Difficulty: Intermediate to Advanced | Prerequisites: Part 1 of these study notes (operations and strategy), basic understanding of overhead rates, direct vs. indirect costs.
This is the analytical heart of the Purdy Call Centers case. Part 1 covered what PCC does and how it is organised. This part focuses on how PCC measures and allocates costs, how it sets fees, and what the four-year budget data reveals. The core tension: PCC's single-pool overhead system treats every operator hour as equally expensive, even though different clients consume different mixes of resources (RRT time, training, communications, equipment). Understanding why this system produces misleading client profitability numbers is the gateway to proposing a better one, almost certainly activity-based costing.
PCC allocates all indirect costs using a single overhead rate: total budgeted indirect costs divided by total budgeted operator hours. This peanut-butter approach treats every hour the same, regardless of what resources a client actually consumes. The four-year budget data shows that after losing Smith Electronics, total overhead barely fell while operator hours dropped sharply, causing the overhead rate to spike. The fee structure, built on top of this distorted cost base, is now creating pricing conflicts with clients.
Single (plantwide) overhead rate
An overhead allocation method that pools all indirect costs into one total and divides by a single allocation base. PCC uses total budgeted indirect costs / total budgeted operator hours.
In simple terms, this means: one number applied to every hour, regardless of what kind of overhead that hour actually triggered.
Budgeted (predetermined) overhead rate
The overhead rate calculated at the start of the year using estimated (not actual) figures. PCC does not restate the budget mid-year, even if conditions change significantly.
Think of it as: locking in the rate before the year starts, so you can price consistently throughout.
Chart of accounts
A standardised list of all cost categories a firm uses to organise its financial records. PCC's chart has two direct accounts (10000 and 10500) and thirteen indirect accounts (20000 through 60300).
Variance analysis
The process of comparing actual spending to budgeted amounts and investigating the differences. PCC computes variances monthly and expects managers to keep spending in line with budgeted levels.
Death spiral (doom loop)
A self-reinforcing cycle in which lost volume raises per-unit costs, which raises prices, which drives away more volume, which raises costs further. PCC is showing early signs of this pattern.
Think of it as: a vicious circle where losing a client makes you more expensive for the clients who stay, which risks losing more of them.
Cross-subsidisation
When one cost object (client, product) is overcharged while another is undercharged because the allocation system does not reflect the true pattern of resource consumption. A likely outcome of PCC's single-rate system.
Peanut-butter costing
A colloquial term for spreading costs uniformly across cost objects without regard to differences in how those objects actually consume resources. PCC's approach is a textbook example.
Start with the actual time spent on each client last year.
Adjust for expected changes in workload for the coming year.
Example: if 50 FTEs were consumed by a client last year and the client predicts a 10% increase in calls, the budget assigns 55 FTE for the coming year.
The estimate covers both call-handling time and client-specific training time.
The CFO (Mary Roby) and her staff discuss each of the thirteen indirect cost accounts with the relevant department heads.
Example for account 21000 (Support Staff Wages): Roby collects estimates from Jack Ryan (RRT wages), Dick Morgan (buildings and grounds), Cal Purdy (executive assistant), and each departmental manager who has assistants. These are summed to produce the account total. For 2003, account 21000 totalled $1,590k.
Sum all thirteen indirect cost account estimates to get total budgeted overhead.
Divide by total budgeted operator hours.
The result is a single, firm-wide overhead rate per operator hour.
The budget is set once per year and is not restated, even if conditions change significantly.
Variances are computed monthly. Managers are expected to find ways to keep spending within the original budget.
Roby's rationale: "If we allowed people to change it at will, it would lose a lot of its authority."
Mary Roby adapted the cost system from her previous employer, a manufacturing firm (Knapp Industries). She applied a traditional fully-absorbed costing approach, with the main differences being that PCC has no direct materials and costs are assigned to clients rather than products.
Account 10000, Operator Wages: wages of all operators while answering calls. Directly assigned to each client.
Account 10500, Operator Training Wages: wages of operators while training for a specific client. Directly assigned to that client.
Every other cost is treated as indirect and allocated using the single overhead rate. These include:
Operator benefits (20000)
Operator overtime (20100)
Unused operator time (20200)
Support staff wages (21000)
Support staff benefits (21050)
Managerial salaries (30000)
Managerial benefits (30050)
Depreciation, equipment (40100)
Depreciation and property taxes, building (40200)
Communications (50100)
Supplies (60100)
Utilities (60200)
Miscellaneous (60300)
Operators are paid based on experience (experienced operators earn roughly 25% more than new hires). But operators are not assigned to clients based on experience level. Using hours rather than dollars prevents a client from being penalised simply because it happens to have more experienced (higher-paid) operators assigned to it at a given time.
Overhead rate = Total budgeted indirect costs / Total budgeted operator hours
Fully absorbed cost = Direct operator wages (including training) + (Operator hours x Overhead rate)
The computer system tracks time per call by operator by client, down time by operator, and training time by client.
Down time is analysed weekly to adjust operator assignments and keep idle time near 10%.
If a client consistently needs fewer operators than assigned, excess operators are reassigned to other clients.
Fees are based on operator hours expended on each client over the year.
Each contract is negotiated individually. PCC aims to charge an hourly fee that exceeds the fully absorbed hourly cost by enough to generate an acceptable return.
Since all operator hours carry the same reported cost, differences in hourly fees between clients reflect negotiation intensity, not actual differences in resource consumption.
Clients are billed monthly based on the number of FTE dedicated to them.
If FTE levels change during a month, the fee is adjusted accordingly.
At year-end, PCC provides each client with a report showing total down time and training time for all operators dedicated to that client.
Training time is client-specific.
Down time is split 50-50 between the two clients each operator serves, regardless of which client generated the actual call volume. This is a deliberate simplification; PCC considered proportional allocation but chose the simpler approach.
2001: Smith Electronics did not renew. PCC spent six months looking for a replacement, laid off operators, and still had not found a replacement by 2003.
2002: Gunn Computers renewed after "unexpectedly strenuous" negotiations. The new rate was higher but generated a lower profit percentage than the original contract.
End of 2003: Krag Wireless and Mason Software contracts due for renewal. Krag's CFO warned they would seek competing bids. Mason was less concerned, as long as fees were "reasonable."
Purdy noted that PCC was losing money on the Krag Wireless account under the current cost system.
Total revenue peaked in 2001 at $19,432k (five clients), dropped to $16,485k in 2002 (four clients, after Smith Electronics departed), and partially recovered to $18,038k by 2003.
Revenue from each remaining client grew year-on-year, but the combined growth was not enough to replace Smith's $4,024k contribution in 2001.
Total operator wages (including training) fell from $5,464k in 2001 to $4,670k in 2002 (reflecting the lost client), then rose to $5,115k by 2003 as remaining clients grew.
Krag Wireless consistently has the largest operator wage bill ($2,100k in 2003), reflecting its larger scale.
Total overhead remained remarkably sticky: $10,248k in 2000, $11,105k in 2001, $11,109k in 2002, $11,397k in 2003.
Even after losing 20% of revenue (Smith Electronics), overhead barely moved, falling just $4k between 2001 and 2002. This is the hallmark of high fixed costs.
Key overhead increases from 2000 to 2003: managerial salaries rose from $1,525k to $2,010k (+32%); depreciation on equipment rose from $1,750k to $2,050k (+17%); managerial benefits rose from $513k to $750k (+46%).
Unused operator time of $530k appeared in 2002 (the year after Smith left) and disappeared by 2003, suggesting PCC eventually adjusted headcount.
2000: $10,248k / 205.0k hours = $49.99 per hour
2001: $11,105k / 218.5k hours = $50.82 per hour
2002: $11,109k / 186.9k hours = $59.44 per hour
2003: $11,397k / 204.6k hours = $55.70 per hour
The spike in 2002 is dramatic. Even though hours recovered in 2003, the rate stayed elevated because overhead continued to climb.
Call centre profit fell from $2,917k in 2000 to just $706k in 2002, recovering only partially to $1,526k in 2003.
Profit margin: 15.9% in 2000, 14.7% in 2001, 4.3% in 2002, 8.5% in 2003.
The 2002 figure is the crisis year: revenue down, overhead flat, profit collapsed.
Cost system design choices like the one PCC faces show up everywhere in service industries. A consulting firm deciding whether to allocate partner time, admin support, and office space with a single rate or multiple pools is dealing with the same problem. The death spiral scenario is particularly relevant for any business with high fixed costs and a small number of large clients: losing one client can make the economics worse for everyone who remains. ABC implementations in service firms are increasingly common precisely because single-rate systems produce the kind of cross-subsidisation and pricing distortions PCC is experiencing.
Students often assume PCC should simply cut costs to solve the problem. While cost reduction helps, the deeper issue is that the cost system itself allocates costs in a way that may not reflect reality. A client that appears unprofitable under the current system might be profitable under a more refined one, and vice versa.
Some students think the overhead rate rose because PCC became less efficient. The arithmetic is simpler: the numerator (total overhead) stayed nearly flat while the denominator (total operator hours) shrank after Smith left. The rate went up mechanically.
It is common to confuse the budgeted overhead rate with the actual overhead rate. PCC sets the rate at the start of the year using budgeted figures and does not adjust it. Actual spending can differ, creating variances.
Students sometimes think that because Krag Wireless has the largest operator wage bill, it must be the most profitable client. That conflates revenue share with profitability. The case specifically notes that PCC is losing money on Krag under the current system.
Be able to calculate the overhead rate for any year from the Exhibit 5 data and explain why it changed.
Know how to compute the fully absorbed cost per client and the resulting profit or loss per client.
Understand why a single overhead rate can produce cross-subsidisation between clients and be ready to give specific examples from PCC's cost categories.
Be prepared to identify which of PCC's thirteen overhead accounts would benefit from separate cost pools and different allocation bases (the setup for an ABC question).
Expect a question on the death spiral: explain the mechanism and identify the evidence of it in PCC's data.
The 50-50 down time split and the decision to use hours rather than dollars are likely discussion or short-answer targets.
True or false: PCC's overhead rate formula uses actual costs divided by actual hours. False. It uses budgeted indirect costs divided by budgeted operator hours.
Fill in the blank: PCC's overhead rate in 2002 was approximately ______ per operator hour. 59.44 per hour (11,109k / 186.9k hours).
True or false: The budget is restated mid-year if a major client is lost. False. The budget is never restated during the year, regardless of changes.
True or false: Operator overtime is a direct cost charged to the specific client that caused it. False. Overtime is classified as an indirect cost (account 20100) and allocated across all clients.
Fill in the blank: PCC's profit margin fell from 15.9% in 2000 to ______% in 2002. 4.3%.
Q: Calculate PCC's overhead rate for 2003 and explain what it means.
A: Total budgeted overhead for 2003 is 11,397k. Total budgeted operator hours are 204.6k. The overhead rate is 11,397 / 204.6 = 55.70 per operator hour. This means that for every hour an operator spends on a client, 55.70 in indirect costs is allocated to that client on top of the operator's direct wages.
Q: Why did PCC's overhead rate spike in 2002, and why did it not fully recover in 2003?
A: In 2002, total overhead remained at 11,109k (virtually unchanged from 2001's 11,105k) while operator hours dropped from 218.5k to 186.9k because Smith Electronics had left. Dividing a nearly identical numerator by a much smaller denominator mechanically pushed the rate from 50.82 to 59.44 per hour. By 2003, hours recovered to 204.6k, but total overhead also climbed to 11,397k (due to rising managerial salaries, depreciation, and benefits), so the rate settled at 55.70 rather than returning to the pre-2002 level.
Q: Give two specific examples of overhead accounts at PCC where a single allocation base of operator hours is likely to produce inaccurate client costs.
A: First, communications costs (account 50100) are driven by the number of calls and their duration, which varies by client product complexity, not simply by the number of operator hours. A client with shorter, simpler calls would be overcharged relative to one with longer, more complex calls. Second, RRT wages (the largest component of account 21000, support staff wages) are driven by the number of undocumented problems and new product introductions, which differ by client. A client with a mature, stable product line generates fewer RRT demands than one that frequently launches new products.
Q: Explain how the Purdy case illustrates cross-subsidisation between clients.
A: Under PCC's single-rate system, every operator hour carries the same overhead charge. But different clients consume different amounts of RRT time, training, communications, and equipment. A client that generates many undocumented problems (requiring heavy RRT work) is undercharged for that resource, while a client with a stable product is overcharged. The result is that the stable client subsidises the demanding one. This helps explain why Krag Wireless appears unprofitable: it may be absorbing too large a share of overhead it does not drive, or alternatively, it may be driving costs that are being spread to others.
Q: If you were advising PCC to redesign its cost system, what would you recommend as a first step?
A: Identify the major activities that drive each overhead cost category and group them into separate cost pools. For example, RRT-related costs could be allocated based on the number of new resolution procedures written per client. Training department costs could be allocated based on the number of new product introductions per client. Communications could be allocated based on call volume or call minutes. Equipment depreciation might be allocated based on the number of operator workstations dedicated to each client. This activity-based approach would replace the single overhead rate with multiple rates, each tied to a driver that reflects actual resource consumption.
The cost system analysis in this case leads directly into activity-based costing (ABC), which is the standard remedy for the distortions a single-rate system creates. The overhead rate calculations connect to the broader topic of predetermined overhead rates and how over- or under-applied overhead is handled at year-end. The death spiral scenario ties into cost-volume-profit (CVP) analysis, particularly the concept of contribution margin and the effect of fixed costs on break-even volume. Finally, the variance analysis PCC conducts monthly connects to the flexible budgeting and standard costing material most courses cover in detail.
Purdy Call Centers, cost system design, overhead allocation, single overhead rate, plantwide rate, predetermined overhead rate, budgeted overhead rate, fully absorbed cost, chart of accounts, variance analysis, budget rigidity, death spiral, doom loop, cross-subsidisation, peanut-butter costing, activity-based costing, ABC, cost pools, cost drivers, allocation base, operator hours vs operator dollars, direct cost, indirect cost, service firm costing, client profitability, fee structure, contract negotiation, capacity costs, unused capacity, cost-volume-profit, CVP, contribution margin, fixed cost behaviour, managerial accounting, University of Texas at Austin