Activity-based costing (ABC) is a method of allocating costs to products and services. It is generally used as a tool for planning and control. This is a necessary tool for doing value chain analysis.

The concepts of ABC were developed in the manufacturing sector of the U.S. during the 1970s and 80s. During this time, the Consortium for Advanced Manufacturing-International, now known simply as CAM-I (www.cam-i.org), provided a formative role for studying and formalizing the principles that have become more formally known as Activity-Based Costing. Robin Cooper and Robert Kaplan, proponent of the Balanced Scorecard, brought notice to these concepts in a number of articles published in Harvard Business Review beginning in 1988. Cooper and Kaplan described ABC as an approach to solve the problems of traditional cost management systems. These traditional costing systems are often unable to determine accurately the actual costs of production and of the costs of related services. Consequently managers were making decisions based on inaccurate data especially where there are multiple products.

Instead of using broad arbitrary percentages to allocate costs, ABC seeks to identify cause and effect relationships to objectively assign costs. Once costs of the activities have been identified, the cost of each activity is attributed to each product to the extent that the product uses the activity. In this way ABC often identifies areas of high overhead costs per unit and so directs attention to finding ways to reduce the costs or to charge more for costly products.

Activity-based costing was first clearly defined in 1987 by Robert S. Kaplan and W. Bruns as a chapter in their book Accounting and Management: A Field Study Perspective. They initially focused on manufacturing industry where increasing technology and productivity improvements have reduced the relative proportion of the direct costs of labor and materials, but have increased relative proportion of indirect costs. For example increased automation has reduced labor, which is a direct cost, but has increased depreciation, which is an indirect cost.

Traditionally cost accountants had arbitrarily added a broad percentage onto the direct costs to allow for the indirect costs. However as the percentages of overhead costs had risen, this technique became increasingly inaccurate because the indirect costs were not caused equally by all the products. For example one product might take more time in one expensive machine than another product, but since the amount of direct labor and materials might be the same, the additional cost for the use of the machine would not be recognised when the same broad 'on-cost' percentage is added to all products. Consequently, when multiple products share common costs, there is a danger of one product subsidising another.

Like manufacturing industries, financial institutions also have diverse products which can cause cross-product subsidies. Since personnel expenses represent the largest single component of non-interest expense in financial institutions, these costs must also be attributed more accurately to products and customers. Activity based costing, even though developed for manufacturing, can therefore be a useful tool for doing this. This extended use of ABC to financial institutions was presented in 1990 in an article appearing in the Journal of Bank Cost and Management Accounting (Volume 3, Number 2) by Richard Sapp, David Crawford and Steven Rebishcke. There was also a subsequent article in 1991 in the same Journal (Volume 4, Number 1).

Direct labor and materials are relatively easy to trace directly to products, but it is more difficult to directly allocate indirect costs to products. Where products use common resources differently, some sort of weighting is needed in the cost allocation process. The measure of the use of a shared activity by each of the products is known as the cost driver. For example, the cost of the activity of bank tellers can be ascribed to each product by measuring how long each product's transactions takes at the counter and then by measuring the number of each type of transaction.