For a private sector firm, Robert Kaplan (http://www.people.hbs.edu/rkaplan ) argues that traditional cost accounting systems distort cost information by weighting direct labor and materials more heavily than factory support operations, thus giving management inferior information on which to base decisions on pricing, product mix, and process technology. In contrast, ABC is a sophisticated approach that offers business managers accurate information, by delineating support costs and tracing them to individual products and product lines. Designing an ABC system entails collecting data on direct labor and material costs and analyzing the demands made by particular products on indirect resources. However, in the private sector, ABC lost ground in the 1990s after initial enthusiasm for it (see article by D. Katz http://www.cfo.com/article.cfm/3007694).
In the public sector, ABC has been applied to costing at the budget program level or at the level of individual services or activities. At a first stage, ABC costs broadly-defined activities or grouping of work tasks of a similar type. At a second stage, the costs of these activities are allocated to programs (or specific outputs within programs). To ensure accuracy, the cost drivers used in the second stage are based upon the quantity, number and resource-intensiveness of each type of service within a program.
Those who advocate use of ABC in the public sector see it as relevant to performance budgeting in two ways: first, as a tool for better expenditure prioritization; and second, as a means of developing tighter linkages between planned outputs and funding. Supporters of ABC argue that it is essential for the success of performance-based budgeting. Clearly, more accurate program costing improves the quality of expenditure prioritization choices. However, as noted in chapter 4 of Marc Robinson’s IMF book performance based budget (see posting of October 4, 2007 blog-pfm.imf.org/pfmblog/2007/10/making-performa.html), there is a dearth of systematic analysis of experience with ABC and no consensus as to ABC’s success in the public sector.
In the USA, for example, although it has been used by many local governments – and continues to be used at that level and in some federal agencies -- it has also been recognized that ABC is difficult to implement. Several government organizations that experimented with ABC have not continued with it, as ABC has been found to be a very high-cost accounting technology. Installing an ABC system is technically complex, requiring talented personnel and a considerable amount of time and money.
In the United Kingdom, from 2003-04, ABC was a mandatory requirement of England and Wales’ National Policing Plan and was regarded as a key component of the Policing Performance Assessment Framework. It was perceived that ABC would enable accurate comparison between police forces (ABC works best when comparator benchmarks are established) and would provide central assistance to weaker forces. A 133-page Guidance Manual for using ABC in the police was disseminated (http://police.homeoffice.gov.uk/finance-and-business-planning/index).
In an independent review of policing in April 2008 it was however found that ABC, as a comparative tool for improving productivity in the police, did not represent good value for money. Two strong recommendations were made. First, that the U.K.’s Home Office should urgently examine its requirement for each police force to undertake ABC, with a view to this requirement being replaced with an alternative that costs less, is easier to use, and has greater impact on productivity. Second, the police services were encouraged to assess alternative ways of meeting its information requirements regarding the allocation of police funding. The rationale for the recommendations can be found in the Final Report by Sir Ronnie Flanagan which, along with the Home Secretary’s initial response to the recommendations, can be found on http://police.homeoffice.gov.uk/police-reform/flanagan-police-review.
This evidence provides some support to the critics who have argued that in most instances ABC is not appropriate for public sector applications. Is this more extreme view fully justified? Or is there a case for using ABC selectively? For example, could it be successfully applied for the production of relatively standardized services? If yes, should it be largely reserved for use by local governments, where the services are relatively homogenous?
Although these questions remain open for debate, the recent report from the United Kingdom suggests that the putative benefits of ABC need to be strongly justified, given the considerable costs of the setting up and operating an ABC system.
One of the principal objectives of a performance-oriented budgeting system is to link the results and the funding of budget programs. In order to make well-informed allocative choices in the national budget, it is important to obtain the best possible information on program costs. However, it is costly to obtain detailed and accurate program costing information. Compromises need to be made when choosing between simple methods of costing and more sophisticated models. 