Count Your Costs

ver the last few years the General Services Administration has been squeezed hard-the agency's budget and staff have each shrunk by almost 30 percent since 1993. In the face of that kind of pressure, top managers had a pretty logical epiphany: They realized they had better start coming up with some new and creative ways to do what they do more efficiently.

In seeking to do that, though, GSA discovered it didn't actually know what it cost to provide any of its services on a transaction-by-transaction basis.

Yes, GSA could gin up general figures about its annual budget and expenditures. It could produce reams of data on what it accomplished in any given year. What it couldn't illustrate in any meaningful detail was what it cost to do very specific things like lease space or buy property. The agency didn't know, in other words, whether it was-or could be-competitive with the private sector real estate office down the block; or whether it was providing real value to its immediate customer: the rest of the federal government.

But even if GSA guessed-or roughly calculated-that it could be more efficient and competitive, the agency still had a problem: It could not break down the specific expenses that contributed to the overall cost of doing something like finding and renting office space for another agency. How much of the cost was for personnel? How much was administrative overhead? How much could be apportioned to travel or office equipment? How much was just redundant paper pushing? Without specific information, any effort to become more efficient would be flying blind. And without such information, GSA was hard-pressed to argue that it could become competitive.

So GSA decided to join the handful of federal agencies who are pioneering a new style of federal accounting. The agency signed on to pilot several experiments in what is known as activity-based costing, or ABC.

ABC is the practice of focusing on some unit of output-whether it's printing a dollar bill or maintaining peace in the Middle East-and then trying to figure out to as precisely as possible what contributes to the cost of that output, from manhours, to materials and equipment, to administrative overhead and beyond. With such a breakdown in hand, the theory of ABC goes, managers can hone in on key cost drivers that are ripe for reduction.

While it sounds simple, ABC can be a very complicated proposition, requiring a whole new approach to how managers and accountants view and evaluate what an organization does and how it does it. With its roots in private-sector manufacturing, ABC is clearly adaptable to a wide variety of governmental production processes. If the output is relatively straightforward-like printing a dollar bill-then the process of allocating costs to that output may also be relatively straightforward. But given the myriad and frequently complex tasks in which the federal government is engaged, implementing ABC usually takes a lot of work, some best-guess creativity and, at least in the beginning, a leap of faith.

Those who've tried it, though, have found it can be an illuminating exercise, offering managers the equivalent of sandpaper and a jigsaw in making changes in process and organization, when all that was available in the past was a sledge hammer and a hand saw.

"It is a way to understand the true cost of operating," says C. Morgan Kinghorn, who helped introduce ABC at the IRS several years ago as the agency's chief financial officer; he now works for Coopers & Lybrand selling ABC to other branches of the federal government. "It allows you to break out the real cost components of the products or services you deliver."

Beyond the Dancing Tax Dollar

That is a far cry from most accounting systems today in the federal government. Those systems are great at tracking big numbers-like how much money was received, how much disbursed and in what broad categories-but they offer only the roughest idea of the real cost of doing certain types of business.

What departments or programs know about cost these days is blunt stuff: how much money is coming in; what their personnel costs are; and what they spend on rent, utilities, equipment and materials. But when it comes to explaining what portion of each of those goes into producing a single unit of work-processing a social security check or protecting an endangered species, for example-they are pretty much at a loss. It is virtually impossible, argue ABC proponents, for managers to divine where or how processes might be made more efficient without knowing in detail what's really driving the cost of doing specific things.

To succeed at activity-based costing, government has to do more than simply follow the dancing tax dollar, which it does well now. Government has to identify discrete outputs and then take labor costs (including fringe benefits), rent, equipment, materials, administrative overhead (including supervisors' salaries) and so forth, and apportion them to those outputs. It is no surprise that the task is widely viewed as arbitrary, capricious and daunting.

Yet as hard-or arbitrary-as it might seem, a science of activity-based costing is emerging. "Most of the information you need to do this exists in government records right now," contends Ellen Doree Rosen, who came up with an equation for doing ABC while a professor of public administration at the John Jay College of Criminal Justice in New York.

Under Rosen's formula, such factors as the cost of labor, equipment, rent, energy, supervisory salaries and other overhead are accounted for. They are added together to arrive at an overall number that is then divided by units of the activity (outputs). The activity can be anything-processing a tax return, handling a labor grievance or accomplishing an F-111 bombing run.

Rosen has even taken her formula a step further. Obviously, calculating the cost of processing a tax return is just one dimension of a two-dimensional exercise. The other dimension is quality-asking, for example, if the form was processed quickly and accurately. Rosen offers a variety of ways to gauge quality in the overall ABC equation. The resulting number is an indicator of both efficiency and effectiveness.

A lot of basic work has been done on implementing ABC in government. Consultants like Kinghorn are considered old hands at it. And some off-the-shelf computer programs can be adapted to the tactic.

But that doesn't mean ABC is easy, or that everybody in government should try it. "It's not the answer to everything," says Doris Chew, acting executive director of the Joint Financial Management Improvement Program.

But for those who have decided ABC is the best answer right now for calculating the costs of doing business, most contend that the work involved in applying it is worth it. "It was eye-opening in a lot of areas," says Dave Rebich, who heads up the ABC effort at the Treasury Department's Financial Management Service.

As FMS began to look at some contributors to the cost of its outputs and products, he says there were some big surprises. "Just looking at what it cost to do time cards amazed us; it is a lot of money. Realistically we can't do away with our current time card system immediately, but we can certainly look at other systems for handling them, and calculate whether or not they'll save us money down the road."

Missions and Costs

The ability to do detailed analyses-such as what the cost of handling and processing time cards contributes to the overall cost of a product-is the purest and least common reason more federal managers these days are learning about ABC.

The more common reason managers get into ABC is that they have to. With passage of the 1990 Chief Financial Officers Act and the 1996 Federal Financial Management Improvement Act, agencies have to begin calculating-and where appropriate, recovering-the real cost of services provided to customers inside or outside government.

The Immigration and Naturalization Service, for example, is using ABC to recalculate its fees for everything from administering citizenship exams to issuing green cards. In all, the agency has identified more than three dozen product lines for immigrants and prospective U.S. citizens which need updated fees. Some cost drivers identified by INS include the cost of an FBI fingerprint check and what courts charge per head for swearing in new citizens.

The huge amount of work INS is putting into investigating real costs is paying off in one other way besides fees that more accurately reflect the true cost of service, says Mike Natchuras, who directs the agency's ABC effort. INS can now estimate with much more confidence what fees it should charge for new services by using its cost breakdowns for similar or related services.

While INS might sound like it's ahead of the game, Congress views the agency as a laggard. Agencies should have had congressionally approved cost accounting systems in place last September. That deadline may not have been realistic, but it had value, says Monica Congleton, a cost accountant at the Veterans Affairs Department and co-chair of the implementation subgroup of the Governmentwide Cost Accounting Work Group, a network of government accountants, managers and CFOs working to carry out the congressional mandate. "If it weren't mandated I don't think it would be happening. I'd be hard-pressed to think of anybody who initiated ABC a long time ago."

Several executive branch organizations are helping push the cost accounting effort. The Joint Financial Management Improvement Program, supported jointly by Treasury, OMB and the General Accounting Office, is developing training materials for officials interested in implementing ABC, as well as other cost accounting systems. In June 1995, the Federal Accounting Standards Advisory Board released a statement entitled "Managerial Cost Accounting Concepts and Standards for the Federal Government."

Another major piece of legislation is indirectly driving ABC: the 1993 Government Performance and Results Act. As most federal managers are now probably painfully aware, GPRA requires agencies to identify core missions and develop performance measures that will allow them to gauge progress on those missions.

While that is a laudable goal, it is only half the performance equation, argue ABC proponents. The other half is to calculate whether resources are being maximized in achieving the outcomes. Even the most sloppily run department or bureau can achieve great things given unlimited resources. The key-especially in times of dwindling resources-is to perform efficiently. And that is where the analytical power of ABC comes in, argue proponents.

"I've come up with a golf analogy," says OMB controller G. Edward Deseve. "Identifying your overall mission is like driving. Then as you refine the whole process you work your way through your irons until you get down to putting. ABC is putting."

Other factors driving the incipient push to do ABC include the General Accounting Office's list of federal programs at high risk of financial impropriety, spurring those on it to investigate how to more solidly connect budgets and achievements, says Al Hyde, an expert in federal budgeting and a senior staff member at the Brooking Institution's Center for Public Policy Education.

Hard Sell

Despite the growing number of cases illustrating the potential of ABC, it is still a hard sell in government (state and local governments have only just begun to embrace the practice, and only in what are considered management bellwethers like Phoenix and Indianapolis). The usual suspects-bureaucratic inertia, turf protection, fear that the numbers will be used to justify cuts, and plain laziness-are all fingered as chief culprits. But another big problem is that many in government view ABC as a neurotic reemergence of such management trends as planning-programming budgeting systems, management by objectives, total quality management, or any of a host of other "flavor of the month" management trends that have swept over the federal government with frequently unhappy effects.

It makes sense that organizations trying ABC are, like GSA, frequently doing it for enlightened self-interest as much as to satisfy any directive.

The key to successfully implementing ABC in any organization is not magic, says Howard Katz, the point man for ABC at GSA. It takes at least a modicum of support from an organization's chief financial officer, and GSA's CFO, Dennis Fischer, happens to be an enthusiastic supporter. But once a core of top program and budget officials committed themselves to investigating the practice at GSA, it boiled down to plain old hard work. It was a matter of educating those who would be asked to implement ABC, says Katz, and then asking for volunteers. The initial effort involved launching a training program which included a two-day workshop on activity-based costing for 1,000 mid-level managers and half-day workshops for members of the Senior Executive Service. "Our approach," says Katz, "was not to try to sell it to anybody, but to work with people who after learning about it wanted to try it."

GSA's Property Acquisition and Realty Service (PARS) business line-GSA's real estate arm-stepped forward first. The organization's interest was inspired in no small part because it was the section of GSA most under the privatization gun.

While PARS is just beginning a full cost accounting of its various products and services, it has already identified ways to do its job cheaper, faster and smarter. For example, GSA develops space requirements with clients jointly. Yet it still sends the final request back to its customer agency for final approval. "During the whole process of analyzing costs, we realized that didn't make any sense, since we'd worked all that out jointly already," says Julie Wilson, who is handling the ABC pilot for PARS.

The larger question of whether ABC is just another flash-in-the-pan management trend won't be answered for years. Generally, those who've tried it, like it. But already there are cautionary tales to tell. The pioneering ABC effort at the IRS of a few years ago is being buried in the debris of a string of management disasters, in particular the debacle surrounding its multibillion-dollar computer overhaul, along with the resignations of some key ABC advocates.

Interest in ABC is increasing. Much of what the federal government does lends itself to activity-based costing. One day the federal government may even identify all the costs associated with maintaining world peace-from travel to pensions and postage.

What you don't know about your costs can put you at clear risk; what you do know may be the key to your survival.

Jonathan Walters is a senior staff writer for Governing magazine. He has covered state and local public policy for the last 15 years. He is working on a book on performance measurement in state, local and federal government.