Source: PCC Case (A), Exhibits 1–5
Tags: activity-based costing, ABC, cost pool, cost driver, cross-subsidisation, client profitability distortion, overhead allocation critique, PCC case analysis, death spiral, strategic cost management
Difficulty: Advanced | Prerequisites: Parts 1 and 2 of this series; solid understanding of activity-based costing (ABC) concepts, cost pools, cost drivers, and cross-subsidisation.
This is the analytical payoff. Parts 1 and 2 established the facts of PCC's operations and its current costing system. This part asks you to critique that system, propose improvements (primarily ABC), and think through the strategic implications. This is where exam questions live. If you are not comfortable with how ABC works in principle, review that material first.
PCC's single-overhead-rate system almost certainly cross-subsidises clients, making some look more profitable and others less profitable than they truly are. An ABC approach would create separate cost pools for distinct activities (RRT work, training, supervision, communications, facilities) and assign each using a driver that reflects actual consumption. This would give PCC accurate client costs, better pricing, and a clearer picture of which clients are worth keeping.
Activity-based costing (ABC)
A costing approach that identifies distinct activities within an organisation, assigns costs to activity-based cost pools, and then allocates those pools to cost objects using activity-specific cost drivers. In simple terms, instead of one big bucket of overhead spread by one measure, ABC creates many smaller buckets, each spread by the measure that best explains what drives that particular cost.
Cost pool
A grouping of indirect costs that share a common cost driver. Under ABC at PCC, you might create separate pools for RRT activities, training support, supervision, communications, facilities, and general administration.
Cost driver (ABC context)
The measurable factor that causes a cost pool's costs to be incurred. Different activities have different drivers. RRT costs might be driven by the number of new problems documented per client; training costs by the number of product updates per client; communications by the number of call minutes.
Cross-subsidisation
The distortion that occurs when a costing system over-charges one cost object and under-charges another. At PCC, clients who consume proportionally less overhead (relative to their operator hours) subsidise clients who consume proportionally more. Think of it as some clients getting a bargain at other clients' expense, hidden by the averaging effect of a single rate.
Death spiral (downward demand spiral)
A reinforcing cycle where losing volume raises the per-unit overhead rate (because fixed costs are spread across fewer units), which raises prices, which risks losing more volume, which further raises the rate. PCC experienced a version of this after losing Smith Electronics.
Capacity cost
The cost of maintaining the ability to serve clients, regardless of whether that capacity is fully used. At PCC, much of the overhead (building, depreciation, managerial salaries) is capacity cost. Account 20200 (Unused Operator Time) explicitly captures one form of this.
Cost-system refinement
The process of improving a cost system's accuracy, typically by increasing the number of cost pools, using more specific cost drivers, or tracing previously indirect costs directly to cost objects. Moving from a plantwide rate to ABC is a refinement.
The fundamental problem is that PCC uses a single cost driver (operator hours) to allocate thirteen different categories of indirect costs. This works only if every client consumes each indirect resource in the same proportion as it consumes operator hours. In reality, clients differ along several dimensions:
Product complexity varies. Krag Wireless (networking equipment) and Gunn Computers (gaming systems) likely generate different types and volumes of RRT work, supervisor escalations, and training needs compared to Carlin Software (dental accounting) or Mason Graphics (animation software).
Call patterns differ. Some clients may have shorter, simpler calls; others may have longer, more complex ones. Communications costs (phone lines, internet bandwidth) are driven by call minutes and data usage, not simply by operator headcount.
Training intensity differs. A client that frequently updates its products requires more training. Training staff salaries are indirect, so a client that causes more training activity is not charged for it proportionally.
RRT usage differs. A client whose product generates many undocumented problems consumes more RRT resources. Under the current system, this extra cost is averaged across all clients via operator hours.
The result: some clients are almost certainly being overcosted and others undercosted. This is cross-subsidisation.
An ABC system for PCC would create multiple cost pools, each with its own cost driver. A reasonable starting design:
RRT cost pool: Includes RRT wages (the largest component of account 21000), RRT secretarial wages, RRT programmer staff costs, and the portion of managerial salaries attributable to the RRT manager. Potential driver: number of new problem resolutions documented per client, or number of RRT hours logged per client.
Training cost pool: Includes training staff salaries (currently in managerial salaries, account 30000), training secretarial wages, and the portion of operator training time that is indirect. Potential driver: number of product updates per client, or training hours per client.
Supervision cost pool: Includes the portion of managerial salaries for supervisors (10% of operators act as supervisors). Potential driver: number of escalated calls per client.
Communications cost pool: Account 50100. Potential driver: call minutes or call volume per client.
Facilities cost pool: Building depreciation, property taxes, utilities, buildings and grounds staff. Potential driver: floor space used per client (number of operator workstations), or simply operator FTEs (this pool is closest to being legitimately driven by headcount).
General administration cost pool: CEO office, CFO and finance, executive assistant, miscellaneous. Potential driver: total operator hours or revenue (this is the residual pool where a volume-based driver may be acceptable).
Equipment depreciation cost pool: Account 40100 (primarily computer systems). Potential driver: number of workstations or operator FTEs per client.
Without specific data on each client's consumption of each activity (which the case does not provide in full), you cannot compute exact ABC costs. But you can reason directionally:
A client with a complex product that generates many new problems would absorb more RRT cost under ABC than under the current system.
A client with stable, mature products and fewer updates would absorb less training cost.
A small client with few operator hours but high RRT demands would look much less profitable under ABC.
A large client with straightforward calls and low escalation rates would look more profitable.
The case strongly hints that Krag Wireless (which PCC is losing money on) may be consuming more than its share of overhead. If so, the current system understates Krag's true cost and overstates the cost of other clients.
When Smith Electronics left in 2001:
Operator hours dropped from 218.5 to 186.9 thousand (a 14.5% decline).
Overhead barely moved: from 11,084 to 11,107 thousand.
Profit collapsed from 2,884 to 708 thousand.
The overhead rate for 2002 would have been approximately 11,107 / 186.9 = 59.43 per hour, up from roughly 50.74 in 2001 (11,084 / 218.5). This higher rate feeds into higher fully absorbed costs, which feeds into pressure to raise fees, which risks losing another client. This is the death spiral in action.
Implement ABC to get accurate client costs before negotiating renewals with Krag Wireless and Mason Software.
Separate capacity costs from activity costs. Report unused-capacity costs as a period expense rather than allocating them to remaining clients. This prevents the death spiral from distorting client-level profitability.
Use ABC data to inform pricing. If Krag Wireless truly consumes more RRT and supervision resources, the fee should reflect that. If Mason Software is lower-maintenance, PCC can price competitively and still earn a return.
Consider the cost of losing a client explicitly. The thin market means replacement clients are scarce. Sometimes accepting a lower margin is better than losing the volume entirely.
The cross-subsidisation problem in this case is universal in professional services. Law firms, consulting firms, IT service providers, and accounting firms all face the same issue: some clients consume disproportionate resources but are billed using averaged rates. ABC adoption in service industries has grown precisely because of cases like PCC's.
The death-spiral dynamic is visible in industries with high fixed costs and declining volume: airlines, hotels, hospitals, and universities all face versions of the same problem when they lose customers but cannot shed fixed costs proportionally.
Students often think ABC will always show that the largest client is the most profitable. Not necessarily. Under ABC, the most profitable client is the one whose resource-consumption pattern is most favourable, regardless of size. A large client with complex, RRT-heavy products could be less profitable than a smaller client with straightforward support needs.
Students sometimes propose too many cost pools. The goal is not to create a pool for every line item. Some costs genuinely do move with operator hours (like operator benefits), and grouping those into an operator-support pool with an hours-based driver is perfectly reasonable. ABC should increase accuracy without adding unjustifiable complexity.
The case does not provide enough data to compute exact ABC client costs. Exam questions typically ask you to identify the pools and drivers and reason directionally about which clients would look better or worse, not to calculate precise numbers.
Students sometimes recommend PCC drop unprofitable clients. In PCC's thin market, losing a client triggers the death spiral. The recommendation should be to reprice accurately and manage capacity costs separately, not to shed volume.
⚠️ The most common exam question on this case asks you to identify weaknesses in PCC's current cost system and propose an ABC alternative. Be ready to name specific cost pools, specific drivers, and explain why each driver is better than operator hours for that pool.
⚠️ You should be able to explain cross-subsidisation in PCC's context and give a directional example (e.g. "Krag Wireless likely consumes more RRT resources per operator hour than Carlin Software, so the current system undercharges Krag and overcharges Carlin").
⚠️ The death-spiral concept is heavily tested. Be able to trace the chain: volume loss leads to higher overhead rate leads to higher costs leads to higher prices leads to further volume loss.
⚠️ Expect a question about whether PCC should allocate unused-capacity costs to clients. The standard answer: no. Unused capacity should be reported as a period cost, not spread to remaining clients, because doing so inflates their apparent cost and feeds the death spiral.
⚠️ Be prepared to discuss the trade-off between cost-system accuracy and cost-system complexity. ABC is more accurate but more expensive to maintain. PCC is a small firm with four clients, so the system needs to be practical.
True or False: Under PCC's current system, a client that generates many RRT escalations pays more overhead than a client that generates few. (False. All clients pay the same overhead rate per operator hour, regardless of RRT usage.)
Fill in the blank: The phenomenon where losing volume raises the per-unit overhead rate, which raises prices and risks further volume loss, is called the ______. (Death spiral, or downward demand spiral.)
True or False: ABC would use a single cost driver for all cost pools. (False. Each cost pool gets its own activity-specific driver.)
True or False: Unused-capacity costs should be allocated to remaining clients under best practice. (False. Best practice treats unused capacity as a period cost, not allocated to clients.)
Fill in the blank: When a cost system overcharges one client and undercharges another, the distortion is called ______. (Cross-subsidisation.)
Q: Identify at least four weaknesses in PCC's current cost system.
A: (1) A single cost driver (operator hours) is used for all thirteen indirect cost categories, even though these costs have different underlying drivers. (2) The system does not distinguish between clients that consume different levels of RRT, training, or supervisory resources. (3) Downtime is allocated 50-50 between each operator's two clients, regardless of the actual ratio of work. (4) Overtime is treated as indirect and spread across all clients, even when overtime is triggered by a specific client's new product launch. (5) Unused operator time (account 20200) appears in 2002 when Smith Electronics left, and under the current system this cost would be allocated to remaining clients, inflating their reported costs.
Q: Propose an ABC system for PCC. Identify at least four cost pools and an appropriate cost driver for each.
A: (1) RRT pool (RRT wages, programmer staff, RRT secretarial): driven by number of new problem resolutions or RRT hours per client. (2) Training support pool (training staff salaries, training secretarial): driven by number of product updates or training hours per client. (3) Supervision pool (supervisor portion of managerial salaries): driven by number of escalated calls per client. (4) Communications pool (telephone and internet): driven by call minutes or call volume per client. (5) Facilities pool (building depreciation, property taxes, utilities, grounds staff): driven by operator FTEs or workstation count per client. (6) General administration (CEO, CFO, finance, miscellaneous): driven by total operator hours or revenue per client.
Q: Explain how cross-subsidisation is likely occurring at PCC and give a specific example.
A: Cross-subsidisation occurs because clients with different resource-consumption patterns are all charged the same overhead rate per operator hour. For example, Krag Wireless manufactures networking equipment with a selling point of "network stability," which implies complex technical issues that likely generate significant RRT work. Carlin Software makes dental office accounting packages, which are likely more straightforward. Under the current system, both pay the same overhead per operator hour, even though Krag probably consumes far more RRT resources per hour. Krag is undercosted and Carlin is overcosted.
Q: PCC is currently losing money on the Krag Wireless account. Using the concepts from this case, explain why this might be happening and what PCC should do about it.
A: The loss on Krag Wireless is likely a combination of two factors. First, Krag may be consuming more overhead resources (RRT, supervision, training) than the current system reveals, meaning its true cost is higher than the fully absorbed cost PCC reports. Second, the fee negotiated with Krag was based on the averaged overhead rate, which understates Krag's actual cost. PCC should implement ABC to determine Krag's true cost, then use that information in renewal negotiations. If Krag's true cost is significantly higher, PCC needs to either raise the fee (with data to justify it) or accept that the account may not be viable at Krag's price point.
Q: Should PCC drop Krag Wireless if it cannot negotiate a profitable fee? Discuss the trade-offs.
A: Dropping Krag Wireless would remove the largest client by operator hours. The death-spiral risk is severe: PCC's fixed overhead would be spread across only three clients, sharply raising the per-hour cost for everyone. In PCC's thin market, finding a replacement is very difficult, as the Smith Electronics experience demonstrated. PCC should first determine whether the loss is due to genuine underpricing (in which case, negotiate harder with better cost data) or due to cost-system distortion (in which case, ABC may reveal the account is closer to breakeven than it appears). Dropping the client should be a last resort, and only after exhausting pricing and cost-reduction options.
Q: Explain how PCC should handle unused-capacity costs and why the current approach is problematic.
A: Currently, unused operator time (account 20200) is pooled with all other indirect costs and allocated via operator hours. This means that when PCC loses a client and has idle capacity, the cost of that idle capacity is charged to the remaining clients, raising their reported costs and potentially triggering fee increases. Best practice is to separate capacity costs from activity costs. Unused capacity should be reported as a period expense on PCC's income statement, visible to management as the cost of maintaining excess capacity, but not assigned to clients. This prevents the death spiral and gives management clear information about the cost of underutilisation.
This case ties together several major cost accounting themes: plantwide vs departmental vs ABC overhead allocation, cost behaviour (fixed vs variable), cost-volume-profit analysis (how volume changes affect profitability), and strategic pricing. It also connects to relevant cost analysis for the keep-or-drop client decision and to capacity management for how to handle idle resources.
If your course covers the Cooper and Kaplan framework for when to adopt ABC (high product/client diversity, high overhead relative to direct costs, high competition on pricing), PCC ticks every box.
Purdy Call Centers ABC analysis, activity-based costing service firm, cost pool identification, cost driver selection, cross-subsidisation overhead, death spiral cost allocation, downward demand spiral, unused capacity cost treatment, capacity cost period expense, client profitability analysis ABC, overhead allocation critique, plantwide rate vs ABC, strategic cost management service firm, keep or drop client decision, Cooper Kaplan ABC criteria, cost system refinement, PCC case exam questions, cost accounting case study analysis