A cross-sectional analysis of cash compensation paid to CEOs of 713 US firms reveals that the sensitivity of compensation to earnings varies directly with earnings persistence. Additional analysis indicates that this sensitivity is greater for cases where executives are approaching retirement. Such evidence suggests the use of earnings persistence to counterbalance adverse consequences of earnings-based contracting with managers who face finite decision horizons.

Alfred Cowles' test of the Dow Theory apparently provides strong evidence against the ability of Wall Street's most famous chartist to forecast the stock market. Cowles (1934) analyzes editorials published by the chief exponent of the Dow Theory, William Peter Hamilton. We review Cowles' evidence and find that it supports the contrary conclusion. Hamilton's timing strategies actually yield high Sharpe ratios and positive alphas for the period 1902 to 1929. Neural net modeling to replicate Hamilton's market calls provides interesting insight into the Dow Theory and allows us to examine the properties of the theory itself out of sample.