| Technology has been one of the most important sources of economic growth in the past. It is crucial in the new economy. Yet financial statements provide very limited information on firms' technological assets. In fact, the accounting treatment of R&D expenditures, i.e., expensing as incurred, distorts reported performance measures and makes it difficult to evaluate returns from firms' investment in technology and innovation. What is more disturbing is some preliminary evidence that the lack of disclosure in financial reports on companies' innovative activity may be costly to participants in capital markets.;This study examines the potential of nonfinancial indicators as measures of innovative efforts in the valuation context. Specifically, patent statistics are used as measures of firms' technological characteristics related to patented innovation assets, which are hypothesized to have implications for firms' future performance. Given the richness, abundance, and quantitative nature of patent data, it is hypothesized that patent statistics contain value relevant information above and beyond reported R&D expenditures. Secondly, it is hypothesized that, since patent statistics are not products of the accounting system, they are not subject to the limitations of financial accounting information. In other words, in situations where a firm's valuation appears to be unrelated to its financial performance measures, patent statistics are hypothesized to provide useful information for valuation purposes. Thirdly, it is hypothesized that accounting treatment of innovation assets, i.e., non-recognition on the balance sheet, has a negative impact on the information content of earnings. Empirical findings are generally consistent with the above hypotheses.;These results hold after controlling for known confounding factors for which patent statistics may be proxies, such as other intangible assets, firm size, growth rates, spillover of other firms' R&D, or industry effects. Interestingly, the incremental information content of patent statistics remains in regressions that include consensus analysts' forecasts, which have been used as a proxy for other information used by stock markets. |
