Blog. Part 3: In God We Trust – Others Bring Data (Anonymous)
PublishedIf you think that big data and business intelligence are modern terms you need to update your definition of modernity to include 19th century. It was then when a banker named Henry Furnese got an advantage over his colleagues by analyzing available information in a structured manner spearheading data analysis for commercial purposes. Fifteen years later the company that is now known as IBM was founded in the USA – data was now officially useful and money could be made from it. Like any commercial commodity, once found useful – people began to collect and store it leading to exponential growth in its volume.
The lesser-known generalization from the 1955 Parkinson’s Law is a notion that data expands to fill all available storage space. Back in 1944 research speculated that American university libraries double in size every 16 years and by 2040 Yale library will have 6,000 miles of shelves (roughly the distance from London to Hong Kong). Unfortunately for librarians the world invented digitalization and in 1991 the internet. By 1996 it became more cost-effective to hold data digitally than on paper.
Just like Henry Furnese in the 19th century, financial companies were early to the big data party. Paris and London had their financial open outcry (lots of men in a pit shouting and waving contracts trying to strike deals) exchanges in early 1800s. Julius Reuters (the founder of Reuters part of Thompson-Reuters news agency) used a combination of telegraph cables and carrier pigeons to run a successful news delivery system in 1850. First digital stock quote delivery system was in place nine years before NASA sent a man to the moon –banking was a natural home for digitalisation and offered generous returns. Slowly, the rest of the world caught up.
As processing powers increased, and prices for data storage and transmission fell people who could understand and interpret it in the meaningful way got commercial advantage and made quite a lot of money. Generally people rather like to make money, so the race for the cutting edge of data mining and trading got faster and increased in intensity. Algorithmic trading that executes thousands of small trades on electronic exchanges very quickly based on previously analysed data was established by early 2000s and grew exponentially ever since.
Credit risks (potential losses to a bank if some (or several) customers default) relied on collecting holdings data and analyzing exposures – that job was made more and more difficult by the increasing complexity of the financial markets and sophistication and speed of trading. That interrelated web of financial entanglement finally collapsed on itself in 2008 leading to 10 years of economic pain to a lot of people regardless whether they were close to financial markets or not.
People losing money and business going under is not a popular situation for any government when potential voters ask them a reasonable question of what was done to prevent the free trading Formula 1 machine crashing into the barrier. The irony of any regulator is that they are always preventing the previous crisis – predicting a new one is a thankless task, as media will remember all your wrong calls.
Nevertheless, the regulators have now come to the financial big data party. Bank of England and the UK's Financial Conduct Authority announced their first digital regulation vision in 2013 with the idea that they should have enough of good quality data delivered to them by market players to run their own prediction algorithms and using prior knowledge predict (and prevent) a new market crash.
In November 2017 both BofE and FCA held a TechSprint aimed at exploring how technology can be used to make regulatory reporting more accurate, efficient and consistent. The idea is that regulatory reporting requirements for firms will increase in quality, frequency, and complexity to allow for sophisticated analysis. Bad news for firms having to pay for enhanced reporting burden if they want to continue trading (post-crisis fines have now exceeded USD 200bn), but good news for tech entrepreneurs who saw an opportunity to bring sleek modern technology to the marketplace.
RegTech as it is known works on helping all parties to standardize data, to simplify arcane codes, and to build real-time connections for the industry. The trend is to use technology to bring transparency to the market, to understand the risks and to predict the shocks before anyone gets hurt. Big banks can no longer submit a hand-written letter with a seal and a signature confirming that they comply with all the rules – they have to show compliance by sharing vast quantities of increasingly standardized data signed with their unique Legal Entity Identifier, using ISO codes for currencies, products, and transactions. If history shows us anything, where financial services lead, others eventually follow, so the trend for digitalization of online entities and eventually identities are set to continue.