THE FUNDAMENTALLY FLAWED THINKING BEHIND activity based costing II

"We would stand a better chance of success if we gutted a farm animal and read its entrails."

Prof Caspari responds to Tony's article and checks (including through a simulation) whether or not Tony's claim is correct. Read the post, the answer comes towards the end.

OF FARM ANIMALS AND abc

Copyright (c) John A. Caspari, 1998; All rights reserved

PREFACE

John Caspari was an independent educator focused on the accounting measurement issues associated with implementing the Theory of Constraints (TOC). He died about 4 years ago (2007). Prior to focusing on TOC, he taught cost and management accounting and accounting information systems on the undergraduate and graduate levels. He has been a consultant in databased application systems since 1981. In addition to administrative experience as a university

department chair of an accredited accounting program, John served as an administrative contracting officer and has experience as a corporate controller.

John majored in Philosophy, emphasizing logical applications, at Washington University in St. Louis and then pursued graduate studies in accounting, receiving both masters and doctoral degrees in accountancy from the University of Missouri at Columbia. He is a Certified Management Accountant (CMA) and a "Jonahs’ Jonah".

The Theory of Constraints first attracted John’s interest in 1985 when Eli Goldratt made his famous "Cost Accounting: Public Enemy Number One of Productivity" presentation at the IMA convention. This was a time when academic and cost accountants were just realizing that their product costing systems were so flawed as to be causing devastating results in "the new manufacturing environment." After detailed study of the Theory of Constraints, he concluded that a great deal of what he was teaching was fatally flawed – for reasons that had never entered his mind!

He was concerned by the fact that most TOC implementations do not even come close to realizing the potential benefits of the philosophy. For many years he felt that the best thing that the accountants could do for a TOC implementation was to provide a "direct costing" type of information and otherwise keep out of the way.

"... clearly placed the management accountant in TOC as a necessary condition for success*. - Finance and Administration Manager
* for success are the blog owners editing. 

John believed that the full participation of the financial reporting function is a requirement for a successful TOC implementation. His Constraints Accounting seminar provides the opportunity for financial managers to accelerate the process of overcoming the inertia of traditional management accounting thinking. His POOGI Bonus seminar shows how individual goals can be aligned throughout the organization to support a process of ongoing improvement.

John's excellent book (written together with Pamela Caspari), "Management Dynamics; Merging Constraints Accounting to Drive Improvement" is a valuable addition to every manager's library. (ISBN 0-471-67231-9). Chapter 6 of John and Pamela's book goes into the question of pricing, probably a place to go to in order to  understand pricing and the possibilities better.

The article below asks several questions that were asked as a result of Tony Rizzo's article of the same title. The questions are repeated in John's text. Tony Rizzo's article resulted in considerable discussion. The complete discussion can be found here: http://casparija.home.comcast.net/~casparija/dweb/l118.htm

Enjoy!

INTRODUCTION

The issues associated with cost accounting have been an ongoing topic for the TOC discussion group. Three aspects of these discussions are activity based costing (represented by Tony Rizzo's paper entitled "THE FUNDAMENTALLY FLAWED THINKING BEHIND activity based costing"), cost based pricing illustrated by Stephane Mercier's queries), and cost allocation in general (exemplified by Tom McMullen's Detective Columbo and his chorus).

Tony's paper made the recommendation (or perhaps it was a 'speculation' or an 'hypothesis') that "it would be better to slaughter a farm animal and read its entrails than to use activity based costing" for decision-making. This resulted in a few folks saying "Great paper, Tony!" and others suggesting that Tony tone down his language. But NOBODY responded directly to the substance of Tony's recommendation. Neither confirming nor disconfirming instances have been presented.

Thus, there is a need to test Tony's hypothesis, respond to Stephane, and educate Columbo.

So, this post first addresses Detective Columbo's question: Why do we allocate costs to some product -- even when there is no volume linkage between the costs and the product?
Secondly, the excellent questions posed by Stephane will be addressed: What cost should be used to set prices, and ought the margins (of diverse products) always be equal? Finally, Tony's recommendation is placed against the background developed from the foregoing and put to the test.

The three topics are closely related. The sequence of presentation ties the three topics together in a logical progression that takes us to the unavoidable conclusion that . . . Well, read on. Perhaps you will find an interesting surprise. I did!

DETECTIVE COLUMBO'S QUESTION

In Tom McMullen's excellent summary of TOC (Note 1), his Detective Columbo observes that some allocations are required for external reporting, but questions why we allocate labor and overhead costs to products for internal purposes -- even when there is no volume linkage between the costs and the product. The best answer his 'group' can come up with is, "Because we always do."

For those on the list who are sophisticated in cost accounting, please bear with us. I believe that the vast majority of list members are not familiar with the details of cost allocation. I need to explain cost allocation, cost based pricing, and distinguish internal and external users.

For external reporting (audited financial statements and Securities and Exchanges Commission (SEC) reporting), Generally Accepted Accounting Principles (GAAP) are followed when they can be identified. The essential allocation question here is, "What costs and revenues should be recognized as related to current period profits on the Earnings Statement (Sales - Expenses = Net Profit) and which costs and revenues should be deferred to a future time period on the Balance Sheet (Assets = Liabilities and Owners Equity)?" Costs, then, are allocated between the Earnings Statement (as expenses) and the Balance Sheet (as assets).

'Cost accounting' has the allocation process at its core. This is a simple four-step process:

Determine the total cost (or other item) to be allocated, e.g., $100 of overhead -- or 12 ounce jar of peanut butter.

Determine the allocation base, e.g., 6 direct labor hours (DLH) for product A and 4 DLH for product B equals 10 total DLH -- or 24 slices in a loaf of bread.

Make the unit cost calculation, e.g., divide the total cost by the base ($10/DLH) or determine how much peanut butter should go with each slice of bread (0.5 oz per slice).

Put the unit cost with the elements of the base, e.g., $60 for product A and $40 for product B -- or make sandwiches with 1.0 oz of peanut butter per sandwich.

(Optional step) : Make a journal entry reflecting the allocation, make a decision based on the allocation, or eat a sandwich.

Almost all of cost accounting is just a variation of this process. Allocation includes inventory cost flow assumptions (FIFO, LIFO, etc), depreciation calculations, product and project costing techniques (such as activity based costing (abc), 'traditional' volume, e.g., DLH, absorption costing, direct costing, constraint or leverage point costing, what Rob Newbold (Note 2) calls 'throughput pricing', and Tom McMullen's TVA).

Many people interested in TOC use the foregoing terminology in different ways.

As a more responsive answer to Detective Columbo's question, and with respect to management (internal) uses of allocations, McFarland (Note 3) observed that "Cost accounting originated primarily to develop product costs for pricing . . ."

It's a shame that apparently neither Detective Columbo, nor any of hes group, had the opportunity to intrude on a management accounting class at one of the nation's influential universities. At the Harvard University Graduate School of Business Administration, for example, he might have encountered Professor Anthony (Note 4) saying, "Managers normally arrive at the selling price of a product on the basis of its full cost. Full cost includes a fair return on capital employed. . . . empirical evidence supports the premise that prices tend to be based on full costs." This view, which is widely accepted by the 1950s, is built on a foundation originating in World War I, experimented with during the 1920s, and cast into American national policy in the 1930s. (Note 5)

Thus, we see that product pricing is an important purpose of cost allocation.

STEPHANE'S QUESTIONS

Stephane posed the questions: What cost should be used to set prices, and ought the margins (of diverse products) always be equal? We need to explore how cost based pricing works and why we might want equal margins.

Let us examine a simple model of cost based pricing. But first, where do you (or your organization) stand on the gamut from laissez-faire capitalism to socialism? What is your goal? Do you want to earn oodles of boodle? Or, do you want to earn only an average profit? Is there a "stock" Theory of Constraints (TOC) answer to this question?

What are your markets like? Are they orderly and populated with a small number of good ol' boy suppliers? Would you want to disturb such a fine market? Has someone else -- perhaps an "evil" foreign dumper -- already disturbed the market? Or, does your market have many suppliers and rigorous price competition. Do your customers really care what it costs you to make your product (or provide your service)?

Do your good ol' boy buddies claim to price on full cost, but actually engage in marginal pricing? What did you think about the Harvard Professor's comment above?

I ask these questions because the answers are, or should be, reflected in your pricing methodology.

Let us examine an example of product costing and its use in pricing decisions. The data for the example is as follows:

1. A manufacturing firm has $440,000 of net invested assets.
2. The firm has the potential to produce four products: A, B. C, and D.
3. Annual market potential for these products is 200, 200, 300, and 400 units.
4. Unit costs are $200, $150, $100, $50.
5. Unit direct labor hours (DLH) are 10, 20, 20, and 10.
6. The firm has 8 direct labor employees who work 2,000 hours per year and earn $10 per hour, for a total annual direct labor cost of $160,000.
7. Manufacturing overhead is $320,000 per year.
8. Selling, General and Administrative (SGA) expenses are $240,000 per year.
   
   

Now, let us assume that we are sort of leaning toward the socialistic side of the gamut. We desire only an average return. Our costs are of concern only in the broadest sense. We are willing to forgo higher returns in the hope of uninterrupted profitability. We have been educated at either an influential university or an influenced university -- no matter, the result is the same either way. We have been taught to establish a target price by calculating the cost of a product ('product cost') and adding a mark up to cover any expenses not included in the product cost and provide a reasonable profit.

CALCULATING PRODUCT COSTS

Filling in the blanks of our allocation formula, we might want to spread all of our manufacturing overhead and SGA expenses out over a DLHs base. 

This would provide an allocation rate of [($320,000 / year + $240,000 / year = $560,000 / year) / (8 employees * 2,000 DLH / employee / year = 16,000 DLH / year) = $35 / DLH]. Finally, we would put $35 with every DLH. Thus a unit of Product A would cost [(materials = $200 / unit) + (direct labor = (10 DLH / unit * $10 / DLH) = $100 / unit) + (allocated costs = (10 DLH / unit * $35 / DLH = $350 / unit)] = $650 / unit.

Confirm for yourself that total unit costs of Products B, C, and D are $1,050, $1,000, and $500 when calculated in the same manner.

In similar fashion we might have calculated just the manufacturing cost of the products. 

For example, the manufacturing cost allocation rate might be calculated as [($320,000 / year) / (8 employees * 2,000 DLH / employee / year = 16,000 DLH / year) = $20 / DLH]. Thus a unit of Product A would cost [(materials = $200 / unit) + (direct labor = (10 DLH / unit * $10 / DLH) = $100 / unit) + (allocated costs = (10 DLH / unit * $20 / DLH = $200 / unit)] = $500 / unit.

Again, prove to yourself that the manufacturing unit costs of Products B, C, and D are $750, $700, and $350 when calculated in the same manner.

We know how to calculate a product cost; in fact, we know how to calculate two costs for a product. But, we may not know how the cost relates to a pricing decision. So we need to examine the pricing decision vis-a-vis product cost.

PRODUCT COST IN PRICING

How are these costs used in pricing decisions? There are basically four situations.

(1) In some cases the prices are regulated. In this case, one simply follows the regulations (which are almost always based on some measure of full cost)

(2) The market is purely (or quite) competitive and a market price exists. The competitive price is the sales price. The product cost is compared to the competitive price and the decision is whether or not to sell the product in that market.

(3) The market has a "price leader" or some other mechanism to indicate an appropriate price. Then managers use the information available to establish a price. Product cost itself might be such a mechanism. At the same university a decade later Columbo might have heard, "If all companies in an oligopoly use cost plus pricing formulas, then the pricing structure will be stable even during periods of declining demand. At a time when all firms in the industry face similar cost increases due to industrywide labor contracts or materials price increases, firms will implement similar price increases even with no communication or collusion. . .." (Note 6)

(4) Finally, the case of current interest is one in which we have very little information about the market. We may presume that there is a maximum price that the market will tolerate. If we price above that amount, we will not be able to sell our products, or will sell inadequate quantities. Once we set the price, it will be difficult to raise. Here, a cost based approach to pricing (cost plus a percentage of cost mark up) is a means of ensuring a "satisfactory" profit -- not a maximum profit. No attempt to obtain oodles of boodle here. On the positive side, it might result in a price so low that no signals would be sent to potential competitors.

How is this last price set? A trip to the library uncovers a dog-eared copy of an Accounting Handbook (Note 7) which tells us " . . . the cost base is the full unit manufacturing cost . . .. Selling and administrative costs are provided for through the markup . . .." The accountants are diabolical -- one rarely knows whether they mean "all costs" or "manufacturing costs" when they refer to 'full cost'. When I use 'full cost' without modification, I mean all cost (including SGA).

This recommended "full unit manufacturing cost" is the second set of product costs that we calculated. Let's use an -- arbitrarily selected - mark up of 54% of cost That would provide an asking price of (1.54 * $500.00 per unit) = $770.00 per unit for Product A. Prove to yourself that the asking price for Products B, C, and D would be $1,155.00, $1,078.00, and $539.00 respectively.

This budgeted product mix and pricing would result in a pre tax profit of $84,000 or a pre-tax return on investment (ROI) of $84,000 / $440,000 = $19%.

A mark up of 10% on the total unit product cost that we calculated would take us to the same overall profit, but the unit asking prices would be different. Check for yourself to see that the asking prices would be $715.00, $1,155.00, $1,100.00, and $550.00, but that the budgeted profit remains at $84,000. Only if the proportional relationships among the various cost elements (materials, labor, manufacturing overhead, and SGA) were the same for each product would the asking price calculated be the same.

This shows the mechanics of setting the price. We know how to mechanically calculate prices based on both manufacturing unit cost and total unit cost, but we don't know if the customers will buy the products at these prices. Hence, we need to add the customer to the model.

Let us assume that the utility of our products, in conjunction with competitive pressures, is such that the maximum unit amounts that our customers would be willing to pay are $600.00, $2,000.00, $1,700.00, and $2,000.00 respectively for products A, B, C, and D.

If we base our price on 154% of manufacturing unit cost, our asking price for product A ($770.00) is higher than the maximum price that our customers are willing to pay ($600.00). Therefore, we will not sell any product A. We would sell the other products. Unfortunately, this results in an overall loss of $30,000.

Thus it will be necessary to re-price our remaining products. We use the same four-step allocation procedure, but now remove the 2,000 DLH associated with product A so as to have a base of 14,000 DLH. Of course, we will need to increase the mark up percentage (to 63.7%) to compensate for the loss of product A. Carrying out the computations (go ahead and do it, don't trust me) we arrive at selling prices for the remaining products B, C, and D of $1,321.29, $1239.44, and $619.72.

Naturally, we have an extra person (2,000 idle DLH per year), but never mind -- our customers are happy and we are earning a satisfactory (as opposed to large) profit. Besides, we need the protective capacity.

Oh. By the way, note that our prices are well below the maximum amounts that our customers would be willing to pay. This means that we will be able to raise our prices if necessary to maintain our margins. We will be able to have those nice predictable earnings increases that the financial analysts so love. This will work as long as there is a great deal of slack in the markets. That is, as long as we don't have any serious competitors who don't "play the game". It worked in the United States from the 1930s until the 1970s and beyond.

Finally we can answer Columbo's question. Why do we allocate costs? To set prices. And because it works.

We see how cost based prices relate to the customer. But, we have two different ways to calculate "full cost" -- which do we use?

WHICH FULL COST BASE SHOULD WE USE?

We need to explore the ramifications of different cost bases.

The traditional means of setting a target selling price is to add a percentage of product cost to the manufacturing unit product cost. This is a middle ground between full cost (including SGA) and marginal cost (assigning only variable cost to the product and adding a mark up to the variable product cost).

Had Detective Columbo been at that management accounting class, he probably would have heard the recommendation that the full cost model be used unless all products use proportionally the same resources. In that case, it doesn't matter. He might have heard that the greater the amount of indirect cost (manufacturing overhead and SGA) and the greater the diversity in "products, customers and processes" (Note 8), the more important to use a full cost model. Otherwise, we would assign too much cost to "low overhead" products, resulting in their not being competitively priced; and we would assign too little cost to "high overhead" products, resulting in a price too low to recover the overhead which they caused.

Activity based costing, as generally recommended, is an instance of the full cost model. I believe that the activity based costing model has now replaced the "manufacturing cost only" model of earlier years in the academic recommendations.

HOW IS abc USED?

From these basic alternatives, we know that abc is recommended by academics for pricing. But we don't know for what purpose activity based costing is actually being used or recommended. We need to determine for what purpose activity based costing is being used IN ORDER TO COMPARE IT TO Tony's recommendation.

So, in response to discussion on the list indicating that activity based costing/management and TOC work well together -- abc for some things, TOC for others -- I queried the group (on 9-3-98) as to exactly what were these things that abc did better. I received no replies to that query. However, Scott Ward had previously observed (9-1-98), "I have found [activity based costing/management] useful in making better pricing decisions regarding support costs." Since this is consistent with other uses of product cost data, I accept pricing as a thing that abc claims to do better.

TESTING THE abc FULL COST BASE SELECTION

When working on some thinking processes trees for a book about TOC and accounting measurements, I had arrived at the unavoidable conclusion that setting prices on the basis of full costs would be unlikely to work well. I needed a way to test the abc claim against the logic of my trees.

So, I wrote a simple simulation routine using an MS Excel worksheet. In this simulation I created a company having nine products. I established a market potential in terms of units that could be sold -- provided that the asking price was not more customers were willing to pay. The customers' "maximum price" was generated as a uniformly distributed random number.

I then calculated asking prices based on both traditional direct labor based manufacturing cost and on full activity based costs. The "product cost" amounts were increased by mark up percentages to arrive at simulated selling prices. The mark up percentages were selected so as to yield equal net profits if all products were sold (i.e., budgeted profit was the same for both traditional and abc product costing).

If a product's asking price was less than the customers' maximum price, then the product was "sold". If the asking price was above, but within 10%, the price was reduced to meet the customer's price.

This allowed me to compare the dynamics of the abc versus the traditional pricing scheme.

The results after 500 iterations tended to support my original analysis. Abc pricing resulted in average profits of about $105,000. Traditional pricing resulted in average profits of about $235,000. Traditional profit was greater than abc profit on about 60% of the trials.

This leads to the question, "Why?"

Well, it seems that the traditional direct labor based pricing model was making a better use of available resources by doing some "marginal" (less than full cost) pricing. This caught me by surprise. I suppose that traditional full costing might be just a "politically correct" way to do marginal pricing.

TESTING TONY'S RECOMMENDATION

In a copyright article entitled "THE FUNDAMENTALLY FLAWED THINKING BEHIND activity based costing", Tony Rizzo suggested that, for decision making, "We would stand a better chance of success if we gutted a farm animal and read its entrails."

Since I had already built the simulator, I wanted to test Tony's recommendation that we obtain a farm animal and . . .

Well, I didn't know how to simulate the gutting of a farm animal and the subsequent reading of its entrails. I also didn't know if some sort of advanced degree is required for entrails reading.

I needed a way to simulate entrails reading.

I decided that an advanced degree is probably not necessary, and that what Tony means is that any old random thing would be better. So I use the random number generator to generate a price. The random price fluctuates entirely without regard to costs.

EGAD. Average profits (using the "farm animal technique") are now up to about $350,000.

So, I guess that Tony is right.

Note 1: Thomas B. McMullen, Jr., Introduction to the Theory of Constraints (TOC) Management System, St. Lucie Press, 1998, pp. 217-224.

Note 2: Robert C. Newbold, Project Management in the Fast Lane: Applying the Theory of Constraints, St, Lucie Press, 1998.

Note 3: Walter B. McFarland, Concepts for Management Accounting, National Association of Accountants (now the IMA), 1966, p. 61.

Note 4: Robert N. Anthony, Tell it like it was: a Conceptual Framework for Financial Accounting, Richard D. Irwin, 1983, pp. 37-38.

Note 5: C. J. McNair and Richard Vangermeersch, Total Capacity Management: Optimizing at the Operational, Tactical, and Strategic Levels, The IMA Foundation for Applied Research, 1998, Chapter 9.

Note 6: Robert S. Kaplan and Anthony A. Atkinson, Advanced Managerial Accounting, Prentice-Hall, Inc., 1989, p. 187.

Note 7: Joel G. Siegel and Jae K. Shim, Accounting Handbook, Barrons Educational Series, Inc., 1990, p.164.

Note 8: Robert S. Kaplan and Robin Cooper, Cost and Effect: Using Integrated Cost Systems to Drive Profitability and Performance, Harvard Business School Press, 1998, p. 164.

Technorati Tags: ABC, Accounting, Cost, Cost Accounting, Costing, Finance, Full Absorption Costing, Shareholder Value Add, Strategy and Tactics, SVA, Theory of Constraints, TOC