600 --> 00:05:56.120 align:middle line:84% when Nakamoto mined the first block, 00:05:56.120 --> 00:05:57.880 align:middle line:90% known as the genesis block. 00:05:57.880 --> 00:06:03.800 align:middle line:84% And nine days later, Hal Finney received the first Bitcoin 00:06:03.800 --> 00:06:08.920 align:middle line:90% transaction from Nakamoto. 00:06:08.920 --> 00:06:13.960 align:middle line:84% And five months later, in May, there 00:06:13.960 --> 00:06:16.760 align:middle line:84% was a commercial transaction where 00:06:16.760 --> 00:06:21.200 align:middle line:84% Bitcoin was used by a programmer to buy pizza, called 00:06:21.200 --> 00:06:26.600 align:middle line:84% Bitcoin pizza day, and a summary of sorts. 00:06:26.600 --> 00:06:33.080 align:middle line:84% Arvind Narayanan, a computer scientist at Princeton, 00:06:33.080 --> 00:06:36.080 align:middle line:84% is saying that all the individual components of Bitcoin 00:06:36.080 --> 00:06:39.080 align:middle line:84% originated in earlier literature. 00:06:39.080 --> 00:06:42.840 align:middle line:84% There's really nothing original about each of the pieces. 00:06:42.840 --> 00:06:45.880 align:middle line:84% But their complex interplay resulted 00:06:45.880 --> 00:06:50.400 align:middle line:84% in this Bitcoin, the first decentralized, Sybil-resistant, 00:06:50.400 --> 00:06:55.180 align:middle line:84% Byzantine fault-tolerant digital cash system. 00:06:55.180 --> 00:06:59.880 align:middle line:84% And we'll get into, again, Byzantine fault tolerance. 00:06:59.880 --> 00:07:05.220 align:middle line:84% I'm not today getting into the verification or validation 00:07:05.220 --> 00:07:07.860 align:middle line:90% algorithm of the blocks. 00:07:07.860 --> 00:07:13.100 align:middle line:84% Just reminding us all that is part of the system. 00:07:13.100 --> 00:07:21.140 align:middle line:84% Now, again, a peer-to-peer electronic cash system. 00:07:21.140 --> 00:07:25.260 align:middle line:84% So it sounds like money, and it sounds like it's peer to peer. 00:07:25.260 --> 00:07:35.260 align:middle line:84% So I discovered this work of Paul Krugman, the babysitting 00:07:35.260 --> 00:07:40.900 align:middle line:84% economy, where there is a peer-to-peer exchange going on. 00:07:40.900 --> 00:07:45.060 align:middle line:84% It's an exchange of certificates. 00:07:45.060 --> 00:07:49.680 align:middle line:84% So the idea is in 1970s, there was a group of staffers, 00:07:49.680 --> 00:07:53.040 align:middle line:84% young couples with congressional connections. 00:07:53.040 --> 00:07:56.240 align:middle line:84% They agreed to babysit for one another, 00:07:56.240 --> 00:07:59.660 align:middle line:90% so no need for a babysitter. 00:07:59.660 --> 00:08:02.480 align:middle line:84% The parents are doing the babysitting, 00:08:02.480 --> 00:08:08.200 align:middle line:84% and the idea is it's mutually beneficial. 00:08:08.200 --> 00:08:10.960 align:middle line:84% The couple with children doesn't mind necessarily 00:08:10.960 --> 00:08:14.760 align:middle line:84% having a few more kids in the house. 00:08:14.760 --> 00:08:23.320 align:middle line:84% And they earn a certificate for the number of hours babysitting 00:08:23.320 --> 00:08:25.440 align:middle line:90% someone else's kids. 00:08:25.440 --> 00:08:29.040 align:middle line:84% And likewise, they're now armed with the certificate. 00:08:29.040 --> 00:08:34.480 align:middle line:84% So they on an evening out find another parent 00:08:34.480 --> 00:08:39.679 align:middle line:84% who would receive those certificates in exchange 00:08:39.679 --> 00:08:44.080 align:middle line:84% for providing the babysitting service. 00:08:44.080 --> 00:08:48.360 align:middle line:84% So it's kind of like money, but they're certificates. 00:08:48.360 --> 00:08:52.340 align:middle line:84% They're certificates earned by providing a service 00:08:52.340 --> 00:08:56.900 align:middle line:84% and then can be used subsequently in exchange. 00:08:56.900 --> 00:09:02.900 align:middle line:84% Well, cutting the story short a bit or a lot 00:09:02.900 --> 00:09:05.860 align:middle line:84% because there's a whole book about it, 00:09:05.860 --> 00:09:08.420 align:middle line:84% for complicated reasons, involving the collection 00:09:08.420 --> 00:09:13.980 align:middle line:84% and use of dues paid in script, these coupons in circulation 00:09:13.980 --> 00:09:16.060 align:middle line:90% became quite low. 00:09:16.060 --> 00:09:20.540 align:middle line:84% Couples were anxious to add to their reserves 00:09:20.540 --> 00:09:23.020 align:middle line:90% of the certificate. 00:09:23.020 --> 00:09:24.780 align:middle line:84% And they're reluctant to run them down 00:09:24.780 --> 00:09:29.140 align:middle line:84% by going out where they would pass them along 00:09:29.140 --> 00:09:31.000 align:middle line:90% of necessity to someone else. 00:09:31.000 --> 00:09:34.100 align:middle line:84% But one couple's decision to go out 00:09:34.100 --> 00:09:36.000 align:middle line:90% was another chance to babysit. 00:09:36.000 --> 00:09:38.760 align:middle line:84% So not going out means others aren't babysitting, 00:09:38.760 --> 00:09:41.500 align:middle line:90% so they're not earning coupons. 00:09:41.500 --> 00:09:45.460 align:middle line:84% And the whole babysitting market collapsed in a recession. 00:09:45.460 --> 00:09:49.680 align:middle line:90% 00:09:49.680 --> 00:09:52.160 align:middle line:84% Eventually, more script were issued 00:09:52.160 --> 00:09:58.680 align:middle line:84% to compensate for this, creating a different problem, inflation. 00:09:58.680 --> 00:10:00.920 align:middle line:84% So Krugman says this story changed 00:10:00.920 --> 00:10:02.900 align:middle line:90% his career, changed his life. 00:10:02.900 --> 00:10:08.440 align:middle line:84% This is when he started thinking deeply about monetary theory 00:10:08.440 --> 00:10:10.620 align:middle line:90% and business cycles and so on. 00:10:10.620 --> 00:10:14.400 align:middle line:90% 00:10:14.400 --> 00:10:18.480 align:middle line:84% I have this here just to begin to be a bit provocative in terms 00:10:18.480 --> 00:10:21.100 align:middle line:84% of thinking about what do we mean by cash, 00:10:21.100 --> 00:10:22.760 align:middle line:90% what do we mean by money. 00:10:22.760 --> 00:10:24.600 align:middle line:84% I mean, clearly these certificates 00:10:24.600 --> 00:10:32.440 align:middle line:84% function as a kind of money but with a purpose, 00:10:32.440 --> 00:10:34.880 align:middle line:90% namely the babysitting platform. 00:10:34.880 --> 00:10:36.800 align:middle line:90% OK. 00:10:36.800 --> 00:10:39.280 align:middle line:84% Central banks did not take kindly 00:10:39.280 --> 00:10:42.240 align:middle line:90% to the invention of Bitcoin. 00:10:42.240 --> 00:10:44.360 align:middle line:90% And I'll just editorialize. 00:10:44.360 --> 00:10:48.320 align:middle line:84% A bit later, Facebook proposed Meta, 00:10:48.320 --> 00:10:51.280 align:middle line:84% and that was like the ultimate threat. 00:10:51.280 --> 00:10:54.020 align:middle line:90% 00:10:54.020 --> 00:10:55.960 align:middle line:90% So you'll see quotes. 00:10:55.960 --> 00:10:59.260 align:middle line:90% This is 2018 from Hyun Shin. 00:10:59.260 --> 00:11:04.260 align:middle line:84% "Cryptocurrencies cannot scale with transactions demand, 00:11:04.260 --> 00:11:08.340 align:middle line:84% are prone to congestion, and greatly fluctuate in value." 00:11:08.340 --> 00:11:10.820 align:middle line:84% Part of that is an allusion to the validation 00:11:10.820 --> 00:11:15.900 align:middle line:84% algorithms we haven't gone into in detail yet. 00:11:15.900 --> 00:11:16.980 align:middle line:90% His quote continues. 00:11:16.980 --> 00:11:20.240 align:middle line:84% "Overall, the decentralized technology of cryptocurrencies, 00:11:20.240 --> 00:11:24.420 align:middle line:84% however sophisticated, is a poor substitute for the solid 00:11:24.420 --> 00:11:26.960 align:middle line:90% institutional backing of money." 00:11:26.960 --> 00:11:30.220 align:middle line:90% 00:11:30.220 --> 00:11:37.460 align:middle line:84% However, and due to Meta and so on, about 80% 00:11:37.460 --> 00:11:39.140 align:middle line:84% of all central banks in the world 00:11:39.140 --> 00:11:42.760 align:middle line:84% are exploring their own digital currency. 00:11:42.760 --> 00:11:47.620 align:middle line:84% It's kind of maybe a preemption strategy 00:11:47.620 --> 00:11:55.560 align:middle line:84% so that they retain a central role in money issue, 00:11:55.560 --> 00:11:59.400 align:middle line:84% although the case for why central banks are doing it 00:11:59.400 --> 00:12:03.200 align:middle line:90% is sometimes less clear. 00:12:03.200 --> 00:12:04.620 align:middle line:90% So that's 2018. 00:12:04.620 --> 00:12:12.400 align:middle line:84% By 2023, Augustin Carstens and the BIS staff 00:12:12.400 --> 00:12:15.600 align:middle line:84% issued this paper called The Future Monetary 00:12:15.600 --> 00:12:18.640 align:middle line:90% System: From Vision to Reality. 00:12:18.640 --> 00:12:23.560 align:middle line:84% So this is essentially a year and a half ago. 00:12:23.560 --> 00:12:25.920 align:middle line:90% The quotes are revealing. 00:12:25.920 --> 00:12:29.440 align:middle line:84% Financial systems have evolved in separate silos 00:12:29.440 --> 00:12:34.480 align:middle line:84% and in a piecemeal way through minor tweaks of existing systems 00:12:34.480 --> 00:12:35.580 align:middle line:90% and processes. 00:12:35.580 --> 00:12:42.660 align:middle line:84% Legacy systems and outdated products abound. 00:12:42.660 --> 00:12:44.580 align:middle line:84% Even today, money and assets reside 00:12:44.580 --> 00:12:46.620 align:middle line:84% at the edge of communication networks 00:12:46.620 --> 00:12:50.180 align:middle line:84% in separate proprietary databases or ledgers. 00:12:50.180 --> 00:12:52.100 align:middle line:84% These databases are patched together 00:12:52.100 --> 00:12:55.900 align:middle line:84% via third-party messaging system so that messages 00:12:55.900 --> 00:12:58.540 align:middle line:84% have to be sent back and forth in a complex web 00:12:58.540 --> 00:13:01.100 align:middle line:84% of bilateral links, all following 00:13:01.100 --> 00:13:05.020 align:middle line:84% a separate path from one corresponding to money 00:13:05.020 --> 00:13:08.140 align:middle line:84% and the other to financial assets. 00:13:08.140 --> 00:13:12.340 align:middle line:84% Cross-border transactions are even worse. 00:13:12.340 --> 00:13:16.740 align:middle line:84% The best way to knit together transactions and operations 00:13:16.740 --> 00:13:18.820 align:middle line:84% among markets and financial services 00:13:18.820 --> 00:13:25.780 align:middle line:84% is to bring them into a shared programmable platform. 00:13:25.780 --> 00:13:30.160 align:middle line:84% This is what the paper labels as a unified ledger. 00:13:30.160 --> 00:13:33.340 align:middle line:90% 00:13:33.340 --> 00:13:39.280 align:middle line:84% But still in the paper, the view remains that central bank money 00:13:39.280 --> 00:13:41.440 align:middle line:90% is somehow special. 00:13:41.440 --> 00:13:45.200 align:middle line:84% They refer to it as the singleness of money. 00:13:45.200 --> 00:13:52.280 align:middle line:84% Well, they mean the singleness of a central bank, fiat money. 00:13:52.280 --> 00:13:58.520 align:middle line:84% And we'll revisit this idea through the lens 00:13:58.520 --> 00:14:00.520 align:middle line:90% of monetary theory. 00:14:00.520 --> 00:14:04.080 align:middle line:84% But let me just say, if there is such thing as a singleness 00:14:04.080 --> 00:14:08.000 align:middle line:84% of money, then what do we do with all the different monies 00:14:08.000 --> 00:14:11.060 align:middle line:84% existing issued by central banks, 00:14:11.060 --> 00:14:14.040 align:middle line:90% each country one at a time? 00:14:14.040 --> 00:14:17.520 align:middle line:84% It's inherently a multiple money world. 00:14:17.520 --> 00:14:22.600 align:middle line:84% Do we need to create some super currency? 00:14:22.600 --> 00:14:26.200 align:middle line:84% And people have proposed it, that the IMF 00:14:26.200 --> 00:14:33.880 align:middle line:84% should issue, for example, a single convertible currency. 00:14:33.880 --> 00:14:38.220 align:middle line:84% However, in theory, you don't need it in the sense 00:14:38.220 --> 00:14:44.100 align:middle line:84% that in Walrasian equilibria, where the objects being traded 00:14:44.100 --> 00:14:48.180 align:middle line:84% are the different countries' fiat monies, 00:14:48.180 --> 00:14:51.180 align:middle line:84% all you need is a unit of account. 00:14:51.180 --> 00:14:57.380 align:middle line:84% You don't need the store of value or medium of exchange role 00:14:57.380 --> 00:15:00.160 align:middle line:84% to set up a cross-border platform. 00:15:00.160 --> 00:15:02.760 align:middle line:84% And the reference here is to this IMF work. 00:15:02.760 --> 00:15:05.940 align:middle line:90% 00:15:05.940 --> 00:15:10.100 align:middle line:84% So lots of important and interesting views 00:15:10.100 --> 00:15:13.740 align:middle line:84% and controversies in this literature. 00:15:13.740 --> 00:15:18.060 align:middle line:84% But the point of this slide mainly is that it's no longer 00:15:18.060 --> 00:15:20.760 align:middle line:90% "Bitcoin is bad." 00:15:20.760 --> 00:15:24.620 align:middle line:84% People are recognizing the platform aspect 00:15:24.620 --> 00:15:27.540 align:middle line:84% of the blockchain and even proposing 00:15:27.540 --> 00:15:31.500 align:middle line:84% that the domestic and international monetary system be 00:15:31.500 --> 00:15:37.560 align:middle line:90% based on this is technology. 00:15:37.560 --> 00:15:38.680 align:middle line:90% OK. 00:15:38.680 --> 00:15:45.200 align:middle line:84% So let's think about the twin cousin here, financial accounts. 00:15:45.200 --> 00:15:50.680 align:middle line:84% Financial accounts are also a sort of database. 00:15:50.680 --> 00:15:55.880 align:middle line:84% In this book, we enumerate how even for households 00:15:55.880 --> 00:15:59.920 align:middle line:84% to think of households as corporate firms using 00:15:59.920 --> 00:16:06.400 align:middle line:84% standard financial account constructions of the income 00:16:06.400 --> 00:16:11.440 align:middle line:84% and cash flow statements and the corporate balance sheet. 00:16:11.440 --> 00:16:13.900 align:middle line:84% So they come in triples, basically. 00:16:13.900 --> 00:16:17.460 align:middle line:84% You've got a cash flow, income statement, and a balance sheet. 00:16:17.460 --> 00:16:20.400 align:middle line:84% Those are the standard financial accounts. 00:16:20.400 --> 00:16:23.080 align:middle line:84% And then you get the ledger interpretation 00:16:23.080 --> 00:16:29.560 align:middle line:84% of money, which is as a liability of the central bank. 00:16:29.560 --> 00:16:32.840 align:middle line:90% So back to the transactions. 00:16:32.840 --> 00:16:36.340 align:middle line:84% In Bitcoin, we're talking about the exchange 00:16:36.340 --> 00:16:41.380 align:middle line:84% of the coins recorded in blocks and validated. 00:16:41.380 --> 00:16:46.580 align:middle line:84% Here we were running a monthly survey, 00:16:46.580 --> 00:16:49.500 align:middle line:84% and we asked the households a lot of questions. 00:16:49.500 --> 00:16:53.380 align:middle line:84% So for example, what is the total amount of cash payments 00:16:53.380 --> 00:16:57.540 align:middle line:84% you received since the last interview for doing this job? 00:16:57.540 --> 00:17:02.080 align:middle line:84% This is pertaining to wage income, 00:17:02.080 --> 00:17:04.140 align:middle line:84% including blah, blah, blah, blah. 00:17:04.140 --> 00:17:05.520 align:middle line:90% So they got the wage. 00:17:05.520 --> 00:17:07.480 align:middle line:90% So that's an increase in cash. 00:17:07.480 --> 00:17:07.980 align:middle line:90% OK. 00:17:07.980 --> 00:17:09.740 align:middle line:90% Where does it enter? 00:17:09.740 --> 00:17:15.220 align:middle line:84% Cash is an asset, so it enters as a change in the balance 00:17:15.220 --> 00:17:17.700 align:middle line:90% sheet. 00:17:17.700 --> 00:17:23.099 align:middle line:84% It's also revenue received for supplying labor. 00:17:23.099 --> 00:17:27.060 align:middle line:84% So that enters on the income statement. 00:17:27.060 --> 00:17:32.000 align:middle line:84% The same transaction also enters on the third account count, 00:17:32.000 --> 00:17:36.060 align:middle line:90% net income from cash inflow. 00:17:36.060 --> 00:17:39.760 align:middle line:90% 00:17:39.760 --> 00:17:42.160 align:middle line:84% Then you have cash used to pay a telephone 00:17:42.160 --> 00:17:48.840 align:middle line:84% bill, which is like outgoing decrease used for consumption. 00:17:48.840 --> 00:17:50.840 align:middle line:90% It's a cash outflow. 00:17:50.840 --> 00:17:54.400 align:middle line:84% You could deposit cash in a production credit group. 00:17:54.400 --> 00:17:57.760 align:middle line:84% That's like changing, getting rid of your cash 00:17:57.760 --> 00:18:00.700 align:middle line:84% and getting an equivalent credit in value. 00:18:00.700 --> 00:18:02.760 align:middle line:90% So you're swapping assets. 00:18:02.760 --> 00:18:05.720 align:middle line:84% No implication for the income statement. 00:18:05.720 --> 00:18:09.200 align:middle line:84% Every single transaction that they report to us 00:18:09.200 --> 00:18:12.920 align:middle line:84% has been systematically put into the financial accounts. 00:18:12.920 --> 00:18:15.880 align:middle line:84% And we have a way, you'll see, of checking 00:18:15.880 --> 00:18:19.000 align:middle line:84% to make sure we're doing it consistently. 00:18:19.000 --> 00:18:21.000 align:middle line:90% And the rest is kind of fun. 00:18:21.000 --> 00:18:25.420 align:middle line:84% What happens when you sell little animals for cash? 00:18:25.420 --> 00:18:28.080 align:middle line:84% What happens when the mature animals die? 00:18:28.080 --> 00:18:33.340 align:middle line:84% You've got capital appreciation and depreciation and so on, 00:18:33.340 --> 00:18:35.340 align:middle line:90% but I think it's enough. 00:18:35.340 --> 00:18:39.280 align:middle line:84% The book goes on for three or four pages of tables, 00:18:39.280 --> 00:18:41.460 align:middle line:90% but you get the idea. 00:18:41.460 --> 00:18:45.020 align:middle line:84% The basic block is the transaction. 00:18:45.020 --> 00:18:47.040 align:middle line:84% Then where to put it-- in this case, 00:18:47.040 --> 00:18:51.120 align:middle line:84% the financial accounts-- as a database are three accounts. 00:18:51.120 --> 00:18:53.580 align:middle line:84% So we're organizing the transactions 00:18:53.580 --> 00:18:56.460 align:middle line:90% into those three accounts. 00:18:56.460 --> 00:18:58.020 align:middle line:90% OK. 00:18:58.020 --> 00:19:04.100 align:middle line:84% Here's a view of the cash flow account. 00:19:04.100 --> 00:19:08.780 align:middle line:84% So what we want is cash flow from production. 00:19:08.780 --> 00:19:10.380 align:middle line:84% Now, it's a little more complicated 00:19:10.380 --> 00:19:11.840 align:middle line:90% than you might think. 00:19:11.840 --> 00:19:14.460 align:middle line:90% 00:19:14.460 --> 00:19:18.460 align:middle line:84% The basic idea is that when you use 00:19:18.460 --> 00:19:21.700 align:middle line:84% cash-- say, currency-- to buy inputs, 00:19:21.700 --> 00:19:23.480 align:middle line:90% that is an outflow of cash. 00:19:23.480 --> 00:19:27.840 align:middle line:84% And when you sell your crop, for example, for cash, 00:19:27.840 --> 00:19:31.290 align:middle line:84% you have an inflow, a cash inflow. 00:19:31.290 --> 00:19:35.110 align:middle line:84% This starts, however, with an income statement, 00:19:35.110 --> 00:19:37.650 align:middle line:90% which is on an accrual basis. 00:19:37.650 --> 00:19:42.130 align:middle line:84% So if you sell something and get an account receivable, 00:19:42.130 --> 00:19:45.630 align:middle line:84% that counts as income, but you didn't get the cash. 00:19:45.630 --> 00:19:49.410 align:middle line:84% You got a promise to pay, that someone would pay you. 00:19:49.410 --> 00:19:52.450 align:middle line:84% So we actually have to subtract that off from income in order 00:19:52.450 --> 00:19:58.250 align:middle line:84% to get the cash implications, otherwise we're overcounting. 00:19:58.250 --> 00:20:03.930 align:middle line:84% And similarly for inventory, and here's consumption. 00:20:03.930 --> 00:20:06.395 align:middle line:84% Consumption of household-produced outputs. 00:20:06.395 --> 00:20:08.270 align:middle line:84% They just grew the crop and ate it, you know. 00:20:08.270 --> 00:20:11.930 align:middle line:84% So there's no implications for cash, but it's still revenue. 00:20:11.930 --> 00:20:13.210 align:middle line:90% All right. 00:20:13.210 --> 00:20:15.810 align:middle line:84% So we get cash flow from production. 00:20:15.810 --> 00:20:19.130 align:middle line:84% Then you can spend cash on consumption goods 00:20:19.130 --> 00:20:21.850 align:middle line:90% or spend it on investment goods. 00:20:21.850 --> 00:20:23.990 align:middle line:84% Cash flow from consumption and investment. 00:20:23.990 --> 00:20:26.370 align:middle line:84% Then we have the financial side, change 00:20:26.370 --> 00:20:30.030 align:middle line:84% in deposits at financial institutions 00:20:30.030 --> 00:20:35.350 align:middle line:84% or other institutions, borrowing and gifts. 00:20:35.350 --> 00:20:38.990 align:middle line:84% So we have cash flow from financing. 00:20:38.990 --> 00:20:40.510 align:middle line:90% Lots of details. 00:20:40.510 --> 00:20:43.710 align:middle line:84% This stuff turned you off very quickly because there's 00:20:43.710 --> 00:20:45.430 align:middle line:90% so much to absorb. 00:20:45.430 --> 00:20:48.630 align:middle line:84% But the main thing is we're classifying cash flows 00:20:48.630 --> 00:20:53.010 align:middle line:84% into cash flows from production, from consumption, investment, 00:20:53.010 --> 00:20:55.230 align:middle line:90% and from financing. 00:20:55.230 --> 00:20:59.350 align:middle line:84% Also, as a check, the change in cash holdings 00:20:59.350 --> 00:21:01.090 align:middle line:84% from the statement of cash flow-- 00:21:01.090 --> 00:21:06.190 align:middle line:84% this statement should also be reflected in the change in cash 00:21:06.190 --> 00:21:09.230 align:middle line:90% on the balance sheet. 00:21:09.230 --> 00:21:12.390 align:middle line:90% So that's the obvious check. 00:21:12.390 --> 00:21:14.190 align:middle line:90% Rarely done. 00:21:14.190 --> 00:21:18.390 align:middle line:84% We did it in order to make sure we hadn't missed transactions, 00:21:18.390 --> 00:21:22.470 align:middle line:84% and it's very much in the spirit of double-entry bookkeeping. 00:21:22.470 --> 00:21:24.830 align:middle line:90% But I'll come back to this-- 00:21:24.830 --> 00:21:27.510 align:middle line:90% here, the lack of discrepancy. 00:21:27.510 --> 00:21:29.810 align:middle line:84% But in other cases, the discrepancies 00:21:29.810 --> 00:21:31.030 align:middle line:90% are going to be glaring. 00:21:31.030 --> 00:21:33.850 align:middle line:90% We'll come back to that later. 00:21:33.850 --> 00:21:37.810 align:middle line:84% So just to remind you, that household 00:21:37.810 --> 00:21:42.230 align:middle line:84% is a wage-earning household in Lopburi, Thailand. 00:21:42.230 --> 00:21:46.230 align:middle line:84% And this is the income statement of that household. 00:21:46.230 --> 00:21:49.690 align:middle line:84% Most of the revenue is coming from labor. 00:21:49.690 --> 00:21:53.130 align:middle line:84% There's a few bits and pieces of other things. 00:21:53.130 --> 00:21:58.610 align:middle line:84% The revenue is offset in part by expenses. 00:21:58.610 --> 00:22:02.250 align:middle line:84% And what's left, if it's positive, 00:22:02.250 --> 00:22:05.110 align:middle line:84% can be either used for consumption or saving. 00:22:05.110 --> 00:22:07.730 align:middle line:90% 00:22:07.730 --> 00:22:11.170 align:middle line:84% The saving part is, of course, the increment 00:22:11.170 --> 00:22:16.130 align:middle line:84% if it's positive in assets on the balance sheet. 00:22:16.130 --> 00:22:20.610 align:middle line:84% And here's a balance sheet, which in this case a lot of it 00:22:20.610 --> 00:22:24.230 align:middle line:90% is cash currency. 00:22:24.230 --> 00:22:28.310 align:middle line:84% US accounting, you say "cash," but we actually mean by that 00:22:28.310 --> 00:22:30.410 align:middle line:84% demand deposits at commercial bank. 00:22:30.410 --> 00:22:34.530 align:middle line:84% Here, I'm just referring to the currency, the paper stuff. 00:22:34.530 --> 00:22:37.390 align:middle line:90% 00:22:37.390 --> 00:22:42.990 align:middle line:84% Total assets on the left is equal to total liabilities 00:22:42.990 --> 00:22:45.910 align:middle line:90% and net worth on the right. 00:22:45.910 --> 00:22:49.130 align:middle line:84% So the balance sheet balances on both sides. 00:22:49.130 --> 00:22:52.090 align:middle line:90% 00:22:52.090 --> 00:22:59.230 align:middle line:84% So blockchain in the broader view as per the outline. 00:22:59.230 --> 00:23:02.190 align:middle line:84% Single-dimensional as in money for payments 00:23:02.190 --> 00:23:07.990 align:middle line:84% versus multi-dimensional as in, for example, contracting. 00:23:07.990 --> 00:23:11.030 align:middle line:84% So we're embracing for sure because that's 00:23:11.030 --> 00:23:14.750 align:middle line:84% the nature of it, the multi-dimensional aspect 00:23:14.750 --> 00:23:17.330 align:middle line:84% of the blockchain and, for that matter, 00:23:17.330 --> 00:23:21.110 align:middle line:90% of the financial accounts. 00:23:21.110 --> 00:23:23.690 align:middle line:84% I guess I featured blockchain here 00:23:23.690 --> 00:23:27.130 align:middle line:84% and not also financial accounts only because there's 00:23:27.130 --> 00:23:29.450 align:middle line:84% so much misunderstanding in my view of what 00:23:29.450 --> 00:23:32.170 align:middle line:90% the blockchain really is. 00:23:32.170 --> 00:23:35.090 align:middle line:90% So it's not just about money. 00:23:35.090 --> 00:23:39.050 align:middle line:84% And you can see this with payment matrices 00:23:39.050 --> 00:23:42.330 align:middle line:90% that capture velocities. 00:23:42.330 --> 00:23:46.870 align:middle line:84% And then we'll cover individual versus common accounts 00:23:46.870 --> 00:23:50.570 align:middle line:84% and finally have a formal definition of a distributed 00:23:50.570 --> 00:23:51.610 align:middle line:90% ledger. 00:23:51.610 --> 00:23:54.330 align:middle line:84% Blockchains are a distributed ledger. 00:23:54.330 --> 00:24:00.450 align:middle line:84% So multi-dimensional is a database of many objects 00:24:00.450 --> 00:24:03.290 align:middle line:90% and can include contracts. 00:24:03.290 --> 00:24:08.170 align:middle line:84% So we can create payment matrices. 00:24:08.170 --> 00:24:13.610 align:middle line:84% Money would be an object that appears frequently in exchange 00:24:13.610 --> 00:24:18.530 align:middle line:84% or, alternatively, is an object that has a high velocity. 00:24:18.530 --> 00:24:25.430 align:middle line:84% It's an object where the amount traded typically divided 00:24:25.430 --> 00:24:29.558 align:middle line:90% by the stock of it is large. 00:24:29.558 --> 00:24:31.850 align:middle line:84% In other words, there's a lot of turnover in the stock. 00:24:31.850 --> 00:24:35.510 align:middle line:84% Most of the stock changes hands from one period to the next. 00:24:35.510 --> 00:24:38.790 align:middle line:90% That's a high-velocity object. 00:24:38.790 --> 00:24:43.270 align:middle line:84% And ipso facto it's almost surely something appearing 00:24:43.270 --> 00:24:45.990 align:middle line:90% frequently in exchange. 00:24:45.990 --> 00:24:49.990 align:middle line:84% That's the monetary theory definition of money. 00:24:49.990 --> 00:24:53.790 align:middle line:84% That's really the definition, if we had to pick one, 00:24:53.790 --> 00:24:57.030 align:middle line:84% that we should be using in our conversations, 00:24:57.030 --> 00:25:01.270 align:middle line:84% otherwise we're going to get confused if not embroiled about 00:25:01.270 --> 00:25:07.790 align:middle line:84% the controversy of whether Bitcoin is a cash, money or not, 00:25:07.790 --> 00:25:13.910 align:middle line:84% and are going to be inextricably wedded to the financial accounts 00:25:13.910 --> 00:25:17.750 align:middle line:84% point of view that one or more objects in the financial 00:25:17.750 --> 00:25:21.690 align:middle line:84% accounts are, quote, "the money." 00:25:21.690 --> 00:25:24.330 align:middle line:84% This definition covers both blockchains 00:25:24.330 --> 00:25:26.410 align:middle line:90% and financial accounts. 00:25:26.410 --> 00:25:32.530 align:middle line:84% And there can be multiple monies, 00:25:32.530 --> 00:25:37.010 align:middle line:84% and they can coexist with things that do not appear frequently 00:25:37.010 --> 00:25:38.730 align:middle line:90% in exchange. 00:25:38.730 --> 00:25:49.090 align:middle line:84% So here's an example of Village India with my co-author, Yung 00:25:49.090 --> 00:25:51.210 align:middle line:90% Jae Lim. 00:25:51.210 --> 00:25:55.530 align:middle line:84% So there's actually quite a lot going on here. 00:25:55.530 --> 00:25:59.210 align:middle line:84% You've got goods or services going out 00:25:59.210 --> 00:26:02.570 align:middle line:84% from a household or goods and services 00:26:02.570 --> 00:26:05.130 align:middle line:90% coming into the household. 00:26:05.130 --> 00:26:10.590 align:middle line:84% So conventionally, you would think money goes out-- 00:26:10.590 --> 00:26:12.830 align:middle line:90% you use money to buy something. 00:26:12.830 --> 00:26:16.250 align:middle line:84% And indeed that is typically the case. 00:26:16.250 --> 00:26:24.450 align:middle line:84% Money is used to buy grain, clothing, livestock, jewelry, 00:26:24.450 --> 00:26:26.070 align:middle line:90% and so on. 00:26:26.070 --> 00:26:29.630 align:middle line:90% Then we have these IOUs. 00:26:29.630 --> 00:26:35.830 align:middle line:84% IOUs are a promise to pay but not the money 00:26:35.830 --> 00:26:39.070 align:middle line:90% like you're going to owe. 00:26:39.070 --> 00:26:41.350 align:middle line:90% You have an account payable. 00:26:41.350 --> 00:26:42.210 align:middle line:90% Grain. 00:26:42.210 --> 00:26:45.450 align:middle line:84% OK, so now we get to something more interesting. 00:26:45.450 --> 00:26:51.030 align:middle line:90% Grain can be used to pay labor. 00:26:51.030 --> 00:26:56.710 align:middle line:84% This, of all these row-column items, is a fairly large number. 00:26:56.710 --> 00:27:02.890 align:middle line:84% So in these villages in India, people work for compensation. 00:27:02.890 --> 00:27:06.350 align:middle line:84% The compensation comes in form of the grain 00:27:06.350 --> 00:27:10.670 align:middle line:84% as a result of growing the crop or inventory that they're 00:27:10.670 --> 00:27:12.750 align:middle line:90% carrying around. 00:27:12.750 --> 00:27:16.910 align:middle line:90% It's a dual monetary economy. 00:27:16.910 --> 00:27:20.130 align:middle line:90% You've got the real good-- 00:27:20.130 --> 00:27:22.410 align:middle line:90% some people say barter-- 00:27:22.410 --> 00:27:26.730 align:middle line:84% coexisting with the more traditional definition of money 00:27:26.730 --> 00:27:28.030 align:middle line:90% as currency. 00:27:28.030 --> 00:27:30.850 align:middle line:90% 00:27:30.850 --> 00:27:33.570 align:middle line:90% A couple of clarifications. 00:27:33.570 --> 00:27:41.490 align:middle line:84% Labor going out and money coming in. 00:27:41.490 --> 00:27:44.970 align:middle line:84% This table is mixing up households 00:27:44.970 --> 00:27:51.570 align:middle line:84% who are purchasing stuff with producers 00:27:51.570 --> 00:27:56.750 align:middle line:84% who are selling stuff, selling labor or farmers selling grain, 00:27:56.750 --> 00:27:57.510 align:middle line:90% for example. 00:27:57.510 --> 00:27:59.803 align:middle line:84% So that's the reverse transaction. 00:27:59.803 --> 00:28:01.970 align:middle line:84% And the other thing-- it's a little confusing here-- 00:28:01.970 --> 00:28:03.710 align:middle line:90% this is an aggregated account. 00:28:03.710 --> 00:28:07.230 align:middle line:84% It's not household A or something in Thailand. 00:28:07.230 --> 00:28:12.290 align:middle line:84% It's aggregating up over all the transactions 00:28:12.290 --> 00:28:15.390 align:middle line:90% we see in a village. 00:28:15.390 --> 00:28:18.710 align:middle line:90% This is the US. 00:28:18.710 --> 00:28:23.890 align:middle line:84% It's created by the Boston Fed in some joint work. 00:28:23.890 --> 00:28:26.470 align:middle line:90% 00:28:26.470 --> 00:28:31.550 align:middle line:84% A long story short, I came home from Thailand and collaborating 00:28:31.550 --> 00:28:34.910 align:middle line:84% across the river with the Boston Fed. 00:28:34.910 --> 00:28:38.670 align:middle line:84% They were already collecting payments data, as you will see. 00:28:38.670 --> 00:28:44.190 align:middle line:84% And so this led to the creation of analog financial accounts 00:28:44.190 --> 00:28:47.270 align:middle line:90% for US households. 00:28:47.270 --> 00:28:50.710 align:middle line:84% So this is the statement of cash flow, 00:28:50.710 --> 00:28:57.550 align:middle line:84% but let me say it's actually a statement of account flows, not 00:28:57.550 --> 00:28:59.830 align:middle line:90% just cash. 00:28:59.830 --> 00:29:03.350 align:middle line:84% It distinguishes items within cash. 00:29:03.350 --> 00:29:11.990 align:middle line:84% So we do have production flows, flows coming from production, 00:29:11.990 --> 00:29:14.890 align:middle line:84% flows coming from consumption and investment, 00:29:14.890 --> 00:29:17.970 align:middle line:90% flows coming from financing. 00:29:17.970 --> 00:29:21.770 align:middle line:84% And we have a check here on the consistency. 00:29:21.770 --> 00:29:27.050 align:middle line:84% So those are the standard items I showed you previously 00:29:27.050 --> 00:29:29.810 align:middle line:90% in the statement of cash flow. 00:29:29.810 --> 00:29:40.530 align:middle line:84% Here, the featured aspect of it is deposit inflows. 00:29:40.530 --> 00:29:47.770 align:middle line:84% So we're going to distinguish under, quote, deposits currency, 00:29:47.770 --> 00:29:53.170 align:middle line:84% demand deposits, debit accounts, foreign currency, other things 00:29:53.170 --> 00:29:54.290 align:middle line:90% OK? 00:29:54.290 --> 00:29:59.970 align:middle line:84% Multiple accounts for a given household, for example. 00:29:59.970 --> 00:30:03.930 align:middle line:90% And you can see from currency-- 00:30:03.930 --> 00:30:10.710 align:middle line:84% so for example, this is off the diagonal from currency 00:30:10.710 --> 00:30:12.170 align:middle line:90% into a demand deposit. 00:30:12.170 --> 00:30:14.810 align:middle line:84% That's a deposit of cash into the bank. 00:30:14.810 --> 00:30:18.870 align:middle line:90% 00:30:18.870 --> 00:30:24.510 align:middle line:84% Deposits from non-financial accounts, from foreign currency. 00:30:24.510 --> 00:30:28.350 align:middle line:84% So again, from and to are always Xed out here 00:30:28.350 --> 00:30:30.090 align:middle line:90% because they're on the diagonal. 00:30:30.090 --> 00:30:35.510 align:middle line:84% Otherwise one payment account is being converted 00:30:35.510 --> 00:30:40.950 align:middle line:90% to another one or vice versa. 00:30:40.950 --> 00:30:46.510 align:middle line:84% And likewise, withdrawals from one account-- 00:30:46.510 --> 00:30:50.330 align:middle line:84% say the currency account-- into demand deposits. 00:30:50.330 --> 00:30:53.650 align:middle line:84% So there you're cashing out your demand deposit. 00:30:53.650 --> 00:30:59.470 align:middle line:90% 00:30:59.470 --> 00:31:02.390 align:middle line:90% So multiple monies. 00:31:02.390 --> 00:31:06.670 align:middle line:90% Multiple account flows. 00:31:06.670 --> 00:31:09.950 align:middle line:84% Not everything is used by a given household. 00:31:09.950 --> 00:31:14.090 align:middle line:84% But on average this is what US households are doing. 00:31:14.090 --> 00:31:18.970 align:middle line:84% You could say that many of them are, in fact, directly 00:31:18.970 --> 00:31:22.570 align:middle line:84% or indirectly a liability of the Federal Reserve. 00:31:22.570 --> 00:31:25.330 align:middle line:84% Households don't have direct access 00:31:25.330 --> 00:31:28.890 align:middle line:90% to reserve accounts at the Fed. 00:31:28.890 --> 00:31:34.810 align:middle line:84% They do have access to demand deposits, which are-- 00:31:34.810 --> 00:31:37.850 align:middle line:84% and the banks in turn who are issuing the demand deposits 00:31:37.850 --> 00:31:39.850 align:middle line:90% have accounts at the Fed. 00:31:39.850 --> 00:31:43.650 align:middle line:84% So there's kind an indirect backing, 00:31:43.650 --> 00:31:45.930 align:middle line:84% but it's not necessary that everything here 00:31:45.930 --> 00:31:49.350 align:middle line:84% be linked to the US Federal Reserve. 00:31:49.350 --> 00:31:53.810 align:middle line:84% You can see this principle would apply for households transacting 00:31:53.810 --> 00:31:56.010 align:middle line:90% in cryptocurrency. 00:31:56.010 --> 00:31:57.230 align:middle line:90% STUDENT: A quick question. 00:31:57.230 --> 00:32:00.850 align:middle line:84% The flow errors here seem massive between the account 00:32:00.850 --> 00:32:02.450 align:middle line:90% flows and the balance sheets. 00:32:02.450 --> 00:32:04.890 align:middle line:84% That has to do with the difference in timing, 00:32:04.890 --> 00:32:08.418 align:middle line:84% that balance sheets might only be available annually in the-- 00:32:08.418 --> 00:32:10.210 align:middle line:84% ROBERT TOWNSEND: Lots of different factors. 00:32:10.210 --> 00:32:15.110 align:middle line:84% I'm glad you reminded me to emphasize this. 00:32:15.110 --> 00:32:17.710 align:middle line:84% So there was nothing automatic about creating 00:32:17.710 --> 00:32:19.010 align:middle line:90% consistent accounts. 00:32:19.010 --> 00:32:22.450 align:middle line:84% You have a database created for different purposes. 00:32:22.450 --> 00:32:26.350 align:middle line:84% This is coming from the Boston Fed's survey 00:32:26.350 --> 00:32:29.070 align:middle line:90% of consumer payments. 00:32:29.070 --> 00:32:33.270 align:middle line:84% And it wasn't created with the idea 00:32:33.270 --> 00:32:36.610 align:middle line:84% of using it for consistent financial accounts. 00:32:36.610 --> 00:32:38.970 align:middle line:90% So that shows up here. 00:32:38.970 --> 00:32:45.470 align:middle line:84% So something's big, typically missing. 00:32:45.470 --> 00:32:47.650 align:middle line:84% I have even more egregious examples. 00:32:47.650 --> 00:32:50.310 align:middle line:90% 00:32:50.310 --> 00:32:52.370 align:middle line:84% Individual versus common accounts. 00:32:52.370 --> 00:32:55.070 align:middle line:84% So this is another thing to think about when you think 00:32:55.070 --> 00:32:57.230 align:middle line:90% about blockchain and so on. 00:32:57.230 --> 00:33:00.630 align:middle line:84% Call it consensus financial accounts. 00:33:00.630 --> 00:33:04.650 align:middle line:84% Back to the Thai villages, where we, the enumerators, 00:33:04.650 --> 00:33:08.450 align:middle line:84% were recording cash transaction and other transactions 00:33:08.450 --> 00:33:10.210 align:middle line:90% in the previous month. 00:33:10.210 --> 00:33:13.770 align:middle line:90% Save of i with j. 00:33:13.770 --> 00:33:15.310 align:middle line:90% This is similar to Bitcoin. 00:33:15.310 --> 00:33:16.970 align:middle line:84% There's an initial state in Bitcoin, 00:33:16.970 --> 00:33:20.770 align:middle line:84% who owns the coins, modified by a transaction that 00:33:20.770 --> 00:33:23.210 align:middle line:90% alters the state. 00:33:23.210 --> 00:33:26.290 align:middle line:90% Paper currency is like that. 00:33:26.290 --> 00:33:32.630 align:middle line:84% Currency is held as a balance on the part of the household, 00:33:32.630 --> 00:33:36.890 align:middle line:84% although it is not public, obviously. 00:33:36.890 --> 00:33:39.950 align:middle line:84% It's private to the individual unless they show you. 00:33:39.950 --> 00:33:42.730 align:middle line:90% 00:33:42.730 --> 00:33:48.530 align:middle line:84% And currency is, of course, an actual portable physical token. 00:33:48.530 --> 00:33:50.990 align:middle line:90% It's not an electronic entry. 00:33:50.990 --> 00:33:53.650 align:middle line:90% 00:33:53.650 --> 00:33:55.530 align:middle line:84% I kind of wish Bitcoin had not said 00:33:55.530 --> 00:34:02.250 align:middle line:84% it was a coin, because it's clinging to this representation 00:34:02.250 --> 00:34:06.630 align:middle line:84% that we've seen so often when we talk about monies. 00:34:06.630 --> 00:34:09.210 align:middle line:84% Bitcoin isn't really a coin at all. 00:34:09.210 --> 00:34:11.389 align:middle line:90% It's just code. 00:34:11.389 --> 00:34:14.770 align:middle line:84% It's an organized database with transactions. 00:34:14.770 --> 00:34:17.350 align:middle line:90% 00:34:17.350 --> 00:34:22.210 align:middle line:84% But anyway, currency is clearly literally a coin 00:34:22.210 --> 00:34:24.590 align:middle line:90% if it's not paper. 00:34:24.590 --> 00:34:28.830 align:middle line:90% And here it's both. 00:34:28.830 --> 00:34:33.350 align:middle line:84% And you could imagine that these transactions of individual 00:34:33.350 --> 00:34:36.590 align:middle line:84% households-- say if household i was household j-- 00:34:36.590 --> 00:34:39.550 align:middle line:84% are the transaction that ought to be 00:34:39.550 --> 00:34:43.909 align:middle line:84% recorded at the individual level but also could have been, 00:34:43.909 --> 00:34:46.670 align:middle line:84% if it had been like Bitcoin database, 00:34:46.670 --> 00:34:50.030 align:middle line:84% recorded on a common ledger to begin with. 00:34:50.030 --> 00:34:56.870 align:middle line:84% So with a UROP with his help, Chad, 00:34:56.870 --> 00:34:59.530 align:middle line:84% looked at the Townsend Thai data. 00:34:59.530 --> 00:35:07.170 align:middle line:84% And we started to create, well, a common distributed ledger, 00:35:07.170 --> 00:35:10.570 align:middle line:84% but the data weren't collected with that purpose in mind. 00:35:10.570 --> 00:35:13.050 align:middle line:90% So this is like hindsight. 00:35:13.050 --> 00:35:16.250 align:middle line:84% So we have the observed transactions. 00:35:16.250 --> 00:35:18.650 align:middle line:90% Create a ledger. 00:35:18.650 --> 00:35:21.970 align:middle line:84% It could have been a common ledger. 00:35:21.970 --> 00:35:24.370 align:middle line:84% That would have consisted of the enumerators collecting 00:35:24.370 --> 00:35:28.730 align:middle line:84% the data each month, and then pooling the record transactions, 00:35:28.730 --> 00:35:33.310 align:middle line:84% and validating or deciding which transactions are valid, 00:35:33.310 --> 00:35:35.430 align:middle line:84% and then putting it on the common database, 00:35:35.430 --> 00:35:39.050 align:middle line:84% which would then be redistributed to the households. 00:35:39.050 --> 00:35:41.170 align:middle line:84% Of course, we weren't doing that. 00:35:41.170 --> 00:35:45.730 align:middle line:84% In retrospect, we're just organizing individual accounts 00:35:45.730 --> 00:35:50.290 align:middle line:84% and not sharing it publicly or with other households. 00:35:50.290 --> 00:35:52.730 align:middle line:84% But we could have done it this way in principle, 00:35:52.730 --> 00:35:55.890 align:middle line:84% and we would have a common integrated database, 00:35:55.890 --> 00:35:58.650 align:middle line:90% a distributed ledger. 00:35:58.650 --> 00:36:04.230 align:middle line:84% Remember Carsten's quote about the fragmentation and so on. 00:36:04.230 --> 00:36:06.170 align:middle line:84% The alternative here is to have, quote, 00:36:06.170 --> 00:36:08.790 align:middle line:90% centralized it to begin with. 00:36:08.790 --> 00:36:15.510 align:middle line:84% Well, now we come to both a disappointment about the Thai 00:36:15.510 --> 00:36:18.790 align:middle line:84% database but also an important lesson 00:36:18.790 --> 00:36:22.190 align:middle line:84% in terms of the functionality of blockchains. 00:36:22.190 --> 00:36:26.810 align:middle line:84% Namely, when household i reports that it bought something, 00:36:26.810 --> 00:36:28.730 align:middle line:90% we know who they bought it from. 00:36:28.730 --> 00:36:31.510 align:middle line:90% We have recorded i and j. 00:36:31.510 --> 00:36:35.790 align:middle line:84% So we go and look at household j to see if they sold it. 00:36:35.790 --> 00:36:42.030 align:middle line:84% And unfortunately, not infrequently, the transaction 00:36:42.030 --> 00:36:45.670 align:middle line:84% appears only on one person's account and not both. 00:36:45.670 --> 00:36:48.710 align:middle line:90% So those are discrepancies. 00:36:48.710 --> 00:36:54.110 align:middle line:84% So one of the advantages of having a common but distributed 00:36:54.110 --> 00:37:03.410 align:middle line:84% ledger to begin with is to reconcile discrepancies 00:37:03.410 --> 00:37:06.570 align:middle line:90% as quickly as possible. 00:37:06.570 --> 00:37:11.090 align:middle line:84% US financial markets do not do this either. 00:37:11.090 --> 00:37:13.530 align:middle line:90% There's an end-of-day process. 00:37:13.530 --> 00:37:18.070 align:middle line:84% State Street over here will take you 00:37:18.070 --> 00:37:22.210 align:middle line:84% to last couple of hours of a working day 00:37:22.210 --> 00:37:24.730 align:middle line:84% to work with all the correspondent 00:37:24.730 --> 00:37:27.930 align:middle line:84% agents to both parties to a transaction 00:37:27.930 --> 00:37:34.370 align:middle line:84% to reconcile transactions that one party thought it had made 00:37:34.370 --> 00:37:36.190 align:middle line:84% and the other denied that it did, 00:37:36.190 --> 00:37:37.650 align:middle line:84% or the amounts could be different, 00:37:37.650 --> 00:37:40.330 align:middle line:84% or the security could have been different. 00:37:40.330 --> 00:37:44.090 align:middle line:84% It's a very long, cumbersome process. 00:37:44.090 --> 00:37:46.730 align:middle line:84% Not to mention that the prices have 00:37:46.730 --> 00:37:50.430 align:middle line:84% changed since the transaction was recorded. 00:37:50.430 --> 00:37:57.910 align:middle line:84% So there are disincentives that grow with the time lag 00:37:57.910 --> 00:38:00.550 align:middle line:84% in between the actual transactions 00:38:00.550 --> 00:38:02.050 align:middle line:90% and the reconciliation. 00:38:02.050 --> 00:38:09.070 align:middle line:84% So I'm extolling one of the virtues of the distributed 00:38:09.070 --> 00:38:09.570 align:middle line:90% ledger. 00:38:09.570 --> 00:38:11.710 align:middle line:90% Here's a formal definition. 00:38:11.710 --> 00:38:15.990 align:middle line:84% Distributed ledger is a technological infrastructure 00:38:15.990 --> 00:38:20.710 align:middle line:84% and protocols that allow simultaneous access, validation, 00:38:20.710 --> 00:38:25.730 align:middle line:84% and record-keeping across a networked database. 00:38:25.730 --> 00:38:28.990 align:middle line:90% 00:38:28.990 --> 00:38:34.630 align:middle line:84% DLT is a technology blockchains are created from. 00:38:34.630 --> 00:38:39.450 align:middle line:84% The infrastructure that allows users to view changes 00:38:39.450 --> 00:38:43.470 align:middle line:84% and who made them up to the identity of the nodes 00:38:43.470 --> 00:38:48.550 align:middle line:84% reduces the need to audit data, ensures reliability, et cetera. 00:38:48.550 --> 00:38:50.210 align:middle line:90% It's not new really. 00:38:50.210 --> 00:38:52.470 align:middle line:84% Businesses and governments have been 00:38:52.470 --> 00:38:55.470 align:middle line:84% using this concept for several decades 00:38:55.470 --> 00:38:59.770 align:middle line:84% in the form of having multiple computers in different locations 00:38:59.770 --> 00:39:04.450 align:middle line:84% used to solve problems like decentralizing 00:39:04.450 --> 00:39:08.130 align:middle line:84% the problem to multiple computers, parallel computing, 00:39:08.130 --> 00:39:14.290 align:middle line:84% and then consolidating the answers. 00:39:14.290 --> 00:39:18.070 align:middle line:84% Advances in data science, computing, software, hardware, 00:39:18.070 --> 00:39:21.410 align:middle line:84% et cetera have made ledgers much more capable 00:39:21.410 --> 00:39:22.750 align:middle line:90% than they used to be. 00:39:22.750 --> 00:39:25.970 align:middle line:90% 00:39:25.970 --> 00:39:31.150 align:middle line:84% The rest is a repeat of things we've talked before. 00:39:31.150 --> 00:39:33.130 align:middle line:84% This thing at the bottom-- every blockchain 00:39:33.130 --> 00:39:36.410 align:middle line:84% is a distributed ledger, but not necessarily 00:39:36.410 --> 00:39:39.770 align:middle line:84% true that a distributed ledger is a blockchain. 00:39:39.770 --> 00:39:44.530 align:middle line:84% Here the blockchain-- part of the definition of a blockchain 00:39:44.530 --> 00:39:47.090 align:middle line:84% had to do with the validation algorithm used 00:39:47.090 --> 00:39:48.850 align:middle line:90% across the different blocks. 00:39:48.850 --> 00:39:52.130 align:middle line:84% And you could have a distributed ledger as a database 00:39:52.130 --> 00:39:55.430 align:middle line:84% that is not being validated in that way. 00:39:55.430 --> 00:39:57.450 align:middle line:84% So I think that's the distinction. 00:39:57.450 --> 00:39:59.990 align:middle line:90% 00:39:59.990 --> 00:40:04.990 align:middle line:84% Tensions between individual versus community perspectives. 00:40:04.990 --> 00:40:07.190 align:middle line:84% We'll talk about central bank accounts 00:40:07.190 --> 00:40:10.770 align:middle line:84% and real-time gross settlement about the father, so to speak, 00:40:10.770 --> 00:40:15.050 align:middle line:84% of double-entry bookkeeping and community currencies. 00:40:15.050 --> 00:40:19.030 align:middle line:84% So central bank real-time gross settlement-- 00:40:19.030 --> 00:40:24.230 align:middle line:84% quite seminal article is Bech and Garrett 00:40:24.230 --> 00:40:32.150 align:middle line:84% in 2003, which is The Intraday Liquidity Management Game. 00:40:32.150 --> 00:40:39.870 align:middle line:84% Oddly, in some respects, banks know what their obligations 00:40:39.870 --> 00:40:44.110 align:middle line:84% are because their customers tell them 00:40:44.110 --> 00:40:46.150 align:middle line:90% they want to make a payment. 00:40:46.150 --> 00:40:49.550 align:middle line:84% But they get to choose the timing of when 00:40:49.550 --> 00:40:51.910 align:middle line:84% to report that payment obligation 00:40:51.910 --> 00:40:55.890 align:middle line:84% to the real-time gross settlement system. 00:40:55.890 --> 00:41:02.090 align:middle line:84% So the data are consistent with a delay in reporting. 00:41:02.090 --> 00:41:05.410 align:middle line:84% And the idea is that if you report early, 00:41:05.410 --> 00:41:09.010 align:middle line:84% that's cash or reserves out the door, 00:41:09.010 --> 00:41:13.550 align:middle line:84% and you're trying to conserve liquidity in your account. 00:41:13.550 --> 00:41:15.950 align:middle line:84% So you prefer that someone pay you first. 00:41:15.950 --> 00:41:21.130 align:middle line:84% And that results in this Nash equilibrium, 00:41:21.130 --> 00:41:23.790 align:middle line:84% where the payments tend to get bunched at the end of the day. 00:41:23.790 --> 00:41:26.850 align:middle line:84% And that could be potentially inefficient. 00:41:26.850 --> 00:41:28.230 align:middle line:90% They're not the only one. 00:41:28.230 --> 00:41:29.730 align:middle line:90% The Goldstein here-- 00:41:29.730 --> 00:41:32.810 align:middle line:84% I met staff at the Swiss National Bank 00:41:32.810 --> 00:41:37.210 align:middle line:84% who referred me to this classic paper 00:41:37.210 --> 00:41:40.730 align:middle line:84% about payments and financial fragility. 00:41:40.730 --> 00:41:43.250 align:middle line:90% Payments are inherently fragile. 00:41:43.250 --> 00:41:45.410 align:middle line:90% Why? 00:41:45.410 --> 00:41:47.450 align:middle line:84% A theory of payments involves a conflict 00:41:47.450 --> 00:41:49.930 align:middle line:84% between the role of the medium of exchange 00:41:49.930 --> 00:41:52.790 align:middle line:90% and the store of value. 00:41:52.790 --> 00:41:56.230 align:middle line:84% Payments must involve a reciprocal transfer 00:41:56.230 --> 00:41:58.790 align:middle line:84% of the scarce reserve good, which 00:41:58.790 --> 00:42:01.030 align:middle line:84% holds value for other purposes, namely 00:42:01.030 --> 00:42:04.390 align:middle line:84% not just for transactions but as a store of value. 00:42:04.390 --> 00:42:09.270 align:middle line:84% The theory demonstrates payments are made only when 00:42:09.270 --> 00:42:12.870 align:middle line:90% reserves are abundant enough. 00:42:12.870 --> 00:42:16.390 align:middle line:84% Otherwise, we get history-dependent equilibriums. 00:42:16.390 --> 00:42:20.350 align:middle line:84% So Rafa, this is building on global games. 00:42:20.350 --> 00:42:23.950 align:middle line:90% 00:42:23.950 --> 00:42:28.670 align:middle line:84% This is also the conflict we saw with Krugman's babysitting 00:42:28.670 --> 00:42:30.550 align:middle line:90% exposé-- 00:42:30.550 --> 00:42:35.830 align:middle line:84% holding on to the coupons as a store of value 00:42:35.830 --> 00:42:40.110 align:middle line:84% rather than using them in transactions as intended. 00:42:40.110 --> 00:42:41.930 align:middle line:90% It's an inherent conflict. 00:42:41.930 --> 00:42:44.510 align:middle line:90% It's a universal conflict. 00:42:44.510 --> 00:42:46.830 align:middle line:84% People say, what's the role of money? 00:42:46.830 --> 00:42:49.570 align:middle line:84% Store value, medium of exchange, as this 00:42:49.570 --> 00:42:53.890 align:middle line:84% were obvious and innocuous that you are potentially 00:42:53.890 --> 00:42:56.970 align:middle line:90% serving multiple masters. 00:42:56.970 --> 00:42:58.230 align:middle line:90% There are consequences. 00:42:58.230 --> 00:42:59.050 align:middle line:90% OK. 00:42:59.050 --> 00:43:07.350 align:middle line:84% Then we come to Luca Pacioli, who, as I said, 00:43:07.350 --> 00:43:11.550 align:middle line:84% is credited with double-entry bookkeeping. 00:43:11.550 --> 00:43:13.290 align:middle line:84% And actually, I think he was just 00:43:13.290 --> 00:43:17.330 align:middle line:84% consolidating previous work and studies on this 00:43:17.330 --> 00:43:18.890 align:middle line:90% and wrote it all up. 00:43:18.890 --> 00:43:23.930 align:middle line:84% So it's certainly the first published manuscript. 00:43:23.930 --> 00:43:26.650 align:middle line:84% But let me start with something odd, the problem 00:43:26.650 --> 00:43:28.970 align:middle line:90% of negative numbers. 00:43:28.970 --> 00:43:32.450 align:middle line:84% So even in the Renaissance, negative numbers 00:43:32.450 --> 00:43:36.370 align:middle line:84% were thought of as absurd, although 00:43:36.370 --> 00:43:43.690 align:middle line:84% Chinese and Arab scholars understood the concept. 00:43:43.690 --> 00:43:45.550 align:middle line:90% Why did it seem absurd? 00:43:45.550 --> 00:43:47.010 align:middle line:84% Because what does it mean to hold 00:43:47.010 --> 00:43:50.070 align:middle line:90% a negative amount of something? 00:43:50.070 --> 00:43:52.070 align:middle line:90% It's a physical object. 00:43:52.070 --> 00:43:56.910 align:middle line:84% It's either there or it isn't, but it can't be negative. 00:43:56.910 --> 00:43:58.970 align:middle line:90% They were clinging to this idea. 00:43:58.970 --> 00:44:04.990 align:middle line:84% Of course, we know from algebra the basic functionality 00:44:04.990 --> 00:44:07.270 align:middle line:84% of addition, multiplication, et cetera. 00:44:07.270 --> 00:44:08.410 align:middle line:90% You have inverses. 00:44:08.410 --> 00:44:10.730 align:middle line:84% There's an obvious negative number. 00:44:10.730 --> 00:44:13.670 align:middle line:90% 00:44:13.670 --> 00:44:17.830 align:middle line:84% That's a necessary part of algebra. 00:44:17.830 --> 00:44:24.450 align:middle line:84% So here money, which, from the financial account point of view, 00:44:24.450 --> 00:44:27.430 align:middle line:84% you kind of naturally think of as an asset-- 00:44:27.430 --> 00:44:29.270 align:middle line:90% I mean, it's embedded. 00:44:29.270 --> 00:44:33.510 align:middle line:84% It's confusing if it's something else. 00:44:33.510 --> 00:44:40.950 align:middle line:84% But Luca referred to it as a liability, not as an asset. 00:44:40.950 --> 00:44:44.870 align:middle line:84% He has in mind that the reference should not be to the, 00:44:44.870 --> 00:44:48.650 align:middle line:84% quote, owner of it but rather the obligation 00:44:48.650 --> 00:44:54.170 align:middle line:84% to the other party that you owe something to the community. 00:44:54.170 --> 00:44:57.970 align:middle line:84% So to try to substantiate this claim-- 00:44:57.970 --> 00:45:01.030 align:middle line:90% and this is from the book. 00:45:01.030 --> 00:45:06.130 align:middle line:84% But Tomaž Fleischman who I've told you about last time, 00:45:06.130 --> 00:45:09.730 align:middle line:84% is the one who had explored this in detail. 00:45:09.730 --> 00:45:11.950 align:middle line:84% Two expressions are used in the journal, 00:45:11.950 --> 00:45:17.370 align:middle line:84% especially in Venice, one called per and the other A. 00:45:17.370 --> 00:45:20.090 align:middle line:90% And what is understood by them? 00:45:20.090 --> 00:45:25.450 align:middle line:84% These expressions, the per and the other-- 00:45:25.450 --> 00:45:32.130 align:middle line:84% per indicates a debtor, and A is a creditor. 00:45:32.130 --> 00:45:36.850 align:middle line:84% And these cash items, I hesitate to say, 00:45:36.850 --> 00:45:42.130 align:middle line:84% either appearing in transactions or in the balance sheet 00:45:42.130 --> 00:45:43.770 align:middle line:90% are denoted with per. 00:45:43.770 --> 00:45:47.970 align:middle line:90% Per means you're a debtor. 00:45:47.970 --> 00:45:51.030 align:middle line:90% 00:45:51.030 --> 00:45:53.730 align:middle line:84% It's clear that this is what he had in mind. 00:45:53.730 --> 00:45:58.670 align:middle line:90% 00:45:58.670 --> 00:45:59.870 align:middle line:90% I'll come back to that. 00:45:59.870 --> 00:46:06.670 align:middle line:84% Today is probably everything you ever wanted to know or more 00:46:06.670 --> 00:46:09.590 align:middle line:90% about databases. 00:46:09.590 --> 00:46:13.870 align:middle line:84% I put in a reminder that actually-- 00:46:13.870 --> 00:46:19.230 align:middle line:84% and starting the next lecture, we will look at ultimate goals. 00:46:19.230 --> 00:46:22.910 align:middle line:84% So here I just gave you some citations 00:46:22.910 --> 00:46:28.670 align:middle line:84% to database being used for what and evaluating the purpose 00:46:28.670 --> 00:46:30.590 align:middle line:90% and the effectiveness. 00:46:30.590 --> 00:46:34.230 align:middle line:90% We're not doing that today. 00:46:34.230 --> 00:46:40.630 align:middle line:84% Well, then we come to community inclusion currencies. 00:46:40.630 --> 00:46:45.050 align:middle line:84% So these CICs are a medium of exchange 00:46:45.050 --> 00:46:50.670 align:middle line:84% that can be used only within a certain network or community, 00:46:50.670 --> 00:46:54.890 align:middle line:84% like the peer-to-peer stuff, which was registered 00:46:54.890 --> 00:46:56.730 align:middle line:90% babysitting services. 00:46:56.730 --> 00:46:59.110 align:middle line:90% It's a mutual credit network. 00:46:59.110 --> 00:47:03.450 align:middle line:84% That's another term for this community inclusion currency, 00:47:03.450 --> 00:47:05.410 align:middle line:84% because the participants effectively 00:47:05.410 --> 00:47:07.430 align:middle line:84% give each other interest-free credit. 00:47:07.430 --> 00:47:08.710 align:middle line:90% Now, what does that mean? 00:47:08.710 --> 00:47:12.010 align:middle line:90% 00:47:12.010 --> 00:47:16.530 align:middle line:84% For example, you sell your goods. 00:47:16.530 --> 00:47:21.370 align:middle line:84% And at least not right away, you don't get something 00:47:21.370 --> 00:47:26.050 align:middle line:84% that you directly value, like goods in return. 00:47:26.050 --> 00:47:29.170 align:middle line:84% You sell your goods, and you get one of these community inclusion 00:47:29.170 --> 00:47:32.730 align:middle line:84% currencies to denote the value of goods sold. 00:47:32.730 --> 00:47:36.850 align:middle line:84% Then you take that and buy goods with it. 00:47:36.850 --> 00:47:40.570 align:middle line:84% So it's sort of a credit, but that's 00:47:40.570 --> 00:47:42.450 align:middle line:90% the accounting point of view. 00:47:42.450 --> 00:47:45.070 align:middle line:90% 00:47:45.070 --> 00:47:48.310 align:middle line:84% The Pacioli point of view is you sold goods, 00:47:48.310 --> 00:47:49.890 align:middle line:90% and you got community currency. 00:47:49.890 --> 00:47:53.110 align:middle line:84% And now you have an obligation to use it to buy 00:47:53.110 --> 00:47:55.910 align:middle line:90% the production of others. 00:47:55.910 --> 00:48:03.230 align:middle line:84% That's the purpose, and others have to accept it. 00:48:03.230 --> 00:48:07.870 align:middle line:84% In fact, there are certain incentive structures 00:48:07.870 --> 00:48:13.110 align:middle line:84% loaded in that encourage holders to spend the currency instead 00:48:13.110 --> 00:48:14.870 align:middle line:90% of hoarding it. 00:48:14.870 --> 00:48:16.610 align:middle line:90% Back to that tension again. 00:48:16.610 --> 00:48:19.710 align:middle line:90% 00:48:19.710 --> 00:48:24.910 align:middle line:84% And these are used in informal settlements in slum areas 00:48:24.910 --> 00:48:28.790 align:middle line:84% in Nairobi, for example, also South Africa. 00:48:28.790 --> 00:48:31.990 align:middle line:90% The best known is Sarafu. 00:48:31.990 --> 00:48:34.030 align:middle line:84% So the vouchers are promissory notes 00:48:34.030 --> 00:48:38.550 align:middle line:84% as a promise against future production. 00:48:38.550 --> 00:48:39.850 align:middle line:90% What was the idea? 00:48:39.850 --> 00:48:45.130 align:middle line:84% Self-governing community of microbusinesses 00:48:45.130 --> 00:48:48.170 align:middle line:84% committing their future production into vouchers 00:48:48.170 --> 00:48:52.250 align:middle line:84% to accept trade with one another so that the value would not 00:48:52.250 --> 00:48:56.290 align:middle line:90% leave the community. 00:48:56.290 --> 00:48:59.530 align:middle line:90% And there's this notion-- 00:48:59.530 --> 00:49:04.930 align:middle line:84% if everyone goes shop at Walmart or something, that's leakage. 00:49:04.930 --> 00:49:08.930 align:middle line:84% That's value that doesn't recirculate in the community. 00:49:08.930 --> 00:49:12.830 align:middle line:84% These guys are doing very interesting work. 00:49:12.830 --> 00:49:18.050 align:middle line:84% It's not necessarily the way we would approach. 00:49:18.050 --> 00:49:20.450 align:middle line:84% As economists, there's stuff to prove here 00:49:20.450 --> 00:49:23.890 align:middle line:84% that this stuff is really not only fulfilling 00:49:23.890 --> 00:49:26.690 align:middle line:84% the purpose for which they intended it but is actually 00:49:26.690 --> 00:49:28.090 align:middle line:90% efficient. 00:49:28.090 --> 00:49:30.670 align:middle line:90% And we could come back to that. 00:49:30.670 --> 00:49:37.850 align:middle line:84% But the bottom line here is it's running on a blockchain. 00:49:37.850 --> 00:49:43.230 align:middle line:84% Everything that I described here is being run on Ethereum 00:49:43.230 --> 00:49:44.890 align:middle line:90% in a slum in Nairobi. 00:49:44.890 --> 00:49:48.310 align:middle line:90% 00:49:48.310 --> 00:49:52.070 align:middle line:84% Blockchains was the topic of the day. 00:49:52.070 --> 00:49:52.730 align:middle line:90% OK. 00:49:52.730 --> 00:49:58.830 align:middle line:90% 00:49:58.830 --> 00:50:05.830 align:middle line:84% At the risk of extolling the virtues, I promised you balance. 00:50:05.830 --> 00:50:12.910 align:middle line:84% So let me provide some caution in terms of the limitation 00:50:12.910 --> 00:50:13.850 align:middle line:90% of blockchains. 00:50:13.850 --> 00:50:15.830 align:middle line:84% The first has to do with something 00:50:15.830 --> 00:50:19.670 align:middle line:84% called the CAP theorem, which is a theorem in computer science. 00:50:19.670 --> 00:50:24.190 align:middle line:84% And then there are more liberal interpretations 00:50:24.190 --> 00:50:30.490 align:middle line:84% of that to classify current payment systems. 00:50:30.490 --> 00:50:33.310 align:middle line:90% 00:50:33.310 --> 00:50:38.970 align:middle line:84% The CAP theorem, CAP, has to do with Consistency, Availability, 00:50:38.970 --> 00:50:42.370 align:middle line:90% and Partition tolerance. 00:50:42.370 --> 00:50:45.850 align:middle line:84% So, again, think blockchain, databases. 00:50:45.850 --> 00:50:50.890 align:middle line:84% Common database ideally consistent in the sense 00:50:50.890 --> 00:50:56.770 align:middle line:84% that anyone who tries to read the data 00:50:56.770 --> 00:51:00.730 align:middle line:90% will get the same answer. 00:51:00.730 --> 00:51:06.710 align:middle line:84% Availability means that at any time they can request data. 00:51:06.710 --> 00:51:09.610 align:middle line:90% 00:51:09.610 --> 00:51:14.290 align:middle line:84% And yet partition tolerance says that the system continues 00:51:14.290 --> 00:51:20.370 align:middle line:84% to operate even if you get a partition of potentially 00:51:20.370 --> 00:51:21.950 align:middle line:90% inconsistent data. 00:51:21.950 --> 00:51:25.490 align:middle line:90% 00:51:25.490 --> 00:51:28.090 align:middle line:84% By the way, Bitcoin has this, at least 00:51:28.090 --> 00:51:31.430 align:middle line:84% for a while, because the blocks are proposed. 00:51:31.430 --> 00:51:34.150 align:middle line:84% There's multiple conflicting blocks. 00:51:34.150 --> 00:51:37.770 align:middle line:84% They're not immediately consistent with one another. 00:51:37.770 --> 00:51:39.830 align:middle line:84% The whole validation process is to try 00:51:39.830 --> 00:51:42.230 align:middle line:84% to decide at least asymptotically 00:51:42.230 --> 00:51:46.150 align:middle line:90% what's the valid block. 00:51:46.150 --> 00:51:48.010 align:middle line:90% But it's not just about Bitcoin. 00:51:48.010 --> 00:51:52.950 align:middle line:84% It's inherent with distributed ledgers. 00:51:52.950 --> 00:51:55.790 align:middle line:84% So if you're going to allow the system 00:51:55.790 --> 00:51:59.850 align:middle line:84% to keep functioning despite network failures, 00:51:59.850 --> 00:52:03.870 align:middle line:90% you have a hard choice. 00:52:03.870 --> 00:52:09.630 align:middle line:84% You could cancel the operation, which decreases availability 00:52:09.630 --> 00:52:11.210 align:middle line:90% but ensures consistency. 00:52:11.210 --> 00:52:14.850 align:middle line:84% You're only allowing whenever a query is made, 00:52:14.850 --> 00:52:19.590 align:middle line:84% it must be the case that anyone asking will get the same answer. 00:52:19.590 --> 00:52:21.968 align:middle line:84% But then you have to shut down availability. 00:52:21.968 --> 00:52:22.510 align:middle line:90% I don't know. 00:52:22.510 --> 00:52:26.910 align:middle line:90% End of day, end of month. 00:52:26.910 --> 00:52:29.910 align:middle line:84% When are the accounts reconciled? 00:52:29.910 --> 00:52:35.370 align:middle line:84% Or you allow the partition, and you proceed. 00:52:35.370 --> 00:52:37.210 align:middle line:84% So you have availability, but then you 00:52:37.210 --> 00:52:38.750 align:middle line:90% risk this inconsistency. 00:52:38.750 --> 00:52:42.610 align:middle line:84% So this is what [? Breuer ?] is saying is a fundamental choice 00:52:42.610 --> 00:52:43.590 align:middle line:90% to be made. 00:52:43.590 --> 00:52:47.210 align:middle line:90% 00:52:47.210 --> 00:52:49.130 align:middle line:84% So, again, extolling the virtues, 00:52:49.130 --> 00:52:53.370 align:middle line:84% you would think consistency, availability, 00:52:53.370 --> 00:52:56.890 align:middle line:84% and don't worry about partition tolerance. 00:52:56.890 --> 00:53:01.450 align:middle line:84% So now you have to think a little more carefully 00:53:01.450 --> 00:53:04.810 align:middle line:84% about the design and the trade-offs. 00:53:04.810 --> 00:53:09.790 align:middle line:84% And in particular you could use this as a dilemma. 00:53:09.790 --> 00:53:13.370 align:middle line:90% It's really a trilemma. 00:53:13.370 --> 00:53:15.170 align:middle line:90% I said that, right? 00:53:15.170 --> 00:53:18.250 align:middle line:90% You can't have all three. 00:53:18.250 --> 00:53:21.090 align:middle line:90% So that's a dilemma, trilemma. 00:53:21.090 --> 00:53:22.550 align:middle line:90% Which one are you going to drop? 00:53:22.550 --> 00:53:25.610 align:middle line:90% 00:53:25.610 --> 00:53:27.770 align:middle line:90% So here's an interpretation. 00:53:27.770 --> 00:53:37.550 align:middle line:84% We start with two parties to a contract, classic Alice and Bob. 00:53:37.550 --> 00:53:39.330 align:middle line:90% And they don't trust each other. 00:53:39.330 --> 00:53:41.530 align:middle line:84% And they have their own financial accounts, 00:53:41.530 --> 00:53:44.750 align:middle line:84% and they're not necessarily consistent with one another. 00:53:44.750 --> 00:53:50.590 align:middle line:84% That's kind of the starting point over here in the partition 00:53:50.590 --> 00:53:51.450 align:middle line:90% vertices. 00:53:51.450 --> 00:53:54.230 align:middle line:90% 00:53:54.230 --> 00:53:57.070 align:middle line:90% Now, that's awkward. 00:53:57.070 --> 00:54:00.670 align:middle line:84% And so you begin to see in economic history 00:54:00.670 --> 00:54:03.870 align:middle line:90% the creation of intermediaries. 00:54:03.870 --> 00:54:07.470 align:middle line:84% So the intermediary has a representation 00:54:07.470 --> 00:54:12.230 align:middle line:84% of the accounts, at least for some transactions 00:54:12.230 --> 00:54:14.750 align:middle line:90% of the individual participants. 00:54:14.750 --> 00:54:20.070 align:middle line:84% So it could still be inconsistency and partition 00:54:20.070 --> 00:54:23.270 align:middle line:84% tolerance in the sense of multiple accounts. 00:54:23.270 --> 00:54:28.230 align:middle line:84% But if third parties are dealing with the intermediary, 00:54:28.230 --> 00:54:30.550 align:middle line:84% they're getting a consistent answer. 00:54:30.550 --> 00:54:33.570 align:middle line:90% 00:54:33.570 --> 00:54:36.970 align:middle line:84% Yeah, you should be worried if you're thinking 00:54:36.970 --> 00:54:40.550 align:middle line:90% layman or something, but OK. 00:54:40.550 --> 00:54:43.490 align:middle line:90% 00:54:43.490 --> 00:54:44.990 align:middle line:90% So those are intermediaries. 00:54:44.990 --> 00:54:49.730 align:middle line:84% Now the system evolves, and we're on this edge here. 00:54:49.730 --> 00:54:57.650 align:middle line:84% In this case, we've given up on having multiple partitions. 00:54:57.650 --> 00:55:02.730 align:middle line:84% We're on the consistency-availability axis. 00:55:02.730 --> 00:55:10.450 align:middle line:84% And in particular a unique ledger 00:55:10.450 --> 00:55:14.250 align:middle line:84% will always give consistent information 00:55:14.250 --> 00:55:17.730 align:middle line:90% across the different ledgers. 00:55:17.730 --> 00:55:23.910 align:middle line:84% A security depository, the CSDs, are like that. 00:55:23.910 --> 00:55:31.210 align:middle line:84% They're a centralized log of the ownership of financial assets, 00:55:31.210 --> 00:55:35.630 align:middle line:84% typically run by a licensed private party 00:55:35.630 --> 00:55:38.210 align:middle line:84% or even run by a government entity. 00:55:38.210 --> 00:55:41.870 align:middle line:90% 00:55:41.870 --> 00:55:46.430 align:middle line:84% Now, if we start at the top with consistency, 00:55:46.430 --> 00:55:52.870 align:middle line:84% there is a one common, consistent, distributed ledger. 00:55:52.870 --> 00:55:55.790 align:middle line:90% 00:55:55.790 --> 00:55:59.710 align:middle line:90% Common but distributed. 00:55:59.710 --> 00:56:05.510 align:middle line:84% But people have realized with congestion and other things 00:56:05.510 --> 00:56:07.910 align:middle line:84% that sometimes you have to choose. 00:56:07.910 --> 00:56:10.630 align:middle line:84% There's something called sharding, 00:56:10.630 --> 00:56:16.910 align:middle line:84% which basically is splitting the single database into subledgers 00:56:16.910 --> 00:56:20.910 align:middle line:84% in order to facilitate processing, 00:56:20.910 --> 00:56:26.570 align:middle line:84% or the layer 1, layer 2, which, as a reminder, 00:56:26.570 --> 00:56:34.890 align:middle line:84% I showed you last time, which is you have a smart contract 00:56:34.890 --> 00:56:39.130 align:middle line:84% and participants agree that it functions 00:56:39.130 --> 00:56:41.990 align:middle line:90% the way they wanted it to. 00:56:41.990 --> 00:56:45.450 align:middle line:84% So you're implementing a multi-agent smart contract 00:56:45.450 --> 00:56:48.330 align:middle line:84% among the parties, but the only transactions 00:56:48.330 --> 00:56:50.850 align:middle line:84% that are actually executed on the ledger 00:56:50.850 --> 00:56:56.730 align:middle line:84% are the transfer of assets, not running the code. 00:56:56.730 --> 00:56:59.890 align:middle line:84% So that's this layer 2 terminology. 00:56:59.890 --> 00:57:04.850 align:middle line:84% But in the end, it's all about trying 00:57:04.850 --> 00:57:08.990 align:middle line:84% to make something functional that can handle large numbers. 00:57:08.990 --> 00:57:12.130 align:middle line:90% 00:57:12.130 --> 00:57:16.990 align:middle line:84% And finally, we get to so-called compatible ledgers. 00:57:16.990 --> 00:57:21.730 align:middle line:90% 00:57:21.730 --> 00:57:26.160 align:middle line:90% You might say unified ledgers. 00:57:26.160 --> 00:57:28.300 align:middle line:84% So at this point you're like, which is which? 00:57:28.300 --> 00:57:31.120 align:middle line:90% Hmm. 00:57:31.120 --> 00:57:36.820 align:middle line:84% The BIS was referring to unifying multiple ledgers. 00:57:36.820 --> 00:57:40.680 align:middle line:84% That's what they had in mind, as opposed 00:57:40.680 --> 00:57:45.880 align:middle line:84% to having one unique common ledger to begin with. 00:57:45.880 --> 00:57:49.960 align:middle line:84% So that's trying to move from these partitioned 00:57:49.960 --> 00:57:53.400 align:middle line:84% and potentially inconsistent ledgers to something 00:57:53.400 --> 00:57:54.820 align:middle line:90% more centralized. 00:57:54.820 --> 00:57:59.700 align:middle line:84% That's what Carsten was hoping for. 00:57:59.700 --> 00:58:02.600 align:middle line:90% 00:58:02.600 --> 00:58:09.960 align:middle line:84% But we do need to worry about whether or not 00:58:09.960 --> 00:58:13.240 align:middle line:84% you lose the functionality of smart contracts 00:58:13.240 --> 00:58:14.860 align:middle line:90% if they're not consistent. 00:58:14.860 --> 00:58:18.080 align:middle line:84% By the way, I should have said all along up 00:58:18.080 --> 00:58:22.280 align:middle line:84% here under consistency, it wasn't just data consistency. 00:58:22.280 --> 00:58:26.980 align:middle line:84% It's contract completeness because on Ethereum we're 00:58:26.980 --> 00:58:30.860 align:middle line:90% running contracts as a node. 00:58:30.860 --> 00:58:33.860 align:middle line:84% So we have to worry about whether the multiple contracts 00:58:33.860 --> 00:58:38.680 align:middle line:84% that can coexist with each other are consistent with one another. 00:58:38.680 --> 00:58:43.100 align:middle line:90% 00:58:43.100 --> 00:58:49.500 align:middle line:84% So let me turn to current difficulties in addition 00:58:49.500 --> 00:58:52.020 align:middle line:90% to what we've already said-- 00:58:52.020 --> 00:58:54.520 align:middle line:84% legacy technology, trust and fraud, 00:58:54.520 --> 00:58:59.180 align:middle line:84% unmeasured currency again, data silos, and inconsistent 00:58:59.180 --> 00:59:02.260 align:middle line:90% collection mechanisms. 00:59:02.260 --> 00:59:06.940 align:middle line:84% So legacy technology-- this is from the Kansas City Fed, part 00:59:06.940 --> 00:59:10.340 align:middle line:90% of the Federal Reserve System. 00:59:10.340 --> 00:59:17.020 align:middle line:84% So Fed Now is a fast payment system 00:59:17.020 --> 00:59:23.640 align:middle line:84% that the US has finally implemented literally 00:59:23.640 --> 00:59:27.240 align:middle line:84% for fast, almost instantaneous payments. 00:59:27.240 --> 00:59:31.800 align:middle line:84% However, commercial banks, depository institutions, 00:59:31.800 --> 00:59:33.920 align:middle line:84% must make significant changes to how 00:59:33.920 --> 00:59:38.360 align:middle line:84% they manage payments to offer customers this instant payment 00:59:38.360 --> 00:59:39.140 align:middle line:90% service. 00:59:39.140 --> 00:59:43.040 align:middle line:90% 00:59:43.040 --> 00:59:45.200 align:middle line:90% I did not know about this. 00:59:45.200 --> 00:59:48.140 align:middle line:84% I learned this when I was visiting the Bank of Canada. 00:59:48.140 --> 00:59:52.000 align:middle line:90% 00:59:52.000 --> 00:59:56.120 align:middle line:84% Some features of instant payments include 00:59:56.120 --> 01:00:00.720 align:middle line:84% around-the-clock services, instant accessibility-- 01:00:00.720 --> 01:00:03.400 align:middle line:84% and this requires real-time core-- 01:00:03.400 --> 01:00:06.320 align:middle line:84% payment processing and automated connectivity 01:00:06.320 --> 01:00:10.040 align:middle line:84% to the instant payment systems operator. 01:00:10.040 --> 01:00:14.480 align:middle line:90% But this is not automatic. 01:00:14.480 --> 01:00:19.560 align:middle line:84% Many depository institutions-- I'm paraphrasing here-- 01:00:19.560 --> 01:00:26.780 align:middle line:84% need to and have not yet modernized their systems 01:00:26.780 --> 01:00:31.380 align:middle line:84% to be able to use the fast payment service. 01:00:31.380 --> 01:00:34.020 align:middle line:90% So how are they getting by? 01:00:34.020 --> 01:00:40.580 align:middle line:84% Typically, they will subcontract to an aggregator payment system 01:00:40.580 --> 01:00:43.760 align:middle line:84% provider, which, as it turns out, 01:00:43.760 --> 01:00:47.420 align:middle line:84% is an incredibly concentrated industry. 01:00:47.420 --> 01:00:52.120 align:middle line:84% So we start to touch on industrial organization. 01:00:52.120 --> 01:00:53.740 align:middle line:90% And I don't know. 01:00:53.740 --> 01:00:56.240 align:middle line:84% It could be inherent in the technology. 01:00:56.240 --> 01:01:01.660 align:middle line:84% It could be sort of an economic capture. 01:01:01.660 --> 01:01:07.180 align:middle line:84% But that, anyway, is a fact for the legacy technology. 01:01:07.180 --> 01:01:09.400 align:middle line:90% Trust and fraud. 01:01:09.400 --> 01:01:09.900 align:middle line:90% OK. 01:01:09.900 --> 01:01:16.900 align:middle line:84% This is Sweden up here, which has 01:01:16.900 --> 01:01:22.000 align:middle line:84% to do with the value of the amount of currency 01:01:22.000 --> 01:01:26.560 align:middle line:84% left in the system and sometimes divided 01:01:26.560 --> 01:01:29.560 align:middle line:90% by the nominal value of GDP. 01:01:29.560 --> 01:01:34.120 align:middle line:84% Either way, it's kind of going down. 01:01:34.120 --> 01:01:37.880 align:middle line:84% In fact, for quite a while, the Riksbank 01:01:37.880 --> 01:01:44.360 align:middle line:84% was concerned that the central bank liabilities are 01:01:44.360 --> 01:01:48.080 align:middle line:84% going to be disappearing and not used 01:01:48.080 --> 01:01:53.320 align:middle line:84% at all because commercial banks have a net clearing 01:01:53.320 --> 01:01:58.660 align:middle line:84% operation having to do with Swish, which is an odd name. 01:01:58.660 --> 01:02:03.300 align:middle line:84% But Swish is an instant payment system among Swedish households. 01:02:03.300 --> 01:02:07.400 align:middle line:90% 01:02:07.400 --> 01:02:09.840 align:middle line:84% That concern led them to contemplate 01:02:09.840 --> 01:02:13.320 align:middle line:84% having a central-bank-issued digital currency. 01:02:13.320 --> 01:02:15.720 align:middle line:84% But I think a few months ago, they 01:02:15.720 --> 01:02:18.460 align:middle line:84% decided they're not going to do it. 01:02:18.460 --> 01:02:19.520 align:middle line:90% This is cute. 01:02:19.520 --> 01:02:23.260 align:middle line:84% This is from Stefan Ingves, who was the then Governor 01:02:23.260 --> 01:02:24.720 align:middle line:90% of the Bank of Sweden. 01:02:24.720 --> 01:02:28.380 align:middle line:90% 01:02:28.380 --> 01:02:36.180 align:middle line:84% In anthropological writings, the reserve 01:02:36.180 --> 01:02:41.740 align:middle line:84% was a big stone buried at the bottom of a lake, 01:02:41.740 --> 01:02:44.700 align:middle line:84% and people would trade claims on it. 01:02:44.700 --> 01:02:47.340 align:middle line:90% And that was the currency. 01:02:47.340 --> 01:02:53.260 align:middle line:84% So Stefan is displaying this stone being 01:02:53.260 --> 01:02:55.200 align:middle line:90% transferred on a cell phone. 01:02:55.200 --> 01:02:58.420 align:middle line:90% 01:02:58.420 --> 01:03:00.940 align:middle line:90% It's kind of cute. 01:03:00.940 --> 01:03:03.060 align:middle line:90% This is Kenya. 01:03:03.060 --> 01:03:08.740 align:middle line:84% And here the e-money, so to speak, 01:03:08.740 --> 01:03:12.660 align:middle line:90% is basically phone credits. 01:03:12.660 --> 01:03:17.040 align:middle line:84% So they jumped the landline technology. 01:03:17.040 --> 01:03:19.280 align:middle line:84% Almost everyone has a cell phone, 01:03:19.280 --> 01:03:23.320 align:middle line:84% and they have multiple accounts on cell phones. 01:03:23.320 --> 01:03:27.800 align:middle line:84% So you can use the fiat money, Kenyan shillings, 01:03:27.800 --> 01:03:29.840 align:middle line:90% to buy an account. 01:03:29.840 --> 01:03:33.160 align:middle line:84% But then you can transfer your credits, your cell phone 01:03:33.160 --> 01:03:36.120 align:middle line:90% credits, to somebody else. 01:03:36.120 --> 01:03:39.080 align:middle line:90% And that becomes an e-money. 01:03:39.080 --> 01:03:42.200 align:middle line:84% And this is a picture of how it works. 01:03:42.200 --> 01:03:45.980 align:middle line:84% Central bank fiat money denoted with dollars, so to speak. 01:03:45.980 --> 01:03:47.960 align:middle line:90% It's Kenyan shillings. 01:03:47.960 --> 01:03:49.460 align:middle line:90% It was a British colony. 01:03:49.460 --> 01:03:56.360 align:middle line:84% And you can see the stick figure here takes his or her dollars 01:03:56.360 --> 01:04:01.320 align:middle line:84% to an agent, a Safaricom agent from the mobile company, 01:04:01.320 --> 01:04:05.320 align:middle line:84% and gets credited with the phone credits, which then 01:04:05.320 --> 01:04:08.320 align:middle line:90% can be used to buy something. 01:04:08.320 --> 01:04:11.000 align:middle line:84% And the supplier can cash it out. 01:04:11.000 --> 01:04:15.420 align:middle line:84% Goes back to a different agent to cash out. 01:04:15.420 --> 01:04:18.980 align:middle line:84% This is like cashing in, cashing out, or send money 01:04:18.980 --> 01:04:24.780 align:middle line:84% back home from Nairobi to his wife or mother 01:04:24.780 --> 01:04:26.380 align:middle line:90% still in the village. 01:04:26.380 --> 01:04:30.800 align:middle line:84% And she would take it to a local agent and cash out. 01:04:30.800 --> 01:04:34.060 align:middle line:90% It's a dual currency system. 01:04:34.060 --> 01:04:36.940 align:middle line:84% You have the e-money part of it, which are the phone credits, 01:04:36.940 --> 01:04:42.620 align:middle line:84% and the paper currency ultimately backed by accounts 01:04:42.620 --> 01:04:43.840 align:middle line:90% at the central bank. 01:04:43.840 --> 01:04:46.540 align:middle line:90% 01:04:46.540 --> 01:04:48.860 align:middle line:90% It actually works pretty well. 01:04:48.860 --> 01:04:57.620 align:middle line:84% But when you start to think about it, it's kind of risky. 01:04:57.620 --> 01:05:00.180 align:middle line:90% Where is the ultimate cash? 01:05:00.180 --> 01:05:04.620 align:middle line:84% It's in a Safaricom account at a commercial bank. 01:05:04.620 --> 01:05:07.980 align:middle line:90% And Kenyan banks do fail. 01:05:07.980 --> 01:05:09.440 align:middle line:90% So there's that risk. 01:05:09.440 --> 01:05:13.400 align:middle line:90% 01:05:13.400 --> 01:05:17.300 align:middle line:84% But it's not on a distributed ledger. 01:05:17.300 --> 01:05:20.840 align:middle line:84% So the record of your account resides with Safaricom 01:05:20.840 --> 01:05:25.000 align:middle line:84% and what you have shown on your cell phone. 01:05:25.000 --> 01:05:27.480 align:middle line:84% Broadening the discussion out a little bit, 01:05:27.480 --> 01:05:31.600 align:middle line:84% you have multiple cryptocurrencies. 01:05:31.600 --> 01:05:35.960 align:middle line:84% But they are built as exchanges of one currency 01:05:35.960 --> 01:05:39.120 align:middle line:84% for another on trusted broker dealers 01:05:39.120 --> 01:05:44.600 align:middle line:84% or centralized exchanges, not blockchains. 01:05:44.600 --> 01:05:49.040 align:middle line:84% So unfortunately, this is the story of FTX 01:05:49.040 --> 01:05:58.680 align:middle line:84% that diverted customer money into other investments. 01:05:58.680 --> 01:06:00.660 align:middle line:90% So that's the trust aspect. 01:06:00.660 --> 01:06:03.880 align:middle line:90% 01:06:03.880 --> 01:06:06.560 align:middle line:90% This is on unmeasured currency. 01:06:06.560 --> 01:06:09.520 align:middle line:84% First of all-- and I'll come back to this, 01:06:09.520 --> 01:06:12.500 align:middle line:90% hopefully at the end-- 01:06:12.500 --> 01:06:16.780 align:middle line:84% there's a paper for Thailand where paper currency 01:06:16.780 --> 01:06:18.020 align:middle line:90% was so heavily used. 01:06:18.020 --> 01:06:24.060 align:middle line:84% And we actually were able to back out huge measure 01:06:24.060 --> 01:06:29.940 align:middle line:84% the use of it and detected huge welfare losses. 01:06:29.940 --> 01:06:34.140 align:middle line:84% And then this Hamilton picture over here 01:06:34.140 --> 01:06:40.860 align:middle line:84% is what would have been the digital dollar. 01:06:40.860 --> 01:06:47.300 align:middle line:84% And DCI here on campus designed it and implemented the algorithm 01:06:47.300 --> 01:06:51.780 align:middle line:90% jointly with the Boston Fed. 01:06:51.780 --> 01:06:59.740 align:middle line:84% However, they wanted to preserve the privacy of the transactions. 01:06:59.740 --> 01:07:04.320 align:middle line:84% So it was heavily influenced, for good or bad, 01:07:04.320 --> 01:07:09.680 align:middle line:84% by cryptographers who did not want the history of transactions 01:07:09.680 --> 01:07:12.820 align:middle line:84% to be anywhere in sight and basically destroyed. 01:07:12.820 --> 01:07:17.480 align:middle line:84% So it didn't have contract functionality. 01:07:17.480 --> 01:07:19.700 align:middle line:84% And there's a new version that does, 01:07:19.700 --> 01:07:25.160 align:middle line:84% but it wasn't their first inclination. 01:07:25.160 --> 01:07:27.980 align:middle line:84% So then we have the problem of unmeasured currency. 01:07:27.980 --> 01:07:30.760 align:middle line:84% And we're back to earlier discussions. 01:07:30.760 --> 01:07:34.260 align:middle line:84% Do we want to just have an electronic version of the cash, 01:07:34.260 --> 01:07:41.400 align:middle line:84% or do we want more capabilities with potential trade-offs? 01:07:41.400 --> 01:07:46.160 align:middle line:84% And then a slide here on data silos. 01:07:46.160 --> 01:07:50.020 align:middle line:84% And here I was referring to Indonesia, 01:07:50.020 --> 01:07:54.360 align:middle line:84% but it could have been another payment provider where basically 01:07:54.360 --> 01:08:01.920 align:middle line:84% they develop apps for payments for credit lending and so on. 01:08:01.920 --> 01:08:04.840 align:middle line:84% And each is a separate department, 01:08:04.840 --> 01:08:07.940 align:middle line:90% and the database is separate. 01:08:07.940 --> 01:08:12.460 align:middle line:90% There is a common identifier-- 01:08:12.460 --> 01:08:13.840 align:middle line:90% multiple, actually. 01:08:13.840 --> 01:08:18.779 align:middle line:84% It's a lot of work to unify the data in order 01:08:18.779 --> 01:08:23.319 align:middle line:84% to know the transactions belong to a common entity. 01:08:23.319 --> 01:08:26.140 align:middle line:84% So it's not like the basic unit is 01:08:26.140 --> 01:08:30.140 align:middle line:84% the business or the basic unit is the household. 01:08:30.140 --> 01:08:34.300 align:middle line:84% And then there are multiple uses. 01:08:34.300 --> 01:08:35.640 align:middle line:90% It's not designed that way. 01:08:35.640 --> 01:08:37.640 align:middle line:90% So this is another problem. 01:08:37.640 --> 01:08:45.460 align:middle line:90% 01:08:45.460 --> 01:08:49.100 align:middle line:84% So inconsistent collection mechanisms. 01:08:49.100 --> 01:08:52.100 align:middle line:84% So this is what I was alluding to before-- 01:08:52.100 --> 01:08:58.340 align:middle line:84% again, the savings, profits not eaten or given out 01:08:58.340 --> 01:09:01.500 align:middle line:84% as dividends, or savings that should show up 01:09:01.500 --> 01:09:03.920 align:middle line:84% in the change in assets in the balance sheet. 01:09:03.920 --> 01:09:09.720 align:middle line:84% So this is how the accounts are connected in principle. 01:09:09.720 --> 01:09:14.920 align:middle line:84% But we did this with the Boston Fed 01:09:14.920 --> 01:09:18.000 align:middle line:84% for the Panel Study of Income Dynamics, which 01:09:18.000 --> 01:09:21.720 align:middle line:84% is one of the premier US microdatabases 01:09:21.720 --> 01:09:26.399 align:middle line:84% for the consumption expenditure survey, the survey of consumer 01:09:26.399 --> 01:09:28.560 align:middle line:90% finances. 01:09:28.560 --> 01:09:33.880 align:middle line:84% And long story short, we looked at the discrepancy 01:09:33.880 --> 01:09:39.160 align:middle line:84% between cash holdings on the constructed statement of cash 01:09:39.160 --> 01:09:43.439 align:middle line:84% flows from each and every one of these data surveys 01:09:43.439 --> 01:09:46.800 align:middle line:84% with the change in cash holdings in the balance sheet. 01:09:46.800 --> 01:09:50.840 align:middle line:90% And it's typically way off. 01:09:50.840 --> 01:09:53.840 align:middle line:84% I'm not saying these are bad surveys. 01:09:53.840 --> 01:09:55.880 align:middle line:84% They're basically the bread and butter 01:09:55.880 --> 01:10:00.520 align:middle line:84% of research on many aspects in the US. 01:10:00.520 --> 01:10:05.820 align:middle line:84% But they were not constructed with this notion of consistency 01:10:05.820 --> 01:10:09.420 align:middle line:90% across financial accounts. 01:10:09.420 --> 01:10:14.080 align:middle line:84% And likewise for new data or something, 01:10:14.080 --> 01:10:16.420 align:middle line:84% if you were starting from scratch, 01:10:16.420 --> 01:10:20.700 align:middle line:84% you probably want to have that standard as a check on the data 01:10:20.700 --> 01:10:26.020 align:middle line:84% you're proposing to enter into the database. 01:10:26.020 --> 01:10:27.480 align:middle line:90% That was a micro picture. 01:10:27.480 --> 01:10:30.740 align:middle line:84% There's a macro picture, which is very similar. 01:10:30.740 --> 01:10:36.380 align:middle line:84% These are macro aggregated accounts 01:10:36.380 --> 01:10:42.580 align:middle line:84% done by the Department of the Bureau of Census. 01:10:42.580 --> 01:10:46.760 align:middle line:90% And you get inconsistencies. 01:10:46.760 --> 01:10:50.100 align:middle line:90% 01:10:50.100 --> 01:10:52.280 align:middle line:84% So both at the macro and micro level, 01:10:52.280 --> 01:10:55.500 align:middle line:84% unfortunately, we don't have very much consistent data 01:10:55.500 --> 01:10:56.320 align:middle line:90% in the US. 01:10:56.320 --> 01:11:01.660 align:middle line:84% But I've done it with the data we collected in Thailand, where 01:11:01.660 --> 01:11:05.860 align:middle line:84% we do have consistent accounts, and hence we can aggregate up. 01:11:05.860 --> 01:11:09.760 align:middle line:90% 01:11:09.760 --> 01:11:15.360 align:middle line:84% And we'd like to go and do this in the US somehow. 01:11:15.360 --> 01:11:21.260 align:middle line:84% And there is a project underway in West Virginia with Scott, 01:11:21.260 --> 01:11:25.000 align:middle line:84% who was at the Boston Fed, now at the University of West 01:11:25.000 --> 01:11:27.520 align:middle line:90% Virginia. 01:11:27.520 --> 01:11:30.940 align:middle line:90% And so we're creating templates. 01:11:30.940 --> 01:11:39.060 align:middle line:84% The idea would be not to do what we did in Thailand, 01:11:39.060 --> 01:11:46.760 align:middle line:84% where we're using enumerators to have and recall 01:11:46.760 --> 01:11:52.280 align:middle line:84% one month at a time but rather to ask the households if they're 01:11:52.280 --> 01:11:56.880 align:middle line:84% willing to share with us their electronic data, whether it's 01:11:56.880 --> 01:12:01.540 align:middle line:84% commercial bank account or credit card account and so on. 01:12:01.540 --> 01:12:04.460 align:middle line:84% And then we merge that along with an enumeration 01:12:04.460 --> 01:12:06.460 align:middle line:84% of their paper currency transactions 01:12:06.460 --> 01:12:10.980 align:middle line:84% into consistent financial accounts. 01:12:10.980 --> 01:12:12.540 align:middle line:90% It is a labor of love. 01:12:12.540 --> 01:12:19.500 align:middle line:84% It's taking some time, but we're hopeful. 01:12:19.500 --> 01:12:28.200 align:middle line:84% So I just wanted to say at the end two things. 01:12:28.200 --> 01:12:32.120 align:middle line:84% One, this lecture on distributed ledgers, 01:12:32.120 --> 01:12:34.620 align:middle line:84% payments, cash, et cetera-- there's 01:12:34.620 --> 01:12:39.660 align:middle line:90% a parallel lecture for 193. 01:12:39.660 --> 01:12:43.900 align:middle line:84% And some of you are registered actually to take that class, 01:12:43.900 --> 01:12:48.180 align:middle line:84% but it's not a requirement for this class. 01:12:48.180 --> 01:12:51.120 align:middle line:84% The lecture material will be available. 01:12:51.120 --> 01:12:59.920 align:middle line:84% And I just edited, and Deb will release lecture one in 193, 01:12:59.920 --> 01:13:02.180 align:middle line:90% and it's parallel with this. 01:13:02.180 --> 01:13:06.360 align:middle line:84% So you can read about measurement of inefficiency 01:13:06.360 --> 01:13:09.400 align:middle line:90% in cash, et cetera. 01:13:09.400 --> 01:13:14.400 align:middle line:84% And it ends still not written down 01:13:14.400 --> 01:13:17.960 align:middle line:84% with the experience we had in Indonesia, 01:13:17.960 --> 01:13:25.340 align:middle line:84% where it's clear that it's neither a pure digital economy, 01:13:25.340 --> 01:13:30.900 align:middle line:84% despite the digitization, nor is it a pure fiat money economy. 01:13:30.900 --> 01:13:32.640 align:middle line:90% It's a dual economy. 01:13:32.640 --> 01:13:34.640 align:middle line:84% And households and businesses are really 01:13:34.640 --> 01:13:39.320 align:middle line:84% struggling with developing strategies for payments, 01:13:39.320 --> 01:13:43.640 align:middle line:84% with inflows and outflows across these multiple accounts, 01:13:43.640 --> 01:13:47.600 align:middle line:84% including the fact that there are transactions costs. 01:13:47.600 --> 01:13:51.720 align:middle line:84% And if you're using a local informal agent, 01:13:51.720 --> 01:13:57.000 align:middle line:84% there are huge markups and potentially theft and fraud. 01:13:57.000 --> 01:14:02.260 align:middle line:84% So the first goal is documentation. 01:14:02.260 --> 01:14:05.580 align:middle line:84% And I'll just say that if anyone is-- 01:14:05.580 --> 01:14:07.660 align:middle line:90% hopefully, someone will be-- 01:14:07.660 --> 01:14:10.540 align:middle line:84% interested in doing this, it's to take 01:14:10.540 --> 01:14:16.560 align:middle line:84% a version of the material covered here, 01:14:16.560 --> 01:14:20.500 align:middle line:84% along with the details of creating consistent accounts 01:14:20.500 --> 01:14:25.940 align:middle line:84% and create a template that we can use in turn in Indonesia 01:14:25.940 --> 01:14:29.420 align:middle line:84% when we go and interview households and businesses. 01:14:29.420 --> 01:14:31.960 align:middle line:84% So it's a bit like starting from scratch in West Virginia, 01:14:31.960 --> 01:14:36.780 align:middle line:84% except we would be doing it in Indonesia. 01:14:36.780 --> 01:14:42.800 align:middle line:84% So some of you are involved as already research assistants. 01:14:42.800 --> 01:14:48.600 align:middle line:84% And so that's a proposal if you're interested, 01:14:48.600 --> 01:14:50.780 align:middle line:90% but it's not an obligation. 01:14:50.780 --> 01:14:54.900 align:middle line:84% I will try in each lecture, if I can remember 01:14:54.900 --> 01:14:57.920 align:middle line:84% to do it, to form these links to the larger 01:14:57.920 --> 01:15:00.720 align:middle line:90% picture and potential research. 01:15:00.720 --> 01:15:06.160 align:middle line:84% And even if-- you know, otherwise, there's 01:15:06.160 --> 01:15:11.360 align:middle line:84% a whole series of papers in the syllabus, some with double stars 01:15:11.360 --> 01:15:12.420 align:middle line:90% and many others. 01:15:12.420 --> 01:15:15.400 align:middle line:84% So for your project for this class, 01:15:15.400 --> 01:15:21.080 align:middle line:84% feel free to peruse and pick potentially. 01:15:21.080 --> 01:15:22.300 align:middle line:90% See what's in there. 01:15:22.300 --> 01:15:26.320 align:middle line:84% Think about it as a write-up of a two pager or something 01:15:26.320 --> 01:15:31.880 align:middle line:84% to share with your classmates or even potentially the one 01:15:31.880 --> 01:15:35.720 align:middle line:90% longer-term research proposal. 01:15:35.720 --> 01:15:37.640 align:middle line:90% OK. 01:15:37.640 --> 01:15:40.280 align:middle line:84% And the red hand is crossing 12:00. 01:15:40.280 --> 01:15:42.430 align:middle line:90% That's kind of amazing. 01:15:42.430 --> 01:15:56.000 align:middle line:90%