ACCT 601– Managerial Accounting
Investigating Cost Drivers
Part A – Instructions
The object of this exercise to gain a better understanding of overhead costs through the use of regression analysis. In particular, you will investigate which service department activities appear to drive or cause departmental overhead. The data you will use is from a proprietary database which contains detailed departmental (i.e. USED, NEW CAR, SERVICE, BODY SHOP) financial information for approximately 250 U.S. car dealerships. The data is in EXCEL format and is called ABC.xls.
We will direct our attention to the service department overhead which includes the following costs: office supplies; ...view middle of the document...
.... + (7*dEMPL + (8*LOSS
[Check your results: The coefficient on (1 should be 0.04749]
Print your results. Explain your answers. Do your results (i.e. the signs and magnitudes) make sense based on your intuition ? What are the benefits of subcontracting out some of your work?
Generally most of the numbers make sense with the exception of “Loss”. While it is expected for overhead cost to fluctuate based on whether or not the company makes a profit, it does not make sense that “Loss” in and of itself would cause such a substantial change in overhead cost. The other activities fall into “reasonable” range of effect on overhead cost.
A2. Examine your results for the UCRET and UCWHOLE. The retail (new and used car) departments require that the service department fix any problems with trade-ins before they can then be sold. The dealership then keeps the more desirable trade-ins and sells them as used cars (UCRET). The less desirable “clunkers” are sold to wholesalers at auction (UCWHOLE). Does the department spend more overhead fixing the good trade-ins or the “clunkers” ? Is the difference statistically significant ? Can you say that they spend more on one or the other ? You may need to refer to the attached note on hypothesis testing of coefficients.
The difference between UCRET and UCWHOLE is not statistically significant. Therefore although it seems as though UCWHOLE causes more cost, we cannot conclude that the cost are not equal.
A3. What can you tell me about the difference between the overhead spent on internal work-orders (IRO) for the dealership (i.e. warranty work, used car trade-ins, promotional vehicles) versus the work orders that are for the paying customers (RRO) ? Again is there a statistical difference ?
Again the difference between IRO and RRO are not statistically significant. We cannot conclude that the cost are not equal.
Part B – Instructions
B1. Examining the data, you will find that of the 249 departments, 207 are profitable and 42 are operating at a loss. Split the data into two groups and repeat question A1. above for the two groups separately. IMPORTANT: DO NOT INCLUDE THE VARIABLE “LOSS” in your regressions for this part as it is the variable that you are using to separate the data. What can you tell me about profitable and loss-making departments ? What do your findings suggest ? Is there anything that you would like to look at further if you had more data ? If you wish you can also repeat questions 2 and 3 , however it is not required.
In profitable firms it seems that employees bring overhead cost down and profitable firms minimize UCRET and UCWHOLE cost in relation to unprofitable firms. We would like more information employee positions, for example how many are sales related and how many are administrative.
Employees seems to be an extraordinary cost...