Working Paper 258
Academic Leave Proposal

Robert E. Jensen at Trinity University

Table of Contents

Preliminary Notes

Introduction

Project Objectives in Five Phases

Methodology

Current State of This Research

My Qualifications for This Research

Personal and Professional Enhancements from This Research

Glossary

Preliminary Note 1: Some colleagues who pre-screened this proposal urged me to define the technical terms. In order to not break up the reading flow, I have added a short Glossary of technical terms at the end of the proposal.

Preliminary Note 2: Some colleagues who pre-screened this proposal expressed concerns that I could complete all phases of this project in the leave period. I think this is possible. I received a Mellon Foundation Grant for summer support on this project. If I also get a Fall 1998 leave, the leave will enable me to spend full time on the project for about seven months and part time between now and the beginning of next summer. The first four phases will be completed by October 31, 1998 if I get this leave. I added proposed deadlines for each phase of the project. Papers can be generated from each phase along the way. Phase 5 is the most uncertain since it is very difficult to predict the completion of the most complex analytic phase of the project. I think I can complete this final phase before December 31, 1998.

Preliminary Note 3: Although I hope to generate at least one paper in each phase of this project, I should note that by accepting the Mellon Grand summer support for the project, I am obligated to make some of the Accounting Information Systems pedagogy innovations available free at my web site. This is not a problem since my new philosophy is to make almost everything I create immediately and freely available at my web site.

There would be numerous results from such a paradigmatic shift. By focusing on the data itself, rather than on compiling and aggregating information for users at the front end, the concept of “recognition” would assume different significance in accounting. Instead of focusing on recognition as a primary determinant of what is included in financial reports, there would be an emphasis on ensuring sufficient detail in disaggregated data bits to allow users a choice about the type of information most useful for their particular needs. Provided sufficiently discrete data, users could formulate their own financial statements based on their own particular “recognition” criteria.

S.M.H. Wallman, Commissioner of the Securities & Exchange Commission

“The Future of Accounting and Financial Reporting: `Access” Accounting,’

Accounting Horizons, Volume 11, Number 2, June 1997, p. 109

Introduction

My leave proposal topic concerns a new phenomenon that may shake up financial reporting standards and practices around the world. In the past, a major problem has been that there are different standards (rules, laws, measurement rules, etc.) for financial reporting in different nations. (See “Accounting Standards” in the glossary for details on accounting standard setting.) For example, SFAS 52 requirements for accounting for foreign currency transactions apply to U.S. firms, whereas foreign currency accounting standards differ for Canada, the United Kingdom, France, Germany, Japan, and other nations. The U.S. also has a problem in that our standards promulgated by the FASB and SECdiffer in significant ways from the international standards proposed by theInternational Standards Committee (IASC). The IASC has been making a largely futile effort to achieve greater uniformity in controversial financial reporting standards that entail specifications for aggregation into measures of financial performance. For example, the IASC proposes capitalizing research and development costs and estimated goodwill value as assets whereas U.S. firms are not allowed, underFinancial Accounting Standards Board (FASB) and the Securities and Exchange Commision (SEC)rules, to follow those IASC standards. Many other nations follow these IASC standards making it difficult to compare financial reports of foreign companies with U.S. firms. If companies had databases for such items on the Internet, investors could aggregate the outcomes themselves under either FASB or IASC rules.

In 1996, Microsoft Corporation made use of new technologies to put out Internet financial statements under multiple standards. By linking to http://www.microsoft.com/msft/ar97/financial/income.htm, it is possible to choose any one of five nation's standards you want to view the income statements of Microsoft Corporation. What this means is that for Microsoft Corporation it is possible to examine the sensitivity of financial performance to the measurement rules in effect. For example, the sensitivity of French rules versus U.S. rules versus Japanese rules can be compared with respect to net income of Microsoft Corporation in 1996. However, this is only the first step since the financial statements themselves are still highly aggregated by measurement rules of a given nation.

Advances in database technology make it possible, in theory, to totally change the traditional financial reporting systems in virtually all nations. In the future, it may be possible for corporations to simply allow investors to access networked databases containing relational tables. The investors could then choose from an array of alternative measurement rules (e.g., the current-rate versus the temporal-rate method of accounting for foreign currency transactions, a current value hedged-base accounting for interest rate swaps, etc.) and generate a custom set of financial statements for each investor's preferred combination of standards and measurement rules.

This idea extends well beyond differences in accounting rules between nations. Alternative bases for accounting aggregations in general could be chosen by investors if a corporation reported data in a "properly formed" relational database on the Internet. Such databases could contain nonfinancial performance data as well. The two primary areas of my leave proposals are outlined below:

1. My sabbatical leave proposal will center on the task of designing new types of relational databases for accounting for financial instruments derivatives (contracted options, futures, forwards, interest rate swaps, and foreign currency transactions). Unlike the Microsoft Corporation approach which presents financial statements under the reporting rules of five nations, I will develop a relational database approach that will allow investors to compare performance outcomes under hundreds of combinations of reporting rules. In order to narrow the scope of this one project, the focus initially will be on interest rate swap contracts (the predominant type of derivative contract around the world.) These swap contracts are largely used for hedges against future movements of interest rates.

2. My sabbatical leave proposal will entail deriving a way to value interest rate swap contracts that are not actively traded in financial markets. A large proportion of these contracts in practice are custom contracts for which the FASB and SEC have not yet discovered a way to value them for corporate reporting purposes. Interest rate swap contracts are largely behind the following “valuing at zero” complaint of the Chairman of the SEC:

1. The most recent skirmish (between the FASB and CEOs) centers on how companies account for derivatives.
   
2. We are in a situation in which the notional amount of derivatives outstanding has reached sum $70 trillion . . . Accounting for derivatives has simply failed to keep pace with their exponential growth . . . The value of derivatives is certainly not zero, although that is how many derivatives are reported today.

Arthur Levitt, Jr., Chairman of the SEC

Journal of Accountancy, August 1997, Page 11

3. My sabbatical leave proposal will center around developing what may be the first “Bot” in financial reporting and analysis. This research was inspired, in part, by the following statement from Steven Wallman:

These and other advances create new possibilities in information technology. They result in developments in areas such as “expert systems,” programs designed to employ the algorithms or logic of experts in various fields. “Smart or intelligent agents” (also referred to as “Bots”) carry out tasks such as scanning large databases (or Web sites) for particular information and acting on that information in accordance with specified instructions. “Neural sets” allow for advanced pattern recognition, enabling computers to “learn” from previous experiences by identifying patterns or foundations for investigation …

S.M.H. Wallman, Commissioner of the Securities & Exchange Commission

Accounting Horizons, Volume 11, Number 2, June 1997, p. 108

Project Objectives in Five Phases

Phase 1. To design a relational database framework for financial reporting of corporations to investors. (Scheduled completion by July 15, 1998)

Phase 2. To design and illustrate a relational database in Microsoft Access for reporting of financial instrument derivatives. Special focus will be placed upon interest rate swaps that are custom contracts and have no trading markets or markets that are too thin to be used for value estimation. (Scheduled completion by September 30, 1998)

Phase 3. To design a Structured Query Language (SQL) application for the database developed in Part 2 above and extend it into a web-based database application. Scheduled completion by October 15, 1998.

Phase 4. To model a value estimation means for estimating database values of derivative financial instruments such as customized contract interest rate swaps that are not traded or are traded in thin markets that do not provide valid estimates of current values. This part of the research uses analytical methodology from the mathematics of finance. Scheduled completion by October 31, 1998.

Phase 5. To model a Bot for evaluating derivative financial instruments risk. The model will probably be based upon the new Analytic Network Process (ANP). ANP is a network extension of the Analytic Hierarchy Process (AHP) which is a mathematical approach for modelling a decision process.

Methodology

The methodology of this project is largely analytical (mathematical modelling of investment risk and multiple criterion decisions involving interest rate swaps). Prior to this I must design relational database systems for interest rate swap contracts.

In Phase 2 of my project, a general structure for relational databases in financial reporting will be designed. There are many considerations in this design. For example, a major consideration is the development of restricted access “views.” Views allow users to create customized subsets of database tables that contain only the authorized records and fields of database tables. Called virtual tables, views of virtual tables can simplify access to database tables (called base tables). By including selected, insensitive columns or rows that users may "see", users can be prevented from access to information for which they have no need to know. Then, by restricting a group of users to a given view of the database, views can provide a type of database security known as discretionary access control. This is vital since corporate management will be using the same databases as the public in general. For many reasons, the public cannot be allowed to view all records and fields in the database.

In Phase 3 of the project, the methodology entails designing illustrative queries that investors might use when disaggregated relational databases of financial derivative contracts are made available by a business or non-profit organization holding such contracts.

In Phase 4, I will propose a means for valuing interest rate swaps (the major type of derivative financial instruments) that are not traded in markets sufficiently deep for market determination of values and risks. This phase of the project is partly completed as explained below. What remains is some fine tuning of the model and some testing with real-world contracts.

In Phase 5 of the project I will model an Analytic Hierarchy Process AHP Bot for scanning large disaggregated databases for “intelligent systems” evaluation of value and risk of interest rate swap contracts. Actually, AHP has now been extended into Analytic Network Processing (ANP) which I will use for my development of an ANP Bot. The latest books on AHP and ANP can be found at the http://ahp.net/home.htm web site.

As pointed out earlier, part of the research entails analytical methodology under the mathematics of AHP. This applies to estimating values of customized financial instruments derivative contracts that are not actively traded in viable markets.

The idea for this research came after being invited to discuss a paper on foreign currency financial reporting at a recent conference at Rutgers University. My discussion paper is available as Working Paper 255 under Document 18 at my web site at http://www.trinity.edu/~rjensen. Because I have been tracking emerging developments in relational database and SQL technologies, it dawned on me that the paper I was discussing at Rutgers University on empirical comparisons of different aggregation rules (in different countries) for reporting foreign currency could become obsolete issues under emerging database and networking technologies. In the case of foreign currency swaps and related types of financial instruments derivatives, it will be relatively costless in the future for corporations to make relational databases available that will allow investors to derive reporting measurement outcomes under any alternative rules they choose. For example, French investors could make SQL queries under accounting rules of France whereas Germans could use the same database to make German queries.

It is “relatively costless” when the data are already available in internal databases. For example, because foreign currency exchange rates are readily available at any point in time, corporations do not have to estimate market values with costly valuation appraisal systems. Providing market value data for customized contracts that are not actively traded presents more of a problem. My research will propose how to deal with this issue in certain types of contracts such as interest rate swaps that are now being used to finance over $70 trillion dollars worth of investments worldwide.

It is increasingly evident that the traditional process of choosing a single aggregation rule by the FASB will someday be a futile and obsolete task. If a proper disaggregated database containing past and current values is provided by a firm, investors can measure interest rate swap risk using alternative risk measurement reporting rules. In the future, the FASB should shift the focus to regulating what is to be made available in the relational database rather than selecting a single required accounting measurement rule.

Current State of This Research

Petrea Sandlin and I published a paper entitled “The Paradigm Shift in Technology: Financial Reporting Will Never the Same” in Research on Accounting Ethics, Volume 3, 1997, 191-202. That paper lays some of the foundation for my proposed project. We reviewed emerging technologies and proposed disaggregated database reporting of accounting information. We also analyzed reasons why investors want these databases and why corporate managers are hesitant to make such databases available at the moment. The SEC is currently studying whether to make corporations bring disaggregated databases online.

For the start of Phase 4 of the project I have an unpublished Working Paper 231 that lays the broad design of a model for valuing interest rate swap contracts. Working Paper 231 is entitled “Legal Settlement Exit Value Amortization Rate Accounting for Custom Interest Rate Swaps Having No Market Trading.” It can be viewed at my http://WWW.Trinity.edu/~rjensen/#Working web site. What remains is to put the model to actual tests on real-world interest rate swap contracts. Readers interested in the mathematical analysis should print this paper. Several points might be noted about my legal settlement discount (amortization) rates that in my viewpoint make it superior to the only other proposed valuation model in the literature. A paragraph from the abstract of Working Paper 231 reads as follows:

The major purpose of this paper is to propose a legal settlement exit value estimation of fair value. An amortization (discount) alternative is introduced using changes in fair value. It will be argued that the legal settlement rate is more appropriate for the purpose of booking interest rate swap receivables/payables on the balance sheet. This rate is more in line with recent recommendations of mark-to-market or fair value reporting of derivatives. It has the added advantage of societal symmetry between what companies report as swap receivables and what other companies report as swap payables. Present methods for computing these amounts have societal asymmetries. It is also shown how the legal settlement discount rates can be adjusted for changing term structures of interest rates.

Key advantages of the disaggregated relational databases that I hope to design are that investors could compare different valuation models. My proposed interest rate swap valuation model is only one of many that a given investor might dream up or modify from my basic design. The design of relational databases for interest rate swap contracts, to my knowledge, has never be undertaken by anyone. I am now teaching relational database theory and design in ACCT 5342 (Accounting Information Systems). I hope to apply my database and query design skills in phases 1 and 3 of this project.

For Phase 5 of the project, I might mention that a significant amount of my published research centers around mathematical modelling of decisions. I have published over 20 papers on this subject and have made numerous presentations at conferences and on college campuses (including a 1991 seminar presentation for the Mathematics Department at Trinity University). Saaty’s proposed extensions of AHP into Analytic Network Process (ANP) are new to me. This phase of the project is admittedly a fuzzy idea in my head at the moment. What caught my attention is Saaty’s sketch of an ANP approach for commodity markets that is provided in pp. 242-245 of his book (referenced in the glossary of this proposal). I envision fleshing out his creative ideas in financial instrument derivative markets. His sketches of paired comparison matrices for Capital Factors, International Factors, Industrial Output, Macro Economy, and Price seem to be adaptable to my research, although they would have to be modified for my topic areas.

Course materials for ACCT 5342 are linked at http://WWW.Trinity.edu/~rjensen/acct5342/index.htm. In the process of developing this course, it is becoming more and more evident that emerging relational database technologies and networking technologies in general have thrust the current state of AIS research and practice into utter chaos at the moment. I have adopted an on-line textbook for ACCT 5342. The title of the book is Accounting Information Systems: A Database Approach. The authors are U.S. Murthy and S.M. Groomer. Students access the book at http://www.cybertext.com. In Chapter 1 of that book, the authors make the following statement:

There are a number of problems that limit the usefulness of the traditional accounting model given the complexities of the global marketplace and the demands of present day information customers. Briefly, the shortcomings are as follows:

• "transactions" orientation as opposed to an "events" orientation,
• narrow focus on financial data,
• limited accessibility of information,
• too high a level of aggregation, and
• · limited flexibility which prevents answering queries that cross functional boundaries.

Virtually all accounting information systems are struggling with how to reengineer entire systems as networking and database technologies emerge. My research will try to make a contribution toward all of the above limitations of traditional accounting reporting systems presently in place. In particular, it will show how the level of aggregation can be efficiently and effectively lowered by making relational databases available to investors in place of aggregated financial statements.

My Qualifications for This Research

This year I will be teaching AIS and applications of Microsoft Access in building relational databases. This will give me a better background in relational database and SQL technologies. As far as financial instruments derivatives are concerned, I teach these in my ACCT 5341 course called International Accounting Theories. I also have written research papers on derivatives accounting and include my research in a multimedia show that I have delivered in whole or in part in over half of the 160 universities that have invited me in to present my technology work shops. My resume under Document 1 at my web site lists my research papers and campus presentations. I indicated previously that my Working Paper 231 (at http://WWW.Trinity.edu/~rjensen/#Working) provides an excellent foundation for Phase 4 of the project dealing with the valuation of interest rate swap contracts.

I also have built up some contacts with technical experts in AIS consulting for large international accounting firms. These persons can help me in identifying some of the key problems and possible solutions to issues that will arise in my project.

As indicated above, I have published a series of papers on the theory and application of modelling of decisions using the Analytic Hierarchy Process (AHP) first proposed by a mathematician named Thomas Saaty. AHP has inspired thousands of subsequent papers in virtually all academic disciplines. There are many documented applications. My publications on AHP are listed in my resume at http://www.trinity.edu/~rjensen/resume.htm.

Personal and Professional Enhancements from This Research

What I hope to personally achieve from this research is to be on the leading edge of emerging networking and database technologies in accountancy and accountancy education. I hope to inspire my students to look to the future trends of accountancy and see beyond traditions rooted in the mathematical writings of Luca Pacioli in 1494. Pacioli’s elegant algebraic accounting system is analogous to the theorems of Pathagoris that serve us well but are not sufficient for modern navigational systems in outer space. Similarly, under the globalization of business and capital flows and the complexities of international transactions, traditional accountancy is becoming increasingly inadequate. What a thrill it is to be on the leading edge of newer networking and relational database technologies.

In 1990, Professor William Beaver (Stanford University) stated the following in private correspondence to me:

While such questions are not easily resolved, they are considerably different from the issues the FASB now devotes major time and effort ... in attempting to resolve, such as trying to determine which set of accounting methods produces the "best" measure of net income. In the database setting, there would be less concern with how to calculate the "bottom line" and more concern with the nature of disclosure presented from which users can make their own calculations.

In 1990 the technologies were not yet developed for effective and efficient use of database technologies for external reporting to the general public. The World Wide Web commenced in 1990 with about 50 users. The recent Mars landing in July of 1997 generated over 100 million Jet Propulsion Lab hits in less than one week. Commerce is moving to the web at an unbelievable pace. Since 1990, the World Wide Web and relational database technologies have evolved to a point where the main obstacles are now our slowness in seeing how to use these technologies and our reluctance to quickly adapt to change. My research project is aimed at discovery of how these technologies can be better used in the reporting of both financial and nonfinancial data in relational databases.

Nobody to date has combined disaggregated accounting and finance relational databases with Analytic Network Processing extensions of AHP. I hope to be the first researcher to undertake such an effort.

I also hope to learn more about ANP and the underlying mathematics. I understand the eigenvector analysis in AHP. However, some of the mathematics of ANP will take time and effort for me to grasp. At the moment the key Step 9 of ANP outlined on Page 154 of Saaty’s book is a challenge. It reads:

Compute the limiting priorities of the supermatrix according to whether it is irreducible (primitive or imprimitive [cyclic]) or reducible with one being a simple or a multiple root.

My learning how to do this will indeed be a rewarding challenge.

History of My Academic Leaves at Trinity University

Since joining the faculty at Trinity University, I have had two academic leaves. The first was in 1988 and the second was in 1993. My first leave resulted in three published papers. My 1993 leave resulted in two published papers plus invitations to present my research findings at one university in Finland, two universities in Sweden, five universities in the United Kingdom, and ten universities in the United States.