The rise of machines
30 April 2019
David Lewis of FIS explains that while science fiction has represented the machines as a potential danger to humans, the reality is that without them, we simply could not function
Image: Shutterstock
Such headlines evoke fears of dystopian landscapes akin to films like the Terminator series—or even War of the Worlds—but in reality, machines and automation have become the very necessities of modern life. Without machines—by which, of course, I really mean computers and technology—where would we be? Without our location maps and social media services, we would not only be unable to find the restaurant we had booked online earlier that day, having read 32 reviews of it, but we would also certainly fail to inform all our friends of what we had eaten and shown them how it looked.
This is, of course, a more than flippant view of the technologically advanced world around us, but it is used as an illustration of the need we have for information and automation to lead our lives effectively and efficiently. We must also take the same approach in our professional lives. Consider just one of your tasks today and now imagine undertaking that without the assistance of any electronics of any sort.
The physical systems and interfaces we all use throughout our days, whether at work, on our way to work or even at home, rely not only on vast processing power undreamt of a generation ago, they require enormous amounts of information and data. Without the underlying data in the right quantities and level of quality, the outcomes become worthless or, arguably worse, they could be wrong. Data itself has become the cornerstone of automation, without which we would be unable to achieve and produce at the levels we can now, be that in our professional or personal lives.
There are many statistics out there regarding the amounts of data gathered and processed on a daily basis now, and they won’t be requoted here, mostly because by the time this is published that statistic will likely be out of date. It is sufficient to say, however, that vast quantities of data are required to drive the systems we use every day without a second thought. For example, at FIS, all of our data is stored in the cloud. Our laptops carry nothing of note, instead swapping enormous files of data backwards and forward near instantaneously—data required to undertake our every task.
The world of securities finance and collateral management is no different in this respect. Looking across our market and the systems we employ, it is easy to see the impact that data has had on our ability to process and manage complex tasks that would not have been possible just a few years ago. The machine constructed to break the enigma encryption code took on a Herculean task way beyond the capability of the humans that built it, not because the decryption task itself was impossible, but it was impossible in the timeframe available. For those not familiar with the story, the codes changed every 24 hours, so that while the code only had to be broken once, it had to be done in one 24-hour period, and failure meant a restart.
Managing a securities finance and enterprisewide collateral management business is not dissimilar, juggling the needs of the long and short books, while providing all the necessary collateral to satisfy the needs of associated businesses, only to have the supply and/or demand characteristics change almost constantly. Without the underlying data, showing available assets and collateral requirements coupled with acceptability criteria, credit limits, counterparty approvals, corporate actions, and intraday fee and rebate levels, the business could not possibly function. Again, imagine undertaking any, let alone all of these tasks, without any electronic assistance. Our businesses would simply not run without advanced automation, let alone run efficiently and within the requirements of the regulations that govern it.
‘Big data’ is an overused phrase, but one that is commonly used to describe the enormous amounts of new data entering our daily lives. However, here we are using ‘big data’ as an indicator of its importance, the fact that it has taken up a big part of our daily tasks, and without it, our systems would not run, and our industry would not function. Being able to do more with less is a common refrain around big companies, and the reality of that is that automated tasks often mean less people employed. Here, the rise of the machines is often not a welcome advance for some, just as self-driving vehicles seem highly attractive to many, cab and lorry drivers will have a different view.
Automation, in this sense, is clearly not just about undertaking enormous and complicated mathematical challenges that are simply beyond human capabilities in the timescales allowed, but it is also about efficiency. Clients expect services of better quality, higher accuracy and, of course, that could be delivered faster while getting cheaper. Witness the rise of zero-fee investment funds; that kind of offering is simply not possible without high levels of automation and even ‘robot advisors’ to take the place of expensive humans. Such offerings are also likely not possible without securities lending incomes to boost their returns and pay the managers fees, but that is another story altogether.
Efficiency is as much about financial capital as it is about human. Capital adequacy and the returns on capital employed are also vital measurements that create complex inputs into the daily process of managing a securities finance business. Complex and potentially conflicting pressures to provide the best, read cheapest, collateral to satisfy the client and the regulator at the same time, require iterative calculations, constantly rebalancing as demands change minute to minute. The demands of regulations don’t stop there of course. This year is bringing in increased demands for collateral to be posted on uncleared derivatives, placing additional volume and complexity through our allocation algorithms, and by April 2020, all our businesses will need to be able to marshal enormous amounts of data to enable them to tell their regulators all about it.
While science fiction has represented the machines as a potential danger to humans, the reality is that without them, and the data that feeds the automation they provide, we simply could not function. Whether in borrowing that share at the right rate, or booking the right restaurant, the machines are not only here to stay; we need them.
This is, of course, a more than flippant view of the technologically advanced world around us, but it is used as an illustration of the need we have for information and automation to lead our lives effectively and efficiently. We must also take the same approach in our professional lives. Consider just one of your tasks today and now imagine undertaking that without the assistance of any electronics of any sort.
The physical systems and interfaces we all use throughout our days, whether at work, on our way to work or even at home, rely not only on vast processing power undreamt of a generation ago, they require enormous amounts of information and data. Without the underlying data in the right quantities and level of quality, the outcomes become worthless or, arguably worse, they could be wrong. Data itself has become the cornerstone of automation, without which we would be unable to achieve and produce at the levels we can now, be that in our professional or personal lives.
There are many statistics out there regarding the amounts of data gathered and processed on a daily basis now, and they won’t be requoted here, mostly because by the time this is published that statistic will likely be out of date. It is sufficient to say, however, that vast quantities of data are required to drive the systems we use every day without a second thought. For example, at FIS, all of our data is stored in the cloud. Our laptops carry nothing of note, instead swapping enormous files of data backwards and forward near instantaneously—data required to undertake our every task.
The world of securities finance and collateral management is no different in this respect. Looking across our market and the systems we employ, it is easy to see the impact that data has had on our ability to process and manage complex tasks that would not have been possible just a few years ago. The machine constructed to break the enigma encryption code took on a Herculean task way beyond the capability of the humans that built it, not because the decryption task itself was impossible, but it was impossible in the timeframe available. For those not familiar with the story, the codes changed every 24 hours, so that while the code only had to be broken once, it had to be done in one 24-hour period, and failure meant a restart.
Managing a securities finance and enterprisewide collateral management business is not dissimilar, juggling the needs of the long and short books, while providing all the necessary collateral to satisfy the needs of associated businesses, only to have the supply and/or demand characteristics change almost constantly. Without the underlying data, showing available assets and collateral requirements coupled with acceptability criteria, credit limits, counterparty approvals, corporate actions, and intraday fee and rebate levels, the business could not possibly function. Again, imagine undertaking any, let alone all of these tasks, without any electronic assistance. Our businesses would simply not run without advanced automation, let alone run efficiently and within the requirements of the regulations that govern it.
‘Big data’ is an overused phrase, but one that is commonly used to describe the enormous amounts of new data entering our daily lives. However, here we are using ‘big data’ as an indicator of its importance, the fact that it has taken up a big part of our daily tasks, and without it, our systems would not run, and our industry would not function. Being able to do more with less is a common refrain around big companies, and the reality of that is that automated tasks often mean less people employed. Here, the rise of the machines is often not a welcome advance for some, just as self-driving vehicles seem highly attractive to many, cab and lorry drivers will have a different view.
Automation, in this sense, is clearly not just about undertaking enormous and complicated mathematical challenges that are simply beyond human capabilities in the timescales allowed, but it is also about efficiency. Clients expect services of better quality, higher accuracy and, of course, that could be delivered faster while getting cheaper. Witness the rise of zero-fee investment funds; that kind of offering is simply not possible without high levels of automation and even ‘robot advisors’ to take the place of expensive humans. Such offerings are also likely not possible without securities lending incomes to boost their returns and pay the managers fees, but that is another story altogether.
Efficiency is as much about financial capital as it is about human. Capital adequacy and the returns on capital employed are also vital measurements that create complex inputs into the daily process of managing a securities finance business. Complex and potentially conflicting pressures to provide the best, read cheapest, collateral to satisfy the client and the regulator at the same time, require iterative calculations, constantly rebalancing as demands change minute to minute. The demands of regulations don’t stop there of course. This year is bringing in increased demands for collateral to be posted on uncleared derivatives, placing additional volume and complexity through our allocation algorithms, and by April 2020, all our businesses will need to be able to marshal enormous amounts of data to enable them to tell their regulators all about it.
While science fiction has represented the machines as a potential danger to humans, the reality is that without them, and the data that feeds the automation they provide, we simply could not function. Whether in borrowing that share at the right rate, or booking the right restaurant, the machines are not only here to stay; we need them.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
