Cost Estimation.docx

COST ESTIMATION Cost estimation involves the use of past costs to predict future costs. It involves breaking down costs into their fixed and variables elements in order to determine future costs. It also involves ascertaining the activity. A cost driver can be defined as any factor whose change causes, a change in the total cost of an activity e.g. direct labour hours, machine hours etc. Cost estimation methods Engineering methods These are methods of the analysis cost behavior which are based on the use of engineering analysis of technological relationships between inputs and outputs. It is appropriate when there is a physical relationship between costs and the cost driver. Accounts inspection method This approach request that the departmental manager and the accountant inspect each item of expenditure within the accounts of a particular period, and then clarify each item of expense as a wholly fixed, wholly variable or semi-variable costs. The costs estimates are mainly based on arbitrary judgments and may lack the precision necessary when they are used to make decisions that are sensitive. High-Low Method The high-low method uses the highest and lowest activity levels over a period of time to estimate the portion of a mixed cost that is variable and the portion that is fixed. Like the account analysis and scatter graph method, the amounts determined for fixed and variable costs are only estimates. Because it uses only the high and low activity levels to calculate the variable & fixed costs, it may be misleading if the high and low activity levels are not representative of the normal activity. For example, if most data points lie in the range of 60 to 90 percent for a particular accounting test, and one student scored a 20, the use of the low point might distort the actual expectation of costs in the future. The high-low method is most accurate when the high and low levels of activity are representation of the majority of the other points. The steps below guide you through the high-low method: <="" p=""> Step 1: Determine which set of data represents the total cost and which represents the activity. Find the lowest and highest activity points. Step 2: Determine variable costs per unit by using the mathematical formula for a slope where you take divide the change in cost by the change in activity: Y2 -Y1 = Variable cost per unit X2 - X1 Where X2 is the high activity level X1 is the low activity level Y2 is the total cost at the high activity level selected Y1 is the total cost at the low activity level selected Step 3: Plug your answer to steps 2 along with either the high or the low point into the cost formula by replacing the slope (VC) with variable cost per unit, the high activity total cost for the y variable, and the high activity for the x variable. Then solve for fixed costs (FC).

1

Step 4: Plug your answers to steps 2 and 3 into the cost formula by replacing the slope (VC) with variable cost per unit and the y-intercept (FC) with total fixed costs in the following format: y = VC x + FC Regression analysis Simple regression analysis is based on the assumption that total cost is determined by one variable whereas multiple regression considers more than one variable. As far as possible, all the factors related to cost behavior should be brought into the analysis so that costs can be predicted and controlled more effectively. The equation for single regression can be expanded to include more than one independent variable. In case of simple linear regression: Y = a + bX Where: Y is the cost or variable being estimated A is the total fixed costs X is the number of units produced or the independent variable influencing costs B is the variable cost per unit In this case: b = n∑XY - ∑X∑Y n∑X2 - ∑(X)2 and a= ∑Y -b∑X n n If there are two independent variables and the relationship is assumed to be linear, the regression equation will be:Y = a + b 1 X 1 + b2 X 2 Where:

a represents the non-variable cost item. B1 represents the average change resulting from a unit change in X1, Assuming X2 and all the unidentified items remain constant. B2 – represents the average change in if resulting from a unit change in X2 asking that X1 remains constant.

The normal equations for a regression equation with two independent variables are: ∑ y = an + b1 ∑X1 + b 2 ∑ X2 ∑Xy = a∑X1 + b1 ∑ X1 2 + b2 ∑X1 X2 ∑X2y = a∑X2 + b1 ∑X1 X2 + b2 ∑X12 This can be solved through spreadsheet packages.

2

Example of application of multiple regression Consider a plant that generates its own steam and uses this steam for both heating and motive power. The cost of steam generation is likely to be determined by both temperature and machine hours, a multiple regression equation would be:Y = a + b1 X1 + b2 X2 Where y – total cost a – total non-variable cost y1 - # of machine hours b1 – regression co-efficient for machine hours y2 – number of days per month b2 – required co-efficient for temperature Multi collinearity When the independent variable are highly correlated with each other, its is very difficult to separate the effect of each variable on the dependent variables. This occurs when there is a simultaneous movement of two or more independent variable in the same direction and at approximately the same rate. This condition is called multi collinearity. Factors to consider when using past data to estimate cost functions 1. The cost of data and activity should be related to the same period. Costs used to estimate costs functions should not be behind the associated activity for accuracy purpose. 2. The number of observations should be sufficient if acceptable cost estimates are to be produced. 3. Security policies – data must be examined to ensure that the accounting policies do not lead to distorted cost functions. 4. Adjustments for past changes – an analysis of past date will yield estimates of future costs that are based on the cost relationships of previous periods. The approp**** of using past data will depend on the extent to which the future will correspond with the past. Any change of circumstances in the future will require past data to be adjusted in line with future data. Cost estimation when the learning effect is present Changer in the efficiency of the labour force may render past information unsuitable for predicting future labour costs. A situation like this may occur when workers become more familiar with the tasks that they perform, so that the labour time is required for the production of each unit. The phenomenon is common in manufacturing and is known as learning curve effect.

3