 Clearly, one issue of creating monopolies is monopolies exercise monopoly power. If you think about Amazon, Amazon, it's squeezing its suppliers, it's squeezing its workforce. If we think about Facebook and some of the other aggregators, we think, oh, they're not monopolies, they're not abusing their power because we don't pay for those sites, but they are, they're just doing it in different ways. If we think about Uber, in fact, there's a brilliant podcast, if you have time to listen to it, by Freakonomics and they look at the demand curve for Uber and how Uber has essentially moved welfare away from its drivers towards them, they've extracted some of that rent and towards us as consumers. Because we benefit as consumers, we sometimes forget that other people in the value chain are getting squeezed. One issue with aggregators is that they exercise, they become monopolies and they exercise monopoly power. The other issue is that rather than help us to manage the abundance of information and help us to get a higher return on our attention, they're actually stealing our attention. Every time you get an alert to say you've been tagged in a photo, every time you get an alert to say somebody's retweeted your tweet, they are hijacking your attention. They're taking you to that site and then you're probably going to spend an hour on that site before you've even realized that. This here is an example of YouTube. YouTube starts the next video straight after the last one you watched. They don't want you to leave those sites. Every time you go onto a site, they present you with their menu, their timeline, not your timeline. So rather than help us with our attention, they're stealing our attention and they're serving us up fake news. So they're not taking any responsibility for the content on their site. They're serving us up fake news and they're creating filter bubbles. So this is a Wall Street Journal site, I think it's called Blue and Red Facebook. So you can go on there any time you want and you can search for many different topics and you can see what conservatives are saying and what liberals are saying on social media and they're completely different. As we found out, if you lived in the UK during the Brexit vote, I don't know, and a single person who voted to lead the European Union. And so it's just, we are living in two different worlds, two different communities because of filter bubbles. So I think, so I'm going to make a prediction, which is I think we are going to see these advertising based aggregation models being challenged. And they're going to be challenged because of regulation, things like GDPR, which are giving us more control over our data. They're going to be challenged, I think, because ad tech has been oversold. So I read the other day that Proctor and Gamble cut their ad tech budget by a billion dollars and saw absolutely no impact whatsoever. I think, so I think ad tech is overhyped and oversold. And I think the other reason is that as consumers, you know, we're starting to appreciate that we are overloaded and we actually need aggregators who will help us to make better decisions. They'll help us to get a higher return on our finite attention. And it's into this, oh, sorry, in this, and I actually think you're starting to see this play out, right? So this is data from Y Combinator and it's data from all of the applications they get to join the program. And you can already see that the subscription based business models are starting to overtake advertising based business models. So I think you're already starting to see this play out. And it's into this space that I think banks can play, right? Because banks can do the role of aggregation. They can take all of our different data sets, contextual data, our locational data, our financial data, which they already have. And they can marry that with all of the other data that they collect, market data, risk data, and so on. And I think they're in a position to really help us, right? To make better financial and commercial decisions, right? Whether that's saving for a car, whether it's planning and putting aside enough money for retirement, whether it's managing a massive portfolio of investments. I think banks really are in a position to act as an aggregator and one that acts in the interest of its customers. And to talk about what externalities that might bring, right? So if they do pool all that information, they can reduce transaction costs. They can reduce transaction costs because we don't have to shop around. If they introduce us to products and services that we actually need and that would improve our financial welfare, then they could reduce transaction costs which would increase the provision of financial services. And don't forget, there's more than a billion people in the world that don't have access to financial services. Reducing adverse selection. So why can't businesses get credit? Why can't individuals sometimes get credit? Many times it's because there isn't enough information, right? Banks don't want to take the risk because there's no credit data. Banks don't want to take the risk because they don't know the financial performance of that business. More data in those situations can lead to less adverse selection and better outcomes. And then if we're pushing more money to business, if we are helping people get access to financial services for the first time, if we're giving loans to people who need them to create businesses, then we will see positive externalities. Just very quickly, if banks do take this role, they clearly need to make sure that they don't become the next Facebook, right? So they need to take care not to turn all the consumer surplus into producer surplus. So what I mean by that is they can't use our information to charge every single person at a different price for the same service. They will need to do much more vetting at their own sites, right? If we get shipped a book and the book is damaged, sure, we can leave a bad rating. That's not going to work in financial services. It's not going to work when we're talking about mortgages, when we're talking about financial advice. So they would have to do a lot more vetting. And then one other, just to leave one final thought about data, which is there's a lot of bias in the data sets that people use to train algorithms. And so I think banks need to be careful to avoid bias in the data and bias being transferred from bankers who then train algorithms. But anyway, just to leave you with a final thought, there is a big gap for banks here. Banks can step into our lives and they can help us to make better financial and commercial decisions. And there is a way forward to be a much better aggregator than the ones we have, the ones we work with today. So thanks. Thanks, Beck. Thanks, Beck.