Pii: S0950-5849(99)00091-9

Software is valuable when it produces information in a manner that enables people and systems to meet their objectives more effectively. Software engineering techniques have value when they enable software developers to build more valuable software. Software economics is the sub-field of software engineering that seeks improvements which enable software engineers to reason more effectively about important economic aspects of software development, including cost, benefit, risk, opportunity, uncertainty, incomplete knowledge and the value of additional information, implications of competition, and so forth. In this paper, we survey the current status of selected parts of software economics, highlighting the gaps both between practice and theory and between our current understanding and what is needed. The sheer volume of current software costs makes the study and application of software economics techniques a significant area of concern and opportunity. Recent studies [1,2] estimate roughly 2,000,000 software professionals in the US in 1998. At typical salaries of $60–80,000/year and a typical overhead rate of 150%, this translates into a $300– 400 billion annual expenditure on software development in the US alone. A conservative estimate of worldwide software costs is twice the US costs, or $600–800 billion per year. With the kind of expenditures now being made on software, just the economics of more efficient software production are important to understand and apply. Software development is widely seen to be inefficient, which means that there is considerable room for improvement. A 10% reduction in software production costs translates into a $30–40 billion annual savings in the US, and $60–80 billion worldwide. One cause of inefficiency is the inadequacy of economic information available to support decision making. In the absence of reasonable cost estimates, for example, projects are at risk. Improving our basic understanding of software economics and the abilities of organizations to model and analyze economic aspects, can help to make them significantly more productive. A simple example: once organizations appreciate that late defect fixes are more expensive than earlier fixes, they can reduce their costs by reorganizing their process to focus on early defect prevention and elimination. The software engineering activity centers on decision making in very complex situations. What requirements are feasible given certain resource limitations? How much should one bid on a project? How should one modularize a design to create opportunities and to reduce future costs? How can processes be structured to dramatically improve cycle times so as to capture profits in ephemeral markets? Such decisions can be addressed more effectively using economics-based techniques; one definition of “economics” is “the study of how people make decisions in resourcelimited situations.” An important problem is to empower high-level managers to choose the best available economic reasoning techniques for use in their projects. This paper is meant to help researchers and managers by providing an overview of the current state and prospects for software economics. We address three main questions. In Section 2 we discuss how we have done up to now in software economics, and what forces are driving the need for new work. Section 3 focuses on the current state of our knowledge and practice in cost and schedule estimation, in particular, and in decision support, more broadly. Section 4 closes with a more general perspective of future trends.