Cost estimation - Cost behavior What we really want to understand is how spending will vary in a variety of decision settings. Cause-effect relations and costs drivers. Capacity and capacity costs: • • • • • Theoretical = 100,000 Practical = 90,000 Normal = 85,000 Budgeted = 80,000 Suppose fixed overhead is budgeted at $1,000,000; variable overhead is $1 per unit; direct material costs = $3; and direct labor = $3. Overhead is applied based on units of product. Capacity and capacity costs: What does a unit of product cost if overhead is allocated based on theoretical capacity? $17 Practical capacity? $18.11 Normal capacity? $18.76 Budgeted capacity. $19.50 Which measure should the company use? Capacity and capacity costs: Suppose the company allocates overhead based on practical capacity and actual production is 70,000 units. By how much is overhead underapplied? About $222,300 What does that cost represent? The cost of idle or excess capacity. Capacity and capacity costs • Who should pay for excess capacity? • Who should pay for idle capacity? • How is capacity measured? What is the scarcest resource? • Idle capacity and opportunity costs. Cost estimation: overhead • When is it important to understand how overhead behaves? – When pricing, production, process and product design decisions are made. – When bids and make or buy decisions are made. – When we need to answer “what if” questions. Cost estimation: overhead costs • First week’s product costing exercises: applied overhead. – Valuing inventories & costs of sales. – Not for costing individual products – Not for predicting costs What methods are available? • Engineering estimates • Account analysis • Scattergraph and high-low estimates • Statistical methods (typically regression) Cost behavior: linear function by assumption. TC = FC + VC*(level of cost driver) where TC = FC = VC = total cost fixed cost variable cost per unit of the cost driver, and sometimes the cost driver is represented by X. Overhead Costs Volume A B C D Cost estimation: Account analysis • Review each account • Identify it as fixed or variable (or mixed) • Attempt to determine the relationship between the activity of interest and the cost – Cost of building occupancy – Cost of quality inspections – Cost of materials handling Example Suppose management believes that the monthly overhead cost ($5000) in the factory is mixed. It is believed to be 50% fixed and 50% variable. The variable portion is believe to depend on machine hours, which average 10,000 per month. How would you show this as a linear equation? TC = $2500 + $.25(machine hours) Peterson Mfg. in Problem Set #1 will require account analysis. Scattergraph Suppose you have data on overhead costs and machine hours for the past 15 months. Can you easily determine whether the posited relationship exists? Yes, plot the data and look for a relationship. Plot of overhead costs vs. machine hours Scattergram 4000.00 3500.00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00 0.00 30.00 60.00 90.00 Machine Hours 120.00 150.00 High-Low cost estimation Find the variable cost per unit of the cost driver (VC): Overhead at highest activity - Overhead at lowest activity VC Highest activity - Lowest activity High-Low method: Example continued $3,105 - $1,896 VC 142 mhr - 50 mhr $1,209 VC 92 mhr VC $13.14/mhr High-Low cost estimation Fixed cost $3,105 - ($13.14 * 142 mhr) $1,239 Estimate the total overhead cost during a months when 115 machine hours will be used: TC FC VC * 115 mhr TC $1,239 ($13.14 * 115 mhr) TC $2,750 Cost estimation using regression • Y = the dependent variable (total O/H cost) • X = the explanatory variables Y = *X + where X = machine hours and = random error. TC = FC + VC*X + . Regression fits a line through these data points: Scattergram 4000.00 3500.00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00 0.00 30.00 60.00 90.00 Machine Hours 120.00 150.00 Simple linear regression • • • • One explanatory variable Cost estimation equation Coefficient of correlation (R) Coefficient of determination (R2) – Goodness of fit – Measure of importance • F-statistic (hypothesis testing) • p-value Coefficient of determination Measures the percentage of variation in the dependent variable explained by the independent variable. When the predicted values exactly equal the actual costs, R2 = 1. A goodness of fit test: R2 > .3 The F statistic • • • • Goodness of fit hypothesis testing Compute a statistic for regression results Compute the associated p-value, or Look up a critical F-value and compare – 1 numerator degree of freedom – (n-2) denominator degrees of freedom – alpha = .05 The F test: • The hypothesis is: The slope coefficient is zero. • The F-statistic measures the loss of fit that results when we impose the restriction that the slope coefficient is zero. • If F is large, the hypothesis is rejected. The p-value • This is the probability that the statistic we computed could have come from the population implied by our null hypothesis. • Suppose we hypothesize that the slope coefficient is zero. • If the p-value associated with the F-statistic is small, chances are the slope coefficient is not zero. Regression result interpretation Re gre ssion Summary Ove rhe ad Costs vs. Machine Hours Count 15 Num. Missing 0 R . 896 R Squared . 802 Adjusted R Squared . 787 RMS Residual 182. 244 ANOVA Table Ove rhe ad Costs vs. Machine Hours DF Sum of Squares 1 1753772. 049 Residual 13 431765.285 Total 14 2185537. 333 Regression Mean Square 1753772. 049 F-Value P-Value 52.804 <. 0001 33212. 714 Re gre ssion Coeff icients Ove rhe ad Costs vs. Machine Hours Interc ept Mac hine Hours Coeffic ient Std. Er ror 1334.293 162. 913 12.373 1. 703 Std. Coeff . 1334.293 . 896 t-Value P-Value 8. 190 <. 0001 7. 267 <. 0001 Simple linear regression Scattergram 4000.00 3500.00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00 0.00 30.00 60.00 90.00 Machine Hours 120.00 150.00 Overhead Costs = 1334.293 + 12.373 * Machine Hours; R^2 = .802 Results using DM$ Regression Summary Ove rhe ad Cost s vs. Direct Materials Cost Count 15 Num. Missing 0 R .960 R S quared .921 Adjusted R Squar ed .915 RMS Residual 115.087 ANO VA Table Ove rhe ad Cost s vs. Direct Materials Cost DF Sum of Squar es 1 2013351.144 Residual 13 172186.189 Total 14 2185537.333 Regr ession Mean Square 2013351.144 F-Value 152.007 P-Value <.0001 13245.091 Regression Coe fficie nts Ove rhe ad Cost s vs. Direct Materials Cost Inter cept Dir ect Materials Cost Coef ficient Std. Error 1456.586 87.225 .356 .029 Std. Coeff. t-Value P-Value 1456.586 16.699 <.0001 12.329 <.0001 .960 Multiple regression Regression Summary Overhe ad Costs vs. 2 Inde pe nde nts Count 15 Num . Missing 0 R .976 R Squared .952 Adjusted R Squared .944 RMS Residual 93.658 ANOVA Table Overhe ad Costs vs. 2 Inde pe nde nts DF Sum of S quares 2 2080274.802 Residual 12 105262.531 Total 14 2185537.333 Regression Mean Square 1040137.401 F- Value 118.576 P- Value <.0001 8771.878 Regression Coef ficie nt s Overhe ad Costs vs. 2 Inde pe nde nts Intercept Mac hine Hours Direct Mater ials Cost Coeffic ient Std. Err or 1333.960 83.724 Std. Coef f. t-Value P- Value 1333.960 15.933 <.0001 4.359 1.578 .316 2.762 .0172 .258 .042 .697 6.101 <.0001 Forecasting overhead • Predict monthly overhead when machine hours are expected to be 62 and direct materials costs are expected to be $1,900. • Recall = $1,333.96 – Coefficient for mhrs = $4.359 – Coefficient for DM$ = $.258 Predicted overhead Overhead $1,333.96 $4.359(62) $.258($1,900) $2,094.42 Putting together a bid • Calculate a minimum bid for a contract that would use 22 machine hours and $900 in direct materials. This would be a one-timeonly job. • What if there is no idle capacity? • Would your bid change if there were potential for repeated business? Problems with regression • • • • Nonlinear relationships Outliers Spurious relationships Data problems – – – – Inaccurate accounting cut-offs Arbitrarily allocated costs Missing data Inflation Thursday • • • • Cenex and Burd & Fletcher Cases. Use Excel for regression computations We will discuss the problems in class and Work a handout problem in groups.