Predicting CEO Compensation in Non-Controlled Public Corporations with the Canonical Regression Quantile Method

Empirical research on the relationship between CEO compensation and corporate performance has, as of yet, failed to yield meaningful results. Evidence instead points to the decoupling of CEO total compensation (salary, bonus, other payments, and stock options) from the corporate performance. Recent work by Barton et al (2017) suggested a predictive model where public corporations engaging in long-term operational and financial goals achieve superior performance and market results in the long-term. The suggested predictive power of this long-termism could provide a quantitative measure of future corporate performance that is tied to current CEO action. If true, this model can be used as a proxy to establish future CEO compensation. Corporate ownership structure can influence CEO compensation to reflect ownership powers rather than corporate performance. Therefore, we limited our research to non-controlled public corporations – those with a single class of shares and without blockholders. The analysis here uses Canonical Regression Quantiles to produce an Index based on past data in order to predict CEO compensation two years ahead (see Portnoy and Haimberg, 2020). The regressions on Index are significant, but not overwhelmingly so, but provide a useful tool to separate over-compensated from under-compensated CEOs, and are a method to determine current CEO compensation. Results: The use of the Canonical Regression Quantiles’ Index proved that non-controlled companies that invest in long-termism post superior future performance. The Index indicates that current CEO compensation influences future performance. The Index provides a method for determining CEO pay for the next 1-2 year and is a useful method to distinguish over/underpaid CEOs as an unbiased alternative to the peer groups comparison used by most compensation consultants. This determination is statistically weak, but future research using the Canonical Regression Quantiles with a larger data set may lead to increased sensitivity and a powerful unbiased method for replacing compensation consultants who are responsible for the decoupling of CEO compensation and corporate performance