stnflogo.gif (813 bytes)

SBG at STANFORD


Some guidelines, derived from more than 15 years of teaching various versions of the Stanford Bank Game at Stanford, may be of interest to instructors. While these suggestions are not exhaustive, they represent one of a variety of successful formats for presenting and running the simulation package.

Introducing the Bank Game

At Stanford, the simulator is taught as part of the second year elective course called Management of Financial Institutions . The course deals with the principles of managing commercial and investment banks and other non-bank financial intermediaries. The Bank Game is played largely outside the classroom during the term; the amount of specific class time devoted to the game consists of an introduction during the first class, one class devoted to establishing the logistics of the game and the basis upon which the winners will be selected, and a final class to present team reports at the end of the game. During the rest of the term, the students make one decision per week.

To introduce the Bank Game, the instructor offers three reasons for its use:

1. The Bank Game is a dynamic learning exercise that builds on and supplements course material. Because new topics are covered in every class, much of the course work may seem only weakly linked; however, the game involves making a series of similar decisions over time, allowing students to note the difference between long- and short-term strategies. After the first period, no two teams are in the same place; the lead changes over time, making an interesting sequence of decisions and results.

2. The Bank Game is integrative in that everything depends on everything else in the decision-making process and the number of decision combinations is almost limitless. Making decisions that involve both analysis and intuition in an integrative fashion is one of the most important parts of the game. Of course, as the game progresses, the better teams learn to apply more analysis and less blind intuition by watching the way results unfold from previous decisions. Nonetheless, the game is not meant to be a solvable deterministic problem. Rather it is an exercise in decision making with imperfect information. Analysis is not down played; it is simply placed in the context of real world complexity.

3. The Bank Game is a group learning experience in which working as a team is one of the most important dimensions of the game. Even though instructors stress the importance of group participation, some teams come to be dominated by a single individual's personality or effort, often to the detriment of the learning experience of all the team members. Usually teams dominated by single individuals do not do as well as more participatory teams.

After discussing the reasons for playing the Bank Game, the class is assigned to read the player's manual and to prepare to discuss the following questions at the next class:

How should the winning team be chosen?

What criteria are appropriate?

The instructor gives no other assignment the first day except to ask for voluntary team lists. Those who do not appear on voluntary team lists are assigned to teams on a random basis. Optimal team size is five or six members.

Competition

At the second class, team lists are handed out and participants discuss how the winning team should be judged. They may suggest the biggest bank, the most profitable bank, the highest EPS, the most consistent bank (whatever that means), the least risky bank, the bank with the highest stock price (as the player's manual suggests), the bank paying the most dividends, and so on. Students usually carry on a lively and interesting discussion, especially if they are encouraged by the instructor.

After about 45 minutes, the instructor intervenes to suggest that the model's stock pricing function may be the best criterion. Essentially, the students are in total control of their profitability, but to increase profits they must usually take on higher levels of risk. A team can generate short-term profits, but they may introduce exceptionally high risk levels, which they must handle successfully at some future date. The computer program's stock pricing function penalizes a bank for high risk in the short-term (on stock price) because it assumes that high levels of risk reduce the probability of the bank being able to maintain current performance levels. In the early stages of the simulation, students are free to increase the risk; they have the rest of the game to solve the problems this creates. At the end of the simulation, if stock price is the criterion, the students are forced to leave their banks in a reasonable risk position. As long as the bank is paying appropriate dividends and is in a reasonable risk position, earnings (past and current) become the dominant issue in the stock price equation.

Explaining the Model

Some students are much more interested in the model than in banking, and they will-if permitted-attempt to examine every detail of the model in order to "solve" it. Of course, in the extreme, this defeats the combined analytic and intuitive purpose of the game. Also, it is beyond the scope of a single course to dissect the model completely-or even to prove its validity.

In lieu of full explanation, the instructor usually explains that the game is a complex model with parameters that are intuitively appealing. For example, when interest rates charged go up, loan demand goes down; when business development expenses are increased, demand increases. The precise equations for these effects are not revealed. Most are so complex that they would be of limited value in any event.

The Stock Price

Most frequently, students want to know how the stock price in the Bank Game is determined. Here, the instructor can ask the student how they would determine the price if they were to write the game from the beginning. After some discussion, the instructor can indicate the following:

1. The stock price starts with a smoothed EPS, as noted previously.

2. The EPS is multiplied by an exogenous variable from the economic deck called the industry P/E ratio.

3. The resultant stock price is adjusted upward or downward by the K factor. The instructor usually tells the class the individual components of the K factor, but does not reveal the magnitude of their effect. The player's manual provides a few clear-cut guidelines in this area. The K factor components, shown in the Instructor's Report, generally are considered intuitively acceptable to most students, even though some are inclined to argue about their precise effects. But the Bank Game is just a game; while it represents a model of banking that makes sense, it cannot claim to be a complete replica of the real world. For the students playing the game, however, the model is the real world.

Final Presentation

For the final class of the course, both the winning team and the runner-up make a 30-minute presentation of their strategy, performance, and group process in playing the Bank Game. Occasionally, the last-place team is also asked for a presentation. All students write a team report of about 10 pages. Grades for the course at Stanford are not dependent on how well a team does, although about 25 percent of the grade does depend on the quality of the written report. The quality of the reports does not necessarily parallel the final stock price of the teams. In fact, teams that do less well in the game tend to learn more about profitability, capital adequacy, and liquidity than teams that do better. They may prove the old adage that every business failure is a learning experience. In any event, the game tends to function better if results and grades are not connected.

In course evaluations, students cite the Bank Game as a major attraction. The Bank Game has been played at other schools with a greater emphasis on the game as the centerpiece of the course. These teaching suggestions show only one possible alternative that you may find helpful.

HOME BACK