Marshall School of Business
University of Southern California
Controls at the Bellagio Casino Resort
Teaching Note
Purpose of Case
The Controls at the Bellagio Casino Resort case was designed with several purposes in mind.
First, it illustrates a control system that is dominated by action and personnel controls, rather
than results controls. The analysis of this system leads to insights about some of the factors that
limit the feasibility of results controls.
Second, the case can lead into a discussion of what is meant by the term tight control. The
Suggested Assignment Questions
Since the case can be used for several purposes, the assignment can be slanted in several ways.
The following is one list of questions that might be used
1. Focus on three key roles at mainly three levels of authority in the casino blackjack dealers,
pit bosses, and the vice president of table games. How would you characterize the control
strategy (e.g., tight vs. loose) used over each of these roles?
This note was prepared by Professors Kenneth A. Merchant (University of Southern California) and Wim A. Van der Stede (London
School of Economics) for the sole purpose of aiding classroom instructors in the use of the Controls at the Bellagio Casino Resor
t
case. It provides analysis and questions that are intended to present alternative approaches to deepening students’ comprehension o
f
business issues and energizing classroom discussion.
144
2. Prepare a list of the controls described in the case. What control problems are they designed
to address? Are the managers interviewed for the case justified in being proud of their
companys control system? Why or why not?
3. Are any of the control systems in place at the Bellagio suitable for firms in other industries?
Case Analysis
Key Roles and Controls
We suggest starting the class by analyzing each of the organizational roles and the controls over
them. (This addresses Suggested Assignment Question 1.) Start with the lowest level role
described in the caseblackjack dealerand build the chart shown in Attachment TN-1 on the
blackboard.7 (Some students are not familiar with the blackjack game, so some discussion of
how it works and what the dealer is expected to do may be necessary.) One point that comes out
of this chart is that results controls are not important at the dealer and pit boss levels.
One major categorization of control systems is tight or loose. Asking students whether control
Evaluation of Controls
The evaluation of the sets of controls in use has to be based on some consideration of what can
go wrong. (This addresses Suggested Assignment Question 2.) A provisional, but not
exhaustive, list of things that can go wrong at the blackjack tables is presented in Attachment
TN-2. Ask the students to build such a list (supplemented by items on the list shown where
necessary) and think about whether the controls in place were sufficient to reduce the incidence
of each undesirable activity to a reasonably low level.
Merchant & Van der Stede, Management Control Systems, 3rd edition, Instructors Manual
145
taken steps to reduce their incidence. As a consequence, the evaluation of the controls in place
should be very favorable as long as cost constraints are kept in mind.
Another problem that casinos face is skimming. Skimming means that money is taken from
the casino, such as cash from the cage or counting room often requiring cover through forged
documentation and changed audit trails. This problem can be used to lead into a discussion of
Weakness of Results Controls
As part of the discussion of Suggested Assignment Question 2, instructors should point out that
results controls are not very important in the casino, and it is important to spend some time to
understand why this is so. Results controls are simply not feasible for control over the dealers
because results are not measured until the 24-hour drop procedure is complete, and during that
period more than one dealer has worked at a table. It would be possible to measure dealer
productivity by keeping track of deals per hour, but this is not done. Modern technology, such
as RFID, actually would make this quite feasible, but the Bellagio management has so far
decided against implementing such a system.
Instructors may wish to spend some time discussing the advantages and disadvantages of RFID
technology that could be used to track deals. Beyond relatively easily quantifiable costs and
benefits (e.g., the cost of replacing the chips vs. the benefit of being able to design better
Merchant & Van der Stede, Management Control Systems, 3rd edition, Instructors Manual
146
As an illustration of the limitations of the results measures, instructors can refer to Exhibit 12,
which indicates rather large drop variances across the 24 blackjack tables listed (from 28.52%
to +37.71%). Drop is a measure of the gambling activity at the tables, but negative drop
variances cannot be taken to mean that the dealers were not productive. The drop is dependent
on the tables minimum and the number and type of players who chose to play at that table on
The results measures are more reliable, the larger the increment of the casino being measured
(e.g., an entire shift vs. a single table) and the longer the periods of time involved; hence, the
GCBs requirement to investigate, on a monthly basis, all statistical fluctuations by game type in
excess of ±5% resulting from the comparison of the previous calendar year to that of the current
Applicability to Other Industries
Are the controls observed at the Bellagio Casino applicable to other industries? (This addresses
Suggested Assignment Question 3.) The casino industry is in some ways unique, but some of
the principles included in the Bellagio control systems are of use in other firms. Tight security
and good controls over movements of assets are necessary where valuable, liquid assets are
Pedagogy
Our experience with this case suggests that the discussion should be relatively structured. A few
students have never been in a casino and do not know how the game of blackjack works, and
Merchant & Van der Stede, Management Control Systems, 3rd edition, Instructors Manual
147
Attachment TN-1: Analysis of roles and controls
Role Key actions Controls Evaluation (in light
of things that can
go wrong)
Blackjack
dealer
– Accurate, honest, deals
– Productive
– Cheerful
– Licensing (P)
– Hiring (P)
– Training (P)
– Frequent rest breaks (P)
– Supervision (A)
– Surveillance (A)
Good, tight control;
some risks
Code:
R = results control
Merchant & Van der Stede, Management Control Systems, 3rd edition, Instructors Manual
148
Attachment TN-2: Things that can go wrong at a blackjack table
2. Dealer payoff of a losing bet.
4. Player past posting (i.e., adding a bet to hand that looks promising) or copping (i.e.,
decreasing a bet on a hand that looks lost).
5. Dealers stealing chips. Dealers have discovered many devious ways to remove chips from
7. Players introducing counterfeit money or chips.
8. Pit bosses granting bad credit. (The casinos protect themselves by having this decision made
10. Players marking or switching cards.