Understanding Financial Modeling

Understanding Financial Modeling

 

by

 

Sam Choon Yin (2003)

 

Introduction

 The use of spreadsheets in financial modeling has grown rapidly since the late 1979. Thanks to the introduction of the Visicalc spreadsheet program in 1979 by a Harvard Business School student in the United States, more sophisticated computer software like Lotus 1-2-3 and Excel have been developed to cater to the needs of both individuals and organizations. This development is not surprising given the usefulness of spreadsheets in financial modeling as compared to carrying out the tasks manually using hand calculators. It is the intention of this essay to highlight the uses of the spreadsheets in financial modeling and some of the problems associated with the usage. Since financial modeling requires the use of microcomputers, it is essential that one gain an idea on the evolution and uses of computers. I therefore include a short section in the essay on computers.

 

Some definitions

            It may be useful to begin by defining some terms, starting off with the word ‘model’. In some books on financial modeling, the term model is often defined as the ‘representation of a particular reality’ (Pfaff, 1990, p. 2), but ‘built on a scale that can be more easily manipulated’ (Beaman et al, 1997, p. 3). The latter indicates that the models built are meant for manipulation either deliberately or not deliberately therefore rendering it both useful and useless. Allowing the models’ variables to change is deemed appropriate if one is interested to perform scenario analysis or sensitivity analysis. The objective in these analysis is to allow the users to substitute different values into the variable inputs and observe how the outputs would alter under different scenarios. Using the spreadsheets to perform the task is much simpler and time saving as compared to doing the same task manually. However, manipulation makes the models subjective. The users may have the urge and tendency to substitute the values into the variable inputs to make the outputs ‘look good’. The users might not seriously question the reasonableness of the inputs (that is, using the data without a proper reasoning behind them). The other problem is that with manipulation of data, the use of financial models may be more dominantly used but adopted unwisely. The users may be so reliance on the use of the models for decision making that the decisions are made solely based on the outputs from the models without sufficient consideration on other variables left out of the models.

The term ‘finance’ deals with the issues on the sources of funds and the uses of funds. The role and responsibility of the finance managers are to ensure that sufficient funds are obtained and available for the firms. Typically, a firm can obtain funds from three sources; the issuance of shares (for listed companies), borrowing (say through the issuance of bonds, or borrowing from financial institutions) and retained earnings (which is the internal source of funds). With the funds, the next question to address is ‘what is the most effective way to allocate the firm’s scarce funds to yield the best returns for the firms’. The issue on funds allocation requires the financial managers to acquire techniques like the net present value calculations and the internal rate of return computations for the purpose of evaluating the various projects that the firms can choose from.

Combining the terms finance and model, it is now appropriate to provide the definition of financial models. According to Pfaff (1990), financial modeling is a ‘mathematical construct which can be used to help better understand, control, examine or manage the financial sector of a firm, or the financial affairs of an individual or institution’. (p. 19). The definition is self-explanatory. Basically, financial models assist the planners to understand how reality works. Through techniques like sensitivity analysis, the planners can focus on those variables that are more important in explaining the behavior of the reality, and put less focus on the less important factors. With the models, which provide the planned outcome, the planners could use the models as a basis of comparison between the actual outcomes and the planned outcomes. With this knowledge, the planners can take correction actions to correct any deviations that may arise.

The other definition of financial modeling can be obtained from Beaman et al (1997). Beaman et al (1997) defines a financial model as ‘a computer-generated model representing the whole or parts of an organization’ (p. 3). They then noted that a financial model ‘can encompass a few aspects of a company’s financial situation, or all of the thousands of activities of a major corporation’ (p. 3). It is interesting to note that the financial model in Beaman et al’s sense does not incorporate only the activities of the financial sector in the company but extends the focus to a larger section of the organization. The authors recognize that the term ‘financial models’ is too narrow since the models are often used as a planning tool not only relating to the financial aspects of the firm but the physical, operational and human aspects as well. Hence they refer to ‘‘corporate modeling’ as possibly a better description of their nature’ (p. 3). I agree with them. It is widely accepted that the main objective of the firm is to maximize its profits. The attainment of this objective would in turn maximize the share price of the listed firms and the values of the shareholders. While it is not my intention to downgrade the importance of the financial sector, the achievement of the objective does require collaboration between the various functional areas of the organization, not just the finance functional area. Success in the marketing function, operations function, the research function and the human resource function for example is equally important. It is not surprising therefore for some to suggest cross- functional decision making in the organization to highlight the fact that success of any organization requires the co-operations and interactions between the various functional areas (see for example Schroeder, 2000). This is easy to say, but difficult to achieve in practice. One reason is that responsibilities of the functional areas are conflicting with one another (see Sam, 2002).

Despite this deficiency, the term ‘financial modeling’ is nevertheless used in this essay as part of the norm. However, it is important to recognize that the success of using computer-systems to aid decision-making encompasses the co-operations from the various functional areas in the organization. Bill Gates (1999) uses the term ‘digital nervous system’ to describe the essentiality to unite all the systems and processes using information technology, and release the information speedily to allow the companies to achieve quantum leaps in growth, profits and efficiency. Like the human body, a company too needs a ‘nervous system’ – an internal communication mechanism – to coordinate its actions and provide the necessary information and data to the staffs at the ‘speed of thought’.

 

Types of financial models

            There are several ways one can categorize the financial models, which basically make use of the computer system to aid in the decision-making process. Herbert Simon, Nobel Laureate in Economics in 1978 for ‘pioneering the research into the decision-making process within economic organizations’ (http://www.nobel.se/economics), identified two types of decisions. The first type, ‘programmed decisions’ represents those decisions that are repetitive and routine in nature (Simon, 1977). These decisions are usually made on a regular basis to assist in the decision-making. The models built can be used again and again over time, with the specifications remaining more or less unchanged. It is the responsibility of the modeler to update the variables inputs so that a new set of outputs can be obtained. The second type is ‘non-programmed decisions’. These decisions are unique and it is usually not known when one is required to make the decisions. Because of the non-predictability and uniqueness of these decisions, they should be treated separately. Models could be tailor-made to suit the nature of the problem addressed. Beaman et al (1997) called such models ‘ad hoc models’ (p. 9).

            Financial models can be broadly categorized into, first, simulation models. Also known as ‘what if ‘models, the models aim to simulate the effects of different scenarios on the outcomes so that the planners become aware of the different outcomes obtained base on alternative values of inputs. With the information, the planner can foresee to some extent, the likely outcome of the model under different circumstances. Different strategies can then be placed to minimize the negative effects of the outcome or maximize the opportunities present. The technique on sensitivity analysis falls under the ambit of ‘what-if models’. Instead of making changes to a set of values in the case of scenario analysis, the sensitivity analysis involves the alternation of only one variable input at one time. This allows the planner to assess the change on the outcome of the model when one variable input changes. Doing so allows the planner to assess the degree and relative importance of the respective inputs. Noting which of the inputs have a greater impact on the outcome permits the planner to place more concentration on these variables.

The second type of financial models belongs to the group known as the optimization models. The objective of these models is to maximize (or minimize) the objective function subject to some constraints that exist in achieving the objective. For example, the decision to locate the optimal amount of goods X and Y to produce could be for the purpose of maximizing the revenue collection for the firm. The attainability of the objective lies on the production constraints that the operations have to face. For instance, scarcity of resources like labor and certain inputs may pose a limitation on the number of units of X and Y the operations can produce on a typical day. It is the interest of the models to select the optimal points of the decision variables that maximizes the objective function, while at the same time, satisfying all the constraints encountered.

The third type of financial models can be collectively named as forecasting models. Forecasting models can be of two types: time series forecasting models and causal forecasting models. The former involves forecasting a variable using only the variable itself. It is important that historical data of sufficient size is available for proper usage of the models. Using techniques like moving average, weighted moving average and exponential smoothing technique, short-term forecast can be derived for the variable concerned. Relative easy to construct, the task can be performed on most of the spreadsheets in the market. An example is the Excel spreadsheet. Causal forecasting technique on the other hand involves the estimation of the regression equations. With historical data on at least two variables, it is the interest of the forecaster to develop the relationship between the dependent variable (the variable whose behavior is explained by the independent variable(s)) and the independent variable(s) (the variable(s) that explains the behavior of the dependent variable). Once the relationship is derived, short, medium and long-term forecasts can be obtained by substituting the relevant estimated figures of the independent variables into the regression equation. Causal forecasting can be performed on Excel for both the simple regression and multiple regression cases.

It is useful to note that the forecasting models can be used in collaboration with the ‘what-if’ models. For example, in financial planning, the sales forecasts can first be estimated using the forecasting models. Given the sales forecast, the planner could assess the need of external financing to finance and support the growth in sales. ‘What-if’ or sensitivity analysis can then be performed on the spreadsheets (such as altering the retention ratio, tax rates and/or cost-to sales ratio) to assess the magnitude of the external financing needed (if any) under different scenarios.

            In some books like Beaman et al (1997, p. 95) and Shim and Siegel (1989, p. 190), financial models are also categorized as deterministic and probabilistic. Deterministic models are those models where the inputs are imputed with certainty. They do not include any random variables. Historical data on sales are usually deterministic in nature. Probabilistic models on the other hand contain inputs that are uncertain in terms of their occurrences. The variables concerned contain probability distributions where one can attach probabilities to each of the values reported. Forecasted values on sales for example are probabilistic where each of the probabilities of occurrence is usually less than 1.

            The other distinction in the financial models lies on the users of the models. Individuals who wish to develop their own financial plans base on their goals, income and preferences can develop financial models to assist in their planning. Or, financial models can be used and developed by individuals working for a corporation to assist the organization in its planning and decision-making.

 

Uses of financial models

             The ultimate aim of an organization is to maximize profits. This is usually an acceptable objective for the firm from the society’s point of view, although the society requires that the objective is attained via ethical means. To assist in the attainment of the objective, it is useful for the firm to set up reasonably sound financial models to incorporate the inputs from the various functional areas in the organization. For example, the sales projection of the firm should be in line with the industry standards and economic conditions prevailing during that time. It must be acceptable from the marketing’s perspective. Constructing supporting models pertaining to long-term financial planning, capital investment, budgeting and tax planning for instance is appropriate. It is important that the planners are well verse with the concerned concepts in the accounting and finance areas.

Financial models help to present the reality in a more manageable and neat sense, providing us with the essential materials to understand how the reality works. Central locations like mean, mode and median and other descriptive statistical measures can be obtained with ease using spreadsheets. The use of easy-to-read graphs and charts also helps to provide a graphical representation on the data involved. As noted earlier, financial models are manipulative. Alterations of certain variables help the planners to assess the degree of importance of certain variables (so that more attention can be placed on them), and assess the effects on the changes of the outcomes of the models. As a controlling tool, financial models provide the planned (or forecasted) scenario, which could be compared with the actual outcome. The intention is to assess the areas which have gone wrong (if any) and take correction actions to rectify the problems, and prevent similar incidents from taking place again. Last but not the least important, construction of financial models provide enjoyments to the people. It is indeed challenging for one to build up the models, making them work out the way they wanted, and ultimately helping one to perform his or her tasks in a more efficient and effective manner. The satisfaction derived from knowing that the models work is priceless to the modelers.

            In general, financial models provide support for decision-making. Using the computers, financial models provide aid to people in ‘identifying problems, consider alternatives, evaluate uncertainty and make decisions’ (Beaman et al, p. 5), as part of the larger concept of decision support system (DSS). Coined by MIT professors, G.Anthony Gorry and Michael S.Scott Morton, the decision support system uses the computer to support the management in decision-making (see Gorry and Morton, 1971). Depending on the extensiveness of the problem addressed, the demand for more computer power to solve the problem increases as the level of complexity of the problem increases. For example, the decision support system needed could be less sophisticated if one is only interested to retrieve information from some sources. But, should the individual be interested in the preparation of reports from multiple files, the estimation of decision consequences, the proposition of decision and making of decisions, the type of decision support system required becomes increasingly demanding and complex. More functions on the spreadsheets for instance may be required to perform the task(s) (Alter, 1976; see McLeod, 1995, chapter 14 for more details on the DSS).

 

Limitations of the financial models

            While the use of financial models have been increasing substantially over the years particular since the introduction of more advanced and easy-to-use computer software, it is essential for one to recognize the problems associated with the use of financial modeling in decision-making. It is the interest of this sub-section to discuss some of these problems.

            First, it is not possible for models to incorporate all the variables affecting the outcome into the analysis. Failure to take all the variables into consideration renders the models ineffective. Although one may say that the degree of ineffectiveness depends on the importance of the excluded variables, the fact that the outcomes could be altered with any of excluded variables changing implies that the models cannot be used to predict certain events with certainty. An example is illustrated by the Tacoma Narrows Bridge built in the state of Washington. Shortly after it was built in 1940, the bridge collapsed. The designers of the bridge later found that the cause of the collapse was due to their ignorance for not taking into consideration the aerodynamic characteristics of the bridge, which they deemed as unimportant. As it turned out, the analysis was very important indeed (Pfaff, 1990, p. 5). Given that they are numerous variables to be taken into consideration, it is a challenge to the modeler to decide which of the variables are important, and which are less. In fact, financial modelers face a dilemma in terms of the size of the models to be constructed. It is not easy to decide on the right size of the models. While a big model is comprehensive, it may become too complicated and costly to handle. Updating of the model may become clumsy, time consuming and costly given the numerous variables to consider. Models that are too small pose a problem too. Excluding too many variables for example may render the model ineffective with low predictive power. It is therefore not surprising for Pfaff (1990) to conclude that model building is more of an art than a science (p. 5). In the financial and economics contexts, some variables are not included in the financial models simply because they are not quantifiable. This highlights a problem with the financial models, which is over-reliance on quantifiable variables. Important information like quality, culture, and customers’ satisfaction are not easy to be quantified, and therefore are often excluded in the construction of the financial models. They are nevertheless important variables to consider.

            Second, the users of the financial models may end up relying too heavily on the financial models in making the final decisions. This problem is particularly serious in cases where the models considered are too simple. As noted, ignorance of important but non-quantifiable variables may render the models ineffective in predicting the outcomes, particularly if there is high probability for the neglected variables to alter in the near future. In many financial models, even quantifiable variables are excluded from the models. It is therefore essential that the modelers and decision-makers are aware of these deficiencies of using financial models. One should not follow blindly on the predictions made by the models. Instead, careful thoughts and knowledge on the subjective internal and external environments are crucial to supplement the financial models in decision-making.

            Third, financial models tend to over-rely on accounting data as a source of input. Accounting data are constructed on an accrual basis. Cash flows are ignored. For example, salary payments for the month of December are likely to be made in January the next year, but recorded in the financial statements in December (when the financial statements are prepared). As a result, the profit and loss recorded in the financial statements does not reflect the cash flow positions of the firm. The proper method in the evaluation of projects therefore is the use of expected cash flows generated from them, rather than the use of accounting information. Despite these problems, the use of accounting data in financial modeling is still common.

            Fourth, the use of accounting data forces the modelers to deal with average costs and average revenues in the decision-making. This tends to mitigate the effectiveness of the financial models for decision-making is better made with marginal costs and revenues instead. For example, in deciding whether to produce another unit of the output, the decision maker is required to assess the additional revenue the unit contributes to the organization and the additional cost incurred in the production of that unit. It makes economic sense to produce the additional unit if the marginal revenue exceeds the marginal cost of production. Otherwise (when the marginal cost exceeds the marginal revenue), the additional unit should not be produced from the economic point of view. The profit maximizing level of output is reached when the marginal revenue is equivalent to the marginal cost of production. Accounting data does not allow the user to carry out marginal analysis.

            Fifth, while the use of spreadsheets has been extensive over the years and more and more functions have also been added into the software, problems exist. One of the problems faced is the difficulty of transferring the models from one user to another. The transfers may be required when there is a change in the modeler. Taking over the spreadsheet and updating them could lie in the hands of another person. Wastage of time and repetition of work may be resulted. To minimize repetition of work and time, it is essential that the person taking over is capable of interpreting the models in the spreadsheets. To facilitate the transition, it is the responsibility of the organization to provide training to the staffs involved on the tools used.  Recruitment of new staffs should also be screened properly. The other essential task is to inculcate the culture of applying the spreadsheet design principles to the modelers (Beaman et al, 1997, chapter 2). These are simply principles to adopt and practice in the development of spreadsheets for ease of updating, transferring and interpretation. For example, proper specification of formulas in the cells in Excel, careful use of range names, auditing and protecting the worksheets are some guiding principles to adopt to assist the model builder to construct models that are error-free, user-friendly and reliable in support of managerial decision-making (see Beaman et al, 1997, chapter 2 for more details on the principles). Basically, not everyone is capable of churning out useful financial models using the spreadsheets. Without proper training and planning, the spreadsheet model errors may be numerous leading to mistakes made into the cells, and therefore inaccuracies in the prediction of the intended outcomes.


Computers: evolution and their uses

            The first electronic computing machine was completed in 1945. Known as the Electronic Numerical Integrator and Computer (ENIAC), its emergence replaced the mechanical calculating machines using gears and wheels. Although it was able to perform computations quickly (it did 333 multiplications or 5,000 additions per second), it did not assist decision-making in an intelligent way.[i] Users needed to be trained to use the technology. It was also costly to make, costing approximately USD500,000 in 1946 dollars (about USD 5 million to USD 10 million today or SGD 8.5 million to SGD 17 million). Today, it costs less than SGD1,000 for one to purchase a desktop computer! Large in size, the ENIAC was 8 feet tall, 80 feet long and weighing about 30 tons.

With the introduction of transistors, electronic computers became more reliable. In the late 1970s, the integrated circuits (or chips) allowed computers to be built in smaller sizes. The first personal computer (the term ‘personal computer’ was used as early as in 1974) – Apple II – was sold to the public in 1977 by Steve Wozniak and Steve Jobs (Apple Computer, Inc.). The International Business Machines Corporation (IBM) sold its first personal computers to the public in 1981. The growth of software development and other peripherals soon followed in the 1980s. Of particular significance was the establishment of Microsoft (founded by Bill Gates and Paul Allen) – the largest software company in the world – in 1975.

            A typical computer system contains four main components commonly known as the computer hardware. The central processing unit (CPU) contains the processor, the heart of the computer system. A tiny computer chip, the processor is the sub-component of the computer where all the operations like calculations, comparisons and messages sending and receiving are carried out. The processor itself stores very little data. It acts more like the retriever of information from a data storage device called the memory. The most common type of the memory is the RAM (random access memory), which can move data at high speed within the computer’s circuits. RAM contains another chip that store data and are retrievable easily by the processor. A computer memory chip (with a size smaller than a finger nail) can store as much as 64 megabytes of data equivalent to the texts of more than 50 books (Basse, 1997, p. 7). The second component of the computer system is the storage device. Storage of memory in the semiconductor chips in RAM is expensive. Cheaper ways of storing mass amount of data and information are available. Data can be stored in the hard disk, which has emerged as the ‘most efficient solution for storing large amounts of data at low, and declining cost’, (McKendrick et al, 2000, p. 17). Besides the hard disk drive, data storage can be done on floppy disks. The floppy disks (which have decreased in size over the years from 8 inches to 3.5 inches) are capable of storing 2.88 megabytes of information each. The other alternative is to store the data in the compact disks (CDs), which can hold up to 650 megabytes. The third component of a computer system is the input devices, which include the keyboard, joystick and mouse. The main function of the input device is to give instructions and provide data to the microprocessor. The last component, the output device, contains the monitor/screen (that displays the output of the program being run) and the printer (to provide hard copies of the output in a format that is controllable by the users).[ii]

            Besides the hardware, the computer software is essential contributing significantly to the overall usefulness of the computer system. Computer software is a ‘set of instructions that tells the computer what to do in order to accomplish a task or an activity’ (Pfaff, 1990, p. 14). Examples of the computer software are the computer operating system and the spreadsheets. The former is like the conductor of an orchestra, helping us to understand what is in the computer, and aiding in the running of the programs stored. Understanding how the operating systems works enable the users to enjoy the benefits of the computer on a fuller extent. The latter (spreadsheets) are programs written by the respective providers like Lotus 1-2-3 and Excel to assist the users in performing certain tasks like graphing, tabulating, and statistical analysis in a simpler manner. Formulas can be imputed in the spreadsheets to tell the computers to perform the calculations intended. Information can be altered and saved for later used, and they are transferable.

            The rise of computers has posed numerous challenges, both good and bad, to the people around the world. The use of computers at home has been growing tremendously (a computer is found in every three households in the United States). More organizations are relying on the computers to perform spreadsheet calculations, report writing, performance of day-to-day operations, and to aid decision-making. Labor productivity has increased with the use of computers. Things that are done manually can now be completed using the computer at a fraction of the time. In manufacturing organizations, the use of computer has helped to facilitate the operational and production processes. For example, time and monetary savings are enjoyed with the use of computer-aided design (CAD) during the product design stage. Instead of building prototypes, designers can use the CAD software to design the products at different geometrical dimensions on the computer itself. The information can be saved and sent to the clients for approval without the need to build actual models for consideration. It has been recognized that the use of information technology in general, and computers in particular will extend beyond a particular functional area to an integrated form where processes and systems are united across the organization to facilitate planning and release of information at a faster speed to support decision-making (Gates, 1999).

            Despite the pros of using computers, one should acknowledge the downside of the computer revolution. Sara Baase (1997) wrote an excellent book on the legal and social issues regarding the use of computers. Some of the social and legal issues highlighted in Baase (1997) include the following. First, rising unemployment has been resulted from the increasing use of information technology in the working place. Citing examples like reduction in the number of tellers in banks, telephone operators, and electric meter readers, Baase (1997) acknowledged that some jobs have become more redundant with the growth of the information technology. However, the situation may not be as bad as one perceives. This is so as the rise of the information technology sector has increased job opportunities, albeit of different kinds, that had not existed in the early 1980s. Examples include computer programmers, computer engineers and Internet-related jobs. Second, interactions via the computers and machines have led to lower customer contact between the service providers and customers. More transactions are being carried out with the machines (like Internet banking and the Automated Teller Machines, ATMs), creating frustrations to some consumers who desire face-to-face interactions with the service providers. Third, crimes involving the computers like hacking, theft (say from the ATMs), tax evasion, money laundering and vandalism have imposed losses on the individuals and the society in general. Monetary losses and physical injuries may be resulted from thefts at the ATMs. Closure and interruptions to essential services may be resulted from hacking and vandalism on computer-related systems and products, while tax evasion may lead to loss of tax revenue to government therefore reducing the government’s capability to finance its expenditures on essential services (like public and merit goods) to its citizens. The ease of money laundering via the computers could have stimulated more criminal activities from taking place given the easiness for one to remit the earnings to another country. Fourth, the increasing use of computers may lead to loss of privacy. The filing of taxes via the Internet for instance, involves revealing information like income, identity card numbers and other confidential information from the taxpayers. The government has huge databases on its residents. The issues of who should have access to the information and how should it be protected from abuse and errors are important issues to consider and address. Consider another example. The purchase of items over the Internet results in the purchasers’ information being recorded in the computer databases of the sellers (for example, the credit card numbers and the purchasers’ addresses). It is possible (and not uncommon) for others to have access to the database leading to exploitation and abuses on the users of the computer services.[iii]

 

Conclusion

            In this essay, I have discussed several issues pertaining to financial modeling, organized in the following manner. The essay begins with a discussion on the meanings of the term financial modeling - an area of study which uses the computer system to aid decision-making in an organization to achieve its objective of profit maximization. The essay then identifies the different types of financial models exist in the literature followed by a discussion on the uses of the financial models. In that section, I highlighted some of the main uses of the financial models aimed to recognize the areas of contributions that financial models can provide to the organizations. Then, to provide a balance view on the use of financial models, the essay recognizes and discusses some of the limitations exist in using financial models in decision-making. Suggestions to soften the negative effects of these limitations are provided whenever appropriate. The last section of the essay provides a brief introduction to computers containing discussions on the revolution of computers, the components of the computer systems, and some pros and cons of using computers.  The understanding of the computer system is deemed appropriate and relevant in our context given the extensive use of computers in financial modeling.

             


References

 

1.                  Alter, Steven, L. (1976) ‘How Effective Managers Use Information Systems’, Harvard Business Review 54 (November-December), pp. 97-104.

2.                  Baase, Sara (1997) A Gift of Fire: Social, Legal, and Ethical Issues in Computing. Prentice Hall, Inc. (United States).

3.                  Beaman, Ian; Ratnatunga, Janek; Krueger, Peter; and Mudalige, Nihal (1997) ‘Financial Modeling’. Third edition. Quill Press (Australia).

4.                  Capron, H.L. (2000) Computers: Tools for an Information Age. Sixth edition. Prentice Hall, Inc (United States).

5.                  Gates, Bill (1999) Business @ the Speed of Thought: Using a Digital Nervous System. Warner Books (United States).

6.                  Gorry, Anthony, G. and Morton, Michael S. Scott (1971) ‘A Framework for Management Information Systems’, Sloan Management Review 13, pp. 55-70.

7.                  McKendrick, David.G.; Doner, Richard, F. and Haggard, Stephan (2000) From Silicon Valley to Singapore: Location and Competitive Advantage in the Hard Disk Drive Industry. Stanford University Press (United States).

8.                  McLeod, Raymond, Jr. (1995) Management Information Systems. Sixth edition. Prentice Hall, Inc. (United States).

9.                  Pfaff, Philip (1990) ‘Financial Modeling’. Allyn and Bacon (United States).

10.              Sam, Choon Yin (2002) ‘Operations function: Can it work independently?’. Unpublished paper.

11.              Schroeder, Roger, G. (2000) Operations Management: Contemporary Concepts and Cases. Irwin McGraw-Hill (United States).

12.              Simon, Herbert, A. (1977) ‘The New Science of Management Decision’. Revised edition. Englewood Cliffs (United States).

13.              Shim, Jae, K. and Siegel, Joel, G. (1989) Encyclopedic Dictionary of Accounting and Finance. Prentice Hall, Inc. (United States).



[i] See Baase (1997, chapter 1) for an interesting introduction on the history and growth of computers.

[ii] For a fuller discussion on the components of the computer system, read (Capron, 2000).

[iii] There are other interesting issues addressed by Baase (1997), which will not be elaborated in this essay. I strongly recommend the book to anyone who is interested to know more about the legal and social issues pertaining to the use of computers.