Business Strategy Archives

Music2Go Marketing Factory

Question of the Week: Why did I receive a different number of units from what I ordered? | Music2Go Marketing Business Simulation

In Music2Go you make decisions for an entire year, but your factory has a limited ability to adjust the number of units produced to try to meet actual demand during the year. This is called Demand Responsiveness.

Demand Responsiveness allows the actual number of units ordered to increase or decrease by up to 20% to meet the actual demand for your product.

For instance, if you ordered 1 million units of a product, then the actual number of units delivered could vary between 800,000 units and 1.2 million units depending on actual demand.

Product Contribution Report in Music2Go Marketing

In our example above, we ordered 1.9 million units of our Sonic product, but the Actual Units ordered was less than this at 1.5 million because the demand for our products was less than what we anticipated to sell.

Note: Most worlds have 20% Demand Responsiveness enabled, although your instructor may request this to be modified or disabled for your Multi-Player. 

Related Articles
MikesBikes Intro Hall of Famer, Joe Powell

Interview with the Top Hall of Fame Student: Joe Powell

Joe Powell from Des Moines Area Community College (DMACC) managed to land the #1 spot in the MikesBikes Introduction Hall of Fame!

Joe took the Introduction to Business course taught by Professor Josh Daines over at DMACC. He’s a 41 year old student and a father of 4 returning for a degree in Business Administration. He just graduated a degree in General Studies and got to share the experience of receiving his diploma alongside his wife and daughter in the same ceremony (see photo above). He is now currently working on his second degree.

He never won anything, except for an English contest when he was in 4th grade and a Renaissance Festival tickets last year, but now he managed to successfully land the Top position in the MikesBikes Intro Hall of Fame. This is such an incredible achievement!

We have interviewed Joe to learn about his journey on how he managed to achieve the top spot in the Hall of Fame, his experience and advice to future students.

Smartsims: What is your decision making process within the simulation?

Joe: This is actually the toughest question, I think, that you could ask. My decision making process started out simple, but as the rollovers continued, other factors began influencing my decisions. The first couple rollovers were straightforward: based on the advertising arc, which investments paid off the best? Then the additional product lines, how would the competition respond, what was the best allocation of funds outside of the advertising arc, etc.

Smartsims: What was your strategy going into the simulation?

Joe: I wanted to jump out to an early advantage, allocating funds where they would make the highest return. I wanted to maintain high revenues, high cash-on-hand, and minimize lost sales. I also figured the other firms would implement similar strategies, so I needed to be flexible yet maintain consistent growth. Once I began doubling my SHV every rollover, I focused more on my firm’s progress projections and less on the competition.

Smartsims: How did you begin implementing that strategy?

Joe: I knew the minimum SHV numbers that I wanted to hit at each rollover. I would run practice sim after practice sim until I found the right combination to get to that number. Once I did, I ran more sims to see if I could improve on it. Sometimes it worked, sometimes I made myself frustrated.

Smartsims: How did you familiarize yourself with the simulation?

Joe: I started by opening it up and trying everything. At the beginning of the semester, I did single-player simulations and failed spectacularly. I learned a little here and there, and with the help of some professors at DMACC, I was able to get a grasp on the program.

Smartsims: How would you describe the competition?

Joe: The competition was fun, above all else. The other firms made strategy implementation vital because I just didn’t know what they would do. It was a lot more fun with real people playing instead of just a computer with predictable moves.

Smartsims: What resources did you pull on to develop your winning strategy which led you to be a part of the MikesBikes Introduction Hall of Fame?

Joe: I utilized the Single-Player format first. After about a dozen practices, I had a solid opening strategy. I also talked to MikesBikes guru, DMACC’s own Professor Zarr as well as my Business teacher, Professor Daines, and asked a few questions. Without pointers from them, I wouldn’t have reached the numbers I did. I studied the interviews of previous winners, played around with different strategies, even ran rollovers where all I did was change a few dollars here or there to learn where the benchmarks were. I created my own spreadsheet that helped keep track of money allocations, allowing me to look back on what worked and what didn’t.

Smartsims: What challenges did you face? How did you overcome these?

Joe: At first, my numbers were right where I wanted them. Then, I started getting lost sales and I couldn’t figure out why. My numbers were exactly where I wanted them, and the results should have been predictable. I turned to the offline mode, and it was a life-saver. I strongly suggest anyone in multi-player competitions use the offline mode. It will not predict your opponents’ moves, but it will help you figure out your own.

Smartsims: Was there anything in particular you did that you think helped to prepare yourself?

Joe: I obsessed over this simulation. It wasn’t healthy. I talked about it, I played it, I thought of new strategies while walking around Disney World. My kids rolled their eyes when I would bring it up.

Smartsims: How has participating within a course which uses a business simulation to supplement their teaching materials helped you? What do you think of the business simulation?

Joe: I really liked the sim as part of the curriculum. I mentioned that to the professor during the class review. Learning about the Advertising Mix as well as the Promotion Mix helped understand the importance of allocating funds to specific areas within the sim. Studying it, then putting it into direct action was really a cool experience. I really enjoyed having Mike’s Bikes as part of the class.

Stack of money.

External Funding Options in MikesBikes: Debt or Equity?

In MikesBikes, funding large investment options such as purchasing additional plant capacity and development projects may require additional capital. Raising Debt or Equity are ways companies can raise additional funds to finance these large projects or investments. In deciding whether to increase Debt, issue Shares or a combination of both; companies should assess their current situation and the impact of each financial decision. Continue reading External Funding Options in MikesBikes: Debt or Equity?

economic value created

What is Economic Value Created (EVA)?

financial results for all firms

An increasingly popular way of measuring the financial performance of a firm is by looking at the Economic Value Created (often called EVA*) over a specified time span. See the Financial Reports section of the All Reports Menu for the firm’s current Economic Value Created report.

In essence, this measure views the business as an investment which must produce a certain return on the capital invested in it. If it produces more than the required return, then the difference is the economic value created or added (EVA).

The “actual return” is calculated by adjusting the net operating profit after tax to exclude the effects of interest.

The “required return” (or cost of capital) is calculated by adding together the interest charges on debt with the return required by the shareholders. The return required by the shareholders will vary according your firm’s level of risk and will be composed of required dividends and/or increases in share price.

If the actual return is higher than the cost of capital, then the difference is the economic value created. From an economic viewpoint this extra return must be due to some competitive advantage. The question is: How long can this competitive advantage be maintained before competitors come along and copy it?

An example of a simple EVA report is below:

EVA Report in MikesBikes


*EVA is a registered Trademark of Stern, Stewart & Co.

Distribution in MikesBikes

Question of the Week: How do we get more stores to stock our products?

MikesBikes Distribution Decision

Distributors look at the total dollar amount of margin earned from stocking each firm’s products last year. Based on that, each Distribution Channel will decide how many stores will stock products from each firm.

The total amount of margin a Distributor receives is affected by your:

  • Retail Price
  • Retail Margin and;
  • Volume of Sales

So all other things being equal, if you sell more products, or increase your Retailer margin then more Distributors will be willing to stock your products and your distribution indexes will increase.

The number of each type of distributor willing to stock your products is then combined with the shopping habits of each segment to give a segment level distribution index. This is why you see different distribution indexes in the Market Summary / All Product Details report.

Have a Question of the Week suggestion? Send this to us here or message us on Facebook.

Student Looking at Advertising and PR Reach Curve, MikesBikes

Question of the Week: How does the Advertising and PR Reach by Media Curve Work?

Advertising and PR Reach by Media Curve

The reach curves and media viewing habits reflect what proportion of the population you can reach with a given Advertising or Public Relations (PR) mix.  These are the same for both Advertising and PR partly because it is reasonable that the advertising reach and viewing habits should be similar for both product advertising and PR.

The sensitivity to Advertising and PR reflects how much each market segment is influenced by any Advertising or PR they are exposed to.  If a segment has low sensitivity to PR, then even if they are exposed to a lot of it, there will be a much smaller influence than if they are highly sensitive to PR.

If a segment is highly sensitive to Advertising, then there is probably going to be a reasonable market share difference, but if there was low sensitivity to PR then you would probably only see a small market share difference (ie. you are probably wasting a lot of your PR spend and you should look at more effective ways of using that money).

The market segments have different sensitivities, so you need to find balance between Advertising Awareness and PR in each market.

2017 MikesBikes World Champs Winning Team

Interview with the 2017 MikesBikes World Champs Winner: Exodus

The 2017 MikesBikes World Championship winners Siyi Tang , William Setiawan , Jiedong Chen, Jayca Y. Siddayao, Mingchang Liang and Robbi A. Harnass of Exodus from the University of Auckland have joined the MikesBikes Wall of Champions with their impressive performance in the competition!

Team Exodus competed against the best teams from around the globe and have achieved a Shareholder Value of $327.81.

To wrap up the 2017 MikesBikes World Champs, we have interviewed the team and ask them about their experience in the competition and what they thought about the simulation.

What is your decision making process within the simulation?

First, we analysed the market scenario then we looked at trends in pricing, marketing, financial performance and so on. We then compared these trends with our previous decisions and results.

Finally, we were able to determine whether we would have an offensive or defensive strategy. The most important thing for us is that the final decisions entered have been approved by all team members.

What was your strategy going into the simulation?

Our strategy was focused on developing low prime cost bikes. This strategy did not require a high investment in plant or workforce.

We then focused on our marketing campaign. We developed a marketing mix suitable for our strategy.

Knowing how to calculate the optimum marketing mix, we were able to reach more customers than if we just allocated a high expenditure on every media channel.

Finally, with regards to our financial decisions, we made sure to repurchase shares as soon as we had the additional funds and no profitable use for our extra cash. We were also able to pay out dividends to shareholders.

What challenges did you face? How did you overcome these?

Our competitors undoubtedly did their best to compete. They presented us with fearsome competition during the Championship. Our team also felt that some of the firms just gave up along the way and purposefully destroyed the market by dumping their prices. Pity moves indeed. But then again, it was one of the “unexpected” things that you would not expect to happen, especially during a fierce competition such as this.

Was there anything, in particular, you did that you think helped to prepare yourself?

One of the things that helped us win the world championship was the opportunity to compete with our fellow classmates. In a way, we were used to playing in such a competitive environment, with fearless competitors. However, even with such experience, there were moments during the world championships that we felt very overwhelmed by our competitors’ moves. Thus, once again, kudos to all who fought until the end.

What do you think of the business simulation?

We think MikesBikes simulation is a great way to teach people who want to learn Business Management. It has both practical and theoretical application that gives us real world experience that we would never get from attending traditional lectures and doing written tests.

For example, when we were looking to enter a more competitive market, we were intrigued to find more detailed information. Reports such as the Income Statement, Cash-Flow Statement, Balance Sheet, and Equity Movements Statements were extremely helpful. We made sure to analyze these reports every rollover.

Comments on your experience with the simulation itself

We had a great experience with the simulation. We are glad that we have used this simulation, especially for those of us who have never worked in the industry. By competing in the course competition at the University of Auckland and entering the MikesBikes 2017 World Championship, we learned so many new things, such as:

  • How to allocated our Marketing Expenditure in more attractive media channels
  • How to turn a significant investment in project development into something more profitable in the long run
  • How to calculate the cause and effect of the decision of repurchasing shares and dividends.

Nevertheless, we are honored to be called the winners of MikesBikes 2017 World Championship. There are many decisions that we made, which definitely could be further improved. We hope that newcomers will join the next world competition of MikesBikes, to stand beside us, and the rest of the MikesBikes World Champ Winners.


Check out the 2017 Competition Results here and the previous winners here.

MikesBikes World Champs Smartsims

2017 MikesBikes World Championship Competition Results

“Success is not final, failure is not fatal: it is the courage to continue that counts.” – Winston Churchill

The 2017 MikesBikes World Championship has finished with a bang! Team Exodus from The University of Auckland has secured the top position as 2017’s MikesBikes Champs and have joined the MikesBikes Wall of Champions. Team GOFAST, also from The University of Auckland, came in close second with an impressive performance! These top two teams certainly had the will to succeed and win the competition.

These two teams were neck-and-neck in the last few rounds of the competition battling for first place. Both teams constantly putting pressure on each other to adapt to their market and make growth-orientated decisions.

Our MikesBikes World Champs winning team, Exodus had the following to say:

2017 MikesBikes World Champs Winners
Siyi Tang , William Setiawan , Jiedong Chen, Jayca Y. Siddayao, Mingchang Liang. Robbi A. Harnass of Exodus from The University of Auckland

Our experience with the simulation has been unimaginable. We are glad that we have done the simulation, most especially since none of us have worked in the business industry. By doing the course competition at the University of Auckland, and entering the 2017 MikesBikes World Championship, we learnt so many new things. Such as, how to wisely allocate our marketing budget, how to turn a significant investment in project development into something more profitable in the long run, how to calculate the cause and effect between the decision of repurchasing shares and dividend, and many more. Nevertheless, as much as we are honoured to be called the winner of the 2017 MikesBikes World Championship, let us restate the fact that we are far from perfect. There are many things that we’ve done, which can be further improved. We hope that newcomers will aim to enter the next MikesBikes World Champs and to stand beside us, and the rest of the winners.

A huge congratulations to all the teams who have participated in the 2017 MikesBikes World Champs!

Here are the Final Round’s Rollover Results:

Team Members
Exodus $327.81 Siyi Tang | William Setiawan | Jiedong Chen | Jayca Y. Siddayao | Mingchang Liang | Robbi A. Harnass University of Auckland
GOFAST $313.41 Harriet Zhang | Golf Tantulaphongse | Xu Wang University of Auckland
Infinity $152.80 Sunil Ravindran | Thi Thu Vu | Manfei Cheng | Zhaonan Liu University of Auckland
Pedal to the Medal $88.06 Peter Bojsza Drexel University
Scooty Puff Legends $86.79 Jamison stewart | Zachary Khan | Nathan Jarrad Linn Benton Community College
Square Tires $2.80 Chris Mitchell | Nick Dawson | Darrell Davis Indiana University Southeast
Apex Bikes Co. $0.01 Zachary Abel Des Moines Area Community College
K2K $0.01 Kimberly Dowe Linn Benton Community College
SB Corp. $0.01 Kaushik Shah Midwestern State University
Van der Zwan AJ Inc. $0.01 AJ van der Zwan Ithaca College

You can view the current and past winners here.

how to create a great company name image

How to Create the Best Company Name for your Business

Naming your company is one of the most important decisions when establishing a new business. It forms the foundation of your company’s brand and is key to a lasting first impression. However, coming up a name for your business is often as challenging as it is important.

So what are the best ways to pick a creative, meaningful or impacting company name?

Here are the Top 5 Methods for Creating the Perfect Name for Your Business:

How to Name Your Company - Personal Identity

Many company names are inspired by the founder’s own name. For example Bayer (Friedrich Bayer was the founder of the company), Boeing (after William E Boeing) and John Deere.

An amalgamation or acronym of the founder’s names could be used. Ben & Jerry’s for example (Ben Cohen and Jerry Greenfield are the founders of this global ice cream empire), 3M (Minnesota Mining and Manufacturing Co) and DHL (Adrian Dalsey, Larry Hillblom and Robert Lynn).

Alternatively, an acronym which combines your name with your location. For example, DKNY (Donna Karan, New York) or IKEA (from the founder’s name, Ingvar Kamprad, and his hometown, Elmtaryd Agunnaryd).

Naming Your Company - Using Language

Take a word that’s symbolic for your company and invent a new word through a mash-up, alternate spelling or misspelling. This can give your company an internationally unique identity.

Mozilla, the internet browser, is a word mash-up, being Mozaic and Godzilla. Reebok is an alternate spelling of “rhebok”, the African antelope. Famous misspellings are Flickr and Tumblr, where the founders simply dropped the last vowel of the word.

Using a foreign word can also provide a unique flavor outside of the country of origin. For example, Acer is the Latin word for “sharp and able”.

Naming Your Company - Use of Humor

If you wish to be bold you can use humor to create an immediate reaction from consumers. A successful funny company name will provide a lasting first impression and encourage customers to tell their friends about you.

For example, in the highly competitive frozen yogurt market, the company name Spoon Me, was quirky and funny enough to catch people’s attention that consumers now proudly don the company’s t-shirts, bumper stickers and other merchandise.

Also, names that are a play-on-words or that utilize puns can be a fun way for a company to create its own distinct image. Everyone remembers the coffee shop in the New York-based sitcom Friends, Central Perk. And our local seafood restaurant is named The Codfather.

Naming Your Company - Making it Descriptive

Descriptive business names are upfront about the kind of products and services they offer. Their strength is they immediately communicate a key message describing what your company offers. This could be your company’s story, an inherent company value or a particular quality linked to your products.

For example, Freeset (a fair trade business whose goal to is set locals free from slavery and poverty), Best Buy and Staples.

More creative options are Greyhound (to convey speed), the Ford Mustang (evokes the emotion of power and freedom) and Nike (the winged goddess of victory).

Alternatively, you can combine two descriptive words to create a new meaningful word. For example Smartsims (smart & business simulations), Evernote (forever & note) and Groupon (group & coupon).

Naming Your Company - Keep it Simple

Sometimes the best company names come from simple ideas and concepts.

These include choosing a word from the dictionary (like Twitter), picking an inanimate object (Steve Jobs chose Apple for its friendly appeal and common usage) or even a geographical location (like Amazon or Fuji).

Alternatively, take a word that has special meaning for your business, but just use a portion of the word, like Cisco (from San Francisco where the company was founded).

Do you know of a Clever, Unique or Funny Company Name?

Have you come across a particularly meaningful, creative or humorous company name? This article is now posted on our Facebook page. Click on the link below and comment to share these with us:

Click Here to Comment and Share



Title image for JITF Case Study on MikesBikes

JITF Case Study of MikesBikes Simulation within an Introduction to Business Course

15 July 2009



The Institute of Finance Case Research (IFCR) is an academic organization committed to the production and promotion of case research and instructional techniques in all areas of finance.

The Spring 2009 edition of the Journal of Instructional Techniques in Finance (JITF) presents a case study about how the MikesBikes Simulation is used by Quinnipiac University to aid teaching of Capital Structure and Dividend Policy in an Intro to Business Course.


Capital Structure and Dividend Policy in an Intro to Business Course
Sean Reid, Len LaBonia, Ben Shaw-Ching Liu, Patrice Luoma, and Anthony Asare
reproduced from JITF Spring 2009

At the undergraduate level, capital structure and dividend policy are generally introduced in a basic finance class and further developed in advanced courses in corporate finance. Exposure to the concept of shareholder wealth maximization earlier in the curriculum would be beneficial for student understanding of business decision-making. It is difficult to grasp the complexities of the process without some basic appreciation of the financing aspect of those business decisions. This paper outlines a pedagogical method for incorporating capital structure and dividend policy decisions into an Introduction to Business course through the use of a business simulation.


Educators are being challenged with applying new pedagogical approaches that satisfy the needs of the next generation of students who have grown up with immersive, computer mediated experiences (Lynch and Tunstall 2008; Nadolski et al. 2008). A growing body of evidence suggests that well-designed and relevant simulations can help students learn complex materials relatively easier (Lynch and Tunstall 2008). Recognizing the ability of computer simulations to serve as effective learning tools for complex material and also the fact that students in this generation feel comfortable utilizing immersive computer related tools, this paper examines the use of a business simulation tool to teach complex materials in an undergraduate curriculum.

This paper explores how the complex topics of capital structure and dividend policy can be effectively introduced early in the typical undergraduate curriculum in an Introduction to Business course using a business simulation. It also explores some of the challenges and opportunities facing schools that might decide to adopt the use of business simulations in their undergraduate finance curricula. The paper is organized as follows: we describe the introductory course; we then describe the simulation; we next describe the capital structure and dividend policy decision-making strategies required; finally, we conclude with a summary and discussion of challenges and opportunities.


At Quinnipiac University all incoming freshmen and new business majors are required to take a sequence of courses in their first semester that includes Introduction to Business. The Introduction to Business (SB101) course along with Introduction to Information Technology (ISM101) and Personal Effectiveness (SB111) make up the first semester core business sequence. The students take this sequence of three courses as a cohort with projects and assignments that are interrelated across courses. SB101 is taught by a cadre of four to five tenure-track professors with representation from the management, marketing, accounting, and finance departments. The course objectives include exposing the students to the four primary functional areas of business (operations management, marketing, accounting, and finance), emphasizing the interdependency of the functional areas, and developing basic skills required by business students (team-building, leadership, decision-making, business writing, and presentation delivery). While each professor teaches the entire course to their assigned sections, weekly team meetings among the cadre ensure that the respective subject matter expert emphasizes the key goals and objectives for the upcoming topics.

The interdisciplinary course is structured around a computer business simulation called MikesBikes Intro® created by SmartSims Inc. The integrated business simulation ensures that students are forced to apply textbook concepts in an experiential learning exercise shortly after those topics are discussed in class. The sequence of topics is presented in Table 1 with the topic followed by the functional area designated as the subject matter expert responsible for the content of each.

Table 1. Topics and Responsibility.


Teamwork and Team-Building
Planning, Organizing, Leading and Controlling
Financial Statements
Market Segmentation
Demand Forecasting
Production and Inventory Management
Promotion Strategies
Ratio Analysis
Capacity Planning
Debt Capital
Equity Capital
Dividend Policy



A comprehensive survey of literature related to the use of business simulations in courses is Faria (2001). As Faria notes, research has shown that team member personality traits are a major factor in performance in business simulations (Armenakis, Field, & Holley, 1974; Johnson & Landon, 1974; Napier, 1974). Early in the course as part of the teamwork and teambuilding topic, the students complete a modified version of the Herrmann Brain Dominance exercise (Herrmann, 1981) and are divided into seven teams of four to six students. The size of the teams is determined by the size of the section with the constraint that the simulation allows a maximum of seven teams per class. The goal is that each team has at least one student from each of the personality trait groups. For many students, this is their first experience in a team-related educational project.

As a result, the freshman business program emphasizes interpersonal behavior and group dynamics through workshops and course material. These teams work together throughout the SB101 course and the other two courses in the freshman business sequence. Competition within the business simulation is the primary activity for these student teams throughout the semester.


The course is structured so that the topics covered are immediately followed by an application of the concept through a decision- making challenge in the business simulation. The course relies on a custom textbook where the topics are arranged in the order of the required decisions within the simulation. This approach allows the students to immediately relate the concepts covered in the course to actions required to run the company within the simulation.

The MikesBikes Intro® business simulation package allows the student teams to compete among the seven teams within each section. The number of sections (known as “worlds” within the simulation) ranges from twelve to fifteen depending on enrollment levels in the course. Each world contains seven bicycle manufacturing companies in a multi-period simulation. The simulation begins with all seven teams having identical companies and an identical mountain bike product to “sell.” The teams must initially make very basic decisions such as naming their company and devising a promotion strategy for their existing product. With each subsequent period the decisions become increasingly complex and numerous. The exercise culminates with the launch of a new product (either a highpriced road bike, a redesigned mountain bike, or a low-priced youth bike). By the fourth fiscal year in the simulation each team must make a full set of corporate decisions that include promotion expenditures, product selection, product specifications, production quantity, production capacity, plant efficiency and quality, capital structure, and dividend policy. The number of periods can vary but we elect to end after the seventh “rollover” (eight fiscal years).

The winner of the simulation contest is that team with the highest shareholder value. Shareholder value is calculated through a proprietary algorithm developed by SmartSims Inc. described in Equation 1:

SHV = f(EPS, D/E, DIV)

The variables in Equation 1 are defined as follows:

  • SHV is shareholder value defined as market share price plus cumulative dividend payments
  • EPS is earnings per share defined as net income divided by shares outstanding
  • D/E is the debt to equity ratio defined as book value of debt divided by book value of equity
  • DIV is the dividend payment history of the company.


In the MikesBikes Intro® simulation, the student teams launch a new product in the fourth period. During the third period, students must determine whether the company has adequate plant capacity to produce the new product. In addition to the capacity decision, there are significant product development, quality improvement, and promotion strategy costs that are also associated with the new product launch.

At this stage of the course and at this point in the simulation, we introduce the concepts of capital structure. In accordance with the Pecking Order Hypothesis (Myers 1984), the simulation algorithm rewards those student teams that are able to finance their new product launch with internally generated funds. Very few teams are capable of financing the new product launch without raising external capital. The basic tradeoff the teams must make is a comparison of the cost of expanding capacity to launch a new product versus the opportunity cost of lost sales and unsatisfied demand. The nature of the shareholder value algorithm forces the students to thoroughly consider the pros and cons of debt and equity financing and assess the impact of each on the resulting share price.

The first option available to the students is debt. Debt financing is available in the short-term (through an overdraft facility that must be paid back in one year at approximately 20% interest) or in the longer-term (through a three-year maturity bond with an 8% annual coupon rate). The student teams are limited in the amount of debt capital they are able to raise based on the financial condition of the company and a maximum amount available on the decision screen. The bond decision screen can be seen in Exhibit 1:

Exhibit 1. The Debt Capital Decision Screen


As the company takes on additional debt and the debt-to-equity level increases, the students will see an immediate negative impact on the share price within the simulation. Also, as the company’s financial condition changes, the required rate of return on debt changes as well and bonds will sell at a premium or discount. For example, the bonds in Exhibit 1 are selling at a discount indicating that the firm’s cost of debt has increased from 8% since the bonds were issued. This is likely due to the increased riskiness the simulation applies to a firm with the relatively high debt level illustrated in the example.

In the period after debt financing becomes available, the students are given the option of raising equity capital within the simulation. The student teams are also limited in the amount of equity capital they are able to raise based on the financial condition of the company. The decision screen for equity allows for equity issuance, share repurchase, and dividend payments and can be seen in Exhibit 2:

Exhibit 2. The Equity Capital Decision Screen


Companies may issue up to 50% of the market value of existing equity during any period but may repurchase only 10% in each decision period (called a “rollover” in the simulation). Further, stock is issued at a 5% discount to current share price (representing flotation costs and market reactions to equity issuance) and equity is repurchased at a 5% premium to current share price. Raising capital through an equity sale has a positive effect on the shareholder value through lowering firm’s debt-toequity ratio, but a negative effect on the shareholder value through lowering earnings per share (as the number of shares outstanding increases). These effects partially offset each other, but the overall impact on share price to an equity issue is generally negative. The other option available on the equity decision screen seen in Exhibit 2 is the ability to pay a dividend. The simulation limits the amount of dividends that can be paid to 50% of the firm’s retained earnings. The fact that dividends have a positive effect on share value tends to support the Gordon (1963) or Lintner (1962) proposition that dividend payments increase firm value as opposed to the dividend irrelevance theory of Modigliani and Miller (1961). Generally, teams only issue a dividend when there is no other more productive use of the cash.

The choice between debt and equity for external financing affords the opportunity to explore several additional capital structure considerations. Perhaps the most important distinction between the two sources of capital that is made obvious to the student is the discretionary nature of dividend payments on stock compared to interest payment obligations incurred with a bond issue. Next, interest payments are a taxdeductible expense in the simulation while dividends are paid from after-tax profits. Finally, students observe the impact of leverage and cost of capital through analysis of the financing decisions within the simulation.

Once the new product is launched, students will have three additional periods to refine their strategy and compete for the highest shareholder value. Teams can (and do) go bankrupt if they have an unsuccessful product launch or make serious errors, but the instructor has the option of providing an “emergency equity injection” to allow every team the opportunity to complete the game. For the successful teams, financial strategies become as critical as product marketing and capacity/ inventory management strategies to winning the game. Indeed professors teaching the course observe that teams often overemphasize the importance financial strategies designed to improve shareholder value at the expense of other value-creating operational decisions.

Within each world, the teams with adequate cash spend the last three periods making decisions that are intended to drive the share price as high as possible. For many teams, this will involve launching additional new products, improving existing products, potentially selling off excess capacity (at a 50% discount), and implementation of cost control measures. Further, many teams will use available cash to repurchase shares, pay off outstanding debt, and pay dividends in an effort to increase earnings per share, reduce the debt-to-equity ratio, and increase cumulative dividends paid. MikesBikes Intro® has restrictions on the capital structure and dividend decisions to avoid teams being able to “game the system” at the end of the simulation. Recall the 5% premium on repurchased shares, and restriction that teams may not buy more than 10% of the outstanding shares in any one period. Once the maximum number of shares has been repurchased (and earnings concentrated as much as possible), the amount of excess cash that can be paid out as dividends is limited to 50% of the value of retained earnings account from the balance sheet. The winning team is typically a team that had a successful product launch, excellent demand forecasting ability, and a thorough understanding of capacity planning and inventory management. Further, the winning teams always employ at least one, if not all, of the capital structure and dividend policy strategies described above.


In the seven years that Quinnipiac University has used the MikesBikes Intro® simulation package as part of the Introduction to Business course, we have found it to be an effective way to introduce many complex business topics that are often not fully understood by the students until much later in their undergraduate curriculum. In this paper we have focused on the key financial considerations of capital structure and dividend policy. The students are introduced to the concepts as a topic in Introduction to Business and the concepts are reinforced during the later Corporate Financial Management course. For finance majors, the topic is explored yet again in the required Intermediate Corporate Finance course. At the end of the course, the professors in the course conduct a survey of the students. One of the questions seeks to gauge the student perception of understanding in each of the functional areas of business. The survey asks the following question: “Indicate how much you feel you learned about each of the functional areas of business through the simulation.” The responses are on a 5-point scale with 5 being the highest (“A Great Deal”) and 1 being the lowest (“Nothing”). The results of this survey question are presented in Table 2.

The survey indicates that almost 90% of the students in the course feel that they learned some or a great deal about finance through the use of the simulation. A similar question asks them to rate their perceptions on learning with respect to specific learning goals. The survey asks the following question: “Using the rating scale below, indicate how well you feel you learned the concepts through the simulation.” Again, the responses are on a 5-point scale with 5 being the highest (“Learned Very Well”) and 1 being the lowest (“Didn’t Learn at All”). The results of this survey question are presented in Table 3.

For purposes of this study, we focus primarily on the dividend policy and debt/equity financing concepts where the vast majority of students (approximately 85% of respondents) feel they learned the concepts in either the “learned a great deal” or “learned somewhat” categories. While student perceptions of learning are often biased, we interpret these results to mean that the students generally view the simulation as a valuable part of the course and useful learning experience.

The use of the simulation to introduce the capital structure and dividend policy concepts does present challenges. First, a student could possibly learn to correctly change the numbers in the decision screen within the simulation with little to no understanding of the underlying concept behind the impact on shareholder value. Indeed, many of the students that perform best in the simulation are not the same students that perform best in the course. Student grades are determined by exam scores, assignment grades, and a course project that requires synthesis of the course content with simulation outcomes. Another major concern is that students equate success in the simulation with earnings manipulation and accounting “shell games.” In this era of rampant accounting fraud, an academic exercise that rewards potentially questionable financial decision-making is a serious concern. Finally, the nature of the team decision-making does little to ensure that every student on each team fully understands the application of the concept. Often, one or two students on a team make a majority of the decisions and “free riders” are allowed to coast along. We attempt to mitigate this situation by weighting the course project grade by a peer evaluation grade, but invariably some team members “slip through the cracks.” Unfortunately, it is impractical to run the simulation with companies run by individual students. However, every student in the course takes a common final exam specifically designed around the MikesBikes Intro® simulation. This evaluation helps ensure that all students are knowledgeable in the concepts developed during the exercise.

Overall, the MikesBikes Intro® simulation is perceived by students and faculty alike as a positive experience for the new business students. First and foremost, it exposes the students to the different functional areas of business within a company, shows how decisions within those functional areas are interrelated, and demonstrates the importance of each for a successfully managed company. For finance in particular, the positives far outweigh the negatives. Even for students that may not fully grasp the concept during the Introduction to Business course, instructors in upper-level finance courses are frequently able to refer back to the shared MikesBikes Intro ® experiences when these topics are revisited later in the curriculum. Instilling the concept that maximizing shareholder value leads to success in business at an early stage of the curriculum establishes a solid foundation for students in the finance major. Understanding by beginning business students of the impact of capital structure and dividend policy on shareholder value, even at a very basic level, makes the simulation worth consideration for inclusion in an undergraduate business curriculum.

Table 2: Survey Results – Functional Areas of Business

A Great Deal Some A Little Very Little Nothing Rating Average Response Count
Management 152 210 30 9 1 4.25 402
Accounting 110 203 64 22 4 3.98 403
Marketing 209 164 22 7 1 4.42 403
Finance 180 181 31 8 2 4.32 402
answered question 403 403
skipped question 15 15


Table 3: Survey Results – Specific Business Concepts

Learned very well Learned Somewhat Neutral Didn’t Learn Much Didn’t Learn at All Rating Average Response Count
Integration of functional areas 133 207 55 6 1 4.16 402
Industry analysis 154 203 42 2 1 4.26 402
Competitive Advantage 204 164 30 2 1 4.42 401
Competitor Analysis 189 169 40 4 1 4.34 403
Creating an effective mission statement 147 179 62 8 5 4.13 401
Analyzing and using financial data 186 173 38 2 1 4.35 400
Forecasting demand 194 163 34 9 2 4.34 402
Capacity Planning 187 165 40 7 2 4.32 401
Debt versus equity for financing 197 141 52 8 4 4.29 402
Dividend policy 174 161 53 11 3 4.22 402
Effective decision making 225 142 33 1 1 4.47 402
Impact of decisions made on firm outcomes 221 143 36 1 1 4.45 402
answered question 403 403
skipped question 15 15



Armenakis, A., Feild, H., & Holley,W. (1974). Correlates of satisfaction, learning and success in business gaming. Simulations, Games and Experiential Learning Techniques, 1, 272-277.

Faria, A., (2001). The changing nature of business simulation/ gaming research: A brief history. Simulation & Gaming, 32:1, 97-110.

Gordon, M. (1963). Optimal investment and financing policy. Journal of Finance, 18:2, 264-272.

Herrmann, N., (1981). The creative brain. Training and Development Journal, 35:10, 10-16.

Johnson, G.,& Landon, L. (1974). Identifying successful game participants. Simulations, Games and Experiential Learning Techniques, 1, 295-299.

Lintner, J. (1962). Dividends, earnings, leverage, stock prices, and the supply of capital to corporations. Review of Economics and Statistics, 44:3, 243-269.

Lynch, M., and R. Tunstall (2008). When worlds collide: Developing game-design partnerships in universities. Simulation & Gaming , 39:3, 379-398.

Miller, M. and F. Modigliani (1961). Dividend policy, growth, and the valuation of shares. Journal of Finance, 34:4, 411- 433.

Myers, S., (1984). The capital structure puzzle. Journal of Finance, 39:3, 575-592.

Nadolski, R.J., Hummel, H.G.K., van den Brink H.J., Hoefakker R.E., Slootmaker A., Kurvers H.J., and Storm., J (2008). EMERGO: A methodology and toolkit for developing serious games in higher education, Simulation & Gaming, 39:3, 338-352.

Nadolski, R.J., Hummel, H.G.K., van den Brink H.J., Hoefakker R.E., Slootmaker A., Kurvers H.J., and Storm., J (2008). EMERGO: A methodology and toolkit for developing serious games in higher education, Simulation & Gaming, 39:3, 338-352.

Napier, H. (1974). Autocratic vs. democratic decision making. Simulations, Games and Experiential Learning Techniques, 1, 291-294.

Sean Reid (Finance), Len LaBonia (Marketing),Ben Shaw- Ching Liu (Marketing), Patrice Luoma (Management), and Anthony Asare (Marketing) are professors at Quinnipiac University. Together, they team teach the freshman integrated business course described above.

This article has been reproduced in its entirety from the Spring 2009 edition of the Journal of Instructional Techniques in Finance (JITF)

Published by the Institute of Finance Case Research (IFCR)

Mikes Bikes Case Study Article PDF