A NEW TOOL FOR MANAGING COSTS The claims for ABC -- activity-based costing -- might suit a patent medicine. Trim waste! Improve service! Increase productivity! But it does do all that -- and more.
By Terence P. Pare REPORTER ASSOCIATE Tricia Welsh

(FORTUNE Magazine) – JAMES DONLON, Chrysler's controller, is thrilled by it. James Flynn, general manager of the coatings division at Union Carbide, thinks it is one of the most important innovations in his 27-year career. Ken Schrader, a product manager at Hewlett-Packard, says that from a financial point of view, it is heaven. It is not a cure for the common cold or the Holy Grail. These executives and many more like them are raving about a new kind of -- accounting. During the 1980s many companies woke up to the fact that their antiquated accounting systems were leading them to charge too much or too little for their products. As managers raced to modernize their bookkeeping, they discovered an innovative system that helped them estimate manufacturing costs more accurately and manage their companies better. Says William M. Herman, manager for service finance at GE Medical Systems: ''The new system helped us increase productivity 8% a year two years running.'' The last word in accounting goes by the first three letters in the alphabet -- ABC, for activity-based costing. Companies that have gotten the ABC religion are reaping earthly rewards in better product design and factory layouts. They are using the new math to trim waste, improve service, evaluate quality initiatives, and push for continuous improvement. The traditional accounting system looks at businesses the way Charlie Chaplin did in Modern Times. The enterprise is a huge machine made of components, or ''departments'' such as human resources, purchasing, or maintenance. They operate together, with employees as cogs, to make products or provide a service. ABC, by contrast, sees a business as a company of individuals performing all manner of activities -- training employees, processing purchase orders, fixing machines -- to satisfy customer demands. The point of view is more like, say, Howard Hawks's in Red River. Employees are cowhands who share common tasks in the arduous cattle drive to market. To see ABC in action, consider Dana, the auto-parts maker in Toledo. Two years ago it installed an activity-based accounting system at its plant in Plymouth, Minnesota. First, during a six-month stretch, employees in all areas of the plant broke down what they did each day in order to define their basic activities. For example, in the material-control department, which buys components for the factory and fills customer orders, a customer service rep spends about 30% of his time taking orders and 50% actually processing them.

Based on that information, the number crunchers took the expenses tallied by traditional bean counting -- fixed costs, supplies, salaries, fringe benefits -- and spread them over the activities. Result: a new ledger that reported the cost of performing specific tasks involved in turning out the product (see chart). Both the traditional and the ABC methods reach the same bottom line. But ABC allows accountants to charge costs to products much more accurately than the conventional method because it breaks down overhead far more precisely than old-fashioned systems do. Traditional accounting focuses on valuing a company's inventory for financial reporting purposes. To satisfy the Internal Revenue Service and shareholders, accountants need hit only the target, not the bull's-eye, for each product. So they simply take the cost of materials and add in direct labor. Then they compute overhead from rent to R&D expenses, based on the number of direct labor hours it takes to make a product. To get a unit cost, they divide the total by the number of items made during the period under consideration. This system works fine when direct labor accounts for most of total costs and a company produces just a few products requiring the same processes. Most outfits fit that description in the 1920s, when cost accounting was first devised. Allocating costs at Henry Ford's River Rouge auto plant, for example, came as naturally as making whoopee -- and nailed costs down to the penny. Raw materials such as iron ore went in one end of the plant, and Model T's emerged from the other 28 hours later. Today most companies run far more complex shops than the Rouge plant, which manufactured one kind of vehicle in any color, as long as it was black. General Motors' menu of cars and options is so extensive that, according to management consultants at Bain & Co., a plant producing a car a minute would take several trillion years to produce one of every variation. Long menus produce big overhead. When a company makes an assortment of products, it must finance and maintain large inventories. Production lines stop and then have to be set up to turn out all the styles, which means costs for downtime and reprogramming the computers that control the machinery. You are soon over your head in overhead. Says ABC expert Earl Landesman, a principal with Chicago consultants A.T. Kearney: ''Once labor was the major cost component in the manufacture of products. Nowadays overhead may be as much as 70%.'' Business has been steadily replacing people with machines, and in some high- tech companies direct labor may account for as little as 5% of cost. Faced with such a huge structural change, even the most energetic accountant using the traditional methods can do little more than estimate and hope for the best. Says John Hoffecker, a cost-accounting specialist at KPMG Peat Marwick in Chicago: ''It's better to have no cost system than a standard cost system because it is so inaccurate.'' Imagine a production line in a pen factory where two kinds of pens are being made, black in high volume and purple in low. Assume that it takes eight hours to program the machinery to shift production from one kind of pen to the other. The total costs include supplies, which are the same for both types of pens, the direct labor of the workers on the line, and factory overhead. The most significant piece of overhead in this instance would be the cost of reprogramming the machinery whenever there's a switch from black to purple or back to black. If the company produces ten times as many black pens as purple pens, ten times the cost of reprogramming will be allocated to the black as to the purple. Obviously this undershoots the cost of producing the low-volume product. The virtue of the ABC approach is that it comes much closer than standard costing to showing managers where the expenses of manufacture are embedded. Once employees break down pen manufacturing into its activities, managers can recognize that the activity of changing pens triggers the cost of retooling. The ABC accountant then calculates an average cost of setting up the machinery and charges it against each batch of pens that requires retooling regardless of the size of the run. Thus, a product carries the freight for only the overhead it actually consumes. Advanced Micro Devices, the big semiconductor maker, installed ABC accounting at its Penang plant in Malaysia and realized that the plant was overestimating the cost of making high-volume chips and underestimating on the low-volume variety. Based on old-fashioned accounting, AMD thought it cost no more than $2 to assemble and test its most expensive products and that 60% of the company's goods came in at just 25 cents apiece. Not so, as ABC demonstrated. The more complicated, low-volume chips, which are more difficult to assemble and require more testing, were far more pricey to produce. ABC showed that 54% of the company's products cost at least 75 cents to make, and some chips were as much as $3.50. The company quickly began altering its product mix and adjusting prices. The ability to make better product-mix and pricing decisions is just ABC's low-hanging fruit. In fact, if companies focus too narrowly on costs, they can miss far greater benefits down the road. Says Harvard accounting professor Robert S. Kaplan, widely credited as one of the creators of ABC: ''If I could ) play back the tape, I would take out the word 'cost.' ABC is really power for strategic management.'' It turns out that the immense amount of information collected when creating an ABC database also reveals a lot about how the business processes operate. Smart managers can use this knowledge to make their processes faster, higher quality, and more efficient. Kaplan suggests that salespeople could use an ABC analysis to show customers that their high-priced product actually costs less than the competition's because fewer units have to be returned or because the firm can deliver orders more reliably. Thus, the gang in the green eyeshades can metamorphose into management information sources vital to the competitive success of the company. Financial people see the value that they can add. And, says Flynn of Union Carbide, ''the operating guys like me view ABC as a way to get better information about the cost of doing business.'' Chrysler is ''dynamiting the old financial system,'' as the company's president, Robert Lutz, puts it, and is replacing it with ABC. The fuse was lit last year by a pilot project that examined the designs for wiring harnesses for Chrysler's popular minivans. The harnesses, as the name implies, yoke together bundles of wires like those you see under the dashboard just before you bang your head searching for the quarter you dropped as you entered the exact-change lane at the tollbooth. Nine departments, from design to assembly to finance, set out to reckon the optimum number of wiring harnesses for the new line of minivans. The assembly people favored using just one kind of harness, the design group wanted nine, and so on. When ABC was used to cost out activities across the entire production of the vehicles, everyone saw the optimum number was two. Says controller James Donlon: ''We eliminated a lot of the normal bickering that takes place when everyone fights for the answer that is best for their individual department's scorecard.'' Hewlett-Packard has used various forms of ABC since the mid-1980s. In H-P's fast-changing world of high tech, the life cycle of a new product can be briefer than young love. The ABC data, which tally the cost of all the activities required in manufacturing and distribution, quickly found its way into the desktop computers of H-P's designers, enabling them to run sophisticated cost estimates of their ideas even as they hatched. The company's hit products, new models of the HP 3000 and HP 9000 midrange computers, were created by designers who had real-time cost information at their fingertips. When ABC showed that testing new designs and parts was extremely expensive, engineers changed their plans on the spot to favor components that required less testing, thus lowering costs. Says Ronald Foster, H-P's controller of manufacturing for computer systems: ''They used to hand the design over to accountants, who needed a couple of days to estimate the production cost. Then in some cases the designers would be told what they wanted was too expensive.'' H-P is not stopping there. Right now Foster is developing ABC techniques to assess the risk of innovation in a market with a short product life. The company usually spreads the excess costs of design innovation over all its products. But Foster hopes to trace them back to the individual design. ''We don't want to discourage innovation,'' he says, ''but we want to get a better idea of whether a particular game is worth the candle. Maybe it's better to let somebody else make that iteration in design.'' Or maybe H-P can make money by scaling back the innovation, using more readily available parts, and getting to market faster. At General Electric Medical Systems, which manufactures and services huge imaging machines for hospitals and clinics around the world, William Herman finds ABC especially helpful when deciding to fund projects to increase the company's productivity and customer satisfaction. Come budgeting time in the days before ABC, employees besieged the finance department with bright ideas about improving the business. Says Herman: ''A guy would come in and say, 'Give me a million dollars, and I'll save the company two million.' If he was convincing, he got the money, but we had no way of knowing whether that million would be well spent. Now we do.'' A CASE IN POINT: The field engineers who maintain the imaging machines were lugging around on each service call a trunkful of manuals that weighed about 200 pounds. While on the job, an engineer might have to make several trips to his car to check procedures or get information to help diagnose problems. If he hadn't gotten a chance to update the manuals, which he was supposed to do periodically, he might have to call the office for information. All in all, the technicians estimated that they spent as much as 15% of their time during a service call shuttling back and forth to their cars. The remedy, as the folks in the field saw it, was to issue laptop computers with CD-ROM readers that would allow them to carry all their reference materials right to the job site. But with 2,500 field engineers in the U.S. alone, switching to laptops was a major capital expenditure. Without compelling evidence that footing the bill for laptops was cost-effective, Jack Welch's managers were reluctant to part with the bucks. Once they completed an activity analysis of the service operations, the decision was easy. Switching to laptops helped GE Medical raise productivity 9% -- the equivalent of a $25 million increase in sales with no increase in cost. One unexpected benefit: The field engineers' cars are 200 pounds lighter, which improves gas mileage. Herman says that eventually he hopes the laptops will help technicians channel service information back to the manufacturing side of the business so that design improvements can be done with an eye to the cost of servicing. So now accountants will be helping design million-dollar pieces of medical equipment. Hold the legumes! You can't call them bean counters anymore.