The Engineering Economist: A Journal Devoted to the Problems of Capital Investment

(1)

This article was downloaded by: [Columbia University]

On: 14 February 2015, At: 16:56 Publisher: Taylor & Francis

Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Engineering Economist: A Journal Devoted to the Problems of Capital Investment

Publication details, including instructions for authors and subscription information:

http://www.tandfonline.com/loi/utee20

Activity-Based Management: Past, Present, and Future

H. Thomas Johnson ^a

a Portland State University Published online: 06 Apr 2007.

To cite this article: H. Thomas Johnson (1991) Activity-Based Management: Past, Present, and Future, The Engineering Economist: A Journal Devoted to the Problems of Capital Investment, 36:3, 219-238, DOI: 10.1080/00137919108903046 To link to this article: http://dx.doi.org/10.1080/00137919108903046

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic

reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/

terms-and-conditions

(2)

A<ctivit)/-BasedMa:nagement:

Past, .Pres·ent,andFuture

H. Thomas Johnson Portland State University

ABSTRACT

The flood of publications and semina:rsonactivitycosting in the last few years suggests that companies suddenly encountered a need for better management accounting information in the 1980s. In fact, management accounting's relevance to most business decision-making deteriorated steadily since the 1950s.However, most companies did not perceive these inadequacies .in management accounting until the early 1980s. Current interest inaetivitymanagementand activity costing reflects the convergence around 1980 of two forces: a .long-undetected set of problems in management accounting that seems to originate in the 1950s; and thegrow·thof new competitive pressures in the 1970s thatmadecornpanies acutely aware of these problems. Alt;houghthese two forces arose independently, their interaction in the 1970s led to the rise of activity-based management thinking. This paper describes thepast,present, and future of activity-based .management: past uses of financial accounting information that confounded companies'efforts to plan marketingstrategies-andto control operations after the 1950s; present ideas for solving these problems with activity-based management concepts such as activity costing; and the likely future direction of activity-based management thought.

INTRODUCTION^l

Over the last two centuries,businesseshave used financial and nonfinancial information to direct management xlecislonsut tthree levels: to control the work of individuals, to control subordinate production units, and to plan the extent and financing of the enterprise as a whole.² Until about forty years ago they generally used financial accounting information to plan the extent and financing of the firm .asawhole. They used nonaccounting information,both financial and nonfinancial, to .control the .work of individuals and production units. .In the 1950s,however,businessesbegan to use financial accounting information todireetmanagementdecisionsat all three .levels: to control workers and subunits in addition to planning the extent and financing of the <enterprise as a whole.Usingfinancialacconnting .informationtocontrol people as well as to plan financial consequences is what present-day accountants refer' to as management accounting.³

219

Downloaded by [Columbia University] at 16:56 14 February 2015

(3)

The Engineering Economist

Accountants since the 1950s consider it natural and inevitable that a company's financial accounting system should be its primary source of management information.

They believe that businesses first created accounting systems to collect and report information on financial transactions; that businesses began to use that accounting information to manage internal activities when those activities reached a requisite level of complexity, which in most cases occurred, apparently, after World War 11. However, recent historical research casts doubt on the idea that financial accounting systems provided the first source of sophisticated financial management information.

Businesses, especially in manufacturing, created very sophisticated financial and nonfinancial management information systems between the early 1800s and the 1920s.~

The financial management information referred to here was not necessarily derived from accounting records - even though it sometimes was reconciled with account data. Rather, it consisted of cost and margin information derived primarily from data about work processes and other activities, and it was used to control workers and to evaluate the performance of companies' subunits.

Historians now believe that following World War I1 financial accounting information intruded upon and distorted the financial and other information companies had used for decades to manage not only operating activities at the worker and the business unit levels but also strategic product choices at the enterprise level. Contrary to popular opinion, new management accounting developments after World War I1 do not indicate financial accounting's increased relevance to decision-making inside complex organizations;

instead, they reflect a fall of management accounting from re~evance.~ Indeed, many authorities now believe that using financial accounting information to plan and control business activities contributed to declining competitiveness and profitability in many American manufacturing companies after 1960. ⁶

ACTIVITY-BASED MANAGEMENT

Problems caused by management accounting's lost relevance triggered the development of activity-based management concepts in the 1970s and 1980s. Basically, activity-based management asserts that the causes of profitability and cost in most businesses are too complex to know or control by referring to financial information recorded in accounting records or reports. Instead, companies must manage financial results and track costs with information about activities (or work); in particular, with information on how activities that consume resources and satisfy customer wants promote competitiveness and profitability.7

(4)

An allegory from a classic of Western philosophy provides a powerful image with which to portray the relationship between activity-based management and post-1950s financial management.' In the Republic, Plato articulates a theory of knowledge - that is, how humans go from an unenlightened state in which they take appearances at face value to a state in which they understand reality. Activity-based management concepts address a similar question; namely, how do people in companies sort out appearance from reality?

Specifically, what information do companies need to identify genuine sources of competitiveness and profitability?

In the Allegory of the Cave, Plato considers the condition of men who have lived all their lives underground in a cave: ⁹

Imagine the condition of men livin in a sort of cavernous chamber underground, with an entrance open to t

% .

e light and a Ion passa e all down the cave. Here they have been from childhood, chained y the

7

eg and also by the neck, so that they cannot move and can see only what is in front of them

...

At some distance higher up is the light of a fire burning behind them; and between the prisoners and the fire is a barrier] like the screen at a

\

puppet-show, which hides the performers whi e they show their puppets over the top. Now behind this barrier] imagine persons carrying along various artificial objects, including

I

igures of men and animals in wood or stone or other materials, which pro'ect above the [barrier]. ... Prisoners so confined

l

would have seen nothing o themselves or of one another, except the shadows thrown by the fire-light on the wall of the Cave facing them. ...

Plato asks us to imagine a prisoner breaking away from his chains and walking out of the cave, passing by the "puppet show" along the way. After overcoming the pain and difficulty of adjusting to light, the freed prisoner would move from the shadow world of appearances, past the barrier where artificial likenesses of real objects cast shadows, to the outside world where he would see real objects for the first time. After living for a very long time in the outside world, the freed prisoner eventually might perceive the highest knowledge, the spiritual reality permeating all objects, represented by sunlight.

We can relate this allegory to post-1950s financial management practice by equating the cave and the ground above with a company and the shadows cast on the cave's wall with financial accounting numbers. Think of those numbers as money-denominated shadows cast by the resources a business consumes and the products it sells. Resources (e.g., labor, raw material, buildings, machinery, and so forth) and products are the "artificial" objects that cast monetized shadows from behind a barrier.

People behind the barrier - perhaps management accounting - manipulate products and resources and stoke the fires that cast shadows on the wall. On the floor of the cave, chained in their places, are other people

-

perhaps finance-oriented executives and

(5)

222 The Engineering Economist

financial analysts - who believe that the finance and accounting shadows passing before their eyes are reality itself, not mere appearances.

Above ground, outside the cave entrance, are the real objects represented below by artificial likenesses - products and resources - that cast financial shadows on the wall.

Those real objects above ground are people, especially customers seeking satisfaction of wants and company associates (employees, managers, suppliers, distributors, etc.) performing activities that consume resources, hopefully to satisfy customer wants. Thus, a freed prisoner who might break out of the cave and walk above ground would progress from watching accounting shadows on a dark wall, to seeing consumed resources and products that create those shadows, and, finally, to observing people performing activities that consume resources and cause spending to occur.

Activity-based management advocates argue that top managers must not use financial shadows either to control operating activities - the work that links employees and other company associates with final customers -or to plan marketing strategies. The gap between appearance and reality - between financial shadows and the activities causing those shadows - is too great, say activity-based management authorities, to assume that traditional management accounting will point a company toward competitive and profitable outcomes. Top managers may, of course, believe that they control reality by manipulating financial variables - especially if their knowledge of business comes primarily from studying financial shadows, not from studying customers or operating activities.

Undoubtedly that was the belief of a great many executives who ran American businesses in the 1970s and 1980s. Companies often lost market share and profitability because financially-oriented managers let the cart

-

studying the results -get ahead of the horse - understanding how resource-consuming activity satisfies customer wants.

Managing businesses with financial accounting information impaired companies' efforts to compete and profit in the past thirty years for two reasons. First, manufacturing cost accounting systems give misleading signals about costs of products. Specifically, traditional cost accounting information does not accurately reveal the high costs of complexity caused when companies proliferated their product lines in the last forty years. ¹⁰ Evaluating product costs with this financial cost information prompted many manufacturers in the 1970s and 1980s to make erroneous strategic marketing decisions, leading often to diminished profitability and, in some cases, bankruptcy.

Secondly, using financial accounting information to control operations divorced managers increasingly from their companies' markets and operating activities." By the 1970s top managers too often were selected for their skill at financial management, not for their knowledge of markets and operations. Most companies led by these

(6)

financially-oriented managers failed to understand and adapt to the new terms of competitiveness that overtook global markets in the last 21 years or so. They lost markets to companies, located generally outside the United States, in which financial management practices had not blinded top management to the new imperatives of global competition.

The following two sections describe the past and the present of activity-based management: past uses of financial accounting information that confounded companies' efforts to plan marketing strategies and to control operations after the 1950s; and present ideas for using activity-based management concepts to solve these problems. A concluding section discusses the future of activity-based management concepts. Some ideas discussed here are controversial. Although virtually all authorities believe today that activity costing resolves the product costing deficiencies inherent in traditional management accounting, opinion is divided on whether activity cost information also helps operating managers reduce costs and improve long-term profitability. Presumably managers who understand these controversies are better equipped to distinguish the shadows of financial management from the reality of global competitiveness.

STRATEGIC PLANNING INFORMATION

As a guide for planning, and to choose among alternatives, businesses need information about the financial consequences of intended actions. They especially need reliable cost information. Cost information serves in many planning and decision support roles, such as estimating profit margins of products and product lines, evaluating decisions to make or buy components, preparing departmental cost budgets, and charging administrative services to production departments.

An important source of cost information in American business since the 1950s has been the financial cost accounting system. Cost accounting systems were designed in the early 1900s t o attach production costs to manufactured goods in order to divide an accounting period's total production costs between products sold and products still unfinished or unsold at the end of the period. These systems were designed to facilitate preparation of financial reports required by regulators, taxation authorities, and participants in capital markets. They were not intended to provide information about costs of individual products. Moreover, companies rarely used financial cost accounting information to gauge individual product's costs before World War 11. In the late nineteenth and early twentieth centuries, companies often managed complex multiproduct strategies with nonaccounting information or by creating focused organizational profit centers.12 But companies everywhere used information from the financial cost accounts to evaluate costs of products after the 1950s.

(7)

Financial accounting systems provide poor information to evaluate modern manufacturers' product costs.13 The manufacturing cost accountants' traditional approach to allocating overhead costs - using volume-sensitive cost drivers such as direct labor hours, machine hours, or material dollars - systematically distorts the costs of individual products. Using such volume-sensitive drivers to attach overhead costs to products is a convenient and economical way to insure that production costs are properly matched against revenues at a macro level in financial statements. But at the micro level of the individual product this exclusive reliance on volume-sensitive drivers provides reliable cost information only if we assume most overhead costs are triggered by or vary in proportion to units of output.

However, the fastest growing overhead costs in American manufacturing companies after the 1950s were caused by drivers that are triggered by batches put into production and by number of product lines, not by units of output.14 Setting-up and material handling, for example, are drivers of overhead cost that are triggered by batches of output, regardless the number of units in a batch. A setup costs just as much for a batch with one unit as for a batch with 100 units. Engineering change notices and the number of part numbers in stock also drive overhead costs, but both are triggered by introducing new product linee, not by the number of batches or number of units produced. By using drivers triggered by units of output to allocate overhead caused by batch and product-line drivers, companies systematically undercost low volume products that cause overhead growth (e.g., capital-intensive products that are custom-made with newer, less familiar, and more expensive materials and equipment; as well as rapidly-proliferating varieties of new products that demand expensive design, scheduling, and rework time - all sources of overhead costs). Moreover, they systematically overcost high volume products that do not cause overhead to grow (e.g., established lines of commodity-type products that are mass-produced with older labor-intensive technologies).

These systematic distortions tend to cancel out at the macro level and therefore do not affect income and asset totals reported in financial statements. But they give a misleading picture of individual product's margins, as many American and European manufacturers discovered in the 1970s and 1980s when, using financial cost accounting information to measure product costs, they erroneously assumed they could improve their company's profitability by abandoning overcosted commodity-type product lines and by proliferating undercosted varieties of newer "high-tech" lines. In fact, that strategy usually depressed earnings and, in several cases, generated a "death spiral" that led companies to the edge of bankruptcy. ¹⁵

(8)

Recognition of problems with traditional product cost accounting grew during the late 1970s. Activity-based costing (ABC) began to appear in the early 1980s as a solution to the problem with product cost information.16 Advocates of ABC tell companies, in effect, to cost products differently for financial reporting information than for planning and decision support information. For financial reporting they recommend companies continue allocating overhead using the volume-sensitive drivers they have used since the early years of this century. For more reliable strategic planning information, however, proponents of ABC tell companies to trace costs to both volume-sensitive and nonvolume-sensitive driver pools. ABC estimates costs of products by adding up costs of the actual drivers that each product consumes.

Simple in concept, ABC was a practical impossibility until the advent of low-cost microchip technologies in the 1970s made it economic to collect and compile large amounts of nonvolume-sensitive cost driver information. In principle, a two-stage approach is used to conduct an ABC analysis. The thrust of the design, which was codified originally by Robin Cooper of Harvard Business School, is to identify a relatively small set of both volume-sensitive and nonvolume-sensitive overhead cost drivers (say 6 to 12) in stage one and to trace indirect costs to each driver. in stage two, the company determines the percentage of the drivers consumed by each product or service. The result is an estimate of the indirect costs of each product based on nonvolume-sensitive drivers such as engineering change notices (ECNs), setups, and inspections, as well as the traditional volume-sensitive drivers such as direct labor hours and material dollars.

For example, consider purchase orders, a nonvolume-sensitive overhead cost driver.

Suppose it costs $100 to process purchase orders for the total output of a plant

-

say ten cars. If each car requires the same direct labor, the old accounting methods would distribute the cost equally to all units at $10 a car. However, one car might be a custom model in which half the parts are non-standard and have to be ordered individually. Using the new methodology of ABC to assign costs per purchase order, that car might be assigned

$50 in purchasing overhead while the remaining $50 is allocated to the other nine cars.

This kind of cost-driver information is excellent for making marketing decisions. In evaluating prices, for instance, it helps marketing managers confirm their suspicions that a company has been overestimating the cost of standard products. It also helps in evaluating the profitability of product mix. With this kind of information, managers in tough competitive situations can "know when to hold and know when to fold." Products with lower margins will be an easier discard. Those with higher margins will clearly be worth fighting for.

(9)

Two-stage, multiple-driver ABC analysis reliably costs a wide array of products that consume resources in diverse ways. As it's currently used by most companies, however, activity-based costing is frequently a misnomer: the "drivers" the tool costs are seldom "activities1'. The difference is important. Activity connotes people doing work and it properly is described with phrases that contain a verb and a noun -for example, move a cart, file a folder, or drill a hole. Activities often are defined at high levels as linked series of steps that create a specific result or output (usually for an identifiable "customer") -for example, process an order, insert components in a circuit board, process a payroll check, or assemble a part. The term "driver," occasionally referred to as a cost generator, sometimes refers to activities but more often refers to things described by a noun - for example, purchase orders, product parts, setups, or slots in a circuit board. Drivers initiate activities and, therefore, cause costs. But they are not necessarily activities. As shown in the next section, steps to economize on drivers may not lead to proper management of activities to achieve global competitiveness and long-term profitability.

Information about costs of drivers, such as the cost of a purchase order, can be an efficient and very effective way of differentiating unit, batch, and product-level costs among products. Manufacturers often have easily available driver information in production control and MRP systems. Bills of material and product routing data give counts by product of the numb& of setups, the number of material moves, the number of part numbers, and so forth. Most companies that have implemented activity-based costing systems to date sort costs into pools defined by such drivers, count the drivers incurred during a production interval, and count the drivers in each product line during the same interval. In concept, a t least, it is a simple matter to trace indirect costs to products using the average cost per driver in each product. While not perfectly accurate," it is widely believed that this technique provides a better "rack and stackv1'* of indirect costs than traditional systems that allocate all indirect costs with one or a few volume-sensitive drivers.

Return briefly to Plato's Allegory of the Cave and consider how driver-based activity costing helps managers progress from appearances - shadows on the wall of the cave

-

to understanding reality. Sorting and tracing costs by volume-sensitive and nonvolume-sensitive drivers helps managers look "behind the barrier" and clarify differences in the resources a company consumes to design, make, and deliver diverse products or services. ABC's power to clarify relationships among products and resources resembles the power of other sophisticated accounting tools that managers and financial analysts have used since the 1950s to "get behind the shadows". Examples of these other tools include: DuPont's ROI (return on investment) formula, originated at DuPont by

(10)

Donaldson Brown in 1912, and widely publicized after 1950 as a tool for analyzing the impact any element in the balance sheet or income statement has on ROI"; variance techniques for assessing the impact on net income of changes in price, volume, and mix of products and resource inputs that became widely taught in American business schools after the 1 9 5 0 s ~ ~ ; and shareholder value analysis tools that stress the importance of using cash flow and debt leverage information to roll back clouds of obfuscation created by GAAP-style accounting inf~rmation.~' Activity-based costing, as a technique for eliminating distortions in traditional cost accounting information, ranks among these major accounting innovations of the twentieth century.

However, tools that analyze ROI, variances, shareholder value, and activity-based cost simply look "behind the barrier" at "artificial objects - products and resources -that create accounting "shadows". While their use reflects managers doing much more than merely reacting passively to accounting signals, their use still reflects managers focusing on artificial objects in the cave, not real objects above the ground. To go outside the cave requires direct analysis of activities and customer wants.

OPERATIONAL CONTROL INFORMATION

Companies before the 1950s used accounting information, in the words of General Motors chairman Alfred Sloan, "to control the pursestrings, not guide the hands of the

artisan^."^'

A major cause of management accounting's lost relevance after the 1950s is the habit companies developed of using financial accounting information to control operating activities

-

"the hands of the artisan." Why this habit appeared when it did is not entirely clear at this time. Probably it originated in the multidivisional organization structures that proliferated after the 1950's. Large-scale organizations divisionalized in order to economize on the high costs of information needed to manage the complexity of diverse product lines. In multidivisional organizations top managers unequivocally use financial accounting information to control the actions of subordinate managers for the first time. After 1960 companies increasingly used accounting results such as costs, net income, or return on investment (ROI) to evaluate and motivate the performance of operating personnel at all levels.

Although their efforts were not widely publicized, a few companies and consulting firms in the 1960s and 1970s looked beyond financial accounting shadows, resources, and products to analyze operating activities directly. In General Electric, perhaps the first company to undertake activity analysis, financial executives by the late 1960s advocated using activity information tools to identify cost-cutting opportunities. Within a decade, several consulting firms bad developed similar activity management tools. Using these

(11)

228 The Engineering Economist

tools, companies could discover opportunities for productivity improvement that traditional financial management tools seldom uncover. Until very recently, no one seems to have developed sophisticated tools to analyze customer wants or to link customer wants with work. By 1980, however, at least a few companies had used activity management tools "outside the cave," t o analyze the resource-consuming work, or activity, that truly causes cost -something rarely done in most ABC studies described to date.

One popular mode of activity analysis begins by building an extensive dictionary of key activities under each of an organization's functions.23 The supervisor of each department then creates a table showing all the people involved in each of the key activities and estimates the percentage of time they spend on each activity. Combining all departments' tables yields a composite cross-functional matrix that portrays what people do in a company far better than any traditional management accounting information.

Indeed, managers analyzing a completed cross-functional matrix for the first time are amazed at how much time is spent on certain activities, such as complaint handling or rework. Moreover, they are surprised to find how certain activities, such as budgeting, redundantly consume time in almost every part of the organization. Here are examples of surprises that often surface in a cross-functional analysis, never in the usual analysis of cost reports:

1. Cost accounting information often confounds efforts to manage costs simply because it shows only where money was spent, and how much, not why it was spent. A company's production department in Cleveland recorded costs in two separate lines for resin and maintenance incurred in running extrusion machinery. Dumpsters full of defective extrusions and extra maintenance to unclog gummed-up extrusion machines showed up in the accounts as extra costs of production in Cleveland. A cross-functional analysis of activities revealed, however, that resin and maintenance consumed in the production department reflected a policy, carried out by the company's purchasing department in Baltimore, to "buy in large quantities from vendors that quote the lowest price." Previously unsuccessful efforts to manage cost "shadows"

in Cleveland had almost led the company's financially-oriented managers to

"reduce costs" by outsourcing extrusion to a Third World country. The outcome of cross-functional activity analysis was to recommend, instead, changing the purchasing policies executed in Baltimore. ²⁴

2. A large telecommunication company for years had managed operating units with monthly cost budgets. New competitive pressures in the mid-1980s

(12)

caused them to consider cross-functional activity analysis as a tool to reduce costs and improve productivity. Activity analysis in a pilot shop revealed that people spent almost 30 percent of their time in various types of rework activity. No one had ever questioned the necessity of any of this work, but then no one looking at the shop through the customary cost reports had ever really examined the nature of the work. Once the rework was identified and its causes traced the company was able to eliminate it and redeploy almost one-quarter of the shop's work force. ²⁵

3. In 1978 a manufacturer of stainless steel milk cooling tanks used activity analysis to reduce costs. With this analysis, management of the company observed that at one plant the second shift required more workers (53 as opposed to 47) and more overtime to produce the same amount of product as the first shift. They began by defining, tracing and estimating the time spent on activities during both shifts, such as receiving and stocking, setting up and changing over, inspecting and testing, reworking, waiting, housekeeping, expediting, and maintaining equipment. To their surprise, they found that the second shift had about the same number of setups, inspections, and reworking. But they spent three times as much of their time performing these and other indirect activities such as stocking, handling materials, maintaining equipment and looking for parts. Why? First, supervisors and support personnel went home at 5:00 every day and left the second shift to fend for themselves. Second, the company always started inexperienced workers on the second shift. Instead of taking the traditional cost-oriented approach of laying off workers, the company hired two materials expediters and one supervisor for the second shift. Even after the additional $80,000 in salaries, increased productivity led to an annual reduction in total labor costs and overtime premiums of $145,000 (in 1978 dollars).26

In each of these cases companies perceived opportunities to save costs by analyzing information about operating activities, not information about costs of operations. These cases suggest the possibility of r u ~ i n g a company's operations without the traditional cost accounting control information. Indeed, leading operations management authorities for years have advocated dispensing with cost control information. Thomas Vollmann, co-director of Boston University's Manufacturing Roundtable, equates most current efforts to reform manufacturing accounting systems with ancient warriors' attempts to untie the massive knot on the yoke of King Gordius' chariot.27 An oracle had prophesied that

(13)

230 The Engineering Economist

whoever loosed the Gordian Knot would rule the world. After countless challengers failed in their attempts to untie the knot, Alexander the Great came up to it with his sword and cut it in two. "Companies," say Vollmann, "have similarly discovered that the simple, bold move of cutting the knot between accounting and performance measurement is much more effective than trying to untie it."28

In a similar vein, Richard Schonberger says that a company using continual improvement and total involvement to control the causes of cost "does not need to rely on cost information for management control." Schonberger believes most companies are reluctant to accept this idea because they have "relied on costs so long that [they] cannot yet imagine using costs just for making product-line and pricing decisions and not for cost control.112g Finally, Robert Hall believes that "financially-oriented management must be weaned from thinking that detailed financial goals stimulate operating improvement.

[Over time] it becomes obvious that actions improving quality or reducing leadtimes will reduce operating costs."30

Although authorities such as Vollmann, Schonberger, and Hall advocate using nonfinancial performance indicators of time, space, distance, and quality rather than cost information to manage operating activities, they are not opposed to collecting cost information for planning and evaluating events in the workplace. Obviously companies must keep financial score. But the financial score is like the score in a tennis match. As tennis players know, it is necessary while playing to keep one's eye on the ball, not on the scoreboard. But nonfinancial measures of operating performance, unlike traditional cost measures, are the ball, not just the score. Controlling activities in order to improve those nonfinancial measures undoubtedly will improve a company's competitive position. But controlling activities in order to manage costs may not improve competitiveness.

Indeed, companies often impaired their competitiveness after t h e 1950s by using standard cost targets to control the performance of operating personnel.31 Almost all American manufacturing companies for the past forty years have used cost targets from top-level planning budgets to set standards for operating personnel. These cost targets are seen as an important tool to control the operating performance of plant managers and department supervisors. Like the setting for desired room temperature on a thermostat, cost targets are a setting to compare against actual costs. Variances between actual and desired costs provide "feedback" that is supposed to prompt operating personnel to adjust what they are doing, as a furnace adjusts in response to feedback from the thermostat.

Standard cost variance systems monitor costs in each and every process of a company's production system. For direct costs, labor and machine tracking schemes report direct costs per hour or per unit of output. For overhead costs, reporting schemes track the

(14)

percentage of overhead "covered" or "earned" by units produced. The goal of these reporting schemes is to have all recorded direct labor or machine hours go toward production of standard output and thereby "absorb" or "cover" direct and overhead costs - a condition referred to as "efficient."

This system encourages department managers to schedule production in long runs, so workers charge less time to categories of indirect or "nonchargeable" time such as changeovers or setups. Because output enables a department to "earn" the direct hours incurred each reporting period, supervisors keep workers and machines buy producing output. Ironically, these efforts to achieve high standard cost efficiency ratings tend over time to increase a company's total costs and to impair competitiveness. Achieving standard direct cost efficiency targets leads to larger batches, longer production runs, more scrap, and rework

-

especially if incentive compensation is geared to controlling standard-to-actual variances. Pressure to minimize standard cost variances, by encouraging department supervisors to keep machines and people busy producing output, regardless of market demand, often causes unnecessary inventories of finished and in-process merchandise to accumulate, product lead times to increase, and dependability at keeping schedules to decrease.

Managing costs with accounting information in standard cost systems impedes companies' competitiveness and long-term profitability primarily because it motivates people to sustain output in order to achieve cost targets. It encourages managers to achieve financial cost targets by producing output for its own sake, instead of encouraging them to focus on the one key to competitive operations and long-term profitability - namely, empowering people to efficiently satisfy customer wants. Indeed, customers scarcely fit into the world of standard cost performance. The customer is merely someone the company persuades to buy the output that managers are driven to 'produce, at prices it is hoped exceed variable costs.

One solution to the dilemma created by using standard cost information to set performance targets is to decouple the collection of cost information needed for financial reporting purposes from the information that was used to control performance. Recent proposals for "backflush" accounting have that decoupling as their goal.32 However, backflushing itself does not answer the need for operating cost information.

Perhaps the most logical way to compile operating cost information is to cost activities. When viewed acoss functions, activity cost information offers managers a snapshot view of costs from an entirely different perspective than the view presented in traditional cost reports and budgets. Compiling this activity-based cost information begins with the table of activities described above, showing the time people spend in key

(15)

232 The Engineering Economist

activities by function or department. Fully loaded payroll costs and controllable nonpayroll costs are used to cost the time people spend on activities. A completed cross-functional activity cost table can "tie-out" with a department's or entire company's total costs as shown in budgets or other reports. Moreover, it often is used to estimate the cost of drivers, opening the way for defining costs not just by function, but also by product-line, customer, distribution channel, and so forth.

Cross-functional activity cost analysis (CFACA) is much more detailed and informative than the driver-based activity costing (ABC) discussed in the previous section.

As we noted before, "activity-based costing" is a misnomer when applied to driver-based cost information. Driver information improves the reliability of product costs for strategic decision-making, but it is not in fact activity information. If it is used to control operating activities it can cause companies to make decisions that impede their ability to compete and profit in the customer-driven global economy. Here are two examples of the problems caused when driver-based ABC information is used to make operating decisions.33

1. An auto component manufacturer called in outside experts to advise them on product costing. Using ABC, they found that the company was abandoning its most profitable product lines and replacing them with lines that barely broke even, if at all. The results of ABC were eye-opening for this company:

they found that product costs varied by as much as 300 percent from what they had thought.

Then they started to manage the driver costs. One insight the new ABC data revealed was how much it really cost to set up machines every time an order for a batch of components was released to the shop floor. The old cost accounting system pooled setup costs with all other indirect factory costs and spread them over components in proportion to direct labor hours.

Components ordered in large lots that keep machines running steadily for long periods of time absorbed the same overhead cost per unit (including setup costs) as components ordered in small lots that required more frequent and costlier setting-up. Obviously, the old costing system did not reveal the true cost of handling small lots produced in short runs. The ABC analysis, however, by pooling setup costs separately from other indirect costs and applying them to components in relation to their demands for setup time, eliminated that distortion and put the costs where they belonged

-

on the small lots.

(16)

The new information prompted management to alter its pricing and operations practices. They cut prices of components produced in large lots and charged a premium to buyers ordering small lots. They also discouraged the sales staff from taking orders from buyers who insisted on frequent delivery of small quantities.

2. A manufacturer of personal computers and electronic measurement equipment followed a similar path in applying ABC information to operations. Management had always costed products by pooling all factory overhead and allocating it using direct labor hours. Thus, the cost accounting system did not reliably differentiate between costs of printed circuit boards that were fabricated through very different processes, such as three different types of component insertion (dip, axial, and manual) and two types of soldering (wave and manual). Consequently, costs of boards with substantial numbers of manually-inserted components were not significantly different from costs of boards with components inserted automatically. This was counter-intuitive to what every design engineer believed to be the true costs of printed circuit boards.

The company performed an ABC analysis to give product design engineers more reliable information about the cost of design decisions. In this anlysis, indirect costs (essentially all costs except materials and purchased components) were pooled separately for automatic and manual insertion.

ABC confirmed that manual insertion procedures cost several times as _much as auto-insertion.

However, auto-insertion machines could not operate reliably in spaces as small as human hands could. Therefore, designs for products made on auto-insertion machines had to space components further apart and further from the edge of a board. That meant trading off lower auto-insertion costs for somewhat larger boards. Nevertheless, design engineers proceeded to design manually-inserted boards out of existence.

In both of these cases, the ABC information helped management see ways to reduce costs and improve short-term profits by altering product mix or process mix, not by altering the way activities are performed. They improved the bottom line either by moving out of products that consume a disproportionately high share of costly drivers, such as setups, or by designing products to use processes that consume less costly drivers, such as auto-insertion. But long-term, both companies made choices that were likely to impair their competitiveness and profitability. To achieve competitive and profitable operations

(17)

in a customer-driven global economy, companies must give customers what they want, not persuade them to purchase what the company presently produces at lowest cost. If customers favor small lots and smaller electronic products, then companies must respond, even when it presently costs more.

In both cases management failed to ask two crucial questions:

1. Do consumers really want the product that the ABC analysis says is the most profitable?

2. If not, what type of analysis can help us identify changes in the way we work so we can produce the product customers want efficiently and effectively?

To compete well today, companies need information about the work people do and the time they take to do it -activity information. Achieving global competitiveness, identified with total customer satisfaction, means reducing activities that actually constrain the company's ability to be responsive to customer needs. To accomplish that goal, companies must really understand what activities are taking place in their organizations and why. As the experience of these companies shows, it's easy to get the cart before the horse in using driver-oriented ABC analysis to manage operating costs. To improve operations, companies should assess activities first

-

in t e n s ^ofsatishing customer wants

-

and then cost products, customers, channels, etc. Companies need to do both, but in the right order.

Once a company's activities have been mapped, management has a much better data base for managing activities and for making strategic product mix decisions. For example, if the auto components manufacturer had used activity analysis to question its processes it might well have discovered how to produce products in small-batches. It still would have information to properly cost products for strategic marketing decisions, but now that information would not interfere with and impede efforts to make competitive operational decisions. Similarly, activity analysis might have led the computer manufacturer to assess the tradeoff between automatic and manual insertion of circuit boards in terms of customer satisfaction, not cost.

A comprehensive cross-functional activity cost analysis really focuses on changing activities, not just managing drivers. All the ABC analysis in the world isn't going to help in the long run if companies mindlessly seek scale economies with high utilization rates and high output. Indeed, with that mindset at work, managers looking at the costs of "drivers"

will tend to economize on drivers by using them sparingly in hopes of cutting costs as such.

The results often diverge greatly from what customers find satisfactory. On the other hand, if a company's mindset is to satisfy customers by improving flexibility, then it will take steps to achieve small lots, short leadtimes, and defect-free output. It follows that

(18)

those steps probably will lead to faster changeover, less work, and - incidentally - lower costs.

CONCLUSIONS

Activity management tools

-

activity analysis and activity costing -undo many of the flaws in financial management information that have beset companies in the past 40 years. They go far toward providing business both the strategic costing information and the operational control information it needs to be competitive and profitable in the global economy. Driver-based product cost information - ABC

-

overcomes distortions inherent in traditional financial cost accounting information. However, ABC information does not necessarily help companies achieve continuous improvement of globally-competitive operations. Cross-functional activity cost analysis (CFACA), on the other hand, provides more of both types of information lacking in traditional management accounting

-

strategic product costs and relevant operational control targets

-

in a form that does not necessarily jeopardize a company's efforts to become a world-class competitor.

Much historical research remains to be done before we have a definitive understanding of how companies' efforts to cope with management accounting's lost relevance after the 1950s led to the development of activity costing and activity analysis tools. Companies that implemented pioneering ABC product costing systems in the early 1980s were apparently unaware of cross-functional approaches to activity management that GE and others had used since the 1970s. Similarly, the modes of activity analysis pioneered by GE were seldom used, until recently, to compile more reliable product cost information.

Existing activity management tools still do not complete the task of restoring management accounting's lost relevance. Much remains to be done. Only recently have scholars and consultants begun to explore the relationship between activity analysis and activity costing.34 Driver-based ABC tools - especially those pioneered in companies such as John Deere and Kanthal and codified in the writings of Robin

-

restore relevance to product cost information and thereby help companies avoid costly marketing blunders. However, I do not believe better product cost information itself could have prevented most American manufacturers from losing long-term market share and profitability in the 1970s and 1980s. It might have been enough if their competitors, especially from Japan, had not changed fundamental assumptions about how to organize work to satisfy customers. That is the point

-

how to discover and adopt globally-competitive ways of organizing work, not how to be more profitable by shifting product mix in companies that continue to follow traditional mass-production practices.

(19)

ABC simply gives a better "rack and stack" of costs, it does not drive companies to change their fundamental mindset about how to organize work to efficiently satisfy customers.

The same criticism can be made of existing modes of activity analysis - they do not magically reveal how to organize work to profitably satisfy global terms of competitiveness.

However, they do provide vital information about activities and work that managers need to continuously improve operations once they understand and adopt the mindset of global competitiveness. Moreover, focusing on activity information, rather than traditional or ABC cost information, should make it easier for companies to make the transit from the old to the new management mindset. That is now the most important challenge - to shift companies from a management mindset focused on products, resources, and costs to one focused on people, time, and customers. Turning once more to the metaphor of Plato's cave, the challenge is to lead managers away from financial management in the cave and toward quality-driven activity management in the outside world of associates and customers.

Global competitors that move away from managing numbers in the cave toward nurturing the talents and fulfilling the desires of people above the ground will still give consideration to accounting and financial shadows. Indeed, they will develop and use variance tools, activity-based costing, and other tools to analyze financial and accounting information. Companies will of course compile such information for public reporting, for

"what if" analyses, and for budgeting. Management accounting information has filled those purposes for nearly two centuries of industrial history, and it will continue to do so for many years to come. However, companies that come out of the cave will no longer use accounting and financial information to dictate people's activities.

FOOTNOTES

his

section, with minor changes, is from H. Thomas Johnson, "Managing By Remote Control: Recent Management Accounting Practice in Historical Perspective," in Inside the Business Enterprise: The Use and Transformation of Information, ed. by Peter Temin (Chicago: University of Chicago Press, forthcoming).

' ~ a n i e l M.G. Raff and Peter Temin articulate these three levels of decisions in "Business History and Recent Economic Theory: Imperfect Information, Incentives, and the Internal Organization," in Inside the Business Enterprise, ed. by Peter Temin.

his

definition is implicit in virtually all management accounting textbooks written since t h e 1950s.

4 ~ . Thomas Johnson, "The Decline of Cost Management: A Reinterpretation of 20th-Century Cost Accounting History," Journal of Cost Management (Spring 1987), 5-12.

%.

Thomas Johnson and Rabert S. Kaplan, Relevance Lost: The Rise and Fall of Management Accounting (Boston: Harvard Business School Press, 1987).

(20)

'Robert H. Hayes and William J . Abernathy, "Managing Our Way to Economic Decline," Haruard Business Review (July-August 1980), 67-77.

7 ~ . Thomas Johnson, "Activity-Based Information: A Blueprint for World-Class Management Accounting," Management Accounting (June 1988), 23-30.

he

following section is adapted from H. Thomas Johnson, "Beyond Product Costing: A Challenge to Cost Management's Conventional Wisdom," Journal of Cost Management (Fall 1990), 17-19.

he

Republic of Plato, translated by Francis M. Cornford (NY: Oxford Univ. Press, 1945), pp. 227-229.

lopeter Drucker foreshadowed this problem in "Managing for Businesl Effectiveness," Haruard Business Review (May-June 1963), 59-62.

his

divorce is the central theme in Hayes and Abernathy, "Managing Our Way to Economic Decline."

I2Johnson and Kaplan, Relevance Lost, Ch. 6.

13~ohnson and Kaplan, Relevance Lost, Cb. 8; Robin Cooper and Robert S. Kaplan, "Measure Costs Right: Make the Right Decisions," Hnruard Business Review (SeptlOct. 1988), 96-103.

1 4 ~ h e distinction between unit, batch, and product--level cost drivers, articulated originally by Robin Cooper of Harvard Business School, is the conceptual foundation that supports modern activity-based product costing. The definitive statement of this distinction is found in Robin Cooper, "Cost Classification in Unit-Based and Activity-Based Manufacturing Cost Systems," Journal of Cost Manngemenl (Fall 1990), 4-14.

1 5 ~ o r a detailed example, see Robin Cooper, "Schrader Bellows," Harvard Business School Case, numbers 6-386-050 et al. (1985).

16Robin Cooper, "The Two-Stage Procedure in Cost Accounting Part One," Journal of Cosl Managemenl (Summer 1987), 43-51; Johnson and Kaplan, Relevance Lost, Ch. 10.

" I ~ r i c Noreen analyzes the accuracy of ABC product cost information in, "Are Costs Strictly Proportional t o Their Cost Drivers?" Unpublished working paper (October 1990).

18Steven Player provided this phrase.

"T. C. Davis, "How the DuPont Organization Appraises Its Performance,"AMA Financial Management Series No. 94 (American Management Association: New York, 1950).

2 0 ~ . Shank and N. Churchill, "Variance Analysis: A Management-Oriented Approach," Accounting Review (October 1977), 950.57.

" ~ l f r e d Rappaport, Creating Shareholder Value (New York: The Free Press, 1986).

2 2 ~ l b e r t Lee, Call Me Roger, (Chicago: Contemporary Books, 1988), 90.

2 3 ~ . Thomas Johnson, Thomas P. Vance, and R. Steven Player, "Pitfalls in Using ABC Cost-Driver Information t o Manage Operating Costs," Corporate Controller (January/February 1991), 26-32.

24From H. Thomas Johnson, "Managing by Remote Control: Recent Management Accounting Practice in Historical Perspective," in Inside the Business Enterprise: The Use and Transformation of Information, ed. by Peter Temin (Chicago: University of Chicago Press, forthcoming).

2 5 ~ r o m H. Thomas Johnson, Gail J . Fults, and Paul M. Jackson, "Activity Management and Performance Measurement in a Service Organization," in Performance Ezcellence in Manufacturing and Service Organizations (Sarasota, FL: American Accounting Association, 1989), 7-20.

2 6 ~ r o m Johnson, Vance, and Player, "Pitfalls in Using ABC Cost-Driver Information to Manage Operating Cost."

(21)

2 7 ~ h e Gordion Knot analogy is discussed in J. Robb Diion, Alfred J. Nanni, and Thomas E. Vollmann, The New Peformance Challenge: Measuring Operafions for World-Class Competition (Homewood, IL:

Dow Jones-Irwin, 199O), 21-22 and 32-36.

28~bid, 22.

2 g ~ i c h a r d J . Schonberger, Building a Chain of Customers (New York: The Free Press, 1990), 187.

3 0 ~ o b e r t W. Hall, "Measuring Progress: Manufacturing Essential," Target (Summer 1987), 7.

3 1 ~ h e evils of standard cost performance measurement systems are recounted both in H. Thomas Johnson, "Performance Measurement for Competitive Excellence," Measures for Monufacfuring Ezcellence, Robert S. Kaplan, ed., (Boston: Harvard Business School Press, 1990), 63-90 and in Robert S. Kaplan, "Accounting Lag: The Obsolescence of Cost Accounting Systems," The Uneasy Alliance:

Managing the Producfiuity-Technology Dilemma, Kim B. Clark, Robert H. Hayes and Christopher Lorenz, ed., (Boston: Harvard Business School Press, 1985), 195-226.

3 2 ~ i l ~ i a m T. Turk, "Management Accounting Revitalized: The Harley-Davidson Experience," Journal of Cod Management (Winter 1990), 28-39.

3 3 ~ o t h examples and paragraphs that follow are from Johnson, Vance, and Player, "Pitfalls in Using ABC Cost-Driver Information to Manage Operating Costs."

3 4 ~ h i s relationship is explored in recent work, by James Brimson, C. U. McNair, and others

3 5 ~ . March and Robert S. Kaplan, "John Deere Components Work," Harvard Businem School Case, Number 91-87-107 (1987); Robert S. Kaplan, "Kanthanl," Harvard Business School Case, Number 9-190-002 (1989). Robin Cooper. Journal of Cost Management. four Dart series on "The Rise of

~ctivity-Ba&d &ting," Part o n e (Summer i988), Part '&o (Fall 1988j, Part Three (Winter 1989), Part Four ( S ~ r i n n 1989) and "Cost Classification in Unit-Based and Activity-Based Manufacturing Cost

I

1991 Edition Engineering Economy Absttaeta

The annual edition of Engineering Economy Abstracb contains abstracts of the relevant articles screened from all precedin year issues of about 35 periodicals.

The 1991 issue o t ~ n g i n e e r i n ~ Economy Abstracts is available after May 1, 1991. To obtain the 1991 issue or issues alter 1987 send your request to Hamid Parsaei, lndustrial Engineering, University of

Louisville, Louisville, KY 40292. (Make a $2.00 check payable to: Department of Industrial Engineering, University of Louisville.)

1981 and earlier issues of Engineering Economy

i

Abstracts continue to be available from Professor G.

W. Smith, lndustrial Engineerin lowa State

University, Ames, lowa 50011. @or epeh issue rnake a

$2.00 check payable to: ISU Englneerlng Extension.)

-

ECONOMIST

is AmilaMe in

I ^.LC.'# ^.",.I^I^.^. COI.,", YIO."..~~" I I