216 Understanding the Numbers
In our case, the budgeted fixed costs were $4,000 and the actual fixed costs
were $4,680. The variance was simply $680, which means a 17% increase.
“Fixed costs are more difficult to control than variable costs because one
cannot create an illusion of control through the elaborate computation of price,
mix, and usage variances or indices.”
“How, then, does one control fixed costs?” asked Tom.
“First,” Jane replied, “one must recognize that if costs are truly fixed,
there is no reason to control them. Consider depreciation costs as an example.
Once one has purchased an item, the total depreciation costs are set—unless
one disposes of the machinery when a disposal cost will substitute for the de-
preciation cost. No control is possible here. The control in this case has to be
exerted when the machinery is purchased. Thereafter, it is a sunk cost that
cannot be controlled. In other words, controlling fixed costs is in the first place
a matter of timing.
“Traditional variance analysis uses one cost driver only, the volume of
production. More modern variance analysis, such as that in activity based cost-
ing,uses multiple cost drivers.^4 For example, setup costs may not vary with vol-
ume of production but might vary with the number of batches. What appears
at first glance to be a fixed cost may just be variable with respect to some other
driver. The analysis of variance proceeds exactly as before except that one
changes the driver from units produced to number of batches. One converts
the fixed cost into a quasi-variable cost by finding and using the appropriate
cost driver.
“Controlling fixed costs is also a matter of scale. Consider the machine
again. Assume one has just one machine with a capacity of 1,000 boxes of
greeting cards per day. Its cost is certainly fixed within this range. However, if
the analysis is being done in terms of tens of thousands of boxes, and if the
corporation has a hundred of these machines, then it is possible to think of ma-
chine costs as being a variable. One can ask, in other words, what the cost
would be to produce an additional ‘unit’ of 1,000 boxes.
“This last question points to the fact that most fixed costs are usually only
fixed within the context of a particular analysis. Consider, for instance, the ink
you use in production. Assume its price is reset by a cartel every three months.
Assume also that its planned usage is reset at the same time. A budgetary con-
trol system that computed variances every month and set the budgeted price
and quantity to those of the latest quarter might show a variance of zero each
month. This might lead everyone to believe that they were dealing with a fixed
cost. However, were the same analysis to be done on an annual basis, with
prices and quantities set at the start of the year, a substantial variance could
arise. The example points up the old truism that all costs are variable in the
long run.
“The example above also points up the need to set your net large enough
to catch the fish you want. Many fixed costs cannot be controlled by a monthly,
or even annual, budget system because they change too slowly. One needs a
coarser net, that is, an annual, triennial, or even longer budgetary system to