Share Price Levels and Beta

TLDR: In this paper, the authors considered combined timeseries and cross-sectional data for two groups of stocks: a non-split group and a split group, and found that only the nonsplit group will show a negative relationship between share price and risk when risk is properly defined as systematic.

Abstract: In their classic guide for security analysis, Graham and Dodd warn that stocks with a very low price per share are more risky than stocks that sell at higher prices [6, p. 649]. The earliest empirical finding on the relationship of risk and share price was consistent with their warning [5]. Subsequent work has suggested that the warning and the early empirical work may be misleading and that there is no relationship between risk and share price. Strangely enough, both those who warn of a relationship and those who claim there is none may be correct. If a company experiences an unexpected change in its circumstances that substantially increases the systematic risk of its common stock, then, accepting the Capital Asset Pricing Model, the price of its shares will fall. If the change substantially reduces systematic risk, then share price will rise. The observed result for a group of different companies with different changes could be just what Graham and Dodd suspected: lower share price associated with higher risk. If, in contrast, some managements searching for the optimal price range lower share price by splits, then risk should not differ between stocks with low and with high share prices. In this study we have considered combined timeseries and cross-sectional data for two groups of stocks: a non-split group and a split group. Our hypothesis is that only the non-split group will show a negative relationship between share price and risk when risk is properly defined as systematic. The findings are consistent with our hypothesis.