The Evolving Role of Technology in Financial Services - Essay Example

THE IMPACT OF ALGORITHMIC TRADING AND DERIVATIVE MARKET Customer Inserts His/her Name Customer Inserts Grade Course Customer Inserts Tutor’s Name 9/05/2011 Since the advent of the earliest adding machines and mainframe computers, technology, and the innovations it has helped to bring about, has played an increasingly important role in the evolution of the financial services industry (Fabozzi, 2008). Traditionally, the pace and impact of these technological advances has been gauged using easily quantifiable metrics such as storage capacity and processing speed (Goldberg, 2008). Today, however, when it comes to the application and potential of technology in the financial services industry, we’re witnessing the emergence of a number of new and rapidly accelerating trends that promise to usher in an entirely new paradigm — one in which Information Technology (IT) is not simply an “add-on” at the periphery of the business function, but rather deeply embedded at its very core (Schapiro, 2010). The lines between what we’ve traditionally viewed as the “technology” part of an organization and the rest of the business are blurring like never before (Pezzutti , 2008). The implication of convergences on the business cannot be ignored by the players in the industry (Fabozzi, 2008). In response to clients’ demands for more and faster information, greater transparency and improved risk management, providers are applying the vast computing power at their disposal toward an increasingly complex, sophisticated and integrated array of tasks (Kendall Kim, 2007). This has enabled the investor to instantly retrieve the details they require at any time. They’re able to consider billions of scenarios for millions of investment positions (Gregoriou Greg, 2010). And the models they’re considering are being built by an entirely new breed of financial practitioner one who possesses a keen understanding not only of critical business processes, but also of the technology that drives them (Morgan, 2009). And underlying all of these changes is a fundamental shift in the industry’s perception of the intrinsic value of the raw data itself (Fabozzi, 2008). Rather than a mere commodity, this data is increasingly viewed as the invaluable business asset it truly is (Corcoran, 2007). The evolving role of technology in financial services That gives it exponentially greater value in the eyes of our clients, allowing them to make more informed investment decisions than ever before (Fabozzi, 2008). Looking ahead, continuing advances in technology will allow the financial services industry to deploy increasingly sophisticated, forward-looking analytics to help clients make more informed investing decisions (Corcoran, 2007). Even at their most detailed, the financial reports of today can only provide the industry with a glimpse in the rearview mirror. The financial reports of tomorrow, however, promise to help the industry better understand the actual precursors of performance (Fabozzi, 2008). In the not-too-distant future, rather than simply providing clients with a simple description of their risk position, we will be able to provide them with detailed insights into the actual factors contributing to those risk positions (Corcoran, 2007). The implications of this shift cannot be overemphasized, as they will have reverberating effects on the habits, business processes and decision-making processes of institutional investors around the globe (Skinner, 2007). Another critical advancement fueling innovation in the industry is the deployment of cloud computing (Corcoran, 2007). Although cloud computing is not new, and has a well-established track record in other industries, it is just beginning to take hold in financial services (Fabozzi, 2008). Operating in a cloud environment brings a range of client benefits, from automation and capacity on demand, to accelerated time to market, real-time data infrastructure and strengthened client service (Corcoran, 2007). Cloud environments are also advanced platforms for product and service innovation, including custom analytics and data, as well as risk and control, performance, compliance and advisory services (Corcoran, 2007) Importantly, working in a private cloud environment also ensures data security (Fabozzi, 2008). In this Vision report, we examine technology’s increasing role in the financial services industry by bringing together the expertise of four of our executives across our investment servicing, investment research and trading, and investment management businesses (Corcoran, 2007). The report spans three distinct areas of the use of technology in financial services: analytics, electronic trading and regulation, and portfolio allocation and modeling (Skinner, 2007). William Pryor is senior vice president and head of State Street Investment Analytics (SSIA), which provides our clients services including performance measurement, attribution, universe comparison, ex-post and ex-ante risk, transaction analytics, compliance, information delivery and data warehousing (Fabozzi, 2008). In “Technology with a Purpose: The Next Generation Today,” he describes asset managers and asset owners’ increasing need for more detailed portfolio analytics, process transparency, risk management systems and dashboards to improve the kind of information they are receiving and their access to it (Corcoran, 2007). He explains that by integrating risk and return technology, investment service providers can give their clients the information they need to invest successfully and better manage their portfolios (Goldberg, 2008). Full-Scale Optimization Portfolio optimization involves determining the allocation among available assets that maximizes an investor’s utility (Fabozzi, 2008). The specific characteristics of the utility function and of the asset returns often interact to yield a computationally challenging maximization problem (Skinner, 2007). Lack of computing power in 1952 made simplifying assumptions unavoidable in Nobel laureate Harry Markowitz’s solution to this maximization problem — his now well-known mean variance optimization (Schapiro, 2010). Traditional asset allocation methodologies based on mean variance optimization are adequate if at least one of two conditions holds: either asset returns are distributed normally or investor utility is not affected by non-normality (asymmetries and fat tails) in returns (Fabozzi, 2008). The global financial crisis reminded us that, in fact, investors are much more sensitive to downside deviations than upside deviations and return distributions can be highly non-normal(Corcoran, 2007). Fortunately, advances in computer processing have made the original maximization problem tractable, rendering Markowitz’s simplifying (and often unrealistic) assumptions unnecessary (Skinner, 2007). Investors can now use full-scale optimization to calculate portfolio utility for every period in a sample, considering as many asset mixes as necessary to identify allocations that yield the highest expected utility for their particular utility function (Fabozzi, 2008). The computational challenge of this type of numerical search procedure is substantial, but modern computing capabilities permit comprehensive analysis of thousands of investments and millions — or even billions — of hypothetical investment scenarios (Pezzutti , 2008). While mean variance optimization only takes into account expected returns, volatilities and correlations, full-scale optimization implicitly takes into account all features of the return distributions, including skewness, fat tails and correlation asymmetries, thereby better reflecting the inherent risks (Corcoran, 2007). This approach may be particularly appropriate for investors with an aversion to losses below a specified threshold — for example, those facing reserve requirements, loan covenants, or the risk of insolvency or termination (Fabozzi, 2008). For these investors, a kinked utility function, rather than a quadratic utility function, best describes their concern for breaching a particular threshold (Schapiro, 2010). Their aversion to losses larger than that critical threshold is very high (hence the utility function is steep), while they are much less concerned about losses smaller than the threshold (so the utility function abruptly becomes much less steep). The management of this tail risk is critical as investors are deploying a larger proportion of assets to hedge funds, private equity and other alternative investments, which generate non-normal return distributions (Kendall Kim, 2007). Optimal Rebalancing Once optimal allocations have been determined, whether by traditional mean variance optimization or by full-scale optimization, those carefully chosen portfolio weights begin to drift almost immediately as asset prices change (Gregoriou Greg, 2010). Investors face a dilemma: as soon as they implement an optimal allocation, the portfolio’s various components gain or lose value, rendering the portfolio suboptimal 7 In a perfect world without transaction costs, investors would simply set up a trading algorithm to continually rebalance the portfolio to the optimal weights (Sussex, Morgan, 2009). But trading costs are substantial and they vary in accordance with the type of security being traded, the size of the position, and the time and place of trade execution (Fabozzi, 2008). At the other extreme, investors could save on transaction costs by never rebalancing the portfolio; that no one does this suggests that there is a cost — albeit an implicit one — associated with deviating from the optimal weights.8 Traditional portfolio rebalancing methodologies, such as calendar-based strategies that periodically rebalance to target allocations, and tolerance-band approaches, which trigger trades upon breach of predetermined thresholds, are relatively easy to implement and are certainly preferable to both no rebalancing and continual rebalancing (Fabozzi, 2008). However, a new approach using dynamic programming allows an investor not only to explicitly weigh the tradeoff between sub optimality costs and transaction costs but also to account for the fact that a rebalancing decision made today affects the rebalancing decisions available at future times (Fabozzi, 2008). This optimal rebalancing approach uses multi-period optimization technology to generate trading rules for a specified time horizon.9 using massive parallel processing to drive dynamic programming, optimal rebalancing is based on an algorithm that creates a roadmap of rebalancing decisions (Corcoran, 2007). Any number of asset allocation scenarios may be considered, though the computational challenge rises sharply with the number of variables used in the model (Corcoran, 2007). This computational challenge — involving the consideration of literally billions of portfolios — is called the curse of dimensionality (Corcoran, 2007). On a regular workstation, it would be impossible to undertake the number of calculations needed to map all of the possibilities for a portfolio of just a few assets (Fabozzi, 2008). Even using 28-processor grid computing and parallel processing to speed up complex computations does not supply enough power to determine an optimal rebalancing schedule for a portfolio with as few as 10 assets (Skinner, 2007). Fortunately, Markowitz and quantitative investment manager Eric van Dijk created an algorithm that reduces the computational complexity of the problem and makes rebalancing solutions feasible for up to 100 assets 10 Dynamic programming (in conjunction with the Markowitz and van Dijk improvement) substantially reduces rebalancing costs compared to the simple heuristics typically employed by investors such as monthly, quarterly or semi-annually rebalancing and tolerance bands ranging from 1 to 5 percent 11 For example, an investor with a $1 billion portfolio allocated among four assets could save roughly $600,000 n trading costs by employing dynamic programming to determine his rebalancing schedule as opposed to using calendar or range-based heuristics (Skinner, 2007). Reference: Chris Skinner, (2007). The future of banking in a globalised world. John Wiley and Sons. Clive M. Corcoran, (2007). Long/short market dynamics: trading strategies for today's markets. John Wiley and Sons. Frank J. Fabozzi, (2008). Handbook of Finance: Financial markets and instruments. Wiley. Greg N. Gregoriou, N Gregoriou Greg, (2010). The Handbook of Trading: Strategies for Navigating and Profiting from Currency, Bond, and Stock Markets. McGraw-Hill Professional. John Sussex, Joe Morgan, (2009). Day One Trader: A Liffe Story. John Wiley and Sons. Kendall Kim (2007). Electronic and algorithmic trading technology: the complete guide. Academic Press. Mary L. Schapiro, (2010). Examining the Causes and Lessons of the May 6th Market Plunge: Congressional Testimony. DIANE Publishing. Paolo Pezzutti , (2008). Trading the US Markets: A Comprehensive Guide to US Markets for International Traders and Investors. Harriman House Limited. Richard S. Goldberg, (2008). The battle for Wall Street: behind the lines in the struggle that pushed an industry into turmoil. John Wiley and Sons. Read More