ACCT2019 Chap.2 Cost Behaviour and Support Cost Allocation
Cost Behaviour and Support Cost Allocation
Cost estimation translates observed activity into a usable planning relationship. A linear function separates estimated fixed cost from a variable rate, but a good fit is not enough without a plausible driver. This chapter compares account analysis, high-low and regression, then connects cost behaviour to support-department allocation.
Direct, step-down and reciprocal methods recognise different amounts of service between support departments, affecting product costs, incentives and later decisions. Account analysis contributes operational judgment, high-low supplies a quick two-point estimate and regression uses the broader observation set with diagnostics. None removes the need to inspect the relevant range, outliers, step changes and causality.
Allocation raises a related modelling choice: which department consumes the service and which method recognises mutual support. A complete reciprocal calculation may improve representation, while a simpler method may be adequate when internal support flows are immaterial. The answer should therefore state not just a rate or allocation, but also what the model ignores and which management decision could be distorted.
Forecast review should compare predicted with actual cost and distinguish random residuals from structural change. A new shift pattern, technology or lease can invalidate the historical intercept and slope even when the earlier regression fitted closely. The estimate and the allocation should each be accompanied by a practical limitation, because a clean number can still rest on a poor operating representation.
What this chapter covers
- 01
Linear and nonlinear cost behaviour
- 02
Fixed intercept and variable slope
- 03
Account analysis and high-low
- 04
Regression fit and causality
- 05
Direct and step-down allocation
- 06
Reciprocal support services
Worked example · free
Estimating maintenance cost with high-low
- +1Original AskSia practice weighting. Activity difference is 700 hours and cost difference is $8,400.
- +1Variable rate is $8,400 / 700 = $12 per machine hour.
- +1Fixed cost is $31,200 - ($12 x 1,800) = $9,600.
- +1At 1,500 hours, predicted cost is $9,600 + ($12 x 1,500) = $27,600.
- +1Machine hours must still be checked as a cause of maintenance demand.
Key terms
- Cost Function
- An equation relating expected total cost to one or more activity drivers.
- Fixed Intercept
- Estimated cost when the modelled activity is zero within the fitted relationship.
- High-low Method
- A two-point estimate using observations at the highest and lowest activity.
- Regression Analysis
- A statistical method that estimates a relationship using the available observations.
- Reciprocal Method
- An allocation method that recognises mutual services between support departments.
Cost Behaviour and Support Cost Allocation FAQ
Why can a high R-squared still mislead?
Fit measures how closely the line describes observed variation; it does not prove the activity causes the cost. Seasonality, time trends or omitted drivers can produce a close historical relationship that fails when operations change.
When is high-low useful?
It is useful as a quick estimate when the endpoints are representative and data are limited. Because it discards the middle observations, it should be checked against a scatterplot, regression result or operational account analysis before guiding a material decision.
Which support allocation method is most complete?
The reciprocal method is most complete because it recognises mutual services among support departments. Step-down recognises only the chosen sequence, while direct allocation ignores support-to-support service and assigns pools straight to operating departments.
Exam move
Practise rebuilding a cost function from two points, then sketch the observations and explain what the intercept and slope mean operationally. Add an outlier to see how high-low can change while a regression based on all observations remains more stable. For allocation, draw the support departments and service arrows before selecting a method.
State what each method ignores and how the omission could affect a production department. Keep fit, causality and incentives as three separate checks: a model can fit well but use a weak driver, and an accurate allocation can still create poor behaviour. In worked answers, finish with the decision that uses the estimate, the relevant range and the operational factor that could make the historical relation fail.
Build a small table of monthly activity and cost, calculate the high-low line and then compare every observation with its predicted amount. Label each residual and ask whether a step cost, season or data error explains it. Re-estimate after removing an unusual endpoint so you can see how dependent the shortcut is on two records.
For support departments, solve the same diagram with direct, step-down and reciprocal logic, focusing on which service arrows disappear under each method. Add a dual-rate interpretation by separating capacity cost from usage cost and explain how that changes a receiving manager's incentive. End with a forecast just inside the observed activity range and another outside it.
The second needs an explicit warning because the fitted line may cross staffing, lease or technology thresholds that the historical slope does not capture.
Working through Cost Behaviour and Support Cost Allocation in ACCT2019? Sia is AskSia’s AI Accounting tutor — ask any ACCT2019 Cost Behaviour and Support Cost Allocation question and get a clear, step-by-step explanation grounded in how ACCT2019 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.