Chapter 6 Cost Behavior

Engineering method of cost estimation

Engineering method of cost estimation is not based on historical data, or how much the cost was last period. It studies the process that results in cost incurrance. So for example, it would ask how much material would be needed this period, and how much that will cost. Industries such as genetic engineering and electronics are evolving so rapidly that historical data is irrelevant.

Multiple Regression Cost Estimation Method

Estimation based on multiple independent variables or cost drivers. Y=a+b1X1 + b2X2 y=dependent variable. x=cost driver, independent variable a=intercept of the line. b=slope of the line.

Cost Prediction

Forecast of cost at a particular level of activity. Using knowledge of cost behavior to forecast the level of cost at a particular level of activity. Focus is on the future.

Visual-Fit Cost Estimation Method

a method for determining cost behavior of a semivariable cost or when the analyst has no clear idea about the behavior of the cost. Plot recent observations about the cost at the various activity levels. The resulting scatter diagram helps analyst visualize the relationship between the cost & the activity (or cost driver). Draw back: not very objective. 2 different cost analysts may draw 2 different visually fit cost lines.

Account-Analysis Cost Estimation Method

a method for determining cost behavior that is based on a manger's judgment in classifying each general ledger account as a variable, fixed, or semivariable cost. Also known as "Account-Classification" cost estimation method

marginal cost

cost of producing the next unit.

Engineered Costs

costs having direct, observable, quantifiable cause and effect relationship between the level of output and the quantity of resources consumed. i.e. tasty donuts: direct material costs rise as more donuts are produced. (You can't produce more donuts without more food ingredients.)

curvilinear cost

costs that fluctuate up and down - i.e. utility costs Low levels of activity show decreased marginal costs. For increases in activity, the marginal costs also increase.

time and motion studies to determine cost behavior patterns

determine steps required for people to perform the manual tasks that are part of the production process. Cost behavior patterns for various types of costs are then estimated on the basis of the engineering analysis.

Explain engineered costs, committed costs, discretionary costs

engineered cost bears strong physical relationship to the activity measure. Direct material cost is an example. A committed cost results from an organizations ownership or hte use of facilities and its basic organization structure. Examples include property taxes and depreciation on buildings and equipment. A discretionary cost arises as a result of a management decision to spend a particular amount of $ for some purpose. i.e. research & development, advertising & promotion.

semivariable (or mixed) cost behavior pattern

has both a fixed and a variable component. ie. cost of delivery trucks. fixed cost is rental payment, variable cost is gasoline & maintenance. It is generally a curvilinear cost within the relevant range.

fixed

remains unchanged in total as the activity level (or cost driver) varies. i.e. salaries, taxes, depreciation.

Discretionary Costs

result from decisions by management to spend in certain areas, such as advertising,research and management development programs, and contributions to charitable organizations. These costs are easier to adjust than committed fixed costs.

Relevant range

the range of activity within which management expects the company to operate.

Learning Curve & Experience Curve on Cost Behavior

when efficiency increases with experience, and the average labor time per unit declines, that is due to the learning curve.. The learning idea applied to costs other than direct labor is called the Experience Curve. Usually applied to labor intensive manufacturing operations.

Describe the behavior of a variable, step-variable, fixed, step-fixed, semivariable (or mixed), and curvilinear costs.

A variable cost total changes in direct proportion to the change in activity level. A step-variable cost is nearly variable, but increases in small steps instead of continuously. A Fixed cost remains unchanged in total as the activity level changes. A step-fixed cost remains constant over a wide range of activity, but jumps to a different amount for activity outside the range. A semivariable (or mixed) cost has both variable and semivariable parts. A curvilinear cost behavior pattern has a curved graph.

Problems encountered in data collected for cost estimation

1. Missing data-misplaced doc or failure to record transactions 2. Outliers-represent errors or highly unusual circumstances. They should be eliminated from the set 3. Mismatched time periods-units of time for dependent & independent variables do not match. 4. Trade-offs in choosing the time period-conflicting objectives in choosing time period. 5. Allocated and discretionary costs-fixed costs allocated on a per unit of activity basis. Fixed costs such as depreciation if allocated to units of production may appear to be variable. Discretionary costs such as advertising if allocated based on sales dollars also will appear variable. 6. Inflation-during periods of inflation, historical cost data may not reflect future cost behavior.

Committed Costs

1. infrastructure capacity, investment in plant, equipment, and basic administrative structure. long term commitment. 2. investments in facilities, equipment, and basic organizational structure that can't be significantly reduced even for short periods of time without making fundamental changes. i.e. Property taxes, depreciation on buildings and equipment, and salaries of management are committed fixed costs.

Describe the following cost estimation methods: account classification, visual fit, high-low, and least squares regression.

All are based on analysis of historical cost data observed at different activity levels. Account Classification method examines organizations ledger accounts. Visual-fit creates scatter graph by plotting observations of costs. High-low method a semivariable cost is estimated using 2 data points. The high and low activity levels. In the least-squares regression method, the cost line is estimated to minimize the sum of the squared deviations between the cost line & the data points.

Relationship between cost estimation, cost behavior & cost prediction

Cost behavior is the relationship between cost and activity. Cost estimation is the determination of a cost's behavior. Cost prediction is the forecast of a cost at a particular level.

Cost Behavior Patterns

Cost functions

Outlier

Data point that falls far away from the other points in the scatter diagram and is not representative of the data.

Describe the multiple regression, engineering, and learning-curve approaches to cost estimation

Multiple Regression is a statistical method that estimates a linear relationship between one dependent variable and two or more independent variables. The engineering method is based on a detailed analysis of the process that results in cost incurrence. The learning-curve approach, the labor cost is estimated by studying hte relationship between the cumulative production quality and the average labor time required per unit. When this approach is applied to costs other than labor, it is refered to as the experience-curve approach. Engineering method of cost estimation is not based on historical data, or how much the cost was last period. It studies the process that results in cost incurrance. So for example, it would ask how much material would be needed this period, and how much that will cost. Industries such as genetic engineering and electronics are evolving so rapidly that historical data is irrelevant.

step-variable

Nearly variable. Increase in small steps instead of continuously. Usually includes inputs that are purchased and used in small increments.

step-fixed

Remain the same over a wide range of activity, but jump to a different amount for activity levels outside that range. i.e. Supervisory salaries: fixed cost until activities warrant hiring an additional person, then the cost jumps levels.

Hi-Low Cost Estimation Method

Semivariable-cost approximation is computed using 2 data points: the highest & lowest activity level points. More objective than the visual-fit method. Weakness is that only 2 data points are used. (subtract lowest from highest costs)/(subtract lowest from highest activity levels)

Least Squares Regression Line Cost Estimation Method

The least-squares regression line may be represented by the equation of a straight line. y=a+bx. x=activity level or cost driver (independent variable), y is dependent variable. a=intercept of the line, b=slope of line. Objective method of cost estimation but requires more computations.

Cost Estimation

The process of determining how a cost behaves. Often focuses on historical data.

Cost Behavior

The relationship between cost and activity.

variable cost

Total variable cost increases in proportion to the activity level (or cost driver). But the unit remains the same as activity changes.