The Intangible Costs and Benefits of Computer Investments: Evidence from the Financial Markets

Abstract

We show how the financial market valuation of firms can be used to estimate the intangible costs and benefits of computer capital and we present several new empirical results based on this model. Using eight years of data for 820 non-financial firms in the United States, we find that an increase of one dollar in the quantity of computer capital installed by a firm is associated with an increase of about ten dollars in the financial markets' valuation of the firm. Other forms of capital do not exhibit these high valuations. Our model suggests that adjustment costs and intangible assets may provide an explanation for the high market valuation found for computers in this study as well as the high returns found for computer capital in firm-level productivity studies. Costly investments in software, training and organizational transformations that accompany computer investments can be regarded as creating intangible assets. These intangible assets do not appear on firms' conventional balance sheets but they can produce both higher market valuations and apparent "excess" returns. The empirical evidence suggests that the vast majority of the costs and benefits of computerization are embodied in otherwise unobserved intangible assets. We thank Brownyn Hall, Robert Hall, Lorin Hitt, Stephen Oliner, and participants of NBER Productivity Seminar for valuable comments. The National Science Foundation, the MIT Center for Coordination Science, the Stanford Computer Industry Project, NationsBanc Montgomery Securities and IBM Research provided generous financial support and International Data Group and Computer Intelligence Corporation provided valuable data. The authors are responsible for any remaining errors. This paper subsumes a previous version that was presented at the Internat...