Transcript of the talk with slides and introduction by Vinay Gupta our Medium blog:
First, let me just tell you a little bit about myself, I’ll take five minutes to talk about Sweetbridge, and then I want to talk about something completely different. Just to tell you a little bit about my background, I’ve been writing smart contracts that weren’t known as smart contracts, but we’re actually taking commercial agreements between the largest corporations of the world, automating the commercial agreements and their supply chains so that they could conduct supply chain logistics transactions, and audit and validate that the transactions that were occurring between parties actually held up to what the parties agreed to. The reason they did this is because over the course of a five-year period of time – just as an example, although this was for 20 years – we actually executed and automated the contractual transactions of over two billion trades – commercial trades, not financial trades – and I learned some really interesting things in that process.
For example, guess what the most common dispute reason was? It was that the parties didn’t agree on the same version of the contract or agreement, they didn’t actually have the same version. The vast majority of time when they didn’t have actually the same version, one party or the other had not executed it through their internal process and returned it, but the other party had executed it and just assumed that it had been enforced. And when it hadn’t been executed, the most common reason for not being executed was that it was sitting in one of the eight or nine or ten email boxes of some executive in the corporation who was supposed to sign it, and it got stuck in their email box when they were out of town or on vacation, all sorts of other stuff got laid on top of it, and their admin, if they even had one, forgot that it was there, and it just basically went to the bottom of the pile because there wasn’t anybody poking them, saying, “Have you signed the agreement?”
I want to try to make some points, I’ll give you some other cases. The most common problem was being billed for something that you had already paid for by another means, either another invoice or another thing, and it was just pure error. We went in and looked at what’s driving this, why is the world this broken, we did a really huge study of the data, and over two billion trades – commercial trades that were for goods and services, not financial trades – if you looked at the information that was being passed between the parties, all the messages and data that was going back and forth, and you just implied the standard that the parties have agreed to use for that, 99% of all of the messages had at least one data defect. Yet, it all works. Just think about that for a second, because we’ve been hearing about all the problems, and yet it all works. The iPhones for Apple still got made, they still got shipped to the right places – Apple, by the way, was a customer – the HP laptops still got made, they still got shipped to the right places, the vendors still got paid – it all works
Well, it turns out that the world is a messy place, and we’ve been hearing about all the problems with the mess, and the messy place has a bunch of messy bits. What it is about blockchain that’s really intriguing to me is not that it’s going to make the world less messy, I think the analogue-to-digital conversion is always going to have points where there’s value. Any time we go from the analogue to digital you have to go through a conversion process: whether that’s the example of the three-year-old pushing a button or it’s the self-driving car or whatever, whenever we’re dealing with the analogue-to-digital boundary, we have to understand that the digital boundary is an approximation of reality. It is not reality; it is approximation.
I’m going to approach this thing from a completely different perspective than everybody else, but it’s a little bit because of the experiences I’ve had. The question that is really interesting to me is if there’s all these problems, why is the blockchain actually a good idea? If there’s all these nascent things and if it’s so undeveloped and it’s so immature, then why is it something we should actually be doing, except as maybe a research project or a prototype or something like that? Can we actually do real things with it? Up until March-April time frame of this year my answer was no, my answer was no, but March-April time frame of this year I became convinced that that was not the right answer anymore, and I thought it would be helpful for me to just give you this little bit of understanding of why, with the background I just gave.
If you think of the blockchain as a transaction machine that allows you to have the same thing, have the transaction that is the legal transaction, the value transaction and the state change in the recordkeeping system all either happening or not happening, that is I think very fundamental. Irregardless of whether that is enforceable in every case, irregardless of whether that actually gets the right person paid in every case, irregardless of whether it’s perfect, the question is: is it better? Let me explain why it’s better in financial terms.
These errors that we’ve talked about that were occurring in these organisations – the largest companies on the planet, the largest supply chains in the world, companies that are the size of midsized countries, that do $100-billion businesses – what does it cost them today to run the world the way the world works today? It’s a staggering number. Just to validate that they are built correctly… I want you to get your head around this: just to validate that they’re built correctly so they can actually put together a valid account of the business, which they’re required to do by law if they’re publicly traded, just to do that costs them between 12 and 20 basis points of their revenue, 12-20 basis points. What is a basis point? A basis point is a hundredth of a percent. Just to figure out where does it go, on which account… This is after all of the automation work we’ve done, this is after all of the automated messaging and EDI infrastructure, this is after all of the billions we’ve spent on putting SAP or Oracle or other ERP systems in place, it still costs that much money.
And then how often is it wrong? Well, I ran a company that made money by auditing these things for error. The level of error is about 50 basis points, half a percent. That’s a pretty stunning number. That means if you’ve got $100-billion corporation, a half a percent is a really material number, it’s a really, really big number. So regardless of all the problems, regardless of all the things, can we use what we have to actually make things marginally better? I want to argue that we can. But first I’d like to talk to you a little bit about what Sweetbridge is and what the supply chain is, just to level-set people so we have a common understanding.
Sweetbridge is about being a blockchain alliance for the supply chain. We are not just a project; we are a project of projects, we’re a community much like Mattereum, that’s trying to bring together all of the people that are trying to do blockchain-type things on supply chains and global commerce.
Most people when they think about a supply chain have an image like this in their head. This is a very tiny part of a supply chain; this is a piece of the supply chain, but it is a very, very small part of the supply chain. When we talk about supply chains, we actually mean something much bigger than this, and the easiest way I’ve found that you can understand this is that it’s either Mother Nature, or the supply chain.
Which means that everything you touch, use, sit in, live in, consume in any way in your day either was created where it is or grown where it is by Mother Nature, or it came to you through a supply chain. Everything you’re wearing right now, the seats we’re sitting on, the building we’re in, these are all products of a supply chain, and supply chains actually manage commerce.
If you take a look at the things that aren’t financial services, this is $54 trillion of GDP so it’s a very big number, it is two-thirds of everything that produce GDP. Just to make sure people understand GDP, because I didn’t and so I won’t assume that you do, GDP is not the amount of sales activity; it’s the amount of profit. It’s the value that’s been added in the system as a result of the activities that have been performed, so the actual number of transactions moved is much larger.
We’re about trying to make this liquid. Why do you want to make this liquid? I’ll give you three reasons real quick and then we’ll get into the practical thing I wanted to talk about. There’s three reasons you want to make supply chains liquid, and the first is capital. At any given time in the supply chains of the world there are trillions of dollars held up between the order being placed and the payment being provided, and this is because in the real world – not the financial world, the real world – things have to actually move through space and time, and this doesn’t happen immediately. They also have to be created. When you go to the store, you don’t give them your credit card and then it gets charged immediately and you pick up your goods; in business it doesn’t work that way.
The other thing is that the assets of the supply chain of the world – the factories, the warehouses, the trucks, the planes, the ships – have become horribly utilised as a result of hypercompetition. In fact, according to the US Federal Reserve, 50 years ago 90% of the assets in the supply chains of the world were utilised at any given point in time; today it’s less than 75% and it is continuing to drop. This means that we have more factories sitting idle, more warehouses basically empty and more trucks that don’t have anything in them… It doesn’t mean that that’s happening in all places, because right now for example trucks in Europe, there’s a driver shortage coming and there’s a truck shortage coming, because we lost a lot of the assets and the fleets of the world during 2008. So it’s not that there aren’t some cases where there’s stuff that’s highly utilised; it’s that in general the assets are underutilised.
The last thing is that the change and pace of change in supply chains is so great that no organisation can keep an optimised supply chain, because they literally cannot afford the talent to do it, it doesn’t matter who you are. Again, in my last business we ran big data analytics systems that would actually look at massive sets of data that we gathered through all this subtle activity, and we’ve found in every single case, no matter who the organisation was, that they were underperforming versus what they could be performing, and we’ve never found a company that that number was less than 5%. What this means is that at any given time businesses have a lot of money tied up in things they’re doing, but they either have to rent money for it, pay interest on it, or they have to do without the money, known as working capital. It means that at any given time we have lots of assets that are underutilised, and that the supply chains that are actually delivering us all this stuff basically are unoptimised. If you add all this stuff up, it actually equates to something between 12-15% of that $54 trillion number, which is meaningful to the GDP of the world. That means that if you could fix these problems, you could actually solve some problems that would actually cause the global economy to grow by something in the neighbourhood of 5-10%.
The reason this is important, the thing I wanted to talk about is why should we care? Why should we care about that it’s hard, it’s messy, it’s still new? Well, the answer is because we can do something really significant about it. To demonstrate what we can do, I want to talk a little bit about the UK and manufacturing, because Digital Catapult, which we are a part of and I’m trying to help here in the UK, has a focus beyond the things we heard about already today, in music, and that is in helping UK manufacturing. If you look at UK manufacturing, it’s 10% of the UK GVA, which is the value add that’s occurring as a result of the manufacturing process. Once you take the raw materials that are coming in and you take the cost of that from the amount of sales that goes out, that’s the GVA.
This is happening on about £364 billion. A one percent, if you could just lower the cost of the average weighted cost of capital for manufacturers, that’s what WACC means, is a £3.4 billion increase in the UK GVA. This can very easily be done with the blockchain today by being able, and this is going to probably sound freakish – if you want to understand how this works, download the white paper, because I don’t have time today to tell you how it works – but you can basically, as an organisation, loan yourself money off your own assets, without a bank, without a counterparty, interest free using the blockchain and smart contracts that promise to sell your property if you do not repay yourself. This is a very doable thing, it is how banking and financial services actually worked in the world for about 5,500 years. A one percent reduction in the average weighted cost of capital for the British government would be a massive increase in value to the citizens. And 99% of the manufacturers in Britain are SMEs, they’re small companies, so their access to capital is probably poor and expensive. I actually tried to find the numbers last night online and I couldn’t.
Why is this a problem? Well, this is the problem: it’s the fact that you place an order from the shop to the factory, the factory has to order from the raw materials supplier, the good comes into the factory, it gets shipped to the shop, and the shop basically makes their payment and the factory makes their payment. This whole cycle on average takes 90 days. I don’t know what it is here, but around the world the average is 90 days, and from shipment of goods to payment the average is 52 days.
The problem here is that the assets that businesses have – invoices, orders, inventory, etc. – they have to be able to be controlled by the blockchain in order for you to create liquidity from them without another counterparty or interest free, which you need to do to lower the average weighted cost of capital, which you could probably do by a significant margin, maybe above 1%, but in our thought experiment it was just 1%.
The need here to make this work is that you actually have to have smart legal contracts. Notice I didn’t use the word “smart contracts”; you have to have a legal contract, a real-world contract that basically is a Ricardian contract, that basically takes its state information about ownership, so the party that owns it, and the current state of the property, you have to be able to control that in a blockchain, but very little else. You don’t need to be able to perfect all of the clauses in the blockchain, although that would be nice; you just need to control that. The other problem you have to have is you have to have some concept of identity that deals with authentication and authorisation, in addition to just whether you know who that party is, so you have to have some mechanism for this. Authentication means not only do I have an identity but I’m actually the one using the identity – so think about that, that’s kind of important – and authorisation is I have a business for example and who in the business actually has the right to bind me to this thing, I have to have that too.
You can solve some of these problems with legal contracts by the way, this is a problem in legal contracts today, so you use the same mechanisms. As we’ve just heard, you have to have dispute resolution. Why? Because for this whole thing to work you have to know how the disputes are going to be settled. If you leave that open ended, then you can’t actually ascertain the risk of what happens if you actually try to repossess the inventory because somebody didn’t pay back themselves from a loan that they took out on the inventory for themselves. Then you have to have how you’re going to do the enforcement, you have to have this because you have to be able to tell people in advance what’s going to happen. You have to tell them, “Listen, if you don’t do these positive things for getting this right, then this is the negative thing that’s going to happen,” and you have to be able to tell them in advance of that, or frankly they’re not going to be very supportable in legal environments, and you have to do this by jurisdiction. So, quite a bit of work for all the lawyers, quite a bit of work for some smart contract developers, quite a bit of work for us to do, but it has some pretty interesting opportunities.
The opportunity isn’t just for manufacturing, that we’ve already talked about, but now let’s just take a look at global trade. Let’s say that in doing this you can solve 1% of cost to be taken out of the settlement process for import or export transactions, and let me just explain to you why that 1% number is relevant. If you’ve ever gone to the bank and you have exchanged your pounds for another currency, or went to another country and withdrew with your credit card, you probably noticed that you got a fee. Companies, when they actually sell things or buy things, have to do that in currencies that are going to be accepted between the buyer and seller, and when they do this they have to actually exchange money for fees. This causes about a 2-4% fee to be charged, depending on how big the company is, that’s about what it’s going to pay. If you can cut that by half – which you could, using a settlement process that ran on the blockchain – this would be worth £28 billion to Great Britain in GVA. Very big targets, very big things, and this is why I think we should care, is that we only need to make things marginally better. We don’t need to make things perfect; we just need to make them better and that can make a very big deal.
Why law firms should care is let’s just say that this is a new industry being able to do this, and that you could extract from that new industry just one basis point, just one basis point is… The import/export business of the UK is £1.4 trillion, so just one basis point – one hundredth of a percent, that was what was available – is £138 million a year of revenue. This a very nice target, and one-hundredth of a percent would be such a low fee that it would hardly be felt and it’s a very big whale to go after.
So, what I wanted to suggest is that though it’s messy, going from analogue to digital, and though there are obstacles, that the rewards are quite large for those who will participate in trying to make this happen, both in terms of social good, improving an economy in a meaningful way, and for those who basically make this a business activity, by tapping into revenue streams and opportunities that don’t already exist.
To make all of this happen is going to take a community. That’s why we are trying to build an alliance of people that are interested in doing this, that’s why Mattereum is trying to put together a group of people trying to do this, because they understand the same thing. It’s going to take us all to make this happen, but it’s worth doing. Thank you! [applause]