Читать книгу Real-Time Risk - Aldridge Irene - Страница 6

Have you ever missed opportunities in the markets because you felt you were disrupted? We have been in a unique and fortunate position to be immersed in the heart of fintech innovation and to observe first‐hand the extent of what is becoming a true disruption to businesses that, in turn, disrupted financial markets in the late 1970s and 1980s. Think of this as Finance 3.0. The possibilities are endless, and the new players are already embedded in most facets of traditional finance. These new players are not boiler rooms – most founders have advanced degrees and the most recent scientific innovations at their fingertips.

According to the Conference Board, investment in financial technology, trendily abbreviated into fintech, grew by 201 percent in 2014 around the world. In comparison, overall venture capital investments have only grown by 63 percent. The digital revolution is well underway for banks, asset managers, and customers. The impact on the financial institutions from the many startups that are trying unproven ideas is beginning to crystallize. Venture capitalists are betting that the once‐stodgy financial industry is about to experience a considerable transformation.

The pace of change for the financial world is speeding up, and startups and venture capitalists are hardly alone in the fintech craze. Apple, Amazon, and Google, among others, have already launched financial services platforms. They have aimed at niches where they can establish a strong position. Threatened by these new entrants, traditional financial stalwarts are hearing the pitch: Adapt to the new environment or perish.

Banks are launching their own internal funds and hiring significant numbers of developers for internal builds. Why now? In his latest annual letter to shareholders, Jamie Dimon, CEO of JPMorgan Chase, wrote that “Silicon Valley is coming.” While this statement went unnoticed by the news, it reflects the torrent of venture capital flowing into fintech. Estimates by the Economist, shown in Figure 1.1, suggest that 2014 was the watershed year for fintech startups.

Figure 1.1 Global fintech investment

Source: Economist, May 19, 2015.

The Current State of Big Data Finance

What is big data finance? For many financial practitioners, big data is still just a buzzword, and finance is business as usual. However, looking at the hottest‐financed areas of business, one uncovers particular trends that move beyond buzz into billion‐dollar investments. According to Informilo.com, for instance, the fastest‐growing areas of big data in finance in 2015 were:

● Payment services

● Online loans

● Automated investing

● Data analytics

Each of these areas, in turn, translates into automation. The payment services businesses, such as TransferWise, harness technology to commoditize counterparty risk computations. Counterparty risk is a risk of payment default by a money‐sending party. Some 20 years ago, counterparty risk was managed by human traders, and all settlements took at least three business days to complete, as multiple levels of verification and extensive paper trails were required to ensure that transactions indeed took place as reported. Fast‐forward to today, and ultra‐fast technology enables transfer and confirmation of payments in just a few seconds, fueling a growing market for cashless transactions.

Similarly, the loan markets used to demand labor‐intensive operations. Just 10 years ago, the creditworthiness of a bank's business borrowers were often judged during a round of golf and drinks with the company's executives. Of course, quantitative credit‐rating models such as the one by Edward Altman of New York University have proved invariably superior for predicting defaults over most human experts, enabling faster online loan approvals. Online loan firms now harness these quantitative credit‐modeling approaches to produce fast, reliable estimates of credit risk and to determine the appropriate loan pricing.

Can anyone issue loans over the Internet or facilitate payments? According to recent industry reports, yes, the founders of many loan startups that originated during the credit squeeze of 2009 – have little prior background in lending.

The key issues in lending are (1) having capital to lend, and (2) estimating credit risk of the borrowers correctly. The pricing of the loan service, interest, is then a function of the credit rating. If and when a borrower defaults, the loan should be optimally paid out from the interest. More generally, the average loan interest should exceed the average loan amount outstanding in order for the lender to make money.

The lending business is central to banking, and banks have had a near monopoly over the lending business for a very long time. New approaches to lending have emerged that compete with banks. Banks fund loans with deposits, whereas peer‐to‐peer lending is funded by investors. The leading players in this new approach to lending are the LendingClub and Prosper in the United States and Funding Circle and Zopa in the United Kingdom. In 2015, Zopa passed the Great Britain pound (GBP) 1 billion mark. Zopa's growth is shown in Figure 1.2.

Figure 1.2 Zopa originations by month

Source: p2p‐banking.com

With peer‐to‐peer lenders prospering with their new model, not only have banks noticed, but in some cases, started to acquire the upstart companies. SunTrust Bank acquired FirstAgain in 2012, later rebranding it LightStream.

New technologies are making their presence felt in wealth management as well. The topics of the robo‐advising and a broad group of analytics are the most diverse and least exact. Robo‐advising takes over the job of traditional portfolio management. The idea behind robo‐advising is that a computer, programmed with algorithms, is capable of delivering portfolio‐optimized solutions faster, cheaper, and at least as good as its human counterparts, portfolio managers. Given a selected input of parameters to determine the customer's risk aversion and other preferences (say, the customer's life stage and philosophical aversion to selected stocks), the computer then outputs an investing plan that is optimal at that moment.

Automation of investment advice enables fast market‐risk estimation and the associated custom portfolio management. For example, investors of all stripes can now choose to forgo expensive money managers in favor of investing platforms such as Motif Investing. For as little as $9.95, investors can buy baskets of ETFs preselected on the basis of particular themes. Companies such as AbleMarkets.com offer real‐time risk evaluation of markets, aiding the judgment of market‐making and execution traders with real‐time inferences from the market data, including the proportion of high‐frequency traders and institutional investors present in the markets at any given time.

Not only are the changes aimed at managing the portfolios of the retail investor but also in the way companies are raising capital from these same investors. Crowdfunding has become a popular way for ideas to turn into projects with real funding. Kickstarter is one of the more popular sites.

And companies like Acuity Trading, Selerity, and iSentium are trying to harness data from platforms like Twitter to give an indication of investor “sentiment,” which, in turn, gives them an idea of which way to trade.

The information‐driven revolution is changing more than the investing habits of individuals. Institutional investors are increasingly subscribing to big data information sources, the more uncommon or uncorrelated is the data source, the more valuable it is. Each data source then drives a small profit in market allocations, and, when combined, all of the data sources deliver meaningful profitability to the data acquirers. This uncommon‐information model of institutional investing has become known as Smart Beta or the Two Sigma model, after the hedge fund that grew 400 % in just three years after the model adoption.

Underlying all these developments are the advances in scalable architecture and data management. Ultra‐fast computation and data processing are critical enablers of other innovative forms of financial research and investing. Several companies have lately generated multibillion‐dollar valuations by providing analytics in the software‐as‐a‐service (SaaS, pronounced “sass”). For instance, Kensho is delivering the power of human‐language queries in customers' data, which have been rolled out across Goldman Sachs.

Risk managers face a daunting challenge. Finding a risk event is the needle in a haystack. With automation and big data, the haystack becomes a mountain, and that mountain is virtual. The potential to catch issues could never have been stronger, but the ways of doing so are drastically novel.