An Experiment in Federal Cost Accounting and Performance Measurement
Geiger, Dale R., The Government Accountants Journal
Recent events place management accounting at the intersection of a number of emerging agenda at the world's largest and most complex organization. The 1992 presidential election focused attention on governmental management. The National Performance Review is pursuing management improvement, and Congress is debating performance based budgeting. Federal agencies are reviewing management practices, systems and accounting in response to the Chief Financial Officer Act (H.R. 5687) mandate for "fundamental reform in financial management requirements and practices."
A role of increased importance seems clear since management accounting supports the information needs of those "engaged in managing the undertaking (Crowningshield, 1969)." Many accounting academics, however, feel that management/cost accounting has little to offer public sector organizations unconcerned with profit making.
This research investigates the usefulness of cost accounting in government by installing an activity based cost system at a federal organization operating without cost data, but with strong non-financial measurement systems. The study documents three pre-experiment management practices, describes the informational intervention and reports changes in management actions and perspectives. The results suggest that cost information provides management with significant new insights into cost reduction, resource deployment, process improvement and line-to-staff interfaces.
THEORY AND BACKGROUND
Explanations for the current lack of governmental management accounting emphasize the differences between public and private sector due to the government's service nature and lack of profit motivation (Schattke and Jensen, 1981 and Anthony and Herzlinger, 1975). Horngren and Sundem (1990) argue that the lack of clear organization goals, the presence of professional employees and the difficulty in measurement of accomplishments create an inhospitable environment. They conclude that "control systems in nonprofit organizations will never be as highly developed as in profit seeking organizations (page 289)." They recommend the study of management accounting to nonprofit managers as an aid in their personal investments and interactions with businesses!
Yet, many private sector organizations do use management accounting and the value of cost computation, containment and control is increasingly apparent in the public sector. Birnberg and Gandhi (1976) and Helmi (1987) cite the important contribution of cost accounting in the administration of social programs. The value of accurate cost information has been recognized in privatization decisions (London, 1990), program control and performance improvement (Nestor, 1979), planning, programming and decision making (Caldwell and Welch, 1989) and fee setting (Dierks, 1978).
Why do organizations use management accounting? Accounting theorists cite the advantages of dollar based metrics in explaining the existence of accounting systems. Sterling, (1961) recognizing accounting's role in enabling comparison, notes the greater precision of numbers, and the additive nature of monetary values. Johnson (1991) also notes the ability of internal accounting systems to simulate external market forces thereby capturing market efficiencies within the organization.
Simon, et al, (1954) found monetary evaluation useful in promoting increased cost consciousness and promoting continuous improvement. This study extends this literature by comparing "before" and "after" observations of a single, public sector organization's management decision making after the introduction of a cost accounting system.
SITE SELECTION AND DESCRIPTION
The Examinations Division of the Internal Revenue Service (IRS) Boston District was selected for this research. The IRS met criteria for having a typical funding environment and well developed non-financial information systems. Non-typical funding environments, such as revolving funding, seem to have greater incentives for cost system needs (Geiger, 1993). Existence of non-financial information increases confidence that changes resulting from the experiment were due to the incremental value of financial information to management.
Collecting the nation's tax is big business requiring roughly 120,000 employees. Only 70 or so corporations on the Fortune 500 list post annual sales greater than the IRS's spending level of roughly seven billion dollars per year.
The lowest level of financial control at IRS is the regional, or second, level (see Figure 1).(Figure 1 omitted) District Managers, who report to Regional Commissioners, receive dollar budgets for only about 15 percent of their total cost. Payroll and benefits, by far the greatest cost, are managed only indirectly at district level through a budget of staff years. Divisions, the organizations that report to districts, receive even less financial input and are held responsible for only travel dollars and staff years. The lowest level organizations, branches and groups, receive no financial goals and are not held accountable for cost even though these organizations spend roughly five and one million dollars per year, respectively.
Lack of financial measurement, however, does not imply lack of measurement. The IRS uses extensive non-dollar denominated measurements for performance evaluation and management control. Tax return data is segmented by type of return and by filer's asset value or income level to generate many operating statistics. Formal annual management by objective goals are established for each manager using this operational performance data. For example, the Boston Examinations Division Manager's goal for 1991 included the following agreed upon commitment known as a "critical success factor."
"After considering plus and minus three standard deviations and excluding training returns, achieved June 30, 1991, accomplishment rates are as follows:
A. 23 hours for TPI 034 returns; B. 39 hours for TPI 037 returns; C. 59 hours for Activity Code 213 returns; D. 78 hours for Activity Code 215 returns; E. 126 hours for Activity Code 219 returns."
The majority of the division's 700 people work in audit groups reporting to one of six audit branches (see Figure 2).(Figure 2 omitted) Each group, the lowest unit on the organization chart, typically contains 10 to 12 auditors, a secretary and a manager.
Branch 1 handles Coordinated Examinations Program (CEP) audits. Branch 1 audits the 48 "large corporate" returns from companies headquartered in Massachusetts. Branch 2's responsibilities include all "other" audit programs such as excise, employment, inheritance and gift taxes. Branch 4 handles the "simple individual" examinations known as office audits since taxpayers must travel to the IRS office.
Branches 3, 5 and 6 split the remaining "general" examination programs of the state geographically. These audits review "corporate" and "complex individual" returns. The audits are conducted at field sites and include most Massachusetts based corporations, partnerships and sole proprietorships.
This study employs a simple quasi-experimental design to test the null hypothesis that management accounting is not useful in public sector decision making. The "one group pretest-posttest" design (Campbell and Stanley, 1963, page 7) takes the form of "O sub 1 X O sub 2 ." "O sub 1 " represents the initial state of the organization or variable under study. "X" denotes the introduction of an experimental treatment and "O sub 2 " describes the organization after treatment. To the extent that changes in "O sub 2 " are not caused by other factors, inference can be drawn as to the effect of the treatment "X".
Interviews with over 35 managers documented the initial state, "O sub 1 ," of decision making on three key areas of management decision making: auditor staff assignment, facilities utilization and overhead support levels.
The treatment "X," consisted of introducing a basic responsibility cost system and extending it to provide "product cost measurements through activity based cost system techniques. Subsequent management reaction to the system output provided the basis for describing the resultant state, "O sub 2 ," of the organization.
DATA: PRETEST MANAGEMENT PRACTICE
Analysis of the Boston District's management practice focused on three key cost areas. Salary and benefits represent the largest single cost line item and decisions concerning auditor staff levels, program assignment and geographical disposition are key management responsibilities. The high cost of staff make audit or staffing assignment a key management decision. Facilities constitutes the second largest expense line item and the importance of efficient facilities utilization is heightened by the high rent levels in the Massachusetts area. Overhead support levels represent a final key cost area since it equals 40 percent of the direct costs of audits.
AUDITOR STAFFING ASSIGNMENT:
Pretest assignment of audit staff resources involved several decisions. The examinations division conducts many different types of audits throughout the state of Massachusetts. Besides deploying resources by audit program, decision makers must also determine the geographical disposition of auditors.
A lengthy and complicated procedure known as a workload analysis" determined pre-installation staffing decisions. The workload analysis was initiated by a negotiation with regional officials concerning the number of returns of each type to be examined. Workload is then spread geographically based on "DIF" scores of returns previously processed at the service center. DIF, discriminate indicator function, evaluates the likelihood of noncompliance based on detailed audits from a national sampling known as the Taxpayer Compliance Measurement Program. To determine the geographical distribution of workload, the analysis simply notes zip codes on the tax returns with highest DIF scores.
Historical data for average time per case then converts workload to staff years. The mix of auditor capabilities required by location is calculated from rules and standards relating likely case compleity to auditor job grade.
Field interviews corroborated the importance of workload and budgeted staff years in the staffing equation. A senior regional manager confirmed that districts, divisions, branches and groups received work goals that specified how many of each type of return were to be processed by a given date. He related that workload measures dominated management concern and that "nobody knew and therefore nobody cared what the costs were."
A survey (see Figure 3) of branch and group level managers further verified the lack of responsibility and authority for cost control at the lowest two management levels of the organization. (Figure 3 omitted) The survey asked the importance of cost control to the manager's performance evaluation and the manager's ability to determine spending priorities within already approved budget levels. The responses indicated moderate levels of cost responsibility and little authority for cost management.
The survey also established that branch mangers possess little intuition concerning the cost structure of the examination process. Salary and benefits constitute 85 percent of the examinations division's cost, by far the greatest expense line item in this labor intensive operation. Yet, the branch managers, who on average manage operations that spend more than five million dollars per year, estimated in the survey that salary and benefits comprised only 60 percent of the division's cost of doing business.
FACILITIES UTILIZATION: Facilities cost comprises the division's second largest expenditure. The General Services Administration (GSA) manages federal buildings and charges the IRS rent based on market rates whether the building is federally owned or leased. Facilities needs are based on staff levels and regulation. GSA standards establish a 125 square foot limit to insure equity among federal employees.(1) While technically a maximum, the regulation has become a defacto entitlement replacing a resource allocating management decision.
OVERHEAD SUPPORT LEVELS: Overhead support for audit operations also represents a major resource allocation decision. The division's support resources are divided among several specialized groups and three branch organizations: Planning and Special Programs, Quality Assurance and Exam Support.
The Planning and Special Program Branch functions like a production control department. The Planning Section physically distributes and monitors audit work. Classifiers travel to the processing center in Andover to build the audit backlog based on desired characteristics of grade level of work, type of return and likelihood of assessment The section also maintains an inventory control system to track the flow of returns through the audit process. The Special Program Section coordinates special audit requests such as the Taxpayer Compliance Monitoring Program.
In its primary function the Quality Assurance Branch audits the auditors to determine if approved procedures were followed, the law correctly applied and proper customer service was maintained. A statistical sampling plan suggests cases for review.
The Exam Support and Processing Branch employs 70 people for data input on completed audits, back-end clerical work such as computation of interest charges and taxpayer correspondence. Along with the other support branches the Exam Support and Processing branch provides some specialized services to other, smaller New England states.
The staffing level of overhead activities has responded to workload requirements of the overhead area. Little interaction existed with audit managers to coordinate support levels with audit plans. Staffing needs were based on historical staffing levels and are adjusted by recent overtime requirements.
SUMMARY OF INITIAL STATE, O sub 1 :
Rules, regulations and restrictions preempted management decision making in the Boston District's Examinations Division. Physical processing of workload has dominated resource deployment without regard to cost or benefit. One manager summed up the initial state of management decision making at the IRS. "Nobody worried about the costs. Somebody later would maybe add them up and try to figure out what happened but the operating people just worried about getting the job done."
DATA: INTERVENTION THROUGH INTRODUCTION OF NEW INFORMATION
The experimental treatment, "X," required development of cost information by responsibility, product and location. Activity based cost (ABC) system techniques were used to distribute indirect overhead responsibility costs to audit programs, locations and branch organizations.
ABC system design typically requires a two stage process (Cooper and Kaplan, 1991, page 267). First, responsibility costs are driven to cost centers. Then cost centers with similar activities are aggregated. The IRS, however, possessed little of the needed data. Responsibility costs had to be roughly estimated in some cases with compromises and efforts not usually required in ABC implementation. The second stage of a ABC system distributes cost pools to product or program based on causal relationships. The system developed for this study required a third stage as shown in Figure 4.(Figure 4 omitted) The added stage was needed to associate costs geographically to the district's 12 offices.
FIRST ABC STAGE COMPILES COST ELEMENTS BY ACTIVITY: While the examinations division possessed virtually no cost information, the District Financial Plan provided a starting point for the first stage. The complete plan included staff years by division, travel budget by division, vehicle expense for the criminal investigations division and District wide rent, data processing, telecommunications, training, transportation, printing, service, supplies and equipment budgets.
All budgets were treated as incurred annual cost and distributed to the five divisions on a staff year basis. Budget figures generally are close approximations to actual cost given the lack of reprogramming authority and incentives to spend authorized budgets completely.
Audit groups and support branches represented the organization entities selected as cost centers. Cost determination for these organizations began with an approximation of payroll based on annualized May data. Benefits cost was based on an average percentage of payroll. All other examinations division cost except rent was distributed to group or support branch on the basis of payroll cost. Analysis of the Facilities Detail Report calculated examinations division rent cost per site. When more than one cost center occupied a site, facilities cost was divided on the basis of staff level, a close approximation of occupied space.(2)
SECOND ABC STAGE TRACES RESPONSIBILITY COSTS TO PROGRAMS: The second stage of a typical ABC system distributes costs of similar activities to products or programs on the basis of causal cost drivers. Two types of cost are considered: direct and support or indirect. The IRS's operational performance measurement systems track direct exam time by program. Direct cost for groups that work on more than one audit program was distributed on the basis of reported direct exam time by program.
Indirect costs were distributed using a level of effort analysis for each support organization. The level of effort exercise required discussion with relevant support management to determine resource consumption by audit programs. Level of effort analyses were performed for the support branches and the computer audit, international, white collar crime, engineering, industry specialist and computer support groups. Process supporting costs such as district overhead and examinations management were allocated on the basis of all other cost.(3)
THIRD STAGE ASSOCIATES COSTS WITH LOCATIONS: The desire for geographical cost and performance measurement necessitated addition of a third stage to the ABC model as shown in Figure 4. Direct audit groups costs were easily distributed to POD since each audit group was physically associated with a location. Support costs were distributed to group/POD on the basis of reported direct exam time per group for each program. This procedure provided a fully burdened cost for each audit group.
IRS operational systems already provided revenue assessment data by group. A performance metric simulating a cost benefit relationship was created by dividing taxpayer assessments by audit cost. This return on investment (ROI) measure was obtained at group level by dividing group revenue assessments by fully burdened group cost. Branch ROI was calculated by adding revenue for all groups in the branch and dividing by the sum of the same groups' fully burdened cost. ROI for an office location resulted from similar addition of revenue from all groups at a location and division by cost from the same groups. Table I shows ROI results of the new system expressed by program and by office. (Table I omitted)
DATA: POSTTEST CHANGES IN MANAGEMENT DECISION MAKING PROCESS
Accounting information from the cost system provided managers with new insights into operations. Information from the decentralized measurement of ROI prompted numerous questions from the division and branch managers. Perceptions of the auditor staffing procedure, facilities utilization and overhead level changed.
AUDITOR STAFFING LEVELS: Variation in performance among audit programs, branch organizations and office locations was noted, questioned and analyzed. For example, the data showed that one location was "losing money." The Fitchburg/Greenfield office posted audit assessments of $.69 for each $1.00 of audit cost. Subsequent review concluded that start up costs, low productivity and high compliance were responsible for the poor performance. The office had been opened recently based on a workload analysis. Management, however, now indicates that the site will be closely watched and closed if ROI continues at current levels.
Performance variation between and within branches and locations suggested additional staffing shifts. For example, Branch 6 Boston and Branch 6 Brockton posted ROI of 1.57 and 7.97, respectively. Moreover, the Branch 4 individual office audit group in Boston recorded a 4.95 ROI. Such variations suggest efficiency improvement by shifting staff from Branch 6 Boston to either Branch 4 Boston or Branch 6 Brockton.
Management review identified two primary reasons for variance in ROI achievement: audit performance and taxpayer compliance. Analysis of the relatively low ROI performance at Branch 4 Lawrence determined that several tax auditors had been diverted to non-audit clerical duties. The non-audit responsibilities were transferred to a support branch improving audit performance by increasing revenue assessments through performance of additional audits.
Recognition of the impact of taxpayer compliance variance encouraged redirection of the audits selected toward local non-compliance issues. For example, the Boston District initiated special audit programs in the fishing industry based on specific, local knowledge of unreported cash transactions. The District also ignored national's suggestion of an excise tax audit of bow and arrow sales due to the low probability of relevance in Massachusetts.
To continue to receive the cost information valuable to audit staffing decisions the division manager further planned the following changes.
1. ROI data would be produced quarterly and trends plotted.
2. Branch managers would explain ROI results quarterly.
3. ROI measures would be added to the annual management by objective goals for branch and group managers.
4. Workload analyses used to determine staffing levels would consider ROI performance.
FACILITIES UTILIZATION: Quantification of facilities costs stimulated interest in reducing facilities costs in favor of other more needed resources. Reduction was considered possible because the division's roughly 300 field agents spend 80 to 90 percent of their time at the taxpayer site. These agents visit their IRS office only one day every week or two. Office sharing or work-from-home programs could save roughly a million dollars per year. To pursue better utilization of facilities space the division manager planned the following actions.
1. With higher approval, branch and group managers would receive increased ability to reprogram spending.
2. A team would make recommendations for reducing facilities space and reprogramming savings into more needed areas.
OVERHEAD SUPPORT LEVELS: In the pre-treatment, initial state, support levels were assumed to be based on the workload measure: number of returns processed. Level of effort analysis, however, developed a causally related base of distribution that reflected the great differences in audit complexity within the division's audit programs. Figure 5 contrasts cost distribution by audit program in the Quality Assurance Branch for the two cost distribution techniques.(figure 5 OMITTED) The level of effort analysis estimated individual office audits' resource consumption at 16 percent instead of the 45 percent indicated by the number of returns processed. The analysis also documented almost 13 percent of effort as supporting small districts outside the Boston District, a fact previously unappreciated.
Support managers commented that the analysis helped them to understand the impact of their work on the audit process. Formerly little distinction existed between line and staff functions since both processed negotiated levels of workload independently of each other. Only when considering the customer relationships inherent in the level of effort analysis did the staff versus line difference become clear. To expand the benefit of this interaction, management planned to extend level of effort analysis to District level support organizations.
RESULTS AND DISCUSSION
The results of the experiment reject the hypothesis that management accounting will not induce change in a public sector organization. This section discusses the changes observed in four areas after the informational intervention: cost awareness of managers, interfaces between organizations, validation of strategy and evaluation of performance.
COST AWARENESS OF MANAGERS: It is difficult to imagine effective cost reduction programs without managerial cost awareness. In the pre-experiment state managers had little awareness of cost. Managers responsible for multi-million dollar per year organizations had no idea that salaries and wages comprised 85 percent of their costs. They estimated only 60 percent.
Management practice reflected the lack of awareness of personnel cost. One manager reported that "we used to always want the highest job grade level people, but now we are moving to lower cost para-professionals." In the past, "we just wanted the best people available" and "a $55,000 per year employee 'costs' as much as a $18,000 per year employee" since each required one staff year of the budget.
Lack of facilities cost information also precluded cost reduction opportunity. Previous to this study the Division Manager did not know that facilities cost almost three million dollars per year. He merely complied with the GSA rules and regulations limiting space per person. Awareness of this cost information led to thinking about cost reduction. While the inefficient use of facilities had existed for years, cost reduction effort began only with cost awareness introduced by the cost accounting system.
Since the demand curve familiar to all who have taken an economics course indicates infinite demand for a free good, it should not be surprising that managers will use more resources if the cost of resources is perceived to be zero. Rules and regulations, such as those limiting space per person, represent an attempt at constraining resource consumption. Unfortunately, rules are a poor substitute for informed decision making. By awakening cost awareness, the cost accounting system stimulated managers to make common sense, situation specific decisions that rules did not, and probably could not, anticipate.
INTERFACES BETWEEN ORGANIZATIONS: Support groups and audit groups operated with little management interaction during the pre-experiment period. Each had a different, work-related, non-financial measurement system unique to its function. The level of effort analysis created a forum for interaction by forcing support managers for the first time to think of the line audit organizations as their customers. Quantification of staff support cost provided line audit managers an opportunity to critically evaluate and influence support practice.
VALIDATION OF STRATEGY: Cost data provided management a new tool for the validation of strategy. For example, the pre-experiment measurement systems had not hinted at the poor financial performance posted by the Fitchburg/Greenfield post of duty. The cost system questioned, for the first time, the deployment of audit resources to the Fitchburg/Greenfield office.
The cost data also introduced a number of other questions concerning resource distribution. It might be asked, for example, why certain locations or audit programs consistently achieved higher ROI levels. If the difference does not appear due to variance in audit execution it may be that there is an intrinsic difference in compliance level that could be attributed to business cycle or demographic differences. When such situations exist, mission effectiveness might be enhanced by shifting audit resources to the locations where "business is better."
Understanding the economics of the audit process might also influence audit strategy if financial measurement determines that a diminishing return to audit labor exists, as some audit managers intuitively believe. For example, if 80 percent of assessments are determined in the first 20 percent of the audit, a strategy of making five times as many quick, "20 percent", audits would increase assessments by a factor of four over current practice.
EVALUATION OF PERFORMANCE: If variance in ROI results cannot be attributed to environmental differences in taxpayer compliance rate, then it might expose differences in management and auditor performance. The Lawrence office, as mentioned earlier, posted half the ROI of some of its counterparts until management review found and quickly resolved inefficient practices. Management attention had not been directed to the problem prior to the introduction of dollar denominated measurements.
The attractiveness of cost-based performance measurement, however, does not appear equal for all levels of management. Some managers saw little added value to their management information needs. Non-financial information appeared adequate for lower level managers focused on a single task at a single location. They also recognized the hazard of providing a superior manager with an additional measurement of their relative performance. The value of dollar denominated measurement may accrue primarily to middle and top level management needing to compare and judge performance of diverse, multifunction, or multi-location operations.
This research asked the question "Does introduction of management accounting information change management decision making in a public sector organization?" In this case it would seem that management accounting prompted considerable change in management decision making.
No claim is made that the system installed here is ideal and exhibits perfect external validity. Indeed, the ROI measure may be justifiably criticized due to the somewhat abstract nature of the revenue assessment measurement. No revenue actually flows to the IRS, and assessments alone do not guarantee collections. Appeals can be filed, assessments can be modified and collection can be avoided.
The ROI measurement, however, provides a powerful synthesis of the IRS audit mission. Similar measurements, however, may have broad external application since almost any government organization could employ similar arbitrary, but reasonable, techniques to generate surrogate, but relevant and situation specific, cost-benefit metrics.
New issues have been added to the Boston District's management agenda in the areas of auditor staff allocation, facilities space utilization and overhead support levels. In adopting and proliferating the system to additional operations under his authority the District Manager offered the following conclusion:
"Our previous method is not the way I'd run my own business. [I] have a responsibility to spend money wisely and I'm not sure that I was. I was only sure that I was spending it."
1. General Services Administration: 41 CFR Part 101-17 "Assignment and Utilization of Space; Temporary Regulation," Section 101-17.200, Federal Register, Volume 56, Number 165, August 26, 1991.
2. No attempt was made to determine depreciation expense. Capital expenditures represented only .2% of total cost and the difference to depreciation expense is undoubtedly even less.
3. Level of effort analysis may require adjustment if significant diversity exists within an overhead function. For example, large salary differences between employees or large non-salary cost differences between programs could have distorted results. According to the managers interviewed such conditions did not apply.
Anthony, Robert and Herzlinger, Regina; Management Control in Nonprofit Organizations, 1975, Irwin Inc., Homewood, Illinois.
Birnberg, J. and Gandhi, N; "Toward Defining the Accountant's Role in the Evaluation of Social Programs," Accounting, Organization and Society, June, 1976.
Caldwell, Charles and Welch, Judith; "Applications of Cost-Volume Profit Analysis in the Governmental Environment," The Government Accountants JOURNAL, Summer Quarter, 1989.
Campbell, Donald and Stanley, Julian; Experimental and Quasi-Experimental Designs For Research, 1963, Houghton Mifflin Company, Boston, Massachusetts.
Congress of the United States, "Chief Financial Officers Act of 1990," 1990, House of Representatives Bill 5687, Library of Congress, Washington, DC.
Cooper, Robin and Kaplan, Robert; The Design of Cost Management Systems: Text, Cases, and Readings, 1991, Prentice Hall, Englewood-Cliffs, New Jersey.
Crowningshield, Gerald; Cost Accounting: Principles and Managerial Applications, Second Edition, 1969, Houghton Mifflin, Boston, Massachusetts.
Dierks, Paul; "Applying Cost Accounting to Transit System Financing," Management Accounting, December, 1978.
Geiger, Dale; Federal Management Accounting, Determinants, Motivating Contingencies, and Managerial Relevance, 1993, unpublished doctoral dissertation, Harvard Business School, Boston, Massachusetts.
Helmi, Medhat: "Deciding On Cuts In Social Programs: An Accounting Framework," The Government Accountants JOURNAL, Summer Quarter, 1987.
Horngren, Charles & Sundem, Gary; Introduction to Management Accounting, 1990, Eighth Edition, Prentice Hall, Englewood-Cliffs, New Jersey.
Johnson, Thomas; "Managing By Remote Control: Recent Management Accounting Practice in Historical Perspective," 1991, in Inside the Business Enterprise: Historical Perspectives on the Use of information, edited by Peter Temin, University of Chicago Press, Chicago, Illinois.
London, Rosanne; "Determining Privatization Feasibility," The Privatization Review, Summer, 1998.
Nestor, Joseph; "How Cost Accountants Can Improve Public Housing Programs," Management Accounting, October, 1979.
Schattke, Rudolf and Jensen, Howard; Managerial Accounting: Concepts and Uses, 1981, Allyn & Bacon, Boston, Massachusetts.
Simon, Herbert; Guetzkow, Harold; Kozmetsky; and Tyndall, Gordon; Centralization vs. Decentralization In Organizing The Controller's Department, 1954, Scholars Book Co., Houston, Texas.
Sterling, Robert; Theory of the Measurement of Enterprise Income, 1961, University Press of Kansas, Lawrence, Kansas.
Dr. Geiger is an Assistant Professor of Accounting at California State University, San Marcos, and a member of AGA's San Diego Chapter.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: An Experiment in Federal Cost Accounting and Performance Measurement. Contributors: Geiger, Dale R. - Author. Journal title: The Government Accountants Journal. Volume: 42. Issue: 4 Publication date: Winter 1994. Page number: 39+. © Association of Government Accountants Fall 1996. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.