CHAPTER TWO

Cost Concepts, Behavior

and Estimation
 2.1. Introduction
 The eventual aim of costing is to determine the cost of producing/providing a
product/service; for profitability analysis, selling price determination, stock
valuation purposes, etc.
 Organizations generate reports containing a variety of cost concepts and terms.
Managers/accountants must understand these concepts and terms to effectively
use the information provided.
 This chapter discusses cost concepts and terms that are the basis of
accounting information used for internal and external reporting. It also covers
Cost Classification and Behavior and Cost Estimation techniques.

MAF-562 2
 2.2. Cost Definition and Concepts
 Cost: is a resource sacrificed or forgone to achieve a specific objective. A cost is
usually measured as the monetary amount that must be paid to acquire goods and
services.
 Cost object: is anything for which a separate measurement of costs is desired.
Example: product, project, service, department, customer, etc.
 A cost system determine the costs of various cost objects in two basic stages:
accumulation, followed by assignment.
 For example, a company can collect (accumulate) costs in various categories such as
different types of materials, different classifications of labor, and costs incurred for
supervision. Managers/management accountants then assign the accumulated costs to
designated cost objects.
MAF-562 3
 Cost assignment: is a general term that includes:
a. Tracing accumulated costs: For direct costs
b. Allocating accumulated costs: For indirect costs
 Cost pool: A cost pool is a grouping of individual cost items possessing
identical nature.
 Cost driver is a variable, such as an activity level or volume, the change of
which causally affect costs over a given time span. That is, there is a specific
cause-and-effect relationship between change in level of activity or volume (i.e.
the cost driver) and change in level of cost.

MAF-562 4
 Relevant range is the normal activity level or volume in which there is a
specific relationship between the level of activity or volume and the cost in
question. For example, a fixed cost is fixed only in relation to a given wide
range of total activity or volume (at which the company is expected to
operate) and only for a given time span (usually a particular budget period).
 Other cost terms and concepts:

• Sunk cost • Unit/total cost

• Opportunity cost • Marginal/average cost
• Incremental cost • Prime Costs and Conversion Costs
• (un) avoidable cost
• Differential cost
MAF-562 5
 Cost Classification and Behavior

 Cost classification is the arrangement of cost items into logical

groups/common characteristics. For example: by their nature (materials,
wages etc.); function (administration, production etc.)….
 Different cost classifications are used to develop cost information for a given
purpose.
 Thus, different classification of costs provide different information that helps
for decision making.

MAF-562 6
 There are various criteria on the basis of which costs can be classified.
 The classification can be based on:
1. Natural/elements/analytical: refers to essential components or parts of the
total costs a cost object. These are: a. materials b. labor and c. overheads
(FOH/MOH)
2. Function/operation/purposes: this includes: production (Manufacturing)
consisting of direct materials, direct labor and manufacturing overhead),
and Non-manufacturing costs (selling, distribution, administration, R&D)

MAF-562 7
3. Traceability to a particular cost object:
 Direct: costs that have a direct relationship with the cost object and can be
traced to that cost object in an economically feasible (cost effective) way.
 For example, the cost of steel or tires is a direct cost of a vehicle/car of say
model X5 or Z4. The cost of the steel or tires can be easily traced to or
identified with each product. The workers on the product line request
materials from the warehouse and the material requisition document
identifies the cost of the materials supplied to the specific product.
 In a similar vein, individual workers/labor record the time spent working on
the product on time sheets. The cost of this labor can easily be traced to the
product and is another example of a direct cost. The term cost tracing is used
to describe the assignment of direct costs to a particular cost object.
MAF-562 8
 Indirect: costs that have a direct relationship with the cost object but
cannot be traced to that cost object in an economically feasible way.
 For example, the salaries of plant administrators (including the plant
manager) who oversee production of the many different types of cars
produced at a plant are an indirect cost of a product such as X5s. Plant
administration costs are related to the cost object (X5s) because plant
administration is necessary for managing the production of X5s.

MAF-562 9
 Plant administration costs are indirect costs because plant
administrators also oversee the production of other products, such as the
Z4, another product. Unlike the cost of steel or tires, there is no
requisition of plant administration services and it is virtually impossible
to trace plant administration costs to the X5 line. The term cost
allocation is used to describe the assignment of indirect costs to a
particular cost object.
 Cost assignment is a general term that encompasses both (1) tracing
direct costs to a cost object and (2) allocating indirect costs to a cost
object.

MAF-562 10
 Several factors affect the classification of a cost as direct or indirect:
 The materiality of the cost in question.
 Available information-gathering technology.
 Design of operations.
 A specific cost may be both a direct cost of one cost object and an
indirect cost of another cost object. That is, the direct/indirect
classification also depends on the choice of the cost object.

MAF-562 11
4. Behavior pattern:
 Fixed (are total costs that remain constant regardless of the level of
activity up to a certain relevant range but unit costs vary according to
the level of activity.)
 Variable (are total costs that changes in direct proportion to changes in the
level of activity but “unit costs remain constant”).
 Semi-variable or semi-fixed costs /stepped fixed costs:
 Costs are defined as variable or fixed with respect to a specific activity
and for a given time period(relevant range).

MAF-562 12
5. Controllability:
 Controllable : costs that can be influenced by management action
 Uncontrollable: cost beyond the control of management.
6. Normality:
 Normal: costs incurred in the normal course of activity for producing
normal level of output under normal circumstances.
 Abnormal: costs incurred on account of abnormal conditions or abnormal
activity.

MAF-562 13
7. Based on timing they are charged against revenue
 Product costs: are costs that are necessary and integral part of producing
(acquiring) the finished product. They are considered as an asset/inventory
when they are incurred. Under the matching principle, these costs do not
become expenses until the finished goods inventory is sold.
Example: Cost of direct material
 Period Costs: are costs of income statement other than cost of goods sold.
They are treated as expense of the period in which they are incurred because
they are expected to benefit revenue in the current period.
Example: Advertising costs

MAF-562 14
Cost Estimation
 A cost estimate, the product of the cost estimating process,  is the
approximation of the cost of a program, project, or operation.
 Cost estimation involves measurement/estimation of past relationship
based on data from past costs and the related level of an activity to
understand/determine how costs behave.
 This activity requires gathering information about costs and how they
behave so that managers can predict what they will be in the future.

MAF-562 15
The cost predictions are used in each of the management functions:
1. Strategic Management: Cost estimation is used to predict costs of
alternative activities, predict financial impacts of alternative strategic
choices, and to predict the costs of alternative implementation strategies.
2. Planning and Decision Making: Cost estimation is used to predict costs so
that management can determine and make a decision on the desirability of
alternative options and to budget expenditures, profits, and cash flows.

MAF-562 16
3. Management and Operational Control: Cost estimation is used to
develop cost standards, as a basis for evaluating performance.
4. Product and Service Costing: Cost estimation is used to allocate costs
to products and services or to charge users for jointly incurred costs.

MAF-562 17
Cost Estimation Methods
 Methods of cost estimation include: the account analysis method, the
conference method, the industrial engineering method and the quantitative
analysis methods .
 These methods differ with respect to how expensive they are to implement, the
assumptions they make, and the information they provide about the accuracy
of the estimated cost function. They are not mutually exclusive, and many
organizations use a combination of these methods.
 Each of the methods will be discussed one by one. However, before
considering these methods, it is better to see Basic Assumptions and
Examples of Cost Functions next.

MAF-562 18
 Basic Assumptions and Examples of Cost Functions
 Managers are able to understand cost behavior through cost functions.
 A cost function is a mathematical description of how a cost changes with
changes in the level of an activity relating to that cost.
 Often cost functions are estimated from past cost data.
 Cost functions can be plotted on a graph by measuring the level of an
activity, such as number of batches produced or number of machine hours
used, on the horizontal axis (called the x-axis) and the amount of total costs
corresponding to—or, preferably, dependent on—the levels of that activity
on the vertical axis (called the y-axis).

MAF-562 19
 Managers/Accountants often estimate cost functions based on two
assumptions:
1. Variations in the level of a single activity (the cost driver) explain the
variations in the related total costs.
2. Cost behavior is approximated by a linear cost function within the relevant
range.
 For a linear cost function represented graphically, total cost versus the level
of a single activity related to that cost is a straight line within the relevant
range.
 Note that not all cost functions are linear and can be explained by a single
activity i.e. there are cost functions that do not rely on the above assumptions.
MAF-562 20
 Basic types of linear cost functions
 To understand three basic types of linear cost functions and to see the role
of cost functions in business decisions, consider the negotiations between
Cannon Services and World Wide Communications (WWC) for exclusive
use of a videoconferencing line between New York and Paris.

MAF-562 21
 Alternative 1: $5 per minute used. Total cost to Cannon changes in
proportion to the number of minutes used. The number of minutes
used is the only factor whose change causes a change in total cost.
Under alternative 1, there is no fixed cost. We write the cost function in
this case as Y=$5X, where X measures the number of minutes used (on
the x-axis), and y measures the total cost of the minutes used (on the y-
axis) calculated using the cost function. The $5 is slope coefficient, the
amount by which total cost changes when a one-unit change occurs in
the level of activity (one minute of usage in the Cannon example).

MAF-562 22
 Alternative 2: Total cost will be fixed at $10,000 per month, regardless of
the number of minutes used.
In this, there is only fixed cost for Cannon Services. We write the cost function
in Panel B as y = $10,000
The fixed cost of $10,000 is called a constant; it is the component of total cost
that does not vary with changes in the level of the activity. Under alternative 2,
the constant accounts for all the cost because there is no variable cost.
Graphically, the slope coefficient of the cost function is zero; this cost function
intersects the y-axis at the constant value, and therefore the constant is also
called the intercept.

MAF-562 23
 Alternative 3: $3,000 per month plus $2 per minute used. This is an
example of a mixed cost. A mixed cost—also called a semi variable cost—
is a cost that has both fixed and variable elements. We write the cost
function in this case as Y = $3,000 + $2X
Cannon’s managers must understand the cost-behavior patterns in the three
alternatives to choose the best deal with WWC. Suppose Cannon expects to
do at least 4,000 minutes of videoconferencing per month. Its cost for 4,000
minutes under the three alternatives would be as follows:

MAF-562 24
Alternative 1: $20,000 ($5 per minute 4,000 minutes)
Alternative 2: $10,000
Alternative 3: $11,000 [$3,000 + ($2 per minute 4,000 minutes)]
Alternative 2 is the least costly. Moreover, if Cannon were to use more than
4,000 minutes, as is likely to be the case, alternatives 1 and 3 would be even
more costly. Cannon’s managers, therefore, should choose alternative2.

MAF-562 25
 A) Account Analysis Method of Cost Estimation

 Estimating costs using account analysis involves a review of each

account making up the total costs being analyzed.
 Then, each cost should be identified as either fixed, variable or mixed
with respect to the identified level of activity i.e. depending on the
relation between the cost and some activity.
 Fixed costs are those costs that are fixed in total regardless of the
activity level and Variable costs are those costs that vary in total as
activity changes.

MAF-562 26
 Typically, managers use qualitative rather than quantitative analysis when
making these cost-classification decisions.
 The account analysis approach is widely used because it is reasonably
accurate, cost-effective, and easy to use.
 Example: Consider indirect manufacturing labor costs for a manufacturing
company. Indirect manufacturing labor costs include wages paid for
supervision & maintenance. During the most recent 12-week period, the
company ran the machines for a total of 862 hours and incurred total
indirect manufacturing labor costs of Br. 12,501.

MAF-562 27
 Using qualitative analysis, the manager and the cost analyst determine that
over this 12-week period indirect manufacturing labor costs are mixed costs
with only one cost driver—machine-hours.
 As machine-hours vary, one component of the cost (supervision cost) is
fixed, whereas another component (maintenance cost) is variable.
 The goal is to use account analysis to estimate a linear cost function for
indirect manufacturing labor costs (Dependent variable) with number of
machine-hours as the cost driver (Independent variable).

MAF-562 28
The cost analyst uses experience and judgment to separate the total indirect
manufacturing labor costs (Br. 12,501) into costs that are fixed (Br. 2,157,
based on 950 hours of machine capacity over a 12-week period) and costs
that are variable (Br. 10,344) with respect to the number of machine-hours
used. Variable cost per machine-hour is Br. 10,344 ÷ 862 machine-hours =
Br.12 per machine-hour. The linear cost equation, Y=a+bX, in this example is
as follows: Indirect manufacturing labor costs = Br. 2,157 + (Br. 12 per
machine-hour * Number of machine-hours).

MAF-562 29
 Management at the company can use the cost function to estimate the indirect
manufacturing labor costs of using, say, 950 machine-hours to produce carpet
in the next 12-week period. Estimated costs equal Br. 2,157 + (950 machine-
hours * Br. 12 per machine-hour) = Br. 13,557.
 To obtain reliable estimates of the fixed and variable components of cost,
organizations must take care to ensure that individuals thoroughly
knowledgeable about the operations make the cost-classification decisions.
 Supplementing the account analysis method with the another method improves
credibility.

MAF-562 30
B) Conference Method of Cost Estimation (cont.)
 The conference method estimates cost functions on the basis of analysis of
information gathered from various departments opinions about costs
(purchasing, process engineering, manufacturing, employee relations, etc.). It is
based on the consensus of estimates from personnel.
 The conference method encourages interdepartmental cooperation. The pooling
of expert knowledge from different business functions of the value chain gives
the conference method credibility.
 Because the conference method does not require detailed analysis of data, cost
functions and cost estimates can be developed quickly.
 However, the emphasis on opinions rather than systematic estimation means
that the accuracy of the cost estimates depends largely on the care and skill of
the people providing the inputs.
MAF-562 31
C) Engineering/Industrial Engineering method of Cost
Estimation
 This method estimates through detailed study of the different operations in the
production process. Engineering cost estimates are based on measuring and
then pricing the work involved in a task.
 The method, also called the work-measurement method, estimates cost
functions by analyzing the relationship between inputs and outputs in
physical terms.
 For example, consider that a company uses inputs of cotton, wool, dyes,
direct manufacturing labor, machine time, and power and production output is
square yards of carpet. Time-and-motion studies analyze the time required to
perform the various operations to produce the carpet.

MAF-562 32
 For example, a time-and-motion study may conclude that to produce 10 square
yards of carpet requires one hour of direct manufacturing labor. Standards and
budgets transform these physical input measures into costs. The result is an
estimated cost function relating direct manufacturing labor costs to the cost
driver, square yards of carpet produced.
 The industrial engineering method is a very thorough and detailed way to
estimate a cost function when there is a physical relationship between inputs
and outputs, but it can be very time consuming.
 Physical relationships between inputs and outputs are difficult to specify for
some items, such as indirect manufacturing costs, R&D costs, and
advertising costs.

MAF-562 33
D) Statistical or Mathematical or Quantitative Analysis
method of Cost Estimation
 Statistical or mathematical is quantitative measures of the model, usually in
the form of Y = a + bX; where, Y is the cost estimate, b is the slope of the
line (variable cost per unit), a is the Y intercept (fixed Cost) and bX is the
total variable cost.
 The methods, usually in estimating costs, high-low and regression analysis
use a formal mathematical method to fit cost functions to past data
observations. These two quantitative cost estimation methods are considered
next after looking the steps involved.

MAF-562 34
Steps in Estimating a Cost Function Using Quantitative Analysis

 Step 1: Choose the dependent variable- the cost to be predicted and managed (Y)
 Step 2: Identify the independent variable, or cost driver (X)
 Step 3: Collect data on the dependent variable and the cost driver
 Step 4: Plot the data
 Step 5: Estimate the cost function
 Step 6: Evaluate the cost driver of the estimated cost function

MAF-562 35
Illustration

 Using the following data on a weekly Indirect Manufacturing

Labor Costs and Machine-Hours for a company, estimate a cost
function using two techniques—the high-low method and
regression analysis method.

MAF-562 36
 A B C
Week Cost Driver: Machine-Hours Indirect Manufacturing Labor Costs

(X) (Y )
1 68 $ 1,190
2 88 1,211
3 62 1,004
4 72 917
5 60 770
6 96 1,456
7 78 1,180
8 46 710
9 82 1,316
10 94 1,032
11 68 752
12 48 963
Total 862 $12,501
MAF-562 37
High-Low Method

 The simplest form of quantitative analysis to “fit” a line to data points is the
high-low method.
 It uses only the highest and lowest observed values of the cost driver within
the relevant range and their respective costs to estimate the slope coefficient
and the constant of the cost function.
 The slope coefficient, b, is calculated as follows:
Slope coefficient =Difference between costs associated with highest and
lowest observations of the cost driver/Difference between highest and lowest
observations of the cost driver 
a = y – bX

MAF-562 38
 To compute the constant, we can use either the highest or the lowest
observation of the cost driver. i.e.
y = a + bX
 Thus, the high-low estimate of the cost function is as follows:
y = a + bX
y = $23.68 + ($14.92 per machine-hour * Number of machine-
hours)
 The estimated cost function is a straight line joining the observations with the
highest and lowest values of the cost driver (number of machine-hours).

MAF-562 39
 There is a danger of relying on only two observations to estimate a cost
function.
 The advantage of the high-low method is that it is simple to compute and
easy to understand; it gives a quick, initial insight into how the cost driver—
number of machine hours— affects indirect manufacturing labor costs.
 The disadvantage is that it ignores information from all but two
observations when estimating the cost function.

MAF-562 40
Regression Analysis Method

 Regression analysis is a statistical method that measures the average amount

of change in the dependent variable associated with a unit change in one or
more independent variables.
 In the example, the dependent variable is total indirect manufacturing labor
costs. The independent variable, or cost driver, is number of machine-hours.
 Simple regression analysis estimates the relationship between the dependent
variable and one independent variable.

MAF-562 41
 Multiple regression analysis estimates the relationship between the dependent
variable and two or more independent variables. In Multiple regression analysis
we might use as the independent variables, or cost drivers, number of machine-
hours and number of batches.
 In case of simple regression (i.e. one dependent and one independent variable
case):
Y = a + bX, where y is the predicted cost value, Y is the observed cost
value, X the cost deriver.

MAF-562 42
 There are alternative techniques for computing/ estimating the values of “a” and
“b”. One method is the least-squares technique where for estimating the
regression line, it minimizes the sum of the squares of the vertical deviations
from the data points to the estimated regression line (residual terms). Thus, the
formulae for a and b is as follows:

MAF-562 43
In the example, where: n = number of data points = 12
= sum of the given X values = 68 + 88 + ... + 48 = 862
= sum of squares of the X values = (68)2 + (88)2 + ... + (48)2 + 4,624 +
7,744 + ... + 2,304 = 64,900
= sum of given Y values = 1,190 + 1,211 + ... + 963 = 12,501
= sum of the amounts obtained by multiplying each of the given X values
by the associated observed Y value=(68) (1,190) + (88) (1,211) + ... + (48)
(963) = 80,920 + 106,568 + ... + 46,224 = 928,716
So, for our example, a = $300.98 and b= $10.31 and the estimated cost
function (also called regression function/line) is:
y = $300.98 + $10.31X
MAF-562 44
 Once the estimation is completed, different statistical tests and analysis can be
conducted. For example, Goodness of fit measures how well the predicted
values, y, based on the cost driver, X, match actual cost observations, Y.
 For more detail discussion on the statistical tests, you can refer introductory
statistics or econometrics books.

MAF-562 45
 In the simple regression, the estimate of the slope coefficient, b, indicates that
indirect manufacturing labor costs vary at the average amount of $10.31 for
every machine-hour used within the relevant range.
 Management can use the regression equation when budgeting for future indirect
manufacturing labor costs. For instance, if 90 machine-hours are budgeted for
the upcoming week, the predicted indirect manufacturing labor costs would be :
y = $300.98 + ($10.31 per machine-hour * 90 machine-hours) = $1,228.88
 The regression method is more accurate than the high-low method because the
regression equation estimates costs using information from all observations,
whereas the high-low equation uses information from only two observations.
The inaccuracies of the high-low method can mislead managers.

MAF-562 46
Using Regression Output to Choose Cost Drivers of Cost Functions
 Consider the two choices of cost drivers for indirect manufacturing labor costs
(y):
y = a + (b * Number of machine-hours)
y = a + (b * Number of direct manufacturing labor-hours)
 In this case we need to estimate the regression lines for the two alternatives.
Then after, using regression results obtained, we can make a choice.
 Four important criteria of using regression output to choose cost drivers of
cost functions are discussed next. Of the criteria available, the economic
plausibility criterion is especially important.

MAF-562 47
Criteria for choosing among alternative cost drivers
1. Economic plausibility – Check for the existence of positive relationship
between the cost to be estimated (the dependent variable) and the cost driver
(the independent variable).
2. Goodness of fit – the higher the value the better.
3. Significance of independent variable(s) – check for statistical significance
that the b value is different from 0.
4. Specification analysis of estimation assumptions – test for Linearity
within the relevant range , Constant variance of residuals (Constant
variance is also known as homoscedasticity). Violation of this assumption is
called heteroscedasticity.), Independence of residuals (serial correlation
(also called autocorrelation) and Normality of residuals….

MAF-562 48
 Multiple Regression (one dependent/estimated cost and two or more
independent variables/cost drivers case)
 There are cases where basing the estimation on more than one independent variable (that is, multiple
regression) is more economically plausible and improves accuracy of estimation. For example,
 The most widely used equation to express relationships between two or more independent variables
and a dependent variable are linear in the form.
y = a + b1X1 + b2X2 + ... + u. Where y = Cost to be predicted
Where: X1,X2, ... = Independent variables on which the prediction is to be based
a, b1, b2,... = Estimated coefficients of the regression model

u = Residual term that includes the net effect of other factors not in the model as well as measurement
errors in the dependent and independent variables.

 A major concern that arises with multiple regression is multicollinearity. it exists when two or
more independent variables (Xis) are highly correlated with each other.
MAF-562 49
 Simple illustration on Multiple Regression
 Consider that a company’s analysis indicates that indirect manufacturing
labor costs include large amounts incurred for setup and changeover costs when
a new batch of carpets is started. Management believes that in addition to
number of machine-hours (an output unit-level cost driver), indirect
manufacturing labor costs are also affected by the number of batches of carpet
produced during each week (a batch level driver).

MAF-562 50
 The company estimates the relationship between two independent variables,
number of machine-hours and number of production batches of carpet
manufactured during the week, and indirect manufacturing labor costs (dependent
variable). The estimated model is :
y = $42.58 + $7.60X1 + $37.77X2
 where X1 is the number of machine-hours and X2 is the number of production
batches.
 Class room discussion: how do you interpret the results?
 For more discussions on multiple regression estimation and analysis, you can
refer basic econometrics or statistics books.

MAF-562 51
The Cause-and-Effect Criterion
 Managers are interested in estimating past cost-behavior functions primarily
because these estimates can help them make more-accurate cost predictions,
or forecasts, of future costs.
 Better cost predictions help managers make more-informed planning and
control decisions, such as preparing next year’s marketing budget. But better
management decisions, cost predictions, and estimation of cost functions can
be achieved only if managers correctly identify the factors that affect costs.

MAF-562 52
 The most important issue in estimating a cost function is determining whether
a cause-and-effect relationship exists between the level of an activity and
the costs related to that level of activity.
 Without a cause-and-effect relationship, managers will be less confident about
their ability to estimate or predict costs.
 When a cause and- effect relationship exists between a change in the level of
an activity and a change in the level of total costs, we refer to the activity
measure as a cost driver.
 Understanding the drivers of costs is crucially important for managing costs.
 The cause and-effect relationship might arise as a result of the following:

MAF-562 53
 A physical relationship between the level of activity and costs. An example
is when units of production are used as the activity that affects direct material
costs. Producing more units requires more direct materials, which results in
higher total direct material costs.
 A contractual arrangement. In alternative 1 of the example used to illustrate
the types of linear cost functions described earlier, number of minutes used is
specified in the contract as the level of activity that affects the telephone line
costs.
 Knowledge of operations. An example is when number of parts is used as the
activity measure of ordering costs. A product with many parts will incur
higher ordering costs than a product with few parts.

MAF-562 54
 Managers must be careful not to interpret a high correlation, or connection,
in the relationship between two variables to mean that either variable causes
the other.
 Consider direct material costs and labor costs. For a given product mix,
producing more units generally results in higher material costs and higher labor
costs. Material costs and labor costs are highly correlated, but neither causes
the other. Using labor costs to predict material costs is problematic.
 Labor costs are a poor predictor of material costs. By contrast, factors that
drive material costs such as product mix, product designs, and manufacturing
processes, would have more accurately predicted the changes in material costs.

MAF-562 55
 The Cause-and-Effect Criterion (cont.)
 Only a cause-and-effect relationship—not merely correlation—
establishes an economically plausible relationship between the level of
an activity and its costs.
 Economic plausibility is critical because it gives analysts and managers
confidence that the estimated relationship will appear again and again
in other sets of data from the same situation.
 Identifying cost drivers also gives managers insights into ways to reduce
costs and the confidence that reducing the quantity of the cost drivers
will lead to a decrease in costs.

MAF-562 56
To identify cost drivers on the basis of data gathered over time, always
use a long time horizon. Why? Because costs may be fixed in the short
run (during which time they have no cost driver), but they are usually
variable and have a cost driver in the long run.

MAF-562 57
Nonlinear Cost Functions
  In practice, cost functions are not always linear. A nonlinear cost function is
a cost function for which the graph of total costs (based on the level of a
single activity) is not a straight line within the relevant range. It is rather that
the cost function is graphically represented by a line that is not straight.
 Economies of scale in advertising may enable an advertising agency to
produce double the number of advertisements for less than double the costs.

MAF-562 58
 Even direct material costs are not always linear variable costs because
of quantity discounts on direct material purchases.
 Step cost functions are also examples of nonlinear cost functions.
 A step cost function is a cost function in which the cost remains the
same over various ranges of the level of activity, but the cost increases
by discrete amounts—that is, increases in steps—as the level of activity
increases from one range to the next. Examples are step fixed-cost
function and step variable-cost function.

MAF-562 59
 Learning Curves
 Nonlinear cost functions also result from learning curves.
 A learning curve is a function that measures how labor-hours per unit
decline as units of production increase because workers are learning
and becoming better at their jobs. As workers become more familiar
with their tasks, their efficiency improves and the related cost per unit
of output decreases.
 Managers use learning curves to predict how labor-hours, or labor
costs, will increase as more units are produced.

MAF-562 60
 Managers have extended the learning-curve notion to other business
functions in the value chain, such as marketing, distribution, and
customer service, and to costs other than labor costs. The term experience
curve describes this broader application of the learning curve.
 An experience curve is a function that measures the decline in cost per
unit in various business functions of the value chain—marketing,
distribution, and so on—as the amount of these activities increases.
 Here, we will consider two learning-curve models: the cumulative
average-time learning model and the incremental unit-time learning
model.

MAF-562 61
Cumulative Average-Time Learning Model
 In the cumulative average-time learning model, cumulative average time
per unit declines by a constant percentage each time the cumulative quantity
of units produced doubles.
 For example, if a company has an 80% learning curve, then the 80% means
that when the quantity of units produced is doubled from X to 2X, cumulative
average time per unit for 2X units is 80% of cumulative average time per unit
for X units. In this case, average time per unit has dropped by 20% (100% –
80%).
 Note that here the additional time required to produce the last X units
of the 2X units and the total time required to produce the 2X units are
0.6 and 1.6, respectively, of the time required to produce the first X
units.
MAF-562 62
The mathematical relationship underlying the cumulative average-time
learning model is as follows:
y = aXb
Where,
y = Cumulative average time (labor-hours) per unit (at Xth
level/units of production)
X = Cumulative number of units produced
a = Time (labor-hours) required to produce the first unit
b= Factor used to calculate cumulative average time to produce units

MAF-562 63
 The value of b is calculated as : ln (learning-curve % in decimal form) /ln2
 For an 80% learning curve, b = ln0.8/ln2 = –0.2231/0.6931 = –0.3219
 For example, when X = 3, a = 100, b = –0.3219 (i.e b-value computed above
for an 80% learning curve),
 y = 100 × 3–0.3219 = 70.21 labor-hours is Cumulative average time (labor-hours)
per unit of producing 3 units
 The cumulative total time when X = 3 is 70.21 × 3 = 210.63 labor-hours.

MAF-562 64
 Illustration on Cumulative Average-Time Learning Model assuming
80% Learning Curve and 100 hours of Time (labor-hours) required to
produce the first unit.
Cumulative number Cumulative Average Time Cumulative Total Time: Individual Unit Time for
of units produced (X) per Unit (y ): Labor-Hours Labor-Hours Col C = Col A × X th Unit: Labor-Hours
(Col A) (Col B) Col B (Col D)

1 100 100 100

2 80 = 100*0.8 160 60= 160-100
3 70.21 210.63 50.63=210.63-160
4 64=80*0.8 256 45.37
5 59.56 297.82 41.82
… …. …. …
15 41.82 627.3 28.67
16 40.96=51.2*0.8 655.36 28.06

MAF-562 65
Incremental Unit-Time Learning Model

In the incremental unit-time learning model, incremental time needed

to produce the last unit declines by a constant percentage each time
the cumulative quantity of units produced doubles. For example, if a
company has an 80% learning curve then the 80% here means that
when the quantity of units produced is doubled from X to 2X, the time
needed to produce the last unit when 2X total units are produced is
80% of the time needed to produce the last unit when the first X total
units are produced.

MAF-562 66
The mathematical relationship underlying the incremental unit-time
learning model is as follows:
y = aXb
Where,
◦ y = Time (labor-hours) taken to produce the last single unit (i.e. the Xth unit)
◦ X = Cumulative number of units produced
◦ a = Time (labor-hours) required to produce the first unit
◦ b = Factor used to calculate incremental unit time to produce units = ln(learning-
curve % in decimal form) /ln2

MAF-562 67
For an 80% learning curve, b = ln0.8 ÷ ln 2 = –0.2231 ÷ 0.6931 = –0.3219
For example, when X = 3, a = 100, b = –0.3219,
y = 100 × 3–0.3219 = 70.21 labor-hours is incremental time (labor-hours) (per
unit) of producing the 3rd unit
The cumulative total time when X = 3 is 100 + 80 + 70.21 = 250.21 labor-
hours.

MAF-562 68
 Illustration on Incremental-Time Learning Model assuming 80%
Learning Curve and 100 hours of Time (labor-hours) required to
produce the first unit
Cumulative Individual Unit Time for X th Cumulative Total Time: Labor- Cumulative Average
number of Unit (y): Labor-Hours Hours (Col C) Time per Unit : Labor-
units (Col B) Hours
produced (X) (Col B=Col C / Col A)
(Col A)

1 100 100 100

2 80 = 100*0.8 180 90
3 70.21 250.21=180+70.21 83.4
4 64=80*0.8 314.21 78.55
5 59.56 373.77 74.75
… … … …
… …. …. …
15 41.82 851.05 56.74
16 40.96=51.2*0.8 892.01 55.75
MAF-562 69
Choice of the Model - How do managers choose which model and
what percent learning curve to use?
 Managers make their choices on a case-by-case basis. For example, if
the behavior of manufacturing labor-hour usage as production levels
increase follows a pattern like the one predicted by the 80% learning
curve cumulative average-time learning model, then the 80% learning
curve cumulative average-time learning model should be used.
 Engineers, plant managers, and workers are good sources of
information on the amount and type of learning actually occurring as
production increases.
 Plotting this information and estimating the model that best fits the
data is helpful in selecting the appropriate model.

MAF-562 70
Incorporating Learning-Curve Effects into Prices and Standards- How
do companies use learning curves?
 The effects of the learning curve could have a major influence on
decisions.
 For example, managers might set an extremely low selling price to
generate high demand.
 As production increases to meet this growing demand, cost per unit
drops. The company “rides the product down the learning curve” as it
establishes a larger market share. Although it may have earned little
operating income on its first unit sold—it may actually have lost money
on that unit— the company earns more operating income per unit as
output increases.

MAF-562 71
Alternatively, subject to legal and other considerations, managers might
set a low price on just the final N units if the total labor and related
overhead costs per unit for these final N units are predicted to be lower
than the previous ones.

MAF-562 72
 Incorporating Learning-Curve Effects into Prices and Standards (cont.)
 Many companies incorporate learning-curve effects when evaluating performance.
 The Nissan Motor Company expects its workers to learn and improve on the job
and evaluates performance accordingly. It sets assembly labor efficiency standards
for new models of cars after taking into account the learning that will occur as
more units are produced.
 The learning-curve models examined assume that learning is driven by a single
variable (production output).
 Other models of learning have been developed by companies that focus on how
quality— rather than manufacturing labor-hours—will change over time,
regardless of whether more units are produced.
 Studies indicate that factors other than production output, such as job rotation and
organizing workers into teams, contribute to learning that improves quality.
MAF-562 73
Data Collection and Adjustment Issues
The ideal database for estimating cost functions quantitatively has two
characteristics:
1. The database should contain numerous reliably measured observations
of the cost driver (the independent variable) and the related costs (the
dependent variable). Errors in measuring the costs and the cost driver are
serious. They result in inaccurate estimates of the effect of the cost driver on
costs.
2. The database should consider many values spanning a wide range for
the cost driver. Using only a few values of the cost driver that are grouped
closely considers too small a segment of the relevant range and reduces the
confidence in the estimates obtained.

MAF-562 74
Data Collection and Adjustment Issues (Cont.)
Unfortunately, cost analysts typically do not have the advantage of
working with a database having both characteristics. Frequently
encountered data problems and steps the cost analyst can take to
overcome these problems are the following.
1. The time period for measuring the dependent variable (for example,
machine-lubricant costs) does not properly match the period for
measuring the cost driver. This problem often arises when accounting
records are not kept on the accrual basis. Consider a cost function with
machine-lubricant costs as the dependent variable and number of
machine-hours as the cost driver.

MAF-562 75
Assume that the lubricant is purchased sporadically and stored for later
use. Records maintained on the basis of lubricants purchased will indicate
little lubricant costs in many months and large lubricant costs in other
months. These records present an obviously inaccurate picture of what is
actually taking place. The analyst should use accrual accounting to
measure cost of lubricants consumed to better match costs with the
machine-hours cost driver in this example.

MAF-562 76
2. Fixed costs are allocated as if they are variable. For example, costs
such as depreciation, insurance, or rent may be allocated to products to
calculate cost per unit of output. The danger is to regard these costs as variable
rather than as fixed. They seem to be variable because of the allocation methods
used. To avoid this problem, the analyst should carefully distinguish fixed
costs from variable costs and not treat allocated fixed cost per unit as a
variable cost.

MAF-562 77
 Data Collection and Adjustment Issues (Cont.)
3. Data are either not available for all observations or are not uniformly
reliable. Missing cost observations often arise from a failure to record a
cost or from classifying a cost incorrectly. For example, marketing costs
may be understated because costs of sales visits to customers may be
incorrectly recorded as customer-service costs.
Recording data manually rather than electronically tends to result in a
higher percentage of missing observations and erroneously entered
observations. Errors also arise when data on cost drivers originate
outside the internal accounting system.

MAF-562 78
For example, the accounting department may obtain data on testing-
hours for medical instruments from the company’s manufacturing
department and data on number of items shipped to customers from the
distribution department. One or both of these departments might not
keep accurate records. To minimize these problems, the cost analyst
should design data collection reports that regularly and routinely obtain
the required data and should follow up immediately whenever data are
missing.

MAF-562 79
4. Extreme values of observations occur from errors in recording costs
(for example, a misplaced decimal point), from non representative
periods (for example, from a period in which a major machine
breakdown occurred or from a period in which a delay in delivery of
materials from an international supplier curtailed production), or from
observations outside the relevant range. Analysts should adjust or
eliminate unusual observations before estimating a cost relationship.

MAF-562 80
 Data Collection and Adjustment Issues (Cont.)

5. There is no homogeneous relationship between the cost driver and the

individual cost items in the dependent variable-cost pool. A
homogeneous relationship exists when each activity whose costs are
included in the dependent variable has the same cost driver. In this case,
a single cost function can be estimated. When the cost driver for each
activity is different, separate cost functions (each with its own cost
driver) should be estimated for each activity. Or, the cost function
should be estimated with more than one independent variable using
multiple regression.

MAF-562 81
6. The relationship between the cost driver and the cost is not stationary. That
is, the underlying process that generated the observations has not remained
stable over time. For example, the relationship between number of machine-
hours and manufacturing overhead costs is unlikely to be stationary when the
data cover a period in which new technology was introduced. One way to see
if the relationship is stationary is to split the sample into two parts and
estimate separate cost relationships—one for the period before the technology
was introduced and one for the period after the technology was introduced.
Then, if the estimated coefficients for the two periods are similar, the analyst
can pool the data to estimate a single cost relationship. When feasible, pooling
data provides a larger data set for the estimation, which increases confidence
in the cost predictions being made.

MAF-562 82
7. Inflation has affected costs, the cost driver, or both. For example,
inflation may cause costs to change even when there is no change in the
level of the cost driver. To study the underlying cause-and-effect
relationship between the level of the cost driver and costs, the analyst
should remove purely inflationary price effects from the data by dividing
each cost by the price index on the date the cost was incurred. In many
cases, a cost analyst must expend considerable effort to reduce the effect
of these problems before estimating a cost function on the basis of past
data.

MAF-562 83
Chapter End Exercises

The Helicopter Division of GLD, Inc., is examining helicopter assembly

costs at its Indiana plant. It has received an initial order for eight of its
new land-surveying helicopters. GLD can adopt one of two methods of
assembling the helicopters. Given data, work out the requirements.
Required:
1. How many direct-assembly labor-hours are required to assemble the
first eight helicopters under (a) the labor-intensive method and (b) the
machine-intensive method?
2. What is the total cost of assembling the first eight helicopters under
(a) the labor intensive method and (b) the machine-intensive method?

MAF-562 84
 Given data for the class work exercise
Labor-Intensive Assembly Method Machine-Intensive Assembly
Method

Direct material cost per helicopter $40,000 $36,000

Direct- assembly labor time for first 2000 labor-hours 800 labor-hours
helicopter

Learning curve for assembly labor 85% cumulative average time (– 90% incremental unit time
time per helicopter 0.234465) (–0.152004)

Direct -assembly labor cost $30 per hour $30 per hour

Equipment-related indirect $12 per direct assembly labor-hour $45per direct assembly labor-
manufacturing cost hour

Material-handling-related indirect 50% of DM cost 50% of DM cost

manufacturing cost
MAF-562 85
 Solution for the class work exercise
1.
a. The labor-intensive assembly method based on an 85% cumulative average- time learning
model:
Cumulative average-time per unit for the Xth is calculated as y =aXb; For example, when X = 3, y =
2,000 3–0.234465 = 1,546 labor-hours.
b. The machine-intensive assembly method based on a 90% incremental unit-time learning
model:
Individual unit time for the Xth unit is calculated as y = aXb; For example, when X = 3, y = 800
3–0.152004 = 677 labor-hours.
2.
Total costs of assembling the first eight helicopters: The machine-intensive method’s assembly
costs are $66,342 lower than the labor intensive method ($892,692 – $826,350). The
computations are as follows:

MAF-562 86
 Solution for the class work exercise (cont.)

Labor-Intensive Assembly Method Machine-Intensive Assembly Method

(using data from part 1a) (using data from part 1b)

Direct materials: $320,000 $288,000

8 helicopters × $40,000; $36,000 per
helicopter

Direct-assembly labor: 294,780 157,740

9,826 hrs.; 5,258 hrs. × $30/hr.

Indirect manufacturing costs 117,912 236,610

Equipment related
9,826 hrs. × $12/hr.; 5,258 hrs. ×
$45/hr.

Materials-handling related 160,000 144,000

0.5 0 × $320,00 0; $288,000

Total assembly costs $892,692 $826,350

MAF-562 87
Suggested exercises

Work out the exercises and problems on the text book (Cost
accounting : a managerial emphasis / Charles T. Horngren,
Srikant M. Datar, Madhav V. Rajan. -- 14th ed.) at the end of
chapter 2 and 10.

MAF-562 88
End of Chapter Two

MAF-562 89