ACCT2002 Chap.2 Cost Behaviour and Cost Estimation
Cost Behaviour and Cost Estimation
Cost estimation converts observed activity and cost into a usable relationship, but arithmetic alone does not establish causality. A plausible cost driver should explain resource consumption, remain operationally measurable and produce a stable enough relationship for the decision horizon.
This chapter separates linear and nonlinear behaviour, contribution-style variable and fixed components, the high-low method, regression interpretation, learning-curve effects and common data defects. The core discipline is to inspect the process before fitting the line: remove mismatched periods, recognise capacity steps, investigate outliers and avoid treating correlation as proof.
A model is useful only inside the activity band and operating conditions from which it was estimated.
What this chapter covers
- 01
Linear and nonlinear cost functions
- 02
Causality and choosing a cost driver
- 03
High-low estimation
- 04
Regression output and economic interpretation
- 05
Learning-curve effects
- 06
Data comparability, outliers and structural change
- 07
Evaluating an estimated cost function
Estimate and test a maintenance-cost function
- 3Variable cost per machine-hour is the change in cost divided by the change in activity: ($38,000 − $28,400) ÷ (2,700 − 1,900) = $12.
- 2Fixed cost from the high observation is $38,000 − ($12 × 2,700) = $5,600. The low observation gives the same amount.
- 2Predicted cost at 2,300 hours is $5,600 + ($12 × 2,300) = $33,200.
- 3High-low uses only two observations and assumes both are representative. It also establishes a numerical line, not proof that machine-hours cause maintenance cost; operating changes or a step in capacity could invalidate the estimate.
Key terms
- Cost function
- A mathematical description of how total cost changes with one or more drivers.
- Cost driver
- A factor whose change is expected to cause a change in the related cost.
- High-low method
- A two-point technique that estimates variable cost from the highest and lowest activity observations.
- Intercept
- The estimated fixed-cost component in a linear cost function.
- Slope
- The estimated change in total cost for one additional unit of the driver.
- Learning curve
- A nonlinear relationship in which labour time per unit or batch declines as cumulative experience grows.
- Outlier
- An observation that differs sharply from the rest and requires investigation rather than automatic deletion.
Cost Behaviour and Cost Estimation FAQ
Does a high R-squared prove the cost driver causes cost?
No. It measures how much observed variation is associated with the fitted relationship, not whether the driver consumes the resource. Operational reasoning and process evidence are still required.
When is high-low acceptable?
It is a quick estimate when the two endpoints are representative and the relationship is plausibly linear. It should not replace inspection of the full data set when regression or engineering analysis is available.
How do learning effects change a linear estimate?
Early production may use more labour time than later production as workers improve. A single constant slope can overpredict later cost or underpredict early cost if cumulative experience is ignored.
What data checks should come before calculation?
Align the periods, units and accounting definitions; identify unusual shutdowns or price changes; confirm the activity range; and plot the observations. Data cleaning should be explained, not hidden.
Exam move
For every estimation question, write four headings before calculating: economic driver, time period, relevant range and data warning. Practise high-low quickly, then ask how a regression or engineering estimate could disagree. Sketch the observations and label capacity steps or learning effects. When checking an answer, distinguish an arithmetic error from a model error: a perfectly calculated line can still be inappropriate.
For oral practice, defend one driver choice and explain why a correlated alternative may not be causal.
Working through Cost Behaviour and Cost Estimation in ACCT2002? Sia is AskSia’s AI Accounting tutor — ask any ACCT2002 Cost Behaviour and Cost Estimation question and get a clear, step-by-step explanation grounded in how ACCT2002 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
