April 2000 Banking on Analysis Mathematical models make good business at Tel Aviv's Bank Hapoalim By Mordecai Greenberg When you walk down the third-floor corridor at the headquarters of Bank Hapoalim — Israel's leading bank — in Tel Aviv, the sights and sounds in the rooms do not resemble your average bank office. Mathematical formulas and graphs are scribbled on white marker boards and bank officers carry on a discussion on the solution of stochastic differential equations. These bank officers hold advanced degrees in operations research, mathematics, physics and computer science. It's not a university lounge, but rather the bank's Department for Analytic Development. Don't let that academic facade mislead you. Those are real bankers solving real banking problems, using "state-of-the-art" mathematical models for appraisal and development of financial products. It's not your traditional, conservative banking, yet it's an inventive and innovative one. It is here that modern finance theory relying on mathematical models meets the world of banking and assists in making better decisions. Bank Hapoalim Goes High-Tech American and European commercial and investment banks began aggressively exploring new ground in the 1980s by hiring people with advanced degrees in the quantitative sciences (mathematics, physics, engineering). The revolution they have introduced into banking and the financial world can be best appreciated by viewing the widespread use of financial derivatives for hedging and speculative purposes. Derivatives are financial products that derive their value from the value of an underlying asset, such as a physical commodity, foreign exchange or plain stock. The risks associated with unexpected changes in the market price of the underlying assets can be offset by including appropriate derivatives in one's portfolio. The pricing of derivatives and the optimal management of portfolios that contain derivatives require complex mathematical models and computation methods. In the 1980s, Israeli banking institutions were fighting to survive inflation, which reached an unprecedented level of hundreds of percentages a year, in an economy tightly controlled by the government. The day-to-day hectic economic environment, aggravated by constant political conflicts (internal and external), kept Israeli banks behind their overseas counterparts with little or no room for a technological revolution. The 1990s saw inflation tempered and political turmoil subside. Government control over the economy loosened. Liberalization was the key word. Laws and regulations, especially those concerned with foreign currency, investments in overseas markets and the presence of banks from abroad in the local market, were abandoned for a more liberal government policy. Thus, in 1995, there was some lost ground to recover. Bank Hapoalim made a strategic decision, which, in its turn, brought about the introduction of analytic methods such as OR, statistics and probability theory, to banking business. The bank embarked on a major reorganization program based upon a customer segmentation strategy and a realignment of its divisions according to customer types. New divisions were formed, while some existing ones were made redundant. Putting to use what had already been explored and tested in practice by American and European institutions, the bank also reached a decision to upgrade the financial products offered to customers and to better manage risk exposure. These tasks, as was already mentioned, demanded that the bank harness the power of analytic and computational methods. Consequently, the bank decided to establish an operations research-oriented Department for Analytic Development. Searching the Ivory Tower Looking for the right person to pioneer such a venture wasn't that straightforward at first. The existing bank personnel did not offer a viable option for the mission. The search committee was familiar with the practice of known Wall Street firms and investment banks to look in the academic world for professors who can translate finance theory to the everyday functioning of foreign exchange, portfolio management and the like. Operations research as a discipline seemed to be the background of choice of many successful finance professionals, including Nobel Prize winners Harry Markowitz and Bill Sharpe. Many professors found their way to executive positions in banking and quite successfully so. Some kept their academic appointments and some left the ivory tower altogether. The members of the search committee had their work cut out for them. Mordecai Avriel, who held the operations research chair at the Technion-Israel Institute of Technology and who gained past experience in U.S. and Israeli financial institutions, was an obvious candidate. Professor Avriel accepted the challenge to initiate the introduction of advanced mathematical techniques to Bank Hapoalim's financial decision-making processes. The professed mandate given to the new department was "to find and identify those finance areas within the bank's operations that could benefit from quantitative methods, develop the appropriate mathematical models and computational tools, and follow their implementation at the user's level." One telling indication of the importance the bank attached to the new department: Avriel reported directly to the member of the bank's Management Board in charge of finance. The management of the bank anticipated a long introductory phase for the new department. Professor Avriel was expected to sit in his offices with little to do at first, while the bank's establishment gradually learned the new tools and technologies that became available. In reality, Avriel barely had time to take off his jacket, let alone settle into his new offices, before phone calls started flowing in from different units of the bank expressing curiosity and eagerness to learn and employ the new tools. Recruiting staff was not an easy task either. The search for Israeli candidates with scientific and computational backgrounds as well as financial/banking expertise was long and strenuous, but it led eventually to the creation of a small, highly skilled "commando" team of dedicated and very able people. Staff member Alex Arensburg, for example, has a Ph.D. in physics plus an MBA in finance. He abandoned a physics research position working on free electron lasers to join the department. Naomi Shpirer has a Master's degree in OR from Technion, supervised by Avriel. They have a number of joint publications on the efficient stowage of containers in container ships, the subject of her M.Sc. thesis. She is currently an external Ph.D. student in finance at Tel Aviv University. Avi Peretz has an MBA in finance and was recruited to the department following his participation in an internal managerial training program. Alon Raviv came from the research department of the Bank of Israel and is also an external Ph.D. student in finance. The most recent recruit to the department is Iddo Eliazar, arriving from the Statistical Laboratory of Cambridge University in the United Kingdom. He holds a Master's degree in OR and a Ph.D. in stochastic analysis. Today, the staff of the department has its hands full now that virtually all of the bank units have learned the benefits that can be achieved using their services. Good Banking Though inflation has subsided, it remains an important factor to be considered in Israeli economical behavior. The first project undertaken by the Department of Analytic Development was to deal with the inflation problem and come up with advanced solutions. How do you report your business activities in an inflationary environment? How do you read a company's annual balance sheet or profit/loss report, if monetary values are eroded throughout the year? The Israeli accounting system decades ago adopted CPI (Consumer Price Index) linkage for financial reporting purposes, and the government regularly issues unique financial instruments - CPI-linked treasury bonds - are traded on the Tel Aviv stock exchange. These relatively long-term, interest-bearing bonds provide (almost) full protection against inflation, if held to maturity. Incidentally, the U.S. market became familiar with similar instruments only by 1997, when the federal government came up with Treasury Inflation Protection Securities or TIPS. Although the Israeli instruments are linked to the CPI (published once a month), the market price of the bonds, or their yield, fluctuate upon trading, reflecting supply/demand conditions based mainly on inflationary expectations. Suppose that an investor was interested in having a tradable financial asset, whose value at any given time would be linked to the CPI, bearing in mind that the CPI itself is not tradable. The bank, in its turn, was interested in investing a portion of its financial equity capital in just such an instrument. The department was asked to recommend a portfolio consisting of publicly traded instruments, whose value would follow the value of the CPI. Solving the problem required a modification of Harry Markowitz's mean-variance portfolio optimization model. The CPI model is a nonlinear program and in order to solve it, the department purchased its first software package, the GAMS/MINOS system. The optimal solution obtained from the system contained CPI-linked and unlinked instruments; it was very different from the investment strategy that had been used by the bank at that time. Professor Avriel was quite concerned about how the new strategy recommendations would be accepted by the top management. He regarded the presentation of the CPI model results as a test case for the future. The case was first presented to the bank's Asset and Liability Committee, which is in charge of implementing the bank's investment strategy. The committee consists of finance professionals with many years of hands-on experience in the Israeli fixed income and money markets. In spite of the differences between the existing and proposed strategies, the committee unanimously approved the latter. Several members of the committee expressed their contentment with the introduction of analytical tools, such as optimization models, into the bank's everyday investment decision process. Having passed the first hurdle, the proposed strategy was then brought before the bank's Board of Directors. An important feature of the new strategy was that it involved lower risks than the existing one. The Board was especially pleased with this aspect and passed a resolution to modify the bank's investment policies regarding its equity capital. The lack of opposition from the bank's top management to the new ideas introduced by the Department of Analytic Development in their first project was a very encouraging sign that the bank's mind was open for innovations. Asset Allocation Soon after the completion of the CPI project, it occurred to Avriel that the methodology used for that project could be also applied to other investment strategies in the bank. An example could be the investment portfolios of wealthy customers, those who are serviced by the Private Banking units. Coincidentally, exactly the same thought occurred to one of the members of the bank's Board of Management in charge of investment services, who read the report of the CPI project. The two got together and on the spot decided to develop an investment advice system for customers of private banking, a system based on modern portfolio theory. Retail banking in Israel is a very competitive business. While creating the Private Banking units five years ago, Bank Hapoalim defined the units' mission as delivering the best and most innovative methodology and technology for investment advice - a service unavailable at competitor banks. A key aspect of this mission is the role played by the customer relations manager (CRM), a highly trained bank officer and certified investment advisor, who gives personal and individual solutions to every banking problem of the customer. In particular, the CRM advises private banking customers in their investment planning. Whereas principles of diversification and portfolio selection reflecting risk attitude were already employed by the CRMs, no quantitative tools were available for providing systematic advice. The development of the investment advice system, an asset allocation decision support system called OptiMoneyTM, was a joint effort of the department and the Securities Division. Asset allocation is the process of selecting the proportions of assets in a portfolio so that a specified performance target is best achieved. Such a target may be the optimal combination of portfolio risk and expected return on the portfolio. The optimal combination depends on, among other factors, the investment horizon, the investor's risk aversion and benchmark selection, as well as tax and liquidity considerations. In OptiMoney, risk can be defined in terms of return variance, symmetric or downside deviations from the return performance of a benchmark or a combination of several benchmarks. The optimization is carried out within the framework of a nonlinear mathematical programming model of a few hundred variables and constraints by using the GAMS/MINOS system. The asset universe in the model consists of indexes of instruments traded in Israel and abroad, along with non-tradable savings and investment products offered by the bank. The available benchmarks are the CPI, the exchange rate of the U.S. dollar, and the short-term interest rate offered by the Bank of Israel. Inputs to the model are raw historical performance data of the assets, expected returns and client-specific constraints. The output of the model is an individualized asset allocation. It should be emphasized that OptiMoney does not replace the advisory aspect of the CRM's work; rather it supports the advisor by recommending optimal portfolios on the asset class level. The CRM is still responsible for providing the detailed composition within the asset classes; thereby his/her investment preferences are retained. The bank strongly believes in such a combination of formal and creative approaches to investment practice. As of the early part of 1998, OptiMoney became operational accompanied by an aggressive advertising campaign in the electronic and written media. Conservative estimates suggest that in its first year of operation it brought $10 million worth of new business into the Private Banking unit alone, without considering the added value of marketing appeal. Exotic Products Finance textbooks classify derivatives as "plain vanilla" (i.e. simple) or "exotic" (i.e. not so simple). Within these classes there are further subdivisions. The valuation of exotic derivatives and the measurement of the risks involved in buying or selling them require mathematical models of complicated stochastic processes. As financial markets become mature, more and more exotic products become available to suit the specific needs of customers. The Department of Analytic Development has worked on a number of exotic product models, some of which were invented within the department. For example, suppose you have a portfolio of corporate bonds. The yield on these bonds may provide you with an attractive income, but there is a risk that some time in the future one or more of these corporations will default. In this unfortunate case they will be unable to pay coupons or repay the principal. Financial institutions offer now a derivative product that will compensate the investor in case the corporation is unable to fulfill its obligations to the bond owners. This product is called a credit derivative. A more advanced type of credit derivative provides insurance against default of a "basket" of corporations — that is, it will pay compensation in the first instance that one of the corporations in the basket defaults (and only in the first instance). Figuring out the premium (the fair price) of such an instrument involving firms from different sectors and/or different countries can be quite difficult. Since the bank is involved in this line of business, the department was asked to provide price quotes for this type of exotic derivatives. Another example of an exotic product the bank is offering is a U.S. dollar-denominated savings plan, in which the customer receives interest at double the regular rate, provided that during the term of the plan (say, six months) the fluctuating Israeli Shekel/U.S. Dollar exchange rate never reaches or crosses some predetermined upper and lower bounds. Otherwise, the customer receives the principal at the end of the plan. The department built a model that determines the actual levels of the bounds, as well as the optimal hedging policy against the risk involved in offering this product. Note that the bank's risk can be almost entirely eliminated, if it also offers the complementary product — namely, a savings plan that pays double interest if the exchange rate does reach or cross the same bounds. For every dollar invested in the first plan, there is a dollar in the complementary one. Today, both products are offered to the bank's customers. After having been in operation for more than four years, the Department of Analytic Development is more ready than ever before to pursue the technological revolution within Bank Hapoalim. True to the moment this article is being written, Bank Hapoalim is still the only bank in the country to employ an elite group of quantitative professionals. The Largest Bank in Israel Bank Hapoalim is the largest bank in Israel. The bank has more than 300 full-service branches, eight business centers and a corporate center for serving large corporate customers. Overseas, the bank has branches in North and Latin America, Europe and the Far East. In 1999, total consolidated assets reached 48 billion U.S. dollars and net return on equity was 13.5 percent. The bank's long-term rating is the highest possible as it matches the sovereign rating of the State of Israel. Bank Hapoalim holds an average of 30 percent of market share. In 1997, a group of private investors acquired a controlling share of the bank's stocks, following the government's decision to hand over control of the Israeli banks to the private sector. Mordecai Greenberg is a free-lance writer based in Tel Aviv. This article was written with the aid of staff of the Department of Analytic Development at Bank Hapoalim. Table of Contents OR/MS Today Home Page OR/MS Today copyright © 2000 by the Institute for Operations Research and the Management Sciences. All rights reserved. Lionheart Publishing, Inc. 506 Roswell Street, Suite 220, Marietta, GA 30060, USA Phone: 770-431-0867 | Fax: 770-432-6969 E-mail: lpi@lionhrtpub.com URL: http://www.lionhrtpub.com Web Site © Copyright 1999, 2000 by Lionheart Publishing, Inc. All rights reserved.