How do you apply the high-low method?


The high low method is a method of estimating cost behavior using the two most extreme activity levels on the relevant range to estimate variable and fixed costs. It consists in selecting the highest and the lowest volumes of the cost driver observed on the relevant range, retrieving the corresponding costs, and computing the slope of the straight line connecting these two points to obtain the unit variable cost (i.e. the change in costs divided by the change in volume between these two observations). The fixed costs are then obtained by subtracting from the total costs of the high (or low) volume the estimated variable costs for that volume (i.e. that volume multiplied by the estimated unit variable cost):

\[ V_c = \frac{\Delta TC}{ \Delta Q} = \frac{TC_{Q.high}-TC_{Q.low}}{Q_{high} - Q_{low}} \\ \quad \\ FC_{Q.high} = FC_{Q.low} = TC_{Q.high} - Q_{high} \times V_c = TC_{Q.low} - Q_{low} \times V_c \]

where \(V_c\) is the unit variable cost, \(TC_{Q.high}\) is the total cost for the highest volume, \(TC_{Q.low}\) the total cost for the lowest volume, \(Q_{high}\) the highest volume, \(Q_{low}\) the lowest volume, \(FC_{Q.high}\) and \(FC_{Q.low}\) the fixed costs. Note that this technique always relies on the highest and lowest volumes on the relevant range, even if these observations are not associated with the highest or lowest total costs.



The high low method can be applied with only two periods, so it does not require much in term of information. Moreover, it lends itself to easy manual computations.


Since it relies on only two data points, it is extremely sensitive to the representativeness of these two observations. This makes the method highly unreliable. Moreover, it can only be used when the underlying costs have only one cost driver, or at least when one cost driver has a dominant influence on the costs variation.

Please indicate how clear and understandable this page was for you: