Three methodological perspectives are used to investigate the common stock price response to the bond reclassifications of 180 New York Stock Exchange firms for the period 1960-1975. They comprise a (traditional) residuals approach, a multivariate comparison-of-returns approach and a rational distributed lag-transfer function approach. The transfer function technology represents a general linear structure in which to view the analysis of information effects. Due to estimation and specification problems associated with the use of market models, the residuals-based tests were unable to detect significant information effects. However, the multivariate T2 tests, utilizing differences in return distributions, point unambiguously to the conclusion that the rating reclassification reveals important information. A two-parameter transfer function model is developed and tested. By the use of finite lead polynomials, the model is able to relate current output (returns) to conditioning agents on that output (information) in the current period and in all future periods but with decaying emphasis.