This thesis examines the usefulness of estimates of systematic equity risk derived using accounting variables. It is the first study in this area to combine accounting variables derived from a theoretical model with the most appropriate estimation technique, developed by Rosenberg and McKibben (1973). Using a sample of 53,471 US company-quarter observations, this thesis shows that the fundamental equity beta, which is based on the coefficient of variation in sales, and the degrees of operating and financial leverage, is a useful measure of systematic equity risk. Specifically, the fundamental beta is found to be substantially more stable over time, and has fewer extreme observations, compared to the standard beta estimate. The fundamental beta also provides a more accurate estimate of systematic equity risk compared to the standard estimate for small firms. Therefore, there are a number of important applications of the fundamental beta. It can be used as an alternative risk measure for small firms. Even for larger firms, the fundamental beta may be useful, as many applications of beta require an estimate that is steady over time. Further, the fundamental beta can be used for unlisted and newly listed businesses, which is not the case for the standard estimate. The fundamental betas are also potentially useful for government regulators in price-setting decisions due to their stability.