T-Space at The University of Toronto Libraries >
School of Graduate Studies - Theses >
Please use this identifier to cite or link to this item:
|Title: ||Categories and Evaluation Bias in Valuation of Technological Innovation|
|Authors: ||Geng, Xuesong|
|Advisor: ||Silverman, Brian S.|
|Issue Date: ||3-Mar-2010|
|Abstract: ||This dissertation examines the perceptual bias of investors and securities analysts (the “audience” in the stock market) in their valuation of public firms’ innovative activities. I suggest that such bias occurs because the audience views a firm’s innovation through the prism of the firm’s categorization in product markets – its industry category – which may only loosely conform to the technological interrelationship among firms in knowledge space. I explore theoretically the conditions under which evaluation bias is most likely to occur – notably, due to innovations that defy the existing categorical structure used by the audience.
Based on this theoretical framework, I develop hypotheses for empirical tests. I first argue that both technological opportunities and technological threats residing outside a firm’s industry are more likely to be underestimated by the stock market than those residing within the industry. In a sample of large U.S. manufacturing firms covering the years 1980 through 2000, I collected patent data to reflect innovative activities of firms. I compare the effect of a firm’s innovative activities on its current market valuation and its future cash flows. Consistent with my predictions, I find that firms capitalizing on cross-industry opportunities are more likely to be undervalued, while firms facing cross-industry technological competition are more likely to be overvalued.
I further argue that reliance on categorical identity for information cues may reduce the audience’s ability to adequately assess the value-relevance of innovations that deviate from the technological norm in an industry. By analyzing the absolute forecast errors in security analysts’ reports, I find confirming evidence that deviant innovations increase the forecast bias. I further argue and demonstrate that such bias is less prevalent for diversified firms and in industries with less stable categorical “norms,” two conditions in which the audience is less likely to rely on categorical information and more likely to employ firm-specific information. Finally, I discuss the contribution and implication of the findings to studies of categorization and value-relevance of technological innovations.|
|Appears in Collections:||Doctoral|
Joseph L. Rotman School of Management - Doctoral theses
This item is licensed under a Creative Commons License
Items in T-Space are protected by copyright, with all rights reserved, unless otherwise indicated.