Importance of measures of past performance: Empirical evidence on quality of e-service providers

We investigate the relative importance of alternative measures of past performance that signal the quality of managers, workers, or other. service providers and determine the fixed (noncontingent) component of their compensation. We appeal to a model of adverse selection to motivate empirical hypotheses about differences in the importance of various signals depending on the nature of the service. Growth of web-based markets enables the collection of a set of quasi-experimental data to test those hypotheses. Specifically, we evaluate the importance of different quality signals in an online auction market for accounting services in the buyer's assessment of service provider quality. We identify various quality signals about accounting service providers, who bid on bookkeeping and tax service projects. posted at an e-marketplace. We find that signals based on service providers' past performance have a significant impact on buyers' award decisions. Our results support theoretical predictions that signals pertaining to service providers' technical competence, such as professional certification, are more important for tax services than for bookkeeping services, whereas signals pertaining to service providers' quality in interacting with clients are important for both bookkeeping and tax services.