We develop a series of hypotheses that predict that the appropriateness of different business strategies is conditional on the firm's relative performance, or distance to the "industry frontier." We use data on three 2-digit high-tech manufacturing industries in the United States over the period 1972-1999, and apply semi-parametric quantile regressions to investigate the contribution of firm behavior to market value at various points of the conditional distribution of Tobin's q. Among our results, we observe that innovative activity, measured in terms of R&D expenditure or patents, has a strong positive association with market value at the upper quantiles (corresponding to the leader firms), whereas the innovative efforts of laggard firms are valued significantly less. Laggard firms, we suggest, should instead achieve productivity growth through efficient exploitation of existing technologies and imitation of industry leaders. Employment growth in leader firms is encouraged, whereas growth of backward firms is not as well received on the stock market. © The Author 2011. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved.