It usesmore observations, it allows a visual, informal analysis of goodness of fit, and it allows users to spot outliers or problematic patterns of cost such ascurvilinear behavior or ki
Trang 1of capacity will reflect inefficiencies from sacrificing efficiency to
increase output. The company will hire inexperienced workers, expedite deliveries of materials, and take other actions that will increase costs. Ifthey can sell all that they can make, the company will be very profitable, and cost control is not likely to be a high priority
33 Methods of Cost Behavior Analysis
The highlow method is quick and easy, but uses only two observations and so is seriously deficient. It assumes that the two selected points are representative. It gives no indication of how accurate predictions are likely to be
The scatterdiagram method is better than the highlow method. It usesmore observations, it allows a visual, informal analysis of goodness of fit, and it allows users to spot outliers or problematic patterns of cost (such ascurvilinear behavior or kinks that indicate two or more cost functions)
Regression analysis gives more precise results than scatterdiagrams, gives formal measures of goodness of fit, and permits the use of more than one independent variable. By itself, it does not allow the user to spot outliers or problematic cost patterns
Trang 2It needs to be made clear with respect to what decision a cost is
avoidable. You might wish to ask for other decisions or types of decisions for which each cost might be avoided.
(f) Avoidable and direct
The pattern that emerges is that avoidable costs are typically direct, indirect costs usually unavoidable, but not necessarily vice versa. Chapter
5 discusses situations in which indirect, common costs might be avoided, essentially when dropping a segment so greatly reduces workload that a
Trang 3Revenues 0.02902 0.00871 3.333 0.013 0.00843 0.04960 The results are quite different. The regression line is $11,928 + 0.029 x
Trang 41.
Cost of
Goods Sold Selling Administrative Cost at high volume $54,000 $ 8,800 $9,400 Cost at low volume 48,000 8,500 9,200 Change in cost $ 6,000 $ 300 $ 200 Divided by change in volume $10,000 $10,000 $10,000 Equals variable cost percentage 60% 3% 2% Total cost at sales of $90,000 $54,000 $ 8,800 $ 9,400 Variable cost portion ($90,000 x
variable cost percentage) 54,000 2,700 1,800 Fixed portion of cost $ $ 6,100 $ 7,600
You might wish to make the point that, for a nonmanufacturing company, cost of goods sold should be wholly variable (not mixed, as it can be for a manufacturer). Of course, cost of sales might not be exactly the same
percentage of sales from period to period even if selling prices are
constant. Changes in the percentage of cost of sales to sales between two periods could result from a change in purchase prices or sales mix (covered
Trang 5The standard error of the estimate, tells us how close our predictions are likely to be to the actual results. In this case, we expect predictions
to be within $9,329 about 68% of the time, and within $18,658 (2 x $9,329) about 95% of the time. (This is a bit rough and does not tell us the
confidence interval for a single prediction but for the average. This point
is probably not important to most classes.)
It is also helpful to understand what the results do not tell you. The equation is not necessarily the best available. Some other factor might predict better. Multiple regression with some other factors might give better results in the form of a higher rsquared and a lower standard error. The intercept, $101,187, is not the estimate of total cost at zero
revenue. The data were collected in the range of $800,000 to $1,200,000, and
it is unsafe to extrapolate outside that range
Note to the Instructor: Students often ask what a good value is for r2. The best answer, that it depends on the data, is not too satisfying. Anythingabove .50 is probably quite good for cost data in a complicated environment. Evaluating the standard error requires examining the relationship of the error to the total cost. For instance, at $1,000,000 hours (midpoint of range) predicted cost is $190,987. The standard error of $9,329 is about 5%
of predicted cost, which is a pretty good fit.
311 Interpreting Behavior Patterns (1015 minutes)
1. The first step is to determine just what the behavior is. The first set
of observations shows a relatively low fixed cost and a rapidly increasing total cost. This indicates that variable cost is relatively high. The second segment shows a jump above the level of the first segment, with a flattening of the total cost line, indicating a decline in variable cost per unit of activity. The third segment shows much the same: a jump in the level of costs, with a further flattering of the slope of the cost line. One possible explanation for the observed behavior is that "variable" costs per unit drop as volume increases, with increases in stepvariable costs accounting for the jumps. If the cost were total manufacturing cost, apossible explanation is that materials were subject to quantity discounts andwere a large proportion of total costs, with jumps in cost occurring because
of stepvariable costs such as supervision
Another possible explanation is that the company has three alternative methods of production, with increasing amounts of machinery causing the jumps
in cost and increased efficiency in the use of labor and materials causing
Trang 6actually operate from near zero to near full capacity in this manner at shortnotice, unless the machinery could be rented at short notice. Hence, the cost behavior under this explanation should be viewed as relatively longterm
2. Planning for the costs should be relatively simple if the range within which the company expects to operate was relatively certain. Three differentlines would be drawn and used in prediction, depending on the range in which volume was expected to occur. A single line could not be a good predictor.
Manufacturing at 30% $330.0 $348.0
S&A at 20% 220.0 550.0 232.0 580.0 Contribution margin 550.0 580.0 Fixed costs:
Manufacturing 340.0 340.0
S&A 150.0 490.0 150.0 490.0 Income $ 60.0 $ 90.0
Several comments apply here. First, some students do not understand thatrecasting income statements does not change profit, only the form of the statement. Second, the contribution margin format allows us to do CVP
analysis, which we could not with the functional income statements. We can, for example, determine the breakeven point because we know that contributionmargin is 50% (100% 30% 20%) and total fixed costs are $490.0:
$490.0/50% = $980.0
We can also calculate sales volumes required for target profits and do other planning that is impossible without knowledge of cost behavior
313 Relationships (15 minutes)
Trang 7(a) $2.00 $8 selling price less $6 contribution margin per unit ($600,000/100,000)
(c) $230,000 $200,000 current income + additional contribution margin of $30,000 (5,000 x $6), or 105,000 x $6 =
$630,000 total contribution margin less $400,000 fixed costs = $230,000
rounded
Plus expected variable cost per unit 4.80
Required price $13.54
4. $9,600 increase in profit. Either the total or incremental approaches could be used here. Using the total approach,
Expected total contribution margin ($7.20 x 83,000) $597,600 Expected fixed costs ($432,000 + $12,000) 444,000 Expected total profit 153,600 Profit expected without additional expenditure (80,000 x $1.80) 144,000 Increase in profit $ 9,600Using the incremental approach,
Trang 8of a breakeven point.
$12,000/$7.20 = 1,667 units
Because 1,667 is well below the expected 3,000 units, the company is probablywelladvised to go ahead. Had the indifference point been, say, 2,800 units,
a reasonable manager might believe that the risk is too great because a relatively small shortfall would wipe out the additional profit
Total costs at $800,000 sales $800,000 $80,000 profit $720,000 Total variable costs ($800,000 x 70%) 560,000 Total fixed costs $160,000
3. $50,000
Contribution margin ($700,000 x 30%) $210,000 Fixed costs 160,000 Profit $ 50,000
Or, the decreased sales of $100,000 decrease profit by $30,000 ($100,000 x 30% CM%), from $80,000 to $50,000
4. 5.55% The easiest way to approach this requirement is to use the basic profit equation. Cost of sales remains at $480,000, ($800,000 x 60%)
Trang 91. C 2. F 3. K 4. B 5. A 6. D 7. J 8. E or H
9. L (Item 9 is different from the original CPA problem.) 10. G
Many of the answers assume that the use of the cost element is at least partly variable with production. Item 3 is an example. The cost of water asthe use of water increases is described by graph K. It is assumed that increases in production cause proportional increases in the amount of water consumed. It is possible, but unlikely, that the use of water is relatively constant whatever production is. It is also possible that 1,000,000 gallons
or more is the base amount, with additional water being related to
production.
Note to the Instructor: Although we did not show the vertical segments
of stepvariable costs in the text, students have had little difficulty with cost graphs such as item 7. You might wish to point out that graph J is technically incorrect because the cost is discontinuous, jumping from one level to another. This poses no real problem in a practical situation
because the portions of discontinuity are quite small. In this case, a single machinehour at the breaking point gives the jumps, and it is unlikelythat any company could be so precise in its hiring practices
Some students will wonder why the second segment of the line in graph H (item 8) tilts upward instead of being parallel to the horizontal axis. The reason seems to be that although the labor force is "constant in number," it could be changing in composition because of turnover. It is also possible that some workers earn annual wages of less than $8,500. Graph E is a good answer if (a) there is no turnover and all workers earning more than $8,500
317 CVP Review (20 minutes)
1. 183 sweaters, rounded ($5,000 + $6,000)/($100 $40) = $11,000/$60 Use the highlow method to determine fixed and variable costs:
$100 $30 (10% x $100) = $60 new contribution margin, same as now ($5,000 + $4,000)/$60 = 150
The proposed agreement gives the supplier the same total compensation at the
$100 selling price. At higher selling prices the supplier will take a largershare. In all likelihood, Mia will raise prices in the future, making the proposed arrangement more attractive to the supplier, less attractive to Mia
Trang 10With no change in fixed or variable costs, a $60,000 increase in profit ($100,000 desired vs. $40,000 earned last year) requires a $60,000 increase
in contribution margin at a volume (current level) of 100,000. Hence, an increase of $0.60 ($60,000/100,000) is necessary
Competition will determine whether we can achieve the expected volume with the higher price
(b) Reduce variable cost per unit to $5.40, a decrease of $0.60 per unit.The logic here is the same as in requirement (a). Contribution margin
must increase by $0.60 per unit, and with a constant selling price of $10the perunit variable cost must decline $0.60 from $6
We might change suppliers to reduce variable costs, but such a step couldreduce the quality of the product. We should look for activities/costs that do not add value to the product
(c) Fixed costs must decrease by $60,000, to $300,000
If profit is to increase $50,000 and contribution margin is to remain thesame, fixed costs must be reduced by an amount equal to the desired
increase in profit
We can easily reduce some fixed costs, but again the question is whether we might run into other difficulties. We can always reduce discretionary costs such as advertising, but perhaps at the cost of reduced sales. We could cut other costs that might harm us in the long run. Such costs include employee training and maintenance
(d) Increase sales to 115,000, an increase of 15%.
Here again, if profit is to increase $60,000 without a change in fixed costs,total contribution margin must also increase by that amount. If selling price ($10) and variable cost per unit ($6) remain constant, contribution margin remains at $4, and it will require 15,000 more units ($60,000/$4) to produce the desired increase in profit
Increasing unit sales without increasing costs could be difficult. An expanding market would help, as would better service to our customers and a higherquality product. Achieving these improvements without increasing costs might not be possible
Trang 11The assistant merely connected the high and low points with his line, notconsidering the intervening observations. Moreover, the high and low points are at volumes far removed from the other observations, so we should questionwhether they are within the relevant range. Because the cost is maintenance,
we might even expect a relatively higherthannormal cost at low activity because there is then more time for performing the work. Similarly, at the high point we might expect lowerthannormal costs because of the inability
to perform work then, as well as managers' unwillingness to take limos in forservice during a peak period.
The assistant's line shows observations in almost equal numbers above andbelow, but the line would fit the majority (all but the high and low) of the observations better if it were tilted up and pushed down on the vertical axis. A line hitting the vertical axis at $100 and with a slope of $1.55 fitsnicely. At 500 hours, the cost is about $875. This line ignores the extremepoints. If the extreme points are to be considered, the line would tilt lessand the fixedcost component would be higher than $100. Putting a couple of alternatives on the board will help students see the differences that would arise from differing interpretations of the particular observations
320 Delta Airlines CVP Relationships (20 minutes)
1. The key is to find revenues and costs at breakeven to be able to use thehighlow method
Revenue and cost at breakeven = $14,881 ($16,741/.729) x .648
So, $15,003 $14,881 = $ 122 = 6.56% variable component $16,741 $14,881 $1,860
$1,738 = $215) to operating income. Of course, each drop reduces operatingincome by the same amount.
Trang 121. $75,992 $49,272 + ($1.78 x 12,000) + ($2.68 x 2,000)
2. $30.90, calculated as follows:
Materials $ 6.00 Labor, 2 hours x $10 20.00 Variable manufacturing overhead (2.0 x $1.78) + (.50 x $2.68) 4.90 Total variable cost $30.90
3. $40.815 (45 x $0.907) per batch of 100, or $0.40815 per unit
4. This question refers to measures of goodness of fit. The r2 of 79.25% indicates quite a good fit because 79.25% of the variation in power cost is associated with changes in machine hours. The specific requirement of the question gets at the meaning of the standard error. Actual cost should be within $9,497 of predicted cost about 68% of the time and within $18,994 (2 x
$9,497) about 95% of the time. At 100,000 hours, $18,994 is only 9.1% of predicted cost of over $209,345 (part 1).
5. No. There could be another simple regression equation or a multiple regression equation that predicts better. Only if the correlation is